Nonlinear material-plasticity

[human123]

I would like to know if defining more plasticity "points" (fy-ep) makes analysis more nonlinear and consequently more expensive.
For example:
If I put 4 rows instead of 20 in plasticity model, will my calculations be faster?
Thank you.

 

[Drej]

"I would like to know if defining more plasticity "points" (fy-ep) makes analysis more nonlinear..."

"More" nonlinear? For a standard hardening model (isotropic?), no. Plasticity is in itself a "nonlinearity" and once defined forms part of a non-linear analysis (assuming the material goes beyond yield). Introducing "more" nonlinearity would involve defining contact or enabling non-linear geometry (NLGEOM), for example.

"...and consequently more expensive."

No. Why would it? By defining more points you are simply defining a better resolution of the plasticity behaviour of the material. Again, for a standard hardening model the analysis becomes more expensive if you choose a different constitutive model or you refine the time step control, or change the loading (amongst many, many other things depending on what it is you're doing).






------------
See faq569-1083 for details on how to make best use of Eng-Tips.com

 

[Drej]

I should also mention, however, that in the case of a piecewise continuous material definition you need to be careful with the number of points you choose, since this could potentially have an impact on the accuracy of your analysis. Of course, this depends on the hardening behaviour of the material itself and the extent of the nonlinearity of the material. If the material shows little hardening behaviour and can be easily defined as a series of small straight lines, then you can likely get away with defining a few points (or by just defining a tangent modulus if applicable).


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com

 

[human123]

Thank you for all your answers.
The reason I tought that more points would mean more expensive analysis is that:
if I use only a few points I tought abaqus just makes the response linear between this two points (linear relationship stress-strain). Therefore, when applying next increment in nonlinear analysis, it is more likely that it will stay on the stress-strain curve and won't have to make iterations.
Sorry, if I am totaly wrong
Thank you again