Machine Learning of Laser Forming Simulator for Multi-stage Forming

Abstract

Thermo-elasto-plastic deformation models cannot predict the final shape in multi-stage laser forming. This is because that laser irradiation varies material properties, posture, and stiffness of metal plates during the process. Regarding this problem, authors develop a laser forming simulator, which employs the artificial neural network to correlate the process parameters and the metal plate deformation. In this study, prediction of the deformation demonstrates the feasibility of proposed method in multi-stage laser bending experiments with stainless-steel strips. The influences of network structure on machine learning are investigated, and the influences of conditions are discussed in aspect of prediction accuracy.