Statistical analysis of electrical discharge machining performance variables to measure form deviation and machinability using non-dominating sorting genetic algorithm-II and fuzzy logic system

Abstract

The main objective of this experimental work is to examine the geometric deviations and machinability of the electrical discharge machining (EDM) process on nickel-based Inconel-718 alloy using square profile copper tool electrodes. The significance of three major EDM process parameters, that is, peak current (IP), pulse on time (Ton) and pulse off time (Toff), have been analyzed to measure the material removal rate (MRR), electrode wear rate (EWR) and squareness deviation, respectively. The dominance of each parameter was measured using design of experiments and the Taguchi method. Non-dominating sorting genetic algorithm (NSGA-II) method was utilized for multiobjective optimization. Empirical modeling has been done using the artificial intelligence-based fuzzy logic system to measure the performance variables. The developed model using fuzzy logic system has an accuracy of 96.45%, that was calculated as the difference between predicted and experimental results; therefore, the developed models can be utilized for more reliable results. The optimized results obtained through this experimental work and empirical modeling will facilitate the aerospace, defense and advanced machining industries to improve the machinability and squareness deviation of nickel-based Inconel-718 alloy with higher MRR and dimensional stability. Finally, the geometric deviations obtained on the square slots of two plates by the previously described EDM process have been transferred to the assembled product in order to verify its assemblability.