Investigation on hybrid optimization approach for minimizing surface roughness in the machining of Mg/TiC metal matrix composites using spark EDM

Abstract

Abstract Electrical discharge machining (EDM) is a non-traditional machining technique in which material is extracted as debris from the workpiece due to a spark produced at the interface of the workpiece and electrode (tool). This paper presents the machining of Mg-TiC composites in a servo-controlled spark EDM using a copper tool as electrode material by varying the input process parameters. Surface roughness (SR) was modeled in terms of four input process variables viz. Mg/TiC composites (WP), pulse on time (T on ), pulse off time (T off ) and input current (I) using artificial neural network (ANN). The predictive performance of the ANN model was better with 93.05% model accuracy. A hybrid optimization methodology integrating ANN with the Jaya algorithm is proposed for the optimization of EDM process parameters to obtain minimum SR. The proposed optimization methodology obtained an optimum value of SR 2.89 µm at optimal process parameters $$WP=8.05\text{\%}$$ , $${T}_{on}=21.00\mu s$$ , $${T}_{off}=75.00\mu s$$ , $$I=8.00A$$ . The performance of the ANN-Jaya integrated optimization methodology was found highly accurate and consistent during the consistency test with lower standard deviation. The increase in pulse on time and input current turns in a rougher surface as observed from surface roughness tester. A study on the surface morphology of the machined component is also presented using scanning electron microscope (SEM) images.