因應全球永續趨勢，以及配合政府發展再生能源政策，風力發電為關鍵的發展項目，然而風的波動性與間歇性造成風力發電供應不穩定，將直接衝擊電網的運行以及安全，不僅增加電力調度難度更是使電網運轉操作成本負擔沉重，因此研究風力發電的預測技術，有助於改善風電對於電網的不良影響，亦是發展風力發電的重要配套機制。
本文透過歷史數據建構風力發電模型，藉由預測海象風速與結合本文建構的發電模型進行風力發電預測驗證。首先應用K-means聚類分析技術，將大量的氣象資訊以聚類方式選出較有意義的訓練資料，藉以減少建模資料筆數與提高運算效率。接著根據挑選出的風場進行發電模型建構，以相同的聚類方法減少樣本訓練空間，提高樣本有效性。在預測模型建構上，以支撐向量機原理為基礎，建構迴歸模型，執行一時與一日的風速預測後，結合發電迴歸模型，實現風力發電的間接預測。為了降低預測誤差，本文以最低誤差為目標，使用改良具時變加速度係數粒子群最佳化演算法解決支撐向量機參數設置的問題，不僅增加預測模型的完整性，亦能提高預測準確性。

Wind power is a key development for the purpose of the global sustainability trends in renewable energy. However, influenced by factors of geography, circumstance and climates, the wind power has the characteristics of intermittency, volatility and uncontrollability. To ensure the reliable operation of a power system which is significantly fed by wind power, therefore, the study of wind power forecasting technology for assisting power system operation is becoming important.
In this thesis, based on the historical marine weather and corresponding wind power output data, a short-term wind power forecasting model for future one day is presented. Due to lots of history data of marine weather data and wind power, we divide data into clusters using K-means clustering algorithm to get the meaningful training data so as to reduce the number of modeling data and improve the efficiency of computing. Besides, based on the principle of Support Vector Machine (SVM), the regression model is constructed. We have been carried out wind speed forecasting for one hour and one day and with the correspondence between marine wind speed and the corresponding wind power regression model, the indirect wind power forecasting model is realized. In addition, proper parameter settings of support vector machine (SVM) is important to its efficiency and accuracy .In this paper, we enhance particle swarm optimization with time varying acceleration coefficients (PSO-TVAC) to perform the parameter optimization for SVM, which not only improve the forecast model availability , but also improve the forecasting accuracy .

論文審定書 i
誌謝 ii
摘要 iii
Abstract iv
目錄 v
圖次 viii
表次 xi
第一章 緒論 1
1.1 研究背景與動機 1
1.2 風力發電預測文獻回顧 2
1.3 研究目的與方法 5
1.4 論文架構 5
第二章 風力發電建模基礎 7
2.1 風速與發電量分析 7
2.1.1 風力發電原理 7
2.1.2 測站選取評估 8
2.1.2.1 風速 8
2.1.2.2 風力發電場 12
2.2 風電場實測數據預處理 14
2.2.1 風場中運行數據的檢驗 14
2.2.2 風電場實際發電量預處理 15
第三章 研究方法 17
3.1 K-means聚類分析基本原理 17
3.1.1 群聚前後風速的差異性 20
3.2 支撐向量機 21
3.2.1 分類與回歸 22
3.2.2 核心函數 25
3.3 風速預測模型 28
3.3.1 風速預測流程 29
3.4 發電模型建構 29
第四章 改良具時變加速度係數粒子群最佳化演算法 30
4.1.1 傳統粒子群演算法 30
4.1.2 時變性質加速係數粒子群演算法(PSO_TVAC) 33
4.1.3 改良時變性質加速係數粒子群演算法(EPSO_TVAC) 34
4.1.4 演算法的強韌性分析 37
第五章 案例測試與結果分析 38
5.1 前言 38
5.2 海象風速預測結果 38
5.2.1 風速歷史資料聚類分析 38
5.2.2 一時預測結果 41
5.2.3 一日預測結果 42
5.2.4 一週預測結果 44
5.3 風力發電建模 46
5.3.1 聚類分析應用於發電建模 46
5.3.2 建模效果比較 52
5.4 風力發電預測 52
第六章 結論及未來方向 58
6.1 結論 58
6.1.1 風速預測 58
6.1.2 建構發電模型與風電預測 58
6.2 未來發展方向 59
參考文獻 60
附錄A 風力發電預測結果 64

[1] GWEC,Global Wind Report Annual Market Update 2016.
[2] 台電網站：資訊揭露-發電資訊-再生能源發電概況-再生能源發電簡介http://www.taipower.com.tw/content/new_info/new_info-b36- 1.aspx?LinkID=8.
