Symbolic regression analysis of cutting force and delamination in drilling and punching of GFRP composites

This study investigates the effects of drilling and punching parameters on cutting forces and delamination factors in ±45° oriented glass fiber-reinforced polymer (GFRP) composites. The GFRP laminates were produced using a hand lay-up method with a 600 g/m ² woven biaxial fabric and Polives 701 vinyl ester resin, resulting in 2.3 mm thick laminates. The study aims to optimize machining processes by examining the influence of feed rates, cutting speeds, and punch types on the cutting forces and delamination. Symbolic regression was used to analyze the experimental data and identify key parameters. Results show that in drilling, the delamination factor increases with feed rate and exhibits a non-linear behavior with cutting speed. Feed rate was identified as the primary factor influencing cutting forces, while cutting speed significantly affected delamination at the hole entrance. In punching, higher cutting forces were observed with flat punches, and punch type was the main factor affecting both cutting forces and delamination. Delamination factors at the hole exit were higher than at the entrance, increasing with both feed rate and cutting speed in drilling, and with feed rate in punching. Symbolic regression models accurately predicted the experimental results.