Multi-Condition PMSM Fault Diagnosis Based on Convolutional Neural Network Phase Tracker

Abstract

In many industrial systems, symmetry is the key to ensuring efficiency and reliability. For example, in electric vehicles, the driving system often requires high symmetry. As widely used motors, permanent magnet synchronous motors (PMSMs) are often used in highly symmetrical structures as the driving devices. Consequently, maintaining the symmetry of the system relies on the normal and stable operation of the PMSM, and it is necessary to diagnose faults in the PMSM in a timely manner. In PMSM fault diagnosis methods, frequency domain features of the stator current are extensively used. However, these features change with the switching of motor operating conditions, leading to difficulty of diagnosis in multiple operating conditions. Therefore, a fault diagnosis method based on a convolutional neural network (CNN) phase tracker is proposed in this paper. Through phase tracking and angular domain resampling, the fundamental frequency of stator currents in different operating conditions are aligned, so as to fix the distribution of frequency domain features and solve the problem of features changing with operating conditions. Experimental results show that the proposed method can resample the stator current signals with a small error, detect faults in a relatively short time with high accuracy, and diagnose fault type and severity level under multiple operating conditions.

Fault Diagnosis and Fault-Tolerant Control of PMSM Drives–State of the Art and Future Challenges

“…The main representatives of DNNs used in diagnostic processes are CNN [219], [220], [222], [229], [230], [233], [237] and LSTM structures [223], [228]. The CNN structure allows higher-order features to be extracted from the input information using a mathematical convolution operation.…”

“…The main applications of CNN, on the other hand, result from the high efficiency of feature extraction when the input data has a specific structure (matrices) or repeating sequences. The possibility of direct extraction of the input signal features in the diagnostics of PMSM stator windings faults has been described in [220], [222], [223], [229], [237]. Nevertheless, the analysis of the literature shows that deep neural structures can also be based on pre-processed information [219], [228], [230], [233].…”

“…However, the direct analysis provides a significant reduction in the reaction time of the diagnostic system to the occurring damage, which is its unquestionable advantage. Moreover, it is characterized by the high precision of detection and assessment of the degree of damage, both in the case of single PMSM defects [220], [229], [237] and mixed defects [222].…”

Classification of Optoelectronic Rotary Encoder Faults Based on Deep Learning Methods in Permanent Magnet Synchronous Motor Drive System

“…Currently, AC drives diagnostics [1] are met with great interest in research centers [1,2]. This is due to the use of AC motors [3] in many sensitive industries wherein failure can be a threat and lead to serious consequences [4,5].…”

Identification of Parameters and States in PMSMs

“…In industrial automation and robotics applications, where precision and control are essential, PMSMs are also well suited. PMSMs are perfect for use in applications that call for accurate positioning and motion control because they provide outstanding speed and torque control [6]. Furthermore, PMSMs can operate at high rates, which qualifies them for high-speed machining applications.…”