The model performs sensor selection in composites processing. It particullary provides the best location to put the limited number of sensors and limited available area. We have uncertainties in:
- number of available sensors
- uncertainty in measurements of sensors
- available area to put the sensors
- all material properties (18 material parameters)
- air temperature inside the autoclave
- initial temperature of the composite and tool (uncertainty in the room temperature)
We are prediction the best location(s) to put the sensors.
Predicted best location for the case of 2 available sensors for in processing of a 30 mm composite part on a 15 mm tool material
For stochastic modelling of the processing, the proposed bootstrap filtering model is used:
https://github.com/saniaki/stochastic_modelling
The solver for generating data is a FEM model on thermochemical heat processing:
https://github.com/saniaki/FEM-heat
For optimization, Scikit-Optimize, a sequential model-based optimization library in Python is used:
https://scikit-optimize.github.io/stable/
https://scikit-optimize.github.io/stable/auto_examples/bayesian-optimization.html
https://scikit-optimize.github.io/stable/modules/generated/skopt.gp_minimize.html#skopt-gp-minimize
Acknowledgement
This model is developed with support of
