FEAP User Forum

FEAP => FEAPpv => Topic started by: wjs on January 10, 2016, 06:19:01 AM

Title: Some question about numerical method of J2 model in feappv
Post by: wjs on January 10, 2016, 06:19:01 AM

Dear Prof. Taylor/ FEAP admin/users of feappv:
    Sorry to bother you again.My question is as follows:
    1,does feappv use N-R method in calculation of J2 model?This method is some complecated.....
    2,when judging yield or not, I found that it seems not to think about the peroid of transition which is from elasticity to plasticity.Is there a elastic factor "m" for calculation of stress-increment in feappv?
    3,in your paper CONSISTENT TANGENT OPERATORS FOR RATE-INDEPENDENT ELASTOPLASTICITY, it refers a ‘tangent moduli’ consistent with the radial return algorithm,whether this moduli is the tangent matrix in each N-R iteration?
   

   

Title: Re: Some question about numerical method of J2 model in feappv
Post by: Prof. R.L. Taylor on January 10, 2016, 07:44:29 AM

1. Yes it uses a Newton solution to find the plastic parameter.

2. A good reference on the computation of plasticity problems is
author    = {J.C. Simo and T.J.R. Hughes},
title     = {Computational Inelasticity},
series    = {Interdisciplinary Applied Mathematics},
volume    = {7},
publisher = {Springer-Verlag},
address   = {Berlin},
year      = 1998 }

In this algorithm there is a trial elastic solution followed by the plastic solution if yielding is detected.  One does not split the strain increment into elastic parts and elasto-plastic parts as earlier algorithms proposed.

3. The tangent moduli returned is consistent with the Newton solution process.