Short-Term Wind Power Forecasting Based on SVM with Backstepping Wind Speed of Power Curve

Accurate wind farm power prediction can relieve the disadvantageous impact of wind power plants on power systems and reduce the difficulty of the scheduling of power dispatching department. Improving accuracy of short-term wind speed prediction is the key of wind power prediction. The authors have studied the short-term wind power forecasting of power plants and proposed a model prediction method based on SVM with backstepping wind speed of power curve. In this method, the sequence of wind speed that is calculated according to the average power of the wind farm operating units and the scene of the power curve is the input of the SVM model. The results show that this method can meet the real-time needs of the prediction system, but also has better prediction accuracy, is a very valuable short-term wind power prediction method.

[1]  R.W. Dunn,et al.  One-hour-ahead wind speed prediction using a Bayesian methodology , 2006, 2006 IEEE Power Engineering Society General Meeting.

[2]  Xiyun Yang,et al.  Research on modeling of wind turbine based on LS-SVM , 2009, 2009 International Conference on Sustainable Power Generation and Supply.

[3]  Dai Hui-zhu,et al.  A Combination Forecasting Model for Wind Farm Output Power , 2009 .

[4]  G. Sideratos,et al.  Using Radial Basis Neural Networks to Estimate Wind Power Production , 2007, 2007 IEEE Power Engineering Society General Meeting.

[5]  Gao Yang,et al.  Short-term Power Prediction of a Wind Farm Based on Wavelet Analysis , 2009 .