[3] Landberg, L., “Short-term prediction of local wind conditions,” Journal of Wind Engineering and Industrial Aerodynamics, vol. 89, no. 3, pp. 235-245, 2001.
[4] Nielsen, T.S., Tofting,John , and Madsen,Henrik, “Experiences with Statistical Methods for Wind Power Prediction,” 1999.
[5] Beyer HG, H.D., Mellinghoff H, Monnich K, Waldl H-P, “Forecast of regional power output of wind turbines,” Proceedings of the European Wind Energy Conference, pp. 1073.1999.
[6] Landberg, L. , “Short-term prediction of the power production from wind farms,”Journal of Wind Engineering and Industrial Aerodynamics, vol. 80, no. 1, pp. 207-220, 1999.
[7] Focken, U., Lange, Matthias and Waldl, Hans-Peter, “Previento - A Wind Power Prediction System with an Innovative Upscaling Algorithm,” Conference Proceedings, 2001.
[8] Pocken U.M, a.L.H.P.P. , “A Wind Power Prediction System with an Innovative Upscaling Algorithm,” in Proceedings of the European Wind Energy Conference, pp. 826-829. 2001.
[9] Bailey B, M.C.B., and J. Zack, “Short-term wind forecasting Proceedings of the European,” in Wind Energy Conference, 1999.
[10] I, M.P., “Wind forecasting activities,” in Proceedings of the First IEA Joint Action Symposium on Wind Forecasting Techniques, Norrkoping, Sweden, Published by FOI - Swedish Defense Research Agency, pp. 11-20, Dec. 2002.
[11] Giebel, G., et al., “The Zephyr-project. The next generation prediction system (poster) ,” European Wind Energy Association (EWEA), 2002.
[12] Jorgensen, J.C.M., B. O Gallaghoir, K. Sattler, and E. McKeogh, “ HIRPOM: Description of an operational numerical wind power prediction model for large scale integration of on and offshore wind power in Denmark,” in Poster on the Global Wind Power Conference and Exhibition, Paris, France, Apr. 25, 2002.
[13] Gu Xinkai, F.G., and Wang Xiaorong, “A review on wind power forecast technologies,” Power System Technology, vol. 31, no. 2, pp. 335-338, 2007.
[14] Yan Li, a.S.G., “A review on short term of wind power forecast technologies,” The twenty-fourth Annual Meeting Proceedings of China Electric Power System of Higher Education and its automation professional.
[15] Lijie Wang, X.L., Yang Gao, and Shuang Gao, “A review on modeling and forecasting of wind power,” Protection and Control in Power System, vol. 37, no. 13, pp. 118-121, 2009.
[16] Gong, W., et al., “Temporal Filtering of InSAR Data Using Statistical Parameters From NWP Models,” IEEE Transactions on Geoscience and Remote Sensing, vol. 53, no. 7,pp. 4033-4044, 2015.
[17] Carpinone, A., et al. , “Markov chain modeling for very-short-term wind power forecasting,” Electric Power Systems Research, 2015. 122, pp. 152-158.
[18] Karadas, M., et al., “Multiple regression analysis of performance parameters of a binary cycle geothermal power plant,” Geothermics, pp. 68-75, 2015.
[19] Cadenas, E., O.A. Jaramillo, and W. Rivera, “Analysis and forecasting of wind velocity in chetumal, quintana roo, using the single exponential smoothing method,” Renewable Energy, vol. 35, no. 5, pp. 925-930, 2010.
[20] Babazadeh, H., et al, “ An hour ahead wind speed prediction by Kalman filter,” in 2012 IEEE Power Electronics and Machines in Wind Applications, 2012.
[21] Zhang, J. and C. Wang, “Application of ARMA Model in Ultra-short Term Prediction of Wind Power,” in 2013 International Conference on Computer Sciences and Applications, 2013.
[22] Tagliaferri, F., I.M. Viola, and R.G.J. Flay, “Wind direction forecasting with artificial neural networks and support vector machines,” Ocean Engineering, vol. 97, pp. 65-73. 2015.
[23] Wang, X., et al., “Short-Term Wind Power Prediction Based on Wavelet Decomposition and Extreme Learning Machine,” in Advances in Neural Networks – ISNN 2012, 9th International Symposium on Neural Networks, Shenyang, China, pp. 645-653, 2012.
[24] Zhang, Q., et al., “A Fuzzy Group Forecasting Model Based on Least Squares Support Vector Machine (LS-SVM) for Short-Term Wind Power. Energies,” vol. 5, no. 9, 2012.
