AI‐Based Predictive Modeling of Delamination Factor for Carbon Fiber–Reinforced Polymer (CFRP) Drilling Process

Carbon fiber reinforced plastic (CFRP) materials have played an important part in the domains of aerospace, sports, etc because of its various characteristics like better modulus, specific, fatigue strength, and also tensile strength, CFRP Drilling is one of the crucial processes in the making of components of CFRP. Delamination can be said to be one of the greatest challenges in the machining process because of its major effect on the structural integrity of CFRP and its application. The delamination factor may decrease the load-carrying ability of the joint. Often, this damage is not detected upon visual inspection because of the nature of the material. Traditional methods of estimation of delamination which is using lab instruments like optical microscopy, digital scanning, ultrasonic C-scan, X-ray have proven to be highly inefficient as these instruments take a long time to measure components and are also subject to human errors. Moreover, these instruments are expensive and very difficult to maintain. Another major disadvantage is that some of the instruments cannot be taken in the field for testing. Machine learning has made a mark in every industry and the machining industry is no different. Machine learning approaches can help optimize the process. This chapter presents several machine learning algorithms like Random forests, linear regression, XGBoost, and Support Vector Machine (SVM), which can help make the process of estimating delamination factor more efficient based on known inputs like Feed Rate, Point angle, and spindle speed,