Condition monitoring for roller bearings of wind turbines based on health evaluation under variable operating states

Abstract

Condition monitoring (CM) is used to assess the health status of wind turbines (WT) by detecting turbine failure and predicting maintenance needs. However, fluctuating operating conditions cause variations in monitored features, therefore increasing the difficulty of CM, for example, the frequency-domain analysis may lead to an inaccurate or even incorrect prediction when evaluating the health of the WT components. In light of this challenge, this paper proposed a method for the health evaluation of WT components based on vibration signals. The proposed approach aimed to reduce the evaluation error caused by the impact of the variable operating condition. First, the vibration signal was decomposed into a set of sub-signals using variational mode decomposition (VMD). Next, the sub-signal energy and the probability distribution were obtained and normalized. Finally, the concept of entropy was introduced to evaluate the health condition of a monitored object to provide an effective guide for maintenance. In particular, the health evaluation for CM was based on a performance review over a range of operating conditions, rather than at a certain single operating condition. Experimental investigations were performed which verified the efficiency of the evaluation method, as well as a comparison with the previous method.