1DCNN-Residual Bidirectional LSTM for Permanent Magnet Synchronous Motor Temperature Prediction Based on Operating Condition Clustering

With the rapid development of electric vehicle (EV) technology, accurate prediction of motor stator temperatures is essential to ensure safe operation and extend the service life of motors. However, traditional prediction methods have limitations in dealing with nonlinear and complex temperature data, making it difficult to provide highly accurate prediction results. To address this issue, this paper proposes a deep learning model based on one-dimensional convolutional neural network (1DCNN) and bidirectional long and short-term memory network (BiLSTM) with operating condition clustering for learning and accurately predicting the temperature of the motor stator directly from the raw data. Through training and validation with the existing dataset of the motor, the experimental results show that the proposed model has a significant improvement in prediction accuracy compared to the traditional methods, which provides a strong support for the health monitoring and maintenance of the motor.