Diagnosis of stator short circuits in brushless DC motors by monitoring phase Voltages

Two adaptive fuzzy systems are developed for identification and location of short-circuit faults on one phase of the phase-modulation brushless dc motors. Motor performance is computed by a discrete-time numerical model. The waveform of line-to-neutral voltages summation is monitored for feature extraction. Diagnostic indices are derived by processing the characteristic signal using discrete Fourier transform (DFT) and short-time Fourier transform (STFT). Testing results show an acceptable diagnosing performance of the proposed scheme.