太陽光伏發電系統可以分為小型與大型發電系統，小型發電系統會直接由太陽光伏板接逆變器後連接負載或是市電，大型發電會由多個小型發電系統與逆變器連接後，再經由變壓器進行升壓後併聯到較高電壓等級的電網傳送功率。由於太陽光伏發電系統的輸出特性，其設備常運轉於輕載，而變壓器輕載效率較差，故長期而言將導致大型太陽光伏發電系統於變壓器的損耗能量過多。此外太陽光伏發電主要受照度大小影響發電量，而每天照度為不連續且變化快速，故其發電量難以精準預測。
本論文提出一考量可靠度之太陽光伏發電系統併網變壓器的最佳設計與控制，文中利用一改良式的情境縮減法，找出具代表性的太陽光伏發電系統的發電量數據樣態，於減少數據樣態同時也不至於過度降低精準度，故可提升分析的速度與精度。此外，為避免太陽光伏發電系統於任一台變壓器損壞時，導致無法併聯運轉，本論文在變壓器N-1可靠度條件下，利用基因演算法進行太陽光伏發電系統併網變壓器的最佳設計與控制。藉由本文所提出之變壓器最佳設計與控制，選定變壓器容量與數量後，進行變壓器併網排程，可以提升太陽光伏發電系統併網的輕載效率。於模擬結果中，以一額定3200kWp太陽光伏發電系統為例，透過本論文之變壓器最佳設計與控制後，大致可提升0.6%的轉換效益。以台灣目前規劃的2025年20GWp的太陽光伏發電裝置容量，如平均每天發電80GWh，則可提升0.48GWh，一年就能多提供約175.2GWh的電量。如以2017年度台灣太陽光伏發電系統躉售價格4.4098元/kWh為例，每年額外產生的售電效益約為7.73億元。

Photovoltaic Generation System (PVGS) can be divided into small-scale and large-scale PVGSs. A small-scale PVGS is composed of photovoltaic panels and inverters and is directly interconnected to loads or low-voltage power grids. A large-scale PVGS is composed of many small-scale PVGSs and then is interconnected to high-voltage power grids by transformers. The PVGS devices are usually operated at light loads due to the output characteristic of PVGS. The transformers have a poor light-load conversion efficiency; therefore, it will cause significant energy loss for a large-scale PVGS in the long run. In addition, PVGS output is mainly affected by solar irradiances that are not continuous and rapid changes, the PVGS output is difficult to accurately predict too.
An optimal design and control of interconnected transformers for PVGSs considering reliability criterion is proposed in this thesis. A modified scenario reduction method is adopted in this thesis to find out the representative data patterns of PVGS outputs. The proposed scenario reduction method can be used to reduce the data patterns within acceptable accuracy; therefore, the analysis speed and accuracy can be improved. To prevent the interconnection failure of PVGS due to a single transformer fault, an optimal design and control of interconnected transformers for PVGSs based on genetic algorithm considering N-1 reliability criterion is implemented. With the proposed design and control, the number and capacity of transformers can be determined. The interconnected transformers can then be scheduled to enhance the light-load conversion efficiency of PVGSs. Simulation results show that 0.6% conversion efficiency improvement can be realized for a PVGS with rated capacity of 3200kWp by the proposed optimal design and control of interconnected transformers. PVGSs with a total rated capacity of 20GWp will be installed in Taiwan in 2025. If the average daily power generation is about 80GWh, then the generation increases of 0.48GWh and 175.2GWh can be obtained daily and yearly respectively. Using the wholesale price of 4.4098 NTD/kWh for PVGS in 2017 in Taiwan as an example, the additional benefit of annual electricity sales is about 7.73 billion NTD.

論文審定書 i
誌謝 ii
摘要 iii
Abstract iv
目錄 v
圖目錄 viii
表目錄 xi
第一章 緒論 1
1.1研究背景 1
1.2研究動機 3
1.3章節概要 4
第二章 太陽光伏發電系統輸出特性分析數據 6
2.1 太陽光伏發電系統介紹 6
2.2 太陽光伏發電輸出特性 7
2.3 情境縮減方法介紹 10
2.3.1 情境縮減簡介 10
2.3.2 本文使用之情境縮減法 13
第三章 考量可靠度之變壓器最佳設計與控制 17
3.1變壓器介紹 17
3.1.1 變壓器效率計算 17
3.1.2 變壓器保護 18
3.2變壓器併聯之最佳控制 21
3.3考量可靠度之變壓器最佳設計 23
3.4基因演算法介紹 26
3.4.1複製 27
3.4.2交配 28
3.4.3突變 29
3.5本文基因演算法設計 29
3.5.1變壓器併聯之最佳控制之基因演算法 30
3.5.2考量可靠度之變壓器最佳設計之基因演算法 34
第四章 模擬結果與討論 37
4.1模擬規格 37
4.1.1變壓器可控與未控效率比較 38
4.2情境縮減結果及探討 40
4.2.1情境縮減流程實例 41
4.2.2情境縮減冬月為例 46
4.2.3情境縮減夏月為例 52
4.2.4情境縮減整年為例 59
4.3變壓器最佳設計與控制案例分析 62
4.3.1裝置容量3200kWp之太陽光伏發電為例 62
4.3.2不同太陽光伏發電系統之變壓器最佳設計與控制 65
第五章 結論與未來研究方向 68
5.1結論 68
5.2未來研究方向 68
參考文獻 70
附錄A 2015年情境縮減資料 73

