近年來因全球資源的耗竭及環保意識的抬頭，再生能源的應用越來越趨於重視，而將再生能源考慮進各相關議題的研究尤其重要。機組排程為電力系統多年來持續在致力於的一個問題，為電力市場價格決定的先決條件。以往傳統機組排程考慮之機組大多為火機、核能、或水力這些傳統的發電機組，並未將再生能源考慮進排程中。由於近年來技術的成熟，再生能源的發電量已逐步迎上傳統發電機組之發電量，又其核能發電機組的停用，再生能源併入機組排程的考量已孰不可等。然而，由於再生能源有著不穩定的特性，無法準確的精算出每小時再生能源的發電量，沒有準確的可靠性。因此如何將再生能源的不穩定性考慮後數值化，一直是近年來研究努力的目標。
本文採用近期提出的不穩定數值分析法，將再生能源的不穩定性數值化，以建立可靠度高的再生能源發電量，用於機組排程中以提高總發電量，又由於加入更多限制式及判斷條件，本文採用純人工智慧演算法，以有效率的求得所需各機組開關狀態，及應供應的發電量。最後經案例分析，在再生能源不穩定時，如何安全的調度，以持續穩定的供應所需再生量，此為本文最重要的貢獻。

Due to the global resources depletion and conception of Environmental awareness, application related to renewable energy were gradually being valued. The researches considering the renewable energy have increased in the recent year. Unit Commitment (UC) is a very important issue in the power system, which is the key to deciding electric price in the electric power market. Traditional generators in Unit Commitment included thermal units, nuclear units, or hydraulic power plants, excluding the renewable energy. However, according to the developed technique and development, the power generation would eventually be enough to consider it. Moreover, the abolition of the nuclear unit increased the demand for renewable energy. But the unstable power generation of renewable energy made the UC problem more complicated and would increase the difficulty computing solution in UC problem. Therefore, researches were working on enumerating the instability of renewable energy in recent years.
In this thesis, some recent methods of numerical analysis were adopted to analyze the instability of renewable energy, in order to apply for UC. Because of the more constraints and condition determination, in this thesis we use the pure Artificial Intelligence Algorithm to solve the UC problem more efficiently by separating the problem into two parts, one is the on/off the unit matrix and the other is 24-hour power generation. At the end of case discussions, we simulate the improper power generation forecasting, end in the similar solution after the situation happened, proving the stability of the UC combining numerical renewable energy in.

論文審定書 i
摘要 ii
Abstract iii
目錄 iv
圖目錄 vi
表目錄 vii
第一章 緒論 1
1.1 研究背景及動機 1
1.2 文獻回顧 3
1.3 研究目的與方法 4
1.4 論文架構 6
第二章 機組排程問題 7
2.1 機組排程 7
2.1.1 目標函數 7
2.1.2 機組運轉限制條件 8
2.2 優先順序表列法 10
2.2.1 發電機組優先順序選用評估 10
2.2.2 機組開關狀態評估 12
2.2.3 經濟調度與成本計算 14
第三章 演算法之理論與分析 15
3.1 蜂群演算法 15
3.1.1 演算法數學模型 16
3.1.2 改良式蜂群演算法 17
3.1.3 演算法流程圖 18
3.2 基因演算法 20
3.2.1 基因演算法演化機制 20
3.2.2 利用基因演算法求解機組排程 23
3.2.3 演化機制決策流程圖 24
第四章 再生能源發電量不確定性探討 26
4.1 再生能源發電模型 26
4.1.1 太陽能發電模型 26
4.1.2 風力發電模型 27
4.2 風險值之估算方法 28
4.2.1 歷史模擬法之風險評估 29
4.2.2 蒙地卡羅模擬法之風險評估 31
4.3 再生能源的信心水準與風險評估 33
4.3.1 風力發電之風速風險評估 33
第五章 案例分析與討論 39
5.1 10發電機組系統案例 39
5.1.1 10發電機機組參數 39
5.1.2 案例模擬結果與演算法強韌度分析 42
5.2 26發電機組系統案例 45
5.2.1 26發電機機組參數 45
5.2.2 案例模擬結果與演算法強韌度分析 49
5.3 澎湖尖山發電機組案例 55
5.3.1 發電機機組參數 55
5.3.2 風速與日照量之風險值估算 57
第六章 結論及未來研究方向 63
6.1 結論 63
6.2 未來研究方向 64
參考文獻 65

