Research on condition-based maintenance decision-making method for rolling bearings based on the Weibull proportional hazards model

Abstract

To address the issues of resource wastage and excessive maintenance caused by traditional preventive maintenance, which overlooks the actual operating conditions of mechanical equipment, a condition-based maintenance decision-making method for rolling bearings based on the Weibull proportional hazards model is proposed. First, aiming at the problem of complex calculation and poor convergence of traditional Weibull proportional risk model parameter estimation method, a step-by-step estimation method of model parameters combined with an improved genetic algorithm is proposed. Second, by integrating covariates representing the actual operating conditions of the bearing, a Weibull proportional hazards model is developed to reflect the degradation state of the bearing. Finally, considering the issue of delayed maintenance that arises from using a single failure threshold model, the traditional maintenance threshold is optimized. Using bearing operational data as a case study, an optimized condition-based maintenance decision-making strategy, aimed at maximizing availability, is proposed. The results show that the proposed method can comprehensively account for the actual operating conditions, develop appropriate maintenance strategies, and enable condition-based maintenance decisions that balance operational safety and resource efficiency. Moreover, this method can be further extended to related fields, providing valuable reference for research on maintenance decision-making methods for other mechanical equipment.