近年來隨著再生能源的興起，使得直流電力系統越趨重要，因此在直流電力系統中發生的電弧故障安全性也越趨重要。因電弧產生的火花與高溫現象容易導致火災，國內外均有直流電弧故障而引發火災與設備故障之案例，所以美國與台灣針對太陽能光電系統，制訂電弧故障保護的相關規範。為探討較精準與快速的直流電弧故障檢測技術，本論文建立一個直流電弧故障實驗平台，進行正常運轉與串聯電弧故障的測試。首先將測試的數據用離散小波轉換取得能量累積值，再使用支持向量機與倒傳遞類神經網路分別訓練檢測技術，訓練出檢測技術後再套用於FPGA開發板。最後將量測之實驗數據使用本文檢測方法與商用AFD進行比較。本論文提出之檢測方法均能檢測出電弧故障。支持向量機的優點有解決非線性問題、於小樣本與非線性等分類應用場合具有獨特優勢、分類思想較簡單、可採用多級支持向量機的形式樣本進行多級判斷。若未來能夠將硬體實現，將有助於提升直流電力的安全性。

In recent years, with the increasing of renewable energy, DC power systems have become more and more important. So the safety about arc faults occurring in DC power systems has become more and more important. Sparks and high temperature phenomena caused by arcs would cause fire events. There are several cases in our country and abroad about the DC arc faults that cause fires and equipment failures. Therefore, the United States and Taiwan have formulated specifications of arc fault protection in solar photovoltaic systems. In order to investigate more accurate and rapid detection technology of DC arc faults, this thesis establishes a DC arc fault test platform for tests under normal operation and series arc fault. Firstly, the test data is analyzed by using discrete wavelet transform to obtain the accumulation energy value, and then the support vector machine and the back propagation neural network, Respectively, used to train the detection technology. After the detection technology is trained, it is applied to the FPGA development board. Finally, the experimental test by using the detection method of this thesis was compared with the commercial Arc Fault Detector. The detection methods proposed in this thesis can detect arc faults more accurately. The advantages of support vector machines are that they solve nonlinear problems, have unique advantages in small sample and nonlinear classification applications, and have simple classification ideas. Multi-level judgment is performed by using a form sample of a multi-level support vector machine. If the hardware can be implemented in the future, it will help to improve the safety of DC power.

[1] “UL 1699B, Outline of photovoltaic(PV) dc arc-fault circuit protection, Issue Number 1” Underwriters Laboratories Inc January 14 2013.
[2] H. Haeberlin and M. Real, “Arc detector for remote detection of dangerous arcs on the DC side of PV plants” in 22nd European Photovoltaic Solar Energy Conference, Milano, Italy, 2007, vol. 200.
[3] 「北市自來水廠大火，疑太陽能板自燃」https://www.jdauspice.com/tw/ 2016.
[4] G.-S. Seo, H. Bae, B.-H. Cho, and K.-C. Lee, “Arc protection scheme for DC distribution systems with photovoltaic generation” in 2012 International Conference on Renewable Energy Research and Applications (ICRERA), 2012: IEEE, pp. 1-5.
[5] “NEC Section 690.11 Arc Fault Cricuit Protection” National Electrical Code 2011.
[6] “A New Look At PV Supply Stations” Electrical contractor 2016.
[7] 「屋內線路裝置規則第三百九十六條之三十」中華民國經濟部 一月 民國104年.
[8] D. A. Dini, P. W. Brazis, and K.-H. Yen, “Development of arc-fault circuit-interrupter requirements for photovoltaic systems” in 2011 37th IEEE Photovoltaic Specialists Conference, 2011: IEEE, pp. 001790-001794.
[9] Y. Cao, J. Li, M. Sumner, E. Christopher, and D. Thomas, “Arc fault generation and detection in DC systems” in 2013 IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC), 2013: IEEE, pp. 1-5.
[10] T. Taufik, C. Aarstad, and A. Kean, “Arc Fault Characterization System for the Low Voltage DC Arc Fault Circuit Interrupter” in 2017 25th International Conference on Systems Engineering (ICSEng), 2017: IEEE, pp. 106-112.
