Blade health monitoring and diagnosis method to enhance operational safety of wind turbine

• Published in: 29th Conference on Precision Electromagnetic Measurements (CPEM 2014) pp. 314 - 315
• Main Authors: Ki-Yong Oh, Jae-Kyung Lee, Joon-Young Park, Jun-Shin Lee, Epureanu, B. I.
• Format: Conference Proceeding Journal Article
• Language: English
• Published: IEEE 01-08-2014
• Subjects: Algorithms; Blade health monitoring; Blades; Condition monitoring; condition monitoring system; Damage; Diagnosis; failure diagnosis; Health; Monitoring; Monitors; Optical sensors; Optical variables measurement; Real-time systems; Rotors; Statistical methods; structural health monitoring; Wind turbines
• ISBN: 1479952052; 9781479952052
• ISSN: 0589-1485; 2160-0171
• DOI: 10.1109/CPEM.2014.6898385

In order to monitor blade health and detect any damage efficiently, a new diagnosis method for wind turbine blades was proposed. In consideration of harsh environments of a wind turbine rotor, high-resolution real-time blade condition monitoring was realized with the use of optic sensors and a wireless network. A hybrid algorithm, which merges a statistical method with model information, was introduced to overcome the weakness of each method. In addition, alarm limits are determined through a machine learning algorithm to enhance its reliability. The proposed algorithm was embedded in the Blade Health Monitoring and Integrity Evaluation System and was verified at a 3MW wind turbine of the Yeongheung wind farm.