Optimization not working properly

[Franzi]

Hello,
I am trying to fit my reaction force values to an experimental curve, see op.feb. However, the optimization runs the same combination of parameter values a couple of times and stops although not having found a sufficiently good answer, see op.log. I tried multiple values for f_diff_scale (from default value to max. 1) and obj_tol (from default value to max. 0.1), wihtout much influence.
Also, what does "Warning: clamping fem.material[0].solid.solid[0].emu to max (was 0.4)" signify?
As the material I am using is user defined, I cannot share the input.feb here (if needed I could send it directly via mail).
Looking forward to your help.
Best regards
Franzi

[maas]

Hi Franzi,

It sounds like you are running into the limits of the parameter range as defined in your input file. In fact, your initial values for all your parameters are set to either the min or max value in the range, which would explain the warnings. Try to modify the ranges on these parameters to allow the optimization to try a wider range of values and see if that helps.

Best,

Steve

[Franzi]

Hi Steve,
I tried that and adjusted the ranges. Unfortunately, it didn't help and I am still getting these "clamping" notifications. Weird now is however, that sometimes the values in brackets are outside of what I had defined (f.e. "Warning: clamping fem.material[0].solid.solid[0].v to max (was 0.5)" even though v only goes to 0.4). I attached the changed optimization file and the corresponding log file.
Best,
Franzi

[maas]

Hi Franzi,

I think you'll have to expand your ranges even further. The optimization seems to think that the solution lies outside the bounds you specified. If it continues to try to use a Poisson's ratio outside the physical range, perhaps it might be best to fix that parameter to a relatively high value (e.g. 0.499) and remove that from the list of variables to optimize.

Best,

Steve