[11] A. Shekhar, L. Ramírez-Elizondo, S. Bandyopadhyay, L. Mackay, and P. Bauera, “Detection of series arcs using load side voltage drop for protection of low voltage DC systems” IEEE Transactions on Smart Grid, vol. 9, no. 6, pp. 6288-6297, 2018.
[12] M. Dargatz and M. Fornage, “Method and apparatus for detection and control of DC arc faults” ed: Google Patents, 2016.
[13] J. Zuercher, J. Hastings, E. Hetzmannseder, J. Pardee, and C. Tennies, “Detection of arcing in dc electrical systems” ed: Google Patents, 2004.
[14] Y. Ohta and H. Isoda, “Arc detecting device and aircraft equipped therewith” ed: Google Patents, 2012.
[15] J. A. Momoh and R. Button, “Design and analysis of aerospace DC arcing faults using fast fourier transformation and artificial neural network” in 2003 IEEE Power Engineering Society General Meeting (IEEE Cat. No. 03CH37491), 2003, vol. 2: IEEE, pp. 788-793.
[16] G. Yunmei, W. Li, W. Zhuoqi, and J. Binfeng, “Wavelet packet analysis applied in detection of low-voltage dc arc fault” in 2009 4th IEEE Conference on Industrial Electronics and Applications, 2009: IEEE, pp. 4013-4016.
[17] 蘇俊銘,「直流電力系統線路串聯電弧故障檢測」碩士學位論文，國立臺灣科技大學，臺北 民國105年.
[18] J. D. Johnson, S. S. Kuszmaul, W. I. Bower, and D. A. Schoenwald, “Using PV module and line frequency response data to create robust arc fault detectors” Sandia National Lab.(SNL-NM), Albuquerque, NM (United States), 2011.
[19] X. Yao, L. Herrera, and J. Wang, “A series DC arc fault detection method and hardware implementation” in 2013 Twenty-Eighth Annual IEEE Applied Power Electronics Conference and Exposition (APEC), 2013: IEEE, pp. 2444-2449.
[20] X. Yao, L. Herrera, S. Ji, K. Zou, and J. Wang, “Characteristic study and time-domain discrete-wavelet-transform based hybrid detection of series DC arc faults” IEEE Transactions on Power Electronics, vol. 29, no. 6, pp. 3103-3115, 2014.
[21] M. Naidu, T. J. Schoepf, and S. Gopalakrishnan, “Arc fault detection scheme for 42-V automotive DC networks using current shunt” IEEE Transactions on power electronics, vol. 21, no. 3, pp. 633-639, 2006.
[22] 張大凱,「電的旅程」天下遠見出版股份有限公司 民國90年.
[23] 「能源思考題（六）：直流電再戰交流電」http://www.thinkingtaiwan.com/content/1566 2013.
[24] 「太陽能發電系統」銓泰環能科技股份有限公司，https://www.inpower999.com.tw/solar-energy-generation-system.htm.
[25] 「中興電工微電網整體解決方案」http://www.chem.com.tw/products/微電網整體解決方案.
[26] 張存德,「中興電工機械公司混合式微電網發展現況」電力人：台灣旅沙電力協會會刊 pp. 第16期，第77-80頁, 十一月 民國103年.
[27] K. George, “DC power production, delivery and utilization” Electric Power Research Institute White Paper, 2006.
[28] 黃隆洲、王正健,「台灣及國際智慧電動車發展現況與挑戰」pp. 能源報導，第5-10頁, 八月 民國101年.
[29] S. Soylu, “Electric vehicles-modelling and simulations” 2011.
[30] 曾元超,「防範住家電器火災的新技術」pp. 台電月刊，第549期，第26-31頁, 民國97年.
[31] 陳又琨,「應用小波轉換於直流電力系統之串聯電弧故障偵測」碩士學位論文，國立臺灣科技大學，臺北 民國103年.
[32] 陳金龍,「配電盤弧光閃絡保護與危險等級分析之研究」碩士學位論文，國立臺灣科技大學，臺北 民國98年.
[33] “VAMP 120&121 Arc Flash Protection Units” Schneider Electric March 2012.
[34] 史明哲,「應用模糊理論與類神經網路於低壓線路電弧故障檢測」碩士學位論文，國立臺灣科技大學，臺北 民國104年.
