Demagnetization Prediction Method of Permanent Magnet Synchronous Motor Based on Digital Twin Model

For the issue of local irreversible demagnetization faults in the permanent magnet of the PMSM (permanent magnet synchronous motor) affected by factors such as high temperature and high current. A method for predicting the demagnetization trend of permanent magnets based on digital twins has been proposed. Firstly, establish a digital twin geometric model based on motor parameters; Secondly, based on the finite element analysis method, a digital twin behavior and rule model is constructed, and a composite model of CNN (convolutional neural network),LSTM( long short-term memory neural network), and self attention mechanism is introduced into the digital twin model to achieve the prediction of the demagnetization trend of permanent magnets; Finally, the effectiveness of the method is validated using the experimental dataset. The experimental results show that compared with traditional digital twin technologies such as CNN,LSTM and so on, the prediction method proposed in this study significantly improves prediction accuracy and speed.