This paper presents a cointegration-based method for condition monitoring and fault detection of wind turbines. The proposed method is based on the residual-based control chart approach. The main idea is that cointegration is a property of some sets of nonstationary time series where a linear combination of the nonstationary series can produce a stationary residual. Then the stationarity of cointegration residuals can be used in a control chart as a potentially effective damage feature. The method is validated using the experimental data acquired from a wind turbine drivetrain with a nominal power of 2 MW under varying environmental and operational conditions. Two known abnormal problems of the wind turbine are used to illustrate the fault detection ability of the method. A cointegration-based procedure is performed on six process parameters of the wind turbine where data trends have nonlinear characteristics. Analysis of cointegration residuals —obtained from cointegration process of wind turbine data—is used for operational condition monitoring and fault/abnormal detection. The results show that the proposed method can effectively monitor the wind turbine and reliably detect abnormal problems.
CITATION STYLE
Dao, P. B., Staszewski, W. J., & Uhl, T. (2018). Operational condition monitoring of wind turbines using cointegration method. In Applied Condition Monitoring (Vol. 9, pp. 223–233). Springer. https://doi.org/10.1007/978-3-319-61927-9_21
Mendeley helps you to discover research relevant for your work.