Investigations on LM25–SiC–Al₂O₃–graphite composites for high-performance marine engineering using machine learning algorithms

This study focuses on the fabrication and performance evaluation of hybrid LM25 aluminum metal matrix composites (HAMMCs) reinforced with silicon carbide (SiC), alumina (Al₂O₃), and graphite for advanced marine engineering applications. Three composites Specimens A, B, and C were developed using the stir casting process with reinforcement combinations of 10%, 15%, and 20%, respectively. After T6 heat treatment, the composites were subjected to mechanical testing and microstructural analysis. Among the three, Specimen C (10% SiC, 5% Al₂O₃, 5% graphite) exhibited the highest performance, showing a 40% increase in hardness and a 35% reduction in wear rate. SEM and EDS analyses confirmed uniform reinforcement dispersion, minimal porosity, and the presence of essential elements such as Si, Al, O, and C. EBSD and XRD analysis revealed refined grain structures and good phase stability in Specimen C. To predict key machining parameters are Material Removal Rate (MRR), Tool Wear Rate (TWR), and Surface Roughness (Ra) using machine learning algorithms including Decision Tree, Support Vector Machine (SVM), and Random Forest were employed. The models were trained on experimental datasets to reduce manual effort and improve predictive accuracy. Random Forest outperformed other models, achieving up to 93% prediction accuracy, with the lowest RMSE (0.09) and MAE (0.06), particularly for Specimen C. The integration of experimental results with intelligent machine learning models proved effective in optimizing composite design and performance assessment. The study concludes that LM25–SiC–Al₂O₃–Graphite composites, especially Specimen C, are highly suitable for marine applications due to their enhanced mechanical strength, wear resistance, corrosion resistance, and predictive machining accuracy.