Detail setup of ESLM considering non-linear dynamic case

Altair Forum User

Hello,

  I wanna optimize the radius and thickness of cylindrical, the rigidwall is dropped to impact the cylindrical with an initial velocity of 10m/s. I have conducted the model in LS_DYNA and obtained time-displacement curve of a certain node. Now I wanna use ESL method in Optistruct and have set up the loadsteps as Geom non-linear (impl static). So is it right? If so, how to apply the load obtained from LS_DYNA into Optistruct? I have checked a lot of Q&A about ESLM, but still have no idea about the detail setup in Optistruct. 

  Thank you in advance. 

Altair Forum User

Anyone knows the above Question? 

Rahul Rajan_21763

Did you try to check Optistruct examples.See attached screenshot.

<?xml version="1.0" encoding="UTF-8"?>

Altair Forum User

Dear Mr. Rahul R,

  Thank you for your reply.

  Actually I have checked the examples, but all of them are about MBD problems, not non-linear problem. So I have no idea about detail setup.

  I have tried the following two methods:

1.   Geom non-linear(impl dynamic) . The detail steps are as follows:

1) Choosing geom non-linear(impl dynamic) as loadstep, then establish NLOAD1 in which choosing existed SPCD as EXCITEID, DISP as TYPE, TABLED1 as TID. Finally, set up TSTEPNL to control parameters.

Time-displacement which is obtained from LS_DYNA is imposed to the FEM by using SPCD and TABLED1.

2) Set up optimization-opti control, including DESMAX,ESLMAX,ESLOPT, OPTMETH,NESLIMPD, and then create design variables, constraints, and objectives.

  Through the above analysis, the deformation of the cylindrical is totally different with the result from LS_DYNA. 

2.  Geom non-linear(expl dynamic).

  Create XSTEP, NLOAD1(setup is same with the first method),etc. 

  The deformation of the cylindrical is totally different with the result from LS_DYNA.  

  So could u tell me what is the problem? And what is the setup of ESLM for non-linear problem?    

  Looking forward to your help. Thank you!

  

 

Rahul Rajan_21763

It would be nice if you can share your model using my dropbox available in my signature.

Altair Forum User

Dear Mr. Rahul R,

  I have sent my models using dropbox, please check it. 

  Looking forward to your feedback.

  Thank you in advance. 

Rahul Rajan_21763

Thanks I got your files.I feel SPCD is not correctly defined in your model.

I recommend you to take a look at below existing example file.

Example 51 - Optimization in RADIOSS for B-Pillar (Thickness optimization)

Altair Forum User

Dear Mr. Rahul R,

  Thank you for your fast response. 

  Actually when I checked Q&A in Optistruct part of this Forum, I knew you have suggested others to see the same example, but the problem is that the Hyperworks version I use is 13.0, not 14.0, which means I cannot see the example. So could you please send the PDF and .fem file of this example to me? If so, I will be so appreciate for your help.   

Rahul Rajan_21763

PFA. This was done using Radioss.

Unable to find an attachment - read this blog

Altair Forum User

Altair Forum User said:

Thanks I got your files.I feel SPCD is not correctly defined in your model.

I recommend you to take a look at below existing example file.

Example 51 - Optimization in RADIOSS for B-Pillar (Thickness optimization)

Can you share some example file on optimization in RADIOSS (topology optimization)

Rahul Rajan_21763

I too dont have any public model.will share once it is available.As of now refer Radioss userguide.

Altair Forum User

Looking forward to your news & updates!

Thank you.