A Combination Model for Short Term Wind Power Forecasting

Abstract:

For analyzing the accuracy of wind power prediction, an analyzing model combined with multi-leaner and dynamic weight distribution is proposed. With this method, Numerical Weather Prediction (NWP), Wind power data (historical) and weather data (historical) are structured into several sample sets, each set has a different weight value, which determined by the training errors, these sample set is trained by different learner algorithm with a weight too. Finally, using these models to predict the outputs. The experiments indicate the effectiveness of the method this paper proposed. Compared with Single model of Support Vector Machine and Artificial Neural Network, the combination method has better performance in both calculation accuracy and generalization.

• Periodical: Advanced Materials Research (Volumes 791-793)
• Pages: 1220-1223
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.791-793.1220
• Online since: September 2013
• Authors: Ke Wen Liu, Jing Wang
• Keywords: Artificial Neural Network (ANN), Combination Forecasting, Dynamic Weight Distribution, New Energy Generation, Support Vector Machine (SVM)

References

[1] Tu Qiang. Existing Applications of Wind Power Forecasting Technology and Relevant Suggestions(in chinese) [J]. Power System and Clean Energy, 2009, 25(10): 4-9.

[2] Peng Huaiwu, Liu Fangrui, Yang Xiaofeng. Study of short-term wind power prediction based on artificial neural networks(in chinese) [J]. East China Electric Power, 2009, 37(11): 1918-(1921).

[3] Tao Yufei, Li Weihong, Yang Xifeng. Wind Speed Forecast Model for Wind Farms(in chinese) [J]. Power System and Clean Energy, 2009, 25(3): 53-56.

[4] Fan Gaofeng, Wang Weisheng, Liu Chun, Dai Huizhu. Wind Power Prediction Based on Artificial Neural Network(in chinese) [J]. Proceedings of the CSEE, 2008, 28(34): 118-123.

[5] Qi Shuangbin, Wang Weiqing, Zhang Xinyan. Wind speed and wind power prediction based on SVM(in chinese) [J]. East China Electric Power, 2009, 37(9): 1600-1603.

[6] R.E. Schapire. A brief introduction to boosting. In Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence, (1999).

[7] Y. Freund and R.E. Schapire. Experiments with a new boosting algorithm. In Proc. 13th International Conference on Machine Learning, pages 148-146. Morgan Kaufmann, (1996).

[8] R.E. Schapire. The boosting approach to machine learning: An overview. In Workshop on Nonlinear Estimation and Classification. MSRI, (2002).

[9] Y. Freund and R.E. Schapire. Game theory, on-line prediction and boosting. In Proc. 9th Annu. Conf. on Comput. Learning Theory, pages 325-332. ACM Press, New York, NY, (1996).

[10] R. Meir and G. Rätsch. An introduction to boosting and leveraging. In S. Mendelson and A. Smola, editors, Advanced Lectures on Machine Learning, LNCS, pages 119-184. Springer, 2003. In press. Copyright by Springer Verlag.

[11] R.E. Schapire and Y. Singer. Improved boosting algorithms using confidence-rated predictions. Machine Learning, 37(3): 297-336, December (1999).

DOI: 10.1145/279943.279960