Fuzzy logic model for bending angle in laser forming

Abstract

Laser forming is a thermoelastoplastic and complicated process. Finite element model simulation is time consuming and the analytical model is cumbersome for computing the bending angle. A model for bending angle in laser forming is constructed using adaptive fuzzy logic called adaptive network fuzzy inference system (ANFIS). The ANFIS model is trained with the published experiment data, in which the laser forming process parameters include thickness of the plate, laser power, laser beam diameter and scanning velocity. Trained ANFIS model is also tested on experiments not pertaining to previous training data. The performance of ANSIF model is optimised as a function of a type of membership function and number of membership functions. Optimised ANFIS model well predicts the results compared with the experiment data. Based on the established model, the analysis results of the process parameters show that a maximum bending angle can be achieved by choosing a reasonable laser beam diameter given other laser forming parameters.