Gearbox fault detection by wavelet and spectrum analysis of the induction motor power

This paper presents a new diagnostic method based on the induction motor three-phase power analysis that is able to detect gear broken teeth. A gearbox mathematical model was developed to simulate the healthy and the broken tooth conditions. The simulation and experimental tests proved that the wavelet and spectrum analysis of the induction motor (IM) three-phase power are valuable techniques to diagnose gearbox broken teeth. However, the wavelet analysis method has the advantage of being able to detect the defect of broken teeth even in situations where the mechanical load is not constant, a situation in which the spectrum analysis method is not effective. Another advantage of the wavelet analysis method is the fact that it can diagnose broken teeth without having the knowledge of the healthy condition data. This remote diagnostic method can be useful in systems difficult to access.