http://www.caeda.com.cn/blog/ zh-cn PBlog2 v2.4 http://www.caeda.com.cn/blog/images/logos.gif http://www.caeda.com.cn/blog/ CAE Blog http://www.caeda.com.cn/blog/article.asp?id=153 caeda@vip.sina.com(admin) Tue,02 Mar 2010 10:11:58 +0800 http://www.caeda.com.cn/blog/default.asp?id=153
Qontix Computational Libraries provide optimized suite of computational support libraries routines for technical and scientific applications. Qontix Computational Libraries currently feature computational libraries including:

HDF5
netCDF
These libraries have been tested and proven on a wide range of architectures.  Qontix Computational Libraries have been engineered to support 64-bit data access (ILP64) where applicable and compatible with MPI implementations including HP MPI, Intel MPI, Open MPI, amongst others.
--------------------------------------------------------

Qontix Numerical Libraries
Qontix Numerical Libraries provide highly optimized, extensively threaded mathematical routines for technical and scientific applications. Qontix Numerical Libraries feature a comprehensive set of mathematical libraries, including BLAS, LAPACK, ScaLAPACK, SuperLU, PETSc, PeIGS, and FFT.  

Qontix Numerical Libraries have been tested and proven to deliver high performance on a wide range of architectures.  Qontix Numerical Libraries have been engineered to support 64-bit data access (ILP64) and 256-bit precision (quad-precision), compatible with MPI implementations including HP MPI, Intel MPI, Open MPI, amongst others

---------------------------------------------

Qontix Performance Tools
Qontix Performance Tools provide commercially supported versions of performance monitoring and analysis tools to assist developers in the optimization process for large scale high performance computer system.  Qontix Performance Tools include:

Open|SpeedShop
Valgrind
PAPI
DynInst
KOJAK
SCALASCA
These performance monitoring and analysis tools have been proven on high end supercomputer systems.  Developers and end users can rely on Qontix support in critical, strategically and tactically vital areas of operations 点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=152 caeda@vip.sina.com(admin) Tue,16 Feb 2010 09:06:46 +0800 http://www.caeda.com.cn/blog/default.asp?id=152

The main enhancements / changes:



TOSCA ANSA environment:

- Update to Version 13.0.3 of TOSCA ANSA environment

- Import of TOSCA Structure Parameterfiles to Task Manager

- Optimization Monitor including Convergence Plot, VTF generation and TOSCA.OUT viewer

- Improved Highlightning

- Tool Tips in preprocessing dialogs

- Support of multiple cutting splines in one smooth run



TOSCA Structure.topology:
- Improvements of manufacturing constraints for sensitivity based topology optimization



TOSCA Structure.shape:

- Improved Mesh Smooth algorithm

- Inner nodes are allowed in design area to ease problem definition



Plattforms:

- Support of Windows 7.


Solver changes:

The interface to PERMAS and MARC was adapted to the functionality of TOSCA Structure Release 7.0.0. This includes the support of contact definitions and displacement constraints for the sensitivity based topology optimization.

For subsequent versions of TOSCA Structure these solver interfaces will be frozen with functionality of

TOSCA Structure 7.0.1. The interfaces will be available in the current version as as-is interface. No further developments and enhancements will be made for this interface.



For more detailed information please see the attached release and platform notes.
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=151 caeda@vip.sina.com(admin) Mon,01 Feb 2010 10:03:23 +0800 http://www.caeda.com.cn/blog/default.asp?id=151
==================
Multicriteria design optimization software - IOSO NM.
IOSO NM - Universal software for complex real-life optimization problems.
Unique capabilities of IOSO NM allow:

to increase considerably the efficiency of the object under optimization and obtain technical solutions having no analogues by means of multifactor (up to 100 variables and up to 100  constraints) and multiobjective (up to 20 objectives) optimization;

to reduce considerably the terms and expenditures of the research;

to create supporting versatile systems for decision makers;

to link and solve tasks being computed by different program modules (multidisciplinary optimization) within the common project.
  


IOSO NM software can be easily integrated with different applications for engineering analysis both in-house and commercial, such as NASTRAN, ANSYS, StarCD, FineDesign, Fluent etc.

Novelty and distinctive features of IOSO NM:

multiobjective optimization for large-dimensionality problems (up to 100 independent design variables and up to 100 constraints), which allows to reach the increase of efficiency by 2 - 7 times higher than that of middle-dimensionality optimization tasks (20…40 design variables);

low expenditures for optimal solution search (reduction of the number of analysis code direct calls calls up to 20 times in comparison with traditional approaches and genetic algorithms (GA), depending on the complexity and dimensionality of the task);

full automatic optimization technology algorithms with easy to use procedure of task setting;

the possibility to solve multidisciplinary optimization problems;

multiobjective optimization for stochastic problems (up to 100 independent design variables), having complex topology of objective and the large number of constraints. Now it is well-known that many methods are capable of solving the tasks having up to 10 - 20 variables, and it is not known the analogues to IOSO  optimizer that is designed for large-dimensional multiobjective tasks (100 independent design variables and 20 objectives);

solving all classes of optimization problems including stochastic, multiextreme and having non-differential peculiarities.




IOSO NM is successfully used for solving practical tasks for such companies as: NPO “Saturn”, Sukhoy’s Design Bureau and other enterprise companies. The software is implemented in these company's technological process of the development of complex, competitive engineering objects, it is used for solving following problems:

obtaining the most efficient technical solutions by means of the optimization of the models of large-dimensional systems providing extremum for the one or several efficiency indices of the system;

determining optimal control laws for complex devices and for their different operational modes;

comparative analysis of alternative optimum variants and the substantiation of choice of technical solution.
  

The design optimization software package IOSO NM operates on PC under MS Windows NT/2000/XP operation systems and can be integrated within the network with analysis modules running under different operational systems, such as Windows, families of Unix or Linux (using TCP/IP data transfer protocol).
]]> http://www.caeda.com.cn/blog/article.asp?id=150 caeda@vip.sina.com(admin) Sun,17 Jan 2010 08:53:56 +0800 http://www.caeda.com.cn/blog/default.asp?id=150 Geometry

new module, introducing new functionality with potential for many further features
new STEP and IGES file format import capability
automatic mesh conversion with surface sew function
automatic Element Set creation for imported surfaces, for easy selection

Design
added draping warp and weft angle deviation graphics
implemented Element Set picking in graphics
corrected problems with Reverse ply application in Layup
re-introduced STL format import node merge tolerance, added edge sew tolerance

Manufacturing
added support for generation of exact innermost / outermost layup surfaces as point clouds
improved exact flat pattern calculations by preserving surface normals at ply edges

Analysis
implemented surface offset extrusion for Nastran solid elements

supported sandwich configurations for solid element extrusion with 3 elements through thickness

automated the 3D solid element stiffness property calculation and export

enabled Nastran/Optistruct exports to merge with any Nastran data file
upgraded Optistruct SET export format

forced laminate angles within -90..+90 for the Nastran First Ply option

improved support for Ansys LOCAL axis systems and Ansys Existing laminate generation

Post-processing
improved stiffness calculations for anisotropic materials, for interactive calculations
improved automatic min-max range for stress or strain contour plots
User Interface
enabled all windows to be considered when optimising desktop layout

added gradient background colour in graphics windows

allowed the opening of multiple files via Windows Explorer, in separate threads

added diagnostics for licencing operations

implemented licence update via user interface

improved status bar update for progress messages

implemented custom toolbar positioning
implemented html help in addition to standard Windows help

added numerous visual and efficiency improvements

PlyMatch
relaxed tolerance of camera movement for graphics update, for smoother animation
]]> http://www.caeda.com.cn/blog/article.asp?id=149 caeda@vip.sina.com(admin) Fri,15 Jan 2010 15:35:36 +0800 http://www.caeda.com.cn/blog/default.asp?id=149


History
Main article: NASTRAN
The original NASTRAN program came out of NASA’s need to develop a common generic structural analysis program that would be used by all of the centers supporting the space program. A specification was written and a contract was awarded to Computer Sciences Corporation for development of NAsa STRuctural ANalysis (NASTRAN) software. NASTRAN was released to NASA in 1968. In addition to NEi Nastran, commercial versions of are also available from Siemens PLM Software (NX Nastran) and others.

Improvements
Main article: Finite element analysis
In the late 1960’s, Finite Element Analysis software was confined to run on expensive mainframe computers and highly trained specialists were needed to apply the program. In this environment, the aerospace industry was the typical user because they had critical projects which could justify the resources FEA demanded. With improvements to the software and wider use of mainframes, FEA technology gradually spread to large corporations that could afford funding the huge investment in hardware, software, and a dedicated FEA staff. Usage spread from primarily aerospace and military applications to the automobile and maritime industries.

The microprocessor revolution and the advent of Personal Computers (PCs) in the 1980’s brought tremendous improvements in computing power, significant reductions in computing costs, and the steady development of numerical methods and algorithms. In the mid 1980’s, Noran Engineering recognized the long term advantages and impact that the PC hardware revolution could have on the engineering analysis field and embarked on a project to significantly enhance and modernize the original NASTRAN code and port it to PCs.

The first commercial version of NEi Nastran for use on PCs was released in 1990. The new code had a number of changes in architecture and programming language compared to legacy Nastran written originally for mainframes. These differences were intended to take advantage of the dramatic changes in computer hardware taking place and provide the code with key strategic advantages for the new PC platform. For example, since the cost of memory was dramatically reduced it was feasible to perform many operations faster in memory that normally were only done on disk.

Another example concerns Nastran’s DMAP (Dynamic Matrix Abstraction Process). DMAP altars were originally intended to provide flexibility and customization capabilities. However, the method has lost much of its initial intent and advantage since it now requires familiarity with arcane computer code and maintenance revisions must be written each time a new Nastran software version is released. With NEi Nastran, DMAP routines are easily added to the main body of the program and maintenance revisions are not necessary.

Present day

NEi Nastran ScreenShotNEi Nastran V9.2 was released in March 2009. It incorporates over 85 customer driven enhancements including the following additions: nonlinear composite Progressive Ply Failure Analysis (PPFA), concrete material model, direct enforced motion, bolt preload, enhanced rigid element features, visualization support for various entities, automatic dynamic plots during nonlinear analysis, transparent max/min, and a new look and feel for its Editor tool.[1]

Applications
Main Products
NEi Nastran
NEi Nastran is a powerful, general purpose finite element analysis tool with an integrated graphical user interface and model Editor which is used to analyze linear and nonlinear stress, dynamics, and heat transfer characteristics of structures and mechanical components. With specialized functions for Automatic Surface Contact Generation and Progressive Ply Failure Analysis, the software is frequently used for aerospace, maritime and automotive industries, particularly for analyzing composites. Features include control over mesh size and sensitivity, and the repair and editing of CAD geometry. NEi Nastran also offers Automated Surface Contact Generation and Automated Edge Contact Generation that allow separate meshing of components, mid-surfaced models, and off-set welds, then combine the meshes via proximity and user-defined settings.[2]

NEi Works
NEi Works provides SolidWorks users with an embedded finite element modeling tool. It features the SolidWorks look and feel for all menus and functions, as well as integration between design and analysis. NEi Works claims true geometry associativity, which updates loads, boundary conditions and meshes interactively whenever changes are made in SolidWorks.

NEi Fusion
NEi Fusion combines a 3D CAD modeler powered by SolidWorks, pre- and post-processing capabilities, and Nastran solvers to create an analysis package that is geared towards designers, consultants, and engineers.

Solutions
NEi Nastran
FEMAP Modeler
NEi Fluid Dynamics
NEi Thermal Basic
NEi Thermal Advanced
NEi Fatigue
NEi Aeroelasticity
NEi Optimization
LS-DYNA
NEi Motion
NEi Acoustics
SmartBrowser
NEi Nastran for Pro/Engineer 点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=148 caeda@vip.sina.com(admin) Wed,13 Jan 2010 08:49:15 +0800 http://www.caeda.com.cn/blog/default.asp?id=148 Setting up and launching a parallel simulation has become much simpler and easier in CFD-FASTRAN V2007. This note discusses some of these developments.

Let’s start with things that have not changed. There are still two versions of CFD-FASTRAN solvers for parallel cases. The difference between the two versions is the underlying parallel communication mechanism, the choice of which is decided by the type of mesh used. Structured mesh models use MDICE (Multi-Disciplinary Computing Environment) and unstructured mesh models use MPI (Message Passing Interface). MDICE allows different applications to “talk” to each other and hence allows multi-disciplinary simulations such as coupled flow-structure simulations possible without integrating all features (disciplines) into a single code.

The following steps have been removed from V2007:

Need to run special scripts as part of the installation procedure
Need to launch MDICE registry and daemons
Need to launch parallel job only from the command prompt using different scripts such as CFD-FASTRAN-SOLVERP and fastran-mpi
These operations are automatically performed by the script behind the scenes, and hence the user does not have to perform these steps.

The following options have been added to setup parallel cases for CFD-FASTRAN:

Can setup both MDICE-based and MPI-based simulations using the improved parallel configuration panel (located on the Run tab)
Can launch a parallel job (MDICE or MPI) from CFD-FASTRAN-GUI or command prompt
A single command-prompt command for all (serial and parallel) FASTRAN job submissions (CFD-FASTRAN-SOLVER –dtf <file name>)
Can easily edit available hosts through CFD-FASTRAN-GUI
Now, we look at some of the settings available in the new Parallel Setup Panel. This panel can be reached through the Configure Parallel Run option available in the Run tab when parallel is selected as the Run Type. Depending on the type of mesh used, the MDICE or MPI panel will become available. The Host List can be edited by clicking on the Edit button on the left hand side of the parallel setup panel.

In MDICE-based parallel simulations, the Distribute Zones button in the Zone Assignment box performs load balancing based on the number of cells. This just assigns available zones to different nodes and no domain decomposition is performed. The load factor column gives an indication of the maximum speed up that can achieved through a particular distribution. The user can review this and reassign the zones to get as close to a load factor of 1 for all zones. Therefore, it is advisable that the user plans ahead to avoid the possibility of having large zones that tilts the load factor unfavorably. Also note that the chimera and stress zones should be assigned to an exclusive processor. At this time, chimera and stress calculations are performed in serial only.

In MPI-based parallel simulations, domain decomposition is performed using dtf_decompose. Several options for domain decomposition are available. The Multilevel Graph Partitioning Scheme provides high quality partitioning with less CPU time and performs work load balancing. If you would like to know more about domain decomposition, please follow this link.
]]> http://www.caeda.com.cn/blog/article.asp?id=147 caeda@vip.sina.com(admin) Fri,11 Dec 2009 08:42:34 +0800 http://www.caeda.com.cn/blog/default.asp?id=147
- The data can be output using SALT in any format required (text or
binary). Documentation of the SALT<->RTM-Worx interface is almost
finished; it will be on the download page within a few weeks.

- 3D will be finished soon.

- Capillary force on resin flow is very small and does not have
significant effect because it is very localized (near the flow front).
However, we may put a hook into RTM-Worx to allow extension of the model
through SALT.

- For curing we have an extension module. This can be combined with SALT
scripts to generate powerful solutions.

Note on SALT: SALT is a fast scripting language that supports
imperative, object oriented and functional programming. It is easy to
learn and very powerful and we have embedded this scripting engine in
RTM-Worx. The Reference Manual can be downloaded from:
   http://www.polyworx.com/upgr/rtwx.php

In the current release several scripts are included, like generation of
midplanes (e.g. 2.5D surface model from a 3D b-rep model), interface to
import ANSYS CDB files, animate the simulation etc.

]]> http://www.caeda.com.cn/blog/article.asp?id=146 caeda@vip.sina.com(admin) Thu,10 Dec 2009 08:39:26 +0800 http://www.caeda.com.cn/blog/default.asp?id=146
Faster engineering, securer data handling, convenient modeling and flexible licensing are now at hand with the new upgrade of the FRIENDSHIP-Framework. Release 2.1 includes new functionalities, operational refinement and strategies for efficient design, which help you to perform better and to create superior products. Functionalities that boost your design contribute to an overall slimmer design process. A design process that raises the bar for efficiency, user-friendliness and applicability—and that strengthens your and your designs' competitive performance.

In brief, the FRIENDSHIP-Framework 2.1 provides

Faster engineering through remote computing

Safer file handling through enhanced data management

Individualized work through the Configuration Wizard

Boosted surface design through hands-on point manipulation and validation

Flexible operation through extended licensing modes

Fully-parametric design of turbomachinery blades

A number of additional solutions for quicker, better and handier design

Clearly, each application yields major savings in time, cost and design resources. Yet their combination will boost your design process to top performance and superior products.

What does that mean, what do you gain?

Remote computing or distributed computing, as it is also referred to, is a technology that allows you to effectively include available hardware resources in your design process, thereby speeding up the overall computational time and consequently minimizing total development time. How does it work? The idea behind remote or distributed computing is to split and dispatch open computational tasks among idle hardware resources. These may be remote computers, far workstations or distant office hardware. Remote computing can employ virtually all available processing units. The distribution and result management  is handled through Secure Shell (SSH) protocol, which connects remote computers. Generally, the computation time of processes supported through remote computing is reduced by the factor of additional powerful hardware components. Especially for computationally intensive tasks, the enormous relief and the savings in time and cost this technology yields are overwhelming.

Applied to the FRIENDSHIP-Framework, this technology allows the parallel execution of integrated tools on your local computer or on remote machines. You can use available multi-core processors (CPU) to execute multiple instances of the extern tools that are integrated into the FRIENDSHIP-Framework. You may well start the external tools on remote computers through the SSH protocol. With respect to large operations like variant calculation or hydrodynamic optimization, distributed computing streamlines and consolidates your workflow massively.



Enhanced data management

Handling and storing data is a key challenge in software development. Release 2.1 answers to this quest with few smart solutions, which are designed to make you feel relaxed while operating the FRIENDSHIP-Framework.
With release 2.1, the current project file is updated and kept consistent to the project state. That helps managing system failures. Crash recovery for projects is a new functionality, which will recover your project to its latest state in case of the odd system collapse.
The incremental saving of project data additionally eases your work flow as it spares you long saving procedures, which can be quite straining especially when operating large projects. For such another development is a real asset: the FRIENDSHIP-Framework now runs on reduced memory. This is an obvious plus especially if you sample numerous designs in one project. Enhanced data management also means that projects are altogether loaded faster, which again reduces idle time significantly.  




Configuration Wizard

A new development for faster data exchange is the Configuration Wizard. This feature supports the smart and easy set up of input data for external tools, which  run within the FRIENDSHIP-Framework. The content and order of each wizard page can be configured by the provider of the corresponding definition and be customized to individual preferences accordingly. The Configuration Wizard translates the concept of the FRIENDSHIP-Framework as the designer's configurable workbench meaningfully into your  benefit.


Extended license modes


Matching the requirements of project engineering, release 2.1 comes with extended licensing modes. Now the FRIENDSHIP-Framework can be licensed with different temporal and functional scope. Float licenses are a flexible solution and cost saving alternative for mobile and group work. Licensed for one or more workstations, the FRIENDSHIP-Framework can be run on several work stations. In comparison to regular licenses, which are hardware-bound and which allow the installation and use of the software on the very registered computer, the float license enables the software use on several computers, even mobile devices, depending on the license scope. The FRIENDSHIP-Framework may now also be rented for a limited time frame. The rent license is particularly attractive for temporary project based work.




Point manipulation
Point management and the general visualization of geometry are easier and more convenient in release 2.1. For instance, the principal plane can be made visible even if the user does not operate in main view. Here, he can create and handle points along the displayed plane. Point creation is handier, work steps are consolidated and altogether faster. This is true also for the establishment of curves such as B-Spline and NURBS, which can now be generated through simple mouse operation. Using the menu bar is nonetheless available. Set your preferences and personalize the FRIENDSHIP-Framework, make it work the way you want it to work: facilitating an excellent engineering environment of utmost comfort is our prime principle and guides our development. Hence, a number of extra improvements have been added, which will surely make your operation of the FRIENDSHIP-Framework more convenient, more relaxed, more efficient.


Special functionality for turbomachinery
The Stream Section, a new curve type for the design of efficient blade surfaces like compressor blades, allows fully-parametric section design in length-preserving and angle-preserving systems. It facilitates the fully-parametric modeling of stream surfaces in comprehensive and convenient steps.

General
The User Guide is now available as a comprehensive handbook in printer-friendly format (PDF). It provides a consolidated overview of the scale and scope of functionality of the FRIENDSHIP-Framework and helps you get a read and too a quick start by showcasing numerous exemplary applications  for hull design and turbomachinery components. Access the latest User Guide here.
]]> http://www.caeda.com.cn/blog/article.asp?id=145 caeda@vip.sina.com(admin) Fri,20 Nov 2009 08:32:06 +0800 http://www.caeda.com.cn/blog/default.asp?id=145 In certain applications, different regions of the computational domain experience flow conditions that are so different that it is very difficult for any solver to produce sufficiently accurate results at the limits. In many situations, such problems can be separated and solved using loosely coupled solvers. Each solver is chosen to provide highly accurate solutions for the prevailing flow regimes. This kind of flexibility and fidelity is necessary when users are moving towards simulation-driven design solutions.

ESI's CFD-FASTRAN, a compressible flow solver, is ideally suited for high-speed external aerodynamics problems while the multi-physics solver CFD-ACE+ is best for heat transfer problems involving conduction, convection (natural and forced) and radiation. Several users may already be aware of FASTRAN/ACE+ coupling for fluid-structure interaction (FSI) simulations. From v2009.2, users can take advantage of a new FASTRAN/ACE+ coupling for heat transfer simulations. A typical application example (see figure 1 below) on thermal environment simulation is presented to help you get started on applications where you can use this feature.



Figure 1.  Re-entry capsule at angle of attack. Mach number contours are displayed for the external flow;
internal flow is represented by velocity vectors; capsule thickness and internal component show temperature distribution.

A large amount of aerodynamic heating is generated over hypersonic vehicles during re-entry. Thermal Protection System (TPS) materials are employed to prevent the heat from conducting into the internal cabin, which holds electronic devices, passengers, and other vital components. As a time-dependent process, the material making up TPS is at a low temperature and “soaks up” the heat – the conductivity of the material transports the heat (from the vehicle surface) through the thickness and into the internal volume. The material will also radiate some of the heat back to the flow – the amount depending on the emissivity of the material. A primary concern is to estimate the effects of aeroheating on the internal volume of the capsule, and its effect on electronic devices, passengers and cooling systems. In this application, typically the external flow is hypersonic in nature, whereas the flow within the capsule is a very low-speed flow dominated by natural convection. In addition to hypersonic aerodynamic heating, several other physics including heat conduction, natural convection and radiation have to be accurately modeled. CFD-FASTRAN solves for the external hypersonic flow and CFD-ACE+ solves for internal flow, heat conduction, convection and radiation. Exchange of heat flux/temperature data between the CFD-FASTRAN and CFD-ACE+ solvers occurs at defined interfaces. A simple demo case can be downloaded here.
]]> http://www.caeda.com.cn/blog/article.asp?id=144 caeda@vip.sina.com(admin) Wed,04 Nov 2009 08:59:57 +0800 http://www.caeda.com.cn/blog/default.asp?id=144
Version 2009.2 includes new enhancements of significant benefit for automotive, aerospace, fuel cell, and electronics customers.

Shorter prototyping process
The ACE+ suite allows the acceleration of the product development process by accurately simulating multiple physics in a single solution. Built around the CFD-ACE+ multiphysics solver and including specialized technology tailored to solve difficult applications, the tools within the suite can analyze complex physical phenomenon at the macro, micro, and even nano scales. The suite, also comprised of advanced automated meshing using CFD-VisCART’s adaptive mesh generation, substantially reduces the time needed to prepare complex geometries compared to traditional hex or tetrahedral meshing methods. This unique combination of specialized solver technology and advanced productivity tools makes the ACE+ Suite a powerful asset for customers.
Version 2009.2 includes significant updates to the CFD-ACE+ Advanced CFD and Multiphysics solver, CFD-FASTRAN density-based solver for high-mach flows, CFD-GEOM modeler and mesh generator, CFD-VisCART adaptive automated mesh generator, and CFD-VIEW visualization tools. Those enhancements have been made to increase usability, robustness and accuracy.

Benefits to select industries
Automotive customers will receive an increased benefit from the implementation of a new advanced turbulence model and the inclusion of automated boundary layer meshing in CFD-VisCART, which has dramatically improved both the speed and accuracy for large aerodynamic simulations.

New aerospace applications were addressed in this release with the ability to allow thermal coupling between CFD-FASTRAN and CFD-ACE+ to simulate high-mach re-entry heating of spacecraft. Fuel cell design engineers continue to benefit from a multi-year investment by the US Department of Energy to refine the multi-phase and porous media capability of CFD-ACE+, allowing ESI customers to simulate more accurately the electrical, structural, and water management performance of these complex electrochemical energy conversion devices.

Finally, the semiconductor manufacturing process simulation toolset received improvements for analyzing capacitive coupled plasma (CCP), which is one of the most common types of industrial plasma sources.

“The latest release of the ACE+ Suite includes many new features and developments designed to further improve our value proposition of high productivity coupled with highly accurate physics for the applications we have targeted”, said Joseph Strelow, Director of Virtual Environment Solutions at ESI. “This release benefits significantly from the expansion of our global CFD services activity which is becoming the driving force behind this application-focused approach to development”.
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=143 caeda@vip.sina.com(admin) Sat,24 Oct 2009 08:42:57 +0800 http://www.caeda.com.cn/blog/default.asp?id=143 http://www.iosotech.com/product.htm#Third-party



George S. Dulikravich, Florida International University
AFOSR - Air Force Office of Scientific Research, US:

Page 2
Currently, a Russian commercially available software named IOSO is the most efficient and the most robust multi-objective optimization software… IOSO, which involves concepts of neural networks, radial basis functions, and self-adapting response surface methodologies, requires the minimum number of the objective function evaluations and that is the most versatile and robust multi-objective optimizer.  Details…



Timothy W. Simpson, The Pennsylvania State University, US
Vasilli Toropov, University of Leeds, UK

Page 13
IOSO offers unique state of the art optimization algorithms that are based on self-organizational strategy and efficiently combine traditional response surface methodology with gradient-based optimization and evolutionary algorithms in a single run. The offered algorithms are equally efficient for the problems of complex and simple topology that may include mixed types of variables. Details...


Carlos A. Coello and Ricardo Landa Becerra
Evolutionary Computation Group Departamento de Computaci髇, Mexico

6.5 Design of alloys
IOSO consists of two stages. In the first stage, an approximate model of the objective functions is created. In the second stage, this approximate model is optimized. IOSO incorporates evolutionary algorithms, and artificial neural networks with radial basis functions that are used to build the response surfaces. The idea is to use this metamodel (or approximate model) to perform a very reduced number of evaluations of the actual objective functions of the problem. Details...

-最近在做一个燃气轮机透平叶片冷却优化设计工作，在这个研究过程中发现IOSO软件相对ISIGHT软件的一个绝对的优点，由于本优化是个多目标优化，由于ISIGHT不具有真正意义上的求解多目标优化问题的能力，它都是把多目标问题通过目标函数的加权求和转化为单目标问题进行处理，而在这个问题的处理上IOSO软件是直接求解多目标优化问题的Pateto解，这种优化方法才是真正意义上求解多目标优化问题的有效方法，它可以通过优化给出多个最优解，工程应用中可以根据实际需要，选择适合当前设计的一个最优解。]]> http://www.caeda.com.cn/blog/article.asp?id=142 caeda@vip.sina.com(admin) Tue,20 Oct 2009 08:11:58 +0800 http://www.caeda.com.cn/blog/default.asp?id=142 When using cfd-ace-mpi or CFD-SOLVER for parallel computations on a Linux/Unix cluster, you will be prompted for a password for each node you are using in the parallel computation. This will also be the case when running a parallel simulation from the GUI since it uses cfd-ace-mpi or CFD-SOLVER. To prevent the prompt for a password, you can use ssh authentication to log into each machine with no password.

The basic idea is to use the ssh connectivity to avoid the request for the password. While there are a couple of different ways to do this, this tip will cover how to generate authentication keys and add the remote hosts to your known hosts file on the master machine. Once you have generated these keys, you will have password-less access to each of the remote machines that have the proper authentication keys.

Here are the steps to create password-less access if you have a shared home directory:

The authentication keys need to be created and they will be in the $HOME/.ssh directory.
ssh-keygen –t dsa


The public key should be copied into the authorized_keys file.
cd $HOME/.ssh

cat id_dsa.pub >> authorized_keys
chmod 600 authorized_keys


After updating the authorized_keys file, login to each remote machine in the cluster to test the connection and update the known_hosts file. For each machine, you will be prompted to add each machine to the known_hosts file on the host machine.
ssh node-01 pwd
ssh node-02 pwd
...

Once these steps have been performed, you can run a parallel job without being prompted for a password for each process. The password request is handled through the authentication keys and the known_hosts file.

Note: If you do not have a shared home directory, you need to "scp" the id_dsa.pub file to the remote machines and append it to the authorized_keys on the remote machine.

Further information can be found here:

man ssh-keygen
http://en.wikipedia.org/wiki/Secure_Shell ]]> http://www.caeda.com.cn/blog/article.asp?id=141 caeda@vip.sina.com(admin) Mon,12 Oct 2009 08:33:51 +0800 http://www.caeda.com.cn/blog/default.asp?id=141 开发，能提高各种复杂系统和过程的优化设计效率。它可以
把工程师从十分复杂和繁重的设计参数寻优工作中解放出来，
从而使工程师能集中精力于核心问题。

软件特性：
针对大规模计算任务的高效而独特的算法
易于操控的全自动优化过程
较低的资源消耗
并行优化过程
良好的设计优化健壮性

软件优势：IOSO NM能从容地处理大规模优化任务，最大优化目标可达20，最大的优化变量为100. IOSO NM采用了开放式消息传递架构，可集成不同的商用CAE和CFD软件，如ABAQUS, MSC Nastran，ANSYS, CFX, Fluent，StarCD, Numeca等。
有关2.0版，请参考附件
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=140 caeda@vip.sina.com(admin) Wed,23 Sep 2009 08:52:55 +0800 http://www.caeda.com.cn/blog/default.asp?id=140
Blue Ridge Numerics is committed to the development of student engineers and would like to empower
your competitive engineering team with the leading upfront CFD software. With CFdesign you can
aggressively drive innovation by conducting fluid flow and heat transfer design studies as part of your
CAD-driven design process. Unlike traditional CFD, CFdesign is not a career path… it’s a tool designed
specifically for multi-tasking engineers working against tight deadlines.

We are currently accepting applications from student engineering teams participating in:

Baja SAE Series EcoCAR Challenge
Formula Hybrid Formula Student
Formula SAE Series Solar Decathlon
North American Solar Challenge  

Take  the next step…
Fill out and submit the CFdesign sponsorship request form below. We will contact you to obtain any  
additional information needed. If your team is approved, we will provide access and instructions on how
to download CFdesign from our customer portal.

点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=139 caeda@vip.sina.com(admin) Wed,16 Sep 2009 09:07:03 +0800 http://www.caeda.com.cn/blog/default.asp?id=139


Sigma technology collaborates with dassault systemes solidworks corp. Sigma Technology, international provider of the indirect optimization method based on self-organization (IOSO) technology, today announced it has joined the Solution Partner program of DS SolidWorks Corp. Sigma Technology is starting the procedure of integration between its well proven optimization code IOSO and SolidWorks® 3D CAD software, that is developed and marketed by DS SolidWorks Corp.



Details: http://www.eng.fea.ru/FEA_news_634.html
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=138 caeda@vip.sina.com(admin) Thu,03 Sep 2009 09:21:54 +0800 http://www.caeda.com.cn/blog/default.asp?id=138 To solve the problem, a 3D CFD code and multilevel IOSO technology was used. The search for the optimum design was carried out in the multi-objective statement (the compromise between the efficiency levels for core and bypass contours was searched).
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=137 caeda@vip.sina.com(admin) Mon,17 Aug 2009 15:30:30 +0800 http://www.caeda.com.cn/blog/default.asp?id=137 Flexcom是爱尔兰MCS公司开发的专门对海底工程结构进行非线性有限元仿真的软件工具，其强大功能允许对任意的海洋工程系统进行建模和仿真。Flexcom软件得到了全球油气公司、软管和设备制造商、工程承包商和咨询商的一致肯定与推荐。
软件特征
采用高度精确和得到广泛验证的有限元算法
承受轴向、弯曲和扭转高度耦合载荷的混合梁有限单元
对大位移、小应变运动采用一致的3D动力学算法
兼具时域疲劳分析和频域疲劳分析功能
丰富的实用建模工具：铰链、柔性连接、导轨面、减震器、绞盘、质点、浮标、弯曲加强筋和锥形刚性连接等
诸多提高工作效率的选项，高效自动化的时间步算法
可应用于非线性材料模型，包括有滞后效应的非线性弯曲
提供管套管（Pipe-in-pipe）建模功能
考虑土壤结构的相互影响

稳健的海底接触算法：对平坦、倾斜或任意的刚性及弹性海底结构，均采用各向异性的库伦摩擦
提供不同的海水状况建模选项：包括正常的Airy, Stokes V, Pierson-Moskowitz, Jonswap, Ochi-Hubble, 用户自定义
人性化的用户界面，即时的在线帮助，强大的数据前后处理功能
软件优势
通用的有限元算法和丰富的建模功能保证了对任何海洋结构都可以进行分析
强大直观的图形化建模器允许利用模块库快速建立3D模型
提供更精确地内管和外观独立模拟功能（管套管模型）
带宽优化和先进的编译器保证模拟高效进行
强大的多元化数据后处理功能
一流的客户支持服务——90%的请求在1个工作日内解决
提供专业的软件培训教程，用户还可根据特殊需要定制不同软件版本 点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=136 caeda@vip.sina.com(admin) Mon,17 Aug 2009 11:08:01 +0800 http://www.caeda.com.cn/blog/default.asp?id=136
LAP程序可以用来对任何承受面内载荷和力矩的复合材料层合板进行深入分析。由于平板层合板材料除厚度外外没有尺寸的限制，所以对任意复合材料部件，在已知载荷或变形的情况下都可以进行进行分析。
通常，该软件应用于复合材料部件的初步设计，首先定制层合板各层的堆叠次序，然后用有限元等其他方法对部件进行分析，最后通过逐层检查层合板的行为特征，对设计进行优化。

Basic模块
Basic模块（基本模块）提供了线性分析功能：
•    支持固化、吸湿、热学及力学等载荷形式，且载荷的效果可以单独进行观察。
•    力学载荷可以通过随意施加顺序的载荷和/或形变分量进行定义。
•    名义纤维体积分数的制造误差效应可被独特地进行考虑。
•    可得出结果包含有效刚度、有效湿热系数、求解矩阵和向量，各层应力/应变，偏移变形，极坐标绘图，以及用多种失效理论得出的失效指数等。
LAP图形化的用户界面包含一个高度专业、水平先进和高技术含量的软件产品的所有特征：
•    利用undo/redo方便数据输入。
•    数据导入和导出的接口。
•    对文本和图形可进行专业水准的打印输出。
•    交互式的帮助和打印文档，包含所有理论背景知识。
•    可以定制工作任务的单位

Non-Linear模块
在Non-Linear非线性模块中，层合板的力学刚度性能可以作为局部应力或应变的分段函数进行定义，还可以对层失效的刚度折减系数进行设置。
深入理解复合材料部件在高应力状态下的响应情况，提高了设计工作的信赖度。

Design模块
可供用户选择的Design设计模块，为LAP引入了一种功能：设计出一种最优化的对称层合板，来满足一组包含各自刚度和强度常数的力学载荷。用户在设计过程中可控制包含材料和纤维角度等选项，LAP对备选的铺放顺序进行全面检查，最后给出满足条件的有限集合，供进一步的筛选和优化。该方法的独特价值在于这种概念和应用。

Additional Failure Criteria模块
Additional Failure Criteria附加失效判据模块，可以应用BFS (Budiansky-Fleck-Soutis) 失效判据来计算无缺口及有缺口层合板的纵向压缩强度。
通过用户提供的函数，用户可以结合线性或非线性强度分析结果定义个性化的失效判据，如软件内建的失效判据具有相同效果。
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=135 caeda@vip.sina.com(admin) Mon,10 Aug 2009 09:08:17 +0800 http://www.caeda.com.cn/blog/default.asp?id=135
点击下载此文件
点击下载此文件
点击下载此文件
点击下载此文件
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=134 caeda@vip.sina.com(admin) Sat,18 Jul 2009 09:41:24 +0800 http://www.caeda.com.cn/blog/default.asp?id=134 ESI is proud to announce the V2009.2 software release of our ACE+ Suite of CFD and Multiphysics software. This release comes just seven months after our last major release, and includes many new features that our customers have requested. We welcome and encourage all of our customers to download and use the latest versions, which include numerous enhancements.

Included in the V2009.2 release are significant updates to:

CFD-ACE+ Advanced CFD and Multiphysics Solver
CFD-FASTRAN CFD Physics Solver
CFD-GEOM Geometry Modeler and Mesh Generator
CFD-VisCART Cartesian Mesh Generator
CFD-VIEW Solution Analysis and Visualization Tool
Enhancements have been made to increase usability, robustness and accuracy, and we are sure that these new versions will make for a worthwhile upgrade to your CFD toolset.

The remainder of this note gives you some information about the V2009.2 software release. Please read on to learn more about the updates and let us know if you have any comments or questions.

IMPORTANT NOTE: If you are running V2008.x or earlier software versions, then you may require a new license to run V2009.x versions! If you are currently running V2009.0, then you will NOT need a new license to run V2009.2. ESI licensing policy requires licensing to the version of the software. This means that if you currently have a V2008.x (or earlier) license file, you will need to replace it with a V2009.x license file. These are available upon request from your ESI license or sales representative. If you are not sure who that is, you should contact support.cfd@esi-group.com. Please include your existing V2008.x license file for the quickest service.

V2009.2 Release Highlights

As with previous releases, there are many improvements to features, performance, and robustness. Below is a short list of some of the top reasons why you should consider upgrading to V2009.2. For a more detailed description of the new and improved features, please consult the relevant release notes document, which is included in the package download files and can also be found here (login required).

CFD-ACE V2009.2 (view full release notes - login required)
New Realizable K-e turbulence model
New Dielectric Charge leakage for CCP Plasma
Removal of limitation: Thin-Walls in contact with Arbitrary Interfaces
New User Subroutine option to specify Porous media resistance direction
Resolution of stack overflow and memory corruption issues on MS-Windows 64-bit platform
CFD-FASTRAN V2009.2 (view full release notes - login required)
New Detached Eddy Simulation (DES) Turbulence model
Thermal Coupling with ACE+
CFD-GEOM V2009.2 (view full release notes - login required)
Improved Advancing Front Surface Meshing
Improved Discrete Surface Meshing
Improved Pyramid Generation
Improved DXF reader
Improved Boundary Layer Meshing
Improved CAD/IGES import
New Shrinkwrap Surface Meshing
New DTF Simplifying Writer
New Solid Boolean operations
New "Create All Domains" Tool
New Thin Wall Surface capabilities
New Exposed Boundaries Visualization
New Customizable ToolBar
45 New Script Functions
CFD-VisCART V2009.2 (view full release notes - login required)
Parallelization of Single Domain Mesher
New Surface Mesh Coarsening option for Shrink-Wrap Mesh
New Manual Node Editing Tool
New Interior Node Location Optimization tool to improve mesh quality
Improved Domain Marker Placement approach
Better handling of 0-cell domains during Multi Domain mesh generation
New option to Not Save Skewed Cells to DTF file
New Localized Feature Preservation option in Single Domain Mesher
Improved Boundary Layer Meshing
New options to Smooth Boundary and Interior Nodes
New option to Refine Cells near Features in Single Domain Mesher
Improvements in GUI
CFD-VIEW V2009.2 (view full release notes - login required)
Improved DTF Import capabilities
Improved Particle Trace Operator
Improved Legend
Improved Scripting and addition of many new Scripting functions
Miscellaneous minor improvements
V2009.2 Installation
The V2009.2 software will be installed parallel to your existing V2009.0 (or earlier) installation. This means that you will have access to all versions should you desire. However, only V2009.2 will be supported going forward, so you should use V2009.2 to get access to future software issue resolutions and feature improvements. Step-by-step instructions for the download and installation of V2009.2 software can be found in the Getting Started Guide (located in the Knowledge Base section of the ESI CFD Portal - login required).

Regards,

]]> http://www.caeda.com.cn/blog/article.asp?id=133 caeda@vip.sina.com(admin) Thu,16 Jul 2009 08:24:47 +0800 http://www.caeda.com.cn/blog/default.asp?id=133

Results are quite interesting; they were assessed by Saint-Petersburg State Marine Technical University and Ship Design Company from Finland Foreship.

点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=132 caeda@vip.sina.com(admin) Fri,10 Jul 2009 09:04:53 +0800 http://www.caeda.com.cn/blog/default.asp?id=132 In engineering systems, the presence of obstacles in the flow path provides resistance to flow accompanied by a pressure drop across the component. Often, the flow paths within such components are too complex to be resolved by a mesh (assuming it is known) and only the macro effects are of interest to design engineers.

Porous media routinely encountered in many engineering applications is a common example and refers to a multiphase matter consisting of a solid matrix with interconnected pore space allowing fluid flow (flow through membranes, filtration products, packed/fluidized beds, etc…). Similar characteristics can be exhibited by individual components of a large system (automobile radiators, showerheads of semiconductor process equipments, tubebanks in heat exchangers, screens/turning vanes in wind tunnels/HVAC etc…) and it is cheaper and desirable to represent them through a momentum loss model.


Momentum losses can be modeled by adding appropriate source (sink) terms to the momentum equations. In CFD-ACE+, this can be done using one of the following methods:

Momentum resistance model

The pressure drop is characterized using lumped parameters (momentum resistance coefficients) and assigning them to the defined momentum resistance region. These coefficients are obtained from experiments or detailed numerical simulations of a representative section of the geometry. This model can be applied over volumes and defined geometries (cylindrical/polyhedral). This model would be appropriate for cases where a separate volume cannot be generated for the resistive region and you have access to the resistance coefficients through flow data such as pressure drop vs. velocity curves.

Full porous media model

The overall effect of the pore space is represented through local volume averaging and the resulting momentum equation is solved. This allows a more physics based treatment of the problem and predicts the physical velocities and correct residence times within the porous region. It can be coupled with other physics including multiphase flows. This model can be applied only on separately meshed volumes and is useful for cases where you only have access to data related to the porous matrix such as porosity and permeability. It is appropriate to use this model when the problem involves other physics within the porous media. The user subroutine template ‘uporous’ can be used to control the porous media parameters.

Momentum sources

Simple or complex momentum sources can be directly specified and added to the standard source terms in the momentum equations that are being solved. Sources can be applied only on separately meshed volumes. It is recommended for cases where you do not have access to much data about the porous matrix but able to estimate the source term based on empirical relations.

The following table provides a comparison of the three methods and can be used to identify and apply an appropriate method for the problem at hand.



Table 1.  Momentum loss models and corresponding parameters


If you have any questions about this tip or would like us to discuss other topics in the future, please let us know.
]]> http://www.caeda.com.cn/blog/article.asp?id=131 caeda@vip.sina.com(admin) Thu,09 Jul 2009 08:10:09 +0800 http://www.caeda.com.cn/blog/default.asp?id=131
Some highlights:
- workflow based optimization setup for topology, shape and bead optimization
- graphical interaction with the FE-model
- visualization of manufacturing constraints
- improved topology manufacturing constraints
- support of the newest FE-solver releases
o Abaqus 6.9.1
o Ansys v12.0
o NXNastran 6.1

For more detailed information on the new features please refer to the TOSCA Structure 7.0 Release Notes. 点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=130 caeda@vip.sina.com(admin) Fri,03 Jul 2009 08:11:29 +0800 http://www.caeda.com.cn/blog/default.asp?id=130
The 2009.2 Beta version of CFD software products was released on May 5, 2009. A number of new features were made available in this version to cater to the demands from the various industry sectors.

DTF version 8 and rollback to DTF version 7

In response to the multiple requests for reduction in DTF file size, V2009.2 Beta codes were linked and made available with DTF version 8 (an upgrade from DTF version 7). DTF version 8 uses a data compression (zlib) algorithm that results in reduced DTF file size. Benchmark runs indicated the average increase in write/read time to be around 10%. In certain extreme cases though, a much higher increase in time was observed. We were concerned that this may not be ideal for some of our customers whose main objective is to reduce their overall CFD solution process time. With this in mind, and after detailed consideration, we have made a decision to rollback to DTF v7. The V2009.2 Beta codes linked to DTF v7 will be available for download from our portal and ftp site today, July 2, 2009 onwards.

How this affects our customers

This rollback affects the usability of DTF files generated with V2009.2 Beta codes till date. All such (DTF v8) files will need to be converted to DTF v7 - we will provide a free service to convert all existing v8 DTF files to v7.
This rollback does not affect access to new features - all new features introduced in V2009.2 Beta will continue to be available to customers.
Customers who have not used V2009.2 Beta codes are in no way affected by this rollback.
How to request the free conversion service (DTF v8 to v7)
All conversion service requests should be sent to support.cfd@esi-group.com - please include your organization name and COIN.
A secure ftp account will be created and detailed information regarding the ftp account and upload procedure will be sent to the customer.
Once the files have been received, we will convert and upload the corresponding DTF v7 files to the same ftp account within 2 business days.
As always, all customer files will be handled with utmost confidentiality and will be deleted from our servers as soon as the converted files have been picked up by the customer.
There is no limit to the number of DTF files that a customer can upload for conversion.
How to check if a DTF file is v7 or v8?
The easiest way to check the version number of a DTF file is to run the following command:

DTF -i filename.DTF

For example:

>DTF –i flow.DTF

Title       : Flow Simulation
origin      : <none>
Created by  : CFD-VisCART, ver. 2009.2.6
Created on  : Tue Jun 30 16:37:02 2009
Modified on : Tue Jun 30 16:39:23 2009
DTF version : 8.0.4
Scaling     : 0.001000


1 simulation(s):
- simulation 1: Projected Cartesian Grid Generated With CFD-VisCART 2009.2.6
1 zone(s):
- zone 1: Polyhedral Zone Type
   is_sorted= 1
   138771 vertices
   128443 cells (by type     Hexahedron: 127971     Poly-Cell: 472)
   18 data array(s)


Please note that if you run the above command on a v8 DTF file using v7 DTF utility, you will get an error message as shown below. This would also confirm that the DTF file is v8.

