Fault diagnosis for five‑phase PMSM drive using residual current fusion and probabilistic neural network

Quan Wang; Qiang Tan; Cong Guo; Bing Tian

Fault diagnosis for five‑phase PMSM drive using residual current fusion and probabilistic neural network

Vol. 26, No. 4, pp. 781-793, Apr. 2026

10.1007/s43236-025-01105-2

Electric faults

Five-phase permanent magnet synchronous motor drive

Health indicator

Probabilistic neural network

Abstract

Among the fault types in five-phase permanent magnet synchronous motor (PMSM) drives, open-phase faults (OPFs), openswitch faults (OSFs), and inter-turn faults (ITFs) are the most common. In addition, they all require prompt detection. These faults typically cause a power imbalance across phases. Accurate diagnosis is crucial for applying an effective fault-tolerant control strategy. Thus, this study proposes a method to construct a health indicator by analyzing residual currents using maximum and minimum functions. The fundamental frequency of the health indicator is extracted and compared against a threshold to detect fault occurrence. A probabilistic neural network (PNN) is then used to identify the fault types and the faulty phase. The effectiveness and reliability of the method are validated through both simulation and experimental results.

Statistics

Show / Hide Statistics

Statistics (Past 3 Years)
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.

Cite this article

[IEEE Style]

Q. Wang, Q. Tan, C. Guo, B. Tian, "Fault diagnosis for five‑phase PMSM drive using residual current fusion and probabilistic neural network," Journal of Power Electronics, vol. 26, no. 4, pp. 781-793, 2026. DOI: 10.1007/s43236-025-01105-2.

[ACM Style]

Quan Wang, Qiang Tan, Cong Guo, and Bing Tian. 2026. Fault diagnosis for five‑phase PMSM drive using residual current fusion and probabilistic neural network. Journal of Power Electronics, 26, 4, (2026), 781-793. DOI: 10.1007/s43236-025-01105-2.