Delamination factor and cutting force optimizations in end-milling of carbon fiber reinforced polymer composites using backpropagation neural network-ant colony optimization

Abstract

Delamination factor (FD) and cutting force (FC) are the critical-to-quality characteristics (CTQs) in the end milling process of carbon fiber reinforced polymer (CFRP) composites. The target of these two CTQs is smaller the better. The best combination of end milling parameters, i.e., depth of cut (Aa), feeding speed (Vf), and spindle speed (n), were used to attain the smallest FD and FC by integrating backpropagation neural network and ant colony optimization methods (BPNN-ACO). BPNN is utilized to model the FD and FC, while ACO has two purposes. The first purpose of ACO is to determine the BPNN network architecture that can produce a minimal mean square error (MSE) to precisely predict FD and SR. The second purpose is to find the best setting of end milling parameters (spindle speed, feeding speed, and depth of cut) that can minimize FD and FC simultaneously during the end milling process of CFRP composite. This study uses a randomized full-factorial 2 × 3 × 3 experimental design. The depth of cut of end milling parameters have two levels, while spindle speed and feeding speed each has three levels. The best BPNN network structure consists of three neurons in input layers, five neurons in the first hidden layer, seven neurons in the second hidden layer, and two neurons in the output layer; with tangent sigmoid as the activation function. The setting of the end milling parameters that can minimize FD and FC simultaneously on end milling process of CFRP composites by using BPNN-ACO is spindle speed of 3712 rpm, feeding speed of 29 mm/min, and depth of cut of 1.2 mm.