A Hybrid Algorithm for Modeling and Optimizing the AISI 304L Stainless Steel Electrical Discharge Machining Parameters

Abstract Electric Discharge Machining (EDM) is known for its ability to machine hard and brittle conductive materials, as it can melt any electrically conductive substance irrespective of its hardness and geometric intricacy. Machining technologies have progressively advanced from a basic tool and die manufacturing process. The purpose of this paper is to investigate the performance in achieving the maximum material removal rate (MRR) while minimizing surface roughness (Ra) and tool wear rate (TWR) together and to optimize the parameters’ operation utilizing ANOVA, RSM, and the hybrid BPNN-PSO. The AISI 304 L stainless steel is machined under different parameters, such as peak current (IP), pulse-on time (Ton), and pulse-off time (Toff). The ANOVA was used to identify significant components in single-objective optimization, whereas the Response Surface Methodology (RSM) was used for multiple-objective optimization of the parameters. The hybrid BPNN-PSO algorithm was used in single and multiple-objective modes to improve the parameters that were worked on in two scenarios. The results showed that the electrical current had a 75% impact on the output. From the analysis, the hybrid BPNN-PSO result achieved significant improvement over both the ANOVA and the RSM algorithms. More specifically, IP was optimized by 1.25 times on the quantity of MRS technique, whereas Ton was 58.18 μs in the second scenario. Based on these data, the second scenario is capable of achieving even better results in terms of optimization.