A Hybrid Model Flux Estimator for Diagnosis of Inter-Turn Short Circuit Fault in Traction Motors Considering Inverter Nonlinearity

Stator inter-turn short circuit (ITSC) faults are among the most severe malfunctions encountered in traction motors, significantly compromising the power output of high-speed trains. This paper presents an effective fault diagnosis method for detecting ITSC faults in traction motors, utilizing a voltage-current hybrid model flux estimator. The fault characterization signal is obtained using the PI module within the flux estimator, specifically designed to compensate for flux errors associated with pure integration and stator resistance measurements at low speeds. Firstly, a mathematical model of the induction motor under ITSC faults is established, and a high-precision voltage-current hybrid flux estimator based on deadbeat stator current observer is proposed. Secondly, the impact of ITSC faults and the characteristics of the inverter nonlinearity on the fault characterization signal is analyzed. Furthermore, an assessment is conducted on the influence of various parameter mismatches in the motor on the fault characterization signal. Finally, the efficacy of the proposed fault diagnosis method is validated through simulations and experiments conducted under diverse operating conditions.