Artificial intelligence-based condition monitoring and predictive maintenance framework for wind turbines

The global wind power capacity continues to grow at a fast pace. However, the profit margins from wind power are being compressed in many countries. Thus, many wind farm owners seek to reduce their operational expenses, including those for maintenance work. In this study, an artificial intelligence-based condition monitoring and predictive maintenance framework for wind turbines is presented. The purpose of this framework is the automated early detection of operational faults in wind turbine systems and subsystems. The early detection of anomalies enables further diagnosis, condition-based maintenance and better planning of repairs. It can prevent consequential damage, lead to fewer turbine downtimes and extend the service lives of the monitored turbines. We present validation results from two onshore wind farms and demonstrate 97% accuracy for a 2-month detection horizon of developing fault events that require attention from maintenance staff.