Research on wear of Ni-Cr alloy milling based on residual network

With the development of the manufacturing industry and information technology, the quality requirements of products are getting higher and higher. A cutting tool is one of the important factors affecting product quality, so it is of great significance to study cutting tool wear. In this paper, the influence of Ni-Cr alloy on milling cutter wear was studied. Deep learning is widely used in the neighborhood of signal recognition. In this paper, a convolution neural network with residual structure is proposed to classify the wear state of cutting tools. The input of the model is the collected vibration signal, and the output is the classification of tool wear. A convolution neural network can automatically extract the characteristics of signals and identify different types of wear signals. The experimental results show that the convolution neural network with residual structure can converge faster and have higher accuracy than the traditional convolution neural network and the accuracy of tool wear classification is about 98.5%. The loss rate of the model is only about 0.25%.