Intelligent multiperiod wind power forecast model using statistical and machine learning model

With the rapidly increasing integration of wind energy into modern grid system, prediction (WPP) is playing an important role in planning and operation electrical distribution system. However, time series data always has nonlinear non-stationary characteristics, which still a great challenge to be accurately predicted. This paper proposes intelligent power forecast model evaluates long term, short term medium power. It uses statistical machine learning approach for finding best multiperiod forecasting. The been tested on Sotavento farm historical data, located Galicia, Spain. experimental results show that random forest better accuracy than other models proposed evaluated RMSE, MAE metrics. shown forecast. improved by 72.12% case 50.49% term.