簡易檢索 / 詳目顯示

研究生: 林以樵
Yi-Chiao LIN
論文名稱: 應用決策樹於交流電路串聯電弧故障檢測 與FPGA晶片設計
Application of Decision Tree for Detection of Series Arc Fault in AC Circuit and FPGA-Based Chip Design
指導教授: 吳啟瑞
Chi-Jui Wu
口試委員: 辜志承
Jyh-Cherng Gu
連國龍
Kuo-Lung Lian
郭明哲
Ming-Tse Kuo
學位類別: 碩士
Master
系所名稱: 電資學院 - 電機工程系
Department of Electrical Engineering
論文出版年: 2019
畢業學年度: 107
語文別: 中文
論文頁數: 112
中文關鍵詞: 電弧故障小波轉換類神經網路決策樹FPGA
外文關鍵詞: Arc Fault, Wavelet Transform, Neural Network, Decision Tree, FPGA
相關次數: 點閱:240下載:0
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  • 根據國外調查指出,電弧事故是引起電器火災的主要原因之一,當低壓線路發生電弧故障時,電弧所產生的高溫可能點燃附近的易燃物,而導致火災的發生。本論文藉由電弧故障檢測平台,針對線路中供應不同特性的家電負載進行實驗,使用快速傅立葉轉換分析線路電流之時域和頻域特性,作為開發電弧故障檢測方法的基礎。針對電弧故障檢測,本文結合離散小波轉換和高頻能量累積法得到線路電流每週期的特徵向量,再分別結合倒傳遞類神經網路與決策樹組合出三種檢測方法。利用這三種方法進行線路正常運轉、串聯電弧故障和發生開關電弧的測試,並與商用電弧故障斷路器(Arc-Fault Circuit Interrupter, AFCI)的檢測結果相比較。由測試結果顯示,所設計的三個方法,誤判情形低於商用AFCI。而決策樹在硬體消耗及檢測準確度均優於倒傳遞類神經網路。最後,以FPGA實現三種檢測方法,驗證本文所提出之電弧故障檢測方法於硬體上實現的可行性。


    According to foreign investigations, arc faults are one of the main causes of electrical fires. When an arc fault occurs in a low-voltage system, the high temperature generated by arcing may ignite nearby flammable materials and lead to a fire. In this thesis, the arc fault experiment platform is used to test the power line which feeds the load of household appliances with different characteristics. The fast Fourier transform(FFT) is used to analyze the time domain and frequency domain characteristics of the line current, which are the basis for developing arc fault detection methods. For the arc fault detection, this thesis combines the discrete wavelet transform(DWT) and the accumulation high frequency energy method to obtain the characteristic vectors of the line current per power cycle. These vectors are applied to back propagation neural network(BPNN) and decision tree, repectively. Three detection methods are tested for the line under normal operation, series arc fault, and switching arc. The test results are compared with the commercial arc-fault circuit interrupter (AFCI). From the test results, it is showed that the detection accuracy is higher than the commercial AFCI. The decision tree is superior to the back propagation neural network in hardware consumption and detection accuracy. Finally, three detection methods are implemented on FPGA to verify the feasibility of the arc fault detection method proposed in hardware.

