Fault diagnosis of offshore wind turbine gearboxes using a dynamic Bayesian network

Abstract

ABSTRACT The gearbox system is one of the most critical subassemblies in offshore wind turbine (OWT) drivetrains whose failures could lead to long downtimes and high repair costs. Therefore, it is crucial to accurately diagnose and predict the gearbox faults at an early stage of development. This study develops a new dynamic Bayesian network (DBN) framework for fault diagnosis and reliability analysis of OWT gearbox systems by incorporating components’ degradation information and condition-based maintenance (CBM) strategy. The reliability, availability, and mean-time between failures (MTBF) as well as the failure criticality index (FCI) for each subassembly are estimated. The results identified the loss of function in the bearing subassembly as the most likely underlying cause of a failure in the gearbox system.