Impact Performance Enhancement of Nano‐Clay‐Reinforced Sandwich Panels: A Machine Learning Approach

ABSTRACT This study explores the fabrication and impact behavior of Nano‐clay reinforced aluminum honeycomb Sandwich panels (NSPs), focusing on their mechanical performance under low‐velocity impact conditions. Sandwich panels were fabricated using glass fiber face sheets, polyester resin infused with varying nano‐clay content (0%, 2%, and 4%), and an aluminum honeycomb core. Experimental investigations were conducted at different impact energies (10, 20, 30, and 40 J) to evaluate peak force and absorbed energy. The results indicated that nano‐clay reinforcement significantly improves impact resistance, with the 4% nano‐clay panel showing a 43% reduction in damage area compared to the unreinforced panel at 10 J. Machine learning (ML) models including Polynomial Regression, Decision Trees, and others were used to predict force and energy absorption. The impact of nano‐clay on force and energy absorption was confirmed using SHAP analysis and validated with heat maps. In this case, Decision Trees were the best model with respect to R ² and RMSE. These findings demonstrate that NSPs are viable for advanced applications in the automotive and construction industries.