A convolutional neural network-based method for identifying delamination damage in carbon fiber reinforced resin matrix composites (CFRP) and titanium alloy laminates is proposed to address the issue of delamination defects in CFRP during assembly of holes in CFRP and titanium alloy laminates. Conduct drilling experiments, collect force signals during the drilling process, use continuous wavelet transform to obtain wavelet scale maps of the force signals as input sets, and use layered factors as label sets to construct a convolutional neural network model. The results indicate that there is a positive correlation between the axial force signal and the degree of delamination damage in CFRP. It is feasible to identify the delamination damage state of CFRP by monitoring the drilling force signal through a convolutional neural network model. Build three network structure models: VGGNet-16, LeNet, and AlexNet. Through comparison, it is found that the VGGNet-16 model has better recognition performance, with a recognition accuracy of 91.84%.