Date of Graduation
1-2016
Document Type
Thesis
Degree Name
Master of Science in Electrical Engineering (MSEE)
Degree Level
Graduate
Department
Electrical Engineering
Advisor/Mentor
Jingxian Wu
Committee Member
Jing Yang
Second Committee Member
Roy Mccann
Keywords
Bearing Faults, Quickest Change Detection, Wind Turbines
Abstract
A new online method for detecting bearing faults in direct-drive wind turbines (WT), based on stator current analysis and quick change point detection techniques, is proposed. Bearing faults are common problems for machinery, such as wind turbines. Online methods of fault detection give clear advantages by allowing the diagnosis of wind turbines without physical access. Stator current signal is always available, it is perfect for online analysis, and no additional sensors are required. Faults of bearings supporting the main shaft of direct-drive WT introduce excitations into the stator current spectrum. Change-detection procedures provide a simple and yet efficient way to find that change. In the proposed method, the stator current signal is synchronously resampled, and fault signatures are extracted using FFT. Then, distribution of a fault signal is estimated. After that, the condition of bearings is tested using a quickest change detection method such as CUSUM or Shiryaev-Roberts procedures.
Citation
Romanov, V. (2016). Quickest Fault Detection in Direct Drive Wind Turbine Using Stator Current. Graduate Theses and Dissertations Retrieved from https://scholarworks.uark.edu/etd/4819