Abstract
The cost of operation and maintenance of wind turbines is a significant part of the overall cost of wind turbines. To reduce this cost a method for enabling early fault detection is proposed and tested in this paper. The method is taking advantage of the fact that wind turbines in wind farms are located near similar wind turbines. This is done by generating a model for each turbine, the model is then used to evaluate the performance of that turbine and the nearby turbines. The evaluations from the models are then combined and used as votes to identify the faulty turbines. The method is applied and tested on historical Supervisory Control And Data Acquisition (SCADA) data from nine operational turbines over a testing period of nine months. The performance of the fault detection is found to be acceptable based on the testing period. During the testing period several gear related services were performed, some of these were predicted by the proposed fault detection systems. The advantage of the purposed method is that it applicable for operational turbines without requiring any extra measurements, since the used SCADA data is available from most modern wind turbines.
Original language | English |
---|---|
Title of host publication | 51st AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition |
Publisher | American Institute of Aeronautics and Astronautics |
Publication date | 7 Jan 2013 |
ISBN (Print) | 978-1-62410-181-6 |
DOIs | |
Publication status | Published - 7 Jan 2013 |
Event | 31st ASME Wind Energy Symposium - Grapevine, TX, United States Duration: 7 Jan 2013 → 10 Jan 2013 |
Conference
Conference | 31st ASME Wind Energy Symposium |
---|---|
Country/Territory | United States |
City | Grapevine, TX |
Period | 07/01/2013 → 10/01/2013 |