As the core component of industrial equipment, permanent magnet synchronous motors are essential for accurate fault diagnosis, and intelligent data-driven methods and real-time supervision technology can help formulate accurate maintenance plans to achieve low-carbon and energy-saving operation. Based on the digital twin technology, the efficient operation and maintenance of permanent magnet synchronous motor can be carried out, which can realize virtual and real synchronization and monitor the operation status of the motor, and further combines with signal processing and machine learning technology, a fault diagnosis method based on variational mode decomposition, maximum correlation kurtosis deconvolution and BP-Adaboost is proposed, which can significantly improve the accuracy and efficiency of fault diagnosis of permanent magnet synchronous motor. For the inter-turn short circuit of the permanent magnet synchronous motor, the optimal diagnostic error rate is as low as 1.75%, and the average diagnostic error is controlled to about 3.6%, with high diagnostic accuracy.
