Research on SCADA data preprocessing method of Wind Turbine

Abstract

The installed capacity of wind turbine is gradually increasing, and the cost of operation and maintenance of wind turbine is also gradually increasing. In order to reduce the cost of wind turbine operation and maintenance, using SCADA data for fault early warning and condition monitoring has become one of the hot research directions in recent years. The SCADA data of wind turbine include abandoned wind data, fault data, shutdown data and so on, these data can not correctly reflect the operating status of the unit, in order to improve the accuracy of early warning and facilitate the follow-up research work, it is necessary to preprocess the data. For this reason, this paper proposes a method of combining dispersion analysis and bin algorithm to preprocess the operation data of wind turbine. First of all, the principles of bin algorithm and dispersion analysis method are introduced, and then the improved bin algorithm is used to fit the wind speed-power curve. Finally, based on the power curve, the data are preprocessed by the method of dispersion analysis. The experimental results show that the combination of dispersion analysis and bin algorithm can effectively remove the abnormal data in SCADA data.