Overview and discussion of the methods presented

A wind turbine is made up of several sub-systems or components, such as the blade shimming system, the gearbox (multiplier), the generator and the power converter. Faults in these components have a significant impact on the availability and maintenance costs of wind turbines. In addition, faults in these components generate a large number of alarms in the control center. This increases the mental task of human supervisory operators to analyze a large number of alarms. Consequently, early and reliable fault diagnosis of these critical components can significantly improve the availability of wind turbines and reduce their maintenance costs.

In this article, fault diagnosis methods for predictive maintenance support of wind turbines have been divided into two broad categories: model-free methods and model-based methods. Model-based methods are in turn divided into...