Research on defect classification method of conveyor belt based on millimeter wave radar

Aiming at the traditional conveyor belt defect classification method with poor light anti-interference ability and insufficient generalization ability, this paper proposes a belt defect classification method based on millimeter-wave radar and wavelet packet decomposition energy entropy features. The belt echo signal is collected by millimeter-wave radar, and eight energy entropies are extracted as multi-scale time-frequency features using wavelet packet decomposition to construct the defect feature set. The LSTM model is used to identify and classify the weak features to improve the defect detection rate in complex environments. The experimental results show that this method can effectively overcome the light interference, significantly improve the accuracy and robustness of conveyor belt defect classification, and provide a new technical means for intelligent conveyor belt monitoring.