Study on Comparing Different Methods for Wind Power Real-Time Forecast

Xingjie Liu, Tianyun Cen, Wenshu Zheng, and Junliang Bai

Keywords

Wind Power, RealTime Forecast, ARMA, Neural Networks

Abstract

Accurate wind power forecasting is significant for safe dispatching, power system stability and operation of wind farm. In this paper, ARMA, BP, RBF and SVM, four typical methods for wind power real-time forecast were discussed. Through the test of different input parameters model we got the optimal model with the minimum error, and we made comprehensive comparisons study on overall forecasting performances and different steps forecasting performances of those four kinds of optimal model. The results show that the neural networks of different input parameters can produce different forecasting performances, and their effects are different; BP, RBF and SVM model are better than ARMA; SVM of optimal model has better forecasting performance than BP and RBF.

Important Links:



Go Back