Error: DTF file was written with a newer version than contained in this application


We sincerely apologize for any inconvenience this rollback may cause but we had to make this decision keeping the best interests of all of our customers in mind.
]]> http://www.caeda.com.cn/blog/article.asp?id=129 caeda@vip.sina.com(admin) Fri,19 Jun 2009 08:39:22 +0800 http://www.caeda.com.cn/blog/default.asp?id=129




Important information about the licensing procedure with TOSCA Structure 7.0.:

All customers will need new license files, this means:

Customers under maintenance and annual lease will get new license files once they have downloaded the version and asked for registration

Customers with TOSCA 6.2. or older versions don磘 need new license files but they will not get access to TOSCA 7.0 (unless they are paining for maintenance or renewal annual lease)



For further details concerning the new functionalities and TOSCA ANSA TM environment please see the attached press release and presentation.
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=128 caeda@vip.sina.com(admin) Tue,16 Jun 2009 10:01:17 +0800 http://www.caeda.com.cn/blog/default.asp?id=128
Seminars relating to TOSCA Structure are conducted by us at regular intervals.
There are 3 day Basic Seminars - next dates will be:
June 29-July 1, 2009 (in English)
September 15-17, 2009 (in German)
November  24-26, 2009 (in German).

For experienced users we hold 1 day Advanced Seminar - next dates will be:
July 02, 2009 (in English)
November 27, 2009 (in German)

The seminar handouts for TOSCA Structure Basic and Advanced Seminars are English.
Of course, additional English spoken TOSCA Structure Basic or Advanced Seminars can be arranged on request. Please contact us directly.

TOSCA Structure Basic Seminar
This seminar is a comprehensive introduction to the structural optimization capabilities of TOSCA Structure. Attendees learn how to formulate and solve basic topology, shape and bead optimization problems, submit optimization jobs, and view and evaluate the results using the TOSCA Structure viewer or your own postprocessing tool.
Special emphasis is put on the integration and effective use of optimization in the design process to reduce costs and time in the global product development process

The seminar covers the following topics:
-  Introduction to Structural Optimization
- Optimization with TOSCA Structure:
  General functionalities and operations, possibilities and features
- Getting the best structure:
  Topology Optimization using TOSCA Structure.topology
  Manufacturing constraints, objective functions and constraints
  Optimizations based on linear static and modal analysis, combining both
- Improving Durability, decreasing stresses, increasing frequencies:
  Shape Optimization using TOSCA Structure.shape
  Mesh smoothing, manufacturing constraints, coupling conditions (symmetry
  etc)
- Maximizing stiffness of sheet metal structures:
  Bead Optimization using TOSCA Structure.bead
- Formulating and executing optimization jobs using TOSCA Structure.gui
- Integration of TOSCA Structure in your CAE-Environment:
  Interaction with FE-Solvers and postprocessors
- Integration of structural optimization in the design process:
  Transfering results using TOSCA Structure.smooth for CAD and
  preprocessing
- Effective and quick evalution:
  Visualization of the results using TOSCA Structure.view and TOSCA
  Structure.report
The seminar content is put into practice in several workshops. They are designed to reinforce concepts presented during the lectures and demonstrations and will provide users with the experience of running and trouble-shooting actual TOSCA Structure jobs.
This basic seminar is conceived for all new TOSCA Structure users, regardless of which FE solver and pre-/postprocessing environment they use.
In order for the training to be effective, all participants should have some basic familiarity with finite element technologies. Experience in structural optimization techniques is not required.

The seminar is held by developers and experienced TOSCA Structure users of the FE-DESIGN engineering team, which gives you the opportunity to discuss your specific application aspects in detail with them.

Subscription facsimile form or online registration to the basic seminar.

Beginning of the first day is 9.30 a.m., end of the third day is 5.00 p.m.
Seminar fee: 1.200 EUR net per person.


TOSCA Advanced Seminar
For experienced TOSCA users the solutions of complex optimization tasks are introduced by 1 day advanced seminars.

Advanced Topology optimzation part I:

Topoplogy optimization based on non linear finite element models
- choice of elements in contact analysis
- influence of large deformations on optimsation results
- optimisation with non linear reversible materials
Next date: April 02, 2009
Duration: 09:00 - 17:00
Seminar fee: 400 EUR/day net per person


Advanced Topology optimzation part II:

Complex setups for topology optimzation

- choice of objective functions and constraints for center of gravity, reaction
  and internal forces
- introduction in multi objective optimzation tasks (Min vs. MinMax
  formulation)

Next date: July 27, 2009
Duration: 09:00 - 17:00
Seminar fee: 400 EUR/day net per person


Advanced Shape optimization:

- shape optimzation in contact areas
- shape optimzation including durabilty responses
- shape optimization and complex objective functions (with sensitivity based
  algorithms)
- bead optimzation with acoustic responses
Next date: November 27, 2009
Duration: 09:00 - 17:00
Seminar fee: 400 EUR/day net per person

Experience with the use of TOSCA and with FE-methods is hereby required.
Subscription facsimile form or online registration to the advanced seminar.

Time limits for registration and cancellation
In order to look after all of the attendees optimally, the number of participants of each seminar is limited to 8 persons. We therefore request that you enrol at least 14 days in advance of the seminar. In case of cancellation up to 7 days before the seminar we will charge a fee of 50 EUR net. If you cancel later the whole seminar fee is charged.

The dates of upcoming seminars
can be found at the events site.

On-site seminars and optimization coaching according to your individual requirements
有兴趣咨询：caeda@vip.sina.com

]]> http://www.caeda.com.cn/blog/article.asp?id=127 caeda@vip.sina.com(admin) Fri,05 Jun 2009 08:37:45 +0800 http://www.caeda.com.cn/blog/default.asp?id=127
Currently, a Russian commercially available software named IOSO is the most efficient and the most robust multi-objective optimization software.  Similarly, the second best multi-objective optimization software package, modeFrontier, was also developed in foreign country (Italy) and is licensed by their software company.  The only multi-objective optimization software package that was developed in the United States and is licensed by the U.S. software company is Engineous.  However, performance speed, robustness and overall sophistication of Engineous optimizer are significantly inferior in comparison to modeFrontier and especially to IOSO performance speed, versatility and robustness.  Performance of IOSO, modeFrontier and Engineous multi-objective optimizers published in the open literature clearly indicates that IOSO, which involves concepts of neural networks, radial basis functions, and self-adapting response surface methodologies, requires the minimum number of the objective function evaluations and that is the most versatile and robust multi-objective optimizer.
索取详细报告：caeda@vip.sina.com
]]> http://www.caeda.com.cn/blog/article.asp?id=126 caeda@vip.sina.com(admin) Thu,28 May 2009 09:15:28 +0800 http://www.caeda.com.cn/blog/default.asp?id=126
Blue Ridge Numerics, Inc. today announced that Cornell Pump Company experienced a 100x speed up of a full transient system-level simulation and performance curve generation using the new CFdesign Motion and high performance computing (HPC) Modules, released earlier this month.

Cornell design engineers were able to perform a series of transient simulations to develop a complete performance curve, enabling evaluation of a centrifugal pump design in only two hours, down from 200 hours, resulting in a significant reduction in product development time.

Similar to the Cornell team, multitasking engineers designing high performance fans, blowers, turbomachinery, and many other products can now benefit from the upgraded Motion Module, allowing for more iterations upfront in the design process.

“In the past design teams have typically had to choose between a slower simulation time for a full system-level transient model simulation or a less accurate fixed-rotor approach,” said Derrek Cooper, product manager, Blue Ridge Numerics. “With the new capabilities in the Motion Module, CFdesign now offers a more complete and reliable solution that is faster than any other options out there, so design teams no longer have to sacrifice accuracy for speed.”

View of internal flow patterns of Cornell’s centrifugal pump, modeled with CFdesign, highlighting the pressure of the liquid and the pattern of the flow.

  The significant performance increases the Cornell team experienced using CFdesign will have major implications for development of new pumps. Beyond cutting computer simulation time from weeks to hours, it is expected to reduce the number of pump castings that have to be made for physical testing. Eliminating just one of the physical tests could save the Cornell team as much as three months of development time.

“With much faster running speeds, we will be able to do many more simulation runs in less time,” said Andrew Enterline, design engineer, Cornell Pump Company. “We’ll spend more time designing the product and less time and money in pattern rework, re-pouring castings, and physical development testing, helping us create more optimized product designs from the start.”

In the past, generating a typical pump curve for the Cornell design team required 300 iterations at 1.5 iterations per hour to evaluate a single design, so the team explored options to speed up the development cycle. New rotational algorithms for transient rotating modeling included in the latest release of the Motion Module help speed up simulation time by 20x on a standard desktop computer, allowing Cornell and other design teams to more quickly and accurately evaluate performance curves. In addition, the Motion Module can be combined with the new CFdesign HPC Module to provide an overall 100x speed up. Cornell leveraged the power of the Motion and HPC Modules to speed up their design process by partnering with R Systems, a provider of high-end computing resources for research, to conduct their simulations on an 8 node HPC cluster setup, allowing them to complete 300 iterations now in only two hours.

About the CFdesign Motion Module

Several mechanical engineering applications require understanding of the behavior of liquid and gas flows as they interact with solid objects. The CFdesign Motion Module allows users to create a virtual prototyping environment to better understand these behaviors by simulating the way components interact and respond to prescribed flows. All the physical effects of the motion as well as the time-history are output for data review and to create animations for visual studies.
In addition to the simulation of rotating turbomachinery described, CFdesign also helps engineers model:
•    Flow-Driven Motion
•    orbital Motion
•    Nutating Disk Motion
•    Combined Linear and Angular Motion
•    Sliding Vane Pump Motion
•    Automatic and Adaptive Time Step Determination
•    External Driving and Spring Resistance Models

About the CFdesign HPC Module

The CFdesign HPC Module allows CFdesign users to harness the power and investment of Windows HPC Server 2008, enabling more design studies in less time and reducing the time it takes to achieve solutions for large complex models. The Module can be used with a small cluster, a data center, or service providers, such as R Systems to gain access to HPC resources.
CFdesign, Motion Module

]]> http://www.caeda.com.cn/blog/article.asp?id=125 caeda@vip.sina.com(admin) Tue,26 May 2009 10:01:33 +0800 http://www.caeda.com.cn/blog/default.asp?id=125
Nemo Engineering AS have confirmed the value of Laminate Tools in a demanding Ansys design environment and have increased the number of Laminate Tools licences held at their facility at Lysaker, Norway. According to Kjetil Antonsen, "By using Laminate Tools as a pre- and post-processor to Ansys we have reduced the analysis time significantly for our products. Problem areas are quickly identified while the number of FE solving processes can be reduced to a minimum. Laminate Tools enables us to deliver Protection Covers for subsea applications with very short lead time."



The screen shot shows an example non-standard shape cover in Laminate Tools. Most often these covers are tailor-made to fit the items they are designed to protect. The loading for the covers are various lifting and upending configurations, impact (up to 350 kJ), trawl loads from fishing equipment and various project specific loads.


Nemo Engineering have engineered and delivered approximately 2500 metric tonnes of these protection covers for the last 10-12 years.
]]> http://www.caeda.com.cn/blog/article.asp?id=124 caeda@vip.sina.com(admin) Fri,22 May 2009 13:58:13 +0800 http://www.caeda.com.cn/blog/default.asp?id=124
   而RTMworx可以读取MSC Patran Laminate Moudeler Layup 格式数据。并且进行注射模拟。

   换句话说， Laminate tools 与RTMworx之间是通过MSC Patran Laminate Moudeler Layup 格式进行接口交换的。但只能是Laminate tools ——〉RTMworx，反之不行。
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=123 caeda@vip.sina.com(admin) Tue,19 May 2009 14:22:59 +0800 http://www.caeda.com.cn/blog/default.asp?id=123 该算法将基于梯度的非线性规划方法与进化式响应面方法巧妙结合，可求解非光滑、随机、多目标、混合变量等各类优化问题
不知附件讲义作者是谁？ 点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=122 caeda@vip.sina.com(admin) Wed,06 May 2009 08:29:44 +0800 http://www.caeda.com.cn/blog/default.asp?id=122
ESI's CFD Division is proud to announce that the Spring 2009 release is on schedule and ready for BETA testing. The final version or the Official Release is scheduled for the first week of June 2009. The fact that this release comes just six months after our last major release proves that ESI is working hard to implement the features that the market demands in a timely manner. We welcome and encourage all of our customers to download and use the latest versions, which include numerous enhancements.

Included in the Spring 2009 release are significant updates to:

CFD-ACE+ Advanced CFD and Multiphysics Solver
CFD-FASTRAN CFD Physics Solver
CFD-GEOM Geometry Modeler and Mesh Generator
CFD-VisCART Cartesian Mesh Generator
CFD-VIEW Solution Analysis and Visualization Tool

Enhancements have been made to increase usability, robustness and accuracy, and we are sure that these new versions will make for a worthwhile upgrade to your CFD toolset.

The remainder of this note gives you some information about the Spring 2009 software release. Please read on to learn more about the updates and let us know if you have any comments or questions.

IMPORTANT NOTE: If you are running V2008.x or earlier software versions, then you may require a new license to run V2009.x versions! If you are currently running V2009.0, then you will NOT need a new license to run V2009.2. ESI licensing policy requires licensing to the version of the software. This means that if you currently have a V2008.x (or earlier) license file, you will need to replace it with a V2009.x license file. These are available upon request from your ESI license or sales representative. If you are not sure who that is, you should contact support.cfd@esi-group.com. Please include your existing V2008.x license file for the quickest service.

Spring 2009 (V2009.2) Release Highlights

As with previous releases, there are many improvements to features, performance, and robustness. Below is a short bullet list of some of the top reasons why you should consider upgrading to V2009.2. For a more detailed description of the new and improved features, please consult the relevant release notes document, which is included in the BETA package download files and can also be found here (login required).

A couple of the new features are in the final phase of internal testing and will become available with the Official Release in the first week of June 2009. In the list below, “available with official release only” has been added next to those features.

CFD-ACE V2009.2 (view full release notes - login required)
New Realizable K-e turbulence model (available with official release only)
New Heat Exchanger model (available with official release only)
New Dielectric Charge leakage for CCP Plasma (available with official release only)
Removal of limitation: Thin-Walls in contact with Arbitrary Interfaces
New User Subroutine option to specify Porous media resistance direction
Resolution of stack overflow and memory corruption issues on MS-Windows 64-bit platform
CFD-FASTRAN V2009.2 (view full release notes - login required)
New Detached Eddy Simulation (DES) Turbulence model
Thermal Coupling with ACE+
CFD-GEOM V2009.2 (view full release notes - login required)
Improved Advancing Front Surface Meshing
Improved Discrete Surface Meshing
Improved Pyramid Generation
Improved DXF reader
Improved Boundary Layer Meshing
Improved CAD/IGES import
New Shrinkwrap Surface Meshing
New DTF Simplifying Writer
New Solid Boolean operations
New "Create All Domains" Tool
New Thin Wall Surface capabilities
New Exposed Boundaries Visualization
New Customizable ToolBar
25 New Script Functions
CFD-VisCART V2009.2 (view full release notes - login required)
Parallelization of Single Domain Mesher
New Surface Mesh Coarsening option for Shrink-Wrap Mesh
New Manual Node Editing Tool
New Interior Node Location Optimization tool to improve mesh quality
Improved Domain Marker Placement approach
Better handling of 0-cell domains during Multi Domain mesh generation
New option to Not Save Skewed Cells to DTF file
New Localized Feature Preservation option in Single Domain Mesher
Improved Boundary Layer Meshing
New options to Smooth Boundary and Interior Nodes
New option to Refine Cells near Features in Single Domain Mesher
Improvements in GUI
CFD-VIEW V2009.2 (view full release notes - login required)
Improved DTF Import capabilities
Improved Particle Trace Operator
Improved Legend
Improved Scripting and addition of many new Scripting functions
Miscellaneous minor improvements
V2009.2 Installation 点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=121 caeda@vip.sina.com(admin) Tue,05 May 2009 08:07:44 +0800 http://www.caeda.com.cn/blog/default.asp?id=121 点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=120 caeda@vip.sina.com(admin) Sat,25 Apr 2009 09:00:10 +0800 http://www.caeda.com.cn/blog/default.asp?id=120
with various major improvements and enhancements like improved result extraction method “particle track filter”, new cutoff method “particle track spread” as well as support of the latest STAR-CD v4.x versions.附件详情
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=119 caeda@vip.sina.com(admin) Sun,19 Apr 2009 09:40:23 +0800 http://www.caeda.com.cn/blog/default.asp?id=119
工作地点：本人所在地，为本地区客户提供软件培训、咨询等技术顾问工作，分以下几类专业

Part I: 复合材料分析及工艺仿真（类）软件
工作性质：兼职
要求: 在校硕士在读一年以上，或曾经有相关工作经验者, 必须熟练使用相关商品软件(指类似于RTM-Worx, ESAComp, Laminate Tools, LAP, CoDA, CADpress等, 这些软件就是将来此职位要做技术支持的)

Part II: 优化设计仿真（类）软件
工作性质：兼职
要求: 在校硕士在读一年以上，或有相关工作经验者, 必须熟练使用相关商品软件(指类似于IOSO NM, TOSCA等, 这些软件就是将来此职位要做技术支持的)

Part III: CFD仿真软件
工作性质：兼职
要求: 在校硕士在读一年以上，或有相关之前工作经验者, 必须熟练使用相关商品软件(指类似于CFD-Fastran, CFD-ACE+, CFDesign, Shipflow, CADNexus等, 这些软件就是将来此职位要做技术支持的)

说明：以上特指本公司代理的CAE软件需要培训讲师，这些软件可以提供试用及详细手册资料等，自己准备一段时间，胜任即可。
请将简历E-MAIL至: caeda@vip.sina.com, (建议在发简历或联系前务必访问我公司网站, 熟悉我公司产品, 以便面谈)，或在网站CAE专家银行目录下登记。
有关CAEDA Ltd. 公司情况，请登陆网站: www.caeda.com.cn, www.caefinder.com
联系人：刘先生





]]> http://www.caeda.com.cn/blog/article.asp?id=117 caeda@vip.sina.com(admin) Fri,03 Apr 2009 10:01:32 +0800 http://www.caeda.com.cn/blog/default.asp?id=117 CoDA 3.2      UpdateD VERSION

Component and Composite Design Analysis

The popular Component Design Analysis (CoDA) package from the National Physical Laboratory (NPL) has been enriched with two new modules, FLANGES and PREDICT.   In addition, full 3D properties are now calculated for laminates  in the LAMINATE module.   The software has now sold hundreds of modules to small manufacturers, medium sized companies, consultants, universities and major companies.


FLANGES

There are numerous examples of premature failure of laminated composite  components  due  to  the  general  lack  of  appreciation  of  their relatively low through-thickness  strength. This is  particularly true  for  structures  such  as  flanges  and  curved  beams,  which  are found  in  most  industrial  applications.  The  Flanges  module  uses geometry  and  material  properties  to  calculate  the  stress distribution  and  ultimate  load  capacity  of  simple  curved  beams, subjected  to  moment  and  shear  loading.  Detailed  and  expensive finite  element  analyses  can  be  circumvented  by  the  use  of  the
CoDA  Flanges  module.   The  analysis  has  been  verified  for laminates by comparing finite element analysis, analytical methodsand experimental results.

Ref:- G. Hinopoulos, D.Ferriss, J.Niklewicz and G.D.Sims,
“Procedures for the local design of flanges and curved composite
structures”, NPL Report CMMT(D) 295, 2000

PREDICT
This  module is  mainly  used  to  simulate the  progressive  formation of ply cracks in a single orientation during the monotonic loading of a  general  symmetric  composite  material  laminates.  The  results include  the  variation  of  laminate  mechanical  properties  with increasing  crack  density  or  applied  stress  or  strain,  while  the effects  of  ply  thickness,  temperature  changes,  etc.  on  laminate mechanical  behaviour  can  be  studied.  General in-plane loading can  be analysed, as  would  arise  in  many  engineering  structures.  The tool can easily be used to determine the first ply-failure stress, and  the temperature  changes  needed  to  cause  failure  during cooling following manufacture at elevated temperatures, or for cryogenic applications.

Ref:- L.N.McCartney, “Physically based damage models for
laminated composites”, Proc. Instn. Mech.Engrs. Vol.217, Part l:
J.Materials Design and Applications.

Further  modules  are  being  developed  and  validated  at  both  the  material  and  structural  component  level  (e.g.
fatigue strength, sandwich structures).  NPL and Anaglyph are pleased to receive suggestions for further modules
and facilities.  


National Physical Laboratory
NPL Materials Centre
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=116 caeda@vip.sina.com(admin) Sat,28 Mar 2009 10:04:00 +0800 http://www.caeda.com.cn/blog/default.asp?id=116 ANSYS:
功能模块：Design Space
DesignXplorer™
DesignXplorer VT
各模块的功能：
ANSYS DesignSpace     完成结构的初始有限元分析功能
DesignXplorer™        读取DesignSpace分析结果，实现了结构的优化功能，DesignSpace 合用。
DesignXplorer VT    DesignXplorer™    的扩展功能，主要体现在多目标优化上。而DesignXplorer™为单一目标优化，从算法上看，由传统的DOE算法向VT变分算法扩展。
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=115 caeda@vip.sina.com(admin) Fri,20 Mar 2009 12:22:53 +0800 http://www.caeda.com.cn/blog/default.asp?id=115
Vacuum Infusion is a promising technology for the production of large, high-quality, composite parts. Vacuum Infusion (also known as Resin Infusion, VIP and also regarded one of the specific processes refered to by the more general term 'Liquid Injection Moulding') adds improved worker conditions, emissions reduced to almost nothing and considerable savings on labor costs. In spite of all these advantages, general acceptance has been slow, mainly due to the high risks involved with trial-and-error methods due to just guesswork. Without a control method, all materials and labor are wasted if the chosen injection strategy does not work.

Do your trial infusions on the PC!

A better method exists right now and we have called it 'Controlled Vacuum Infusion' (CVI). The costly trial-and-error phase has been replaced by FEM based simulations on the PC. Trial injections and overall optimization of the process can now be done in a very quick and cost effective way: caeda@vip.sina.com

点击下载此文件
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=114 caeda@vip.sina.com(admin) Thu,12 Mar 2009 09:26:14 +0800 http://www.caeda.com.cn/blog/default.asp?id=114     复合材料分析软件Laminate Tools (LT)是专业的面向复合材料分析的工具软件，其功能涉及到复合材料结构设计、分析和制造领域。该软件由5个基本部分组成，即：显示、设计、分析、检查以及制造工艺5个功能模块，分别涵盖了三维模型可视化显示、材料铺层等信息建立与编辑、层合板性能生成、商业化有限元软件结果可视化检查以及制造工艺信息及下料图形的输出。

    正如开发商所说，Laminate Tools不是CAD软件、不是网格生成器，也不是有限元求解器。换言之，该软件无CAD能力、也没有网格生成功能，当然也不具备有限元求解功能。它可集成设计、分析、制造相关的软件，形成简易的复合材料分析、设计环境，其自身具有铺覆设计及二、三维图形输出的功能。在力学分析方面，该软件仅能根据接口规则，为有限元软件准备材料特性数据，无法实现LAP的主要功能，可结合使用。
    该软件可以作为有限元软件结果察看器，它支持多种有限元软件的数据格式，同时也可以作为商业有限元软件复合材料的材料特性输入编辑功能增强，使得材料性能的编辑更为方便、简洁。

该软件的典型工作流程是：
    
-利用CAD软件定义几何模型曲面，然后用LT进行材料铺覆模拟；
    
-利用预定的有限元前处理器创建网格，然后利用LT工具用复合材料织物对曲面进行铺覆并建立整个结构的铺层；
    
-有了利用LT建立的铺层，可将单层信息转换为层合材料特性，以适应有限元分析的需要；
    
-利用LT显示后处理结果；
    
-输出单层裁切图形样板，用于切割或者激光投影设备


点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=113 caeda@vip.sina.com(admin) Fri,06 Mar 2009 08:43:25 +0800 http://www.caeda.com.cn/blog/default.asp?id=113
Some highlights:
- Support of Windows 64 bit systems: You are now able to optimize Models with more than 2 Million
  Elements on Windows 64 bit systems
- Interface update for Abaqus 6.8-2 support
- New features in TOSCA Structure.smooth to improve mesh quality of the smoothed surface

For more detailed information on the new features please refer to the TOSCA Structure 6.2.1 Release Notes. 点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=112 caeda@vip.sina.com(admin) Fri,27 Feb 2009 10:10:32 +0800 http://www.caeda.com.cn/blog/default.asp?id=112
Tosca items可以索取

-STATIC-VIBRATIONAL DESIGN OF A BONNET WITH FRAME TOPOLOGICAL OPTIMIZATION

-INTEGRATION OF TOPOLOGY AND SHAPE OPTIMIZATION IN THE DESIGN PROCESS

-VIRTUAL VEHICLE DEVELOPMENT IN THE COMCEPT STAGE-CURRENT STATUS OF CAE AND OUTLOOK ON THE FUTURE-

-TOPOLOGY AND SHAPE OPTIMIZATION OF COMPONENTS AND SYSTEMS WITH CONTACT BOUNDARY CONDITIONS

-OPTIMIZATION OF SIMPLE LINEAR AND ADVANCED NON-LINEAR PROBLEMS USING TOSCA AND ABAQUS

-TOPOLOGY OPTIMIZATION WITH CASTING AND MACHINING RESTRICTIONS BY THE EXAMPLE OF BRAKE PARTS

-STRESS REDUCTION IN AN INTERSTAGE AIR SEAL

-MULTIDISCIPLINARY OPTIMIZATION CONSIDERING CRASH AND NVH LOADCASES

-THE EFFECT OF SPOT WELD FAILURE ON DYNAMIC VEHICLE PERFORMANCE

-AUTOMATED TOPOLOGY OPTIMIZATION OF FLEXIBLE COMPONENTS IN HYBRID FINITE ELEMENT MULTIBODY SYSTEMS, USING ADAMS/FLEX AND MSC.CONSTUCT

-NON-PARAMETERIC LARGE SCALE STRUCTURAL OPTIMIZATION FOR INDUSTRIAL APPLICATIONS

-FROM SIMULATION TO OPTIMIZATION OF WIND TURBINE COMPONENTS

-SHAPE OPTIMIZATION BASED ON PARAMETERS FROM LIFETIME PREDICTION

-TOPOLOGY OPTIMIZATION OF A LINEAR GUIDANCE SYSTEM USING TOSCA AND ABAQUS

-TOPOLOGY- AND SHAPE OPTIMIZATION IN THE VIRTUAL PRODUCT PROCESS ILLUSTRATED ON STRUCTURES IN AUTOMOTIVE AND SUPPLIER INDUSTRY

-SHAPE- AND TOPOLOGY OPTIMIZATION REGARDING FATIGUE ANALYSIS

-OPTIMIZATION OF CASTINGS AND FORGINGS AT AUDI AG

-SHAPE AND FORM OPTIMIZATION

-FASTER TO SUPPERIOR CASTING COMPONENTS BY EMPLOYMENT OF THE TOPOLOGY OPTIMIZATION WITH THE SUPPLIER


]]> http://www.caeda.com.cn/blog/article.asp?id=111 caeda@vip.sina.com(admin) Fri,27 Feb 2009 10:09:24 +0800 http://www.caeda.com.cn/blog/default.asp?id=111
Ioso items：以下论文可以索取

-HOW TO EXECUTE ROBUST DESIGN OPTIMIZATION

-MULTI-OBJECTIVE OPTIMIZATION OF TURBOMACHINERY CASCADES FOR MINIMUM LOSS, MAXIMUM LOADING, AND MAXIMUM GAP-TO-CHORD RATIO

-ROBUST OPTIMIZATION OF CONCENTRATIONS OF ALLOYING ELEMENTS IN STEEL FOR MAXIMUM TEMPERATURE, STRNGTH, TIME-TO-RUPTURE AND MINIMUM COST AND WEIGHT

-THE USE OF VARIABLE FIDELITY MODELS IN THE OPTIMIZATION PROBLEMS FOR AIRCRAFT ENGINES PARAMETERS AND CONTROL LAWS

-MULTI-DISCIPLINARY DESIGN OPTIMIZATION

-PARALLEL THERMOELASTICITY OPTIMIZATION OF 3-D SERPENTINE COOLING PASSAGES IN TURBINE BLADES

-MULTI-OBJECTIVE APPROACH FOR ROBUST DESIGN OPTIMIZATION PROBLEM

-OPTIMIZATION OF A LARGE NUMBER OF COOLANT PASSAGES LOCATED CLOSE TO THE SURFACE OF A TURBINE BLADE

-ROBUST DESIGN OPTIMIZATION STRATEFY OF IOSO TECHNOLOGY

-INDIRECT OPTIMIZATION METHOD ON THE BASIS OF SELF-ORGANIZATION

-INCREASING OF FAN EFFICIENCY BASING ON OPTIMIZATION TECHNOLOGY

-THE MAIN FEATURES OF IOSO TECHNOLOGY USAGE FOR MULTI-OBJECTIVE DESIGN OPTIMIZATION

-FAST METHODS OF EXPERIMENTAL CALIBRATING OF MICROPROCESSOR CONTRAL SYSTEM

-OPTIMIZATION OF THE GAS TURBINE ENGINE PARTS USING METHODS OF NUMERICAL SIMULATION

APPLICATION OF OPTIMIZATION PROGRAM IOSO NM AND ABAQUS AT CIVIL STRUCTURES OF NPP

OPTIMIZATION OF CONCENTRATIONS OF ALLOYING ELEMENTS IN STELL FOR MAXIMUM TEMPERTURE, STRENGTH, TIME-TO-RUPTURE AND MINIMUM COST AND WEIGTH

-CALIBRATION OF MICROPROCESSOR CONTROL SYSTEMS FOR SPECIFIED LEVELS OF ENGINE EXHAUST TOXICITY

-SAM 146 FAN STRESS CHARACTERISTICS OPTIMIZATION BY IOSO

-OPTIMIZATION OF TURBINE DISK AIMED TO MASS AND STRESS REDUCTION

-MULTISTAGE AXIAL FLOW COMPRESSORS OPTIMIZATION USING 3D CFD CODE

-THE TECHNOLOGY OF MULTIPURPOSE OPTIMIZATION OF GAS-TURBINE ENGINES AND THEIR COMPONENTS

-MULTI-DISCIPLINARY ANALYSIS AND DESIGN OPTIMIZATION

-SEMI-STOCHASTIC MULTI-OBJECTIVE OPTIMIZATION OF CHEMICAL COMPOSITION OF HIGH TEMPERATURE AUSTENITIC STEELS FOR DESIRED MECHANICAL PROPERTIES

-TWO APPROACHES TO MULTIDISCIPLINARY OPTIMIZATION PROBLEMS

-MULTI-OBJECTIVE ROBUST OPTIMIZATION USING IOSO TECHNOLOGY, PART II: REAL LIFE EXAMPLES

-STOCHASTIC OPTIMIZATION OF PRRAMETERS AND CONTROL LAWS OF THE AIRCRAFT GAS-TURBINE ENGINES – A STEP TO A ROBUST DESIGN

-MULTI-DISCIPLINARY HYBRID AND EVOLUTIONARY OPTIMIZATION

-DESIGN OF ALLOY’S CONCENTRATIONS FOR OPTIMIZED STRENGTH, TEMPERTURE, TIME-TO-RUPTURE, COST AND WEIGHT

-OPTIMIZING CHEMISTRY OF BULK METALLIC GLASSES FOR IMPROVED THERMAL STABILITY

-OPTIMAL DESIGN AND CONTROL OF GAS-TURBINE ENGINE COMPONENTS: A MULTICRITERIA APPROACH

-
]]> http://www.caeda.com.cn/blog/article.asp?id=110 caeda@vip.sina.com(admin) Thu,12 Feb 2009 09:45:48 +0800 http://www.caeda.com.cn/blog/default.asp?id=110
1)Advanced durability:
-Life interface for shape optimization.
Support of DAMAGE and DAMAGE_LC Design responses(one datavloocks 600 and 601)
-Support of temperature analysis
  TEMPERATURE_FILE in FEM_INPUT

2)Advanced nonliner
-optimization based on ninlinear results from the used solver (independent to the optimization type)
-optimization of contact surface to homogenize the contact stress for shape for shape optimization
-shape optimization based upon a geometric non-linear analysis
-shape optimization using non-linear materials
-support of temperature analysis
TEMPERATURE_FILE IN FEM_INPUT

3)Advanced acoustics
-Support of acoustic and frequecy response design responses
TYPE=FS_*

more ask: caeda@vip.sina.com 点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=109 caeda@vip.sina.com(admin) Wed,28 Jan 2009 10:14:43 +0800 http://www.caeda.com.cn/blog/default.asp?id=109
Design

implemented safe draping simulation method, to prevent unexpected crashes
implemented view filter for plies in layup builder window
added Delete Laminates menu command

Manufacturing
added support for generation of exact flat patterns, taking layer offset into account
added Ply Area and Mass to the available plybook export columns

Analysis
implemented new Optistruct optimiser export interface
Post-processing
enabled 3 user-defined spectra for results post-processing
added global ply id and ply name to results summary report
added ply name to plot window popup for a layer
implemented interactive LaRC05 failure criterion
implemented interactive Cowin failure criterion
implemented interactive Hankinson failure criterion
implemented interactive Hoffman failure criterion
implemented partial, user controlled, result type import from op2 or f06
added post-processing support for SOL103 eigen value, SOL111 (incomplete) modal frequency response, and SOL112 modal transient response dynamic results
implemented new failure results generation for a single element set only, as an option
introduced parameter to control the maximum lines in the results list feedback (default=50)
added Show ALL button in setting results scope window, to easily disable view filter
added separate call of user-defined criterion for layer interfaces
User Interface
saved plot windows to desktop layout

implemented 64-bit version
enabled multiple instances on same machine for each licence
added ply node under each layup ply node, for easy graphics and modifications

enhanced Ansys material data import

improved automatic web updates checks

upgraded installation environment

PlyMatch
blocked calibration if coordinates differ by given tolerance
implemented better feedback for error situations
made dialogs large size, to view from a distance
implemented fixed camera option
allowed the option of no tool probe being used
added support for user-defined line thickness
added keyboard shortcut for open calibration file function
enabled targeting the tool probe for easy calibration / check
]]> http://www.caeda.com.cn/blog/article.asp?id=108 caeda@vip.sina.com(admin) Tue,13 Jan 2009 09:26:18 +0800 http://www.caeda.com.cn/blog/default.asp?id=108
Many users who are running steady-state simulations like to watch the results develop. Most of the time users hit the Reload the Same Data File(s) again and again. You will be happy to hear that there are two easier ways to do it.

First, you can ask the solver (CFD-FASTRAN or CFD-ACE+) to write out intermediate results (say, every 50 iterations) to uniquely named files. Then you can animate those files using VIEW's file-based animation. Second, you can tell VIEW to *automatically* reload the same data file(s), using a device known as a "Trigger" file. Each of these options is described below.


Animation of Unique Iteration Stamped Files
In a CFD-ACE+ simulation, choosing the option for writing a unique DTF file at a specified interval of iterations in the Out/Output section of the CFD-ACE-GUI allows a user to output unique iteration-stamped DTF files. For example, if the steady state simulation is to be run for a total of 100 iterations and the user sets the iteration frequency to 25, DTF files will be output at the end of iterations 25 (modelname_steady.000025.DTF), 50 (modelname_steady.000050.DTF) and 75 (modelname_steady.000075.DTF) in addition to the final results at the end of the run that are stored in the file modelname.DTF.

At the end of the run, the steady state files can be animated in CFD-VIEW by using the procedure outlined in the user tip on File Based Animation with CFD-VIEW.

For a CFD-FASTRAN steady state run, unique iteration stamped files can be obtained by specifying a positive cycle value for the Cycles Interval under the Out/Output/Output Cycles/Cycles Interval field in the CFD-FASTRAN-GUI. The procedure for animating these results is the same as mentioned for the CFD-ACE+ files in the previous paragraph.


Automatic Reloading Using Trigger Files
The Trigger File update option in CFD-VIEW allows users to monitor a developing steady state solution in a running CFD-ACE+ or CFD-FASTRAN simulation. If you are interested in learning more about this feature, please read on.

Trigger File Based updating tracks a running simulation by means of tracking the modification time of a trigger file (modelname.RUN for CFD-ACE+ and modelname.TRIG for CFD-FASTRAN). The solver first generates a data set and then activates the trigger file to make CFD-VIEW aware of the new data. Each time the trigger file is changed, the solution from the data files is automatically reloaded into CFD-VIEW. This is the same as performing a manual operation of Reload the Same Data File(s) under the File menu in VIEW.

For the trigger file updating, the user should set the option for writing a DTF file at a specified interval of iterations in the Out/Output section of the CFD-ACE-GUI with the Unique File option turned off. This will ensure that the DTF file will get updated with the results at the specified iteration frequency. Since CFD-VIEW automatically reloads the same DTF file upon a trigger event, it is necessary to overwrite the DTF files with the latest results rather than write unique iteration-stamped DTF files.

For a CFD-FASTRAN simulation a DTF file can be updated with results at a specified cycle frequency by specifying a negative cycle value for the Cycles Interval under the Out/Output/Output Cycles/Cycles Interval field in the CFD-FASTRAN-GUI. For example, specifying a value of -100 for the Cycles Interval in a 1000 cycle simulation will overwrite modelname.DTF with results at the end of every 100 cycles.

To perform automatic updating using the Trigger File in CFD-VIEW, please follow the following steps in CFD-VIEW as the CFD-FASTRAN-SOLVER (or the CFD-ACE-SOLVER as the case maybe) is running.

Select the Import Additional Data File... option from the File menu to load the DTF file of interest.
Configure the model as desired (select variable, create cutting planes, choose color attributes, etc.) in CFD-VIEW.
Click on the Select Trigger File option from the File Menu. Choose the trigger file (modelname.TRIG) from the Open Trigger File dialog. Note: The trigger file will have a *.RUN extension for a CFD-ACE+ simulation.


Select the Enable Trigger option from the File Menu to turn on trigger file monitoring. From this point forward, each time the trigger file (or the RUN file for CFD-ACE+) is touched by the CFD-FASTRAN-SOLVER, CFD-VIEW will reload its data set.


Please click on this link ( updating using trigger files) to view an animated GIF of the pressure variable being tracked over a NACA0012 airfoil at an interval of 100 iterations over the course of 1000 CFD-FASTRAN-SOLVER iterations.
The animated gif was recorded using the procedure outlined in the tip on "File Based Animation with CFD-VIEW".


Note about Transient Simulations
Since the trigger file update option is based upon reloading the same data set during the course of the simulation, use of the trigger file is not suitable for transient simulations. This is so because the user would already be writing out unique time stamped DTF files which can be animated in VIEW to track the solution over time.
]]> http://www.caeda.com.cn/blog/article.asp?id=107 caeda@vip.sina.com(admin) Fri,09 Jan 2009 11:45:22 +0800 http://www.caeda.com.cn/blog/default.asp?id=107
As a result of a new partnership between Componeering Inc. and EVEN - Evolutionary Engineering AG, the post processing capabilities for ESAComp are provided by ComPoLyX software developed by EVEN. (See Press Release, Sept. 26, 2008)

ESAComp add-on modules for FE post processing are replaced by the enhanced capabilities of ComPoLyX. The current version of ComPoLyX supports exchange of material data with ESAComp. Closer integration of ComPoLyX and ESAComp is under way. The ESAComp database and material definition capabilities (Database Module) will be utilized in ComPoLyX for efficiency and ease of use. Further on, the composites failure analysis capabilities will be merged to guarantee consistency in the different phases of the design process. Already now the ComPoLyX failure analyses are compatible with ESAComp.

ComPoLyX is distributed as part of the ESAComp software suite through the ESAComp sales channels.
点击下载此文件
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=106 caeda@vip.sina.com(admin) Mon,22 Dec 2008 08:49:21 +0800 http://www.caeda.com.cn/blog/default.asp?id=106 The work was completed with the FINE/Design3D design and optimization software.
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=105 caeda@vip.sina.com(admin) Fri,19 Dec 2008 13:58:31 +0800 http://www.caeda.com.cn/blog/default.asp?id=105
ESI Group发布CFD-ACE+ V2009.0

Full Porous Media with Two-Fluid Capability Represents Significant Breakthrough for CFD and Multiphysics Applications

Paris, France – December 9, 2008 – ESI Group (ESIN FR0004110310), a pioneer and world-leading supplier of digital simulation software for prototyping and manufacturing processes, announced today the release of CFD-ACE+ V2009.0.  This version strengthens the modular and expandable design of CFD-ACE+ which offers users the highest level of flexibility.  ESI Group’s CFD-ACE+, the industry’s most advanced Computational Fluid Dynamics (CFD) and Multiphysics software, enables coupled simulations of fluid, thermal, chemical, biological, electrical and mechanical phenomena for a variety of applications across a wide range of industries.  

There are over 30 additions and improvements in the V2009.0 release across all applications.

Improved Fast Time Stepping (FTS) Performance
CFD-ACE+ V2009.0 has focused on the market’s demands for reducing costs. Indeed, CPU costs of transient simulations can be reduced significantly using Fast Time Stepping by ensuring convergence of each time step within a few iterations. FTS was first implemented in V2008.2 and has now been improved to allow for auto time stepping.  It has been extended to support additional modules including porous media, deforming grids and free surfaces (VOF).

Improved porous media with two-fluid capability
A major highlight of CFD-ACE+ V2009.0 is the improvement of the two-phase module with a new numerical formulation that is extensively more robust and faster to converge. In addition, two-phase coupling allows the most accurate modeling of two-phase simulations in porous media regions. This new improvement enhances ESI Group’s capabilities for Proton Exchange Membrane Fuel Cell (PEMFC) modeling and helps address common deficiencies of previous model formulations. This is a significant advancement for the simulation of fuel cell water management.  Additionally, a large number of processes and applications can now be addressed efficiently using the enhanced two-fluid model.

Advanced front methodology for unstructured triangular surface meshing in CFD-GEOM
V2009.0 also includes advances in accuracy of the CFD-ACE+ solver and the mesh generation capabilities in CFD-GEOM.  Indeed, the general purpose mesh generation tool CFD-GEOM has added advancing front surface triangulation to its already impressive set of surface meshing tools. Users can now select from Delaunay triangles, advancing front triangles, quad morphing, or quad paving algorithms. ESI Group wants to give to users the ability to select the best surface mesh type for the specific application. Other CFD-GEOM improvements include remeshing of discrete surfaces and boundary layer generation from discrete surfaces which make working with existing (legacy or 3rd party) meshes easy.

CFD-ACE+ V2009.0 also offers a unique and innovative tool – ESI CFD Mobile support – providing remote access to CFD-ACE+ core applications (CFD View and CFD GUI). Indeed, ESI CFD Mobile allows engineers and managers to have handheld access to their post-process solutions remotely from their iPhone or iPod Touch.

“Release V2009.0 includes more than 30 new features and enhancements, including significant advances for our Fuel Cell, Energy, and Electronics industry customers.  This release comes just six months after the last major CFD-ACE+ release, thus underscoring our commitment to quickly addressing market requirements,” said Joseph Strelow, Director of Design and Analysis Simulation at ESI Group. “With the increased usability, robustness, and accuracy V2009.0 provides, we are confident the market will recognize the enhanced value provided by the CFD-ACE+ toolset,” he continued.



Boundary Layer Meshing

About ESI Group
ESI Group is a world-leading supplier and pioneer of digital simulation software for prototyping and manufacturing processes that take into account the physics of materials. ESI Group has developed an extensive suite of coherent, industry-oriented applications to realistically simulate a product’s behavior during testing, to fine-tune manufacturing processes in accordance with desired product performance, and to evaluate the environment’s impact on product performance. ESI Group’s products represent a unique collaborative and open environment for Simulation-Based Design, enabling virtual prototypes to be improved in a continuous and collaborative manner while eliminating the need for physical prototypes during product development. The company employs over 700 high-level specialists worldwide covering more than 30 countries. ESI Group is listed in compartment C of NYSE Euronext Paris. For further information, visit www.esi-group.com.
                