[25] Zeng, J. and W. Qiao, “Short-term solar power prediction using a support vector machine,” Renewable Energy, vol. 52, pp. 118-127, 2013.
[26] 葉泰和, "台灣風場風能評估方法分析," 新能源施工處.
[27] Pandit, Y.P., et al., “Classification of Indian power coals using K-means clustering and Self Organizing Map neural network,” Fuel,. vol. 90, no. 1, pp. 339-347. 2011
[28] Celebi, M.E., “Improving the performance of k-means for color quantization,” Image and Vision Computing, vol. 29, no. 4, pp. 260-271, 2011.
[29] Yiakopoulos, C.T., K.C. Gryllias, and I.A. Antoniadis, Rolling element bearing fault detection in industrial environments based on a K-means clustering approach. Expert Systems with Applications, vol. 38, no. 3, pp. 2888-2911, 2011.
[30] Kalyani, S. and K.S. Swarup, “Particle swarm optimization based K-means clustering approach for security assessment in power systems,” Expert Systems with Applications, vol. 38, no. 9, pp. 10839-10846, 2011.
[31] Burges, C.J.C., “A Tutorial on Support Vector Machines for Pattern Recognition,” Data Mining and Knowledge Discovery, vol. 2, no. 2, pp. 121-167, 1998.
[32] Chih-Wei, H. and L. Chih-Jen, “A comparison of methods for multiclass support vector machines,” IEEE Transactions on Neural Networks, pp. 415-425, 2002.
[33] Cortes, C. and V. Vapnik, “Support-Vector Networks,” Machine Learning, vol. 20,no. 3, pp. 273-297. 1995.
[34] Cristianini,N. and J.Shawe-Taylor, “An introduction to support Vector Machines, and other kernel-based learning methods,” Cambridge University Press , 2000,
[35] Sch, B., et al., “New Support Vector Algorithms,” Neural Comput., vol. 12, no. 5, pp. 1207-1245, 2000.
[36] Lin, W.m., et al., “Classification of Multiple Power Quality Disturbances Using Support Vector Machine and One-versus-One Approach,” in 2006 International Conference on Power System Technology, 2006.
[37] Scholkopf, B. and A.J. Smola, Learning with Kernels, “Support Vector Machines, Regularization, Optimization, and Beyond,” MIT Press, 2001.
[38] Kruif, B.J.d. and T.J.A.d. Vries, “Pruning error minimization in least squares support vector machines,” IEEE Transactions on Neural Networks, vol. 14, no. 3, pp. 696-702, 2003.
[39] Gestel, T.V., et al., “Financial time series prediction using least squares support vector machines within the evidence framework,” IEEE Transactions on Neural Networks, vol. 12, no. 4, pp. 809-821, 2001.
[40] Kecman, V., “Learning and Soft Computing, Support Vector Machines, Neural Networks, and Fuzzy Logic Models,” MIT Press, 2001,
[41] Tucker, H.W.K.a.A.W., “Nonlinear programming, Proceedings of 2nd Berkeley Symposium,” Berkeley, University of California Press, pp. 481-492, 1951,
[42] Aizerman, A., E.M. Braverman, and L.I. Rozoner, “Theoretical foundations of the potential function method in pattern recognition learning,” Automation and Remote Control, vol. 25, pp. 821-837, 1964.
[43] Kennedy, J. and R. Eberhart, “ Particle swarm optimization. in Neural Networks, 1995. Proceedings.,” IEEE International Conference on, 1995.
[44] Eberhart, R. and J. Kennedy, “A new optimizer using particle swarm theory. in Micro Machine and Human Science,” Proceedings of the Sixth International Symposium on, 1995.
[45] Shi, Y. and R.C. Eberhart, “Empirical study of particle swarm optimization,” in Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406). 1999.
[46] F. S. Cheng, J.S.T., C. H. Lv and M. T. Tsay, “A Generalized Regression Neural Network for Solving Economic Dispatch Problem,” ICEE for the 21st Century With focus on Sustainability and Reliability, 2007.
[47] Ratnaweera,A.,S.K.Halgamuge,and H.C.Watson, “Self-organizing hierarchical particle swarm optimizer with time-varying acceleration coefficients,” IEEE Transactions on Evolutionary Computation, vol. 8, no. 3, pp. 240-255, 2004.
[48] Chaturvedi, K.T., M. Pandit, and L. Srivastava, “Self-Organizing Hierarchical Particle Swarm Optimization for Nonconvex Economic Dispatch,” IEEE Transactions on Power Systems, vol. 23, no. 3, pp. 1079-1087, 2008.