SimCenter Forum

Can you please share your model script, in a private message if you would prefer, so that I can reproduce the issue and try a few solutions?

Hello,

Can you please share the "logFileTMCMC.txt" located in "~/Documents/quoFEM/LocalWorkDir/tmp.SimCenter"?

Best regards,
Aakash

Hello Josh,

By observing the results, we think there could be two potential issues:

1) The model might not be numerically stable (going by the large changes in the MaxStrength residuals for small changes in the parameter values, as seen in evaluation 2 vs all the others).

2) A gradient-based (i.e., local) optimization algorithm is used to perform parameter estimation. This algorithm works for continuously differentiable objective functions and will converge to a local minimum. It seems that there are multiple minima even in a very small range of the parameter values (caused by noisy MaxStrength results) - see attachment. That is why the algorithm is taking small steps and converging to a local optimum in the vicinity of the start point of the search.

We recommend that you first perform a parameter study (for e.g. forward propagation using LHS in quoFEM) to identify any such discontinuity/noisiness in the model outputs before performing a parameter estimation study. Sometimes the noisy behavior indicates that the model can be improved (for example by setting stricter convergence tolerance, if feasible). If this is not feasible, you can perform a Global Sensitivity Analysis using quoFEM and decide to not calibrate the values of parameters that are causing this noisy behavior, provided that parameter has a low sensitivity index. If the output is sensitive to the parameter which is causing the noisy behavior, then your option is to use a global parameter estimation method. Currently, you can use the TMCMC algorithm in quoFEM to perform probabilistic parameter estimation, which works for such noisy landscapes too. The trade-off is that it will require a large number of model evaluations, in comparison to a local search method. But you can use DesignSafe's computational resources to overcome this issue. In the future, we might make other global deterministic parameter estimation methods available in quoFEM.

P.S. Please double check if the MaxStrength values are being returned correctly. We ran simulations with the model script that you had previously shared with us over a large range of the parameter values and the MaxStrength residuals were around -4.

Thanks,
Aakash and Sang-ri

Hello Rashad,

1) The parameter estimation methods require data to be provided (either from experiments or otherwise), since the data are necessary to define how 'good' different parameter values are relative to each other.

In the absence of data, if there are no recommendations in literature of good parameter values for your problem, you might want to first explore how the parameter values map to the outputs to get an idea if the parameter values that you have chosen are satisfactory. You can do this by performing either forward propagation or global sensitivity analysis in quoFEM by defining uniform distributions for the parameters in the RV panel and setting wide bounds corresponding to the range of values that you want to explore. Inspecting the scatter plots of the parameter values vs. the outputs in quoFEM's RES panel after the analysis (as in the example shown in the screenshot attached) will show you the range of outputs that are produced corresponding to different parameter values. Sensitivity analysis results will also tell you how much influence each of the parameters in the range that you have defined will have on the outputs from your model. You can check whether the parameter values you have used can produce the outputs that you want. If not, try expanding the range of the parameter values and run the forward propagation/sensitivity analysis again. You can then pick a particular target output value from this plot, provide it as data and perform parameter estimation in quoFEM.

Once you have chosen some parameter values, you can specify a narrow Normal distribution centered at those parameter values and run a forward propagation analysis. You can check the outputs produced and judge if the parameter values are satisfactory.

2) If the finite element analysis is running successfully, then the parameter estimation methods should find parameter values that produce outputs from your model that best match the data. If your analysis is running successfully but you still are having any issues performing parameter estimation in quoFEM, please get back to us with more details (the 'ops.out' and 'dakota.out' files) and we can help resolve the issue.

Thanks,
Aakash

Thank you,
I have download quoFEM 2.3 version, I wanted to check the app before I check my model parameters. then I logged in and hit “run” for a selected example in the category, it showed me this "No dakota. err file - dakota did not run- problem with dakota setup or the applications failed with inputs provided" however, in the old version (2.2) it works well. please what's the problem is?

A solution to this problem is available here - https://simcenter-messageboard.designsafe-ci.org/smf/index.php?topic=191.0

Summary of the issue for users of quoFEM Version 2.3 on Windows:

quoFEM might not find the right path to the Python executable on Windows, resulting in this error message when an example is run - "No dakota.err file - dakota did not run- problem with dakota setup or the applications failed with inputs provided".

The workaround to this bug is to create a registry item for ‘pythonExePath’ under ‘Computer\HKEY_CURRENT_USER\Software\SimCenter\Common’, as described in this chain.

This bug will be fixed in the next release of quoFEM, expected at the end of September/early October 2021. An update will be posted here after the release.

Rashad,

Thank you for sending the screenshot.

Can you please confirm that in the registry editor you changed the python executable path setting under 'Common'?

Because, in the screenshot I see the registry settings for 'Computer\HKEY_CURRENT_USER\Software\SimCenter\quoFEM', not 'Computer\HKEY_CURRENT_USER\Software\SimCenter\Common'.

Aakash