Enhancing Drilling Performance in Self-Healing Composites with Machine Learning Approaches

Abstract

Machining fiber-reinforced plastic (FRP) materials is a big challenge to overcome lot of difficulties in achieving adequate quality for better and acceptable range in assembling process. FRP machining is still a crucial step in achieving quick part assembly while maintaining exact geometric tolerances. However, a variety of machining-induced defects (like matrix -smearing, thermal deterioration and delamination) frequently occur because of the heterogeneous structure. In this paper 3-Gly-cidoxy-propyl-tri-methoxy-silane (KH560) treated carbon fiber-reinforced plastic (CFRP) laminate is investigation for delamination during drilling with L9 orthogonal array and experiments were created by selecting control elements that influence the delamination factor and maximum cutting force. Peel-up and push out delamination was carried out using microscope.1000 rpm cutting speed, 15 feed and 850 point angle showed to be optimum. Machine learning (ML) algorithms exhibiting exceptional performance and concluded Random Forest (RF) algorithms is better among others based on computational complexity.