LAUSR

Abstract A full factorial experiment is performed for the conventional dry drilling of CFRP with spindle speed, feed rate and point angle as drilling parameters, response variables are thrust force and exit-delamination. Artificial neural network (ANN) is developed to express thrust force and delamination factor as a function of drilling parameters. Multi-objective optimization of drilling parameters is accomplished based on Non-dominated Sorting Genetic Algorithm (NSGA-II) with thrust force, delamination factor and material removal rate as optimization objectives, delamination factor also serves as a constraint. The Pareto front of drilling response variables determined by NSGA-II consists of a large number of non-dominated solutions. In order to facilitate the experimental verification of optimization results, fuzzy C-means clustering algorithm is used to narrow down the solutions on the front to several representative ones. Conformation tests are conducted and results show that the representative solutions can give satisfactory performance with achieving a trade-off among thrust force, exit-delamination and material removal rate.