Search

Permanent magnet synchronous machines work with controllers and controller parameters affect the machine's voltages and currents and, thus, the diagnosis of stator winding fault. One such parameter is the current controller bandwidth that affects the harmonic content in these signals. In the literature, the second harmonics in both $dq$ -axis currents and voltages have been used for fault detection, and the main advantage is that interturn short-circuit (ITSC) winding fault with a few shorted turns, specially one turn, can be detected. One indicator that combines the fault information in voltages and currents is the instantaneous reactive power (IRP). However, it has been shown that IRP does not perform robustly with respect to current controller bandwidth. The main contribution of this article is that it proposes and validates a new mathematical function for ITSCs detection, Rayleigh quotient, which combines the fault information present in voltages and currents to make ITSCs detection robust to current controller bandwidth. The validations from both cosimulation and experiment show that the proposed method is robust to the bandwidth and can detect ITSC faults faster than IRP. [ABSTRACT FROM AUTHOR]

With the wide application of inverter-fed motors (IFMs), it is necessary to monitor the groundwall (GW) insulation status to guarantee the reliable operation of motor drive system. This article first proposes a mathematical model of leakage current in IFM to analyze its time domain characteristics, and then an online GW insulation monitoring method is proposed based on the transient characteristics of leakage current. By using the initial oscillation amplitude, oscillation period, and attenuation time of leakage current, the phase, position, and severity of GW insulation degradation can be identified. The proposed method is validated using an IFM, and the experimental results agree well with theoretical analysis. [ABSTRACT FROM AUTHOR]