A GA Approach for Modeling and Optimization of Powder-Mixed Electrical Discharge Machining Process Parameters for Ti-Co Alloy

In recent years, powder-mixed electrical discharge machining (PMEDM) has been successfully employed in manufacturing of different kinds of materials including super alloys. In this paper, mathematical models are proposed, using regression method, to model and analysis the effects of machining parameters on the machining characteristics in the PMEDM process. In this regard, the effects of four machining parameters (grain size of aluminum powder, concentration of the powder, discharge current and pulse on time) on the important process outputs, including metal removal rate (MRR) and electrode wear rate (EWR), have been investigated. To model the machining process, different regression functions have been fitted to the experimental data. Then, using analysis of variance (ANOVA), the best and most fitted set of models are identified. In addition to influence of individual machining parameters, the interactions between these parameters are also investigated. Finally, a genetic algorithm (GA) procedure has been employed to optimize the process parameters for any set of desired outputs. The results show that the proposed solution procedure performs very well in solving such complicated and non-linear optimization problems.