Maximizing Machining Efficiency and Quality in AA7075/Gr/ B4C HMMCs through Advanced DS-EDM Parameter Optimization Strategies

Abstract Hybrid Metal Matrix Composites (HMMC) offer improved mechanical properties crucial for various industrial applications. However, machining these composites via traditional methods is arduous due to their inherent hardness and abrasive nature. The optimization of the Die-Sinking Electrical Discharge Machining (DS-EDM) process presents a viable solution to overcome these machining challenges and unlock the full potential of HMMCs. Despite the potential benefits of HMMCs, the lack of optimized machining strategies hampers their widespread adoption in industrial applications. Traditional machining approaches often result in excessive tool wear, poor surface finish, and suboptimal material removal rates (MRR) when applied to HMMCs. Additionally, the complex interplay between DS-EDM process parameters and material properties necessitates a systematic investigation to identify optimal machining conditions. This study proposes a multi-step methodology to systematically optimize the DS-EDM process for machining HMMCs comprising AA7075 alloy, 5% of graphite (Gr), and 5% of boron carbide (B4C) particles, which is termed as (AA7075-GR-B4C). Beginning with the formulation of the composite material through stir casting, the research progresses to an L16 Orthogonal Array (L16-OA) based Design of Experiments (DoE) to investigate the effects of key process parameters on machining performance. Experimental testing is conducted to measure mechanical properties, followed by statistical analysis using Analysis of Variance (ANOVA) with Regression Analysis to identify significant factors influencing machining outcomes. Furthermore, an Artificial Neural Network with Black Widow Optimization (ANN-BWO) approach is employed to optimize machining performance metrics such as MRR, Surface Roughness Rate (SRR), and Tool Wear Rate (TWR). The relative error (RE) was estimated between experimental MRR (E-MRR), experimental TWR (E-TWR), experimental SRR (E-SRR), and predicted MRR (P-MRR), predicted TWR (P-TWR), predicted SRR (P-SRR) generated by ANN-BWO shows effectiveness of AA7075-GR-B4C.