A SCADA data based anomaly detection method for wind turbines

Abstract

In this paper, a data driven method for Wind Turbine system level anomaly detection is proposed. Supervisory control and data acquisition system (SCADA) data of a wind turbine is adopted and several parameters are selected based on physic knowledge and correlation coefficient analysis to build a normal behavior model. This model is based on Self-organizing map (SOM) which can project higher dimensional SCADA data into a two-dimension-map. After that, the Euclidean distance based indicator for system level anomalies is defined and a filter is created to screen out suspicious data points based on quantile function. Moreover, a failure data pattern based criterion is created for anomaly detection from system level. The method is tested with a two-month SCADA dataset with the measurement interval as 20 seconds. Results demonstrate capability and efficiency of the proposed method.