Problems in DerivativeParsedMaterial

[Yingjie Wu]

Hi~

I recently needed to use DerivativeParsedMaterial to add some material information. My question is about this:

I have a material parameter M that is a function of the variable c (eg. M(c) ).  I just need to define 'args = c     f_name = M '  in DerivativeParsedMaterial and give a function expression.  

The other parameter N is a function of the material parameter M and the variable d (eg N( M,d ) ).  I use the code ' material_property_names = 'M' ' to add the material parameter M.

I wonder if this nested material parameter definition is legal in the JFNK framework ? 

Or I need to specify that the material parameter M is a function of c , like ' material_property_names = 'M(c)' '?

I'm not sure how material parameters that vary with variables iterate through the JFNK framework.

I'm looking forward to your reply~

[Yingjie Wu]

Besides,  I have an unconvergent problem. The misinformation for petsc is:

Linear solved did not converge due to DIVERGED_DTOL iterations 30

2 Nonlinear |R| = 8.735066e+29

Nonlinear solve did not converge due to DIVERGED_LINE_SEARCH lterations 2

I don't know the wrong information. Please help me out.

在 2018年3月21日星期三 UTC+8下午9:24:45，Yingjie Wu写道：

[Alexander Lindsay]

On Wed, Mar 21, 2018 at 8:02 AM, Yingjie Wu <yjw...@gmail.com> wrote:

Besides,  I have an unconvergent problem. The misinformation for petsc is:

Linear solved did not converge due to DIVERGED_DTOL iterations 30

http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/KSP/KSP_DIVERGED_DTOL.html#KSP_DIVERGED_DTOL

 

2 Nonlinear |R| = 8.735066e+29

Nonlinear solve did not converge due to DIVERGED_LINE_SEARCH lterations 2

http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/SNES/SNES_DIVERGED_LINE_SEARCH.html#SNES_DIVERGED_LINE_SEARCH 

Because your linear solve did not converge the Newton update was no doubt bad, leading to that huge non-linear residual. You should be able to look at the history of your linear solve to get an idea of what is happening to prevent convergence. Consider running with -ksp_monitor_true_residual

I don't know the wrong information. Please help me out.

在 2018年3月21日星期三 UTC+8下午9:24:45，Yingjie Wu写道：

Hi~

I recently needed to use DerivativeParsedMaterial to add some material information. My question is about this:

I have a material parameter M that is a function of the variable c (eg. M(c) ).  I just need to define 'args = c     f_name = M '  in DerivativeParsedMaterial and give a function expression.  

The other parameter N is a function of the material parameter M and the variable d (eg N( M,d ) ).  I use the code ' material_property_names = 'M' ' to add the material parameter M.

I wonder if this nested material parameter definition is legal in the JFNK framework ? 

Or I need to specify that the material parameter M is a function of c , like ' material_property_names = 'M(c)' '?

I'm not sure how material parameters that vary with variables iterate through the JFNK framework.

I'm looking forward to your reply~

--
You received this message because you are subscribed to the Google Groups "moose-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to moose-users+unsubscribe@googlegroups.com.
Visit this group at https://groups.google.com/group/moose-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/b99b3da4-3555-4d44-b668-f4ba64f0c4da%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.