    摘要 I Abstract II 致謝 III 目錄 IV 圖目錄 VIII 表目錄 XII 第一章 緒論 1 1.1 研究背景與動機 1 1.2 文獻探討 2 1.3 研究內容 3 1.4 論文架構 4 第二章 串聯電弧故障特性與實驗設備 6 2.1 前言 6 2.2 電弧故障特性與種類 6 2.2.1 電弧 6 2.2.2 電弧故障種類 6 2.2.3 串聯電弧故障特性 7 2.3 人為電弧故障實驗設備 9 2.3.1 實驗平台 9 2.3.2 量測儀器 10 2.3.3 電弧故障斷路器 11 2.3.4 串聯電弧故障產生機台 13 2.4 線路電流量測與商用AFCI測試 14 2.4.1 正常運轉 15 2.4.2 串聯電弧故障 16 2.4.3 開關電弧 17 2.5 快速傅立葉分析串聯電弧故障之頻域特性 18 2.6 小結 22 第三章 串聯電弧故障電流檢測方法 23 3.1 前言 23 3.2 傅立葉轉換 23 3.2.1 離散傅立葉轉換 24 3.2.2 快速傅立葉轉換 25 3.3 小波轉換與高譜能量 26 3.3.1 離散小波轉換 26 3.3.2 小波多層解析與高頻能量累積法 27 3.4 倒傳遞類神經檢測法 31 3.4.1 類神經網路簡介 31 3.4.2 倒傳遞類神經網路訓練流程 31 3.5 決策樹檢測法 36 3.5.1 決策樹簡介 36 3.5.2 決策樹訓練流程 37 3.6 小結 41 第四章 使用FPGA進行電弧故障檢測 42 4.1 前言 42 4.2 FPGA簡介 42 4.3 FPGA硬體開發平台 44 4.4 FPGA設計流程 45 4.5 FPGA電弧故障檢測模組 48 4.5.1 鮑率產生器 50 4.5.2 UART接收控制模組 52 4.5.3 封包組合模組 52 4.5.4 檢測法測試模組 53 4.6 小結 56 第五章 串聯電弧故障檢測結果 57 5.1 前言 57 5.2 各負載條件測試 57 5.2.1 負載條件一:吹風機 58 5.2.2 負載條件二:17顆燈泡 62 5.2.3 負載條件三:吹風機與17顆燈泡 66 5.2.4 負載條件四:7顆燈泡與17顆燈泡 70 5.2.5 負載條件五:吹風機與電鍋 74 5.2.6 負載條件六:17顆燈泡與100μF電容 78 5.2.7 負載條件七:吹風機與100μF電容 82 5.2.8 負載條件八:混合負載 86 5.3 小結 90 第六章 結論與未來研究方向 91 6.1 結論 91 6.2 未來研究方向 92 參考文獻 93

