Advanced Data Mining Approach for Wind Turbines Fault Prediction



Published Jul 8, 2014
Houari Toubakh Moamar Sayed-Mouchaweh


Wind turbine operation and maintenance costs depend on the reliability of its components. Thus, a critical task is to detect and isolate faults, as fast as possible, and restore optimal operating conditions in the shortest time. In this paper, a data mining approach is proposed for fault prediction by detecting the faults inception in the wind turbines, in particular pitch actuators. The role of the latter is to adjust the blade pitch by rotating it according to the current wind speed in order to optimize the wind turbine power production. The fault prediction of pitch actuators is a challenging task because of the high variability of the wind speed, the confusion between faults and noise as well as outliers, the occurrence of pitch actuator faults in power optimization region in which the fault consequences are hidden and the actions of the control feedback which compensate the fault effects. To answer these challenges, the proposed approach monitors a drift from normal operating conditions towards failure condition. To achieve drift detection, two drift indicators are used. The first indicator detects the drift and the second indicator confirms it. Both indicators are based on the observation of changes in the characteristics of normal operating mode over time. A wind turbine simulator is used to validate the performance of the proposed approach.

How to Cite

Toubakh, H., & Sayed-Mouchaweh, M. (2014). Advanced Data Mining Approach for Wind Turbines Fault Prediction. PHM Society European Conference, 2(1).
Abstract 94 | PDF Downloads 75



fault diagnosis, Wind Turbines, Pattern classification, Drift detection and monitoring

Blesa, J., Puig, V., Romera, J. & Saludes, J. (2011). Adaptation of set theoretic methods to the fault detection of wind turbine benchmark . ( 8316– 8321), August 28 - September 2, in Proceedings of IFAC World Congress.
Milan, Italy. doi: 10.3182/20110828-6-IT-1002.01842 Chammas, A., Duviella, E., & Lecoeuche S. (2013). Fault Diagnosis of Wind Turbine Drive Train Faults based on Dynamical Clustering, (5650 - 5655), December 10-13, in Proceedings of CDC Firenze, Italy . doi: 10.1109/CDC.2013.6760779
Chen, W., Ding, S., Sari, A., Naik, A., Khan, A., & Yin, S. (2011), Observerbased fdi schemes for wind turbine benchmark. General Guidelines. (7073–7078), August 28 - September 2,in Proceedings of IFAC World Congress 2011. Milan, Italy. doi: 10.3182/20110828-6-IT-1002.03469
Laouti, N., Sheibat-Othman, N., & Othman, S. (2011). Support vector machines for fault detection in wind turbines. (7067– 7072) ,August 28 - September 2, in Proceedings of IFAC World Congress. Milan, Italy. doi:10.3182/20110828-6-IT-1002.02560
Lecoeuche, S. & Lurette,C. (2003) Auto-adaptive and dynamical clustering neural network . (350–358) June 26–29. in ICA NN’03 Proceedings, Istanbul, Turkey . doi: 10.1007/3-540-44989-2_42
LIU .W, (1999). An extended Kalman filter and neural network cascade fault diagnosis strategy for the glutamic acid fermentation process. Artificial intelligence in engineering A. page 131-140, vol. 13, n°2,
Odgaard, P., & Stoustrup, J. (2009), Fault tolerant control of wind turbines a benchmark model.(155–160), June 06-07,in Proceedings of the 7th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes. Sants Hotel, Spain. IFAC, doi: 10.3182/20090630-4-ES-2003.00026
Odgaard, P., & J. Stoustrup. (2010).Unknown input observer based detection of sensor faults in a wind turbine. (310 –315), September 8-10, in Proceedings of Control Applications (CCA), Yokohama , Japan. doi: 10.1109/CCA.2010.5611266
Ozdemir, A., Seiler, P., & Balas, G. (2011). Wind turbine fault detection using counter-based residual thresholding. (8289– 8294), August 28 - September 2, in Proceedings of IFAC World Congress. Milan, Italy. doi:10.3182/20110828-6-IT-1002.01758
Kim, K., Parthasarathy, G., Uluyol, O., Foslien , W., Shuangwen, S.,Fleming P. (2011). Use of SCADA Data for Failure Detection in Wind Turbines. August 7-10 , ,in Proceedings of the Energy Sustainability Conference and Fuel Cell Conference, Washington, USA. doi: DE-AC36-08GO28308.
Kusiak, A., Li, W. (2011). The prediction and diagnosis of wind turbine faults, in Proceedings of Renewable Energy , volume 36, Issue 1Pages 16-23. doi:10.1016/j.renene.2010.05.014
Kusiak, A.,Verma, A. (2011). A Data-Driven Approach for Monitoring Blade Pitch Faults in Wind Turbines, Sustainable Energy, page 87-96. doi: 10.1109/TSTE.2010.2066585
Simani, S. , Castaldi, P. , & Bonfe, M. (2011). Hybrid modelbased fault detection of wind turbine sensors,(7061–7066), August 28 - September 2, in Proceedings of IFAC World Congress, Milan, Italy, doi: 10.3182/20110828-6-IT-1002.01311
Schlechtingen, M., & Santos, I. F. (2011) Comparative analysis of neural network and regression based condition monitoring approaches for wind turbine fault detection., Mechanical Systems and Signal Processing . Page 1849–1875. doi: org/10.1016/j.ymssp.2010.12.007.
Toubakh; H, Sayed-Mouchaweh, M. ; Duviella, E. (2013) Advanced Pattern Recognition Approach for Fault Diagnosis of Wind Turbines. (368 - 373) Decembre 4-7. in Proceedings of 12th ICMLA, Miami, USA. doi: 10.1109/ICMLA.2013.150
Traore, M. , Duviella, E., & Lecoeuche, S. (2009) Comparison of two prognosis methods based on neuro fuzzy inference system and clustering neural network. (1468-1473) June 06-07. in 7th IFAC symposium on Fault Detection, Supervision and Safety of Technical Processes, Sants Hotel, Spain. doi: 10.3182/20090630-4-ES-2003.00239
Zhang, X., Zhang, Q., Zhao, S., Ferrari, R. M., Polycarpou, M. M. & T. Parisini (2011). Fault detection and isolation of the wind turbine benchmark: An estimation-based approach, ( 8295– 8300), August 28 - September 2, in Proceedings of IFAC World Congress 2011. Milan, Italy. doi:10.3182/20110828-6-IT-1002.02808
Technical Papers