Modeling and optimization of process outputs in electrical discharge drilling using genetic algorithm

Electrical Discharge Drilling (EDD) is a non-traditional machining process used for producing small-diameter holes on electrically conductive materials by means of electrical discharges occurring between electrode and workpiece. Drilling operation is performed with a rotating hollow electrode through which dielectric fluid flows to flush the eroded particles away. Material Removal Rate (MRR) and Electrode Wear Rate (EWR) are significant process outputs, defined as the amount of material removed from workpiece and the corresponding loss in electrode per drilling time. It is difficult to estimate MRR and EWR as they refer to erosion phenomenon between electrode and workpiece, which are depending upon several process parameters. In this study, modelling and optimization of EDD process were performed using Gene Expression Programming (GEP) and Genetic Algorithm (GA). The models for reliable and accurate prediction of MRR and EWR were developed using GEP based on selected process parameters (current, pulse-on time, pulse-off time, and capacitance). GA was used for optimizing the outputs to determine the optimal drilling conditions. The most suitable process parameters for obtaining the greatest amount of material removal with reasonable electrode wear were achieved.