Optimization of Al₂O₃/SS316L composites fabricated via laser powder bed fusion using machine learning and multi-objective optimization

This study investigates the mechanical and tribological performance of Al₂O₃/SS316L composites fabricated via Laser Powder Bed Fusion (LPBF) and optimizes process parameters for enhanced material properties. The effects of layer height, laser power, scanning speed, and Al₂O₃ reinforcement content on composite performance were analyzed using a combination of experimental techniques and computational models. Mechanical and tribological properties, including compressive strength, wear rate, and coefficient of friction, were evaluated. A Gradient Boosting Decision Tree (GBDT) model was developed to predict material behavior, achieving high accuracy (R² = 0.98 for training and 0.93 for testing). Multi-objective optimization using the Strength Pareto Evolutionary Algorithm 2 (SPEA2) identified optimal process parameters, balancing mechanical strength and wear resistance. Microstructural and wear mechanism analyses via SEM and EDS confirmed uniform Al₂O₃ dispersion at 10 wt%, enhancing strength and wear resistance, while excessive reinforcement (15 wt%) led to clustering and performance degradation. Optimized composites exhibited compressive strength up to 762 MPa, wear rates as low as 0.012 mg/km, and reduced coefficients of friction (0.231). This study provides a structured approach to optimizing LPBF-fabricated composites, supporting their application in aerospace, biomedical, and automotive industries.