A Fault Diagnosis Method for Manufacturing System Based on Adaptive BRB Considering Environmental Disturbance

Abstract

Timely fault diagnosis is essential to ensure the reliable performance of manufacturing systems. Aiming at the problems of insufficient prior information and incomplete reliability of monitoring data affected by environmental disturbance during the diagnosis process in manufacturing system, an adaptive belief rule base with index uncertainty (ABRB-u) is proposed. Initially, the adaptive method is used to accurately estimate the initial parameters, facilitating the construction of belief rule base (BRB). Subsequently, considering the limitations of the current model in dealing with uncertain monitoring data, a method for transforming matching degree is introduced, which incorporates the index uncertainty into the model. Finally, the results of the case study demonstrate that this method not only achieves favorable diagnostic outcomes in the absence of prior information but also successfully addresses the challenge of incomplete reliability in monitoring data. This offers a promising solution for fault diagnosis in manufacturing systems.