    [1] 內政部消防署,「102-107年全國火災次數起火原因及火災損失統計表」,https://www.nfa.gov.tw/cht/index.php
    [2] National Fire Protection Association (NFPA), "Home Electrical Fires," https://www.nfpa.org/-/media/Files/News-and-Research/Fire-statistics-and-reports/US-Fire-Problem/Fire-causes/osHomeElectricalFires.pdf
    [3] Arc-Fault Circuit Interrupters, UL Standard 1699-2008, 2008.
    [4] K. Zeng, L. Xing, Y. Zhang, and L. Wang, "Characteristics analysis of AC arc fault in time and frequency domain," in 2017 Prognostics and System Health Management Conference (PHM-Harbin), pp. 1-5, 2017.
    [5] 曾元超,「防範住家電器火災的新技術」,台電月刊,第549期,第26-31頁,2008。
    [6] P. Muller, S. Tenbohlen, R. Maier, and M. Anheuser, "Characteristics of series and parallel low current arc faults in the time and frequency domain," in 2010 Proceedings of the 56th IEEE Holm Conference on Electrical Contacts, pp. 1-7, 2010.
    [7] C. Hong, C. Xiaojuan, X. Wei, and W. Cong, "Short-time fourier transform based analysis to characterization of series arc fault," in 2009 2nd International Conference on Power Electronics and Intelligent Transportation System (PEITS), vol. 1, pp. 185-188, 2009.
    [8] 趙尚程、張認成、杜建華、楊凱、潘冷,「採用小波變換的光伏串聯電弧故障檢測」,華僑大學學報,第38卷,第1期,第7-12頁,2017。
    [9] 劉曉明、趙洋、曹雲東、侯春光、王麗君,「基於小波變換的交流系統串聯電弧故障診斷」,電工技術學報,第29卷,第1期,第10-17頁,2014。
    [10] C.-H. Kim, H. Kim, Y.-H. Ko, S.-H. Byun, R. K. Aggarwal, and A. T. J. I. T. o. P. D. Johns, "A novel fault-detection technique of high-impedance arcing faults in transmission lines using the wavelet transform," vol. 17, no. 4, pp. 921-929, 2002.
    [11] R. Zhang and Z. Song, "Arc fault detection method based on signal energy distribution in frequency band," in 2012 Asia-Pacific Power and Energy Engineering Conference, pp. 1-4, 2012.
    [12] C. E. Restrepo, "Arc fault detection and discrimination methods," in Electrical Contacts-2007 Proceedings of the 53rd IEEE Holm Conference on Electrical Contacts, pp. 115-122, 2007.
    [13] C. Xiaochen, W. Li, S. Qiangang, and M. Zhen, "AC arc fault detection based on Mahalanobis Distance," in 2012 15th International Power Electronics and Motion Control Conference (EPE/PEMC), pp. DS3b. 13-1-DS3b. 13-6, 2012.
    [14] Y. Zhao, X. Zhang, Y. Dong, and W. Li, "Characteristics analysis and detection of AC arc fault in SSPC based on wavelet transform," in 2016 IEEE International Conference on Aircraft Utility Systems (AUS), pp. 476-481, 2016.
    [15] 王奕捷,「低壓線路串聯電弧故障時頻及頻域分析與檢測」,碩士學位論文,國立臺灣科技大學,臺北,2011。
    [16] 柯至遠,「低壓線路串聯電弧檢測之FPGA晶片設計」,碩士學位論文,國立臺灣科技大學,臺北,2014。
    [17] 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,2014。
    [18] 季桂樹、陳沛玲、宋航,「決策樹分類算法研究綜述」,科技廣場,第1卷,第9期,第12頁,2007。
    [19] Ms. G. Priyadarshini, M. Sc. and M. Phil, "Decision Tree Algorithms for Diagnosis of Cardiac Disease Treatment," IJCSMC, vol. 7, no. 7, pp. 138-144, 2018.
    [20] A. Albu, "From logical inference to decision trees in medical diagnosis," in 2017 E-Health and Bioengineering Conference (EHB), pp. 65-68, 2017.
    [21] G. D. Gregory and G. W. Scott, "The arc-fault circuit interrupter, an emerging product," in 1998 IEEE Industrial and Commercial Power Systems Technical Conference. Conference Record. Papers Presented at the 1998 Annual Meeting (Cat. No. 98CH36202), pp. 48-55, 1998.
    [22] G. D. Gregory, K. Wong, and R. F. J. I. T. o. i. a. Dvorak, "More about arc-fault circuit interrupters," vol. 40, no. 4, pp. 1006-1011, 2004.
    [23] K. J. Lippert and T. A. J. I. T. o. I. A. Domitrovich, "AFCIs—From a standards perspective," vol. 50, no. 2, pp. 1478-1482, 2013.
    [24] 劉鈺韋,「運用小波轉換與類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,2015。
    [25] 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,2012。
    [26] V. Apetrei, C. Filote, and C. Ciufudean, "Effective value calculation using wavelet transform," in Recent Researches in Electric Power and Energy Systems, Proceedings of the 13th International Conference on Electric Power Systems, High Voltages, Electric Machines, vol. 22, pp. 204-210, 2013.
    [27] S. Mallat, A Wavelet Tour of Signal Processing, Academic Press, San Diego, third edition, 2009.
    [28] 葉怡成,類神經網路模式應用與實作,「類神經網路應用與實作」,台北,儒林圖書公司,格致圖書公司,1997。
    [29] 翁政雄、洪令莊、呂培豪、陳學瀚、郭家佑、施博惟、謝孟哲,「應用決策樹於心臟病預測之研究」,第19屆資訊管理暨實務研討會,台中,2012。
    [30] J. R. Quinlan, C4. 5: programs for machine learning. Elsevier, 2014.
    [31] 高鴻文、林詩偉、萬書言,「運用決策樹演算法於護理人員離職預測」,醫療資訊雜誌,第29卷,第4期,第15-29頁,2012。
    [32] 林灶生、劉紹漢,「Verilog FPGA晶片設計」,全華圖書,臺北,2004。
    [33] 徐文波、田耘,「Xilinx FPGA開發實用手冊」,佳魁資訊,臺北,2015。
    [34] 鍾崇訓,「使用FPGA晶片設計之電壓閃爍計算」,碩士學位論文,國立台灣科技大學,臺北,2011。
    [35] 薛小剛、葛毅敏,「Xilinx ISE 9.X FPGACPLD設計指南」,人民郵電,2007。
    [36] 陸瑞強、廖玉評,「系統晶片設計使用QuartusII」,全華圖書,臺北,2008。

    無法下載圖示 全文公開日期 2024/06/26 (校內網路)
    全文公開日期 2024/06/26 (校外網路)
    全文公開日期 2024/06/26 (國家圖書館:臺灣博碩士論文系統)
    QR CODE