Motor Front Axle Temperature Forecasting Based on Phase Space Reconstruction and BP Neural Network

Predicting the temperatures of drive motors of important equipment can help to detect motor failure timely, avoiding the losses caused by the motor faults. Against the nonlinear characteristics of the equipment temperature changes, according to phase space reconstruction principle of chaos theory, the motor front axle temperature series were analyzed and the chaotic nature of the motor front axle temperature series is verified. In order to predict the trend of the motor axle temperature more accurately, the prediction based on BP neural network is conducted, and the embedding dimension of phase space reconstruction is chosen to be the number of input nodes. Simulation shows that this method has higher prediction accuracy and can be used to predict the motor axle temperature.