Finite Element Analysis (FEA) Products and Developers

[Astronuc]

These are the one with which I am most familiar. There are others, but these are the biggies. We use ABAQUS as well as our own proprietary methods, particularly for dynamic, large strain and nonlinear FEA.


ABAQUS (HKS) - for nonlinear Finite Element Analysis (FEA)
www.hks.com
http://www.hks.com/about/about.html
http://www.hks.com/products/products_overview.html - Products


ANSYS (Ansys, Inc) - Finite Element Analysis (FEA)
http://www.ansys.com/
http://www.ansys.com/products/default.asp - Products

* ANSYS purchased CFX (CFD code system from AEAT a few years ago)


NASTRAN (MSC Software Corporation (OTC: MNSC) - Finite Element Analysis (FEA)
http://www.mscsoftware.com/about/?Q=135
http://www.mscsoftware.com/products/nastran.cfm - NASTRAN


ALGOR (Algor, Inc) - Finite Element Analysis (FEA)
http://www.algor.com/service_support/about.asp
http://www.algor.com/products/default.asp - Products

 

[Astronuc]

I hope others will add in companies and other sites that I missed.

Here is a subsidiary of ANSYS, ICEM CFD Engineering, which does CFD, or FEA applied to dynamic fluid analysis.

http://www-berkeley.ansys.com/

A nice little comment on CFD - http://www.cranfield.ac.uk/sme/cfd/whatiscfd.htm

 

[PerennialII]

I'll add my favorite repository of links especially if you're after free code:

http://www.engr.usask.ca/~macphed/finite/fe_resources/node14.html

...and my favorite continuum mech FE research code, which I use for about 75% of my related analyses and quite a large junk of new code is written for/in Warp3D in our research group as a whole:

http://cern49.ce.uiuc.edu/cfm/software.html

 

[brewnog]

I remember someone here telling me about Warp 3D, which is supposed to be pretty good although I've never used it, can anyone remember who told me about it?!

Edit: Ahh, that would be Perennial then!

 

[Astronuc]

A couple more, but I am not familiar with these:

http://www.intes.de/ - PERMAS Finite Element Software System.
http://www.intes.de/english/projects
http://www.nafems.de/ (auf Deutsch)

http://www.comsol.com/ - Multiphysics (?)
http://www.comsol.com/products/

And another package we use, but I don't know much about it.

ADINA (Automatic Dynamic Incremental Nonlinear Analysis) - ADINA R & D, Inc.
http://www.adina.com/
http://www.adina.com/products.shtml

The non-linear part is key. Not too many packages necessarily do it right!

 

[PerennialII]

I remember someone here telling me about Warp 3D, which is supposed to be pretty good although I've never used it, can anyone remember who told me about it?!

Edit: Ahh, that would be Perennial then!

... I'll continue my add then (without "too much" of an affiliation btw): completely free, has properties equal and in many areas beyond the "commercial monsters" and complete access to source (along with the usuals like extremely resource efficient & robust & continuously under development).

 

[PerennialII]

http://www.comsol.com/ - Multiphysics (?)
http://www.comsol.com/products/

And another package we use, but I don't know much about it.

ADINA (Automatic Dynamic Incremental Nonlinear Analysis) - ADINA R & D, Inc.
http://www.adina.com/
http://www.adina.com/products.shtml

The non-linear part is key. Not too many packages necessarily do it right!

... if you want to go dynamic, Adina has a rep of being a stable and accurate in complex nonlinear problems (involving usually contact, that being nonlinear as well).

Comsol was previously known as Femlab, which is yes essentially a multiphysics code. Meaning, that it suits very well to problems involving different types of couplings between various PDEs (can be several, unidirectional or totally coupled etc., have used it e.g. in solving coupled mass diffusion-thermal-mechanical-damage-fluid problems, great code for such problems). One of its strengths for a researcher in particular is access to the PDE implementations, can easily implement own modifications (or totally new eqs and solve using the FE kernel), introduce own material models etc. + It's adaptive in many directions (like hp-FEA and some other aspects as well).

 

[Astronuc]

Morbius posted an article from Physics Today in Nuclear Engineering which had an add for PDESolutions. Anybody use this? It seems to be a Multiphysics package. The example shows dynamic meshing - pretty cool.

FlexPDE - http://www.pdesolutions.com/

 

[PerennialII]

... I've considered it kind of like the little-bro of Comsol/Femlab. A scripting multiphysics FEA package with a h-adaptive mesh generator, nice package for numerically solving own PDEs (=customizing for whatever equation and equations with any degree of coupling) and comes with a price tag a tad more decent than its bigger-bro(s).

 

[Astronuc]

TeraGrande

A relatively new product.

TeraGrande represents a single, integrated, general-purpose, finite element application for all your analysis needs. You will find a true seamless multiphysics capability in TeraGrande. You can perform an analysis sequence of mechanics procedures. Each procedure in the TeraGrande library can inherit its state from any previous procedure in the sequence (including restarting from a saved restart file). There is no need to run one program, save a file, and then run another program with a restart file. All the analysis procedures in TeraGrande provide full large deformation/rotation implementations with a complete element library of solids, shells, beams and trusses in two and three dimensions. We provide a robust material library including elasticity, plasticity, concrete and user materials.

http://www.terascale.net/products.php

 

[Mitra]

Random vibration and FEM

I am wondering if deterministic method such as FEM is reliable for computation of random response to dynamic system.

We are developing a method to predict the response of a flat shell element to a pressure filed arising from the fully developed turbulent boundary layer of a subsonic flow. Fluctuating wall pressures induced by fully developed turbulent boundary layer are random, so we must rely on the statistical properties of the turbulent flow, such cross spectral density and cross correlation functions.