MSCNN-LSTM tool wear detection model based on attention mechanism

In order to improve the prediction accuracy of tool wear and realize online monitoring of the machining process, this paper proposes a tool model detection model based on attention mechanism combining multi-scale convolutional neural network and long short-term memory neural network. The model extracts the local and global wear characteristics in the image through multi-scale convolutional neural network, combines LSTM to model the time series to analyze the wear evolution trend, and introduces the attention mechanism to strengthen the key information. Experiments show that the proposed model is better than the traditional method in terms of prediction accuracy, detection efficiency and false alarm rate, which provides a new idea for tool wear detection, and is of great significance for improving machining efficiency and ensuring production safety.