Fault diagnosis and condition monitoring of wind turbines. (24th May 2017)

• Record Type: Journal Article
• Authors: Niemann, Henrik; Kjølstad Poulsen, Niels; Mirzaei, Mahmood; Henriksen, Lars Christian
• Other Names: Puig Vicenç guestEditor.; Odgaard Peter Fogh guestEditor.; Schulte Horst guestEditor.
• Is Part Of: International journal of adaptive control and signal processing. Volume 32:Number 4(2018)
• Page Start: 586
• Page End: 613
• Publication Date: 2017-05-24
• Subjects: blade faults -- model‐free fault diagnosis -- multiblade coordinate transformation -- system monitoring -- signal modulation -- wind turbines; Adaptive control systems -- Periodicals; Adaptive signal processing -- Periodicals; 629.836
• DOI: 10.1002/acs.2782
• Languages: English
• ISSNs: 0890-6327

Abstract

Summary: This paper describes a model‐free method for the fault diagnosis and condition monitoring of rotor systems in wind turbines. Both fault diagnosis and monitoring can be achieved without using a model for the wind turbine, applied controller, or wind profiles. The method is based on measurements from standard sensors on modern wind turbines, including moment sensors and rotor angle sensors. This approach will allow the method to be applied to existing wind turbines without any modifications. The method is based on the detection of asymmetries in the rotor system caused by changes or faults in the rotor system. A multiblade coordinate transformation is used directly on the measured flap‐wise and edge‐wise moments followed by signal modulation. Changes or faults in the rotor system will result in unique signatures in the set of modulation signals. These signatures are described through the amplitudes and phase information of the modulation signals. It is possible to detect and isolate which blade is faulty or has been changed based on these signatures. Furthermore, the faulty component can be isolated, ie, the actuator, sensor or blade, and the type of fault can be determined. The method can be used both on‐ and off‐line.