Research on wind power ramp events prediction based on strongly convective weather classification

Abstract

In this study a forecasting model for wind power ramps based on strongly convective weather classification is presented. First, the dynamics and thermodynamic behaviours of strongly convective weather are characterised by the predictors in the selected region. Then the support vector domain description for ramps scenario classification is introduced to establish an initialised extremum model, and the parameter templates method is used to identify the ramps weather in strongly convective weather library. Meanwhile, the original wind speed data is modified to obtain more accurate wind speed, and a new wind power ramps definition is proposed based on the ramp character itself and its impact on the power grid. Thus the catastrophe detection method (Bernaola Galvan algorithm) used for strongly convective weather forecasting. Finally, the wind power ramps forecasting method based on the discrimination of convective weather is developed. Comparing with the existing wind power ramps forecasting algorithms, the proposed prediction method here gets into meteorologic essence of triggering great fluctuation of wind speed.