Modelling and process optimisation for wire electric discharge machining of metal matrix composites

Abstract

This paper describes the process modelling and optimisation of wire electric discharge machining (WEDM) of SiCp/6061 Al metal matrix composite (MMC) through response surface methodology (RSM) and artificial neural network (ANN) approach. The experiments were planned and carried out based on the design of experiments (DOE). Four WEDM input process parameters namely servo voltage (SV), pulse-on time (TON), pulse-off time (TOFF) and wire feed rate (WF) were chosen as machining process parameters. Two response criteria [i.e., material removal rate (MRR) and cutting width (kerf)] were selected during optimisation. The analysis of variance (ANOVA) was carried out to study the effect of process parameters on response variables and models have also been developed for response parameters. RSM was used to determine the optimal values of input process parameters maximum MRR and minimum kerf. The output of the RSM model was used to develop the ANN predictive model. ANN model was validated through experimentation conducted at the RSM optimal setting of input parameters and results show that ANN predictive model and the actual experimental observations are very close to each other which give a good agreement between the two. Comparisons of ANN models and RSM models show that ANN predictions are more accurate than RSM predictions.