欢迎咨询中国授权代理商CAEDA Ltd.：caeda@vip.sina.com, www.caeda.com.cn     
]]> http://www.caeda.com.cn/blog/article.asp?id=104 caeda@vip.sina.com(admin) Mon,15 Dec 2008 13:46:41 +0800 http://www.caeda.com.cn/blog/default.asp?id=104
一、前言

    航空发动机是精密、复杂的高技术产品，大量零部件在十分恶劣的环境下工作，承受着高温、高压和高转速的工作负荷，以满足高性能的要求，并确保安全科考的运行。发动机的设计参数的微笑变化都有可能极大的影响发动机推力及耗油率，设置时发动机和飞机的制造以及使用成本。推力大、耗油率低是极具竞争力的，但在满足推力大、耗油率低的同时也必须考虑到其他的一些指标（如尺寸、质量等）的限制，从而形成了发动机总体性能方案的多目标优化设计问题。该优化问题的目标函数和约束函数都是非线性函数，并且无法用数学显式来表达，导数求解困难。传统的处理此类多目标优化问题的方法是构建一个效应函数，将多目标优化问题转化为单目标优化问题，然后利用成熟的求解单目标优化问题的方法获得问题的一个解。为了获得多目标优化问题的Pareto最优解集的一个子集，必须求解一系列的单目标优化问题，及时最优化求解很成功，这种“点到点”的寻优过程也不能保证Pareto最优性。 IOSO NM是一个独特的软件包，为解决“复杂” 实际问题提供了新的机会。它被用来改善复杂系统性能，技术设备和工程进程，并基于探索原材料最佳参数，以开发新材料该软件同时使花费较低的研究支出和较短的执行时间成为可能。该系统的主要目的是，将设计者或研究者从探索最优的系统设计参数的非常复杂过程以及繁重的体力劳动中解放出来。针对航空发动机设计这样的多目标优化问题能够提供有效的解决。

附件完整文章： 点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=103 caeda@vip.sina.com(admin) Tue,09 Dec 2008 08:45:19 +0800 http://www.caeda.com.cn/blog/default.asp?id=103
Have you ever wanted to find the differences between two DTF files? If you have, then you will probably be interested in a DTF utility included in your ESI Software installation. The utility is called dtf_diff and can be found in your UTILS_2009.0/bin directory.


The dtf_diff utility can be used to compare two DTF files and provides output on the differences between them. The usage for the dtf_diff utility is

dtf_diff dtf_filename_1.DTF sim_num dtf_filename_2.DTF sim_num
The sim_num part is the simulation number (in most cases this is one) for the DTF file you want to compare. By default, all of the information in the DTF files is compared. However, there exist several options to ignore certain information in the DTF files. The options are:

-igsim: ignore the differences in the simulation data
-igvc: ignore the differences in the volume/initial data
-igbc: ignore the differences in the boundary record data
-igunit: ignore the differences in the units
-igversion: ignore the DTF version number differences
-igguiversion: ignore the CFD-ACE-GUI version number differences
-ig varname: ignore specific name specified by varname
There are also two additional options for more control over the comparison:
-geom: compare the geometry data
-p number: set the float precision for the comparison (default is exact comparison)

As an example, I will compare two DTF files I have setup to run a heat conduction case. The only difference between the files is the temperature I have specified on one of the boundaries, and the filenames. The usage for this case is:

dtf_diff heat_conduction.DTF 1 heat_conduction_1.DTF 1
The output will be written to the command line, or you can pipe the output into a file like this:

dtf_diff heat_conduction.DTF 1 heat_conduction_1.DTF 1 > differences.dat
Keep in mind that you can only use the dtf_diff utility from the command line. The utility output for this example is presented in Figure 1 below. It shows that the filenames are different and the BC record with key 54 have a different temperature.



Figure 1. Output from the dtf_diff utility


You can then open the DTF file in CFD-ACE-GUI, go to the BC tab and sort the BC's by key number in the BC Explorer, and verify that the BC with key 54 is indeed different in the two files.

This is a very useful and easy to use utility. The next time you want to compare DTF files, try using it.


If you have any questions about this tip or would like us to discuss other topics in the future, please let us know
]]> http://www.caeda.com.cn/blog/article.asp?id=102 caeda@vip.sina.com(admin) Mon,08 Dec 2008 09:52:59 +0800 http://www.caeda.com.cn/blog/default.asp?id=102
ESI Group, a pioneer and world-leading supplier of digital simulation software for prototyping and manufacturing processes, announced today its partnership with Optimal Solutions Software to reduce the time required for complex design studies.
With the combination of ESI Group’s CFD-ACE+ computational fluid dynamics (CFD) solution and Optimal Solutions Software’s SculptorTM, the world’s leading Computational-Aided Engineering (CAE) real-time design deformation technology, users benefit from the ideal solution to improve productivity and overall speed.
    

ESI Group’s proven solution CFD-VisCART allows users to quickly and easily mesh complex geometry. This automated, easy to use meshing solution can generate a mesh for a complete car overnight compared to days or weeks using other tools. As design studies are evaluated, Sculptor is then used to quickly deform the mesh in real-time for subsequent runs. This is all completed without having to go back to the CAD model or do any additional re-meshing. The benefit to the end-users is the potential for them to save weeks or even months in a typical automotive aerodynamics analysis for a new car program.
“The partnership with Optimal Solutions Software further solidifies ESI Group’s value proposition of reducing the total time required to execute complex design studies”, said Joseph Strelow, Director of Design and Analysis Simulation for ESI Group. He added: “The solution begins when a CAD model is delivered and finishes when a final design is approved. We’re now able to assist in reducing development times by up to 30% due to recent advances in ESI’s CFD-VisCART and CFD-ACE+ Multiphysics products. Combining these with Sculptor’s morphing technology provides our customers with an even more powerful solution.”
“Our customers are continuously searching for better tools to help them improve their designs for maximum performance. Using traditional tools and techniques for this ongoing improvement effort requires more and more time which translates to more money,” stated John Jenkins, VP Sales & Marketing for Optimal Solutions Software. “This partnership gives our customers a powerful CAE solver linked to a revolutionary deformer/optimizer so that they can quickly and easily improve their designs without the increase in time and money required in the past. Our customers are saving tens of thousands of dollars while coming up with better designs in less than 25% of the time that it would have taken them using their old design improvement approaches.”

]]> http://www.caeda.com.cn/blog/article.asp?id=101 caeda@vip.sina.com(admin) Thu,04 Dec 2008 08:18:24 +0800 http://www.caeda.com.cn/blog/default.asp?id=101
ESI Group's CFD Division is proud to announce the V2009 release. The fact that this release comes just six months after our last major release proves that ESI Group is working hard to implement the features that the market demands in a timely manner. We welcome and encourage all of our customers to download and use the latest versions, which include numerous enhancements.

Included in the V2009 release are significant updates to:

CFD-ACE+ Advanced CFD and Multiphysics Solver
CFD-FASTRAN CFD Physics Solver
CFD-CADalyzer Upfront CFD Design Tool

CFD-GEOM Geometry Modeler and Mesh Generator
CFD-VisCART Cartesian Mesh Generator
CFD-VIEW Solution Analysis and Visualization Tool
Enhancements have been made to increase usability, robustness and accuracy, and we are sure that these new versions will make for a worthwhile upgrade to your CFD toolset.

The remainder of this note gives you some information about the V2009 software release. Please read on to learn more about the updates and let us know if you have any comments or questions.

IMPORTANT NOTE: You will require a new license to run V2009.x versions! ESI Group licensing policy requires licensing to the version of the software. This means that if you currently have a V2008.x license file you will need to replace it with a V2009.x license file. These are available upon request to your ESI license or sales representative. If you are not sure who that is, you should contact support.cfd@esi-group-na.com, and please include your existing V2008.x license file for the quickest service.

V2009 Release Highlights

As with previous releases, there are many improvements to features, performance, and robustness. Below is a short bullet list of some of the top reasons why you should consider upgrading to V2009. For a full description of the new and improved features, as well as usage and limitation information, please consult the relevant release note document, which is included in the RELEASE package download files and can also be found here (login required).

CFD-ACE V2009.0 (view full release notes - login required)
Improved Fast Time Stepping (FTS) performance
Extended Fast Time Stepping (FTS) to VOF and Stress modules
Automated Force and Moment coefficient calculations on specified surfaces

Force and moment information now available for momentum resistance regions
Improved convergence for two-fluid module
Porous media with two-fluid capability
Improved v2f turbulence model
ESI Mobile support (control runs / view residuals remotely on iPhone or iPod-touch)
CFD-FASTRAN V2009.0 (view full release notes - login required)
Spatially varying boundary conditions
AUSM scheme
CFD-CADalyzer V2009.0 (view full release notes - login required)
Improved animation recording
Autodesk Inventor 64-bit Support

CFD-GEOM V2009.0 (view full release notes - login required)
Remeshing of discrete surfaces
Better handling of discrete shells and domains for use in volume meshing
Boundary layers for discrete surfaces
Advancing front methodology for unstructured triangular surface meshing
Ability to split trimmed surfaces
Ability to combine surfaces
Tool to minimize number of curves

CFD-VisCART V2009.0 (view full release notes - login required)
Smooth / delayed transition of cell sizes
Bounding box surface can be boundary for mesh refinement
Faster mesh generation with less memory usage
Batch scripting improvements

CFD-VIEW V2009.0 (view full release notes - login required)
ESI Mobile support (post-process solutions remotely on your iPhone or iPod-touch)
Improved animation quality
Improved macro user interface
V2009.0 Installation
The V2009.0 RELEASE software will be installed parallel to your existing V2008.2 (or earlier) RELEASE installation. This means that you will have access to both versions should you desire. However, only V2009.0 will be supported going forward so you should use V2009.0 to get access to future bug fixes and feature improvements. Step-by-step instructions for the installation of V2009.0 can be found here (login required).

如需要CFD-Fastran/ACE+软件试用，请填写附件表格发至：caeda@vip.sina.com
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=100 caeda@vip.sina.com(admin) Tue,02 Dec 2008 09:20:04 +0800 http://www.caeda.com.cn/blog/default.asp?id=100
As you know we have application examples in turbomachinary(presentations - that you have also) - CFD calculations were done with NUMECA mostly. I think it could be good idea for you to offer our IOSO software to help these companies solve optimization tasks with NUMECA and IOSO.

The main advantages of our IOSO Optimization Technology is that it purposed for "heavy", real-life optimization tasks. We proved our leadership as regards sophisticated optimization tasks applying our IOSO technology in industry: http://www.iosotech.com/cooperat.htm and on world conferences: http://www.iosotech.com/publicat.htm We have experience solving very complex optimization tasks that have no analogous in published literature(the order of task may have 150 independent variables and 15 criteria). About comparison testing results of the IOSO with well-known optimization algorythms you can read from here: http://www.iosotech.com/compare.htm  http://www.iosotech.com/stoch.htm

I think the above information is necessary to explain why our software fits turbomachianry and aviation more than other optimization applications and algorithms.

To understand how NUMECA operates with our software I can give you an example of coupling of our software and NUMECA FINETM/DESING3D.

NUMECA offer turbine developers their product that consists of 3 modules:
1) AUTOBLADE - parametrizator of blades geometry

2) AUTO GRID - automatic mesh constructor for blades

3) FINE/TURBU - solver

All these programs are launched consistently through *.bat file during optimization process.



There are 2 standard text files that one can use to transfer data between IOSO and NUMECA:

1) *.par file consists of geometrical and other parameters of AUTOBLADE model, user can construct one reference model and then change a geometry of that model by changing necessary parameters in *.par file

2) *.mf file consists of solution results, one can analyze this results by reading necessary parameters of that file



Please, find attached to this e-mail a scheme of a some stage optimization by IOSO & NUMECA. Also, find attached in this e-mail some .par and .mf files of a certain NUMECA project.
点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=99 caeda@vip.sina.com(admin) Thu,27 Nov 2008 10:46:04 +0800 http://www.caeda.com.cn/blog/default.asp?id=99
Ø        美国船舶技术部(Ship Technology Division)开发的LAMP(Large Amplitude Motions Program )大型三维时域计算系统，它用于解决平台动力学、波浪载荷及结构响应等动力学及耐波性相关的问题。

Ø        美国造船委员会(American Boat & Yacht Council)1995年开发的PowerSea商业应用软件，用来帮助船舶工程设计人员对滑行艇耐波性分析。

Ø        Skip是意大利船舶研究中心(Italian Ship Research Centre)研究开发的，这是预测单体船耐波性的商业代码，它依据切片理论计算船体在波浪中的六个自由度的运动。

Ø        VisualSMP是加拿大国际船舶软件联合会(The International Marine Software Associates (IMSA))开发的一套预报分析船舶耐波性的工具，它依据切片理论在频域中开发出的软件，而且具有前后处理及船舶在海上运动的仿真及可视化功能。

Ø        澳大利亚Formation Design Systems公司开发成功的MAXSURF船舶设计软件系列中的SEAKEEPER模块是综合性船舶耐波性设计软件。开发的SEAKEEPER船舶耐波性设计软件运用标准的Strip（薄片）理论，可对各种船舶在各种规定航线、海况和载况下的典型摇荡运动响应进行科学的预报。应用SEAKEEPER软件的这些计算结果，设计者就可在船舶后续的详细设计和生产设计阶段对船舶诸多方面的性能进行耐波性船型的优化设计。

]]> http://www.caeda.com.cn/blog/article.asp?id=98 caeda@vip.sina.com(admin) Thu,20 Nov 2008 09:37:56 +0800 http://www.caeda.com.cn/blog/default.asp?id=98
有关产品中文翻译：www.caefinder.com

附件请参考：


点击下载此文件
]]> http://www.caeda.com.cn/blog/article.asp?id=97 caeda@vip.sina.com(admin) Tue,18 Nov 2008 08:58:03 +0800 http://www.caeda.com.cn/blog/default.asp?id=97
In CFD-GEOM, a convenient meshing tool called "Butterfly Faces" allows the user to quickly and automatically create a structured mesh for circular topologies. The resulting mesh consists of 5 structured faces (4 sides and 1 core) inside a circle. Parameters are available to set the size of the core and the mesh density.



Figure 1a. Circle with four edges                                          Figure 1b. Resulting butterfly mesh


To start, create a circle and split it into 4 arcs. Create edges on the arcs, making sure that opposite sides have the same number of grid points. In this case (Figure 1a), 11 grid points have been placed on each arc.

In the "Structured Face Options" panel, select the "Butterfly Face" tool, shown in figure 2.



Figure 2. butterfly mesh tool

Select each edge followed by a middle mouse click. Figure 3 depicts what you should see after picking all four edges.



Figure 3. Edge selection for the butterfly mesh tool

The user has control over the radius of the Major and Minor axes. Note: this radius is measured from the center of the circle. Figure 4 shows the Butterfly faces setting options. Once the major and minor radii have been defined, click Apply and you will get a Butterfly mesh as shown on top of this page in figure 1b.



Figure 4. Butterfly faces setting options

  

If you have any questions about this tip or would like us to discuss other topics in the future, please let us know
]]> http://www.caeda.com.cn/blog/article.asp?id=96 caeda@vip.sina.com(admin) Wed,12 Nov 2008 08:34:07 +0800 http://www.caeda.com.cn/blog/default.asp?id=96
Batch processing through scripting appeared in CFD-VisCART v2006 and this feature kept improving since then. You can see a short description in this past user tip. This can be very useful to create meshes without graphical interaction or to set up similar models over and over. In CFD-VisCART v2009, a new feature will make its use even easier and powerful: You will now be able to dump a batch script from an existing VGD file.

After loading a pre-set vgd file in CFD-VisCART graphical interface, or creating one, you can now output a script file automatically with all the commands needed to recreate this same VGD file in batch mode.
The main advantage obviously is the ability to easily create script files for similar models. After setting up one model, you can dump one script, and modify it for all your similar models. Parameters, such as mesh sizes, components, mesh sources…, can be modified and adapted.
Another advantage of this feature is that you don’t need to know the batch script commands to write the scripts you need. You can simply set up a vgd file as you would normally do in the graphical window and then dump a script that you can reuse, without knowing any scripting commands

You can use this feature from the File menu (Figure 1) or the Toolbar (Figure 2).



Figure 1. Dump Script option from the File menu




Figure 2. Dump Script option from the Toolbar


After clicking on the Dump Script icon, a file saving dialog button will appear to specify the dump script file. The Default extension is .vbt (CFD-VisCART batch text file), but you can specify any name and extension you would like.

]]> http://www.caeda.com.cn/blog/article.asp?id=95 caeda@vip.sina.com(admin) Wed,05 Nov 2008 09:56:51 +0800 http://www.caeda.com.cn/blog/default.asp?id=95
CFD-VisCART is a 3D unstructured adaptive Cartesian mesh generator. It is ideally suited for external aerodynamics and automotive underhood applications involving extremely complex geometries. Models/geometries in SAT, Parasolid, STL, Nastran, 3DS, Plot3d and FAST formats can be imported into CFD-VisCART.

A typical complete assembly/model of a car or an aircraft consists of hundreds of parts and surfaces. While reading in a model into CFD-VisCART, the user is asked if the model should be read in as a single patch or surface (figure 1). This is a recommended option (specially for large models), as it would result in lower memory being used and consequently faster processing.



Figure 1. Option for reading in as a single patch.

While setting up the meshing parameters, some surfaces of the model may need to be assigned as sources. Also, the user may want to blank-out some of the surfaces to be able to see other underlying surfaces. For these and other similar reasons, the user may want to split the model to some extent after reading it in as a single patch. This tip illustrates the various ways in which a model can be split in CFD-VisCART by using some basic tools (see figure 2).



Figure 2. Detection and splitting tools


Split using Outlines

An outline is a curve that links the continued edges of the model if they are only shared by one surface triangle or by more than two triangles. To use this option for splitting the model, use the following steps :

Select the model
Click the Outline Detection button - Outlines appear on the geometry (see figure 3)
Click the Split Selected Geometry button - a pop-up window shows the user the number of surfaces the model will be split into and asks whether or not to proceed. Click Yes. The result is showed in figure 4.



Split using Critical Features

A critical feature is the curve that links the continued edges of the model if the edge is shared by only two triangles and the angle between these two triangles is smaller than that defined by the user. To use this option for splitting the model, use the following steps:

Select the model
Click the Critical Feature Detection button - a Set Feature Detection Options panel pops up, depicted in figure 5. Here, the user can specify the minimum critical angle. Note: The smaller the angle, the larger the number of critical feature lines.


Figure 5. Critical feature detection

Click Detect - Critical feature lines appear on the model
Click the Split Selected Geometry button

Split using Cutting Planes

Cutting planes, normally used to view the grid on a cross-section, can also be used to split a model. To use this option for splitting the model, use the following steps:

Scroll the cutting plane to the desired location. Figure 6 shows the cutting plane as a light blue line.


Figure 6. Position the cutting plane at the desired location

Select the model
Click the Split Selected Geometry button - make sure no outlines or critical features are present, otherwise the model will be split along these lines.Figure 7illustrates the resulting geometry.



Figure 7. Geometry after splitting with a cutting plane

  

If you have any questions about this tip or would like us to discuss other topics in the future, please let us know
]]> http://www.caeda.com.cn/blog/article.asp?id=94 caeda@vip.sina.com(admin) Sun,02 Nov 2008 09:47:01 +0800 http://www.caeda.com.cn/blog/default.asp?id=94
ESI Group's CFD Division is proud to announce that the V2009 release is on schedule and ready for BETA evaluation. The fact that this release comes just six months after our last major release proves that ESI Group is working hard to implement the features that the market demands in a timely manner. We welcome and encourage all of our customers to download and use the latest versions which include numerous enhancements.

Included in the V2009 release are significant updates to:

CFD-ACE+ Advanced CFD and Multiphysics Solver
CFD-FASTRAN CFD Physics Solver
CFD-CADalyzer Upfront CFD Design Tool

CFD-GEOM Geometry Modeler and Mesh Generator
CFD-VisCART Cartesian Mesh Generator
CFD-VIEW Solution Analysis and Visualization Tool
Enhancements have been made to increase usability, robustness, and accuracy and we are sure that these new versions will make for a worthwhile upgrade to your CFD toolset.

The remainder of this note gives you some information about the V2009 software release and the BETA program. Please read on to learn more about the updates and how to participate in the BETA program and let us know if you have any comments or questions.

IMPORTANT NOTE: You will require a new license to run V2009.x versions! ESI Group licensing policy requires licensing to the version of the software. This means that if you currently have a V2008.x license file you will need to replace it with a V2009.x license file. These are available upon request to your ESI license or sales representative. If you are not sure who that is, you should contact support.cfd@esi-group-na.com, and please include your existing V2008.x license file for the quickest service.

V2009 Release Highlights

As with previous releases, there are many improvements to features, performance, and robustness. Below is a short bullet list of some of the top reasons why you should consider upgrading to V2009. For a full description of the new and improved features, as well as usage and limitation information, please consult the relevant release note document, which is included in the BETA package download files and can also be found here (login required).

CFD-ACE V2009.0 (view full release notes - login required)
Improved Fast Time Stepping (FTS) performance
Extended Fast Time Stepping (FTS) to VOF and Stress modules
Automated Force and Moment coefficient calculations on specified surfaces

Force and moment information now available for momentum resistance regions
Improved convergence for two-fluid module (may not be available until final release)
Porous media with two-fluid capability (may not be available until final release)
Improved v2f turbulence model
ESI Mobile support (control runs / view residuals remotely on iPhone or iPod-touch)
CFD-FASTRAN V2009.0 (view full release notes - login required)
Spatially varying boundary conditions (not available until final release)
AUSM scheme (may not be available until final release)

CFD-CADalyzer V2009.0 (view full release notes - login required)
Improved animation recording
Autodesk Inventor 64-bit Support

CFD-GEOM V2009.0 (view full release notes - login required)
Remeshing of discrete surfaces
Better handling of discrete shells and domains for use in volume meshing
Boundary layers for discrete surfaces
Advancing front methodology for unstructured triangular surface meshing
Ability to split trimmed surfaces
Ability to combine surfaces
Tool to minimize number of curves

CFD-VisCART V2009.0 (view full release notes - login required)
Smooth / delayed transition of cell sizes
Bounding box surface can be boundary for mesh refinement
Faster mesh generation with less memory usage
Batch scripting improvements

CFD-VIEW V2009.0 (view full release notes - login required)
ESI Mobile support (post-process solutions remotely on your iPhone or iPod-touch)
Improved animation quality
Improved macro user interface
V2009.0 Installation
The V2009.0 BETA software will be installed parallel to your existing V2008.2 (or earlier) RELEASE installation. This means that you will have access to both versions should you desire. However, only V2009.0 will be supported going forward so you should use V2009.0 to get access to future bug fixes and feature improvements. Step-by-step instructions for the installation of V2009.0 can be found here (login required).
]]> http://www.caeda.com.cn/blog/article.asp?id=93 caeda@vip.sina.com(admin) Tue,28 Oct 2008 08:25:11 +0800 http://www.caeda.com.cn/blog/default.asp?id=93
The mesh has a great influence on the solver convergence and solution of every CFD simulation. The user is strongly advised to check the quality of a numerical mesh. Several parameters that reflect the quality of a mesh are available. In CFD-GEOM, under the tab Mesh/Mesh Query Options, the user can check 2D as well as 3D mesh quality. In figure 1, 3D Mesh Queries for Aspect Ration has been called. See figure 2 for an example for the Centroid to Face Angle.



Figure 1. Mesh Queries tool


Aspect Ratio
If the model contains thin layers, which are mostly used as boundary layer or for the different layers of the MEA inside fuel cells (see figure1), it may require placing several cells normal to the plate (thickness). Due to this, aspect ratios of several thousands can occur (as shown in figure 1). According to our experience, aspect ratios for such models up to 1500 are acceptable, for flow, heat and electrochemistry. If the same geometry is used for a solid stress analysis, this ratio must be reduced down to 15. The reason for this great discrepancy may be in the different solver structure. It is strongly recommended to check this ratio in CFD-GEOM before writing out the DTF file.

Centroid/Face Angle
Especially in unstructured meshes, very small centroid to face angle can occur (see figure 2), mostly inside thin layers but also in arbitrary interfaces. When arbitrary interfaces are called, a common, new interface with additional cells is automatically created the solver. The quality of these cells will be checked by the solver and printed out in the CFD-ACE output file to inform the user about the status of this new interface (see figure 3). Models containing many bad cells as shown in figure 3 will not run successfully. By changing the edges mesh distribution and/or density as well as the number of cells, one can reduce the number of bad cells. As soon as the warning message no longer occurs, the solver will start running satisfactorily.



Figure 2. Centroid/Face Angle quality tool




Figure 3. Arbitrary interface data in output file


Volume
Curved surfaces or splines can create overlapping areas. Due to this, cells with negative volumes can occur. Mesh refinement and/or repairing of geometrical data can resolve this problem. The user can determine the number of such bad cells in CFD-GEOM as well as in the output file.
]]> http://www.caeda.com.cn/blog/article.asp?id=92 caeda@vip.sina.com(admin) Tue,21 Oct 2008 08:21:56 +0800 http://www.caeda.com.cn/blog/default.asp?id=92
In CFD modeling and analysis, there are times you would like to change the grid and rerun the case. These studies require coarsening and/or refining of the grid on the same model. In complicated 2D or 3D structured grid, changing the edge distribution on individual edges can be time consuming. Matching the same edge distribution on all edges in the group by changing the distribution on the edges one by one can become cumbersome. This work can be made very simple and quick by using the Edge Linking tool in CFD-GEOM.

The Edge Linking tool (Figure 1) allows you to link edges together such that when one edge in the group is updated, the changes are applied to all the edges in that group. By using this option, you can specify the complete grid propagation pattern throughout the model to automatically refine or coarsen the mesh.

Edge linking groups consist of two or more edges. The first edge added to the group is the Master Edge. Subsequent edges added to the group are called Secondary Edges. There are three link options available:

Number of points
Distribution
Both


Figure 1. Edge Linking Tool

Consider the simple example shown in Figure 2; the left edge is the master edge while the remaining three vertical edges are secondary edges. Edge linking groups are colored with the 'Picked Set 3' color, as opposed to the regular dark blue edge color.



Figure 2. Master and Secondary edges in example case

Number of Points: When a secondary edge is linked according to the number of points, it will always have the same number of grid points as the master edge, but its distribution type is independent of that of the Master edge.
Distribution: When linked according to distribution, the secondary edge will always have the same distribution as the Master edge, but the number of points on the secondary edge is independent of that of the Master edge.
Both: When a secondary edge has "Both" link type, it will always have the same number of grid points and distribution as the master. Within an edge linking group, secondary edges can have different link types. An edge can belong to one, and only one, edge linking group.

How to Create an Edge Linking Group:
Click the Edge Linking button in the Edge toolbox. The Edge->Link Edges panel appears
Select a master edge with the left mouse button. It will be highlighted with the "picked" color. When creating a new link group, you should not select a master edge or secondary edge from an existing group.
Select all additional edges to include in the group.
Select one of the Link Options in the Edge->Link Edges panel and click the Apply button. This will create the edge linked group. Please note that all edges in the group will get the currently selected link type.

How to Add an Edge to an Edge Linking Groups:
Select the Edge Linking button from the Structured Edge toolbox.
Select an edge in the group to which you want to add edges.
Select any additional edges. These edges should not belong to existing groups.
Select the link type and press the middle mouse button to complete the group. All edges in the group will get the currently selected link type.

How to Remove an Edge from existing Edge Linking Group:
Select the Edge Linking button from the Structured Edge toolbox.
Select the edge you want to remove. If the edge is part of an edge linking group, both it and the master edge will be highlighted.Note that you cannot remove the master edge from the group.
Click the Remove button in the edge linking toolbox. This will remove the edge from the edge linking group. You can repeat this process for other edges in the group.

How to Delete an Edge Linking Group:
Select a member of the group.
Press CTRL-A to select all.
Click the Remove button.

Useful Tips:
For edges that have faces built from them, it is recommended that you only use the link option on them when all edges in the group have the same number of grid points.
When you assign a "Both" or "Distribution" link type to a secondary edge, and the distribution type is hyperbolic tangent, the secondary and Master edges should be of similar length. Otherwise, you may encounter a situation in which the sum of the specified endpoint spacing is larger than the length of the secondary edge. In such situations, CFD-GEOM assigns a uniform power law distribution to the secondary edge. ]]> http://www.caeda.com.cn/blog/article.asp?id=91 caeda@vip.sina.com(admin) Tue,14 Oct 2008 08:23:10 +0800 http://www.caeda.com.cn/blog/default.asp?id=91
Special approaches are needed when dealing with the modeling and analysis of various complex rotating systems having a stator and a rotor, in single or multistage, with change in the frame of reference from fixed to rotating (namely fans, pumps, compressors, turbines, propellers, impellers, diffusers, stators, vanes etc.). Indeed, all parts attached to the rotating shaft will be rotating with a certain angular velocity with respect to the machine axis, while the stationary parts such as stator, casing, inlet(s), exit(s) etc., do not move. When viewed by a stationary observer, the flow around the moving parts is always unsteady, even if the flow is apparently of steady state. When viewed by an observer stationed on the rotor, such a flow appears steady. For unsteady flow problems such as transients, it does not matter whether the observer is situated on the stationary or moving parts.


So, the important issue is the transfer of information from one stationary frame to the moving reference frame and vice-versa. By changing the frame of reference, quantities like normal, shear and deformation stress tensors are invariant. However, quantities like velocity, acceleration and rotating tensors depend on the frame of reference, e.g. static pressure does not change with change in coordinate frame (Pstatic); but total pressure changes with the coordinate frame. Additional complexities in rotating systems arise from three dimensionality, complex shapes with curvature, accelerated/decelerated flows, rotor-wake-stator/stator-wake-rotor interactions, etc.

Hence, special approaches/models come in handy to convert a very complex transient problem into a steady state problem and thus reduce the modeling and analysis time drastically. Proven approaches are:

Body Force Model (also known as lumped parameter model, mostly applicable for fans),
Multiple Reference Frame model (MRF),
Mixing Plane Model (MPM),
Sliding Mesh method.
All the above models are made available in CFD-ACE+. This tip describes the Multiple Reference Frame (MRF) feature. If you are interested in this tip please read on.



The MRF model accounts for complete blade design and fan details to simulate complex turbo machinery. It is a steady state approximation where the fluid zone in the fan region is modeled as a rotating frame of reference and the surrounding zones are modeled in a stationary frame. Contrary to the Body Force Model, the MRF model includes the geometry of the fan blades. The fan blades are modeled stationary but since the fluid domains surrounding them are in a rotating frame, the pressure jump and the swirl components are given by the presence of blades as wall without the need of experimental data as an input.

The major difference between MRF and Mixing Plane Methods is that the MRF directly translates the properties of the flow at the interfaces between the rotating and stationary zones, whereas the MPM averages the properties of the flow circumferentially. The MPM avoids flow field non-uniformities that arise from the fact that the fan blades are modeled stationary.

The steady state approximation of MRF allows individual cell zones to rotate or translate with different speeds. This is achieved by dividing the whole problem domain (say a rotor/stator interaction for rotating machinery) into separate zones where the flow is solved in stationary or rotating coordinate systems. The MRF approach is good when the flow (say between two rotors moving in counter directions or a rotor and a stator) is nearly uniform.



MRF and Navier-Stokes Equations of Motion:
The MRF transforms the fluid velocities from stationary to rotating frames using the following relation:

Stationary Frame



Rotating Frame



Where: Velocity Relative to Rotating Frame = Absolute Velocity - Whirl Velocity




Solving the equations of motion in the rotating reference frame results in additional acceleration terms in the momentum equation. Also, for a reference frame rotating at a constant angular velocity, the body force per unit mass includes both Coriolis and centrifugal forces. MRF solves the equations of motion in both stationary frame (for absolute velocity) and rotating frame (velocity in the rotating reference frame). Therefore, with the general MRF capability, steady state analysis can be performed on various components of the rotating system using local reference frames, either stationary or rotating as appropriate.

As shown in figure 2, CFD-ACE+ offers three options under MRF. They are:




Figure 2: CFD-ACE+ Multiple Reference Frame Options




Frame (Global): Select this option when you would like to solve the system as a Rotating Frame Reference to yield relative velocity solutions. Inputs are: Point on axis of rotation and Angular velocity vector.
Multiple Frame (VC Based): Select this option for more than one rotating frame of reference, like for fans and impellers. Inputs are: Point on axis of rotation and Angular velocity vector.
Multiple Frame (Geometry based, see figure 3): Use this option when the rotating domains are not defined as separate volume conditions. Here, a virtual cylinder will be created to define the rotating region. Two levels adjacent to this region are used as the rotating frame boundaries. Inputs are: center location, normal direction, thickness, radius of virtual cylinder geometry and angular rotation speed.



Figure 3: CFD-ACE+ Multiple Reference Frame - Geometry Based Option


To work with the first two options of MRF, select ‘Rotation’ under ‘MO’ tab. Select ‘Rotation’ under ‘VC’ tab, activate ‘Rotating Frame’, choose Fluid zone to be rotated from Explorer and specify the values for input fields.

Select Multiple frame (Geometry based) option from ‘Rotation’ under ‘MO’ tab. Select ‘MRF’ tab adjacent to ‘BC’ tab in ACE-GUI control panel, hit on ‘Add’ button in the Explorer window to add a multiple reference frame, choose virtual cylinder zone ‘MRF’ from Explorer and specify the values for input fields.

Convention followed in CFD-ACE+ for Angular rotation speed; angle is positive for counter-clockwise rotation and negative for clockwise rotation.

All the boundary conditions values must be specified in absolute frame of reference. Walls that rotate and appear in absolute frame of reference, e.g. blade surfaces must be specified as rotating walls. The ACE-SOLVER will automatically convert all velocities specified from the absolute frame to the rotating frame using the frame information provided in the Problem Type setting.


Try using the MRF feature next time. If you have any questions about this feature or would like us to discuss other topics in the future please let us know.
]]> http://www.caeda.com.cn/blog/article.asp?id=90 caeda@vip.sina.com(admin) Tue,07 Oct 2008 08:16:47 +0800 http://www.caeda.com.cn/blog/default.asp?id=90
While working with CFD-VIEW, the user has the option of saving an mdl file. This file stores all the entities created in CFD-VIEW, from the point the DTF file (or other data file) was imported, to their current state at the time of saving. The user can thus reopen this file anytime later to resume work.

By default, the ‘Save All Data with Model’ option under the File menu is checked (see figure 1), making the mdl file store all parts of the model, including the data that was read from the DTF file. Although this makes the file independently portable, the file size can be large.



Figure 1. ‘Save All Data with Model’ option

If the user does not plan to move the DTF file from its current location, the above option can be unchecked before saving the mdl file. The mdl file will then store only a link to the DTF file instead of the entire DTF data, thus significantly reducing its size.
]]> http://www.caeda.com.cn/blog/article.asp?id=89 caeda@vip.sina.com(admin) Wed,01 Oct 2008 08:45:00 +0800 http://www.caeda.com.cn/blog/default.asp?id=89
Carpet Plots are used for visualization of computed data on 2D surfaces. A carpet plot is created by displacing the 2D surface vertically based on one of the variables computed over that surface. The carpet plot can also be colored by the same or different variable.

To create a carpet plot (see figure 1):

Create a plane cut (Z_cut shown in figure 1).
Click on the Carpet Plot icon - the control panel will switch to carpet plot controls. This panel asks for two inputs - Carpet Plot Elevation and Carpet Plot Color.



Figure 1. Carpet plot icon and controls


Using this tool, two figures (see figure 2) showing temperature and velocity magnitude distribution on a Z-cut can be replaced with one carpet plot as depicted in figure 3. In this example, the carpet plot was created over the Z-cut (a 2D surface). The inputs are also shown.



Figure 2. Velocity magnitude (left) and temperature (right) distribution





Figure 3. Carpet plot using temperature for elevation and velocity magnitude for color

With this type of plot, the regions of high and low temperatures are readily visible, as well as the strong gradient identified by the steep regions of the surface.


The Carpet plot menu (see figure 4) allows the user to change axes labels, the surface properties and take a snapshot of the plot.



Figure 4. Carpet plot menu


If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know.
]]> http://www.caeda.com.cn/blog/article.asp?id=88 caeda@vip.sina.com(admin) Mon,29 Sep 2008 08:31:21 +0800 http://www.caeda.com.cn/blog/default.asp?id=88
敬启者：
诚邀您参加马路科技顾问股份有限公司举办的德国TOSCA软件技术培训班，具体介绍如下：

培训目标
1. 为企业培养在特定工程涉及领域中具有独立设计能力，还具备现代管理知识和方法的国际化人才；
2. 此次培训，是为了让CAE工程师们初步了解结构优化设计在产品设计中的重要价值，如何在产品设计过程中采用全面和有效的优化方法，以降低整体产品开发过程的成本和时间。
3. 学员将能学习到基本拓扑知识、形状和条纹优化问题的构造和解决方法，优化模型求解过程以及采用Tosca前后处理软件如何对结果进行察看和评估。

日期：2008年10月18-19日
时间：上午9：00-下午5：00（两天）
地点：北京市朝阳区东三环南路58号富顿中心B座601（劲松桥东北角）

报名办法：请用EMAIL发至：Amily.zhang@rat.com.tw，电话：86-10-58672523
或发传真至：86-10-58672503，我们将迅速与你们商定。由于人数有限，如报名后而不能如期参加，请尽早通知，以免影响日后参加的机会。

参加对象：本次培训面向所有从事CAE设计分析专业人士，对于已在试用TOSCA软件的客户，及有具体产品合作设计意向的客户优先参加。参加者并不需要结构优化技术方面的相关经验，但须具有基本的有限元结构分析能力。在培训其间，请自带笔记本电脑，以便练习使用我们的软件产品。

培训内容：
•    结构优化设计原理和方法；
•    Tosca.Topology 拓扑优化方法；
•    Tosca.View优化结果可视化显示；
•    Tosca.Smooth优化结果平滑化；
•    Tosca.Report优化结果报表；
•    Tosca.Shape外形优化方法；
•    Tosca.Bead条纹优化；
•    Tosca 同用户CAE环境联接；
•    Tosca同CAE后处理软件的接口。
报 名 表
公司名称        联络人    
E-Mail        电话 / 手机    
地址        传真    
参加人员姓名    (请全部填写)
传真专线: 010-58672503 / 洽询电话: 010-58672523 分机622 张小姐
此培训由CAEDA公司协办，CAEDA公司邀请的客户，请直接和刘先生确认：13901064208, nsliu@vip.sina.com

]]> http://www.caeda.com.cn/blog/article.asp?id=87 caeda@vip.sina.com(admin) Mon,22 Sep 2008 13:34:40 +0800 http://www.caeda.com.cn/blog/default.asp?id=87
Laminate Tools is a powerful Windows application aimed at the professional composites designer. It addresses the entire Design-Analysis-Check-Manufacture group process of structural design, but focuses on the composite material features. It is a tool that links the various disciplines, communicates original data between departments, ensures proper understanding of the structure, offers in-depth checks and saves valuable team time.

"Laminate Tools was used for the design of the full composite ply layup, for the creation of FEA property cards and for all post-processing for the Speedy Racing Team Sebah Lola B08/80 LMP2 Coupe, a car that exceeded all expectations in early races during 2008."
        Paul Rennie, Senior Composites Engineer, Lola Cars International Ltd, UK

"The user-friendly interface of Laminate Tools has proved very useful in the creation of complex laminates. Moreover, the ply book export is an extremely useful feature for ply-by-ply check and lay-up visualization."
        Luca Furbatto, McLaren Racing, UK

"Certainly, on a cost / performance basis Laminate Tools would be my recommendation.  My experience with the code has been quite satisfactory.  Anaglyph has been very responsive in being able to meet my specific needs in the past in a very timely manner."
        Dan Molligan, McAps Inc., USA

Users import surface models from almost any CAD system and define composite materials, plies and layup using a state-of-the-art graphical user interface. Ply producibility can be evaluated immediately using proven draping simulation algorithms to identify potential manufacturing difficulties, meaning fewer change orders downstream. Data for finite element analysis purposes are generated seamlessly, ensuring that analysis models reflect the design intent. Results from the finite element analysis can be checked in detail using advanced composites failure criteria, and weak areas identified rapidly. Manufacturing data such as flat patterns are produced to ensure efficient use of manufacturing data. Finally, Laminate Tools can be used as an electronic ply book to communicate the complex structure of composite components and structures to clients, subcontractors and the shop floor.

Key benefits
It is focused on laminated composites.

It bridges the Design-Analysis-Manufacture domains.

It supports industry standard file formats, through the use of Simulayt's proven Layup Technology.

It offers powerful composites visualisation and becomes invaluable in simply examining a design and pinpointing troublesome areas where composite properties must be reviewed.
It is easy to use.

It offers an effective method to communicate information securely and up to date, both within and across organisations.

It can be used to view past designs for reference purposes, without the need for the original software that was used for the creation, and without running the risk of incurring accidental changes.

Unique features
Hierarchical tree view showing object inter-relations (select a node to show relations elsewhere in tree)

Interactive "select to interrogate" ability across three views (tree view / 3D graphics / text information)

Very easy interface for global Layup definition and modifications, with multi-level undo/redo

Log of Layup changes maintained on file, for quality control

Support for sub-Layups, for easier handling

Easy interrogation of local layup/laminate information (text + graphics), plus global view (thickness/solid/offsets/ply drop-off)

Composites-related model information, such as total material area and weight, or per ply
Directional Stiffness contour plot for entire model

ANSYS support (shell model data, laminated property creation and results post-processing)

Complete tabular element information in one window (geometry, layup, laminate, FEA results layer by layer)

Layer by layer graphical view of FEA results for individual elements (including interlaminar results)

User-defined failure criteria

Interactive calculation of stresses, strains, failure indices, reserve factors, margins of safety for instant estimation of modifications.

Easy FEA results post-processing for a single layer/loadcase/theory or for multiple layers/loadcases/theories together, or sorted by global ply.

Unique instant enveloping of FEA results across multiple loadcases

Unique summary of FEA results for all the layers of the worst loaded elements

Flat pattern export: extend boundary by given amount for later trimming

Flat pattern window for interactive adjustments, working in 2D and 3D views, and automatic generation/export of strips for each flat pattern.

Plybook export to HTML in multiple graphical views

Plybook export to Microsoft Word, with user-customised page layout and content


Plybook export laser projection data from layup file

Import/export STL interface for draping

Excel interfaces
File-open preview image for easy browsing
Stand-alone Windows environment
Floating (network) licence as standard
Comprehensive tutorial on CD]]> http://www.caeda.com.cn/blog/article.asp?id=86 caeda@vip.sina.com(admin) Wed,17 Sep 2008 08:03:32 +0800 http://www.caeda.com.cn/blog/default.asp?id=86
Did you know you can output summaries, such as mass flow summary, force summary etc. at your desired time interval independently of the DTF file save frequency? This option not only saves time but also memory and disk space. Using a “Special DTF Update” command, you can activate the summary output frequency at your choice of iteration interval (for steady state cases) or time steps (for transient cases).

This is a very useful tool, especially when you are running a large transient model (3D case with more than 1 million cells for example) and you need the output summary every time step. Indeed, if you have to save DTF files every time step, it will take time to write and save the DTF files and also fill up the hard disk drive very fast. So you would have to manually clean-up the DTF files or write a script for deleting them. Fortunately, you can avoid these issues by using the “special DTF update” command for summary output frequency.

The command syntax is:   SUMMARY_OUTPUT_FREQ int 1 Frequency
- int: means that the input for this command in an integer
- 1: means that the command SUMMARY_OUTPUT_FREQ requires exactly 1 input (the frequency in this case)
- Frequency: time step interval or frequency to write summary in output file


Consider a transient case running for 1000 time steps (see figure 1) where DTF files need to be written and saved every 100 time steps (see figure 2).



Figure 1. Input for number of time steps




Figure 2. Input for DTF write/save frequency


If you need to write the summary output every 10 time steps, simply write the command in the “Special DTF Update”, as shown in figure 3.



Figure 3. Special DTF update feature in CFD-ACE-GUI




Click OK and save the DTF file. Submit to solver.

Were the simulation steady, the summary output would be written every 10 iterations.




If you have any questions about this tip or would like us to discuss some other topic in the future, please let us know.
]]> http://www.caeda.com.cn/blog/article.asp?id=85 caeda@vip.sina.com(admin) Wed,10 Sep 2008 08:13:56 +0800 http://www.caeda.com.cn/blog/default.asp?id=85
Spatial differencing schemes control the spatial accuracy of the simulation. Several spatial differencing schemes are available in CFD-ACE+ to estimate the convective term in the transport equations. The Spatial tab under SC (Solver Control) allows the user to select a spatial differencing scheme as shown in Figure 1.



Figure 1. Spatial differencing options in the Solver Control → Spatial tab.

The default scheme is the first-order Upwind Differencing scheme. If higher-order schemes are selected, you can also enter a blending factor to blend the higher-order scheme with the Upwind Differencing scheme for added stability.

Given below is list of spatial differencing schemes available for flow:

Upwind (1st order)
Central (2nd order)
2nd order Upwind or 2nd order (2nd order)
2nd order Limiter (2nd order)
Van-Leer or 3rd order (3rd order) [only available for structured meshes]
Smart Scheme [only available for structured meshes]


Tips for selecting spatial differencing scheme for flow:

For steady state problems:
Use second order scheme for higher accuracy.
If you encounter stability problem with second order, but want a more accurate solution than first order, use second order scheme with non-zero blending factor, for instance 0.5, or you can try using smart scheme where the blending factor is not fixed but dynamically determined according to local flow conditions.
Be careful when doing steady state runs on oscillatory flows such as vortex shedding behind an automobile. Convergence with second order will probably be bad, but first order, despite giving a better convergence, is less accurate compared to taking the mean of a partially non-converged second order.
A better way for oscillatory flows is transient mode, which is necessary if one is to obtain the Strouhal number of the oscillation. But for a quick averaged solution in steady state mode, you can smooth out the oscillations by coarsening the grid in the wake region, this way second order can give much better convergence and may still be more accurate to first order with finer grid in the wake region.
Use second order scheme (central or 2nd order upwind) for simulations having rotating body or rotating boundary for steady state and transient problems.
For time-accurate problems like LES, DES, use central scheme or second order upwind.
For shock capturing problems, use third order scheme (van-Leer).


