Wind turbine drivetrain health assessment using discrete wavelet transforms and an artificial neural network

Cost-optimised maintenance of wind turbines is becoming increasingly important due to the remote location and access limitations of offshore wind farms. Maintenance and repair costs increase significantly due to the specialist vessels required and delays due to unfavourable weather conditions. Significant cost reductions can be achieved by making better use of condition monitoring (CM) information to make more efficient and effective maintenance decisions. The key innovation described in this paper is the transfer of CM technologies from the aerospace industry to the wind turbine industry. Algorithms have been developed to enable the detection of faults and transient events in vibration signals. The ability of the new algorithms to identify real-time features has been demonstrated using bearing seeded fault testing.