Time-frequency analysis based on BLDC motor fault detection using Hermite S-method

Fault signals of brushless DC (BLDC) motors typically are non-stationary. Conventional Fourier transform method cannot matching the demand of extraction of such fault signals. Time-frequency analysis (TFA) based motor fault diagnostics, which can identify effectively rotor faults by detecting time-variant frequency components of stator current signal, such as the dynamic eccentricity and the unbalanced rotor fault, have been important signal processing methods. This paper proposes a TFA based BLDC motor fault detection approach using Hermite S-method. Compared with commonly used short-time Fourier transform (STFT) and Wigner-Ville distribution (WVD), Hermite S-method owns better time-frequency concentration and better cross-term suppression abilities, thereby improving the accuracy of BLDC motor fault detection. Taking rotor dynamic eccentricity fault as an example, the validity of method is demonstrated.