全球暖化和資源耗竭是我們現正面臨的兩大難題，隨著問題日趨嚴重，人們對於節能和環保的議題也越來越重視。但是大部分的用戶並不清楚用電的狀況，在這樣的情況下，很容易會錯估節能的方向和效果。然而，研究顯示，持續給用戶用電反饋可以達到一定比例的節電效果。因此，用電分解開始變得越來越重要，除了可以幫助用戶了解用電分布，進而也能達到節能的功效。

在這篇論文中，我們提出了一個新的解決方法，透過群聚迴歸模型實現用電分解，利用電器之間的相關程度將電器分群並建模估計，這個方法是合理而且有效的，因為在很多時候某些用電行為可能形成一種習慣亦或是本來就是一連串的操作程序，所以用電之間會有一定程度的相關。在我們的實驗結果中，我們的排序值 (ranking value) 最少都有達到80%以上。另外，原本需要裝設14個電錶的商店，在使用我們的群聚迴歸模型之後，只需要裝設3個電錶，就可以計算出一個和實際相差不遠的結果。

Global warming and the depletion of natural resources such as oil, coal are some of the most difficult problems we have ever faced. In recent years, people have begun paying more attention to energy saving and environmental issues. Most people do not know what percentage of power consumption was cost in what usage so they would misestimate the direction and the effect of conservation. However, the study shows that continuous feedback to the consumers can reduce energy usage by 10-15\% on average. Therefore, energy disaggregation is more and more important to help consumers understand their electricity consumption distribution and let them know the expenses coming from which appliances. After understanding the distribution, consumers can have their plan on how to achieve power saving.

In this thesis, we propose a novel framework using the relationships of appliances with each other to accomplish energy disaggregation. It is reasonable and effective because the using behavior of appliances is regular or even becomes a fixed series of procedures sometimes. In our results, we are not only arriving at 80\% ranking value but also reducing the number of smart meters from 14 to 3.

[1] Feng Chen, Jing Dai, Bingsheng Wang, Sambit Sahu, Milind Naphade, and Chang-
Tien Lu. Activity Analysis based on Low Sample Rate Smart Meters. In Proceedings
of the 17th ACM SIGKDD International Conference on Knowledge Discovery and
Data mining, pages 240-248. ACM, 2011.
[2] Harris Drucker, Chris J. Burges, Linda Kaufman, Alex Smola, and Vladimir Vapnik.
Support Vector Regression Machines. Advances in Neural Information Processing
Systems, pages 155-161, 1997.
[3] John Fox. Applied Regression Analysis, Linear Models, and Related Methods. SAGE
Publications, Incorporated, 1997.
[4] Jon Froehlich. Promoting Energy Efficient Behaviors in the Home through Feedback:The Role of Human-Computer Interaction. In Proc. HCIC Workshop, volume 9, 2009.
[5] Jon Froehlich, Eric Larson, Sidhant Gupta, Gabe Cohn, Matthew Reynolds, and
Shwetak Patel. Disaggregated End-Use Energy Sensing for the Smart Grid. Pervasive
Computing, IEEE, 10(1):28-39, 2011.
[6] Sidhant Gupta, Matthew S. Reynolds, and Shwetak N. Patel. ElectriSense: Single-
Point Sensing using EMI for Electrical Event Detection and Classification in the
Home. In Proceedings of the 12th ACM International Conference on Ubiquitous
Computing, pages 139-148. ACM, 2010.
[7] George William Hart. Nonintrusive Appliance Load Monitoring. Proceedings of the
IEEE, 80(12):1870-1891, 1992.
[8] Helen Ai He and Saul Greenberg. Motivating Sustainable Energy Consumption in
the Home. In Defining the Role of HCI in the Challenges of Sustainability Workshop,
2009.
[9] Chris L. Holcomb. Disaggregation of Residential Electric Loads using Smart Metered Data. Thesis, The University of Texas at Austin, 2011.
[10] Daniel A. Kelly. Disaggregating Smart Meter Readings using Device Signatures.
Thesis, Computing Science of Imperial College London, 2011.
[11] Hyungsul Kim, Manish Marwah, Martin Arlitt, Geo Lyon, and Jiawei Han. Unsupervised Disaggregation of Low Frequency Power Measurements. In Proceedings of the SIAM Conference on Data Mining, pages 747-758, 2011.
[12] J. Zico Kolter, Siddarth Batra, and Andrew Y. Ng. Energy Disaggregation via
Discriminative Sparse Coding. In Neural Information Processing Systems, pages
1153-1161, 2010.
[13] J. Zico Kolter and Matthew J. Johnson. REDD: A Public Data Set for Energy
Disaggregation Research. In Proceedings of the SustKDD Workshop on Data Mining
Applications in Sustainability, pages 1-6, 2011.
[14] Christopher Laughman, Kwangduk Lee, Robert Cox, Steven Shaw, Steven Leeb,
Les Norford, and Peter Armstrong. Power Signature Analysis. Power and Energy
Magazine, IEEE, 1(2):56-63, 2003.
[15] W.K. Lee, G.S.K. Fung, H.Y. Lam, F.H.Y. Chan, and Mark Lucente. Exploration
on Load Signatures. In International Conference on Electrical Engineering (ICEE),
pages 1-5, 2004.
[16] Yuh-Jye Lee, Wen-Feng Hsieh, and Chien-Ming Huang. "ε-SSVR: A Smooth Support Vector Machine for "ε-Insensitive Regression. IEEE Transactions on Knowledge and Data Engineering, pages 678-685, 2005.
[17] Yuh-Jye Lee and Olvi L. Mangasarian. SSVM: A smooth support vector machine for classi_cation. Computational optimization and Applications, 20(1):5-22, 2001.
[18] Steven B. Leeb, Steven R. Shaw, and James L. Kirtley Jr. Transient Event Detection in Spectral Envelope Estimates for Nonintrusive Load Monitoring. Power Delivery, IEEE Transactions on, 10(3):1200-1210, 1995.
[19] Victor E. McZgee and Willard T. Carleton. Piecewise Regression. Journal of the
American Statistical Association, 65(331):1109-1124, 1970.
[20] Hannu Pihala. Non-Intrusive Appliance Load Monitoring System based on a Modern kWh-Meter. Technical Research Centre of Finland, 1998.
[21] Bernhard Scholkopf and Alexander J. Smola. Learning with Kernels. The MIT Press, 2002.
[22] Taipower. Power generation ratio of various source in calendar year. http://www.
taipower.com.tw/content/new_info/new_info-c37.aspx?LinkID=13, 2013. Online;
accessed 2013.
[23] Bingsheng Wang, Haili Dong, Arnold Boedihardjo, Feng Chen, and Chang-Tien Lu. A Hierarchical Probabilistic Model for Low Sample Rate Home-Use Energy Disaggregation. In SIAM, pages 704-712. SIAM, 2013.
[24] Michael Zeifman. Disaggregation of Home Energy Display Data using Probabilistic Approach. Consumer Electronics, IEEE Transactions on, 58(1):23-31, 2012.
[25] Michael Zeifman and Kurt Roth. Nonintrusive Appliance Load Monitoring: Review and Outlook. Consumer Electronics, IEEE Transactions on, 57(1):76-84, 2011.