Fault diagnosis of wind turbine based on alarm signals and D-S evidence theory

Wind turbine fault diagnosis is important to improve wind turbine reliability and reduce the capital cost of wind power system. The Supervisory Control and Data Acquisition (SCADA) systems contain a large number of wind turbine alarm signals indicating certain fault types. In order to diagnose the wind turbine fault quickly and effectively, a new method of fault diagnosis based on SCADA alarm signals and D-S evidence theory is proposed in the paper. Firstly, the identification frame is constructed based on the fault types which are extracted from the maintenance records. Next, all the alarm signals triggered during the occurrence of faults are extracted as the source of the evidence. Finally, the information fusion based on the improved D-S theory is utilized to realize the fault diagnosis. The results show that the method based on D-S theory is feasible and effective in wind turbine fault diagnosis which provides a new idea for wind turbine fault diagnosis. © 2019, Editorial Board of Acta Energiae Solaris Sinica. All right reserved.