Please use this identifier to cite or link to this item:
http://hdl.handle.net/20.500.11889/6319
Title: | Model-Based Fault Detection in Wind Turbines | Authors: | Basheer Shaheen Ahmed Abu Hanieh - Supervisor Michel Kinnaert - Co-supervisor |
Keywords: | Wind turbines - Monitoring | Issue Date: | May-2017 | Abstract: | An early Model-based fault detection was developed and presented in the basis of WT’s power curve to detect the degradation (faults) in gear box efficiency, resulted from the existing mechanical losses (torque losses) through the low-speed shaft(LSS) and the high-speed shaft (HSS), then to assist in implementing predictive maintenance. The detection was performed on two levels; the first level represents a slight and progressive degradation in the gear box efficiency, and the other one represents a radical (abrupt) degradation in the efficiency. Artificial SCADA data for different measurements (wind speed and active power) in both, fault free and faulty operating modes were generated using FAST-NREL simulator. Two WT power curves’ parameters were estimated; the first one through Least Squares algorithm, and the second one using non-linear optimization through unconstrained function minimization, then power residuals were generated from each power point. Finally, on-line CUSUM statistical change detection algorithm was used to evaluate and detect small changes in power residuals generated from the first model. The presented fault detection system successfully detected faults in both detection levels under realistic wind turbulence, and with fault magnitude of 2% efficiency degradation for the progressive degradation level | URI: | http://hdl.handle.net/20.500.11889/6319 |
Appears in Collections: | Theses |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Final_Model_Based_Fault_Detection_in_Wind_Turbines_revised.pdf | Thesis | 2.53 MB | Adobe PDF | View/Open |
Page view(s)
225
checked on Mar 25, 2024
Download(s)
284
checked on Mar 25, 2024
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.