Real deformation in Finite element analysis

[Mm.Kaiser]

Hi,
Consider two analysis of Finite element model:
1-Linear Elastic (Large deformation is off)
2-Elastic-plastic analysis of a same model (large deformation is on)

Which one shows us the real deformation of a component? it is so important for me according to design the sealing zone of component.

Regards

 

[TGS4]

Are the stresses less than our greater than yield?. If so, then the E-P analysis will be more accurate. Otherwise it won't make a difference.

 

[Mm.Kaiser]

Finally you mean Greater or Less than yield?
Absolutely in Elastic analysis the max stress amplitude is more than E-P analysis although both of them in Concentrations are greater than yield.

 

[TGS4]

In an elastic-plastic analysis, once the stresses exceed yield, it really is inappropriate to discuss stress as the relevant quantity - strain is the appropriate value to discuss.

Since you elastic stresses exceed yield, then your real deformation is more likely to resemble the E-P analysis. This is because beyond the yield stress, the behaviour is no longer linear-elastic, but has a function involving the tangent modulus of the non-linear components of the stress-strain cure. Does that make sense?

 

[Mm.Kaiser]

That's right.
So i understood, when we assign a tangent modulus to our model with large deformation(E-p Analysis), we will have a deformation which in reality will happen and this is the reliable value.
And as you said, stresses in E-p analysis are not reliable but i thought that elastic analysis give us a theoretical value and in reality, stresses in concentrations are self-equilibrium and because of strain-hardening characters, they can't grow up such in elastic analysis happen!