Advanced Data Mining Techniques for Power Performance Verification of an On-Shore Wind Farm
The monitoring of wind energy production is fundamental to improve the performances of a wind farm during the operational phase. In order to perform reliable operational analysis, data mining of all available information spreading out from turbine control systems is required. In this work a Supervisory Control and Data Acquisition (SCADA) data analysis was performed on a small wind farm and new post-processing methods are proposed for condition monitoring of the aerogenerators. Indicators are defined to detect the malfunctioning of a wind turbine and to select meaningful data to investigate the causes of the anomalous behaviour of a turbine. The operating state database is used to collect information about the proper power production of a wind turbine, becoming a tool that can be used to verify if the contractual obligations between the original equipment manufacturer and the wind farm operator are met. Results demonstrate that a proper selection of the SCADA data can be very useful to measure the real performances of a wind farm and thus to define optimal repair/replacement and preventive maintenance policies that play a major role in case of energy production.
KeywordsWind energy Wind turbine condition monitoring SCADA analysis Data mining
- 1.Catmull S (2011) Self-organising map based condition monitoring of wind turbines. In Proceedings of EWEA conference, 14–17 Mar 2011, Brussels, BelgiumGoogle Scholar
- 4.Carvalho H, Gaião M, Guedes R (2010) Wind farm power performance test, in the scope of the IEC 61400-12.3. In Proceedings of EWEC 2010 European wind energy conference and exhibition, 20–23 Apr 2010, Warsaw, PolandGoogle Scholar