Joint temperature prediction method for segment-powered linear motor with working condition based on nonlinear autoregressive with exogenous input neural network

The numerical prediction of the main state of the equipment is the first step in the prognostic and health management (PHM) study. Temperature is one of the most important health indicators of segment — powered linear motor. A nonlinear autoregressive with exogenous input neural network (NARXNN) model is trained to predict the temperature of the segment-powered linear motor. Based on the single — value prediction, the optimal prediction model is introduced. Based on the working condition information, the joint prediction model is designed and trained for the temperature prediction. The model is applied on the multi — stage test data obtained on different dates and the prediction result is compared and analyzed.