If you have any questions about this tip or would like us to discuss some other topic in the future, please let us know.
]]> http://www.caeda.com.cn/blog/article.asp?id=84 caeda@vip.sina.com(admin) Tue,09 Sep 2008 08:28:58 +0800 http://www.caeda.com.cn/blog/default.asp?id=84 一个基于IOSO，新一代优化技术的新的优化策略

“由于宇宙的构造是最完美的，最明智的创作者的工作，都是
发生在并不存在最大最小规则的宇宙中的。“Leonard EULER


Ioso PM
多重标准的并行优化软件包

IosoPM 是一个可以显著缩短解决实际问题所需的时间的工具，
也可以用于在工业，工程， 科学和其他领域解决“额外大“的问题

可以：
•最大限度地利用多处理机系统和局域网的潜力
•有效地利用难以并行化的应用软件和计算模型
•为目前被认为无法实现的复杂问题提供解决方案

特殊功能：
借助于现代多处理器系统，在今天解决的问题巨大的多样性中，
号称“大”甚至“超大”问题扮演了重要角色。
这种问题的特点其解决需要巨大的时间和金钱投入。
然而，此种问题中用到的计算模型目前为止不能总是能如此有效的并行。
这是因为许多算法在并行化方面有困难或者一些计算模型的内部可并行分支
的数量低于可用于计算的处理器的数量。

这种情况下，通过使用并行优化技术，
在不需要对计算模型做出修改的情况下实现重要的作用。
并行优化意味着有多少用于计算的处理器，
计算模型就会有多少次带有不同的输入变量被执行。
计算结果表明，对于大多数问题，整体方案的解决时间
会依据所使用的并行处理器的数量成比例的降低。

应用实例
改善三段风扇在巡航模式下的空气动力特性

目的
确定六叶片风扇的几何，以便在巡航模式下
达到最大空气动力学性能，同时保持设计模式下相同的（或者更高的）效率以及
相同的空气消耗和持续操作时间。

•三维CFD程序用来求解气动特性（需要21小时进行一次迭代）
•三段风扇的叶片几何形状通过65个独立的变量来描述

结果
总迭代次数为340。
风扇空气动力效率增加1.2%。
采用12个计算节点解决并行优化的任务。
同传统优化方法相比，时间消耗节省了超过90%。

集成
IOSO PM与多种不同的CAE/CFD/FEM应用软件相兼容

用户
Ioso nm已成功地为众多俄罗斯和遍布世界各地的行业领先公司和工程中心解决实际任务。

会议参与
ioso技术的高效性被领先的西方专家所认可。
涉及其使用的工程应用论文已经发表AIAA， ASME ，
eurogen，eccomas等著名国际会议上。

没有相似的此类问题的解决被发表。
]]> http://www.caeda.com.cn/blog/article.asp?id=83 caeda@vip.sina.com(admin) Mon,08 Sep 2008 09:57:38 +0800 http://www.caeda.com.cn/blog/default.asp?id=83
LAP is a tool strictly for the analysis of composite laminates of infinite length and width. It lets the user apply various types of hygrothermal and mechanical loading on the laminate, and it gives in-depth information about effective stiffness and coefficients, as well as through-thickness stresses and strains. It can also handle non-linear stiffness properties, offers laminate design functionality and calculates notched compressive strength.

CoDA has 7 independent modules:
- the Panels module analyses composite panels of fixed dimensions and subject to boundary conditions, under out-of-plane loads.
- the Beams module analyses composite beams.
- the Laminates module calculates the effective stiffness and in-plane strength of a laminate. No through-thickness stresses or strains are given.
- the Joints module analyses composite double-lap bolted or bonded joints.
- the Synthesiser allows for the calculation of composite properties based on those of the constituent fibre and resin.
- the Flanges module analyses curved beams.
- the PREDICT module analyses the crack formation and propagation in laminate layers, but also synthesises composite layer properties.
As shown, the only common area between the two packages is in the calculation of effective in-plane stiffness and strength for laminates.
Think of LAP and CoDA as complimentary. They have been designed to work stand-alone, but also together. They can read each other's data files, and data objects can even be dragged and dropped between them.

Can the Laminate Tools modules work independently?
The View module is necessary for the product to function. The other modules are optional and independent of each other.  

Can the LAP modules work independently?
The Basic module is necessary for the product to function. The other modules are optional and independent of each other.  

Can the CoDA modules work independently?
Each CoDA module is independent and does not require input from other modules. Of course, it could be desirable to synthesise material properties (Synthesiser module) and use those properties in other modules. or you could solve for the effective properties of a laminate (Laminates module) and then treat those as a material in the Beams module. But none of this interaction is necessary for the operation of any individual module. If you wish to use other software to generate material and laminate mechanical properties then you can input those directly into CoDA for use with the Panels, Beams or Joints modules.

Is CoDA suitable for conventional materials (e.g. metals)?
Yes, the Panels, Beams, Joints and Flanges modules are perfectly suitable for use with conventional materials.

How can I analyse sandwich configurations?
Both LAP and CoDA can analyse sandwich configurations. The CoDA Panels and Beams modules include separate analyses for sandwich configurations, also covering skin buckling modes. Elsewhere, the skin and core layer properties are simply defined as orthotropic material layers. Transverse shear stress deflections are taken into account only for CoDA beams.

How can I analyse woven fabrics?
Both LAP and CoDA can use fabric layers in laminates if the fabric materials are defined as simple orthotropic materials.

How do I order your software?
To place your order, contact us for a firm quotation. The most popular payment method is by raising a Corporate Purchase order for payment by 30-day invoice. This is normally accepted, subject to approval.
Credit card payment may also be possible depending on products ordered, please call us on +44-20-89876056.
orders are accepted from all countries. Agents may be available locally. Quotations, invoicing and payment in currencies other than pound sterling can be arranged on request.

What are the delivery charges?
Delivery charges vary depending on destination and value of order, please enquire.

What are the terms of your Software User Licence?
The text for the user licence accompanying each product is available on request. In brief, you are allowed to use the software simultaneously on as many computers as the number of licences you have purchased. Site licences do not have this restriction within one site. The user licence is normally indefinite. Optional maintenance for receiving regular version updates is available, please enquire.

Can I install the software on a network?
Yes, the software can be installed on a network for easier access. The number of simultaneous users cannot exceed the number of licences purchased.

I use a previous version of your software, why should I upgrade?
There are significant updates between versions, which are indicated in the detailed description page for each product. There are also significant discounts for upgrades.

I employ custom failure criteria, do you support that?
The LAP Additional Failure Criteria module as well as the Laminate Tools Check module offer this functionality. In addition, we can carry out any kind of customisation for you, including failure criteria, specific results, application interfaces, etc. Just let us know what you need and we'll come back with the details.


]]> http://www.caeda.com.cn/blog/article.asp?id=82 caeda@vip.sina.com(admin) Wed,03 Sep 2008 09:02:09 +0800 http://www.caeda.com.cn/blog/default.asp?id=82
A CFD simulation process starts from an accurate representation of the boundaries that usually originates directly from CAD systems. STL and IGES are two of the most common output formats used as a starting point for mesh generation. STL (StereoLithography) files represent 3D surface geometries using a triangular mesh allowing unambiguous transfer of files from one system to another. Each patch in a STL file is defined by three points and an orientation vector.

CFD-GEOM, an integrated geometry modeling and grid generation tool, supports the export of geometries and meshes in several different output formats. Traditionally,CFD-GEOM users have been able to write out ASCII STL files with consistent orientation from triangulated surface meshes. From V2008.2, CFD-GEOM also allows export of quad surface meshes (generated as structured or unstructured) into STL files. Quad facets are automatically split into triangles as per STL requirements. For structured meshes, only 2D blocks (not faces) can be exported.



Figure 1. 'Save STL files' option from File menu

These files can be opened later in any tool supporting STL files such as CFD-VisCART and CFD-VIEW for mesh generation or visualization purposes.



Figure 2. Examples of exported STLs from triangular and quad surface meshes.

The exported STL files can also be imported into CFD-GEOM as a discrete surface for further manipulations. A newly implemented discrete modeling engine in CFD-GEOM can import and manipulate meshed surfaces from several external grid formats. Some of the available tools include:

Creation of discrete surfaces from lines and CAD surfaces
Extracting outlines and critical features
Split at outlines and critical features
Boolean operations on closed surfaces


Figure 3. Discrete surface operations showing union of different closed surfaces, intersection with a plane, and splitting using outlines and critical features.

In future releases, CFD-GEOM users could expect more powerful features including re-meshing of imported discrete surfaces and tet mesh generation from re-meshed discrete surfaces.

If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know.
]]> http://www.caeda.com.cn/blog/article.asp?id=81 caeda@vip.sina.com(admin) Tue,26 Aug 2008 08:47:00 +0800 http://www.caeda.com.cn/blog/default.asp?id=81
Turbulent flow can be hard to stabilize during the first iterations. Fortunately, a solution control feature exists in CFD-ACE+ and CADalyzer that can aid the convergence of turbulent cases. Consider the case study presented in figure 1.



Figure 1: Example case study



This is a small circular jet entering into a rectangular shaped duct. The inlet conditions for the duct are 30.0 m/s with a turbulence intensity of 0.03 and for the jet are 71.263 m/s with a turbulence intensity of 0.05. The initial conditions for this case are 30.0 m/s for U velocity, 0.1 for turbulence intensity, and 0.203 for the dissipation rate. Using default values for relaxation the simulation will actually diverge due to poor initial conditions. However, using the new Turbulent Start Control feature will allow the case to converge. The residual plots from each of the two cases are shown in figures 2 and 3.



Figure 2: Residuals without turbulence start-up option





Figure 3: Residuals with turbulence start-up option



So what is this option doing? It is giving the user the ability to ramp the values of turbulence calculated by the code. Sometimes during start-up the effective viscosity computed from the turbulence model can produce values which cause problems with convergence. This new method holds the value of the effective viscosity to a more reasonable value for a given number of iterations so that the flow field can start to develop. Once this occurs, the turbulence model will start to produce more reasonable values and a more accurate effective viscosity. There are three inputs for this feature (see figure 4):



Figure 4: Turbulence start control option in CFD-ACE-GUI



Viscosity Ratio: Ratio of effective viscosity to laminar viscosity.
Initial Iterations: The effective viscosity is then used over an initial number of iterations (specified in this second field) instead of the values calculated from the turbulence model.
Transition Iterations: This gives the user the ability to linearly ramp in the values calculated from the turbulence model. For instance, if 20 is used, then at the first transition iteration it will use 95% of the value calculated from the viscosity ratio and 5% calculated from the turbulence model. However if zero is used, then at the end of the Initial Iterations the values from the turbulence model will immediately be used.

Consider using this feature if you have problems getting a turbulent case to start-up without diverging.

If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know.
]]> http://www.caeda.com.cn/blog/article.asp?id=80 caeda@vip.sina.com(admin) Wed,20 Aug 2008 08:53:17 +0800 http://www.caeda.com.cn/blog/default.asp?id=80
As simulation file sizes get bigger and bigger, we are always looking for ways to decrease the amount of data stored. A new option has been added to CFD-ACE+ that allows you to NOT save the restart data to the DTF file. This can reduce file size by as much as 40% for some large models.

Basically, the DTF file has two sets of solution data: one is for visualization (what you see in CFD-VIEW) and one for restart purposes (allowing you to restart the simulation to run further or restart it with different parameters).

Many times, a simulation will be run only one time to convergence, and will never be used as a restart point. Up until now, both the visualization and restart data were written into DTF files, which can make them unnecessarily large.

A brand new option, now available in CFD-ACE+ version 2008.2.9, is the ability to choose whether or not the restart data will be written to the DTF file. This option can be found in the 'OUT' panel, under the 'Ouput' tab, as shown in figure 1.



Figure 1: Restart data check box in CFD-ACE-GUI

By default, the "Write Restart Solution Data" option is activated, meaning that the restart data WILL be written to the DTF file. Uncheck it and the restart data will not be written to the DTF file, thus making it much smaller, up to 40% size reduction. Be careful though, as you will not be able to use the file as initial conditions for steady or unsteady simulations.

If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know.
]]> http://www.caeda.com.cn/blog/article.asp?id=79 caeda@vip.sina.com(admin) Sun,17 Aug 2008 09:04:36 +0800 http://www.caeda.com.cn/blog/default.asp?id=79 Press Release


FE-DESIGN releases Version 6.2 of the Optimization Software TOSCA Structure

Strongly enhanced functions of the software solution for efficient structural optimization

Karlsruhe, 15. August 2008

FE-DESIGN is enhancing the reliable approach of solver independent structural optimization in combination with the commercial finite element solvers ABAQUS, ANSYS, MSC.Nastran, MSC.Marc, NX.Nastran and PERMAS in the area if nonlinear finite element analysis.

In Topology Optimization, several non-linear constitutive materials are allowed to be a part of the design domain. This may from now on be combined with the definition of complex objectives and constraints. In Version 6.2, acoustic measures may be used for topology optimization. For the formulation of objectives and constraints, the surface velocities as well as the pressure level may be used. The support of internal forces allows the application of contact forces or forces of weld elements in the objective or constraints. In the area of manufacturing restrictions, the rib distance was implemented for the optimization of reinforcement ribs for casting parts.

Also in Shape Optimization, enhancements in the area of non-linear analysis have been made. In TOSCA Structure Version 6.2, the plastic strains are allowed to be used as design response for the objective formulation. For the manufacturing constraints, a new sliding constraint is implemented that allows the movement of the surface nodes on arbitrary free form surfaces.

In Bead Optimization, the proven sensitivity based algorithm is implemented that allows the definition of complex optimization tasks for the optimization of sheet metal components. In Version 6.2 compliance, displacements, modal frequencies and results from frequency response analysis may be used for bead optimization.

The smoothing algorithms in TOSCA Structure.smooth were redesigned and the detection and avoidance of self intersections was added. The definition of the smoothing task may now be added to the parameter file of TOSCA and therefore be executed automatically at the end of the optimization run.

The actual TOSCA Structure Version 6.2 was presented at the “2nd Conference for applied Optimization in the virtual Product development Process“ in Karlsruhe.

About FE-Design:
FE-DESIGN is a worldwide leading provider of consultancy services, solutions and software products for structure and process optimization tasks in the field of numerical simulations. FE-DESIGN provides competitive advantage to its customers by improving their products, accelerating innovation, reducing development cycles and by minimizing prototypes and hardware tests.

FE-DESIGN develops and distributes the structural optimization software TOSCA Structure with modules for topology, shape and bead optimization utilized in numerous R&D departments of successful operating companies around the world.
With OPTIMUS, one of the most sophisticated software for parameter-driven design optimization can be offered, which is distributed by FE-DESIGN in Germany, Austria and Switzerland. OPTIMUS is an ideal tool for decisive IT- and engineering teams for process integration and for performing reliability and robustness designs in all areas of numerical simulation.

As a full service optimization supplier, FE-DESIGN covers the entire range of optimization services, e.g.: analysis of single components with TOSCA Structure, all-comprising execution of project developments, integration of TOSCA Structure and OPTIMUS into existing design and analysis environments as well as onsite training and coaching of designers, engineers and technical executive management at your location.

FE-DESIGN is a partner for numerous leading companies working on the field of automotive, aerospace, mechanical engineering, drive technology, electronics, medical engineering and heavy industry.

FE-DESIGN has offices in Karlsruhe, Munich, Husum (Germany) and Copenhagen (Denmark) and is represented by authorized distribution partners around the world.

For more current information please visit our website: www.fe-design.com

Yvonne Schilling
Marketing
FE-DESIGN GmbH


]]> http://www.caeda.com.cn/blog/article.asp?id=78 caeda@vip.sina.com(admin) Tue,12 Aug 2008 09:46:58 +0800 http://www.caeda.com.cn/blog/default.asp?id=78
Here some highlights:

Topology Optimization:
- Internal forces to be used as design response for objectives and constraints
- Design Responses from acoustics and frequency response analysis
- Definition of multiple volume constraints possible
- Support of geometric nonlinear analysis and contact in the design area for ABAQUS and ANSYS
- Support of a number of nonlinear materials in combination with ABAQUS
- Soft Delete
- Definition of rib distances for wall thickness constraints (maximum membersize).

Shape Optimization
- Support of plastic strains as design response
- Check Slide for manufacturing
- Support of CPRESS for the contact surface shape optimization with ABAQUS

Bead Optimization
- New Sensitivity Based Algorithm allows new responses for Bead Optimization: DISPLACEMENTS, ROTATIONS, STRAIN ENERGY, FREQUENCY RESPONSE / ACOUSTICS, EIGENFREQUENCIES

For more detailed information please have a look at the attached release notes.

The TOSCA Structure Release 6.2 will run with your existing Licenses.

With TOSCA 6.2, a new structure of Advanced licensing is implemented. The existing ADVANCED License is divided in 2 new Licenses (ADVANCED LIFE and ADVANCED NONLINEAR).
All Customers who have purchased/leased TOSCA or are under maintenance with the Advanced License (up to TOSCA Version 6.1.1) will get the ADVANCED LIFE and ADVANCED NONLINEAR license features in future. All new customers have to pay for each module separately.
Additionally the ADVANCED ACOUSTICS License feature allows to use Design Responses from Acoustic and Frequency Response analysis for optimization.
Enclosed you can find the new international price list.
Advanced life
        Life Interface for Shape Optimization.
Support of DAMAGE and DAMAGE_LC Design Responses
(ONF Datablocks 600 and 601)
        Support of Temperature Analysis
TEMPERATURE_FILE in FEM_INPUT

Advanced nonlinear
        Optimizations based on nonlinear results from the used solver (independent
        to the optimization type)
        Optimization of contact surfaces to homogenize the contact stress for shape optimization
        Shape optimization based upon a geometric non-linear analysis
        Shape optimization using non-linear materials
        Topology optimization based upon a geometric non-linear analysis
        Topology optimization using non-linear materials
        Support of Temperature Analysis
TEMPERATURE_FILE in FEM_INPUT

Advanced acoustics
        Support of Acoustic and Frequency Response Design Responses
TYPE = FS_*


For further questions please contact caeda@vip.sina.com
        
If you already have a download account for TOSCA Structure at our website, please log in at        
http://www.fe-design.de/en/login.html with your personal password. On the left side in the menu you can get access under TOSCA Structure Software to the download area for TOSCA Structure 6.2. If you do not have your password available, please request it to be sent to your email address again.        

To set up a new account, please register on our website http://www.fe-design.de/en/registration.html.
After having received the registration form we will activate your account immediately and send an email confirmation to you.

If you need a DVD for installation please let us know and inform us about the quantity and the shipping address.
]]> http://www.caeda.com.cn/blog/article.asp?id=77 caeda@vip.sina.com(admin) Tue,12 Aug 2008 09:42:28 +0800 http://www.caeda.com.cn/blog/default.asp?id=77
One of the most common questions that occurs during CFD-GEOM Training sessions is "How do I generate mixed structured and unstructured grids?" Both grid types have their specific advantages and disadvantages, e.g. structured grids are very efficient in resolving the viscous terms, whereas unstructured grids can be used to handle complex geometries.

This example demonstrates complex geometry grids for a 2-dimensional model simulating a microfluidic DNA chamber containing five pillars. Using the following steps, node to node matching at the interface between the structured and unstructured domain is achieved, and a grid such as the one shown below can be created.



1. Create the Geometry Out of Points, Lines and Curves:
Your computational area of interest is now shown as a wire frame.



2. Create the Structured Domain:
Create edges on the lines/curves with the desired number of points and point distribution. Once edge creation is complete, create a structured face and 2D block for the channel.



3. Create Trimmed Surfaces:
Go to Geometry --> Trimmed Surface Creation Options --> Fit Though Curves.




Create the trimmed surface by selecting the outer lines and edge (i.e., do not select the circles).

A trimmed surface can be created from lines, curves, or edges. In order to get point to point matching at the interface between the structured and unstructured domain, make sure that the edge is selected instead of the curve at the interface.




Since the pillars interior will not be meshed, we need to cut the surface so that it does not cover them.
Go to Geometry --> Trimmed Surface Modification Options --> Cut Hole(s).



Select the surface, and then select the lines of the five circles. Click Apply.



TIP: Using edges for trimmed surface creation is a very handy way to control the grid distribution of an unstructured mesh in addition to corner point sources and surface interior sources.

4. Create the Unstructured Mesh.
Once the correct trimmed surface is created you can use the surface meshing tool to generate the triangular unstructured mesh.



The 2D grid shown here could also be extruded to create a 3D system made of a 3D structured block and a 3D prism domain.

If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know
]]> http://www.caeda.com.cn/blog/article.asp?id=76 caeda@vip.sina.com(admin) Sat,02 Aug 2008 08:19:47 +0800 http://www.caeda.com.cn/blog/default.asp?id=76
The Plotter operator in CFD-VIEW supports signal processing for Time History data. A previous user tip – Digital Signal Processing using CFD-VIEW – shows how time history date can be made available and how to perform a PSD on a periodic signal. It is recommended to review this user tip before going further.The aim of this new user tip is to list all options supported by the Plotter Operator.

This tip will discuss the following operations:
- Correlation
- Fast Fourier Transform
- Discrete Fourier Transform
- Integral
- Derivative
- Power Spectral Density
- Smoothing
- Filter
- Sound Pressure Level
- Window


If you are interested in these functions, please read on.



1. Correlation
This operation performs an auto or inter-correlation, depending on whether one or two entities are currently selected (no more than two entities are authorized). The user has to define two parameters, both in number of sampling points:
- Lag length: the length over which the correlation has to be estimated (by default, half the number of curve points).
- Samples number: allows the user to truncate the curves given the number of abscissa points. The default value is the number of points in the curve.

Inter-correlation assumes that both curves have the same number of points.

The result is a curve {Ri, i=1,n_sample}, where n_sample is the value of Samples number, and Ri is defined as:



Where:
- x and y are the point-wise values of the single curve (auto-correlation) or of each of the two curves (inter-correlation),
- N is the lag length,
- <x>M1,M2 is the average of x over the sampling points ranking from M1 to M2




top

2. Fast Fourier Transform (FFT)
The algorithm used is the base 2 FFT, which performs a temporal permutation at the start of the computation. The input signal step must be quasi constant and increasing. Otherwise the transformation cannot be performed.

If the signal has not a number of points of the form 2n, it is re-sampled by linear interpolation. In any case (with or without re-sampling) the FFT result is scaled by the signal time step Δt.

An option is available to convert the FFT magnitude in dB:



The phase, the imaginary and the real part of the FFT is also provided.


top

3. Discrete Fourier Transform (DFT)
This option is applied on the selected data via the list Entities. It computes the Discrete Fourier Transform and normalizes it by 1/Fe (as with the FFT). The DFT has the same output options as the FFT (Magnitude ...) and performs (n ´ N) operations (N is the number of frequencies and n is the number of samples). Its algorithm is not optimized as the FFT (n x ln(n) operations).

The user input are the minimum, maximum frequencies and the frequency step, so that the transform is computed with a limited required frequency bandwidth and a frequency resolution (therefore, this can reduce the time computation).

top

4. Integral
Select one entity in the list of external entities, corresponding to {y1i, i=1,n}. Choose the Integral item in the Operation list, then Create at the bottom of the window: this will create a new curve defined on the set of mid-points of the original set of abscissa, defined as:




where i =1,n-1.

The integral curve is defined on n-1 points only.

top

5. Derivative
Select one entity in the list of external entities, corresponding to {y1i, i=1,n}. Choose the Derivative item in the Operation list, then Create at the bottom of the window: this will create a new curve defined on the set of mid-points of the original set of abscissa, defined as {y2i = (y1i+1 - y1i-1) / (x1i+1 - x1i-1) , i=2,n-1}.

The derived curve is defined on n-2 points only.

top

6. Power Spectral Density (PSD)
The PSD is computed by the previously described FFT so that the signal step must be constant and increasing. If the number of signal points is not of the form 2n, it is re-sampled by linear interpolation.

The PSD is the power density per bandwidth of frequency:



Where N is the number of signal points, Dt is the signal time step, and T is the total time of the signal.



The result is then expressed in (Physical Unit)2/Hz.

The value of PSD magnitude in dB is given by



REF_PRESSURE is equal to 2.0e-5

top

7. Smoothing
Select 1 entity in the list of external entities, corresponding to {y1i, i=1,n}. Choose the Smooth item in the Operation list, then Create at the bottom of the window: this will create a new curve defined on the set of n-2 "interior" original X-points, defined as {y2i= (y1i-1 + y1i + y1i+1) / 3, i = 2,n-1}.

The smoothed curve is defined on n-2 points only.

top

8. Filter
The Window frequency filter is used to block some frequency and pass others. For example, used as a low pass filter, it can erase the numerical noise which can appear in numerical computed results.

Two user defined frequency gives the window parameters. By default, the band pass is [0,υmax]; these parameters give back the original signal.

Time signal process:
The frequency product is performed: X(υ)×H(υ). H(υ)= if υ1 < υ < υ2, else H(υ)=0 and X(υ) is the FFT of the input signal. The Inverse Fast Fourier Transform is then directly applied on the frequency product to give the time filtered signal. The IFFT is performed by the same algorithm than the FFT.

top

9. Sound Pressure Level (SPL)
The Sound Pressure Level is computed by the formula:



where the reference acoustic pressure is 2.10-5 Pa, Sυ is the power spectral density, and [υ1,υ2] is the bandwidth frequencies expressed in octave, 1/3 octave, or user defined band in Hz.

The power spectral density is computed as described previously.

In each band j, the integral



is approximated by the trapezoid method. The SPL is then computed by:



Acoustic filters can be applied to the result in order to obtain SPL A, B, or C.

top

10. Window
The windowing operation is defined by its type and its size. The size is specified by the user through the Lower abscissa and the Upper abscissa input fields, whose default value are the min & max of the abscissa of each entity, separately. Windowing can be used, for instance, to truncate irrelevant parts of a signal before its Fourier analysis.

The size of the window identifies a list of N abscissa index k, where the weight Wk applied to the point-wise signal value will be different from 0. This weighting function Wk applied over the window size depends on the window type, and is defined from the literature as follows:

- Rectangular: Wk = 1

- Hamming: Wk = 0.54 - 0.46*cos(2πk/(N-1))

- Hanning: Wk = 0.50 - 0.50*cos(2πk/(N-1))

- Blackman: Wk = 0.42 - 0.50*cos(2πk/(N-1)) + 0.08*cos(4πk/(N-1))

top

References
CFD-VIEW 2008.2 User Manual, Chapter Plotter Operation
Digital Signal Processing with CFD-VIEW, User Tip # 735, ESI-CFD Portal



For more information on these features, please refer to the CFD-VIEW user manual (available through the WebHelp).

If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know.

]]> http://www.caeda.com.cn/blog/article.asp?id=75 caeda@vip.sina.com(admin) Fri,25 Jul 2008 07:48:21 +0800 http://www.caeda.com.cn/blog/default.asp?id=75
Many users have asked for a method to obtain a solution on a coarse grid and then use it as the initial guess for a finer grid. This option can be done using the Map utility in CFD-Toolkit.

The steps to perform this operation are as follows:

Create coarse grid system: The first step is to create the coarse grid system of the geometry in CFD-GEOM. In this example the following coarse grid is used. Save the coarse grid system as a DTF file, such as coarse_grid.DTF.


Set up and run the problem: Open the coarse_grid.DTF file in CFD-ACE-GUI and set up the problem. Once the setup is complete, submit it to the solver to be run.

Copy the Simulation: Once the simulation has completed running, copy the coarse_grid.DTF file into another file, such as fine_grid.DTF. The fine_grid.DTF has all the settings that the coarse_grid.DTF has, but still has the coarse grid system.

Create the Fine Grid: Go to CFD-GEOM and open the coarse_grid.GGD file if it is not already open. Create the fine grid now by changing the number of grid points. Only the number of grid points are changed and no modifications are made to the geometry shape. Also be careful not to delete or recreate any entity. Save the DTF file as fine_grid.DTF by selecting the Update Grids Only option. Now fine_grid.DTF consists of the finer grid system but retains all the model setup options from the coarse grid simulation.
  
Map Solution Data: The solution data in the fine_grid.DTF file is still from the coarse grid solution and therefore not consistent with the fine grid system. To map the solution data from the coarse grid DTF file to the fine grid DTF file, we can use CFD-Toolkit. The first thing to do is launch CFD-Toolkit. (Note: On Windows systems, choose Start-->Run-->CFD-Toolkit.) Go to the File View/Manipulate --> Map DTF File. Once it has been selected, the following window will be active.



The source file will be coarse_grid.DTF, which contains the solution data for the coarse grid simulation. The destination file will be fine_grid.DTF, which consists of the same settings as coarse_grid.DTF. Once these files have been input, select the Zone Data option, which will map the solution data from one file to another. There are three options for Zone Data: Node, Cell, Face. To map the data, select all of the options (Node, Cell, and Face). Click Map to perform the mapping.  
Run the Fine Grid Model: The last step is to run the fine_grid.DTF file. Open the file in CFD-ACE-GUI. All the settings used to run the coarse_grid.DTF file will be retained. The only modification needed for the model setup is the initial condition. Go to the IC tab (shown right) and select Previous Solution. The restart filename will be fine_grid.DTF, since the solution data was mapped onto fine_grid.DTF. The final step is to submit it to the solver.  


Once the run is complete, you can import both of the files into CFD-VIEW and analyze the results. Below are some results obtained to demonstrate the advantage of using a finer grid. The results on the left are from coarse_grid.DTF and the results on the right are from fine_grid.DTF. As you can see, a much more accurate solution is obtained using the finer grid.



coarse_grid.DTF fine_grid.DTF

The example above showed specifically how to use the CFD-Toolkit Map File function to interpolate solution data from a coarse grid to a fine grid for restart purposes. You can also use the CFD-Toolkit Map File function to map Simulation Data, BC Data, or VC Data between any two DTF files. For example, you could have two totally different grid topologies but you know that all the Sim Data should be the same (Sim Data are items like problem type, model options, and solver control settings). Likewise you could have very complex VC settings that you want to copy from one DTF file to another. When mapping BC or VC data, CFD-Toolkit lets you choose the criteria that define a "match" (e.g., name only, name and type, etc
]]> http://www.caeda.com.cn/blog/article.asp?id=74 caeda@vip.sina.com(admin) Thu,24 Jul 2008 08:09:47 +0800 http://www.caeda.com.cn/blog/default.asp?id=74
CFdesign v10是CFdesign软件历史上的一次大版本升级。 V10使“Upfront CFD”的技术革新又向前迈进了一大步。它兼备现代化环境的实用性及众多出色的全新功能，成功地向设计师和工程师呈现精确到位的模拟工作流，是一套与众不同的 CFD 解决方案。

CFdesign v10 结合了具有前瞻性的创新技术和现今市场流行的解决方案。其中，可模拟空化和热电制冷器的各种新增功能，以及快速执行概念化设计迭代的“快速对流”模式，都是 v10 软件包的主要元素。 新增加的一套几何图形工具大大简化了制作复杂的三维装配几何的过程。CAD 组件 AutoPrep 已直接加到 CAD 环境中，模拟部件的几何创建功能也出现在 CFdesign 环境里。而这两者对于诊断、修复和准备用于分析的设计图形均十分关键。 通过在 CFdesign 环境中保持 CAD 主要元素的可用性，进一步深化了 CAD 与 CFdesign 模拟环境的联系。改进的用户界面使 CFdesign 工作流程更易掌握，更加有。            
    
“CFdesign 分布式计算”概念的引入，让多方案设计同时进行成为可能。随着全球市场压力的不断增大，快速改善模拟结果对于创新、有效的设计至关重要。“CFdesign 分布式计算”可让使用“前端 CFD”技术的产品开发组织充分利用先进的高速网络计算环境，在激烈的竞争中脱颖而出。

请点击我们的官方网站以获取有关新版本的更多信息： caeda@vip.sina.com]]> http://www.caeda.com.cn/blog/article.asp?id=73 caeda@vip.sina.com(lns) Wed,23 Jul 2008 10:05:01 +0800 http://www.caeda.com.cn/blog/default.asp?id=73
Microsoft作为软件行业的领军企业，长期以来致力于帮助各类用户解决在日常办公和科研/生产/管理中的遇到的计算机应用问题，开发和发展Microsoft的HPC软件也是为了更好地服务于用户，让更多的用户透过微软提供的简单易用的高性能计算平台告别单机处理计算的时代，采用多机并行的方式来大大提高工作效率，缩短科研生产周期，进而提升企业的竞争力。

Microsoft的高性能计算软件WCCS是建构在PC集群基础上的，除操作系统外，还集成了功能完善的集群系统管理、监控、MSMPI、作业调度、日志报告等功能，与传统的Linux集群系统相比较，更容易部署和管理。WCCS 2003的基本组件有两张CD：第一张 CD 包含 Windows Server 2003 Compute Cluster Edition， 第二张CD是Microsoft Compute Cluster Pack — 构成 Windows Compute Cluster Server 2003的界面、实用工具和管理基础架构的组合。

Microsoft WCCS 2003 提供了一个运行在商用 x64 计算机集群上的极经济的强大 HPC 解决方案，可使用熟悉的工具和技术，轻松、快速地对其进行扩展和部署。其基本体系架构描述如下：包含单个头节点（也叫主控节点）和多个计算节点（参见图 1）。头节点控制和协调所有对集群资源的访问，实现计算集群的单一管理、部署和作业调度。WCCS 2003 使用 Microsoft Operations Manager 2005 和 Microsoft Systems Management Server 2003 等工具，将现有的公司 Active Directory 基础结构用于安全、帐户管理和总体操作管理。

图 1. 典型的 WCCS 2003 网络

WCCS 2003的安装包括在头节点上安装操作系统，将头节点加入现有的 Active Directory 域，然后安装 Compute Cluster Pack。当 Compute Cluster Pack 安装完成后，将会弹出一个“任务列表”页面，其中显示完成计算集群配置所需执行的步骤。这些步骤中包括定义网络拓扑，使用“配置 RIS”向导配置 RIS，将计算节点添加到集群，以及配置集群用户和管理员等。 通过安装向导，可以指导用户方便地一步一步完成整个集群系统的部署。

微软的WCCS 2003 虽然涉足高性能计算时间较短，但其继承了微软视窗系统的图形化的管理风格，具有部署简单，管理容易，功能齐全，能有效地与企业已有Windows架构的IT系统方便集成，并且有更加简单熟悉的Visual Studio集成开发环境帮助技术人员快速进行并行程序开发设计。因为门槛低，用户可以快速上手，因此微软的WCCS 2003 越来越广泛地应用在工程设计/科学计算/图形图像/金融分析等高性能计算领域。
具体来讲，WCCS的优势集中体现在如下几个方面：

灵活的扩展性可充分保护用户的投资

在信息爆炸的时代，业务量的增长通常是难以预测的和巨大的，往往要求业务处理系统具有强大的扩展能力，以满足这种高速发展的要求。该采用并行计算集群系统的架构方便用于按需所取地扩充，当用户发现计算机系统的处理能力不够用时，可以方便地通过增加计算节点的方式来提升整个系统的处理能力，而不需要从头购买新的设备，从而充分保护用户的投资。采用WCCS 2003 可以方便地增加新节点，不需要停掉系统，从而提高系统的使用率。

方便的集群管理和监控

计算集群系统相对传统的小型机具有更复杂的架构，系统的统一管理和调度变得尤为重要，为了提高系统管理的效率、管理的安全性，其必须有便于使用的管理工具。事实上，一套集群系统是否易于管理已经是判断该系统是否好用的标准之一。WCCS整合的管理和监控软件不仅能大大方便系统的管理和维护，还提供诊断和报告机制以利于系统故障时的快速判断。

简单易用的并行平台

在计算集群系统中需要有提供消息传递接口的MPI，作业调度软件和基本线性代数库函数、并行库函数和通信库函数等等。WCCS提供的整合的一体化并行平台可以大大方便和简化用户的使用。

WCCS 2003 中包含的MS-MPI是 Argonne 国家实验室开放源代码 MPI2 实现的一个版本（现有的 HPC 集群广泛使用了该实现）。MS-MPI 与 MPICH2 参考实现和其它 MPI 实现兼容，并支持一个具有 160 多个函数调用的全功能 API。WCCS 2003同时包含作业调度器和 Compute Cluster Manager，使用户可以调度作业，为作业分配所需的资源，以及更改与作业关联的任务和属性。

较高的性能价格比

微软WCCS高性能解决方案充分采用现有的通用硬件和商用软件的成果，与提供同等处理能力的小型机相比，最大限度的降低花费在系统建设和系统维护上的开销。由于其不仅提供了易用的Windows Server操作系统，也打包提供了相关的并行环境、作业调度软件等，使系统的建设和维护的成本较小。
与现有IT基础架构完美整合

如下图所示，采用微软的WCCS可以很好的和企业现有的IT基础架构整合在一起。方便用户的使用和维护，简化企业的IT架构， 并能能充分利用已有的工程技术人员和计算机管理人员的知识与技巧。试用版请联系：caeda@vip.sina.com


]]> http://www.caeda.com.cn/blog/article.asp?id=72 caeda@vip.sina.com(lns) Tue,22 Jul 2008 15:07:57 +0800 http://www.caeda.com.cn/blog/default.asp?id=72
报告题目（一）:

Fault Tolerance for PetaScale Systems: Current Knowledge, Challenges and Opportunities

报告人：Professor Franck Cappello( INRIA)

时间：2008年7月25日（星期五）上午9：30-10：30

地点：软件所5号楼3层337会议室

简介: Professor Franck Cappello holds a Senior Researcher position at INRIA. He leads the Grand-Large project at INRIA, focusing on High Performance issues in Large Scale Distributed Systems. He has initiated the XtremWeb (Desktop Grid) and MPICH-V (Fault tolerant MPI) projects. He is currently the director of the Grid5000 project, a nation wide computer science platform for research in Grid and P2P. He has authored more than 60 papers in the domains of High Performance Programming, Desktop Grids, Grids and Fault tolerant MPI. He has contributed to more than 40 Program Committees. He is editorial board member of the international Journal on Grid Computing, Journal of Grid and Utility Computing and Journal of Cluster Computing.He is a steering committee member of IEEE HPDC and IEEE/ACM CCGRID.He is the General co-Chair of IEEE APSCC 2008, Workshop co-chair for IEEE CCGRID'2008, Program co-Chair of IEEE CCGRID'2009 and was the General Chair of IEEE HPDC'2006.

报告摘要:

The emergence of PetaScale systems reinvigorates the community interest about how to manage failures in such systems and ensure that large applications successfully complete. Existing results for several key mechanisms associated with fault tolerance in HPC platforms will be presented during this talk. Most of these key mechanisms come from the distributed system theory. Over the last decade, they have received a lot of attention from the community and there is probably little to gain by trying to optimize them again. We will describe some of the latest findings in this domain. Unfortunately, despite their high degree of optimization, existing approaches do not fit well with the challenging evolutions of large scale systems. There is room and even a need for new approaches. Opportunities may come from different origins like adding hardware dedicated to fault tolerance. We will sketch some of these opportunities and their associated limitations



                                             All are Welcome!

Lab. of Parallel Computing, ISCAS





报告题目（二）:

EDGeS: A Bridge Between Desktop Grids and Service Grids

报告人：Dr. Haiwu HE（贺海武 博士）

时间：2008年7月25日（星期五）上午10：30-11：15

地点：软件所5号楼337会议室

简介: Dr.Haiwu HE has an research engineer expert position in INRIA. He got his Master's degree and Ph.D in LIFL (Laboratoire d'Informatique Fondamentale de Lille) of University of Sciences and Technologies of Lille USTL (Lille I ). The main subjects of his Ph.D Program are supercomputing algorithms, high performance P2P computing and Grid computing.He is working on an OpenSource Middleware for Data Desktop Grids: BitDew.He is now also a research scientist of European FP7 project EDGeS(Enabling Desktop Grids for e-Science) aiming to work on bridge from "institutional" grid such as EGEE to desktop grids such as BOINC, and XtremWeb.

报告摘要:

Desktop grids and service grids widely used by their different users communities as efficient solutions for making full use of computing power and achieving loads balances across Intranet or Internet. Nevertheless,little work has been done to combine these two girds technologies together to establish a seamless and vast grid resources pool. In this paper we will present a new European FP7 infrastructure project:EDGeS(Enabling Desktop Grids for e-Science),which aim to build technological bridges to facilitate interoperability between desktop grid and service grid. We give also a taxonomy of existing grid systems: desktop grids such as BONIC and XtremWeb, service grids such as EGEE. Then we describe furtherly our solution for identifying translation technologies for porting applications between desktop grids and service grids, and vice versa. There are three themes in our solution, which discuss actual popular bridging technologies, user access issues, and distributed data issues about deployment and application development.





                                                 All are Welcome!

Lab. of Parallel Computing, ISCAS
]]> http://www.caeda.com.cn/blog/article.asp?id=71 caeda@vip.sina.com(admin) Sat,19 Jul 2008 08:46:47 +0800 http://www.caeda.com.cn/blog/default.asp?id=71
When post-processing results, you often want to compare two different models in the exact same position and size. The ‘Copy and Past Viewpoint’ feature allows you to set the same position and angle to display your model but, in addition to that, you might also want to have the exact same size for the two windows you are comparing. This would ensure that all the pictures in your report have the same original dimensions.

One new Python command will let you do that very easily.

This command is:


SetViewerSize(w,h)
where w is the width of the active viewer and h is the height.

To use it, you can click on the ‘Edit Python script’ icon:



Figure 1: 'Edit Python script' icon

In the new editor window, type: SetViewerSize(350,450)

Then click on one of the ‘Run script’ button to resize your window (Figure 2). The "!" button on the left runs the script using the current model. The "n!" button on the right starts a new model and runs the script using that new model.



Figure 2: 'Run script' buttons

This can easily be made a macro for frequent reuse. For that, save the short Python script with only this line in your working directory. Then, click on Tools/Macro and Rescan, you will then have a new macro available:



Figure 3: Macros panel

Now, just by clicking on that ‘resize’ button, your window will be resized to the dimensions you chose.

Here, we wrote a very short script, just one line, but of course, this command can be added to any Python script.

If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know.

]]> http://www.caeda.com.cn/blog/article.asp?id=70 caeda@vip.sina.com(lns) Tue,15 Jul 2008 07:36:53 +0800 http://www.caeda.com.cn/blog/default.asp?id=70


为培养高性能计算专业人才、推广高性能计算应用、加强与国家科研单位、高等院校及企业的交流与合作，推动区域高性能计算应用与发展，中国科学院计算机网络信息中心超级计算中心在中科院西安分院的大力支持下，将于2008年7月28-29日在西安举办高性能计算培训班。



具体通知如下：

l         培训对象：具备高性能计算和Linux基础知识，已经从事与高性能计算应用相关的科研院所及高校老师、硕博研究生及相关企业工程师，或对高性能计算感兴趣的相关人员。

l         培训时间：2008年7月28日－7月29日

l         培训地点：西安，东方大酒店（西安市小寨西路子午路）

住宿费：200元/标间/天、210元/单间/天

    培训费用：资料费100元/人

l         食宿安排：授课期间，提供免费工作午餐

l         报名方式：

按照报名回执表的要求填写各项内容，然后发邮件到 lhf@sccas.cn

我们将在一个工作日之内给予回复。

]]> http://www.caeda.com.cn/blog/article.asp?id=69 caeda@vip.sina.com(admin) Sat,12 Jul 2008 08:37:16 +0800 http://www.caeda.com.cn/blog/default.asp?id=69
In this user tip, we consider a CFD simulation that contains a volume where a pressure loss occurs due to a porous media in the domain (see Figure 1). If a compressible flow has been assumed, the density could be affected due to that pressure drop.

To control such physical effects it’s helpful to determine the difference of pressure, or density, between two plotted areas (cutting planes). In CFD-VIEW, the user can easily create two cutting planes. For calculating the differences between two planes (topology must be identical), and plotting results on one of them, the user can use a Python script.




Figure 1. Porous media case in CFD-VIEW

The script is doing the following:

Create the first cutting plane and get the value of the desired variable.
Create the second cutting plane and get the value of the same variable as cut one.
Check to see that the two cutting planes have the same number of nodes.
Determination of the difference of the variables in each node.
Create a new variable on the second cutting plane to store the result of the operation.
Showing the new variable at the surface in a convenient view.
To execute the script, select Tools --> Macros to open the script dialog. Once the dialog is open, select the script to run on the current model.

Figure 2 is showing the pressure (P) along the flow direction as well as the pressure drop (DP), the result of this script.



Figure 2. Result after running script. Delta P plotted on Cut 2.

Information about scripting can be found in the Scripting chapter of the CFD-VIEW WebHelp or PDF file in the CFD-VIEW installation.

