A rolling ARMA method for ultra short term wind power prediction

摘要

Wind power is one of the popular renewable energy in the world. And accurate wind power prediction can improve the quality of wind power integration and help to guarantee the safety of power grid system. In this paper, a Rolling Auto-regressive Moving Average(RARMA) method is proposed to improve the prediction accuracy for ultra short term wind power prediction. And for the purpose of illustrating the feasibility of the RARMA method, numerical experiments are conducted. Simultaneously, the Root Mean Square Error(RMSE), the Maximum Error(MAXE), Average Relative Error(ARE), and Prediction Accuracy Rate(PAR) are selected to compare the effect of prediction methods. The numerical results show that the RARMA method exhibits a promising prediction accuracy.