[35] “KB electric LLC” https://kbelectricpa.com/afci-breakers/
[36] 王奕捷,「低壓線路串聯電弧故障時域及頻域分析與檢測」碩士學位論文，國立臺灣科技大學，臺北 民國100年.
[37] F. M. Uriarte et al., “A DC arc model for series faults in low voltage microgrids” IEEE Transactions on smart grid, vol. 3, no. 4, pp. 2063-2070, 2012.
[38] J. Johnson, C. Oberhauser, M. Montoya, A. Fresquez, S. Gonzalez, and A. Patel, “Crosstalk nuisance trip testing of photovoltaic DC arc-fault detectors” in 2012 38th IEEE Photovoltaic Specialists Conference, 2012: IEEE, pp. 001383-001387.
[39] “NEC Section 690.11 Arc Fault Cricuit Protection” National Electrical Code 2014.
[40] “Manual of 5SM Arc Fault Detection Unit for Photovoltaics” Siemens April 2015.
[41] “Datasheet of Firefighter Switch with Arc Fault Detection PVSEC-AF” E-T-A Engineering Technology 2015.
[42] “Datasheet of PVAF-T Series Arc Fault Detection for Solar Applications” Sensata Technologies January 205.
[43] B. Novak, “Implementing Arc Detection in Solar Applications: Achieving Compliance with the New UL 1699B Standard,” Texas Instruments, August 2012.
[44] A. V. O. a. R. W. Schafer, “Discrete-Time Signal Processing. 3rd edition” Prentice Hall 2010.
[45] 江國銘,「以FPGA為設計晶片之電壓閃爍計算」碩士學位論文，國立臺灣科技大學，臺北 民國100年.
[46] 林銀議,「信號與系統」五南圖書出版股份有限公司 民國98年.
[47] 劉鈺韋,「運用小波轉換與神經網路檢測屋內低壓線路串聯電弧故障」博士學位論文，國立臺灣科技大學，臺北 民國104年.
[48] 董長虹、高志、於嘯海,「小波分系工具箱原理與應用」國防工業出版社 民國93年.
[49] I. Daubechies, Ten lectures on wavelets. Siam, 1992.
[50] K. Sayood, Introduction to data compression. Morgan Kaufmann, 2017.
[51] R. M. R. a. A. S. Bopardikar, “Wavelet Transfrom Introduction to Theory and Applications” Addison-Wesley, California 1998.
[52] W. G. Morsi and M. El-Hawary, “Reformulating power components definitions contained in the IEEE Standard 1459–2000 using discrete wavelet transform” IEEE Transactions on Power Delivery, vol. 22, no. 3, pp. 1910-1916, 2007.
[53] 陳慶芳,「心音訊號之分割與特徵擷取」碩士學位論文，義守大學，高雄 民國101年.
[54] 王進德,「類神經網路與模糊控制理論入門與應用」全華圖書股份有限公司 2011.
[55] S. S. Haykin, S. S. Haykin, S. S. Haykin, K. Elektroingenieur, and S. S. Haykin, Neural networks and learning machines. Pearson education Upper Saddle River, 2009.
[56] Y. Liu, R. Wang, and Y.-S. Zeng, “An improvement of one-against-one method for multi-class support vector machine” in 2007 International Conference on Machine Learning and Cybernetics, 2007, vol. 5: IEEE, pp. 2915-2920.
[57] W.-M. Lin, C.-H. Wu, C.-H. Lin, and F.-S. Cheng, “Detection and classification of multiple power-quality disturbances with wavelet multiclass SVM” IEEE transactions on power delivery, vol. 23, no. 4, pp. 2575-2582, 2008.
[58] 薛小剛、葛毅敏,「Xilinx ISE 9.X FPGA/CPLD 設計指南」人民電郵 2007年.
[59] “ML505/506/507 Overview and Setup” Xilinx 2008.
[60] 徐文波、田耘,「Xilinx FPGA 開發實用手冊」佳魁資訊 民國104年.
[61] 陸瑞強、廖裕評,「系統晶片設計使用Quartus II」全華圖書 民國97年.
[62] 柯志遠,「低壓線路串聯電弧故障檢測之FPGA晶片設計」碩士學位論文，國立臺灣科技大學，臺北 民國104年.