Short Term Wind Power Forecasting Using Autoregressive Integrated Moving Average Modeling

Abstract – Wind energy is one of the most promising electricity generating sources as a clean and free alternate compared with the conventional power plants and due to the volatility feature in the wind speeds it will reflect some problems in power systems reliability particularly if the system is deeply penetrated by wind farms. Therefore, Wind power forecasting issue became and is still an important scope that will help in economic dispatch (ED), unit commitment (UC) purposes to get more reliable and economic systems. This paper introduces short term wind power forecasting model, based on autoregressive integrated moving average (ARIMA) which will be applied to hourly wind data from Zaafarana 5 project in Egypt. The proposed model successfully outperforms the persistence model with significant improvement up to 6 hours ahead. Index Terms – Wind forecasting, time series analysis, ARMA