[1] Renewable Energy Policy Network for the 21st Century (REN21), “RENEWABLES 2017 GLOBAL STATUS REPORT,” 2017.
[2] “台電網站”， http://www.taipower.com.tw/
[3] T. Kerekes, E. Koutroulis and D. S´era, “An Optimization Method for Designing Large PV Plants,” IEEE Journal of Photovoltaics, vol. 3,issue: 2, pp. 814-822, Apr. 2013.
[4] International Finance Corporation “Utility-Scale Solar Photovoltaic Power Plants,” 2015
[5] N. Growe-Kuska, H. Heitsch and W. Romisch, “Scenario Reduction and Scenario Tree Construction for Power Management Problems,” in Proc IEEE Power Tech Conference Proceedings, Bologna, pp. 7, Jun. 2003.
[6] J. M. Morales, S. Pineda, A. J and Conejo, M. Carrion, “Scenario Reduction for Future Market Trading in Electricity Markets,” IEEE Transactions on Power Systems, vol 24, no2, pp. 878-888, May 2009.
[7] J. Dupaˇcov´a, N. Gr¨owe-Kuska and W. R¨omisch “Scenario reduction in stochastic programming - An approach using probability metrics,” May 2002.
[8] Holger Heitsch and Werner R¨omisch “Scenario reduction in stochastic programming - An approach using probability metrics,” KU Leuven Energy Institute WP EN2014-15, 2014MSC
[9] Kenneth Bruninx, Erik Delarue & William D‘haeseleer “A practical approach on scenario generation & reduction algorithms for wind power forecast error scenarios,” Tmeworking paper - Energy and Environment Last update: Sep 2014
[10] D. Phaengkieo and S. Ruangsinchaiwanich, “Optimization of Three-Phase Transformer Design using Adaptive Genetic Algorithm,” in Proc IEEE ICEMS, Chiba, pp. 1-5, Nov 2016.
[11] 維基百科“變壓器- 維基百科”，https://zh.wikipedia.org/wiki/變壓器.
[12] 財團法人台灣大電力研究試驗中心，”配電變壓器能源效率測試方法座談會”，民國103年
[13] 楊盛鐘，”變壓器保護簡介”， 台灣電力公司 民國103年
[14] Ed Zientek, PE, Square D Engineering Services, “Loading Considerations When Paralleling Transformers,” Oct. 2011
[15] L. Petkovska, M. Digalovski and G. Cvetkovski, P. Lefley, “A novel approach to multi-objective efficiency optimisation for a distribution transformer based on the Taguchi method,” in Proc IEEE ICEM, Berlin, pp. 2228-2234, Sep. 2014.
[16] “趣味電力：變壓器並列運行及負荷分配計算”，https://kknews.cc/zh-tw/science/eopep4.html
[17] F. H. Tseng, X. Wang, L. D. Chou, H. C. Chao and V. C. M. Leung, “Dynamic Resource Prediction and Allocation for Cloud Data Center Using the Multiobjective Genetic Algorithm,” IEEE Systems Journal, vol: PP, issue: 99, pp. 1-12, Jul. 2017.
[18] T. P. Hong, Y. C. Peng, W. Y. Lin and S. L. Wang, “Migration Effect of Hierarchical Multi-population Genetic Algorithm,” in Proc IEEE CYBCONF, Exeter, pp. 1-4, Jun. 2017.
[19] R. Maniu; L. A. Dumitru, “Genetic Algorithm – Adaptive Crossover Based on Solution Distribution in Search Space,” in Proc IEEE OPTIM, Brasov, pp. 899-904, May 2017.
[20] 百度文庫，“S11及S9變壓器型號規格表_圖文_百度文庫”，https://wenku.baidu.com/view/5aa83c8f360cba1aa911da31.html.
[21] “電力行業：油浸式變壓器產品價格購買指數 - 壹讀”， https://read01.com/zh-my/G40Bk5.html#.WX89EISGMdU
[22] ebay ，“US Ship AC to AC Three 3 Phase Solid State Relay SSR 40A 90-280VAC 380V AC _ eBay” http://www.ebay.com
[23] 東技企業，”日本KOYO真空斷路器VCB價格”，http://www.toyotech.com.tw/products_detail.php?Key=172&cID=16
[24] “NFB_Price”，華城電機，http://www.fortune.com.tw/
[25] “中華民國統計資訊網（專業人士）”， https://www.stat.gov.tw/mp.asp?mp=4
[26] 科技新報，“台政府公布再生能源目標，太陽能定調 20GW _ TechNews 科技新報” ， https://kknews.cc/zh-tw/finance/9lv3lq.html
[27] “台灣銀行”，http://www.bot.com.tw/Pages/default.aspx
[28] 陳盈在，”「太陽光電系統之規劃設計與其電能轉換設備蒐集資料技術」實習”“http://report.nat.gov.tw/ReportFront/report_download.jspx?sysId=C09903834&fileNo=001”