Optimization of wire electrical discharge machining for pure tungsten using a neural network integrated simulated annealing approach

Abstract

This study analyzed variation of cutting velocity and workpiece surface finish depending on wire electrical discharge machining (WEDM) process parameters during manufacture of pure tungsten profiles. A method integrating back-propagation neural network (BPNN) and simulated annealing algorithm (SAA) is proposed to determine an optimal parameter setting of the WEDM process. The specimens are prepared under different WEDM process conditions based on a Taguchi orthogonal array table. The results of 18 experimental runs were utilized to train the BPNN predicting the cutting velocity, roughness average (Ra), and roughness maximum (Rt) properties at various WEDM process conditions and then the SAA approaches was applied to search for an optimal setting. In addition, the analysis of variance (ANOVA) was implemented to identify significant factors for the WEDM process and the proposed algorithm was also compared with respect to the confirmation experiments. The results of proposed algorithm and confirmation experiments are show that the BPNN/SAA method is effective tool for the optimization of WEDM process parameters.