Vibration Based Condition Monitoring of Spur Gears in Mesh using Stochastic Resonance

In recent years, Stochastic Resonance (SR) has evolved as one of the vibration-based condition monitoring methodologies in mechanical systems fault diagnosis. Of peculiar interest is the distinctive ability of SR to amplify a signal using noise. In reality, signals are never obtained in their raw state and as such, they are always mixed with noise. This seems to suggest that SR has the potential for practical applications. This paper examines the application of SR to impact signals generated by a set of spur gears in mesh as shown in Fig. 1. Since the transmission path alters the signals obtained for SR analysis, it is taken into account in this study. The re-scaling frequency pre-processing technique is used to convert the large parameter system to a small parameter system in order to fulfil the requirements needed for the functionality of SR. Furthermore, the SR results are enhanced by employing a resonance curve, which was recently investigated by the authors. Different number of data points (signal cycles), fault severities and noise levels are tested to establish the suitability of SR for vibration-based condition monitoring in gears