Power curve modeling for wind turbine systems: a comparison study

Abstract

The main goal of this study is to determine which one of the current methods is the most appropriate in representing the power curve of commercial Wind Turbine Systems (WTSs). For this purpose, the adequacy of various power curve methods is quantitatively evaluated with reference to three well-known statistical indices; Root Mean Square Error (RMSE), Coefficient of determination (R2) and relative Mean Bias Error (rMBE). The capabilities of these methods in representing power curve of commercial WTSs are also evaluated on predicting the mean power of these WTSs assuming that they were located at a specified wind site. Several WTSs (about 100 wind turbines) with their experimental data are undertaken in the analyses. Results indicate that the methods based on capacitor charging voltage equation (M10) and quadratic form of wind speed (M4) are more effective than the remaining ones in predicting the output power of WTSs. © 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group.