Modeling and optimization of the electrical discharge machining process based on a combined artificial neural network and particle swarm optimization algorithm

In this study, the Electrical Discharge Machining (EDM) process, which is extensively employed in different manufacturing processes such as mold/die making industries, was modeled and optimized using Artificial Neural Network (ANN) and Particle Swarm Optimization (PSO) algorithm. Surface quality, material removed from the workpiece, and tool erosion ratio were considered as the performance characteristics of this process. The objective of this study comprises the optimization of the process in order to find a combination of process input parameters to simultaneously minimize Tool Wear Rate (TWR) and Surface Roughness (SR) and maximize Material Removal Rate (MRR). By establishing a relationship between the process input parameters and the output characteristics, a neural network with back propagation algorithm (BPNN) was used. In the last section of this research, PSO algorithm was used for the optimization of the process with multi-response characteristics. By verifying the accuracy of the proposed optimization procedure, a set of confirmation tests was carried out. Results showed that the proposed modeling method (BPNN) could accurately simulate the authentic EDM process with less than 1% error. Furthermore, the optimization technique (PSO algorithm) is quite efficient in process optimization (with less than 4% error). (C) 2020 Sharif University of Technology. All rights reserved.