Automatic Fault Diagnosis and Prediction of Wind Turbines

Automatic Fault Diagnosis and Prediction of Wind Turbines

Xiang, Jianping (Changsha University of Science and Technology)
Jiang, Nannan (Changsha University of Science and Technology)
Watson, S.J. (TU Delft Wind Energy)

2017-10-15

A Morlet wavelet-based compensated algorithm was proposed to calculate the accurate amplitudes of faulty signals. The specific way is to compute the time range and frequency values of the faulty signals at first, and then to compensate the amplitudes calculated for above faulty signals according to the center frequency values of Morlet wavelet coefficients to further obtain the accurate amplitudes. A Simulink model was used to demonstrate the feasibility and generalization of the algorithm. At the same time, the algorithm was used to analyze the electric power signals of a test rig and large turbines. Results show that this algorithm can automatically find the amplitude trend of faulty components in a time sequence, and indicate the residual service life of wind turbines after faults are generated. Based on the information of the residual service life, the maintenance and repairing plan for wind turbines, especially offshore ones, can be developed to lower the cost of wind power in operation and maintenance.

Compensated calculation
Electric power signal
Fault diagnosis
Morlet wavelet transform
Wind turbine

http://resolver.tudelft.nl/uuid:2b313d60-a440-4abf-9501-c6c41f445781

1674-7607

Dongli Gongcheng Xuebao/Journal of Chinese Society of Power Engineering, 37 (10), 821-828

Institutional Repository

journal article

© 2017 Jianping Xiang, Nannan Jiang, S.J. Watson