Fractional Pareto Degradation Model-Based RUL Prediction of Roller Bearings

Estimation of remaining useful life (RUL) of roller bearings is a critical task for predictive maintenance in manufacturing. Bearings are subject to a variety of complex factors during operation, making their degradation process nonlinear with non-Gaussian characteristics. This issue represents a challenge for estimation of the RUL. However, slow degradation process takes on long-range dependence (LRD) and self-similarity. In terms of Pareto’s probability density function (PDF), we proposed the Generalized Pareto Motion (GPm) and its fractional-order extension (fGPm) as degradation predictive models. Estimation of the Hurst exponent shows that this model has long correlation and self-similarity. Through the analysis of the uncertainty of the end point of the bearing’s RUL and the prediction process, not only verified the high adaptability of fGPm in simulating complex degradation process, but also the criteria for judging self-similarity and LRD characteristics were established. One case study demonstrated the advantages of this method, providing an effective analytical tool for a deeper understanding of the degradation mechanism of bearings.