[1] 台灣電力公司-資訊揭露，https://www.taipower.com.tw/tc/pageList.aspx，2018年
[2] Z.Luo; G.Chen and Lirong Wang, “A Modified Sequence Quadratic Programming Method for Nonlinear Programming”, IEEE Conferences, p458-461, 2011.
[3] Yang Zhang; Fang Yao; Herbert H. C. Iu; Tyrone Fernando and Hieu Trinh, “Operation optimization of wind-thermal systems considering emission problem”, IEEE Conferences, p2199-2205, 2014.
[4] Satyendra Pratap Singh; Rachna Tyagi and Anubhav Goel, “Genetic Algorithm for Solving the Economic Load Dispatch”, International Journal of Electronic and Electrical Engineering, pp. 523-527, 2014.
[5] P. Surekha; N.Archana and S. Sumathi, “An Integrated GA-ABC Optimization Technique to Solve Unit Commitment and Economic Dispatch Problems”, Asian Journal of Scientific Research, p93-107, 2012.
[6] W.Yu; Z.Zeng; B.Peng; S.Yan; Y.Huang; H.Jiang; X.Li and T.Fan, “Multi-Objective Optimum Design of High-Speed Backplane Connector Using Particle Swarm Optimization”, IEEE Journals & Magazines, p35182-35193, 2018.
[7] K.Anoop and M.Nandakumar, “A novel maximum power point tracking method based on particle swarm optimization combined with one cycle control”, IEEE Conferences, p1-6, 2018.
[8] Manisha Govardhan; Ranjit Roy, “Evolutionary computation based unit commitment using Hybrid Priority List approach”, IEEE Conferences, p245-250, 2012.
[9] 李柏彥, 應用量子蟻拓法求解包含碳交易之短期火力機組排程, 民國102年
[10] G. Pampará; A.P Engelbrecht, “Binary artificial bee colony optimization”, IEEE Conferences, p1-8, 2011.
[11] Reza Akbari; Alireza Mohammadi and Koorush Ziarati, “A novel bee swarm optimization algorithm for numerical function optimization”, Communications in Nonlinear Science and Numerical Simulation. p 3142–3155, 2010.
[12] Reza Akbari; Alireza Mohammadi; Koorush Ziarati, “A powerful bee swarm optimization algorithm”, IEEE Conferences, p.1-6, 2009.
[13] T. Senjyu; S. Chakraborty; A.Y. Sabe; H. Toyama; A. Yona and T. Funabashi, “Thermal Unit Commitment Strategy with Solar and Wind Energy Systems Using Genetic Algorithm Operated Particle Swarm Optimization”, IEEE Conferences, p.866-871, 2008.
[14] Md. Sajid Alam; Matam Sailaja Kumari, “Unit commitment of thermal units in integration with wind and solar energy considering ancillary service management using priority list(IC) based genetic algorithm”, IEEE Conferences, p277-282, 2016.
[15] Anup Shukla; S. N. Singh, “Cluster-based Wind-Hydro-Thermal Unit Commitment Using GSA Algorithm”, IEEE Conferences, p.1-5, 2014.
[16] Y. Zhang; T. Fernando and Hieu Trinh, “Operation optimization of wind-thermal systems considering emission problem”, IEEE Conferences, p. 2199-2205, 2014.
[17] Y. Zhang; Yi Feng; W. Cao and Lin Zhang, “Unit commitment considering the effect of load and wind power uncertainty”, IEEE Conferences, p.1324-1328, 2014.
[18] Yongtao Wan; Lutao Zhao; Zhigang Zhang; Dongmei Ai, “An improved historical simulation method to estimate the amount of refined oil retail value at risk VaR”, IEEE Conferences, p.2401-2403, 2011.
[19] J. Chen; Xinjian Zhang, “The Research on Improvement and Application of VaR Model Based on Historical Simulation Method”, IEEE Conference, p.1-5, 2008.
[20] Cui Xiao-hai; Liu Xin-meng, “Uncertainty analysis of microwave power measurement with Monte Carlo method”, IEEE Conference, p.1-4, 2006.