11-03-2006 03:30 PM
11-06-2006
03:50 PM
- last edited on
01-15-2026
03:21 PM
by
Content Cleaner
Don,
I have not seen applications built in LabVIEW that perform finite element modeling. While I am not entirely familiar with the intricacies of FEM, I do know that the process entails a great deal of advanced mathematics. LabVIEW does have VIs for performing various math functions including differential equation solvers. If you wanted to use LabVIEW for FEM, you can also consider the use of the MathScript node and the MathScript window to write m-script code for these mathematical functions.
I may be able to provide additional information if you could provide an example of your FEM application. How intricate is the model for the object that you will be modeling? For example, will you be analyzing forces and torques on a beam, or will you be modeling an asymmetrical collision of a car into a wall? Do you know the sort of calculations will your application entail?
At this time, LabVIEW does not have an easy means of communicating with a 3rd party FEM software. Again I would ask in what respect you would like to interact with alternative software. Do you want to see a visual representation of the physical object that you are analyzing? LabVIEW can easily interact with SolidWorks software for 3D modeling, but not a software package that would perform finite element analysis. Would you like to develop a front panel for an analysis application that is performed in another software? This may be possible if we further discuss the details.
Finally, I can encourage you to submit any product suggestions at the National Instruments Product Suggestion Center. Our LabVIEW developers carefully consider each of these suggestions, and any input you have as to how our application can better serve you would be greatly valued.
I apologize that I do not have a more concrete answer for you, but I hope that this information was useful for you. Please let me know if you would like to discuss this application further. Thanks,
Mike
National Instruments
Applications Engineer
11-07-2006 08:16 AM
11-08-2006
04:59 PM
- last edited on
01-15-2026
03:22 PM
by
Content Cleaner
Hi Don,
Yours sounds like a very interesting application. I can definitely understand why you would want to create a UI as well as a data viewer in LabVIEW.
As you mentioned, LabVIEW has the capability to interact with other software programs primarily through ActiveX or .NET. In order for this to be possible, the other software program must be designed with an ActiveX server. This would depend on the software package, but essentially a set of objects and methods must be registered with the operating system in order for LabVIEW to utilize ActiveX. This is similar to defining how the interaction would work between the software and ActiveX. For a brief description of ActiveX and how it works, please take a look at the ActiveX Overview.
I do not know if there are FEM tools out there that are ActiveX ready, but this can factor into your selection of a tool. I also noticed that the link in my previous post appears to be broken; I would also suggest that you submit your idea for a FEM LabVIEW toolkit to the Product Suggestion Center. Thank you very much for your time. Have a great day!
Mike
11-09-2006
07:42 AM
- last edited on
01-15-2026
03:22 PM
by
Content Cleaner
Hi Mike:
We found this link for at least the visualization aspect of model output using Diadem NI-Insight plug-in but looks anything but straightforward.
http://www.ni.com/swf/presentation/us/niinsitetut/default.htm
Yes, I did go ahead and submit a product suggestion.
Sincerely,
Don
10-05-2012 02:20 PM
Hello , iam from colombia.
i am writing code in labview for develop the finite element method. i will devolop code for calculate estress, strain, forces, and FS wilth element unidimensional beam and truss, bidimensional CST, LST, Q4, Q8 and Q9. it is possible !!! if u know the mathematical formulation.
Sebastian
ING Mechanical.
10-08-2012
05:23 PM
- last edited on
01-15-2026
03:24 PM
by
Content Cleaner
Sebastian,
LabVIEW does not have a specific toolkit used for FEA, as mentioned above. Multiple product suggestions have been made for this toolkit, and I encourage you to make another. You do have the ability to create your own application to run FEA for all of the referenced simulations, but this would require a lot of overhead and development time. If you want to go this route, I recommend you take a look at the Partial Differential Equations VIs. If you do end up creating an application to run FEA, I am sure it would be very coveted.
Best,
10-08-2012 08:02 PM
Hello Jason
it that you told me is really true. Other companies as abaqus, ansys, they have many people developing day a day code for aplications about finite element method, for a one person it is very slow, it take very time.
if you like when i finish my research where i use labview for write code of finite elements, i can send you it.
thank you for the tools for solve differrencial equations
sorry for my english.
10-08-2012 08:36 PM
Sebastian,
I have written a thermal simulation and parts of a fluid dynamics simulation in LV. Neither is general purpose and neither is anything I can publish at this time.
Tips: If the changes are sufficiently small (from time step to time step and cell to cell) you may be able to simplify the differential equations to difference equations or even algbraic equations. If you can do this without loss of accuracy or generality, you may get substantial speed improvements.
Pay attention to memory management so that you do not create unnecessary copies of large datasets. NI has published a white paper on the management of lareg datasets. It is several years old but still very valuable.
Make sure your algorithms test for convergence or timestep errors. If possible try to adapt the calculations to prevent or overcome such errors. There are books which describe methods for doing this.
Please keep us informed of your progress.
Lynn
10-08-2012 09:33 PM
hi how are you , i want give a advance for the structure of code:
When you speak about simplify the differential ecuation is totaly true. my purpose is reduce the computational cost, using the way functions for give numeric solution a diferential ecuation, remove the older, the most amount possible of ceros of the stiffness matriz, further of other artifices.
while less nodes number, less operations, less computational cost, more small the stiffness matriz, it will be more efective and eficient.