Response Surface Modeling and Optimization of Electro-Discharge Machining of Al/Al2O3p

Abstract

In this study, wire electrical discharge machining of aluminium matrix composites has been experimentally investigated. A metal matrix composite Al/Al2O3p/10% has been machined at various combinations of machining parameters such as pulse on-time, off-time, wire speed and wire tension. Empirical models have been developed to predict the cutting rate, surface roughness and kerf width of the machined composite material. The models were developed by using the technique of response surface methodology. The experiments were conducted by using carefully planned response surface design matrix (central composite design). The model predictions are in good agreement with the experimental results. The R2 values of the proposed cutting rate, and kerf width models are above 0.99 and 0.98 respectively, where as it is 0.95 for the surface roughness model. The pulse on-time was found to influence cutting rate and surface finish significantly. Kerf width was significantly affected by pulse on-time, off-time, wire speed, and wire tension. In addition, a model was developed to correlate the multiple performance characteristic called as grey relational grade and the process parameters. Thus, a new combination of response surface method and grey relational analysis is proposed. The grey relational grade was significantly affected by pulse on-time and off-time time. The R2 value for the grey relational grade model was 0.94. Optimal parameter setting was determined for the multiple performance characteristic. The improvement in cutting rate was more than 100% compared to the ‘0’ level experiments, with reasonably smooth surfaces and narrow kerf width.