Modeling localized bearing faults using inverse Gaussian mixtures

Localized bearing faults exhibit specific repetitive vibrational patterns. Due to the constant angular distance between the roller elements, the vibrational patterns occur on regular angular intervals. Under constant operating conditions such patterns become easily detectable as “periodic” events. Slippage or small variation in rotational speed are commonly modeled by introducing normally distributed time variations, which allows for occurrence of “negative” time intervals. In this paper we present an approach which models the occurrences of localized bearing fault patterns as a realization of random point process whose inter-event time intervals are governed by inverse Gaussian mixture. Having support on (0,∞), the random impact times can acquire strictly positive values. The applicability of the model was evaluated on vibrational signals generated by bearing models with localized surface fault.