If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know.
]]> http://www.caeda.com.cn/blog/article.asp?id=68 caeda@vip.sina.com(admin) Fri,11 Jul 2008 08:59:01 +0800 http://www.caeda.com.cn/blog/default.asp?id=68
专为解决船舶设计中的几何建模、数值模拟、系统振动以及形状优化等问题而开发的CAE(计算机辅助工程)系统。

Friendship-frameware是一个适用于先进的船舶水动力学设计的整合环境。它将CAD系统与CFD分析紧密结合起来，可以在船体形状设计流程的关键阶段给予工作人员巨大帮助。该系统支持对变量进行调和一致的生成和评估，并与进一步的约束管理合成一体。海军设计师可根据自己个人的具体设计环境对其量身定做，全面提高自身的生产力和竞争力。

特性
支持通过偏移、平板网格、IGES几何文件等方式导入部分参数化的建模数据，并据此进行高效率的形状改变
支持有效的类型说明和完全参数化几何形状定义
支持通过完整的流体静力学模拟进行尖端精妙的偏移量处理（分析、修正）
支持非均匀有理B样条（NURBS）曲线曲面技术
支持采用高级设计实体方法（例如F-样条曲线，翼型，运动预测点等）
支持将嵌入式优化策略应用于探索（例如采用实验设计，DoE）和开发(例如采用多目标遗传算法，MOGA)工作
支持创建和管理所有手动和自动变量
支持对项目数据和文件的舒适浏览
直接生成报告（数据表格、图表等），方便对各个变量作对比
包含已封装的前后处理器，用于CFD求解器（例如求解兴波阻力的位势流计算程序）
支持直接与Direct coupling to SHIPFLOW船舶流体设计仿真软件, n-Shallo, SEDOS进行耦合分析
完成以上功能不需要额外编制脚本或接口程序
以上功能在LINUX和WINDOWS下均有效

应用
造船厂、咨询机构、模型水池及研究所中的先进水动力学设计
整体调查研究，例如速度范围研究、变量整理和浅水影响研究等
船体、推进器等重要功能性表面的微调
高校中对基本原理设计的教育
进一步资料
caeda@vip.sina.com
www.caefinder.com
]]> http://www.caeda.com.cn/blog/article.asp?id=67 caeda@vip.sina.com(admin) Sun,06 Jul 2008 08:42:13 +0800 http://www.caeda.com.cn/blog/default.asp?id=67
     诚邀您参加2008“产品现代设计与技术创新”论坛！

     本届论坛亮点：设计与创新主题报告、三方面产品设计专题报告、《中国制造业
上市公司产品创新能力排名榜》发布！国内外专家队伍空前盛大。（详细情况如下或见
附件邀请函，论坛网址http://forum.cmdt.cn ; ）

  （直接回复邮件便可报名参加，如传真，请填写附件报名表）



主办单位：教育部现代设计与制造网上合作研究中心

                   中国工程院产业工程科技委员会

                  上海交通大学

                 上海市院士咨询与学术活动中心

论坛主题： 产品创新  现代设计 技术研发 知识服务  

研讨中心： 装备自主研发中的问题---设计与创新

专题研讨： 内燃机设计 巨型重载装备设计  核岛主泵设计

时   间：  2008年7月31日-8月1日（31日为主题研讨、1日为专题研讨）

地   点：  上海市上海交通大学浩然高科技大厦

论坛特色： 主题报告引导下的深入讨论交流；针对具体产品设计设三个子论坛进行专
题讨论。─ 届时发布《中国制造业上市公司产品创新能力排名榜》！论坛后续以“产
品现代设计” 系列培训,同时举办“创新思维与现代设计”师资培训班。



一、主题报告：

1.提高装备制造业自主创新能力 （清华大学教授 柳百成院士）

2.设计思维: 现代设计与技术创新的基础（美国南加州大学IMPACT中心主任 Stephen
C.Y. Lu 教授）

3.支持创新的知识服务(上海交通大学、西安交通大学教授 谢友柏院士)

4.巨型重载操作装备的自主研发（上海交通大学副校长 林忠钦教授）

5.重大机械装备的数字化设计与动态监测研究(武汉理工大学校长 周祖德教授）  

6.制造业上市公司产品创新能力评价与排名榜发布（教育部现代设计与制造网上合作研
究中心上海主管 戴旭东博士）

二、内燃机设计专题报告：

1.专题报告一：发动机Benchmark和概念设计(上海交通大学汽车工程研究院院长 许敏
教授)

2.专题报告二：发动机的设计开发与试验开发(上汽研究院副院长 辛军博士)

3.专题报告三：发动机的CAE开发流程(AVL李斯特技术中心（上海）技术经理 马明堂博
士)

4.专题报告四：轿车柴油机的开发 (长丰汽车发动机有限公司韩志玉副总)

三、巨型重载设备设计专题报告：

1、专题报告一：锻造操作机机构性能分析与设计（上海交通大学重大装备设计与控制
工程研究所所长 高峰教授）

2、专题报告二：重载操作机的力学性能分析与优化设计（大连理工大学，李刚教授）

3、专题报告三: 重型夹持装置的承载能力建模及其驳动控制策略（中南大学 邓华教
授）

4、专题报告四：自由锻造操作机的多领域仿真（上海交通大学 王皓 博士）

四、核电站主泵设计专题报告：

1、专题报告一：核主泵轴密封设计技术（清华大学教授 王玉明 院士）

2、专题报告二：核主泵标准规范（国家核安全局 方庆贤教授）

3、专题报告三：核主泵国产化技术（上海核工程设计研究院 陈鉴墅高工）

4、专题报告四：核主泵虚拟仿真技术（上海交通大学核电中心  张继革 博士）

五、论坛其它特邀研讨专家与领导：

1.中国工程院副院长                    邬贺铨 院士

2.上海市院士咨询与学术活动中心主任    翁史烈 院士

4.浙江大学教授                            谭建荣 院士

5.西安交通大学教授                    蒋德明 教授  

6.清华大学摩擦学国家重点实验室主任    雒建斌 教授

7.河北工业大学副校长                 檀润华 教授

8.上海交通大学重大装备设计与控制工程研究所所长  高 峰 教授

10.西安交通大学润滑理论与轴承研究所所长                陈 渭 教授

11.北京交通大学                          查建中教授              

12.浙江大学                                  潘双夏教授              

13.湖南大学                                  李仁发教授              

14.四川大学                                  李  彦教授                    

15.SolidWorks公司技术总监     C.C,Chen 先生

16.AMT咨询资深顾问                 王玉荣 博士  

论坛还将邀请上海市经委、上海市科委、中国工程院上海院士咨询与学术活动中心、教
育部现代设计与制造网上合作研究中心学术委员会委员，现代设计与产品研究开发网络
理事会理事以及关注本论坛研讨主题的同行专家参加。(详情请访问：
http://forum.cmdt.cn/）

六、“产品现代设计”系列培训：

培训1：产品现代设计与创新理论

培训2：产品三维实体建模（CAD建模）

培训3：结构有限元分析（CAE分析）

培训4：机械系统动力学建模与仿真（数字样机构建）

培训5：计算机辅助制造（CAM）

培训6：产品设计优化

培训7：创新思维与现代设计（师资培训班）

培训8：企业技术研发平台构建策略与方法（根据企业要求开设）

（详细课程内容请参阅：http://202.117.208.15/px/px-11.html）。

七、报名方式: Email、电话、传真报名均可（ 截至时间：2008年7月23日）

八、费用：论坛注册费：人民币￥1000元/人（会员参加优惠费：800元/人）。

          产品现代设计系列培训班（5天40学时）每班2000元、会员每班1600元）。

九、联系方式：

上海交通大学现代设计研究所  教育部现代设计与制造网上合作研究中心（上海）

地址：上海市华山路1954号上海交通大学机械楼605室       邮编：200030

E-mail：zhinanz@yahoo.cn；cmdtforum@gmail.com

电话：021-34206245、021-62933701、    

传真：021-62932252、021-34206245

联系人：孙明杰（会务联系），张执南（13585718558，专业联系）  戴旭东
（13701931304,合作联系）    
]]> http://www.caeda.com.cn/blog/article.asp?id=66 caeda@vip.sina.com(admin) Sat,05 Jul 2008 08:31:50 +0800 http://www.caeda.com.cn/blog/default.asp?id=66
Under relaxation is a constraint on the change of a dependent or auxiliary variable from one solution iteration to the next. It is required to maintain the stability of the coupled, non-linear system of equations. The relax tab in the solver control panel (see Figure-1) allows the user to set under-relaxation factors for each of the solved variables and the auxiliary variables.



Figure - 1. Solver Control - Under Relaxation

The panel contains four columns: the first defines the variable, the second contains a slider bar which can be used to adjust the value, the third contains up/down buttons to adjust the order of magnitude of the value, and the fourth is a field for the under relaxation value itself.

We have different methods for applying under relaxation for the solved and auxiliary variables that are listed below:

Inertial Relaxation (I):
Inertial under relaxation (I) is applied to variables, which are directly solved for (dependent variables as determined by active modules) during the iterative procedure, for example, velocities, pressure correction, enthalpy, etc.

I usually varies from 0.0 to 2.0 with default value of 0.2.
Increasing the value of I adds constraint. It means increasing I increases stability.
Increasing the value of I slows convergence. It means an increase in I will take more time to get the same order of convergence.
Values of I greater than 1.5 are allowed but not recommended.
Linear Relaxation (L):

Linear under relaxation (L) is applied to all variables that are computed during the solution procedure. These variables are called auxiliary variables, which are computed from the solved (dependent) variables, for example, density, pressure, temperature etc.

L usually varies from 0.0 to 1.0 with default value of 1.0.
Decreasing the value of L adds constraint. It means decreasing L increases stability.
Decreasing the value of L slows convergence. It means a decrease in L will take more time to get same order of convergence.

It all can be summarized in Figure-2 and Figure-3 below.



Figure-2. Under Relaxation for Faster Convergence



Figure-3. Under Relaxation for More Stability

Note: Please note that relaxation values can help in getting faster convergence or it may help prevent divergence. For a given problem (identical BC/VC/IC), change in relaxation values may take more or less number of iterations to reach convergence. But as long as problem is fully converged, you will get the same result irrespective of relaxation values.

Tips on troubleshooting your problems:

The following tips are just guidelines that can help in getting a converged solution or faster convergence. The values on relaxation can be problem specific, so there are no hard and fast rules as to which value one should use.

Problem Diverges:
If you see that your problem is diverging, you can try the following:

Make sure that you have applied correct scaling and all input values (BC/VC) are correct or at least in reasonable range.

Check the residual and see what variable diverges or starts diverging first.
Decrease the linear under relaxation (from 1.0 to say 0.7) for that variable.
In order to make it more stable, you can also increase inertial under relaxation for the corresponding solved (dependent) variable.

For compressible flows, decreasing linear under relaxation for density helps.
For problem involving heat transfer, if you see enthalpy is diverging, decreasing the linear relaxation of temperature from default value of 1.0 to smaller value like 0.7 can help in getting converged solution.

For Fluid Structure Interaction problems, decreasing the linear relaxation on pressure helps to moderate the pressure fluctuations seen by the stress solver, reducing the displacement fluctuations and aiding in convergence.

If it is a Fluid Structure Interaction problem and you encounter negative volumes, first try to decrease the linear relaxation for pressure to a value of 0.3 or 0.2. If the problems still exists, then you can try to decrease the linear relaxation for Grid Deformation anywhere from 0.5 to 0.1. This basically restricts the grid deformation in the solid volumes to 50% (if a value of 0.5 is used) of the actual value due to sustained pressures every time you solve for stress during the time step. Upon convergence, you still get the correct grid deformation.

For complex physics, when small changes in relaxation do not work, change the inertial relaxation values to 0.5. Also, reduce the linear factors to 0.3 and rerun. If this does not work, change the inertial factors to 0.9 and the linear ones to 0.1. These factors can be changed up to 1.5 for inertial and 0.01 for linear. Anything higher may result in a solution that has been frozen to the initial field.

Another item that may help is a change to the AMG solver for pressure correction or enthalpy. If convergence problems still persist, look at the residual information especially noting the location of the maximum residual. Next examine the grid closely at this spot in CFD-VIEW and look for skewness. Sometimes problem areas can be isolated by plotting the results every few iterations. The problem area is generally the location where the flow field first becomes unstable.

Slower Convergence:
If you see that convergence is very slow, you can try following:

Check the residuals and see what variable has slow convergence (might also remain flat)
Decrease the inertial under relaxation (from 0.2 to say 0.02) for that variable.

For conjugate heat transfer problems, decreasing the inertial relaxation of enthalpy from default value of 0.05 to smaller number like 1E-05 can help in faster convergence.

When solving for the electric module, decreasing the inertial relaxation of electric potential from 0.0001 to smaller number like 1E-07 can help in faster convergence.

The next time you setup your simulation, trying using these suggestions for setting your Solver Control parameters.
]]> http://www.caeda.com.cn/blog/article.asp?id=65 caeda@vip.sina.com(admin) Tue,01 Jul 2008 09:39:10 +0800 http://www.caeda.com.cn/blog/default.asp?id=65
    以下从AMI产品特点、FASTRAN特点、个人使用感受和对比结论四个方面进行论述，条件所限，难免有以偏概全之嫌，敬请批评、指正。

一、AMI公司产品的特点
完整的AMI公司产品线如下图所示：
  
  
各模块的求解器及其相互关系如下图所示：



    由以上两图可以看出：
1、AMI公司提供了完善的涵盖气、液流体、可压与不可压缩、有粘性与无粘性求解器；
2、AMI产品前、中、后处理工具完善，建模、仿真、显示自成体系，均有相应的工具；
3、AMI软件的各求解器所用的方法以面元法和欧拉法为主，对于有粘性流动问题，大部分的AMI模块是通过加入修正系数求解解决，技术上比较成熟，但不够先进；
4、从整个产品线的发展状况来看，AMI产品更适合处理低速、亚、跨音速问题，以求解不可压缩流动为主，因此，AMI软件非常适合解决液体问题，如水面、水下问题；
5、AMI的FPI（Flight Path Integration）模块可以解决空间多体运动问题，如发射水雷、飞机投弹等问题，可作气动、运动耦合分析，并为自动驾驶仪设计提供仿真结果；
6、部分求解器有并行求解功能，能够满足流场仿真高性能计算要求；
7、AMI产品还有一些专用的设计工具如：翼形设计、优化设计、化学反应气动问题等，以针对特定问题提出解决方案。

二、FASTRAN产品特点：
    由于对FASTRAN软件比较熟悉，对其功能、适用范围等特点这里不再赘述。

三、个人使用感受
    由于工作机会，本人曾经使用过AMI产品中的GRID－GEN(建模)、MGAERO 、USAERO（气动仿真）、MGFPI（运动分析）以及OMINI3D（后处理）模块，在linux操作系统下使用。
1、AMI求解器的求解效率比较高，求解速度比较快，对硬件的要求相对其他CFD软件来说不算很高；
2、AMI产品的界面比较繁杂，尤其是其建模前处理软件，不同于标准的windows界面，也不是传统的linux或Exceed软件界面，习惯了其他CFD软件的工程师必须通过一定的时间，才能熟练使用，如下图所示：

前处理软件――GridGen

后处理软件：Omini3d
      另一方面，在前、后处理软件中有非常多的快捷键，一旦熟练使用，其效率还是非常高的。

3、在FASTRAN中，求解器有图形交互式用户界面（GUI），可以非常方便的设置初始条件、边界条件、定义运动等，完成参数的设置。在AMI软件中，求解器没有图形用户界面，通过编辑指定路径下的文本文件，控制求解参数，继而通过命令行的方式调用求解器，不如FASTRAN方便。
    但是，在另一方面，这种开放的结构却提供了入口，与其他软件共同搭建集成一体化设计平台。曾经见过几个单位，用UG二次开发建模，整合其他的气动、运动、优化软件，作出了适合本行业应用的建模、气动、运动一体化快速设计与评估软件。

四、对比结论
    综上所述，个人认为AMI与FASTRAN都是CFD软件，肯定是竞争的关系，但是在以下方面，其重叠度并不高：
1、主要的应用领域不同：从个人所了解到的国内客户分布也可以看出AMI产品更多的应用于水面船舶、水下舰艇等领域；而CFD－FASTRAN主要应用于航空、航天领域；
2、求解器不同：AMI求解器以面元法、欧拉法为主；FASTRAN以求解可压缩NS方程为主；
3、使用方法不同：FASTRAN使用方便，界面简洁；AMI界面不易上手，但灵活度较高。

]]> http://www.caeda.com.cn/blog/article.asp?id=64 caeda@vip.sina.com(admin) Sun,29 Jun 2008 15:15:20 +0800 http://www.caeda.com.cn/blog/default.asp?id=64 CAD Integrated CFD for Design Simulation

Introduction

CFdesign is one of a new generation of CAD-integrated design simulation tools. Like PTC's Pro/Mechanica it focuses on bringing the power of advanced simulation technology to the design engineer's desktop computer, rather than just being for a small group of highly qualified R&D specialists using high-end workstations and supercomputers. Unlike Pro/Mechanica though, the focus of CFdesign is on the closely coupled areas of fluid-flow and heat-transfer simulation, using the technique known as 'Computational Fluid Dynamics' or CFD.

Almost every mechanical or electronic product involves fluid-flow or heat-transfer at some stage in its manufacture or operation, and often a combination of the two. Every machine or electronic circuit uses power, and that power turns to heat which is dissipated by a combination of conduction, convection and radiation. Anywhere except in the vacuum of outer space, cooling involves air, water or some other fluid. An understanding of fluid-flow and heat-transfer processes thus becomes important for many designs, and can still offer useful insight into the performance and optimal design of less critical products, ranging from toaster-ovens to taps (faucets). Applications in the automotive, aerospace, electronics, medical, materials processing, power-generation, HVAC and fluid power/control industries are too numerous to mention.

For simulation tools to be useful to the design process, they must be designed in a way that they can be integrated directly into that process, and thus CAD integration, parametrics and associativity become key requirements. This differs from the traditional use of CAE and CFD software as 'validation tools' that can only be used as a final check at the end of the design process. By that time it's usually too late for simulation to have any real impact on the design, so what is needed are tools that can actually drive and guide the design process.

Using the Software
Unlike most other CFD software, CFdesign has no geometry creation or editing facilities in its pre-processor, and that is because 100% of the geometry comes from the model created in your own CAD software. CFdesign reads 3D geometries directly from Pro/Engineer and most other CAD programs based on the ACIS or Parasolid kernels, including Inventor, SolidWorks, UG, SolidEdge, and many others. No CAD conversion using IGES or STEP is required, so translation errors and inconsistencies are eliminated. This direct CAD integration also brings the advantages of parametrics and associativity to your simulation model, so modeling is fast, reliable, and easy to learn.

Once you have transferred your geometry to the system your first task is to define the global parameters for the analysis problem and model. This is done in a very similar manner to many CAD applications, using the feature/part browser tree window at the left hand side of the screen. Items to be entered include measurement units (although values can be entered in any units you require) and model materials, which are chosen from the supplied library or from a user-defined custom material with many advanced non-linear fluid types available. The next stage is the definition of loads and boundary conditions including inlets, outlets, heat sources, as well as any planes or axes of symmetry. The final set-up stage prior to building the element mesh, is to give the system a rough guide as to mesh element size, though this is automatically adapted to suit the actual geometry. Meshing in CFdesign uses a unique adaptive mesh-enhancement technology which can automatically produce highly optimized meshes for fast accurate solutions, but using fewer elements/cells than most other CFD tools. Most other aspects of the solution set-up are handled automatically by the intelligence of the software, though advanced users may have access to a full range of solver controls.

The analysis process within CFdesign is very interactive, especially in comparison with other analysis tools (CFD or otherwise). The interface is split into two sections, with the classic convergence graph at top, showing the progress of your analysis run towards an ideal solution. Below this you have the inspection window, which allows you to view the results, as each iteration is completed. This includes interrogation tools such as dynamic slicing, flow vectors, etc. As you would also expect, should you deem that your analysis is veering off course, or indeed you feel that the results are adequate for your needs (for quick test runs), you can stop the process at the next iteration. At any time you can also halt the analysis to change boundary conditions, materials, or even the mesh, and then restart automatically from the previous solution. This facility allows quick multi-parameter design studies without having to restart the analysis from the beginning each time.

Results
Once you have reached your decided goal (either a partial or a fully converged solution), you can then utilize the full range of results inspection and reporting tools on offer. These allow you to display the usual range of properties associated with CFD such as velocity magnitudes, pressures and temperatures, density, stress (on walls), heat flux etc. These are complimented with a range of graphical tools such as the use of iso-surfaces, cutting planes and particle tracing (which can be animated) all of which allow you to extract precisely the data you require from your analysis. You also have the ability to output your results in non-graphical formats such as graphs and text-based listings. Simulation results such as temperature and pressure can be exported to a range of popular structural CAE programs including Pro/Mechanica, NASTRAN and ABAQUS for use as thermal or mechanical loads.

Functionality
CFdesign includes the full range functionality and physics seen in most other CFD, electronics cooling or general heat-transfer simulation tools. Rather than provide a long and detailed listing of the functionality it should suffice to mention that the analysis capabilities include:
•    2D, 3D and axisymmetric domains with Cartesian or cylindrical coordinates
•    Internal and external compressible and incompressible  flows
•    Laminar and turbulent flows (Eddy-viscosity, K-E & RNG)
•    Steady-state and transient analyses with multiple independent time-varying BCs
•    Conjugate heat-transfer with conduction, convection (forced & natural) and radiation
•    Multiple non-linear and non-Newtonian fluids
•    Humidity, steam and dissolved concentration scalars
•    Distributed resistance and porous media
•    Valves and internal or external fans with head-capacity curves

Summary
CFdesign is one of the most powerful and flexible design simulation tools available, with potential applications in a wide variety of products and processes. Designers can use it to simulate their products right from the initial concept stage through to final validation, and gain a much greater understanding of performance than is possible using standard experimental techniques. Simulation can thus be used to drive the design process towards the combined goals of better products, cheaper and faster.

]]> http://www.caeda.com.cn/blog/article.asp?id=63 caeda@vip.sina.com(admin) Sat,21 Jun 2008 06:40:31 +0800 http://www.caeda.com.cn/blog/default.asp?id=63
Multiphase flows occurring in various scientific and engineering fields and application areas like biotechnology, bio-defense, micro-fluidics, chemical processing and environmental research etc., can be modeled using the Macroparticle module in CFD-ACE+. Macroparticles are considered to be large particles that block a significant portion of the flow path and thus affect the surrounding flow field. In such cases, the Spray module method is not suitable and the Macroparticle technique should be used.

Below is a list of features for the Macroparticle module:

Allows usage of multiple particles
Accounts for particle-particle interactions
Coupled with the Flow, Turbulence, User Scalar, and Biochemistry options
Particle motion includes translation and rotation in X, Y and Z directions
Accounts for pressure forces and viscous forces on the particles
User subroutine access for defining particle shape, external forces, etc.
For more information on Macroparticles, please refer to the Macroparticle chapter in the Modules section of CFD-ACE+ WebHelp.

In the example for this user tip, a user subroutine is used to add a body force to the particle immersed in an infinite medium filled with a quiescent fluid.


Example Case


In this example, a periodic body force is added to the macroparticle using the UPBLOCK_BFORCE user subroutine available in CFD-ACE+.

A 50m x 50m x 50m cube with no-slip wall boundary condition was taken. In order to achieve neutral buoyant condition the gravity force was set to zero. The spherical macroparticle was taken to be small compared to the size of the cube and placed at the centre. The initial velocity for the particle was set to zero, but it was subjected to an external periodic force (along X-direction) through a user subroutine called “Upblock_bforce” subroutine, thus taking the external periodic force on the macro particle as a body force.



Figure 1. Initial position of the particle in the domain

The body force applied to the particle is a function of time. The current time in the simulation is obtained using the get_time() subroutine which returns the current time and time step number for the simulation. Using the set_macp_valuer() routine, the body force in the X direction ( variable name in CFD-ACE-SOLVER is "GBFX") is set on the particle.

If you would like the source code for the user subroutine and the DTF file used in this case, please click here.


If you have any questions about this feature or would like us to discuss
]]> http://www.caeda.com.cn/blog/article.asp?id=62 caeda@vip.sina.com(lns) Tue,17 Jun 2008 08:42:33 +0800 http://www.caeda.com.cn/blog/default.asp?id=62
培训时间：2008年6月20-21号两天，上午9：00至下午4：00.
培训地点：北京朝阳区东三环南路58号（劲松桥东北角），富顿中心B座601室培训室。
联系人：刘先生，1390 106 4208，caeda@vip.sina.com （请务必预约参加）

课程目标
    CAEDA公司举办的此次培训是为了让复合材料分析设计工程师掌握先进复合材料分析设计软件，在产品设计过程中采用全面和有效的分析设计方法以降低产品开发过程的成本和时间。学员将能学习到ESAComp软件的基本操作技巧，复合材料单层、层合板的定义和分析，复合材料梁、板、加筋板、连接等结构设计和分析，ESAComp软件与其它有限元软件综合分析流程，以及相关复合材料分析设计理论。

课程特色
    提供免费软件试用，理论学习和实际操作相结合，保证熟练掌握。
    提供最新的软件培训教程和技术文章。

参加对象
    本次培训面向所有从事或将要从事复合材料分析设计的人士，只须具有复合材料的基本概念即可，并不需要有限元技术方面的相关经验。

培训时间及地点
时间：2008年6月20-21日，每天上午9点至下午4点，中午午餐一小时
地点：北京朝阳区东三环南路58号（劲松桥东北角），富顿中心B座601室培训室。

课程安排:
6月20日:
    8:45AM        签到
    9:00-10:00     ESAComp复合材料分析设计软件介绍、数据管理、基本操作
    10:00-10:15    休息
    10:15-11:50    复合材料单层的设计与分析及其相关理论，操作练习
    12:00-1:15     午餐
    1:15PM-2:45   复合材料层合板的设计与分析及其相关理论，操作练习
    2:45-3:00      休息
    3:00-4:00      复合材料层合梁的设计与分析及其相关理论，操作练习
    4:00-4:30      讨论


6月21日
    8:45AM        签到
    9:00-10:30     复合材料层合板与加筋板结构的设计与分析及其相关理论，操作练习
    10:30-10:45    休息
    10:45-11:50    复合材料结构连接的设计与分析，操作练习
    12:00-1:15     午餐
    1:15PM-2:45   有限元接口与综合分析流程，操作练习
    2:45-3:00      休息
    3:00-4:00      上机操作练习、交流讨论

]]> http://www.caeda.com.cn/blog/article.asp?id=61 caeda@vip.sina.com(admin) Sat,14 Jun 2008 07:43:44 +0800 http://www.caeda.com.cn/blog/default.asp?id=61
A Point Net is primarily used for creating an evenly spaced vector field over existing 3D volumes, 2D surfaces (if the domain is two-dimensional) or XYZ cutting planes. The Point Net creates a set of points over the selected entity and the simulation variable values (such as pressure, temperature, etc.,) at these points are derived by interpolation on the underlying mesh.

Figure 1 (a) shows the vectors as they would be on the original grid. You can see that due to the grid density near the walls, the vectors are tightly clustered near the walls. In Figure 1 (b), which uses the Point Net feature, the vectors are evenly spaced leading to a tidier presentation of the results. In each of the figures, the vectors are colored by Velocity Magnitude.



To create a point net

Select an object, which can be a volume or a single XYZ cutting plane
Click on the Point Net Operator






In the Point Net control panel, input the number of points in the X, Y, & Z directions






Check the Vectors On button in the ToolBar

There are additional display options that can be employed at this point:
the vectors can be colored based on a computed variable (e.g. pressure, temperature, etc.)
the Points tab in the Control panel can be used to display the points as pixels, cubes, balls, etc.
the Vectors tab in the Control panel can be used to adjust the size of the vectors
If you have any questions about this topic or would like us to discuss some other topic in the future, please let us know.

]]> http://www.caeda.com.cn/blog/article.asp?id=60 caeda@vip.sina.com(lns) Mon,09 Jun 2008 08:29:00 +0800 http://www.caeda.com.cn/blog/default.asp?id=60     
    软件背景：IOSO NM是由俄罗斯SIGMA Technology公司开发，能提高各种复杂系统和过程的优化设计效率。它可以把工程师从十分复杂和繁重的设计参数寻优工作中解放出来，从而使工程师能集中精力于核心问题。
    软件特性：
    针对大规模计算任务的高效而独特的算法
    易于操控的全自动优化过程
    较低的资源消耗
    并行优化过程
    良好的设计优化健壮性
    软件优势：IOSO NM能从容地处理大规模优化任务，最大优化目标可达20，最大的优化变量为100. IOSO NM采用了开放式消息传递架构，可集成不同的商用CAE和CFD软件，如ABAQUS, MSC Nastran，ANSYS, CFX, Fluent，StarCD, Numeca等。
试用联系：caeda@vip.sina.com, www.caeda.com.cn
]]> http://www.caeda.com.cn/blog/article.asp?id=59 caeda@vip.sina.com(admin) Sat,07 Jun 2008 07:57:57 +0800 http://www.caeda.com.cn/blog/default.asp?id=59
Rotation, zoom, pan and other graphical operations require re-drawing of the model in the new position within the graphics window. When dealing with large models, these operations can slow down considerably due to the huge amount of graphical data that needs to be processed and re-drawn. CFD-VisCART V2008.2 provides the user the option to view the model in a graphics-light wireframe mode, making the graphical operations faster and easier to handle.

In this mode, outlines of the surfaces together with the surface triangles connected to them are displayed. For closed surfaces (no outlines), about a hundred random surface triangles are displayed.

To enable wireframe display, click the wireframe mode button as indicated in Figure 1.



Figure 1: Model displayed in shaded surface mode. Wireframe mode button location indicated.

The display switches to the wireframe mode showing only outlines and some surface triangles (Figure 2).




Figure 2: Model displayed in wireframe mode.

This is one of the many new features included in the V2008.2 release, which can be downloaded from the Software Download section of the ESI CFD Customer Support Portal.

If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know: caeda@vip.sina.com
]]> http://www.caeda.com.cn/blog/article.asp?id=58 caeda@vip.sina.com(lns) Thu,05 Jun 2008 09:23:38 +0800 http://www.caeda.com.cn/blog/default.asp?id=58
       请各位Club的成员务必抽时间出席此次论坛，算作咱们Club的一次活动。会议日程附后。


6月13日下午，西苑饭店四层 鸿运厅

13:10～13:30 会议签到

　 嘉宾致辞

13:30～13:50 中国软件行业协会 陈冲 理事长
　 原信息产业部软件与集成电路促进中心 邱善勤 副主任
　 中国计算机学会高专委副主任 迟学斌 研究员（国家863专家组）
　 主题演讲

13:50～14:15 主题演讲：浪潮天梭高性能计算机系统在科研和商业领域的应用实践  

14:15～14:40 主题演讲：基于惠普新型刀片平台部署高性能计算应用

14:40～15:00 主题演讲：中科院超算中心在科学计算中的应用
　 迟学斌 主任 中科院超算中心

15:00～15:20 主题演讲：上海超算在工程方面的应用
　 李根国 副主任 上海超算中心

15:20～15:45 主题演讲：曙光的国产高性能计算机平台

15:45～16:10 主题演讲：AMD多核处理器加速高性能计算产业发展

16:10～16:30 主题演讲：计算系统虚拟化：机遇与挑战
　 金海 教授 华中科技大学

16:30～16:50 主题演讲：虚拟化技术在高性能计算领域的应用
　 孙毓忠 研究员 中科院计算所

16:50～17:10 主题演讲：高性能计算技术在地球物理中的应用  
　 王宏琳 教授 中国石油集团咨询中心工程技术组专家 东方地球物理公司顾问

17:10～17:30 主题演讲：高性能计算在动漫产业中的应用
　 叶向宇 高级顾问 信息产业部软件与集成电路促进中心

17:30～17:40 主题演讲：澳汰尔公司高性能计算解决方案创新

17:40～17:45 抽奖

17:45 会议结束
]]> http://www.caeda.com.cn/blog/article.asp?id=57 caeda@vip.sina.com(admin) Tue,03 Jun 2008 13:25:56 +0800 http://www.caeda.com.cn/blog/default.asp?id=57
1.算法评价
IOSO致力于多目标优化，求解均匀分布的Pareto最优解，最优解个数可以由用户指定。传统上用加权和法求解多目标优化问题容易丢失一些最优解，无法获得可靠的Pareto最优解分布；而IOSO可以获得较为全面的Pereto最优解。

2.操作评价
操作流程与其他多目标优化软件如iSight类似，由输入文件——分析程序——输出文件三部分构成。操作简单，入门容易。
设置变量类型、范围，约束，多个目标函数都很方便。
算法的设置也比较简单，容易掌握。

3.功能评价
优点：
    可通过参数试验检查优化问题定义和设置是否正确
    可由用户指定求解多少个Pereto最优解
    可简单进行重分析，前面的分析结果作为重分析的起点
    具有多种停止准则，且设置简单
    可实时画出Pereto前端曲线
缺点：
    没有实验设计功能（DOE）
    算法只有一种，无法让用户根据问题选择算法
    没有响应面模型功能
    操作过程反应有些慢
]]> http://www.caeda.com.cn/blog/article.asp?id=56 caeda@vip.sina.com(admin) Sun,01 Jun 2008 08:41:59 +0800 http://www.caeda.com.cn/blog/default.asp?id=56
ESI Group's CFD Division is proud to announce that the Spring 2008 (V2008.2) release is complete and ready for delivery. The fact that this release comes just six months after our last major release proves that ESI Group is working hard to implement the features that the market demands in a timely manner. We welcome and encourage all of our customers to download and use the latest versions which include numerous enhancements.

Included in the Spring 2008 (V2008.2) release are significant updates to:

CFD-ACE+ Advanced CFD and Multiphysics Solver
CFD-FASTRAN CFD Physics Solver
CFD-CADalyzer Upfront CFD Design Tool

CFD-GEOM Geometry Modeler and Mesh Generator
CFD-VisCART Cartesian Mesh Generator
CFD-VIEW Solution Analysis and Visualization Tool
Enhancements have been made to increase usability, robustness, and accuracy and we are sure that these new versions will make for a worthwhile upgrade to your CFD toolset.

The remainder of this note gives you some information about the V2008.2 software release. Please read on to learn more about the updates and let us know if you have any comments or questions.

IMPORTANT NOTE: If you are running V2007.x or earlier software version, then you may require a new license to run V2008.x versions! If you are currently running V2008.0 then you will NOT need a new license to run V2008.2. ESI Group licensing policy requires licensing to the version of the software. This means that if you currently have a V2007.x (or earlier) license file you will need to replace it with a V2008.x license file. These are available upon request from your ESI license or sales representative. If you are not sure who that is you should contact support.cfd@esi-group.com please include your existing V2007.x license file for the quickest service.

Spring 2008 (V2008.2) Release Highlights
As with previous releases, there are many improvements to features, performance, and robustness. Below is a short bullet list of some of the top reasons why you should consider upgrading to V2008.2. For a full description of the new and improved features, as well as usage and limitation information, please consult the relevant release note document which is included in the software package download files and can also be found here (login required).

CFD-ACE V2008.2 (view full release notes - login required)
Fast Time Stepping (FTS)
Fast Time Stepping (FTS) is a new feature to speedup transient simulations. FTS is currently based on the Fractional Step Method (FSM). FTS cuts down the simulation time compared to the standard timestepping by a factor of about 5 for simple flow problems, and even more for practical (industrial) flows.

Spray Flash Vaporization
A new model has been implemented to accurately predict the evaporation of spray droplets injected under superheated state.

Copy/Paste BC/VC Records
Copy and paste boundary condition and volume condition information as well as initial conditions, spray injectors, fans, momentum resistances, etc. Can be copied within the current model or even from one model to another.


BC Integral Output
BC Integral Output ("Out->Print->BC Integral Output") setup dialog automatically detects all unique boundary names and allows you to pick from a list.

Journaling for Database Manager
All the actions you perform in the Database Manager are now recorded in the Python script. You can see the script by clicking on "Tools->Edit Python Script" menu. Journaling must be turned on in order to use this feature.

CFD-FASTRAN V2008.2 (view full release notes - login required)
Incorporation of PETSc Solvers
The use of the PETSc cgs and gmres linear equation solvers from Argonne National Laboratories has been added in this release. These solvers enhance stability for convergence performance.

CFD-CADalyzer V2008.2 (view full release notes - login required)
Support for Pro/E Wildfire 4
CFD-CADalyzer now has support to work with Pro/E Wildfire 4 installations.

CFD-GEOM V2008.2 (view full release notes - login required)
Model Inversion
Given a set of "capping" surfaces, this tool will create the internal fluid domain from the unstructured domains in the current model. Optionally, the original solids can be retained.

Model Splitting
Split the current model at a specified plane. All unstructured domains intersecting the plane are split and those that do not are left unaltered. Optionally, the desired portion of the model (inside or outside or both) can be retained. The tool automatically builds new surfaces along the intersecting plane to close off any new domains.

BC/VC Grouping
BC/VC grouping has been added to facilitate DTF grouping of related BC/VC entities. This is primarily useful when the grouped entities are processed in other applications such as CFD-ACE-GUI as it minimizes the number of BC/VC records that the you typically see and makes the model setup a much easier process.

Python Macros
Allows the execution of Python scripts on the current model. This allows you to add your own custom capabilities to CFD-GEOM and to use those capabilities interactively from the CFD-GEOM GUI.


Export All Surface Meshes to STL
2D structured blocks can now be exported to the STL format. The quadrilateral typically present in such a mesh are automatically split into triangles.

Discrete Mesh Modeling
As a first step to allowing users to work with geometries only defined by meshes such as STL, NASTRAN, PATRAN and other mesh formats, a discrete modeling system has been implemented. The system currently allows eleven different operations including, merging/joining, outline/critical feature extraction, intersections, and boolean operations.

Additional CAD File Import Options
The following CAD file formats have been added: CATIA V4, CATIA V5, SolidWorks, Parasolid, Unigraphics, PRO/E and STEP. All these formats are supported on 32 and 64 bit Windows based platforms while 32 bit Linux platforms only support Parasolid, Unigraphics, PRO/E and STEP. The importers are not available on other Unix or 64 bit Linux platforms. These importers are licensed separately.

CFD-VisCART V2008.2 (view full release notes - login required)
Viscous Boundary Layer Support (BETA)
The viscous boundary layer generation function is brought back with a completely different algorithm. It is added as a post processing step and you have choices to generate boundary layer for required domains and/or patches. The current version handles relative simple geometries. This should be considered BETA capability and we welcome all feedback.

Additional CAD File Import Options
The following CAD file formats have been added: CATIA V4, CATIA V5, SolidWorks, Parasolid, Unigraphics, PRO/E and STEP. All these formats are supported on 32 and 64 bit Windows based platforms while 32 bit Linux platforms only support Parasolid, Unigraphics, PRO/E and STEP. The importers are not available on other Unix or 64 bit Linux platforms. These importers are licensed separately.

Improved ShrinkWrap Mesher
Much better feature capture mechanism and mesh smoothing are added into the ShrinkWrap mesher. Features are better preserved and surface mesh quality is better.

Improved Single-Domain Mesher
The same feature capture and mesh smoothing methods (as used for shrinkwrap mesher) are added into the single-domain mesher. You will see better feature preservation and a smoother surface mesh when you activate the feature preservation mechanism. Note, this algorithm will take longer time to run and final volume mesh may have bad cells which cannot be removed.

Improved Mesh Improvement Method
The new method can remove bad cells more efficiently than before. Many previously irremovable bad cells can now be removed.

Local Cell-size Control per-Domain
You now have control of the local cell size per domain. Each domain can be given a local cell size which overwrites the global cell size control if the local cell size is smaller than global one.

Wind Tunnel Tool
For simulation efforts with a fixed wind tunnel configuration, this tool provides a quick way to setup the outside bounding box to match the wind tunnel configuration.

Wireframe Display of Geometry
For patches with millions of triangle faces, this feature will speedup the drawing of geometries and make easier for you to do graphic manipulations (rotate/pan/zoom).

CFD-VIEW V2008.2 (view full release notes - login required)
Headless Operation (No GUI Display)
CFD-VIEW now supports "headless" operation (no GUI display) for batch processing with Python scripts. This also gives the ability to run CFD-VIEW on server systems (nodes which do not have a display).

Scripting Improvements
CFD-VIEW's scripting system has been greatly improved and simplified, while maintaining compatibility with existing user scripts. Error reporting from Python is much more clear, as well as better function argument checking in CFD-VIEW. Many new simplified functions have been added, with many more to come. The new system also allows newer functions to be added easily, so user requests for additional functionality can be addressed much quicker by ESI.

New Calculator Functions
New calculator functions have been added, with more to come. New for this release are Average, Line Integral, and Mass Averaged functions. User requests for new functions are always welcome.

Font Updates
Font choices have been improved for the Legend, origin, and Axes. All system fonts are now available.

Improved Animation Recording
Animation recording has been greatly simplified. Now, when the record button is clicked on the animation panel, an improved and simplified record dialog is presented, and the animation is then automatically launched and saved. Multiple procedures are not required with the new one step system.
Text Input Improvements
Numerical data entry into text fields has been improved in areas such as the 3D Viewer orientation settings.

V2008.2 Installation
The V2008.2 software will be installed parallel to your existing V2008.0 (or earlier) installation. This means that you will have access to both versions should you desire. However, only V2008.2 will be supported going forward so you should use V2008.2 to get access to future bug fixes and feature improvements. Step-by-step instructions for the installation of V2008.2 can be found here (login required).
咨询或试用联系：caeda@vip.sina.com
]]> http://www.caeda.com.cn/blog/article.asp?id=55 caeda@vip.sina.com(lns) Wed,28 May 2008 08:02:07 +0800 http://www.caeda.com.cn/blog/default.asp?id=55 FE-DESIGN GmbH, Haid-und-Neu-Str. 7, D-76131 Karlsruhe

CAEDA Ltd.
LIU Ning Sheng, Director
Registered address: Unit C. 17/F.
Silvercorp Int’l Tower, 713 Nathan Road, Kowloon, Hong Kong
and
Marketing Center Address: Room #3-0305, Guangda Garden, No.2 Quanzong Road,
Haidian District, Beijing China, 100089

May 22, 2008
Subject:
Evaluation and Implementation of Business Partner for FE-DESIGN in China

Dear Mr. LIU,

FE-DESIGN hereby authorizes CAEDA Ltd., represented by the director Mr. Liu, to act on behalf of FE-DESIGN in China for the following activities:

    to evaluate and select appropriate business partners in China which could be capable to sell TOSCA in this territory

    to propose these business partners to FE-DESIGN as qualified candidates

    to prepare the distribution agreements between the business partner and FE-DESIGN which finally has to be signed between business partner and FE-DESIGN directly

    to provide technical support and training on the software to business partner to the extent reasonably required by FE-DESIGN

This authorization is part of the distribution & promotion agreement between CAEDA Ltd. And FE-DESIGN, signed on May 19th 2008.


Dr. Jürgen Sauter
Managin Director
FE.DESIGN GmbH
]]> http://www.caeda.com.cn/blog/article.asp?id=54 caeda@vip.sina.com(admin) Sun,25 May 2008 08:38:31 +0800 http://www.caeda.com.cn/blog/default.asp?id=54
With the new CAD model import options in CFD-GEOM, it is now easier than ever to read your solid models into CFD-GEOM. Once the model is in CFD-GEOM, you have the solid representation of the part but in many cases the fluid, or negative space, in the CAD model is of importance and the area which needs to be discretized (i.e. meshed). In V2008.2, CFD-GEOM can now "invert" these models to give you just the fluid region or the fluid region plus the solid part with a single click. This new tool is called "Invert Model" and can be found on the Geometry tab under the Solid and Wireframe Tools, as shown in Figure 1.



Figure 1. Invert Model tool in CFD-GEOM V2008.2

When doing Model Inversion, there a few things to keep in mind:

Available for inverting Unstructured Domain(s) not constructed from Edges, Faces, nor Discreet Surfaces
The domain(s) must have defined cap-able inlet/outlet holes.
Except at the caps, the domain(s) must have a finite thickness throughout, i.e. it must be a solid (Unstructured Domain in CFD-GEOM).
Model inversion is not part of the Undo system, so be sure to save the model before inversion.
Ideally, inversion should be done immediately after the model has been read into CFD-GEOM.

Option to discard or keep the original solid.
To perform a Model Inversion, do the following steps:

Read the CAD model into CFD-GEOM
Identify the cap-able inlet and outlet surfaces in the model
Go the the Invert Model tool and select the inlet and outlet surfaces that cap the model.
Click the Apply button to invert the model.
Note that as a part of the process, untrimmed surfaces are created when performing a model inversion. These can be deleted or blanked from the model since they are not needed. In future versions of CFD-GEOM, these untrimmed surfaces may be deleted as a part of the model inversion tool.

To view a short movie demonstrating the inversion capability, please click here. To download the model used in the tip, please click here.

This is one of the many new features included in the V2008.2 Beta release, which can be downloaded from the Software Download section of the ESI CFD Customer Support Portal.

