Dual-scale evaluation of hybrid Al-SiC/graphene composites: Mechanical properties and deep learning-driven machinability insights

• Developed Al 5457 hybrid composites with 6% SiC and 6% Graphene exhibiting 57% increase in hardness and 88% improvement in impact strength. • Achieved a 48% reduction in wear rate and 15% lower coefficient of friction due to synergistic reinforcement effects. • Identified optimal drilling conditions (60 m/min, 0.1 mm/rev, 500 N) using Taguchi DOE for minimal tool wear and surface roughness. • Trained deep learning models (VGG16 and VGG19) on 1,200 augmented tool wear images for high-precision wear classification. • VGG16 outperformed VGG19 with 98.86% accuracy, enabling real-time tool wear monitoring for hybrid MMCs. Growing need for advanced materials in automotive and aerospace applications has led to the development of hybrid aluminium metal matrix composites (AMMCs) reinforced with Silicon Carbide (SiC) and Graphene. While these reinforcements significantly enhance mechanical and tribological properties, their influence on machinability—particularly during drilling—remains insufficiently explored, with limited research addressing the optimization of process parameters alongside predictive modeling of tool wear. The present study uses an integrated experimental and deep learning framework to investigate the drilling performance of Al 5457 alloy reinforced with 6% SiC and 6% Graphene. The research employs a Taguchi-based design of experiments to analyze the effects of cutting speed, feed rate, and thrust force on surface roughness and tool wear identifying optimal conditions for enhanced machinability. In parallel, convolutional neural networks—VGG16 and VGG19—are utilized to classify tool wear types based on augmented image datasets, enabling intelligent condition monitoring. Results revealed that a cutting speed of 60 m/min, a feed rate of 0.1 mm/rev, and a thrust force of 500 N yielded minimum tool wear and surface roughness. VGG16 achieved a superior classification accuracy of 98.86%, outperforming VGG19 at 92.3%, demonstrating its robustness in wear pattern recognition. This dual approach optimizes drilling performance and enables predictive maintenance through image-based wear detection, offering practical applicability in the precision machining of hard-to-machine advanced materials and metal composites (AMMCs). It has the potential for integration into smart manufacturing systems for real-time tool health monitoring and process control.