Optimization of WEDM parameters for machining Mg-Li-RE alloy using CRITIC-COCOSO approach

This study focuses on neural network modelling and optimization of the machining parameters of Mg-Li-RE alloy through Wire Electrical Discharge Machining (WEDM). The Taguchi method is employed to plan experiments by varying WEDM parameters such as pulse ON time, servo voltage, wire feed rate, current, and pulse OFF time. An L27 array is designed to examine the effects of considered parameters on surface roughness, material removal rate and kerf width. Multi-objective optimization, specifically the CRITIC-COCOSO method, is applied to determine optimal parameters for improved output responses. Artificial neural network (ANN) model is employed to predict output responses, showing superior prediction compared to conventional linear regression models with an R value of 99.9%. The CRITIC-COCOSO approach suggested optimal output responses yields surface roughness of 3.671 mu m, material removal rate of 0.048 g/min, and kerf width of 0.322 mu m.