Optimization of Multiple Characteristics of EDM Parameters Based on Desirability Approach and Fuzzy Modeling.

Abstract: Electrical Discharge Machining (EDM) has emerged as an outstanding approach for cutting conductive metals that are otherwise difficult or impossible to be cut with traditional machining [1]. Moreover, a survey of literature has found that a little of work was carried out in the direction of achieving optimal levels of machining parameters for the super alloy Inconel 718, among other super alloys, which are otherwise difficult to machine. The objective of the present work was to investigate the effects of various EDM input parameters as well as the influence of different tool geometry on Material Removal Rate(MRR), Tool Wear Rate(TWR) and Surface Roughness(SR) on machining of Inconel 718 material by using copper electrode. Five EDM parameters, namely pulse on time (TON), pulse off time (TOFF), peak current (A), flushing pressure (P) and electrode tool geometry (Geo), were considered here. Tool geometry of the electrodes was circle (C), square (S), rectangle (R) and triangle (T). Four different levels for the five input parameters were planned as per the L16 orthogonal array. The parameters were optimized using multi-objective optimization technique of desirability approach and the significance of each parameter was analyzed by Analysis Of Variance (ANOVA). In addition, Fuzzy Logic Model (FLM) was used to better understand the input and output responses. With the desirability approach, it was sought to optimize the values for copper electrode for maximum MRR and minimum TWR and SR. Overall, the rectangular tool geometry emerged successful. A comparison of the performances of the electrode by desirability approach and ANOVA showed that the current was the most influencing factor, followed by pulse on time and pulse off time. It was also observed that the rectangular tool geometry provided better results as compared to other tool geometries. Validation tests for FLM were carried out and show closer relationship with the experimental results. [Copyright &y& Elsevier]