If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know.
]]> http://www.caeda.com.cn/blog/article.asp?id=53 caeda@vip.sina.com(admin) Fri,23 May 2008 09:31:15 +0800 http://www.caeda.com.cn/blog/default.asp?id=53
以下CFD专业设计分析类软件，欢迎有兴趣试用的专业人员和我们联系，我们提供短期免费试用：caeda@vip.sina.com, www.caeda.com.cn

CFD-Fastran 先进的空气动力学和热力学应用软件

软件背景:法国ESI Group公司出品的CFD-Fastran是先进的CFD软件，代表了目前CFD方面各种最新的技术 ，他的原开发商是美国CFDRC公司，多年同美国国家航空和宇宙航行局（NACA），美国能源部（DOE），美国国家科学基金会（NSF）等机构合作，使CFD-Fastran不断得到完善和发展。
软件特性
•基于密度的 CFD 求解器，适合从低速到 高超音速 范围内的流动、热交换及非平衡化学反应计算。
•包含多种精确的空间求解格式及高阶限制器。
•功能强大的 动网格 和 重叠网格 技术。
•简单易用的 6DOF 和精确运动模型。
软件优势：专门针对航空航天应用而设计，先进的chimera方法使复杂几何建模更容易，本身包含结构力学求解器，能够解决气动弹性和流固耦合等复杂问题

CFD-ACE+多学科工程解决方案

CFD-ACE+结合了多面体混和单元，结构化及非结构化网格，适用性更广泛，是引领多学科工程仿真发展方向的高级CFD软件。
基本特点
•采用了最新改进的GUI界面，建模时具有无法比拟的灵活性。
•通过拖曳菜单，可直接应用数据库来设置化学反应，物质属性以及边界条件。
•适用于结构化网格以及任意多面体混合单元的非结构化网格，单元表面可匹配任意的计算域交界面（静止，滑动，旋转或移动）。
•计算过程中具有网格自适应功能。
•处理多学科问题时无法比拟的灵活性。
•包含了碰撞外壳模型以计算偏振晶片上所有离子核素离子能量分布。

CFD-ACE流场仿真：
•可压/不可压流场
•稳态/瞬态流场
•层流，并提供非牛顿的流变学作为选择
•湍流，提供众多的湍流模型
•带表面张力的VOF自由表面模型
•随外形及时间变化的边界条件

CFD-ACE几何建模与划分网格：
CFD-ACE充分集成了CFD-GEOM,既可以独立完成几何建模、划分网格以及设置边界条件等功能，也可以从其它CAD/CAE系统中读入这些数据，具体的功能包括：
•结构化和任意多面体（非结构化）单元网格
•可用单元表面来匹配求解区域的界面，从而增加了建模的灵活性与方便性
•通过CFD-ADAPTER有针对性地进行求解，增强仿真的真实性，并提高了求解速度
•滑移的网格边界与任意单元表面的匹配功能
•可设置旋转坐标系，移动网格及壁面
•可仿真流场内部带有固体的情况，求解集中或分散的阻抗（固体和带孔的平板、复合平板、多孔介质）
CFD-ACE热及质量的传输模型：
•自然对流和强迫对流
•多介质，共轭传热
•非灰体辐射传热（离散坐标方法）
•面对面的辐射
•蒙特卡罗辐射模型
•有多元扩散、Soret项及Stefan-Maxell选项的传质
•可指定温度、热流、换热系数及绝热的边界条件

CFD-ACE化学反应和燃烧仿真：
•有限体积，基于压力的强守恒方程
•为提高精确性，对任意交界面及悬挂节点（无插值）采用完全隐式的处理方法
•SIMPLE/PISO数值方法
•代数求解器采用共轭（CGS）和多重网格法（Algebraic Multi-Grid,AMG）
•可采用MPI并行计算

CFD-ACE+适用的平台：
•Windows NT (Intel ,DEC)
•LINUX
•Unix Workstations (SGI , IBM , HP , SUN, DEC)

CFD-ACE+附加的模块：
•CFD-PLASMA：等离子体化学及等离子体增强分解/蚀刻的仿真模块
•CFD-FEMSTRESS：流体结构与热耦合情形的仿真模块；用于静电学、电磁学、电镀、电泳及电渗的电物理模型
•CFD-FILM：薄膜形态及表面形状仿真模块
•CHEMKIN III：用于延伸化学的模块

CFD-ACE+作为CFD-ACE+包的一部分，与VisCART，CFD-GEOM和CFD-VIEW一起提供。
CFD-ACE+与其它ESI CFD Inc 求解器一起可以进行多学科的仿真分析。

ESI公司是世界工业产品数值模拟研究以及软件研发的杰出代表。它创建于1988年，总部位于法国巴黎，拥有强大的研发队伍，其产品已经得到世界范围内的认可。全球使用ESI公司软件和服务的有六百多个国际性的集团公司，如洛克希德-马丁，波音, Caterpillar, 克勒斯勒,福特汽车, Ballard Power,通用, 宝马,霍尼维尔, 惠普, IBM, 英特尔等。详细咨询，请联系中国授权代理商：www.caeda.com.cn

]]> http://www.caeda.com.cn/blog/article.asp?id=52 caeda@vip.sina.com(admin) Tue,20 May 2008 10:02:37 +0800 http://www.caeda.com.cn/blog/default.asp?id=52
1. Laminate Tools专业复合材料设计平台

Laminate Tools是英国Anaglyph软件公司的主打产品，它是专业的复合材料设计平台。 Laminate Tools集成了设计(Design)-分析(Analysis)-校验(Check)-制造(Manufacture)——结构设计的全部过程。它为复合材料设计－分析－制造建立了桥梁。

功能特性
兼容大多数CAD系统曲面定义，进行材料铺覆模拟。
可输入多种有限元软件生产的网格文件。
能够转化铺层信息，使得适应有限元分析。
可对有限元结果进行分析。
可输出便于制造加工的铺层信息。


2. LAP层合板设计分析软件

LAP可应用于任何类型的复合材料层板面内载荷分析。LAP自1988年诞生以来已经被世界20多个国家的设计师所应用，Airbus公司更是将LAP作为室内复合材料层板的标准分析工具。

功能特性
拥有Basic 、Non-Linear、Design、Additional Failure Criteria四个模块。
具有友好的图形界面，支持鼠标托拽功能。
支持固化应力、热应力、机械应力等线性载荷，支持分段函数定义应力和应变。
可定义纤维排布方向，可对铺层进行对称操作。
支持多种失效准则，可自定义失效准则。
软件优势
简化的分析能够提供软件的通用性，强大的解算器使计算过程快速高效。
数据操作和结果显示的图形化使分析过程一幕了然。

3. CoDA复合材料部件设计分析软件

CoDA是复合材料平板、梁、接头、法兰和层板等部件的初步设计辅助工具。

功能特性
整合了Panels、Beams、Laminates、Joints、Flanges、PREDICT、Materials Synthesise7个模块。
支持从简单到复杂的各种几何模型。
可同时打开多个数据文件。
多种计算结果同步显示。
支持鼠标托拽功能。
软件优势
多个模块整合在一起可以及时方便的考察输入参数或铺层参数对结果的影响。
软件简单易学。
]]> http://www.caeda.com.cn/blog/article.asp?id=51 caeda@vip.sina.com(admin) Tue,20 May 2008 09:46:26 +0800 http://www.caeda.com.cn/blog/default.asp?id=51
LAP consists of four modules. The Basic module includes the linear analysis functionality and constitutes the minimum configuration. The other three modules (Non-Linear, Design, Additional Failure Criteria) are optional.

Basic module  
Non-Linear module  
Design module
Additional Failure Criteria module  

User Interface features
The graphical user interface includes all the features expected of a professional, state-of-the-art, technical software application.  

Any type of results can be displayed simultaneously on any number of independent windows.

Interfaces for data quantities can be used to import or export information and communicate with other applications. Interfaces for results can be used to export results to files or to the Windows clipboard for instant transfer to spreadsheets, word processors, or custom application software.  

Generic reports can be defined to obtain professional quality printed output of text and graphics efficiently. The contents of any window can be printed instantly in isolation.  
Drag-and-drop copy operations between LAP files can be used for any database object, such as a material, lay-up, etc.  

Multi-level Undo and Redo functions, as well as a timed Auto-Save feature ensure data safety.  

A choice of working units is available at any stage, for both input and output.  
Quick interactive solutions and results updates save you time.  
Application-wide as well as file-specific user preferences offer a consistent working environment.  

The desktop layout of the various working windows can be saved for easy return to a particular problem, even when working with computer screens set at different resolutions.  
A convenient custom screen font is used to display standard engineering symbols.  
Groups of material properties, stacking sequences, loading conditions and supporting information are saved in database files and restored as needed.  
There are no limits to the database size.  
Multiple database files can be open simultaneously.  
The supplied printed documentation explains how to use the software, offers useful tips and covers all theoretical background.  
Help is available in the familiar Windows hypertext format. ]]> http://www.caeda.com.cn/blog/article.asp?id=50 caeda@vip.sina.com(admin) Sun,18 May 2008 10:43:15 +0800 http://www.caeda.com.cn/blog/default.asp?id=50
Many times post processing results requires computing average quantities of variables over a surface or volume. CFD-VIEW V2008.2 made it easy by adding three new calculator functions under the ‘Miscellaneous’ function category as shown in Figure 1.

average
vaverage
aaverage

Figure 1. New averaging functions added to the CFD-VIEW calculator.


average: Nodal average function

Usage: average(variable_name)
This can be used on any nodal based elements (line, curve, surfaces or volume)
aaverage: Area weighted average function

Usage: aaverage(variable_name)
This can be used on all surfaces
vaverage: Volume weighted average function

Usage: vaverage(variable_name)
This can be used on all volumes
A simple example would be a pipe where we would like to calculate the average velocity or pressure at a particular position downstream from the inlet. To do so, use the following steps:

Create a cut (X, Y, or Z) and position it in the desired location in the geometry, as shown in Figure 2.
Go to the calculator tab on the Attribute tool bar, as shown in Figure 3.
In the Expression window, right click and select "average" from the Miscellaneous tools
Enter the variable (by typing or selecting from the Variables list under right click in the Expression window) for which you would like to get the average value.
Click the calculate button, as highlighted in Figure 3.


Figure 2. Cut through pipe geometry where we will calculate the average pressure.



Figure 3. Average pressure usage in CFD-VIEW calculation.

This is one of the many new features included in the V2008.2 Beta release, which can be downloaded from the Software Download section of the ESI CFD Customer Support Portal.

Regards,
caeda@vip.sina.com

]]> http://www.caeda.com.cn/blog/article.asp?id=49 caeda@vip.sina.com(admin) Fri,16 May 2008 08:23:50 +0800 http://www.caeda.com.cn/blog/default.asp?id=49       Sculptor是美国Optimal Solutions公司基于多年的计算流体力学方面的研究和工作经验，研发出的致力于解决自动化网格变形的应用软件。它能够导入相应的CFD软件的网格文件，如Fluent, Star-CD, Plot3d, Nastran等，按照工程师的要求，对其网格进行变形操作，并能够实时的观察变形的效果。        Sculptor的网格变形技术和CFD软件的结合使用已经经历了很多年的时间，并且由于这一技术的引用进一步开发了CFD的潜力。

    Sculptor使传统的设计过程简单化，在原有模型的基础上进行局部形变，并实现光滑过渡，且可以检查变形后几何体之间是否有干涉问题，大大节省了工程师在需要改变物体外形时重新构建CAD几何体和重新划分网格所需要的时间。

     Sculptor内部嵌入了基于梯度的优化算法，因此它能够根据工程师制定的优化任务，对设计问题进行优化，帮助工程师找到满足设计需求的解决办法。


      Sculptor采用批处理模式，因此它也可以很方便的被其他优化软件集成，从而应用优化软件中的优化算法对设计问题进行优化。为了实现网格变形，Optimal Solutions公司研发了ASD(Arbitrary Shape Deformation)技术，在原有的CFD网格节点外面加一个控制域作为Sculptor的变量，通过对控制域上控制点的位置改变，简单方便的达到网格变形的目的。控制域的创建方法很简单，如下图所示。点击一个Shrink命令后，控制域就将原有的控制体包裹上了。


通过移动控制点，可以实现网格变形，并且保证几何体的光滑过渡。


变形之后得到的仍然是可以被CFD软件读取的文件，设计者可以将其导入到相应的分析软件中进行分析。同时，也可以通过Sculptor调用相应的CFD软件进行分析，如果该设计不能满足要求，再用Sculptor进行网格变形，直至达到满意的设计效果。
         
•    汽车空调管路的压力损失及噪声最小化
•    飞行器阻力最小化及提升力最大化
•    工业上流体流动效率最大化
•    热交换器的传热效率最大化，对模型进行变形，而这个过程无需合并文件
•    燃烧设备中有害化学物质排除最小化
•    化工反应器混合效率最大化

     总之，Sculptor帮助工程师解决基于网格的模型的形状优化问题。尤其对于CFD工程师，大大节省了修改模型设计时重新构建几何体和重新划分网格的时间。并且由于优化方法的引入，提升了CFD的潜力，避免了依赖工程师设计经验来探索几何形状的盲目性，帮助工程师找到符合设计需要的更加有效的解决方法。
         
         
如需更多产品资料请下载产品资料索取



]]> http://www.caeda.com.cn/blog/article.asp?id=48 caeda@vip.sina.com(admin) Fri,09 May 2008 15:09:50 +0800 http://www.caeda.com.cn/blog/default.asp?id=48

    为在全国范围的国家科研单位、高等院校及企业开展高性能计算培训与应用推广，培养高性能计算专业人才、推广高性能计算应用，由中国计算机学会高性能计算专业委员会、中科院计算机网络信息中心、北京大学、中科院数学与系统科学研究院、北京应用物理与计算数学研究所、中美联合科学计算中心、北京泰瑞世纪科技有限公司共同联合举办2008年高性能计算培训班。

    高性能计算培训班自2001年开始，每年举办一期，已经连续成功举办七届，培训学员近500人次。为了更好的普及、提高国内高性能计算水平，根据去年的培训效果调查结果以及学员们的意见和建议，本着"理论指导实践，学以致用、加大中外交流"的原则，利用国家"973"项目"大规模科学计算研究"以及北京其他高性能计算中心提供的高性能计算环境，今年将以往的一个班分拆成普及班和提高班，特邀业界高性能计算知名专家张林波、莫则尧、张云泉、孙毓忠、迟学斌等担任主讲教师（详见老师简介），为学员们带来国内高性能计算培训的饕餮盛宴。

培训对象：

    具备高性能计算和Linux基础知识，已经从事与高性能计算应用相关的科研院所及高校老师、硕博研究生及相关企业工程师，或对高性能计算感兴趣的相关人员。

培训时间：

    2008年6月16日－6月21日，报名截止日期：2008年6月11日，名额有限，报名从速。


培训地点：

    中科院计算机网络信息中心514房间
    详细地址：北京市海淀区中关村南四街四号中科院软件园区2号楼
    地图：http://www.sccas.cn/gb/other/images/situation_02.jpg

培训费用：

高性能计算培训普及班（编号：01）：￥1000元/人（科研教育单位￥600元/人），包括教材费、培训费、上机费和午餐费，其它费用全部自理；
高性能计算培训提高班（编号：02）：￥1800元/人（科研教育单位￥900元/人），包括教材费、培训费、上机费和午餐费，其它费用全部自理；
组合班（普及班＋提高班）（编号：03）：￥2500元/人（科研教育单位￥1200元/人），包括教材费、培训费、上机费和午餐费，其它费用全部自理。
    注：允许同一单位2人共同报名组合班，一人参加普及班，另一人参加提高班

报名方式： (回执表下载)

    按照报名回执表的要求填写各项内容，然后发邮件到sajccs@sccas.cn ，我们将在一个工作日之内给予回复。
    联系人：刘利萍，联系电话：010-82133799-612

付款信息：

电汇（重点推荐）：截止日期2008年6月13日（含）。凭汇款单在签到时领取发票（发票项目：培训费），请在回执中填写正确的发票单位名称。
公司全称：北京泰瑞世纪科技有限公司
开户行名称：中国工商银行北京市海淀支行营业部
账号：0200049609200235564
现场付款：签到当天可以直接用现金或支票付款，并开具发票。
培训内容（如有变更，恕不再通知，以实际培训为准）：



点击图片看大图

培训受益：

与HPC领域知名专家和科研院所、企业等一起分享高性能计算相关理论和实践中的经验和心得，提供一个公共的交流平台，提升单位的人才素质和科研能力。
免费在中国国家网格主结点－深腾6800四万亿次超级服务器进行上机操作练习，做到理论和实践相结合，保证培训效果。
凡是参加培训的学员有机会参观中科院超级计算中心机房；
考试合格的将由中科院计算机网络信息中心颁发相应的证书；
参加培训的学员将可能推荐到相关单位实习或工作。
附：

    老师介绍（点击下载）

    交通住宿（点击下载）
]]> http://www.caeda.com.cn/blog/article.asp?id=47 caeda@vip.sina.com(admin) Tue,06 May 2008 08:33:54 +0800 http://www.caeda.com.cn/blog/default.asp?id=47
ESI Group's CFD Division is proud to announce that the Spring 2008 release is on schedule and now available for BETA testing. The final version is expected to be released on June 1st, 2008. The fact that this release comes just six months after our last major release proves that ESI Group is working hard to implement the features that the market demands in a timely manner. We welcome and encourage all of our customers to download and use the latest versions which include numerous enhancements.

Included in the Spring 2008 release are significant updates to:

CFD-ACE+ Advanced CFD and Multiphysics Solver
CFD-FASTRAN CFD Physics Solver
CFD-CADalyzer Upfront CFD Design Tool

CFD-GEOM Geometry Modeler and Mesh Generator
CFD-VisCART Cartesian Mesh Generator
CFD-VIEW Solution Analysis and Visualization Tool
Enhancements have been made to increase usability, robustness, and accuracy and we are sure that these new versions will make for a worthwhile upgrade to your CFD toolset.

The remainder of this note gives you some information about the Spring 2008 software release. Please read on to learn more about the updates and let us know if you have any comments or questions.

IMPORTANT NOTE: If you are running V2007.x or earlier software version, then you may require a new license to run V2008.x versions! If you are currently running V2008.0 then you will NOT need a new license to run V2008.2. ESI Group licensing policy requires licensing to the version of the software. This means that if you currently have a V2007.x (or earlier) license file you will need to replace it with a V2008.x license file. These are available upon request from your ESI license or sales representative. If you are not sure who that is you should contact support.cfd@esi-group.com please include your existing V2007.x license file for the quickest service.

Spring 2008 (V2008.2) Release Highlights
As with previous releases, there are many improvements to features, performance, and robustness. Below is a short bullet list of some of the top reasons why you should consider upgrading to V2008.2. For a full description of the new and improved features, as well as usage and limitation information, please consult the relevant release note document which is included in the software package download files and can also be found here (login required).

CFD-ACE V2008.2 (view full release notes - login required)
Fast Time Stepping (FTS)
Fast Time Stepping (FTS) is a new feature to speedup transient simulations. FTS is currently based on the Fractional Step Method (FSM). FTS cuts down the simulation time compared to the standard timestepping by a factor of about 5 for simple flow problems, and even more for practical (industrial) flows.

Spray Flash Vaporization
A new model has been implemented to accurately predict the evaporation of spray droplets injected under superheated state.

Copy/Paste BC/VC Records
Copy and paste boundary condition and volume condition information as well as initial conditions, spray injectors, fans, momentum resistances, etc. Can be copied within the current model or even from one model to another.


BC Integral Output
BC Integral Output ("Out->Print->BC Integral Output") setup dialog automatically detects all unique boundary names and allows you to pick from a list.

Journaling for Database Manager
All the actions you perform in the Database Manager are now recorded in the Python script. You can see the script by clicking on "Tools->Edit Python Script" menu. Journaling must be turned on in order to use this feature.

CFD-FASTRAN V2008.2 (view full release notes - login required)
Incorporation of PETSc Solvers
The use of the PETSc cgs and gmres linear equation solvers from Argonne National Laboratories has been added in this release. These solvers enhance stability for convergence performance.

CFD-CADalyzer V2008.2 (view full release notes - login required)
Support for Pro/E Wildfire 4
CFD-CADalyzer now has support to work with Pro/E Wildfire 4 installations.

CFD-GEOM V2008.2 (view full release notes - login required)
Model Inversion
Given a set of "capping" surfaces, this tool will create the internal fluid domain from the unstructured domains in the current model. Optionally, the original solids can be retained.

Model Splitting
Split the current model at a specified plane. All unstructured domains intersecting the plane are split and those that do not are left unaltered. Optionally, the desired portion of the model (inside or outside or both) can be retained. The tool automatically builds new surfaces along the intersecting plane to close off any new domains.

BC/VC Grouping
BC/VC grouping has been added to facilitate DTF grouping of related BC/VC entities. This is primarily useful when the grouped entities are processed in other applications such as CFD-ACE-GUI as it minimizes the number of BC/VC records that the you typically see and makes the model setup a much easier process.

Python Macros
Allows the execution of Python scripts on the current model. This allows you to add your own custom capabilities to CFD-GEOM and to use those capabilities interactively from the CFD-GEOM GUI.


Export All Surface Meshes to STL
2D structured blocks can now be exported to the STL format. The quadrilateral typically present in such a mesh are automatically split into triangles.

Discrete Mesh Modeling
As a first step to allowing users to work with geometries only defined by meshes such as STL, NASTRAN, PATRAN and other mesh formats, a discrete modeling system has been implemented. The system currently allows eleven different operations including, merging/joining, outline/critical feature extraction, intersections, and boolean operations.

Additional CAD File Import Options
The following CAD file formats have been added: CATIA V4, CATIA V5, SolidWorks, Parasolid, Unigraphics, PRO/E and STEP. All these formats are supported on 32 and 64 bit Windows based platforms while 32 bit Linux platforms only support Parasolid, Unigraphics, PRO/E and STEP. The importers are not available on other Unix or 64 bit Linux platforms. These importers are licensed separately.

CFD-VisCART V2008.2 (view full release notes - login required)
Viscous Boundary Layer Support (BETA)
The viscous boundary layer generation function is brought back with a completely different algorithm. It is added as a post processing step and you have choices to generate boundary layer for required domains and/or patches. The current version handles relative simple geometries. This should be considered BETA capability and we welcome all feedback.

Additional CAD File Import Options
The following CAD file formats have been added: CATIA V4, CATIA V5, SolidWorks, Parasolid, Unigraphics, PRO/E and STEP. All these formats are supported on 32 and 64 bit Windows based platforms while 32 bit Linux platforms only support Parasolid, Unigraphics, PRO/E and STEP. The importers are not available on other Unix or 64 bit Linux platforms. These importers are licensed separately.

Improved ShrinkWrap Mesher
Much better feature capture mechanism and mesh smoothing are added into the ShrinkWrap mesher. Features are better preserved and surface mesh quality is better.

Improved Single-Domain Mesher
The same feature capture and mesh smoothing methods (as used for shrinkwrap mesher) are added into the single-domain mesher. You will see better feature preservation and a smoother surface mesh when you activate the feature preservation mechanism. Note, this algorithm will take longer time to run and final volume mesh may have bad cells which cannot be removed.

Improved Mesh Improvement Method
The new method can remove bad cells more efficiently than before. Many previously irremovable bad cells can now be removed.

Local Cell-size Control per-Domain
You now have control of the local cell size per domain. Each domain can be given a local cell size which overwrites the global cell size control if the local cell size is smaller than global one.

Wind Tunnel Tool
For simulation efforts with a fixed wind tunnel configuration, this tool provides a quick way to setup the outside bounding box to match the wind tunnel configuration.

Wireframe Display of Geometry
For patches with millions of triangle faces, this feature will speedup the drawing of geometries and make easier for you to do graphic manipulations (rotate/pan/zoom).

CFD-VIEW V2008.2 (view full release notes - login required)
Headless Operation (No GUI Display)
CFD-VIEW now supports "headless" operation (no GUI display) for batch processing with Python scripts. This also gives the ability to run CFD-VIEW on server systems (nodes which do not have a display).

Scripting Improvements
CFD-VIEW's scripting system has been greatly improved and simplified, while maintaining compatibility with existing user scripts. Error reporting from Python is much more clear, as well as better function argument checking in CFD-VIEW. Many new simplified functions have been added, with many more to come. The new system also allows newer functions to be added easily, so user requests for additional functionality can be addressed much quicker by ESI.

New Calculator Functions
New calculator functions have been added, with more to come. New for this release are Average, Line Integral, and Mass Averaged functions. User requests for new functions are always welcome.

Font Updates
Font choices have been improved for the Legend, origin, and Axes. All system fonts are now available.

Improved Animation Recording
Animation recording has been greatly simplified. Now, when the record button is clicked on the animation panel, an improved and simplified record dialog is presented, and the animation is then automatically launched and saved. Multiple procedures are not required with the new one step system.
Text Input Improvements
Numerical data entry into text fields has been improved in areas such as the 3D Viewer orientation settings.

V2008.2 Installation
The V2008.2 software will be installed parallel to your existing V2008.0 (or earlier) installation. This means that you will have access to both versions should you desire. However, only V2008.2 will be supported going forward so you should use V2008.2 to get access to future bug fixes and feature improvements. Step-by-step instructions for the installation of V2008.2 can be found here
]]> http://www.caeda.com.cn/blog/article.asp?id=46 caeda@vip.sina.com(admin) Sun,04 May 2008 15:37:49 +0800 http://www.caeda.com.cn/blog/default.asp?id=46
Software program

CAE system for the geometric modeling, numerical simulation, systematic variation and formal optimization in ship design

The FRIENDSHIP-Framework is an integration environment for the advanced hydrodynamic design of ships. It facilitates the work flow at the crucial stage of hull form development by closely coupling CAD capability to CFD analysis. The system supports the consistent generation and assessment of variants and offers integrated constraint management. It can be flexibly tailored to the naval architects’ individual design environment, increasing productivity and improving competitiveness.

Features

•Efficient shape variation via partially parametric modeling on data being imported from offsets, panel meshes or iges-files (images)
•Effective type specific and fully parametric shape definitions
•Sophisticated offset handling (analysis, modification, repair) with integrated hydrostatics
•NURBS curve and surface technology
•High-level design entities (e.g. FSpline, Foils, Motion Prediction Points)
•Embedded optimization strategies for exploration (DoE) and exploitation (e.g. MOGA)
•Creation and management of both manual and driven variants
•Constraint management and screening (WatchDog)
•Comfortable browsing of project data and documentation
•Direct reporting (tables, diagrams) and easy comparison of variants
•Encapsulated pre- and post-processing for CFD solvers (e.g. potential flow codes for wave resistance and seakeeping)
•Direct coupling to SHIPFLOW®, nu-Shallo, SEDOS
•No requirement for scripting or further programming
•Availability for Linux® and Windows®


Applications

•Advanced hydrodynamic design at shipyards, consultancies, model basins and research institutes
•Ensemble investigations, for example speed range studies, trim variations, shallow water influences
•Fine-tuning of functional surfaces such as hulls, propellers etc.
•Education in first-principle-design at universities

]]> http://www.caeda.com.cn/blog/article.asp?id=45 caeda@vip.sina.com(admin) Fri,02 May 2008 08:51:14 +0800 http://www.caeda.com.cn/blog/default.asp?id=45
This note discusses a common error encountered by users when trying chimera meshes in CFD-FASTRAN.

Hole-Cutting is Encroaching a Wall/Blocked Region
Entire zone was blanked out while cutting a chimera hole
Such errors are easy to avoid and hopefully this note will assist you.

Let’s start with a brief overview of the chimera methodology using a 2D case of flow over a cylinder. We have a background mesh encompassing the outer boundaries of the computational domain and a chimera mesh to include the cylinder as shown in Figure 1. In order to accurately represent the physics of flow over a cylinder, the region corresponding to the cylinder has to be removed from the background mesh. This, in essence is the hole-cutting operation and the resulting hole in the background mesh is the chimera hole. The errors mentioned above appear during this operation.



Figure 1. Overview of Chimera methodology

These errors usually occur under the following conditions:

Two wall boundaries are intersecting. The wall boundaries may be from a background mesh or any of the chimera meshes. An example shown in Figure 2 where the wall boundaries from two chimera meshes intersect. This condition violates the current limitations on our chimera methodology. Usually one might avoid this in the initial setup. However, when the chimera grid is moving either through prescribed motions or 6-DOF models, one might encounter this. For prescribed motions, the chimera test mode under SC → Adv → Skip Flow-field Solution can be used to test if there is a possibility for wall boundaries to intersect.
The mesh in which a chimera hole is being cut has a wall boundary and during the hole-cutting operation, this wall is exposed (i.e., all the cells up until the wall were removed).In the example shown in Figure 3, the background mesh has only one cell available for hole cutting and hence the wall is exposed resulting in the encroachment error. This mesh would work fine if there were no walls. The solution is to increase the mesh resolution in the background mesh such that hole cutting doesn’t remove all the cells up to the wall. About 5 cell thickness is recommended. Another variation of this error is that an entire zone is blanked out. A similar solution should be able to fix that as well.
A trivial but sometimes overlooked reason for the second error (entire zone blanking) is that a user forgets to specify the overset boundaries.


Figure 2. Chimera errors due to intersecting walls



Figure 3. Chimera errors due to insufficient mesh resolution.

If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know.
]]> http://www.caeda.com.cn/blog/article.asp?id=44 caeda@vip.sina.com(admin) Thu,01 May 2008 11:31:33 +0800 http://www.caeda.com.cn/blog/default.asp?id=44 The 4th China CAE Annual Conference
暨2008全国计算机辅助工程（CAE）技术与应用高级研讨会
2008 National Symposium for the Advanced CAE Technology & Application

会 议 通 知
2008年6月7－9日
陕西•西安

主题：引领仿真革命  驱动企业创新

支持单位：
陕西省国防科学技术工业委员会
中国软件行业协会数学软件分会
上海CAE技术公共服务平台
主办单位：
中国机械工程学会机械工业自动化分会
中国自动化学会制造技术专业委员会
中国力学学会促进工程应用与产业结合工作委员会
陕西省国防科技工业信息化协会

会务承办：
北京诺维特机械科学技术发展中心（IMTDC）
官方网址：www.cmenet.org/CCAC2008

金牌合作媒体：《CAD/CAM与制造业信息化》杂志、《e制造》杂志、《航空制造技术》杂志、《DESIGN NEWS工业设计》杂志、《计算机辅助工程》杂志、e-works中国制造业信息化门户网

支持媒体：《机械工业信息与网络》杂志、《AI汽车志制造业》、icax网、中国CAE联盟网、中国仿真互动网、新浪科技时代网、佳工机电网、CAEfinder电子杂志。


尊敬的                 ：

    计算机辅助工程（CAE）是一种迅速发展的计算技术，是实现重大工程和工业产品设计分析、模拟仿真与优化的核心技术，是支持工程师、科学家进行理论研究、产品创新设计最重要的工具和手段。经过几十年的发展，CAE技术在航空、航天、核工业、兵器、造船、汽车、机械、电子、土木工程、材料等领域获得了成功的应用，正在逐步成为制造业信息化深化应用的关键技术。
中国CAE年会是我国CAE领域一年一届规模和影响最大的专业盛会，为相关领域的科学家、专家、工程师提供交流合作、发布最前沿科研成果的平台，年会已经分别在北京、青岛、大连成功举办过三届，来自国防科技工业、汽车制造、机械装备、电子工业、土木工程、教育等行业的与会代表累计超过750多人，已成为我国CAE领域专家学者和仿真工程师的专业交流平台。
本届年会将由来自科技主管部门、科研院所、制造企业、高等院校的领导、两院院士、知名专家、软硬件厂商代表做主旨报告、主题演讲和技术交流。预计今年会议规模将超过500人，本届年会将是历届会议层次最高、规模最大、影响最广的一届值得期待的盛世大会。
欢迎各位业内领导、专家、工程师踊跃投稿并出席年会，我们期待着与您相约古城西安，共享这一高层次产业界、学术界和工程界隆重聚会的美好时光。

一、年会宗旨
    为增强我国企业的产品开发能力、缩短产品开发周期、提高设计质量及优化产品开发流程、降低产品开发成本，推广CAE技术在产品研发和制造过程中的深入应用，快速提高我国企业的核心竞争力，促使“中国制造”迈向“中国创造”。
    搭建交流平台，促进制造企业、科研院所、高等院校之间合作与交流。

二、年会研讨主题
1、CAE当前研究热点与未来发展趋势
    计算流体力学、结构力学、材料力学、仿生力学、爆破力学等新进展
    新材料与新工艺、生物材料、微纳米、复合材料的CAE应用技术
    高性能计算（HPC）与CAE
    智能化CAD/CAE集成
    多学科、多尺度CAE仿真技术
    可靠性分析与CAE工程稳健设计
    非线性有限元进展及应用
    有限元网格自动生成技术
2、CAE专项技术应用探讨
    产品结构强度分析、疲劳寿命分析、振动及噪音仿真分析、碰撞仿真
    机构动力学、多体动力学与控制仿真技术
    跌落（Drop-Test）以及冲击、多物理场耦合分析
    结构轻量化设计与拓扑优化技术
    先进材料/结构一体化设计技术
3、CAE的平台技术与应用
    虚拟产品开发平台（VPD）
    分布式仿真平台技术与协同仿真
    产品研发仿真流程和数据管理平台建设
    企业级仿真和多学科联合仿真
4、CAE技术的行业应用与解决方案
    CAE在航空、航天、兵器、船舶工业中的应用
    CAE在海洋工程、核工业及特种行业的应用
    CAE在汽车制造、铁道机车行业中的应用
    CAE在装备制造及通用机械工业的应用
    CAE在电子、材料、土木工程、生物科技中的应用
5、CAE技术的人才培养
    社会对仿真分析工程师的需求及要求
    高校CAE课程的设置及人才培养模式
    社会科技中介及培训机构的CAE人才培训项目开发

三、年会活动
1、主旨报告
特邀行业领导、两院院士、知名专家为年会做大会主旨报告。
2、专题分会场
（1）国防科技工业：国防科技领域一直是CAE应用最大和最广的行业，也是最成熟的行业，该分会场包括航空、航天、兵器工业、核工业、特种行业等。
（2）汽车、交通运输：汽车制造业也是CAE技术应用最成熟的行业之一，近年来高速发展的铁路交通运输和船舶工业也对CAE的应用产生了巨大的需求。该分会场包括汽车及零部件制造、内燃机制造、特种车辆、机车制造、数字造船等。
（3）先进制造及综合科技应用：在由“中国制造”进入“中国创造”的过程中，CAE技术将起着不可忽视的重要作用，CAE技术作为创新设计最重要的工具和手段，同样在材料科技、电子工业、海洋工程、石油化工、压力容器、土木工程、桥梁工程、生物技术等领域也获得了广泛的应用。
（4）高性能计算应用：在高性能计算领域，一个根本趋势是将越来越多的CPU连在一起，成为集群，以完成复杂的计算。CAE是典型的高性能计算应用领域，它需要进行许多种参数计算。为了推广高性能计算技术在制造业的应用，本届年会特别开设高性能计算技术应用分会场。
（5）新热点课题研讨：本届年会还将开设“大飞机设计制造的CAE技术”、“航天飞行器热防护、生命保障、动力学与控制”、“协同仿真的CAE平台技术”、 “CAE人才开发与培养”、等专题研讨，组委会开设现场咨询项目：“专家面对面”（即征集企业遇到的复杂仿真分析问题，会务组将组织行业专家现场咨询、答疑）也将是本届年会的一大亮点。
欢迎业内专家学者、学术带头人、科研工作者、仿真分析工程师、国内外知名软硬件厂商针对年会以上主要议题和分场设置进行演讲，介绍和交流国内外研究和应用的最新进展、应用案例。
3、展览展示
年会组委会将在主会场的露天广场举办新技术、新产品、软硬件展览展示，欢迎国内外软硬件厂商、科研院所、高校及应用企业积极参与展示活动。
4、论文评选
本届年会面向全国征集论文，由年会组委会邀请有关专家评审，并选出优秀论文颁发获奖证书和奖品，优秀论文将推荐给专业刊物正式发表。论文征集要求请参考第四点。
5、最佳实践案例征集
为推广CAE技术支持产品创新的成功经验，普及CAE技术的应用，本届年会组委会将组织“2008中国CAE应用最佳实践案例”评选活动，诚挚邀请广大企业积极参加评选。
参评项目时间范围要求在2007年1月后完成的产品分析项目或阶段性成果，规划或实施中的项目不在此次评选之列，此次评选不收取任何费用。
6、欢迎晚宴及文艺演出
为答谢参会代表和合作伙伴对CAE年会的关注和支持，组委会将在会议开幕当天晚上举办招待晚宴，届时将颁发各种奖项和观看具有浓郁陕北特色的文艺演出（也欢迎有文艺特长的代表能参与表演节目，如有意向，请提前与组委会联系，以便早日安排）。
7、观光游览
西安古称长安，是我国黄河流域古代文明的重要发源地之一、世界四大古都，也是我国建都最早、历时最长的古城，距今已有3000多年的历史。自西周起直到唐代，先后共有13个王朝在此建都。西安堪称中国古代社会的“天然历史博物馆”，这里有全世界保存最完整、规模最宏大的古城墙，有总面积达108平方公里的周秦汉唐四大遗址。

四、论文征集要求
(1) 围绕主题内容充实、数据准确、文字通顺，字数在5千字以内，未在正式刊物发表。
(2) 文稿录入请用Word系统，版面为A4纸规格，通栏排版，上下左右页边距均取30mm。
(3) 标题（2# 黑体）居中；作者姓名（5# 楷体）位于标题下方（空一行）居中；作者单位、地区、邮编（小5# 宋体）位于作者姓名下方（空一行）；摘要、关键词（小5# 仿宋）位于作者单位等下方（空一行）；正文用5# 宋体，一、二、三级标题序号分别用阿拉伯数字表示（ 如“1”、“1.1”、“1.1.1” ）；图、表尽量排列紧凑，线条清晰；正文后列出参考文献（小5# 宋体）和第一作者简介及通讯方式（100字左右，小5# 宋体）；文章不留打印页码。
(4) 论文结束页后另附论文全部作者详细信息，包括作者职称、学历、职务及主要专业方向，联系方式，并标明按上述5个专题类作者论文应属的类别。
(5) 会议收录论文，不收取任何费用，仅供业内人士交流参考。
(6) 本次会议属非密级，论文中请不要有泄密内容，文责自负。
(7) 论文请务必在5月24日之前发送到（hb_cao@126.com）邮箱里。

五、会议费用
1)    会务费：1800 元/人（含会议费、资料费、餐饮费，不含住宿费）。
2)    企业对本次会议的冠名、赞助、协办等费用，根据情况具体协商。
3)    以上费用不含参会代表来往交通费和住宿费，费用自理，会务组可统一协助安排。
4)    本次会议会务工作由北京诺维特机械科学技术发展中心承办，并为代表出具正式发票。

六、注意事项及会务联系
★ 凡确认参加会议者请务必将：
① 会议论文及主题演讲题目于5月24日发到（hb_cao@126.com）邮箱内。
② 会议回执表于5月31日之前传真至会务组。（6月1号之后不再接受任何报名）
③ 欢迎访问官方网站了解会议最新进展：www.cmenet.org/CCAC2008
★  咨询电话：（010）88145672            联 系 人： 曹宏博
报名方式：1、邮件：请网上下载报名表，填写好后E-mail至hb_cao@126.com
               2、传真：请将填写好的报名表传真至 010-51717078（24小时接收）









第四届中国CAE工程分析技术年会报名表  编号№              
单位名称    
通讯地址        邮 编    
电   话        传 真        E-mail    
参会代表    姓   名    性 别    职务/职称    手   机
                
                
                
                
                
论文题目    
主题演讲
（需审核）    演讲主题    
    演讲人        职务/职称    
    演讲时间    分钟    需要设备    
是否参加最
佳实践案例    □是    □否（如果是，请向会务人员所取申报表格）
展览及赞助    □赞助年会  □ 具名赞助晚宴  □ 预定展示展位    □ 分发资料或纪念品      
是否参加9日西安兵马俑一日游    □是    □否    是否参加会后旅游    □是    □否
住 宿 预 定
(因旅游旺季必须提前预定)    □ 是
□ 否    □预定标准间（双人合住，每人每晚200元）
        □预定个人包房（每人每晚400元）
是否带有家属或随从    □是
□否    家属：    人，预定房间间数：         间
年会入住酒店
★★★★★    陕西宾馆（又称丈八沟宾馆）：是一座接待党和国家领导人及陕西省委、省政府召开重要会议的五星级园林式国宾馆，被世人美誉为“阆苑仙境”和“陕西的钓鱼台”。
到达时间    航班或车次：                       到达时间：
票务服务
(须提前预定)    □ 是  

□ 否    返程日期：          目的地：           航班：

返程日期：          目的地：           船票或火车：
其他要求    
    请务必将报名回执表于5月31日前传真至：（010）51717078   联系人：曹宏博
    或发邮件到：hb_cao@126.com，会议回执表可登录www.cmenet.org/CCAC2008下载。

]]> http://www.caeda.com.cn/blog/article.asp?id=43 caeda@vip.sina.com(admin) Wed,30 Apr 2008 08:52:06 +0800 http://www.caeda.com.cn/blog/default.asp?id=43
      在过去的十二年里，HyperSizer® 复合材料软件运用快速的结构分析和最优化设计， 大大减轻了多项大型航空航天项目中飞行器的重量。HyperSizer 不同于应用有限元分析（FEA） 的NASTRAN软件， 也不同于应用有限元素（FEM）前后置处理的 FEMAP/PATRAN 软件。HyperSizer 是一款自动化航空工程结构设计和失效分析的软件包，同时适用于金属和复合材料。 HyperSizer 不仅可用于初步设计（优化尺寸以减轻重量），也可进行最后的安全裕度计算，生成可用于适航性认证的应力报告。每个方案从头到尾都会运用到验证与确认（V&V）分析方程 ， 以排除设计过程中不必要的错误。 HyperSizer  一项主要特性是能自动生成应力安全裕度报告的  Word 文 档， 包括数值， 样本计算， 以及安全裕度汇总表。

Material Manager
     使用任意组合的材料形态或材料类型构建复合层压材料。可用剪切、复制、粘粘进行快速的铺层和接合排列。可自动绘制断裂点轨迹和应力应变剖面图。此软件也支持高度互动的 ‘what-if’ 假设性设计更改，并实时查看效果。

Basic
      在 Material ManagerTM 的基础上， 增加可针对50种以上加强筋、 加芯板结构进行应力分析和结构优化设计的功能。可根据给定的载荷优化材料和梁截面。模拟边缘载荷和临界条件，并使用100种以上的失效分析来解决由此产生的应力和结构完整性问题，如板屈曲，局部屈曲，断裂和层压复合材料理论，粘接和螺栓接合，逐步破坏和虚拟裂缝闭合等。

Pro
      在 BasicTM 的基础上， 增加与有限元分析（FEA） 相结合进行系统级的有限元要素模型自动简化和结果显示功能。该软件可检索 FEA 计算所得的内部载荷并自动输入以方便板梁设计。运用 Hypersizer   的板梁和层压板分析也可生成针对整个系统完整详细的分析结果， 适用于诸如机身、航天运载器、船体和火车车体等。另外还可以定义一般化的热弹性刚度项并返回有限元分析模型再次进行内部载荷方程计算。  

世界领先的培训和支援
    我们提供方便实用的指南、用户手册以及详尽的分析方法文件。另外设有 HyperSizer 培训课堂，由精通电脑的执业工程师担任培训师。所有的培训都采取上机操作形式并强调学员与培训师之间的互动。 HyperSizer 团队可指派应用软件工程师协助您的项目， 也可以根据贵公司的项目和要求定制和编写软件。

详细咨询及试用，联系中国授权代理: CAEDA Ltd. caeda@vip.sina.com


]]> http://www.caeda.com.cn/blog/article.asp?id=42 caeda@vip.sina.com(admin) Sat,26 Apr 2008 08:04:36 +0800 http://www.caeda.com.cn/blog/default.asp?id=42 Comment for TOSCA software（来自学员）

      你好，十分感谢你给了我学习的机会，在此对你的慷慨表示谢意。我想如果作为一个在学校中的研究者，能给你以后的工作提供机会和建议将是对你最大的感谢。在此我想谈谈对Tosca软件的一些评价：

    Tosca软件令我十分吃惊，在开始的时候我还对它表示过怀疑，但现在我会对我认识的所有人都提出它以下的一些优点。

    首先一个软件都是有缺陷的，ansys，nastran包括tosca也不例外，但这并不防碍它的广泛应用。这一点不必对人遮掩什么，但它的优点将大大超过缺点。

    首先从技术上说，它能考虑到很多研究工作中无法考虑或很难考虑的问题，比如加工制造方面的考虑；

    其次，目前国内外在结构优化方面所做的工作方面多数是在研究、理论方面，如果要实现的话，需要从底层编程做起，这上面有很大而且很辛苦的工作量，国内除了一些大的机构、研究所或者力量很强的高校能做到。即使是在现有先进有限元结构分析基础上进行二次开发，通用性也只能局限于这个软件，但tosca软件能充分利用各种有限元软件，让用户能十分自由得不必抛弃以前的传统，这一点应该十分强调！

     第三，目前的结构优化局限于一些简单模型，而tosca能做一些复杂、非线性如接触方面的东西，如果以后能更多的拿出这方面的成功例子，我想大家会十分乐意接受它。

     第四，一定要强调tosca比其它优化软件的优势，目前国内它是第一个专业的优化软件。
     第五，工程人员一般接受的是在原先基础上的满意解，不一定是最优解，最优解很难达到而且不现实，而tosca正是做到了这一点。

     第六，tosca的界面十分良好，学习也很方便，这是我见到的最好学，而且我个人认为是最实用的东西。

     这以上是我个人的看法，而且我认为也代表了很多人的意见，如果遇到询问工程技术上的问题，强调以上几点应该会让他们有深刻的映像。以后常联系，如果能我这个软件的推广进一份力量的话。。。


]]> http://www.caeda.com.cn/blog/article.asp?id=41 caeda@vip.sina.com(admin) Thu,24 Apr 2008 08:47:40 +0800 http://www.caeda.com.cn/blog/default.asp?id=41
计算流体力学（CFD）技术已经发展成为了比较可靠和成熟的数值模拟技术，针对目前市场上主流的CFDesign、I-Deas-ESC、Flotherm和IcePak等四种电子散热分析的商业计算流体力学（CFD）软件，其中的每一种都在电子系统的热问题研究中得到了精确的验证，以下的表格将对这四种软件进行一些对比。
    CFDesign    I-DEAS-ESC    Flotherm    IcePak
分析类型    有限元    有限元    有限体积    有限体积
几何模型
文件格式    直接与各主流
CAD绘图软件结合    I-DEAS格式    IGES,STEP以及Pro/Engineer转换    IGES,STEP以及Pro/Engineer转换
复杂几何模型
的支持    是    是    否    是
模型改变
的关联性    是    否    否    否
非计算流体专家适用    是    否    否    否
结构有限元
分析耦合    是    否    否    否
复杂流动    有    无    无    无
如风扇等的
旋转网格    有    无    无    无
一年License    15，000美元    20，000美元    20，000美元    20，000美元
永久License    30，000美元    60，000美元    60，000美元    60，000美元
升级维护(2年)    5，000美元    12，000美元    12，000美元    12，000美元
总价格（5年）    50，000美元    108，000美元    108，000美元    108，000美元

