Instantaneous Phase Estimation Based Single-Signal Diagnosis for Inter-Turn Short Circuit Fault in PMSMs

Abstract

Inter-turn short circuit fault is a commonly occurring electrical fault in permanent magnet synchronous machines (PMSMs) with serious consequences. Diagnostic methods under non-stationary operating conditions lie in removing the impact of speed variations, hence the rotational speed signal is essential. In addition, another signal containing the fault feature information is also needed. However, to reduce the amount of data transmitted from the subordinate device to the principal computer, or deal with signal transmission failure and sensor failure to strengthen the reliability, a single signal based diagnostic method is proposed, which only requires any one of the dq-axis signal, speed, or phase current, even under variable speed conditions. The instantaneous phase is estimated using the single signal based on an intrinsic harmonic feature of the tested machine, and angular resampling is then implemented on the signal based on the fault model. The proposed method is eventually tested on a PMSM using dq-axis signal, speed, and phase current, respectively. The results show that the estimation error of the instantaneous phase using any signal is less than 1.13%, and accurate fault feature extraction is achieved under different speed conditions.