Abstract:The signal of incipient fault in the drilling permanent magnet synchronous motor (DPMSM) is not obvious, and it is easy to be submerged by noise. Moreover, the sensitivity of signals to different faults varies, and the fault features of different fault severity are different. Therefore, the diagnostic accuracy is generally low with a single signal for the incipient fault of the DPMSM. A data-driven information fusion method for the incipient fault diagnosis of the DPMSM is presented. Firstly, the empirical wavelet transform (EWT) method is used for signal analysis. Then, the singular value decomposition (SVD) method and the principal component analysis (PCA) method are used to extract the fault feature of the vibration signal and the torque signal, respectively. Finally, the Bayesian network (BN) method is applied to the fault diagnosis with the vibration signal method and torque signal method, and the improved evidence theory method based on the Dempster-Murphy rule is used as an information fusion method to improve diagnostic accuracy. The diagnostic accuracy of the three methods is compared and discussed, the information fusion method has stable and highest diagnostic accuracy for faults with different severity under different load conditions. Note to Practitioners —This article was inspired by the field experience of the maintenance engineers. Our maintenance engineers have found that the diagnosis method based on a single signal is not stable for the DPMSM faults. For different types of faults, the diagnostic accuracy of the vibration signal method is better for mechanical faults, while the diagnosis effect of the torque signal method is better for excitation faults. For different fault severity, the vibration signal method can hardly correctly diagnose the incipient faint faults, while the diagnosis effect of the torque signal method is better for the incipient faults. For faults under different load conditions, the diagnostic accuracy of the two methods decreases with the up in load condition. Based on the above findings, we want to explore the relationship between the diagnostic accuracy of the two methods and the fault type, fault severity, and load condition, and hope to find a better and more stable diagnostic method for engineering applications. Based on this study, the practitioners can select appropriate fault diagnosis methods according to fault type, fault severity, and load condition.

Vibration-current data fusion and gradient boosting classifier for enhanced stator fault diagnosis in three-phase permanent magnet synchronous motors

Application of AdaBoost for stator fault diagnosis in three-phase permanent magnet synchronous motors based on vibration–current data fusion analysis

Fault Diagnosis of PMSMs Based on Image Features of Multi-Sensor Fusion

Machine Learning-Based Stator Current Data-Driven PMSM Stator Winding Fault Diagnosis

Research and Application of Intelligent Diagnosis Technology in Permanent Magnet Generator for Stress Demagnetization Fault

Vibration and current dataset of three-phase permanent magnet synchronous motors with stator faults

Diagnosis of High-Resistance Connection Faults in PMSMs Based on GMR and Deep Learning

Increasing Interpretability and Feasibility of Data Driven-Based Unknown Fault Diagnosis in Permanent Magnet Synchronous Motors

Data-driven early fault diagnostic methodology of permanent magnet synchronous motor

Demagnetization Fault Diagnosis of Permanent Magnet Synchronous Motors Based on Stator Current Signal Processing and Machine Learning Algorithms

A Wavelet-Based Fault Diagnosis Approach for Permanent Magnet Synchronous Motors

Detection of permanent magnet damage of PMSM drive based on direct analysis of the stator phase currents using convolutional neural network

Application of an information fusion method to the incipient fault diagnosis of the drilling permanent magnet synchronous motor

Feature Extraction for Short-Circuit Fault Detection in Permanent-Magnet Synchronous Motors Using Stator-Current Monitoring

Performance of vibration and current signals in the fault diagnosis of induction motors using deep learning and machine learning techniques

Mechanism-Based Fault Diagnosis Deep Learning Method for Permanent Magnet Synchronous Motor

Fault Diagnosis of Inter-turn Short Circuit in Permanent Magnet Synchronous Motors with Current Signal Imaging and Unsupervised Learning

Optimizing and Analyzing Performance of Motor Fault Diagnosis Algorithms for Autonomous Vehicles via Cross-Domain Data Fusion

Modeling and Fault Detection of Brushless Direct Current Motor by Deep Learning Sensor Data Fusion

Diagnosis of Stator Winding and Permanent Magnet Faults of PMSM Drive Using Shallow Neural Networks

Research on Three-Phase Asynchronous Motor Fault Diagnosis Based on Multiscale Weibull Dispersion Entropy