分析类型
主要有三种有限分析类型，即有限元（FE）、有限体积（FV）和有限差分（FD）。有限差分主要用于简单的学术代码。有限体积（FV）主要是利用较早的计算流体代码。有限元（FE）则是目前研究最透彻，验证和应用最广泛的方法，并且提供了元素质量和形体外形最大的适应性。CFDesign使用了独特的有限元网格生成技术，并自动完善边界层来提高计算精度。

几何模型
CFDesign是唯一的能够直接与目前大多数三维几何建模软件，包括Pro/Engineer接口的程序。CFDesign完全支持最基本的几何外形，这样可以避免由于文件格式转换，如IGES和STEP转换等造成的错误。对于复杂的形体外形，CFDesign也没有任何的限制。最重要的是，CFDesign完全支持随原始几何形体的变化。经过初步的分析之后，外形的变化可以直接在模拟中反映出来。同时，CFDesign可以存储计算结果而不用反复的进行初始计算。在Pro/Engineer方面，CFDesign可以使得用户方便的将CFD计算结果存储于块（.prt）文件中。
Flotherm支持IGES和STEP转换格式，但是仅限于立方体。Flotherm支持简单Pro/Engineer组合体外形的导入和计算。复杂形体将会影响计算的准确性。原始几何外形的改变并不能在Flotherm程序中体现，这样就要求重新开始进行计算。
IcePak可以使用大多数三维几何软件的转换的IGES和STEP格式，同样IcePak/Pro软件在复杂几何形体处理方面较弱，原始几何外形的改变并不能在程序中体现，这样就要求每次的模拟计算必须重新开始，十分的繁琐。

易用性
CFDesign是基于如Pro/Engineer等三维几何软件的。熟练掌握CAD几何建模软件的用户已经具备了80％CFDesign使用过程，CFDesign的用户界面非常的简单和直观。经过两天的教程就可以相当熟练的掌握了。
ESC、Flotherm和IcePak则需要更高层次的计算流体专业知识背景，这些产品的大多数使用者是热问题专家，并且专门从事模拟计算。

与结构有限元分析软件的耦合
CFDesign是唯一可以与Pro/MECHANICA、ABAQUS、NASTRAN、ANSYS和CASMOS等有限元分析软件直接进行耦合分析的软件。CFDesign中计算出来的热结果可以直接在这些结构有限元软件中使用。

计算范围
CFDesign可以用于各种基本流动，以及二相流动、旋转物体和非线性以及非牛顿流体的相关模拟。

概述
CFDesign在以下的这些方面提供了最好的性价比：
1对简单和复杂形体的适应性
2 其他软件所不具备的CAD软件整合特性
3 结构有限元分析软件的耦合
4 基于基本流动模拟的多种新模拟类型
5 适用于最广大的机械设计者使用
]]> http://www.caeda.com.cn/blog/article.asp?id=39 caeda@vip.sina.com(admin) Tue,15 Apr 2008 09:28:25 +0800 http://www.caeda.com.cn/blog/default.asp?id=39 转载

    软件类（入围项目4个）
    获奖企业和项目：
    Precimould 工具设计软件。由Advanced Composites Group(UK)和它的合作者FEA (UK) Ltd., BAE Systems Plc(UK),Bombardier Aerospace (UK)提出。
    Precimould使人们有可能设计出工具的表面形状尽可能接近所需的尺寸和形状设计工具表面形状。这个软件工具使用内部Lusas有限元代码。通过它有经验的CAD操作员可以迅速了解如何使用Precimould，而不需要具有有限元分析广泛的背景。这个软件工具减少了工艺过程中的致使制件不精确的变形或装配过程中不相符的匹配。
发展阶段经历了数年之久。第一个阶段了解制件变形过程，确定变化的参数。随后的软件设计过程包含材料性能的广泛研究。最后，软件通过实验测量和模拟数据的比较进行验证。这项产品现正在三个方面发展：复合材料工具和产品设计营销许可证，在ACG和FEA中使用工具用于咨询工作，用于ACG为他们的客户生产优化工具。
    入围决赛的其他项目：
    Digimat设计软件平台用于复合材料零部件.由e-Xstream Engineering (BE)和它的合作者Rhodia Polyamide(F)和Trelleborg(F)推出。
    Digimat是一个多范围（从微观到整体），而不是线性的。它采用矩阵，增强体，微观结构和形态特性来影响热机械性能。这个平台由6个模块组成，对应不同的制件形态，组成材料，以及设计的数学和物理定律。这有一点类似一个虚拟实验室用于研究材料的微观行为和微观结构，及其破坏。这个软件可用于大部分结构分析项目和注射成型项目。
这个理念是在开始包含模具制造的实际原型实验之前，在设计产品和探索不同的变种中利用计算和分析尽量避免浪费超过计算用的时间。主要优点是：减轻重量，降低成本，更方便的制造（成型），更大的设计自由度和功能集成。
        
    用于复合材料生产人工铺层的PlyMatch视觉辅助系统。由Anaglyph Ltd(UK)和NPL National Physical Laboratory(UK)开发。
    这个系统结合了数码相机，一个PC和一个位置软件程式，以方便制造工艺。这比激光投影系统便宜很多。这个系统在电脑屏幕上显示装配图像的现场视频图像（模具，当前层，操作者的手），由计算机产生模具图像叠加，复合材料铺层被精确放置在模具中。这种制造辅助可用于小型和中型企业。他具有精确的铺层定位和修整、调节性和使用方便特性。这个系统能直接CAD文件输入或与Anaglyph’s Laminate Tools软件一起工作。

    用于小型和中型公司和供应商在发展使用RTM，注射，模压制造工艺中的软件程序。由PPE Pǒle de Plasturgie de I’Est(FR)和LDB(FR)推出。
    这个软件包含众多的模块，用于初始机械计算，模拟，技术和经济估价，最适合工艺选择，工具结构计算。它还包含一个材料数据库模块和一个装配方式选择矩阵。该软件使用材料数据库以模拟所有复合材料结构的发展阶段。它给出一个给定项目的初始看法，并能得出快速结论。材料数据库包括通用数据，数据由合作公司对他们的产品提供，由用户输入。
]]> http://www.caeda.com.cn/blog/article.asp?id=38 caeda@vip.sina.com(lns) Mon,14 Apr 2008 08:10:37 +0800 http://www.caeda.com.cn/blog/default.asp?id=38
2007年4月12日上午，在清华大学汽车研究所301会议室举行Tosca软件的培训会圆满结束。会议由侯之超教授发起组织，博士生苏瑞意负责培训，共有7名博士生和硕士生参加了培训。

本次培训首先介绍了Tosca软件的背景和应用，通过丰富的案例、现场试用展示了Tosca软件强大的优化功能（线弹性、非线性、碰撞工况下的拓扑优化），使大家了解到通过Tosca的拓扑优化能够快速设计出新产品以及对原有产品进行技术升级。

培训中结合8个拓扑优化案例，介绍了Tosca的向导、命令行操作模式，涉及到常用拓扑优化问题（最小柔度问题、频率最大化问题等）、工艺约束（对称性、拔摸角度、最大最小尺寸）、多工况、多目标优化以及网格自适应问题等，中间还就拓扑优化的基本理论和算法进行了说明与讨论。最后，参加活动的部分老师和同学还现场运行有关案例，切实体验了运用Tosca进行设计优化的完整过程。]]> http://www.caeda.com.cn/blog/article.asp?id=37 caeda@vip.sina.com(admin) Sat,12 Apr 2008 08:44:37 +0800 http://www.caeda.com.cn/blog/default.asp?id=37
During manufacture of composite structural components by manual lay-up, it is necessary to position plies of dry or wet fabric or prepreg accurately on a tool, or on preceding plies. Template methods or specialised equipment are often used for this purpose, in order to aid the operator in correctly and accurately locating a ply. Accuracy in ply placement as well as quality assurance regarding the correct ply sequence and orientation are very important factors in the manufacture of high-cost, high-performance composite components.

The purpose of the novel Ply Placement Visual Aid System is to display together on a computer screen a live video image of the "real world" that the operator is interacting with, and a computer-generated image of the structure and composite ply being assembled.  

The data for the computer generated drawings of the plies that are displayed correctly in place, may originate from virtually any CAD application, or from our own Laminate Tools software application, for additional features. Laminate Tools is a powerful design tool for composite material structures, and includes industry-leading draping simulation capabilities for ply definition.  

In this way, ply information comes directly from design and may not require any translation or manipulation whatsoever, for use with the present system.

Details inquiry: caeda@vip.sina.com
Photo Gallery...

]]> http://www.caeda.com.cn/blog/article.asp?id=36 caeda@vip.sina.com(lns) Fri,11 Apr 2008 16:20:01 +0800 http://www.caeda.com.cn/blog/default.asp?id=36

制造行业CAE对高性能计算平台的选择

CAE在制造企业中承担着关键的业务，所以其高性能平台的选择非常重要，这个平台直接影响CAE的运行性能表现、整体成本和系统维护等方面问题。想让CAE能实现“随需应变”，在高性能计算平台的选择上就需要进行全面的考虑。

CAE软件使用SMP系统存在的问题

理论上讲，SMP系统可以胜任CAE应用程序的运行，但是不可避免的存在一些问题。首先是SMP系统价格相对较高。而且CAE的某些应用软件的扩展性不好，如隐式有限元分析（IFEA）类软件通常只能用到最多8个CPU。这个特点决定了配置超过8CPU的大SMP服务器没有太大的必要，反而会造成投资的浪费。

此外，SMP系统由一个操作系统管理，如果没有强大的资源管理软件和作业调度软件配合，很容易造成CPU分时处理多个任务的现象，即在一个CPU上同时运行多个进程，从而影响整个系统的使用效率，同时难于保证关键任务的按时完成。常用的作业调度软件有LSF，PBSpro和OpenPBS，其中开放源代码的OpenPBS很难胜任复杂的CAE应用软件的管理，尤其是对共享内存方式并行模式（OpenMP，pthreads）的管理。而LSF和PBSpro的价格比较高，同样会增加用户的投资；

CAE软件使用Cluster系统存在的问题

目前市场上的Cluster系统通常是用PC服务器作为节点构建的，每个PC服务器内部一般配置2个CPU。这种类型的Cluster系统在承担CAE应用软件运行任务时也存在一些问题，比如共享内存方式的并行模式只能运行在一个节点内部，也就是说，最多只能用2个CPU去处理OpenMP/threads类型的CAE任务，很难满足应用的时效性要求。又因为单CPU的处理能力限制，需要更多的CPU来完成一个计算任务，这样就需要购买较多CPU的软件许可证，增加了用户的投资

IBM eServer p5 575服务器：CAE应用软件的最佳选择

针对SMP系统和Cluster系统在处理CAE应用时存在的问题，我们可以考虑一种综合这两类体系结构优势的高性能计算机系统——SMP Cluster。如下图所示：



IBM Cluster1600系统就是这种类型的体系机构，可以采用多种安装IBM POWER5芯片的服务器作为节点，如p5 575，p5 590，p5 595等，其中p5 575是专为HPC推出的一款高性能的计算服务器。使用p5 575作为节点来构成Cluster1600系统具有以下特点：

 采用SMP服务器作为节点，构成Cluster系统。这样兼有两种体系结构的特点，可以称之为SMP Cluster系统；

 p5 575是一个在2U高度中可以安装8个POWER5处理器的SMP服务器（业界密度最高的非刀片服务器）。在一个p5 575内部运行共享内存方式的并行模式（OpenMP或threads）应用程序，可以保证系统能胜任较大规模的运算模型处理，满足计算的时效性。应用程序也具有较高的并行效率和加速比；

 节点间连接采用IBM为HPC专门设计的高性能交换机HPS，其单通道带宽达到4GB/s，并且每个节点都配置了双通道作为冗余配置。这样在保证跨节点通信的优越性能的同时，提供了较高的稳定性；

 整个系统可以运行消息传递方式的并行任务（MPI或PVM）。同时支持混合并行模式，即在节点内部使用共享内存方式的并行模式，节点间使用消息传递方式的并行模式，如MPI + OpenMP模式；

 完善的集群管理环境，统一进行资源管理和作业调度。管理软件包括：

 CSM集群系统管理软件；

 LoadLeveler作业调度软件，具有强大的作业调度、记帐及断点/续算功能；

 WLM资源管理软件，与LoadLeveler配合使用，满足用户对系统资源管理和作业调度的复杂要求；

 优化的并行使用与开发环境PE，提供高性能的、完善的并行作业运行与管理；

 高性能的并行文件系统GPFS，为整个集群提供稳定的共享文件系统，方便管理与使用；

 较高的性能价格比；

结论：

结合制造行业CAE应用软件的特点，我们可以看到使用IBM eServer p5 575作为节点的IBM Cluster1600系统可以很好地满足这类应用的复杂需求，是CAE应用软件的最佳选择。

首先，使用8 CPU的SMP系统作为节点，可以很好地满足隐式有限元（IFEA）类型的应用，如ABAQUS, ANSYS和MSC Nastran。这类应用以使用共享内存方式的并行模式为主。8 CPU的p5 575具有足够的处理能力满足应用软件处理大系统模型的需求，同时保证系统具有最高的使用效率。在这类应用中，更大的SMP服务器是没有任何必要的。

其次，对于扩展性较好的EFEA类和CFD类应用软件，通过IBM高性能交换机（单通道带宽为4GB/s）连接的Cluster1600系统可以很好地满足节点间的通讯要求。在某些MPI的应用中，SMP Cluster具有比相同数目CPU的SMP系统更好的性能。

再次，IBM Cluster1600系统具有完善的系统管理软件，可以保证同时使用IFEA、EFEA和CFD应用软件的用户方便地规划与管理系统。作业调度软件LoadLeveler和资源管理软件WLM的结合使用可以满足用户对系统资源的有效管理，达到系统的最优化使用。

最后，IBM eServer p5 575采用性能强大的POWER5处理器，用户可以使用较少的CPU达到较高的性能，从而节省应用软件的许可证费用。另外，IBM eServer p5 575是专门为高性能技术运算和大规模并行处理开发的产品，大缓存、高缓存带宽和高内存带宽保证了处理复杂数学模型和复杂运算时的性能。下面是一些已公布的CAE应用程序的测试结果：

ANSYS的测试结果



MSC.Nastran的测试结果：

(T112)



]]> http://www.caeda.com.cn/blog/article.asp?id=35 caeda@vip.sina.com(lns) Fri,11 Apr 2008 16:17:35 +0800 http://www.caeda.com.cn/blog/default.asp?id=35
作者：IBM 互联网服务器部 么石磊博士 来源：赛迪网 字体[放大、标准、缩小]
  
高性能计算（HPC）正逐步进入制造行业，承担诸多关键的计算应用。该领域中用户主要分成两类，一类是实际制造企业，如汽车设计制造厂商、航空工业企业、电力企业及消费产品生产商等。这一类用户通过高性能计算技术来提高产品的性能，减低成本，同时缩短产品的设计、生产周期，以使企业在市场上更具竞争力，另一类是研发单位，如政府、国防和大学中涉及制造行业的部门或专业。这一类用户的目标是利用高性能计算技术改善设计方法，提高设计水平从而为实际生产服务。

下图给出了制造行业中采用计算机进行产品开发的流程，包括建模、前处理（模型修改和网格生成）、计算分析、交叉学科综合及后处理几个部分。其中高性能计算主要应用于计算分析部分，统称为计算机辅助工程（CAE）。



制造行业CAE应用程序的特点

制造行业CAE的应用可以分为隐式有限元分析（IFEA）、显式有限元分析（EFEA）和计算流体动力学（CFD）三个子学科。几乎所有的制造企业的高性能计算都依赖于独立软件开发商（ISV）提供的商业软件，只有计算流体动力学中结构网格计算类型的部分软件是用户自己开发的。因此制造行业中的用户在购买硬件平台的同时通常会购买相应的科学计算软件产品。而在某种程度上，往往是应用软件的特性决定了硬件平台的选择。

下表中给出了CAE常用的应用软件，并列出这些软件的特点，包括并行方式和可扩展性。



从上表中我们可以了解到CAE应用软件具有以下特点：

特点1：IFEA类应用软件（如ABAQUS, ANSYS和MSC Nastran）的可扩展性不是很好。当使用超过8个CPU来处理一个任务时，通常不会再有性能上的提升；

特点2：IFEA类应用软件通常使用共享内存方式（pthreads或OpenMP）进行并行处理，其中ABAQUS不支持消息传递方式（MPI）的并行；

特点3：EFEA类应用软件（如LS-DYNA, PAM-CRASH和RADIOSS）和计算流体动力学软件（如FLUENT, STAR-CD和PowerFlow）的扩展性相对较好；

特点4：EFEA类应用软件和CFD软件以采用消息传递并行方式（MPI）为主。

高性能计算（HPC）服务器体系结构分类及特点

目前市场上常用的高性能计算服务器大致可以分为以下3种体系结构，即：

并行向量处理机（PVP）：

PVP系统含有为数不多、功能强大的定制向量处理器（VP），定制的高带宽纵横交叉开关及高速的数据访问。由于这类系统对程序编制的要求较高，价格很昂贵且难于管理，因此，这种类型计算机主要集中在一些大型国家关键部门，在本文中不再赘述。

对称多处理机（SMP）：

SMP系统采用商品化的处理器，这些处理器通过总线或交叉开关连接到共享存储器。今天的市场上常见的机型有IBM p系列服务器、HPQ的SuperDome、Alpha的ES、GS系列及SGI公司的Altix系列。SMP系统通常具有以下特点：

1) 系统内的CPU共享并可以直接访问所有的内存；

2) 由一个操作系统管理整个系统；

3) 支持共享内存方式的并行模式，如OpenMP，pthreads等；

4) 支持消息传递方式的并行模式，如MPI，PVM等

5) 系统的价格相对较高；

6) 为提高系统的使用效率，需要有功能强大的资源管理软件和作业调度软件配合进行系统管理。如LSF、PBS及IBM的WLM和LoadLeveler等；

工作站集群（COW，Cluster Of Workstation, 或简称Cluster）：

Cluster结构是近年来发展势头很好的一种体系结构。这类机型的技术起点比较低，用户甚至可以自己将一些服务器或微机通过以太网连接起来，配以相应的管理、通讯软件来搭建Cluster。但是如果要构造高性能、结构合理并具有好的RAS特性的Cluster却不是一件容易的事情。几乎所有的国内、外计算机厂商都有自己的Cluster集群产品，如IBM的Cluster1350、联想的深腾系列及曙光的天潮系列等。Cluster系统通常具有以下特点：

1) 系统由多个独立的服务器（在Cluster概念下称为节点）通过交换机连接在一起。每个节点拥有各自的内存，某个节点的CPU不能直接访问另外一个节点的内存；

2) 每个节点拥有独立的操作系统；

3) 需要一系列的集群软件来完成整个系统的管理与运行，包括：

 Cluster系统管理软件，如IBM的CSM，xCat等；

 消息传递库，如MPI，PVM等；

 作业管理与调度系统，如LSF，PBS，LoadLeveler等；

 并行文件系统，如PVFS，GPFS等；

4) 支持消息传递方式的并行模式，如MPI，PVM等；

5) 只能在单个节点内部支持共享内存方式的并行模式，如OpenMP，pthreads等；

6) 性能价格比好；


]]> http://www.caeda.com.cn/blog/article.asp?id=34 caeda@vip.sina.com(admin) Thu,10 Apr 2008 08:44:38 +0800 http://www.caeda.com.cn/blog/default.asp?id=34 邀请函


主办单位：教育部现代设计与制造网上合作研究中心
中国工程院产业工程科技委员会
上海交通大学
上海市中国工程院院士咨询与学术活动中心
支持单位：上海市经济委员会
上海市科学技术委员会
承办单位：教育部现代设计与制造网上合作研究中心（上海）
          上海交通大学机械与动力工程学院现代设计研究所        

论坛主题： 产品创新  现代设计 技术研发 知识服务  
研讨中心： 装备自主研发中的问题  民营企业的作用和问题
专题研讨：内燃机设计 巨型重载装备设计 核岛主泵设计
─ 届时将召开教育部现代设计与制造网上合作研究中心学术委员会和现代设计与产品研究开发网络理事会会议
时   间：  2008年7月31日-8月1日
地   点：  上海市上海交通大学浩然高科技大厦
论坛特色： 主题报告引导下的深入讨论交流；领域专家现场提供问题咨询；针对具体产品设计设三个子论坛进行专题讨论。论坛后续以“产品现代设计” 系列培训,同时举办“创新思维与现代设计”师资培训班。
           ─  届时发布《中国制造业上市公司产品创新能力排名榜》！

产品设计与技术创新能力是企业的核心竞争力。企业如何选择适合自身企业发展战略的产品设计开发道路，如何在产品设计开发中应用各种现代设计方法，如何在全球范围内获取（或者提供）设计知识服务及如何科学客观地评价企业产品开发与技术创新能力是政府、企业家与学者都持续关注的主题。2008“产品现代设计与技术创新”论坛将围绕企业产品创新、设计开发、设计理论与方法、产品知识管理，特邀国内外设计专家、管理学专家、政府领导、学者与企业家进行主题研讨。并具体以内燃机设计、巨型重载装备设计及核岛主泵设计为专题深入讨论产品创新与设计开发中的相关问题。同时现场提供企业问题咨询。
一、    论坛特邀主题报告专家
1.    清华大学教授                     柳百成院士
2.    上海交通大学副校长               林忠钦教授
3.    美国南加州大学IMPACT中心主任    Stephen C-Y. Lu 教授
4.    上海交通大学、西安交通大学教授   谢友柏院士
5.    发布中国制造业上市公司产品创新能力排名榜
二、    论坛拟邀领导及专题研讨专家（排名不分先后）
1.    中国工程院副院长                           邬贺铨 院士
2.    上海市中国工程院院士咨询与学术活动中心主任 翁史烈 院士
3.    清华大学教授                               王玉明 院士
4.    浙江大学教授                               谭建荣 院士
5.    西安交通大学前校长                         蒋德明 教授  
6.    武汉理工大学校长                           周祖德 教授
7.    清华大学摩擦学国家重点实验室主任           雒建斌 教授
8.    河北工业大学副校长                         檀润华 教授
9.    上海交通大学汽车工程研究院院长             许  敏 教授
10.    上海交通大学重大装备设计与控制工程研究所所长       高 峰 教授
11.    西安交通大学现代设计与制造网上合作研究中心副主任   陈 渭 教授
12.    美国Engineous 公司技术总监                 Wei-Shan Chiang 先生
13.    SolidWorks公司技术总监                     C.C.Chan 先生
14.    AMT咨询资深顾问                            王玉荣 博士  
15.    AVL李斯特技术中心（上海）技术经理          马明堂 博士
论坛还将邀请上海市经委领导、上海市科委领导、中国工程院上海院士咨询与学术活动中心、教育部现代设计与制造网上合作研究中心学术委员会委员，现代设计与产品研究开发网络理事会理事以及关注本论坛研讨主题的同行专家参加。
三、    论坛专题讨论介绍
在第一天设计与创新大会论坛后，第二天将设三个专题讨论，分别为内燃机设计相关技术、巨型/重载设备设计相关技术、核岛主泵设计相关技术。以具体的产品对象研讨产品现代设计与技术创新的具体应用。
四、    会议日程
第一天：主题报告及主题讨论
        2007“中国制造业上市公司产品创新能力排名榜” 发布
第二天：专题报告及专题讨论
（详细日程于会议二周前确定）
五、    论坛征文范围及时间
本届论坛将出光盘版论文集，优秀论文拟由《计算机辅助工程》专刊登载。
征文范围：
设计理论与方法  设计知识服务  设计知识管理
内燃机设计相关技术  重型/重载设备设计相关技术  核岛主泵设计相关技术
            企业产品开发与技术创新能力评价　
论文截止时间：2008年5月30日   投稿信箱：zhinanz@yahoo.cn  
六、    “产品现代设计”系列培训
制造业的竞争实质就是产品设计的竞争。产品设计是一个集需求分析、知识获取、概念设计、详细设计、性能仿真、加工制造及试验等于一体的复杂的过程。设计创新理论、三维实体建模、结构有限元分析、机械系统动力学建模与仿真（数字样机构建）、计算机辅助制造、产品设计优化等技术得益于日新月异的计算机软件工具的支持，已被广泛地应用于产品设计与技术研发领域。为了推动产品现代设计理论、方法与工具在产品开发与技术创新中的应用，考虑各种形式的短期研讨会无法真正深入到产品设计方法与支持软件的具体使用方法层面，弥补单一的软件使用培训缺乏和理论的具体结合及应用案例介绍。教育部现代设计与制造网上合作研究中心（上海）论坛后续以一年一度的“产品现代设计”系列培训活动。包括：
培训1：产品三维实体建模（CAD建模）
培训2：结构有限元分析（CAE分析）
培训3：机械系统动力学建模与仿真（数字样机构建）
培训4：计算机辅助制造（CAM）
培训5：产品设计优化
培训6：产品现代设计与创新理论
培训7：创新思维与现代设计（师资培训班）
培训8：企业技术研发平台构建策略与方法（根据企业要求开设）
（详细课程内容请参阅“产品现代设计”系列培训大纲或访问网址：http://202.117.208.15/px/px-11.html）。
七、    报名方式
    Email、电话、传真报名均可（ 截至时间：2008年6月30日）
八、    费用：
论坛会务费：人民币￥1000元/人（会员参加优惠费：800元/人）。
产品现代设计系列培训班（每班5天40学时）每班2000元、会员每班1600元）。
九、    联系方式：
上海交通大学现代设计研究所  教育部现代设计与制造网上合作研究中心（上海）
地址：上海市华山路1954号上海交通大学机械楼605室       邮编：200030
联系人：戴旭东（13701931304） 张执南（13585718558）
电话：021-62933701、021-34206245    
传真：021-62932252、021-34206245
E-mail： zhinanz@yahoo.cn；cmdtforum@gmail.com
网址：http://forum.cmdt.cn http://www.chinamoderndesign.com
                  
              

教育部现代设计与制造网上合作研究中心
中国工程院产业工程科技委员会  
上海交通大学
上海市中国工程院院士咨询与学术活动中心
二ΟΟ八年三月十四日
]]> http://www.caeda.com.cn/blog/article.asp?id=33 caeda@vip.sina.com(lns) Tue,08 Apr 2008 22:27:56 +0800 http://www.caeda.com.cn/blog/default.asp?id=33
    来自CAE年会秘书处的消息：第四届中国CAE年会即将于2008年6月7－8日在古城西安举行。中国CAE年会是我国CAE领域一年一届规模和影响最大的专业盛会，为相关领域的工程师、专家学者提供交流合作、发布最前沿科研成果的平台，推动我国CAE产业的发展，促进CAE技术在我国的应用。会议的研讨内容及展览涵盖仿真、分析、高性能计算、数据模拟与仿真、流体计算等多个领域。年会已有来自航空、航天、汽车、机械、船舶、兵器、电子、土木工程、教育等行业累计超过750多位专业人士进行深入研讨和交流。

    据悉，本届年会将由科技部“十一五”制造业信息化科技工程办公室、中国机械工程学会机械工业自动化分会、中国自动化学会制造技术专业委员会、陕西省国防科技工业信息化协会共同主办，北京诺维特机械科学技术发展中心承办。并得到了《e制造》杂志、《CAD/CAM与制造业信息化》杂志、《航空制造技术》杂志、《中国制造业信息化》杂志、《DESIGN NEWS工业设计》杂志、《计算机辅助工程》杂志、《机械工业信息与网络》杂志、中国仿真互动网、中国CAE联盟网、e-works中国制造业信息化门户网、CAEfinde网、icax网、新浪科技时代网、佳工机电网等众多媒体的大力支持。

据主办单位介绍，本届年会将由来自科技主管部门、科研院所、制造企业、高校的领导、两院院士、知名专家学者、软硬件厂商代表进行演讲和交流，预计本届的与会代表规模将达到500余人，相信本届年会也将是历届会议层次最高、规模最大、影响最广的一届值得期待的盛世盛会。

年会秘书处正在向广大专家和工程师征集论文，投稿信箱：51meeting@163.com

如需了解年会详细信息，请致电010-88145675 / 88145672咨询。

往届回顾：http://www.cmenet.org/CAE3

]]> http://www.caeda.com.cn/blog/article.asp?id=32 caeda@vip.sina.com(admin) Mon,07 Apr 2008 10:09:36 +0800 http://www.caeda.com.cn/blog/default.asp?id=32
曙光航空航天高性能计算FASTRAN解决方案

摘要：CFD-FASTRAN软件是由CFDRC公司与美国NASA联合开发的专门用于航空航天领域空气动力学计算的CFD软件，该软件可广泛应用于飞行器的亚、跨、超和高超音速的气动力学计算和一些特殊气体动力学问题如直升机旋翼、导弹发射、座舱弹射、投弹、机动和气动弹性等。本文针对航空航天高性能计算提出了一套完整的集群解决方案，并在此平台上对FASTRAN并行性能进行了详细的测试。测试结果表明，FASTRAN软件可以方便的部署和运行于曙光集群系统，并能够获得令人满意的加速比。

关键字  CFD，FASTRAN，并行，航空航天

1.FASTRAN介绍

    CFD-FASTRAN是CFDRC公司与美国NASA联合开发的专门用于航空航天领域空气动力学计算的CFD软件，在当前所有为航空航天设计的计算流体力学软件中位于前列，被广泛应用于飞行器的亚、跨、超和高超音速的气动力学计算和一些特殊气体动力学问题如直升机旋翼、导弹发射、座舱弹射、投弹、机动和气动弹性等。CFD-FASTRAN 具有挑战性的功能是将基于密度的可压缩欧拉方程和N-S方程同多体运动力学、有限反应率化学和非平衡传热学耦合起来，解决一系列极为复杂的航空航天问题。

CFD-FASTRAN具有强大的技术优势，它体现在：

    ·多年同美国国家航空和宇宙航行局（NASA），美国能源部（DOE），美国国家科学基金会（NSF）等机构的合作，使CFD-FASTRAN不断得到完善和发展。
  
    ·CFD-FASTRAN专门针对航空航天设计，可以计算包括超高速流动，移动体，气动热化学和气弹等复杂艰辛的课题。

    ·CFD-FASTRAN充分考虑了行业的需要，开发出航空航天工程师所需要的前处理、后处理程序。

    正是基于上述原因，使得FASTRAN在全球拥有巨大的用户数量并获得了广泛的认可。在美国军方，CFD-FASTRAN软件一直用于先进战斗机的设计，其中著名的应用案例包括F-16战斗机翼身气动弹性分析、Martin Baker MK16座椅弹射设计、F/A18杰达姆投弹模拟设计以及X34机高超音速激波模拟设计等，如下图所示。

    此外，近期NASA采用CFD-FASTRAN模拟了在9500英尺高空高速飞行(7倍声速)状态下的飞行器分离情况。高速分离现象一直是CFD软件很难模拟的领域，但如下图所示，FASTRAN很好的模拟了飞行器分离过程中整个流场的速度和温度分布。

    随着中国航天航空事业的快速发展，尤其是载人航天技术的巨大成功，我国科技人员对空气动力学的数值模拟研究提出了越来越多的需求，常规的计算能力远远无法满足复杂的大型飞行器设计所带来的巨大需求。因此，采用大规模集群技术成为最佳选择。

    FASTRAN支持大规模并行计算集群系统，它可以采用两种并行计算环境，一种为我们熟知的MPI，适用于非结构化网格；一种为自行开发的MDICE，适用于结构化网格。由于FASTRAN的大多数湍流模型仅针对结构化网格，因此本文仅对MDICE并行环境下的FASTRAN性能进行了测试，并根据测试结果提供了一套完整的集群解决方案。

2.曙光航空航天高性能计算解决方案

    曙光高性能集群由若干台高性能节点机、高速计算网络连接到一起组成，并通过监控系统和一组集群管理软件对其进行有序管理。曙光集群的基本体系结构如下图1所示，主要包括节点、网络、存储、监管和软件等几个部分。根据用户对计算规模需求的不同，可以对节点、网络和存储等进行量身定制。

2.1.节点机

    节点机主要包括三类节点：计算节点、管理/登入节点和IO节点。

2.1.1.计算节点

    计算节点主要用于完成FASTRAN的计算任务，因此要求：

    ·有很强的计算能力，对CFD而言特别是浮点运算能力。

    ·有适量的内存以使前处理和求解时数据能完全驻留在物理内存中。经验表明，一般每个CPU至少分配1GB内存。

    ·支持高速、低延迟的通讯网络

    针对以上要求，我们推荐计算节点采用两路的曙光天阔R210A（可升级到双核）或者4路的曙光天阔R4280A。R210A支持两颗200 Series的Opteron处理器，同时根据需要可以轻松扩展到双核；R4280A则支持4颗Opteron 800 Series单核或者双核处理器，这两种服务器都具有极高的性能，非常适合用于构建高性能集群系统。体现在：

    ·Opteron系列处理器较Xeon架构的处理器具有更高的性能。AMD的Opteron系列处理器采用创新的直连架构，通过在处理器内部集成内存控制器而大幅降低内存延迟，提高数据访问速度。由于消除了Intel固有的共享带宽瓶颈，使得AMD平台的内存带宽随着处理器数量的增加而成倍的增加，例如4路的R4280A配合双通道DDR400内存可以提供最高达4*6.4GB/s的理论内存带宽，这尤其适合浮点密集型和内存密集型的CFD应用程序。此外，Opteron处理器在提供强大的计算能力的同时具有业内出名的低功耗，用户通过选用基于Opteron平台的处理器能够大幅降低在供电与散热系统上的开销，可以最大限度的降低TCO。

    ·采用四路的曙光天阔R4280A可以获得极高的性能。AMD的超传输技术将4颗CPU直接连接在一起，两颗CPU之间可以提供高达6.4GB/s的通讯带宽和纳秒级的延迟，这使得1个四路的SMP系统比采用任何高速网络互联的2个两路系统的性能都要高。此外，在CPU数量需求一定的情况下，采用4路的R4280A可以大幅减少网络设备（网卡、交换机、网络监控卡）从而降低系统成本。

    ·采用升级到双核的曙光天阔R210A作为计算节点可以在获得高性能的同时降低系统成本。采用一颗AMD的双核处理器可以获得50-70%左右的性能提升(参见AMD官方网站测试数据)，这使得其具有几乎相当于两路单核处理器的性能。同时，采用Opteron的DP产品较MP产品的成本更低。用户通过选用双核R210A可以获得性能与价格的平衡。

2.1.2.IO节点

    IO节点由存储设备、NFS文件系统等组成。IO节点作为NFS文件系统主节点，外挂磁盘阵列或者连接其它存储设备，负责文件的I/O操作。其它节点包括计算节点和管理登入节点访问存储设备的请求都要通过I/O节点完成。

    由于IO节点和存储系统密切相关，所以IO节点将在后面存储系统中详细讨论。

2.1.3.管理/登陆节点

    管理/登入节点是外部设备和集群系统之间连接的桥梁，任何用户和系统的管理员都只能通过管理/登入节点才可能登陆到节点机子系统上，事实上，管理/登入节点是一个多种身份于一身的节点，它具有登入、管理和作业提交等功能。但总的来说，系统对管理/登入节点要求不高，只要采用相对经济的配置就可以了，但必需保持和计算节点、IO节点同构。

2.2.网络

    网络的设计是系统中很重要的一部分，网络系统特别是计算网络的性能很大程度上影响着整个系统的性能。由于并行计算时的数据通讯和NFS文件系统的数据传递对网络的要求较高，为了管理、监控等操作不对计算通讯和文件IO操作产生任何干扰，本方案设计了两套网络系统，实现双网分离、专网专用，保证了系统的高效性。同时，两套网络又起着相互备份的功能，当其中一套网络由于某种故障瘫痪时，另一套网络就会担负起其功能，提高了系统的可靠性。

    两套网络包括计算网络和管理网络。计算网络宜采用高速的互连网络，目前的高性能网络包括千兆网、Myrinet和Infiniband等。目前Myrinet和Infiniband价格非常昂贵，并且FASTRAN软件尚未支持这两种网络，因此本方案中的计算网络选用千兆以太网；管理网络对性能的要求不高，选择价格便宜的百兆以太网。

    对于较大规模的集群系统，还可以考虑单独设置数据传输网络，将计算过程中的数据传输与个节点之间的通讯分离开，避免相互干扰。这对于非稳态、有大量中间数据需要保存的CFD应用而言尤其重要。数据传输网络一般采用千兆以太网，根据需要还可以在IO节点上进行双网卡绑定以提高数据传输带宽。

2.3.存储

    目前根据应用规模的不同，存储的解决方案有三种选择：直连式存储DAS、网络接入存储NAS和存储区域网络SAN。DAS主要适合于构建小型存储系统，NAS适合于投资规模不大的部门级存储，而SAN造价昂贵，性能突出，适合于企业级、要求扩展能力强的存储。

    针对高性能计算的特点，本解决方案的存储系统采用简单易维护的DAS方式，系统由I/O节点、磁盘阵列和NFS文件系统等组成。为了保证性能，可单独设置I/O节点（也可与管理/登入节点合并以节省投资），所有计算节点通过NFS文件系统和I/O节点，即可象访问本地文件一样访问保存在存储设备上的数据。

    磁盘阵列采用曙光DS-2120SA或DS-2120。其中DS-2120SA盘阵支持Serial ATA磁盘技术，外部主机通道支持SCSI接口技术，内部可以容纳12颗SATA硬盘，最大提供3TB的存储空间； DS-2120包括2个SCSI内部通道和2个SCSI外部主机通道，最大支持12块146GB的高速SCSI硬盘。两种磁盘阵列在设计上都采用了CableLess（无线缆连接）结构，避免了诸如线缆脱落、信号串扰、线缆受热老化等问题，从根本上提高了磁盘阵列系统的可靠性，同时也带来了更高的性价比、更方便的管理。

3.曙光集群FASTRAN性能测试与分析

    曙光公司和ATE中国有限公司合作，在曙光集群系统上部署了FASTRAN程序，同时对其并行性能进行了测试。结果表明，FASTRAN空气动力学模拟软件在曙光集群上可以取得较高的并行效率。

3.1.测试环境

    本测试环境为曙光公司的内部测试平台，由4个R4280A组成，采用千兆以太网连接。具体配置如下：

3.2.测试模型：DLR-F4翼身组合体

    DLR-F4翼身组合体算例是NASA的一个标准算例，用于测试商业CFD软件计算阻力的准确性。采用CFD-FASTRAN软件进行了结构化网格的计算。DLR-F4外形如图所示：

给定的计算状态为：

    采用多块结构化网格进行翼身组合体计算，计算网格如图所示，图中给出了对称面和物体表面的网格。

    计算网格共80块，网格单元数目为：3,780,480。

    数值计算空间格式采用Roe格式，通过MUSCL限制器扩展到高阶，时间格式采用全隐格式进行时间积分。采用层流模型，迭代30步后记录结果。

3.3.测试结果

    从以上测试结果可以看出，CFD－FASTRAN软件可以非常好的运行在曙光集群系统上，并可以获得非常优良的近乎线性的加速比。正是基于这种特性，曙光集群系统非常适合CFD－FASTRAN等计算流体力学软件的并行计算。由于其良好的可扩展性和加速比，用户可以通过扩展系统的节点数来达到提高系统性能的目的。

4.方案选择

5.附录

5.1.相关编纂人员简介

    ·曹连雨

    ·所学专业：热物理与流体力学

    ·所在单位：曙光公司技术支持部深度技术支持组

    ·所做工作：流体力学、财政软件等相关技术的售前支持工作
]]> http://www.caeda.com.cn/blog/article.asp?id=31 caeda@vip.sina.com(lns) Mon,07 Apr 2008 09:45:27 +0800 http://www.caeda.com.cn/blog/default.asp?id=31
高性能计算在工业工程领域的应用和发展

李根国  桂亚东（上海超级计算中心 上海 201203   ggli@ssc.net.cn）

摘要：本文简要介绍了高性能计算在工程和工业领域的发展状况，介绍了上海地区主要大型制造业的概况，最后说明了上海超级计算中心在工程计算领域开展的工作和发展。
关键词：高性能计算；CAE；CFD

一 高性能计算在制造业领域的应用现状
   近年来高性能计算在工业和制造业领域的应用越来越普遍和广泛。从TOP500的统计信息来看，工业领域所占的比例在不断增加。2005年6月，工业用户使用的高性能计算机占到52.8％。而其中的半导体和制造业用户所占的比例相当可观。其中美国半导体公司大约有70台，美国Boeing有4台, Lockheed Martin有2台，德国BMW有7台，德国VW有2台，德国Siemens有2台。许多国际著名的制造业大公司已实现了产品的虚拟化设计和制造，并实现了全球资源共享,利用全新的理念设计产品。美国GM,美国GE,日本Nissan等公司都拥有总计算能力超过10万亿次的高性能计算机用于新产品的研发。Boeing公司在上世纪90年代就实现了无纸化设计。Boeing公司已宣布利用高性能计算机对航空发动机进行全物理过程的模拟仿真。主导世界制造业方向的大集团公司目前都拥有大量的超级计算机用于产品设计和数据处理，制造业信息化是知识经济时代企业核心竞争的必要组成环节。

下面以计算流体动力学（CFD）为例，说明高性能计算机的发展对工程应用领域的极大推动作用。自二十世纪六十年代以来CFD技术得到飞速发展，其原动力是不断增长的工业需求，而航空航天工业自始至终是最强大的推动力。传统飞行器设计方法试验昂贵、费时，所获信息有限，迫使人们需要用先进的计算机仿真手段指导设计，大量减少原型机试验，缩短研发周期，节约研究经费。如美国战斗机YF-23采用CFD进行气动设计后比前一代YF-17减少了60％的风洞试验量。目前在航空、航天、汽车等工业领域，利用CFD进行的反复设计、分析、优化已成为标准的必经步骤和手段。但是还有大量的问题需要更大规模的计算才能满足需求。如飞机设计中全机气动性能计算，火箭发动机复杂多变的燃烧和跨音速流动模拟，导弹的气动隐身性能评估，低阻力系数高性能汽车外形的设计和分析，透平机械复杂叶型及组合的设计分析，潜艇尾迹模拟，高超音速航天器空气动力学设计分析，核电站水蒸汽两相流流动分析，非定常状态的物理过程如飞机起飞降落、过载下空间推进剂晃动分析等。这些大规模设计计算问题不但单个作业计算量庞大，且需不断调整，重复计算。人们试图在很多场合期望放弃假设直接去求解和探索机理问题。如抛开基于统计理论的湍流简化模型直接模拟（DNS）湍流，这在计算能力有限的过去几乎是无法想象的。当前CFD问题的规模为：机理研究方面如湍流直接模拟，网格数达到了109（十亿）量级，在工业应用方面，网格数最多达到了107（千万）量级。

    工程计算中的难题随着计算技术和计算机的发展而不断解决。与此同时，新的问题又会不断产生。例如，从基本材料的组份与构造到复合材料，从复合材料做成构件，再由构件装配成工程/产品，存在着从微观、细观到宏观的多尺度现象；不同的尺度服从于不同的物理、力学模型；通过对宏观模型的细分，不能导致细观和微观模型；相反，通过微观和细观模型的无限迭加，也难导致宏观模型。因此，在工程/产品的精细分析中，客观地会遇到多尺度模型的耦合问题。目前的CAE软件，都是仅限于宏观物理、力学模型的工程/产品分析。计算机性能和计算方法的限制是一个重要的因素。

二 Cluster体系的高性能计算机推动了工程计算
在超级计算机发展中，随着Cluster架构的日渐成熟及其系统管理技术和软件的逐步完善，并借助于Linux系统的不断发展，其构建的系统规模在不断扩大。在TOP500中，Cluster已经成为主流结构，且发展势头强劲。2006年6月发布的排名中，Cluster架构的机器已有364台，占72.8％。
工业和制造业在计算和数据处理方面的应用，由于其资金、规模以及实际解题的规模限制，对计算机机型并非越大越好，往往采用中型或小型规模的超级计算机达到的效果和加速比可能比大规模的更好。正是这样的原因，制造业新产品研发选择制造成本低，可扩展性极强，容易构造和维护，对运行环境要求低的Cluster体系的高性能计算机。另一方面，Cluster体系的高性能计算机的发展推动了工程计算的发展，工业和制造业使用高性能计算机的门槛不断降低，几乎得到了普遍的应用。
   制造业领域的另外一个主要特点是利用行业公认的商业软件进行分析和计算，特别是涉及到大型工业产品设计，如飞机，船舶，汽车等，所用软件必须得到行业标准的认可。随着高性能计算机的广泛应用和快速发展，近年来CAE软件开发公司纷纷支持高性能计算，例如MSC, ANSYS, Fluent等大型工程分析软件公司，纷纷推出并行计算版本，支持高性能计算机的应用，特别是对基于Linux操作系统的Cluster体系的支持，推动了广大制造业大规模计算应用的发展。

三、 完整的工业体系对高性能计算有大量的需求
以上海为龙头的长江三角洲地区是中国大型制造业的基地，在航天航空，汽车，船舶，核电工程等领域有举足轻重的地位。上海地区完整的工业体系的信息化，特别是产品设计和研发过程对高性能计算的需求非常迫切。
在航天领域，国家已启动中国新一代无毒、无污染运载火箭的研制工作，该火箭研制成功后将全面进入全球商用运载市场的竞争。上海航天局将承担部分总体和分系统研制工作。航天工业是高性能计算应用开展最早、应用最广泛的工业领域，有大量的问题，例如高超音速飞行器外形空气动力学设计和分析、飞行器气动热、结构耦合分析、不同载荷条件下推进剂行为模拟、火箭电磁兼容性分析等都需要大规模的计算和仿真。航天工业是中国大力发展的高新技术产业,它的发展将直接带动相关产业。

    在航空方面，中航商用飞机有限公司和中航一集团第一研究院承担的国家重点项目支线飞机的研发，高性能计算在数值风洞、气动气弹、结构强度、载荷疲劳等方面有大量的应用，进行民用飞机、军用飞机和通用飞机的组件、部件、全机的计算、分析、研究和模拟等。
    在核电工程方面，上海核工程研究院承担的国产100万千瓦压水堆核电站前期设计及国内多个核电站的设计任务，在核反应堆堆芯热工水力分析，核反应堆保护和控制分析，核反应堆主冷系统流固耦合分析，核级设备应力分析与抗震力学分析等方面需要大规模的计算。
    在船舶制造方面，上海年造船能力将从150万吨提高到1200万吨，在自主设计能力方面也必须有大力的提升，高性能计算是必然的选择手段。《经济参考报》2005年04月08日刊登的文章“设计落后拖中国造船业后腿，水平与日本相差15年”指出，计算手段落后是我国船舶设计水平与国外有巨大差距的一个重要原因。
     在汽车制造方面，上汽集团汽车工程研究院承担的具有自主知识产权的汽车设计、泛亚汽车和大众汽车向整车研发的大步推进带来的巨大的汽车设计任务。在汽车设计中整车碰撞仿真，虚拟制造（三维虚拟装配、冲压仿真等），整车空气动力学设计，虚拟试车场等需要高性能计算作为支撑条件。
中国目前是一个制造大国，正在向制造强国的方向努力，工业产品研发和设计能力至关重要，作为必需工具的高性能计算机及其相关的软件和工程化的体系将得到重视，但工程计算的应用需要一个比较长期的积累过程才能与产品设计有机地关联起来，发挥强大的作用。目前从上述工业制造业项目无法直接推算出一个精确的计算能力需求，上海超级计算中心通过市场调研，得到2006年上海地区大型制造业设计和研发对高性能计算机的峰值计算能力需求（非平均计算能力需求）将达到15.7T Flops。
四、上海超级计算中心在工程计算领域的应用和发展
    从TOP500分析可见，前100名大型超级计算机，多数是美国、日本以及欧洲那些安装在国家核心部门和由国家支持的以科研为主的超级计算中心或数据中心，主要用于学术和科研和国家信息安全。而大量的安装在工业和企业的超级计算机规模并不特别大，主要用于处理商业数据和制造业产品设计中的工程计算，由企业根据实际需求投资建设。

   我国特殊的国情决定了企业的科研投入十分有限，目前还没有企业建立高性能计算中心的案例，但企业产品创新和可持续发展，需要现代化的设计分析工具。上海超级计算中心在支持科学研究领域的高性能计算应用的同时，为广大的工业和制造业企业提供工程计算方面的服务和应用。
   上海超级计算中心成立于2000年12月，是2000年上海市一号工程——上海信息港主体工程之一，由上海市政府批准投资建设，座落于浦东张江高科技开发园区内。
   上海超级计算中心是国内第一家集高性能计算应用、科研开发与创新、技术支持与咨询、人才培养、科普教育等多功能为一体的面向社会开放、资源共享、设施一流、功能齐全的高性能计算公共服务平台。上海超级计算中心帮助科研单位提高科技创新能力，帮助企业加快新产品的开发与创新进程，提高国际竞争能力，为上海市“科教兴市”战略的实施提供服务与支撑。
   上海超级计算中心自投入运行以来，本着随需应变的理念，为上海各行各业提供了大量的高性能计算应用服务，软硬件资源得到充分有效地利用，在气象预报、药物设计、生命科学、汽车设计、土木工程、物理、化学、航空航天、船舶设计、环境保护、新材料等十多个应用领域取得了一系列重大成果，充分发挥了公共服务平台的重要作用。上海超级计算中心在拥有峰值计算能力每秒10万亿次的高性能计算机的同时，配备了得到世界上大型制造业认可的，广泛应用于工程计算的大规模商业软件，例如用于CFD的软件Fluent、CFX，用于结构分析的Ansys、Nastran等，用于非线性动力学分析的LS-Dyna、Pam-Crash等。四年多来，已经在航天航空、汽车、船舶、核电工程、钢铁、市政工程等领域开展了大量的应用推广工作，积累了工程计算方面大量的实践经验。根据业务需要，上海超级计算中心正在建设专业领域的综合设计分析平台，将零散的软件集成起来，根据行业的特点，提供功能更加强大，更加专业化的平台。目前正在建设的有面向汽车、飞机和船舶设计的三个专业综合服务平台，将提升上海超级计算中心在工程计算领域的服务水平和能力，更好地发挥公共服务平台的作用，为上海科教兴市主战略服务，为加快我国工业和大型制造业信息化进程做贡献。

    上海超级计算中心以开放，共享的，应用服务供应商的模式为全社会各领域提供高性能计算服务，打破传统计算中心的封闭模式，创造新型科研和生产手段，提供国际水准的现代化科研环境，为跨行业跨学科的技术合作创造机会。

    上海超级计算中心在提供高性能计算资源的同时，开展并行计算方法研究和自主高性能软件开发。由于缺乏并行算法和软件，超级计算机在工程应用中还远未充分发挥作用，上海超级计算中心以应用为突破口，以国产超级计算机为平台，强调并行算法设计、软件开发与工程应用的紧密结合。根据工程应用需求，研究并行计算关键技术；针对超级计算机特点，发展工程应用方法，例如，开发了基于数学模型区域分解的并行算法与软件、基于数学模型时域分解的并行算法与软件、基于力学模型区域分解的并行算法与软件、特大型工程地震安全性评价的并行算法和软件、汽车碰撞事故再现的并行算法与软件、工程应用的全三维非线性建模方法与技术，形成了一套具有鲜明特色与优势，并具有自主知识产权的解决工程应用中大规模复杂系统结构动力学问题的系统技术。在重大工程领域，解决了上海外环线隧道、上海复兴路隧道、上海光源工程、上海世博会变电站、上海液化天然气工程和上海崇明越江隧道等重大工程中采用现有方法不能解决的结构动力学问题。上海超级计算中心在高性能计算方法的研究和软件开发方面还在不断探索和开拓，结合工程和科学计算的要求，结合高性能计算机的特性，开发高效率的计算方法，解决工程和科学问题。

五、结论
   高性能计算已成为计算科学的主流，面向工业和工程的复杂问题，建立数学模型并开发高效能的软件是目前应用的瓶颈之一。超级计算中心的运行模式是一种资源聚集的比较好的模式，是多学科融和，交叉和合作的平台。在工业产品自主创新方面，CAE软件起了很重要的作用，自主CAE软件中包含了大量的产品设计理念和方法，只有自己的CAE系统，才能真正谈得上自主的新产品设计开发体系。在此呼吁国家重视自主CAE软件开发和应用。
参考文献
[1] 谢向辉，胡苏太，千万亿次超级计算机展望，中国计算机学会通讯，2006，2（1）：22～26.
[2] 黄凯，徐志伟，可扩展并行计算——技术、结构与编程. 北京：机械工业出版社，2000.

