於本篇論文中，我們為了節能減碳，在智能電表下探討有關家用能源管理系統的多電器辨識系統。為了能夠辨識出電器目前的狀態，利用動態時間校正的波形辨識演算法來辨識電器，並且為了加快辨識速度，將完整的波形分開分別去做動態時間校正的波形辨識演算法。本論文對於五種電器進行識別，其識別率達到73.72%，而將完整波形分四部份分別做動態時間校正也有70.47%，辨識速度也快一半以上，用在電器辨識系統有一定的可行性。

In order to conserve energy to reduce carbon emission, we probe into multi-appliance recognition system concerning home energy management system based on smart meter. We identify appliance by using dynamic time warping to find out the state of appliance. To increase the speed of recognition, we divide waveform into four parts and use dynamic time warping to find out the state of appliance. In this thesis, we identify five appliances by waveform identification algorithm. The recognition rate can reach 73.72%. And we divide waveform into four parts and use dynamic time warping to find out the state of appliance. The recognition rate can also reach 70.47%, and it also consume 50% of time. So this method for appliance recognition system is plausible.

論文審定書.................................................................................................... i
致謝............................................................................................................... ii
摘要............................................................................................................. iii
Abstract ........................................................................................................ iv
目錄............................................................................................................... v
圖次.............................................................................................................. vi
表次............................................................................................................. vii
第一章 緒論.................................................................................................................. 1
1.1 研究背景與動機........................................................................................ 1
1.1.1 能源與環境議題............................................................................ 1
1.2 研究架構.................................................................................................... 6
第二章 相關背景.......................................................................................................... 7
2.1 背景............................................................................................................ 7
2.1.1 智能電網........................................................................................ 7
2.1.2 智能電表...................................................................................... 10
2.2 波形識別演算法...................................................................................... 15
2.2.1 歐幾里得距離.............................................................................. 15
2.2.2 動態時間校正.............................................................................. 16
第三章 多電器識別.................................................................................................... 20
3.1 波形辨識系統.......................................................................................... 20
3.2 模擬環境.................................................................................................. 22
第四章 模擬結果與分析............................................................................................ 25
第五章 結論................................................................................................................ 38
參考文獻...................................................................................................................... 39

[1] 裝置容量與發電量佔比，台灣電力公司。[Online]. Available:
http://www.taipower.com.tw/content/new_info/new_info-c34.aspx?LinkID=13 [2] 國際能源總署(International Energy Agency), Key World Energy Statistics, 2013 edition, pp.48–57. [Online]. Available: http://www.iea.org/publications/freepublications/publication/KeyWorld2013_FINAL_WEB.pdf
[3] solar works. [Online]. Available: http://www.solarworksca.com/
[4] 潘晴財、朱家齊，全球智慧電網產業環境 (政策、市場、產業與技術) 趨勢整體分析，清華大學電機工程學系，2012年8月
[5] Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC.Official Journal of the European Union, pp. 91, Aug. 2009.[Online]. Available:http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=OJ:L:2009:211:FULL&from=EN
[6] 台灣智慧電網發展願景，台灣智慧型電網產業協會。[Online]. Available:
http://www.smart-grid.org.tw/content/smart_grid/development.aspx
[7] 游宜樺，十年內 英全面換裝智慧電表，經濟日報，A8版, 2009年5月22日。
[8] 張蕎韻，全球智慧電表市場趨勢分析，工業技術研究院，2009年。 [9] 提升電網可靠度及發輸電效率，經濟部。[Online]. Available: http://www.moea.gov.tw/AD/Ad04/content/ContentDetail.aspx?menu_id=4598
[10] 黃怡碩，智慧電表基礎建設之簡介，工業技術研究院綠能所，2011年12月。 [11] 能源產業技術白皮書，經濟部能源局，2012年。[Online]. Available: http://web3.moeaboe.gov.tw/ECW/populace/content/SubMenu.aspx?menu_id=62
40
[12] 陳瑋哲、裴鴻達、賴益祺、黃悅民，應用輕量化波形識別方法於平行多電器辨識系統，2012年10月。
[13] J. S. Jang, "Data Clustering and Pattern Recognition," 8-4 Dynamic Time Warping.[Online].Available:http://mirlab.org/jang/books/dcpr/dpDtw.asp?title=8-4 Dynamic Time Warping [14] S. Salvador and P. Chan, “FastDTW: Toward Accurate Dynamic Time Warping in Linear Time and Space”, KDD Workshop on Mining Temporal and Sequential Data, pp. 70–80, 2004.