Shell File Iteration on FEAP

[mohany]

Hi,
I was working a drawing tension problem with Iterative Method (several Loop test) by using shell file. The first inverse analysis works totally fine, but the second inverse analysis gave me cracked model. I assume there are some problems due to nodal displacement overwritten, But I am not sure whether my direction is correct, or any people face same kind of issue before?

Regards

[Prof. R.L. Taylor]

Can you post your file(s)?

[mohany]

In this file, Icir and IcirInv are the first forward and inverse analysis, Icir2 and Icir2Inv are the second forward and inverse analysis. To run the shell file (for example in Xquartz (mac)), just go into the local feap directory and type "sh filename.sh". As I mentioned before, the second inverse analysis make the model pretty wired, I would like to know how I am going to fix it. Thank you!

[Prof. R.L. Taylor]

The sparse solver in feap does not do unsymmetric problems.  You will have to download one of the solvers listed at the web site if you want to do an unsymmetric problem using sparse.  You do not  use OPTI with a sparse solver.  The profiles solver will do your problem fine and can do unsymmetric problems.

You are also using some of your own commands that I do not know about.  What do each of the commands do?

port
import
inverse

are there any others?


It is unclear to me what you are trying to do.  The material is a linear elastic material, thus no need to recompute the tangent each iteration.  If the dt is constant you can do it once and then use

loop,,3
  form
  solve
next

You cannot use the above if you are using anything nonlinear.

 

[mohany]

Sorry for the confusion.

Port is used to collect the history variable, the Import is used for adding the displacement, and the inverse is the macro command used to do the inverse analysis. Since I am doing a displacement-driven problem, I need also to change the sign of the displacement; the inverse is just the macro command for performing the inverse analysis.

Thank you, I will check back my material model.