A hybrid ANN–GA framework for multi-objective optimization of surface integrity in Nd: YAG laser percussion drilling of 4.5 mm Al/SiC composites

This study presents a hybrid modelling and optimisation framework integrating Artificial Neural Networks (ANNs) and Genetic Algorithms (GAs) to enhance laser beam percussion drilling (LBPD) of aluminium–silicon carbide (Al/SiC) metal matrix composites (MMCs). Experiments were performed on 4.5 mm-thick Al/SiC plates using an Nd: YAG laser to investigate the effects of lamp current, assist gas pressure (AGP), and pulse width on key quality characteristics—surface roughness, microcrack width, and recast layer thickness. A Central Composite Design was employed to plan the experiments systematically. The trained ANN model was integrated with GA for multi-objective optimization. The optimized parameters resulted in a 26.21% improvement in SR, a 29.68% reduction in MCW, and a 35.93% decrease in RLT compared with initial conditions. Experimental validation using SEM and surface profilometry confirmed the predictive capability of the ANN–GA model, establishing its effectiveness for precision drilling of Al/SiC composites.