An optimized fuzzy-genetic algorithm for metal foam manufacturing process control

The present investigation deals with the proposal of a combined fuzzy-genetic algorithm (F-GA) model able to describe the inherent uncertainties related to the manufacture of open-cell aluminum foams by using the dissolution and sintering process (DSP). The use of the F-GA method allows to take into account, within the same model, both the uncertainty related to the model and the statistical manufacturing process variability. The developed model is aimed at controlling the capability of this material at absorbing energy in compressive deformation, for a different set of process parameters. In particular, the use of genetic algorithms allows the optimization of the support of the fuzzy numbers defined in the model in order to take into account most of the experimental data in combination with the smallest uncertainty. Then, the input uncertainty, related to both the process variability and the chosen model, is propagated to the output variables by the Transformation Method. The fuzzy results are then compared with the measured data and the membership level of the dataset to the uncertain model is evaluated. The process maps generated allow to select the operational parameters in order to obtain a desired process output, in combination with the lowest uncertainty level, providing, as additional information, how much the uncertainty of the model and the process varies by changing those operational parameters.