Rotor broken-bar fault detection using a combined method of MUSIC and orthogonal projection

It is difficult to detect the rotor broken-bar fault feature frequency components as they always hide behind the strong supply frequency component and noise in the spectrum of the stator current when the induction motor is in steady running state. A new combined detection method of multiple signal classification (MUSIC) and orthogonal projection is presented in this paper. At first, the strong supply frequency component in the stator current is filtered out but the other harmonics retained by orthogonal projection. Then MUSIC is employed to analyze the filtered signal. Without the influence of the supply frequency component, the fault feature frequency components are easy to be found in the spectrum. The results of simulation and experiment show that the proposed method is feasible and effective.