System Health Monitoring of Wind Turbines Using SCADA Data and Gaussian Mixture Models
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Published
Jul 14, 2017
Akihisa Yasuda
Jun Ogata
Masahiro Murakawa
Hiroyuki Morikawa
Makoto Iida
Abstract
Wind turbines are the major driving force to produce renewable energy, but there is a strong need to reduce the costs of operation and maintenance. To detect anomalies of wind turbines, this paper proposes a method which uses data collected by the Supervisory Control And Data Acquisition (SCADA) system and is based upon building the normal behavior model of wind turbines. This is achieved by using supervised data and Gaussian mixture models with filtering SCADA data from the macroscopic point of view. The method is validated with SCADA data collected from actual 2-MW wind turbines. The result shows the potential of detecting anomalies and the effectiveness of filtering
conditions for building the model.
##plugins.themes.bootstrap3.article.details##
Keywords
PHM
References
C. Lins et al. (2015). Renewables 2015 global status report, REN21. http://www.ren21.net/renewables-2015-globalstatus- report-full-report-2
Chris Bishop (2006). Pattern recognition and machine learning. Springer.
Duda, Richard O., Peter E. Hart, & David G. Stork (2001). Pattern classification. 2nd., edition. New York Hamilton, James Douglas (1994). Time series analysis, Vol. 2. Princeton university press
Rohatgi, J. S., & Nelson, V. (1994). Wind characteristics: An analysis for the generation of wind power, Alternative Energy Institute, West Texas A&M University .
Kim, K., Parthasarathy, G., Uluyol, O., Foslien, W., Sheng, S., & Fleming, P. (2011). Use of SCADA data for failure detection in wind turbines. In ASME 2011 5th International Conference on Energy Sustainability (pp. 2071-2079), January, American Society of Mechanical Engineers
Martin, E. B., & A. J. Morris (1996). Non-parametric Confidence Bounds for Process Performance Monitoring Charts. Journal of Process Control, 6 6, 349-358
Yu, Jianbo (2011). Bearing Performance Degradation Assessment Using Locality Preserving Projections and Gaussian Mixture Models. Mechanical Systems and Signal Processing, 25 7, 2573-2588
Chris Bishop (2006). Pattern recognition and machine learning. Springer.
Duda, Richard O., Peter E. Hart, & David G. Stork (2001). Pattern classification. 2nd., edition. New York Hamilton, James Douglas (1994). Time series analysis, Vol. 2. Princeton university press
Rohatgi, J. S., & Nelson, V. (1994). Wind characteristics: An analysis for the generation of wind power, Alternative Energy Institute, West Texas A&M University .
Kim, K., Parthasarathy, G., Uluyol, O., Foslien, W., Sheng, S., & Fleming, P. (2011). Use of SCADA data for failure detection in wind turbines. In ASME 2011 5th International Conference on Energy Sustainability (pp. 2071-2079), January, American Society of Mechanical Engineers
Martin, E. B., & A. J. Morris (1996). Non-parametric Confidence Bounds for Process Performance Monitoring Charts. Journal of Process Control, 6 6, 349-358
Yu, Jianbo (2011). Bearing Performance Degradation Assessment Using Locality Preserving Projections and Gaussian Mixture Models. Mechanical Systems and Signal Processing, 25 7, 2573-2588
Section
Regular Session Papers