Process modelling of electric discharge machining by back propagation and radial basis function neural network

Abstract

This work is an attempt to model electric discharge machining (EDM) process using neural networks. In this work, Back propagation neural network (BPNN) and Radial basis function neural networks (RBFNN) have been employed for process modelling of EDM. Training has been done on experimental data generated by conducting experiments on EDM by taking Inconel 718 as work piece. Prior to this, experiments were designed by Taguchi’s orthogonal array. Prediction ability of the trained networks has been verified experimentally. The mean absolute percentage error (MAPE) have been obtained as 2.74% and 11.70% for BPNN and RBFNN respectively. Modelling of EDM by RBFNN and its comparison with BPNN model is the novelty of work.