作者简介：
李根国，男，1969年2月出生，专业方向是计算数学，力学，工学博士，高级工程师，现在上海超级计算中心工作，担任中心副主任。
主要从事高性能计算研究和应用，特别擅长高性能有限元计算方法在工程和工业产品设计中的应用。先后在国际和国内重要学术刊物上发表20余篇论文，大部分被SCI、EI等检索或收录。
]]> http://www.caeda.com.cn/blog/article.asp?id=30 caeda@vip.sina.com(lns) Sun,06 Apr 2008 10:06:36 +0800 http://www.caeda.com.cn/blog/default.asp?id=30
越来越多的迹象表明，微软已经在高性能计算领域站稳了脚跟。其擅长的“跟随”战术，让所有人都很难漠视微软的存在。
　　《互联网周刊》陈琼
　　大约两年多前，比尔·盖茨曾经非常令人意外的出现在“2005年超级计算机大会”(SC05)的会场，并做了主题发言。这是世界高性能计算(HPC)领域内年度最高级别的两场大会之一，另一场是自2001年以来，每年6月固定在德国小城海德堡举行的ISC(InternationalSupercomputerConference)大会。最引人瞩目的是，在每年的SC和ISC上，都会发布一份名为TOP500的超级计算机排行榜，根据运算指标，将这个星球上运行速度最快的前500台计算机排一个顺序。截至2007年11月最新的榜单出炉，TOP500已经发布了30次。
　　盖茨当年能赏光出席SC大会和微软当时一项野心勃勃的产品发布计划有关。这个产品就是微软针对高性能计算领域内的集群系统所推出的操作系统Windows Compute Cluster Server 2003(以下简称WCCS2003)。盖茨的身体力行为WCCS2003在2006年的正式亮相做了最有效的预热。
　　两年过去了，WCCS 2003一直表现得不温不火，但是，越来越多的迹象表明，微软其实在WCCS 2003身上寄予了很多厚望，并且也为此煞费苦心。
　　2007年3月，高性能计算领域内的老牌厂商SGI宣布，开始销售基于WCCS2003的AltixXE集群系统，这也是SGI首次在Altix系列中使用微软的操作系统，这之前，SGI一直是Linux和Unix的拥趸；7月，微软与惠普(www.hp.com.cn)的“水平组合”面向企业和中端市场，推出了极具灵活性和高可用性的超级计算集群，由此扩展了双方所签署的面向高性能计算市场的全球销售和营销协议。
　　最新的例子出现在2007年11月底，微软和曙光签署了合作备忘录，双方将基于微软WCCS产品，针对高性能计算领域在市场、技术、渠道、销售层面进行合作，同时建立“曙光-微软软件技术测试实验室”，共同培训软件技术人员并定制行业解决方案。曙光多年来一直是国内高性能计算领域内的标兵，能被微软选中几乎是一种必然。根据合作协议，双方很快就会推出基于WCCS2003的机群产品曙光4000W，并准备针对4000W联手研发适用于各垂直行业的解决方案。
　　这是微软觊觎已久的一大块市场，从某种角度讲，WCCS也是微软打开这块市场的钥匙。在2007年11月底发布的那份TOP500榜单仍是Linux的天下，有381台超级计算机采用了Linux，基于微软WCCS 2003的只有6台，苹果的MacOSX系统有2台。IBM(www.ibm.com.cn)的两台“蓝色基因”超级计算机仍牢牢占据着排行榜的前两位，就像IBM对于企业市场的控制力一样。微软面前看似困难重重，但这些都不是问题的关键。
　　微软很早就已经明确表示过，其进入高性能计算市场的目标之一就是要为中小型用户带来更便宜和更易用的超级计算机，而传承自Windows的使用习惯无疑是微软手中的杀手锏之一。“对于普通的非计算机专业用户，如需要进行计算的生物学家，Linux集群则不太友好。”康奈尔大学计算机科学教授RonElber的观点就是一个典型的例子。
　　这就是微软最擅长的“跟随战术”的可怕之处，一旦微软想要跨入的领域，所有人都很难漠视其存在。根据微软的路线图，WCCS 2003的继任者Windows HPCServer2008(WHS2008)也将很快登场。届时，微软在高性能计算领域内的技术部署将会进一步立体化，微软也将很快完成从一名“参与者”到“竞争者”的蜕变。
]]> http://www.caeda.com.cn/blog/article.asp?id=29 caeda@vip.sina.com(lns) Sun,06 Apr 2008 10:03:24 +0800 http://www.caeda.com.cn/blog/default.asp?id=29 分析：2008年高性能计算的十大趋势

1.高性能计算市场仍将保持强势增长
2008年，高性能计算市场销售收入将超越过去的120亿美金（2006年的销售收入为100亿美金）。集群服务器将占据高性能计算整体份额的70%之多。集群服务器的销售也将逐渐向刀片服务器领域偏移。工作组系统（售价低于50000美金）和企业级服务器（售价为50000美金到250000美金）的销售收入将占据服务器整体收入的60%以上。全球年度服务器增长速度平均约为9%。高性能计算存储市场的全球平均增长速度将略高于服务器市场，大约为11%。高性能计算服务器和服务的销售总收入将超过190亿美金（软件应用程序，人员配置，设备和能耗的成本不计在内）。

2."Petaflop Club"将初次亮相
追寻享有盛誉的petaflops这种里程碑式的运算能力的公司数量在不断增长。"PetaflopClub"并不需要成为全球性的超级组织：瑞士联邦可能会对其投资。他们也不必非坐落在美国或者日本，可能是欧盟的某个国家或者中国，印度等，具体地址目前还是待定。在2008年看起来又会有一个或者更多系统将达到Linpack基准的Petaflop级别运算性能。这将为美国劳伦斯.贝克利国家实验室的HorstSimon开创Petaflop运算时代铺平道路，来实现颇有挑战的64位应用程序并行代码的大量petaflops运算性能的目标。美国和日本的petascale项目目前看起来在这方面都进展顺利。

3.高端市场将更具人气
随着定价高于5000000美金的高性能计算计算机市场日渐萎缩，新的IDC超级计算机种类（售价高于500000美金的系统）继续呈现增长态势，也为大型高性能计算市场和生产厂商的整体业务的技术革新提供了绝无仅有的大好时机。从2007年11月超级计算机Top500排行榜中排名前10位的榜单可以看出，在最昂贵的超级计算机分类中地理区域和生产厂商也呈现出多样化趋势。在2007年11月排名前十位的计算机系统中，德国、印度和瑞典开始与美国并驾齐驱，IBM，Cray和SGI也与惠普强强联手。Borat会在SC09大会上成为关键词吗？

4.低端市场的固有模式将开始变化
工作组系统和企业级服务器是高性能计算市场中最具增长势头的动力所在。但他们还远远不能与使用高端产品的用户相提并论。IDC2007年的研究结果显示针对低端高性能计算系统的设置比之前认为的更加变化多端，会随着数万亿美金的工业企业项目的启动而有所不同（有些项目要使用超过1000台工作组系统）。
这项研究还显示虽然这些项目于没有在科技和工程方面取得突破性进展，但他们的工作负载却涵盖了容量与创新的工作内容。低端市场的经济吸引力，特别是对于英特尔、AMD和微软这样的大型软硬件系统生产厂商来说，都会逐渐将注意力聚焦于此来吸引他们的用户。(IDC机构计划在2008年发布其对于2007年低端市场的研究报告）

5.低端市场的增长将带动便捷解决方案的发展
我们对低端市场已经知之甚多，特别是售价低于50000美金的工作组技术服务器市场更是趋于白热化。有些方面已经非常肯定：那就是这些用户都希望有一种简便快捷的解决方案--从购买，安装，操作（包括软件兼容性和性能）和升级。无论是首次进行技术服务器迁移的小型工程服务公司，还是具备多年高性能计算数据中心经验的大型企业部门，他们都缺乏真正富有高性能计算经验的IT专业人员。
2007年，高性能计算厂商将新系统的价格定位于15000美金到25000美金之间（诸如SiCortex Catapult，HPCluster Platform Express, IBM等）预测试的参考体系架构目标就是提供能满足工作组要求的便捷解决方案。PlatformComputing公司（在英特尔ClusterReady体系架构上与英特尔和戴尔的合作伙伴）的持续性增长证明了中间件的价值所在。IDC预测新型工作组平台将成为2008年颇具影响的新生力量。

6.2008年将成为功效的突破之年
在高性能计算面临的三大环境挑战之中（能耗、冷却和设备占地面积），其中之一的电力消耗将变的最为重要。2008年会把重点更多的集中在低能耗的高性能计算系统和更RFP的额定功率之内。这种趋势也将带动加速器的增长需求，对于某些问题加速器能以较低的时钟周期进行更多的并行运算来传递更高的性能。随着能源价格的上升，2008年应该会推动这种趋势以相对低的能耗向功能最为强大的计算设备领域发展。这种趋势也将最终导致高精尖科技人才将注意力集中在同样的领域来充分利用最佳的计算资源。

7.2008年蕴含加速器的发展机遇
随着X86单核频率壁垒成为过去，高性能计算厂商和用户都在多核处理器的应用程序性能方面取得了进步。但是在许多情况下这种进展还是不够的。那就意味着加速器将在2008年和不久的将来逐渐担当重要的角色。有个关键性的未知问题就是使用何种加速器，高性能计算加速能带来的商业机遇又有多大？HyperTransport和PCIExpress的互联方式是否已经足够了？如果机遇足够诱人，那么微处理器的生产厂商可能就会提供直接嵌入加速器的人性化芯片产品。
2007年，GPUs, FPGAs, ClearSpeed, IBM Cell和矢量处理器厂商(Cray,NEC)都取得了进步。2007年英特尔备受AMD的技术冲击，而AMD也缄默的固守着它高性能计算的舞台。AMD将在2008年反戈相击吗？加速器又将成为AMD反击的武器吗？

8.软件许可证授权成本将接近临界点
IDC研究数据显示，许可证授权在2007年对于许多用户而言开始到达临界点，这是受主要软件厂商继续使用针对核心数量授权的定价模式，而并没有调整他们的定价模式来适应多核硬件并行计算的快速增长的驱动。一些独立软件开发商也引进了可选性模式诸如Altair,Allinea，但是其成效还需要时间来证明。同时传统的独立软件开发商定价模式还在逐年增长的硬件系统和应用软件性能方面扮演着至关重要的角色。

9.互联方式将开始向光纤推进
200年我们将看到高性能集群计算和存储互联方式开始向光纤转变。Luxtera公司聘请了有关专家格雷格.杨并借助其QDRInfiniBand产品向市场暗示他们已经为2007年第四季度投入生产而准备就绪。来自英特尔和其他厂商的DDR每秒达到20G容量的光导电缆目前已投入市场，它比24 AWG铜芯有更大的优势（覆盖范围更广，带宽更高，错误更少）。

10.存储和计算网络将日渐一体化
随着多核处理器的迅速扩张，存储网络远远的落后于高性能计算领域的发展。这就意味着对存储容量的需求比以往任何时候都显得重要。IDC预测高性能计算存储市场的销售收入在2008年能达到48亿美金（2006年为38亿美金）。诸如EMC和NetApp这样的大型商业存储设备供应商也将更多重点放在高性能计算领域。Mellanox公司也随着DDRInfiniband互联架构的腾飞把关注的目光聚焦在高性能计算市场，这些都促使高性能计算存储和计算网络日渐一体化。]]> http://www.caeda.com.cn/blog/article.asp?id=28 caeda@vip.sina.com(admin) Sat,05 Apr 2008 08:04:11 +0800 http://www.caeda.com.cn/blog/default.asp?id=28
Although the process of developing and using a user subroutine is straightforward, it is not easy at first. Many users have trouble getting started. If you fall into this category, this tip is for you.

This tip will take you through a specific example of making and using a UCOND user subroutine. This example is a simple heat conduction problem in a square, where the conductivity is higher in a circular region inside the square. UCOND is similar to UVISC or UDENS and many others --- one of many user subroutines that let you change specific material properties inside a given volume.


--------------------------------------------------------------------------------
Prerequisites: In order to compile user subroutines, you must have a Fortran compiler. For Windows machines, we use the Compaq Visual Fortran V6.5 compiler. For Linux, we use NAGWare Fortran 95 V5.0. For other Unix platforms, we use that system's native compiler. We do not guarantee or recommend other compilers, but they may work too.

Step 1 – Model Creation. Using CFD-GEOM, create the 2D square grid shown in the figure (21 x 21), and save a clean DTF file.

Step 2 – Model Setup. Set up the model in CFD-ACE-GUI. Pay special attention to steps (c) and (e) below:

On the PT panel, turn on Heat
On the BC panel, pick the left wall and set it to ‘Heat Flux’ and specify q = 100 W/m2. Similarly, pick the right wall and set it to ‘Isothermal’ and specify T = 300K.
On the VC panel, pick the volume and change its conductivity to ‘User Sub(ucond)’
On the SC panel, set the number of iterations to 1000. On the Relax tab, set the enthalpy relaxation to 5E-06.
On the Run panel, indicate the name of the file containing the user subroutine you will be writing. Let's call the user subroutine "my_UCOND" .









--------------------------------------------------------------------------------
Step 3 – User Subroutine Development. Follow these steps to create the user subroutine:

Copy the template file libUserAce.f over to your working directory, and rename it to my_UCOND.f The template file libUserAce.f is located in the bin directory of your software installation.
Edit the file my_UCOND.f with a non-formatting text editor or programming editor. On Windows, Notepad will work fine, but don’t use Microsoft Word or Wordpad, because these add extra hidden formatting into the file. You could also use the editor that came with your Fortran compiler.
Find the UCOND user subroutine. The template file starts by defining a MODULE named cfdrc_user at the top, containing some general parameters. Then, the file contains 50 empty templates for user subroutines. UCOND is the 5th one. In version 2003, UCOND starts on line 448 and ends on line 499. You are free to delete those user sub templates that aren't being used.
Type in the coding shown below inside the UCOND user subroutine template:


In order to understand what each line is doing, you can refer to the User Subroutine chapter in the CFD-ACE+ User Manual. Here I'll just give a short description.

The purpose of lines 15 through 17 is to get identifiers for each of the variables you need to get or set.

Line 19 gets the ‘active cell index’. Each iteration, the solver will call your UCOND for every grid cell in the volume, one cell at a time. Thus, there is a concept of the ‘current’ or ‘active’ cell. Some other routines work the same, but others work differently. For example, a UBOUND is called for a given boundary one face at a time, but a UINIT or USOURCE is called for an entire volume at a time. These things are well-documented both in the user manual and in comments inside the template user subroutine file.

Lines 21 through 29 determine what conductivity value to use, depending on the location of the current cell. Finally, line 31 sets the conductivity for the current cell.

--------------------------------------------------------------------------------
Step 4 – Compile the User Sub. Now open up a DOS or Unix shell, change directories to your working directory, and type the following command at the shell prompt:

build-userlib my_UCOND

Note the absence of the “.f” on the filename. You must use the build-userlib script to compile the user subroutine, because it links the appropriate libraries for the solver.

Step 5 – Run the Simulation. Go back to the Run panel of CFD-ACE-GUI and press Run. Wait a few seconds and then launch CFD-VIEW to look at the results. You should get the results shown below. Because of the higher conductivity region, the overall resistance to heat flow is decreased, resulting in a lower temperature on the left wall. Also, the temperature contours are no longer vertical.


So what would you do with this user subroutine? In some situations, it can be a coarse but easy alternative to creating a mesh with two distinct volumes, especially if the position of the high conductivity region changes with time. For Flow problems, you could use block off certain regions by using Porous Media and writing a UPERMEABILITY user subroutine.


The files used for this example can be found by clicking here.


If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know.

]]> http://www.caeda.com.cn/blog/article.asp?id=27 caeda@vip.sina.com(lns) Tue,01 Apr 2008 14:56:59 +0800 http://www.caeda.com.cn/blog/default.asp?id=27
China HPC Club 首次聚会报告

到会人数20人，另有5人申请成为会员但因各种原因未能到场，详见如下：



会议议题：

1.成员基本情况介绍

2.中国HPC发展机遇和机会

3.HPC沙龙后期活动方式讨论

会议决议：

建立基本组织，以班级为单位，首次选举班组委成员如下：

班主任：张云泉

班长：覃燕

组委：陈健

宣委: 王玲

CFO：付鸿雁

支书：刘利萍

Club长期议题：

如何推进HPC在中国的发展，如何解决用户应用HPC的问题，如何将市场做大，让从业人员都能从中受益

让用户认识到软件是应该付费的，服务是有意义的，促进行业的健康发展，在用户成功的同时让厂商和从业人员也有长足收益

建议今后主题议题：

近3-5年内，HPC市场的增长点，需要班主任提供分析报告

热点技术讨论，大家可以提出关注的技术热点，我们投票表决热点议题



未来组织：

每2-3月组织一次活动，基于大家兴趣将以Outing为主（地点投票之后班组委决定），同时每次准备1-2个热点议题供讨论

平时的交流主要通过chinahpc@yahoogroups.com邮件交流，如果有长期的讨论可以依托网站论坛进行，目前已经提供的两个网站是

http://www.winhpc.org.cn

http://www.caeda.com.cn

大家有什么建议和意见请提出，如果愿意单独交流的也请发邮件给相关的班组委员

BTW

请收到邀请的同学接受邀请邀请就可以加入俱乐部成为会员]]> http://www.caeda.com.cn/blog/article.asp?id=26 caeda@vip.sina.com(lns) Sun,30 Mar 2008 08:50:34 +0800 http://www.caeda.com.cn/blog/default.asp?id=26

CFD-FASTRAN， CFD-ACE+, 技术支持工程师
工作性质：兼职
要求: 硕士在读一年以上，曾有相关工作经验者优先, 必须曾熟练使用此软件，或对相关CFD商品软件很熟练的。
有个人笔记本电脑熟悉者, 有实际CFD仿真项目经验者优先


多目标设计优化软件(IOSO NM), 技术支持工程师
工作性质：兼职
要求: 硕士在读一年以上，或有相关工作经验者, 必须曾熟练使用相关商品软件(指类似于iSight, Modefronteir, 等)
有个人笔记本电脑熟悉者, 有实际仿真项目经验者优先

请讲简历E-MAIL至: caeda@vip.sina.com, (建议在发简历或联系前务必访问我公司网站, 熟悉我公司IOSO NM产品, 以便面谈)
公司网站: www.caeda.com.cn, www.caefinder.com






]]> http://www.caeda.com.cn/blog/article.asp?id=25 caeda@vip.sina.com(lns) Sun,30 Mar 2008 08:47:51 +0800 http://www.caeda.com.cn/blog/default.asp?id=25
随着CAE/CFD软件在工程设计中的大量应用，越来越多的设计分析工程师发现，对于产品设计仿真流程的掌握不是一件十分容易的事。不单是各种软件使用、软件接口问题，而且牵扯到不同学科、及设计方法的技巧和经验问题。

CAE应用课程在国外发展很快，而其中的Know how也形成技术各种文件广为流传。CAEDA正是将这种典型应用收集起来，并和相关国内外CAE专家一起，将这些方法和经验编辑成“A+B”的课程，传授给大家。
通过“A+B”课程学习和实际操作，可以相当程度上理解掌握分析仿真方法，充分利用好软件的功能，并有机会和专家及同行建立关系，交流学习。

欢迎预约参加CAEDA公司2008年4-5月间举办的以下CAE软件专业技术培训：

课程名称    培训班（北京）    E-learning         远程培训    客户现场培训

ESACOMP4.0 + Patran
复合材料设计分析软件培训
培训类别：标准培训
课程：1.5天
时间：4、5月                （选择）           （选择）    （选择）

RTM-Worx 2.9+ Patran
RTM树脂模注工艺仿真软件培训
培训类别：标准、应用
课程：1.5天
时间：4、5月            
CFDesign9.0 + CAD

流体分析仿真软件培训
培训类别：标准、应用
课程：2天
时间：5、6月        
    
TOSCA 6.1.1+ FEA（UGS/NX）
结构拓扑优化设计软件培训
培训类别：标准
课程：4天
时间：4、5月            
说明：

1）以上培训班具体时间、地点、课程内容等，预约后，再通知确认。
2）培训费用：标准培训班费用：按每人600元/天（含资料、上机、午餐、证书）。高校老师学生半价，食宿请自理。E-Learning及“客户现场培训”按具体情况协商。
3）对于本公司已购买软件客户，免2人培训费。如即将购买软件客户，参加此培训后（2人免费），即等同于完成售后标准培训。
4）培训讲师：厂商授权本公司工程师，及“CAE专家银行”成员，TSP (Tosca School Program)、RTM School Program合作伙伴讲师。
5）有关培训具体事宜，请联系：caeda@vip.sina.com, 电话：010-6847 7947，短信：1390 106 4208，传真：010-8257 0138，刘先生。我们可根据客户具体需求调整充实培训内容。
6）上期报名的客户，因我们没有回复确认，而没参加成的客户，免费参加这期培训。


您的联系办法：

姓名：__________,  单位：________________________________，预计参加人数：_______

联系电话：_________________________________, Email（重要）：______________________
]]> http://www.caeda.com.cn/blog/article.asp?id=24 caeda@vip.sina.com(admin) Sat,29 Mar 2008 08:28:50 +0800 http://www.caeda.com.cn/blog/default.asp?id=24


Figure 1. Levels Editor in CFD-GEOM

Fuel cells are a good example where there are different sections of the geometry that could be added to Levels. In a fuel cell, there are many different layers and each layer could be added to a different level, as shown in Figure 2.



Figure 2. Fuel cell with each layer on a different Level. Each Level has been colored differently.

There are several features to the Levels Editor:

New – Create a new level
Delete –Delete a Level (Note that Levels are only a visual grouping, and thus no entities will actually be deleted.)
Add Entities – With entities selected, they can be added to a Level.
Remove Entities – With entities selected, they can be removed from a Level.
Hide – Visually hide the selected Level.
Show – Visually show the selected Level.
Color – Below the Level name text field, a color-well is available for specifying a Level color.
Entities can belong to multiple levels which allows shared entities between blocks/domains to be shown on multiple levels. The selected Level is also known as the "Active Level". Any newly created entities will be placed on the "Active Level". The "Active Level" can be changed from the Level Editor, or by selecting the desired Level from the Levels drop-down dialog located near the bottom of the graphics area, as shown in Figure 3.



Figure 3. Levels drop-down dialog in CFD-GEOM.

In order to see the different Levels in their assigned color, the color mode must be set to "Level". The color modes can be found in the Entity Colors toolbar, as shown in Figure 4. There are three available color modes: Color by Assigned Color, Color by Level, Color by Entity Type.



Figure 4. Available color modes in CFD-GEOM.

Levels can be very helpful when working with geometries with multiple parts. The next time you use CFD-GEOM, try using Levels to visualize and work with different parts of your geometry.

If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know.

]]> http://www.caeda.com.cn/blog/article.asp?id=23 caeda@vip.sina.com(流浪) Thu,27 Mar 2008 23:30:23 +0800 http://www.caeda.com.cn/blog/default.asp?id=23
SALT是内置于RTM－Worx的应用程序接口（API）。通过它可以实现：
※模块基本控制
※生产变量模块
※实现动态边界条件模拟
※参数自动统计分析
...............]]> http://www.caeda.com.cn/blog/article.asp?id=22 caeda@vip.sina.com(lns) Sat,22 Mar 2008 09:47:51 +0800 http://www.caeda.com.cn/blog/default.asp?id=22
以下是在德国做结构优化设计咨询的准备文件，给大家分享，CAEDA公司也可以提供算例。

此文本的目的就是来描述为什么初始项目和现场培训是我们销售策略的一个决定性的重要部分，和伴随战略执行和实施我们怎样继续进行下去。

对我们的前景来说，建议初始项目和定点培训被重新包装（软件允许期限，培训和咨询相结合）以便于：

1.    只要开始使用TOSCA就可以让客户马上产生生产力
2.    开始TOSCA的实施阶段就会伴有对设计过程产生影响的现实问题出现
3.    降低对客户端的工程量
4.    避免免费的标准试用版
5.    避免在销售环节出现折扣纠纷
6.    通过现场展示TOSCA的能力来激发客户需求
7.    更好的了解客户的痛苦、存在的问题和组织问题
8.    降低销售费用

初始计划和现场培训进行重新打包的建议是一种特洛伊木马策略以便于了解客户的痛苦和通过为客户提供特定需求的相关培训来嵌入到个体中去，这种培训是基于一个固定费用比率来取代单独的提供培训、评估许可证和咨询服务。（通过可见的和有可比性的价格）


初始项目的首要事项＆规则：
根据第一印象最深刻的原理，同客户的第一个项目将是有最深刻的影响（积极的或者负面的），因此第一个项目必须要成功
成功的关键问题在于有一个详细的计划，也就意味着，项目的选择和现场实施的准备工作必须要做的非常详细和具有策略化。

一些最重要的首要事项＆原则：

1.    客户/用户必须有FEA的经验
2.    客户端至少有一个全值人员专注于这个项目
3.    我们紧紧选择那些我们确信能够传递积极影响的和对客户能够产生可以测量的价值的项目
4.    定义那些能够通过项目受到影响的部门和个人以便于来评估对他们的价值。
5.    识别他们的痛苦（必须要改变的）并跟踪他们的痛苦链（通过项目的效果其他的那些也可以受益或者收到影响的人）
6.    是否存在唯一的原因（压力）来开始这样的一个项目，如果答案是的，为什么不是我们来做呢？
7.    避免“我认为我们可以比别人传递更好的效果。。。”这样的项目因为浪费时间在教育他们而不是让他们陷入进来
8.    在投递我们的建议（对理想顾客）前就要定义对这个项目而言可以测量的成功标准，或者至少在开始这个项目前定义
9.    获得更好层次管理的委托，项目已引起他们的关注和我们被允许简短的来报告结果。
10.    在投递报价前，同客户匆匆过一遍“咨询＆初始项目清单”
11.    传递一个包含附加的执行协议的确认信和一个完整的报价而不是一个用剪切/粘贴式的价格列表条目的标准格式的信件。

咨询＆初始项目点检表：
1）    项目的主要目的是：
a).目标最优化
  ⅰ）减轻重量（减轻多少？）
   ⅱ）应力多高
   ⅲ）耐久性能多低
b) 设计所要考虑的限制条件：
   ⅰ）重量目标
   ⅱ）应力限制
c) 应该所做的事情（可能基于我们的建议）：
   ⅰ）有限元分析（线性，非线性）
ⅱ）拓扑优化
ⅲ）形状优化
ⅳ）条纹优化
ⅴ）完成设计，在载荷和限制条件下的设计空间开始
2）    项目里程碑
a)    项目的尺度是什么（通过技术方面，预算和时间验证）
b)    有没有需要特殊批准的预算限制
c)    项目的时间框架是什么（开始日期，持续时间，几个阶段）
3）    对这个项目来说可以利用的数据
a)    一个有限元分析是否已经被做
  ⅰ）有限元模型的产生所用的时间和努力
  ⅱ） 有限元求解器（线性，非线性）的计算时间
b)    是否载荷和边界条件已知
c)    材料数据是否已知和可用（ASCII表首选）
  ⅰ）钢材、铝
  ⅱ）塑胶
  ⅲ）橡胶
d)    用哪种格式原始数据是可用的
  ⅰ）3D CAD数据（来源于任何系统、任何界面格式）
  ⅱ）有限元的输入格式（来源于任何求解器）
4）    项目怎么被实施
a)    客户现场（所有的时间、部分、几个阶段）
b)    FE-DESIGN内置（不需要被推荐）
c)    需要哪种类型的培训（基础，进阶）
5）    对于优化任务来说：对模型来说哪些允许修改和哪些是固定限制的（要么客户通过自己来做或者以书面的形式同意我们来做）
a)    CAD系统中设计空间的定义（怎么样？）
b)    FEA模型中设计空间的定义（形状优化：节点数，对拓扑优化：网格数）
  ⅰ）对设计部件所考虑的区域应该被收集在一个“DESIGN”群组中或者必须被重新编号
ⅱ）不被允许修改的区域应该被收集在“FROZEN”群组中或者必须被重新编号
ⅲ）在特定加工限制情况下允许改动的区域应该被收集在一个群组中或者重新编号。
6）    给客户汇报结果的过程（格式，程序）
a)    谁来进行CAD设计（同委托人一起或者客户自己完成）
b)    哪种数据模型必须被转换
c)    哪种报告程序和格式（内容和容量）
d)    局部项目的叙述和总结
7）建议事项
  a) 是否需要一个保密协议
  b) 从客户端哪种数据是可用的以便于提交一个准确可靠的建议（原则：完整的数据－固定的价格，仅有少量数据－变动的价格）
     ⅰ）问题的屏幕捕捉
     ⅱ）CAD 数据
     ⅲ）FEA模型
  c) 提交建议的时间框架
  d)建议内容和里程碑
  e)在时间框架里，谁在对客户端做出决定


对一个小的但在风能事业领域具有全世界领导地位的公司的初始项目的例子：


XXX
德国
项目建议：MM系列的支撑架的优化
Dear Mr.

基于通过电话和同你们小组会面的相关交流，在交流过程中你们向我抱怨在设计和计算风能特征时在当今社会和全球竞争下必须要较短的产品设计开发周期，现在我很荣幸给你提供一个关于优化MM系列铸铁支撑架的建议。
这个项目主要目的是：在一个非常短的时间内（整个过程大概4－6周）来开发一个最佳重量的支撑架，但是必须满足设计标准限制和计算限制（例如不允许有塑性变形，高的循环疲劳）
我们项目建议包含详细的关于问题定义的描述，服务规格，从你那里的所要求的相关介入如同我们建议的经济型报价一样。
我们希望我们的建议能够满足你的期望。
至于更深层的东西请毫不犹豫的联系我们。
我正渴望能同你和你的公司进行一个成功的合作。


Best regards.
XXX


对风能系统公司的“MM系列的支撑架优化”的项目规格书
（以下称客户）
A)项目描述
初始状况/问题定义
风力发动机元件尺寸的不断增大，更高性能的要求和客户要求，还有被德国爱派克（Lloyd）所要求的更加苛刻的设计和制造限制，已经导致了一个越来越复杂的风力发动机特征的开发周期。
目前最要的开发目标是发现一个最可能的折中方案，其中设备的寿命要长，但是维护间隔要大，并且有显著的重量降低甚至更高的安全标准（例如运输和近海岸工厂的组装）
今天模拟还分析软件的使用允许我们甚至在设计早期阶段都可以获得相当的精度和可靠性设计标准以至于可以预测设计的耐久性。由于设备重量的重要性和对不同部件产生的正的/负的影响，在过去传统的用时间和金钱的花费来最优化零部件。由于这些方面的原因，一直存在较长的开发周期和高昂的开发成本。
对于一些原因而言，手工设计优化方式是非常不现实和不理智的，因为：
1)    时间和造价非常高
2)    在设计阶段由于预算和设计限制，在设计阶段利用最可能的变化中较优的设计被中止，但没有达到最佳优化。
解决方法的建议：
基于有限元分析和耐久性预测的自动优化程序将导致元件和设备设计过程的基本性的简化和降低。通过使用优化系统TOSCA，你可以利用初始的设计空间和耐久性的限制条件下来开发一个非常接近目标的较低重量的结构设计建议。利用拓扑优化的第一步，材料就会被从设计空间移除和一个新的设计建议被产生。在转换成CAD设计后，下一个优化链将被执行通过较少的修改使得表面应力均匀化。你也可以从耐久性分析中优化损伤因子。通过这些开始的联合项目你可以在将来自己运行和实施类似项目。

真个项目被分割成下面将被定义的几个步骤：

阶段一：拓扑优化的设计建议的提出：

开始现场TOSCA培训，接下来的辅导期的目标是开始于给定的设计空间来进行拓扑优化设计。对于这些目的，FE-DESIGN的员工将在客户工厂的现场以便于主持培训（两天）和完成辅导活动（三天）。回到家后，将被FE-DESIGN给予两天额外的技术支持，这些支持可以通过电话，MAIL或者WebEx来进行。

拓扑优化的首要条件：

客户必须预先设计空间的FEA模型。计算时间应该适合指导期限，不应该超过1.5小时。当然，也可以生成其他不同的几中模型，这些模型由于不同的网格大小和不同的工况就会有不用的计算时间。这些在耐久性分析必须被考虑的工况，在拓扑优化时必须作为单独的静态工况被定义。这些输入条件也必须事先被客户准备好。
在优化环节中，所谓的优化程序文件将被生成，这些文件在后面的分析中容易被较大的FEA模型所采纳。这样当辅导期间过后，客户有可能自己运行较大的模型。

第一个阶段完成后，就会有一个可以被客户使用和执行的设计建议。

阶段二：详细的形状优化

阶段二开始前，客户必须阐明在耐久性分析中的最大损伤位置能否被单独的静态工况所表达。依靠这个客户的阐明，在下一个优化环节中将会考虑或者不用考虑耐久性模拟。这些准备和优化是现场指导项目的一部分，并且将在客户端的场所实施。总共有五天的现场技术支持的计划安排和两天的可以通过电话、email和WebEx方式的时候技术支持。

形状优化成功实施的首要条件：
客户必须设计出加载有相关载荷和边界条件的FEA模型。客户决定是否有耐久性综合分析的必要性。（如果是≥Pos.2,如果否≥报价可供选择的Pos.3）

客户必须关注FEA模型合理的离散化，也就意味着在比较光滑的表面轮廓和指导期间合理的计算运行时间找到一个最恰当的平衡。

在优化环节中，所谓的优化程序文件将被生成，这些文件在后面的分析中容易被较大的FEA模型所采纳。这样当辅导期间过后，客户有可能自己运行较大的模型。

如果耐久性分析直接被集成在优化环节中（参照可选择Pos3）,客户必须确认耐久性的代码能够以批处理的方式运行（意思是：交互式开始不被允许和运行过程中不允许有干扰）

另外下面的工作需要FE-DESIGN来完成：
1）    开发和执行命令解释程序脚本，这些程序用来驱动所涉及的一些代码的程序步骤。
2）    所需要的结果文件作为一个节点和相关损伤因子的表格来产生，所有这些都要基于客户的耐久性系统导出的ASCII码结果文件。

对于这个项目的执行过程中，在项目开始后，FE-DESIGN给予一个两个月使用期限的TOSCA的许可证以至于保证执行的无缝和顺畅。

更深层次的FE-DESIGN的服务通过努力将会收费。

文件编制
这些程序和结果文件按照要求被递交并且将被分别收费。
A)    所有数据将通过电子方式传送。

B)    里程碑和时间表
   基于我们的经验，项目所花费的时间要比纯工作时间高，因为在过去的项目运行时间中，FE-DESIGN将不能连续工作在项目上而是紧紧针对客户的特定需求。
整个项目所用的时间严重依赖每个FEA工作运行的时间。
预计的项目所用时间是根据平均每个FEA工作2个小时的运行时间来估算的。显著长的FEA计算时间，项目所持续的时间也就较长。
情况    描述    FE-DESIGN的工作时间    每个过程所有的时间
1    执行拓扑优化    7天    大约2周
2    无耐久性分析的形状优化    7天    大约2周
3（可选择）    有耐久性分析的形状优化    11天    大约3周

XXX
德国                                                             时间：XXX

报价单

情况    描述    价格（欧元）
1    MM系列的支撑架的拓扑优化    
2    未集成耐久性分析的MM系列的支撑架形状优化    
可选择        
3    集成耐久性分析的MM系列的支撑架形状优化    

小计不包含可选项情况3：XXX欧元

所有价格不包括增值税

所有差旅＆住宿费用将另行收费


对于这个项目的执行过程中，在项目开始后，FE-DESIGN给予一个两个月使用期限的TOSCA的许可证以至于保证执行的无缝和顺畅。

FEA的许可代码或者耐久性分析将有客户提供

FE-DESIGN被允许在每个项目情况完成后对客户开出清单

上面提供的价格一个月有效

根据我们最好的知识水平来估计提交时间，最终的投单日期后面有客户来定

所有不合理的或者由客户引起的不能接受的增加的努力，FE-DESIGN保持要么重新谈判契约条件要么收回购买订单的权利。



]]> http://www.caeda.com.cn/blog/article.asp?id=21 caeda@vip.sina.com(lns) Fri,21 Mar 2008 08:20:27 +0800 http://www.caeda.com.cn/blog/default.asp?id=21
The Group/Ungroup tool can be found under the Geometry explorer list, as shown in Figure 1.



Figure 1. Group/Ungroup icons in CFD-VisCART

Creating a Group:
To create a Group in CFD-VisCART, simply select the set of surfaces to be grouped and click the Group icon, as shown above in Figure 1. Next, you will be prompted to specify a name for the group. Once a name has been specified, select Ok and the group will be shown in the Identity Groups explorer, as shown in Figure 2.



Figure 2. Identity Groups in CFD-VisCART

In this example, there are four groups: inlet, outlet, hot_walls, cold_walls.

When a group is selected, only entities that are in that group are displayed in the graphics area. This is can be very helpful since you can do outline detection, critical feature detection, splitting, etc. on the entities in this group.

To delete a group, simply select the group in the Identity Groups window and hit the Delete key. The group will be destroyed, but the surfaces in the group will still be in the Geometry explorer.

If you haven't already upgraded to the latest version, this is one of the many new features available in CFD-VisCART V2008.0.

If you have any questions about this feature or would like us to discuss some other topic in the future, please let us know.

]]> http://www.caeda.com.cn/blog/article.asp?id=20 caeda@vip.sina.com(lns) Tue,18 Mar 2008 09:18:45 +0800 http://www.caeda.com.cn/blog/default.asp?id=20
在ESAComp的主窗口中集成了用于数据库浏览的新的图形化用户界面, ESAComp的数据库、算例内容(数据文件)和当前活动项目均通过树形结构显示，可以通过拖拽来实现目标复制。

在新的数据库浏览器中集成了新的增强的数据库搜索能力。

依赖于温度和湿度环境的材料特性。需要通过内插法和外推法来求解。基于给定的项目环境, 在所有ESAComp的分析中采用适当的数值。

层的强度设置：A值和B值，用户自定义的设置(疲劳极限设定等)。用于分析的强度设置的选择将结合环境的选择来完成。

在板和梁分析中可以施加多点载荷。

新的多目标分析：板、梁、连接接头分析(例如单板多载荷、多板单载荷或者多板多载荷)。

对单向单层和其他的增强单层来说可以单独选择失效准则。允许默认使用高级单向失效准则(例如PUCK准则)，其他的增强类型，甚至在同一层合板中存在不同的增强类型时都可以采用适当的失效准则来处理。

增强了多种分析类型：夹层板的载荷响应/失效分析、层合板的概率分析、机械连接接头分析等。

新增了与LS-DYNA的接口：可以从ESAComp输出层合板构造和相关材料数据到LS-DYNA。

增强了与ANSYS的接口：支持ANSYS 11.0版本新增的SHELL 281单元，在后处理中可以进行层间剪切强度计算。

从其他应用软件输入到ESAComp时可以自动对单层进行分类。在ANSYS和MSC PATRAN后处理界面中，可以识别输入的芯层材料，相关的夹层板分析在ESAComp的后处理中完成。同样地，在适当的时候将采用