進階搜尋


下載電子全文  
系統識別號 U0026-1610201809053600
論文名稱(中文) 具電器辨識功能之輕量化電源資訊監控模組
論文名稱(英文) A Lightweight Power Monitoring Module for Appliance Recognition
校院名稱 成功大學
系所名稱(中) 工程科學系碩士在職專班
系所名稱(英) Department of Engineering Science (on the job class)
學年度 107
學期 1
出版年 108
研究生(中文) 柯佑霖
研究生(英文) Yu-Lin Ke
學號 N97041081
學位類別 碩士
語文別 中文
論文頁數 54頁
口試委員 指導教授-鄧維光
口試委員-胡誌麟
口試委員-侯廷偉
口試委員-王明習
中文關鍵字 能源管理系統  電器辨識  遠端控制  MQTT 
英文關鍵字 energy management system  appliance recognition  remote control  MQTT 
學科別分類
中文摘要 因科技的發達而開發出許多新產品,為生活帶來更多便利性,但也面臨能源需求增加的情況,以電力能源來看,增加發電設施是一種方法,或是透過節約能源降低對環境的負擔。一般使用者可以透過能源管理系統監控日常能源損耗,降低電力成本的支出。本研究設計出體積小、低成本和低網路頻寬需求的輕量化電源資訊採集模組,可直接安裝於電源插座內部。並使用MQTT作為通訊協定,降低頻寬需求。結合資料分析,達到辨識電器種類的功能。本論文比較SVM與隨機決策森林兩套電器辨識模型,辨識準確度皆超過90%。此系統可自動統計不同電器的用電資訊,分析家庭電器的使用狀況,透過行動裝置清楚地掌握家庭用電情況。結合物聯網的概念,不僅可以監控電源資訊,還可遠端控制插座,關閉待機狀態的電器,提升用電安全。
英文摘要 The effective energy management system can reduce energy waste, especially electricity energy management is a key part of the future. In this thesis, design a lightweight power monitoring module with small size, low cost and low network bandwidth requirements, which can be directly installed inside the power socket box. And apply the MQTT protocol to reduce bandwidth requirements. Combine with the cloud data analysis to identify different types of appliances. This thesis compares two sets of electrical identification models between SVM and random decision forest, and the identification accuracy is high more 90%. This system can automatically count the electricity consumption information of different electrical appliances, analyze the using status of household electrical appliances, and clearly grasp the household electricity consumption through mobile devices. Combined with the concept of the Internet of Things, not only monitor power information, but also remote control sockets, turn off appliances on standby, and improve power security.
論文目次 摘要 I
Extended Abstract II
誌謝 VIII
目錄 IX
表目錄 XI
圖目錄 XII
第一章 緒論 1
1-1 研究背景與動機 1
1-2 研究目的與貢獻 2
第二章 文獻探討 4
2-1 家庭能源管理系統簡介 4
2-1-1 非侵入式負載監測 6
2-1-2 侵入式負載監測 6
2-2 電源資訊採集設備比較 8
2-3 電源感測器簡介 9
2-3-1 採樣電阻 10
2-3-2 電流互感器 10
2-3-3 霍爾元件 11
2-4 微控制器簡介 13
2-5 IoT簡介 14
2-5-1 感測器傳輸技術比較 15
2-5-2 網路層簡介 16
2-5-3 應用層 16
第三章 研究方法與系統規劃 18
3-1 系統架構與流程圖 18
3-2 電源資訊採集模組架構 20
3-2-1 電源監控IC 21
3-2-2 硬體通訊協定 22
3-2-3 微控制器 26
3-3 網路通訊傳輸 27
3-4 電器辨識方法 30
3-4-1 支援向量機 32
3-4-2 隨機森林 33
第四章 研究結果與分析 35
4-1 設計與實作電源資訊採集模組 35
4-1-1 控制器運作邏輯 37
4-1-2 電源採集流程 38
4-2 MQTT無線通訊傳輸 39
4-3 電器辨識 40
4-3-1 支援向量機 46
4-3-2 隨機決策森林 47
4-4 遠端控制 48
第五章 結論與展望 50
參考文獻 51

參考文獻 [1] "EIA residential report". 2017; Available from: https://www.eia.gov/consumption/residential/index.php.
[2] Meyers, R.J., et al., Scoping the potential of monitoring and control technologies to reduce energy use in homes. 2010. 42(5): p. 563-569.
[3] Ahmadi-Karvigh, S., et al., Real-time activity recognition for energy efficiency in buildings. 2018. 211: p. 146-160.
[4] Basu, K., et al., Nonintrusive load monitoring: A temporal multilabel classification approach. 2015. 11(1): p. 262-270.
[5] Gillis, J.M., S.M. Alshareef, and W.G.J.I.T.o.S.G. Morsi, Nonintrusive load monitoring using wavelet design and machine learning. 2016. 7(1): p. 320-328.
[6] Fernández-Caramés, T.M.J.I.J.o.D.S.N., An intelligent power outlet system for the smart home of the Internet of Things. 2015. 11(11): p. 214805.
[7] Message Queue Telemetry Transport (MQTT). Available from: http://mqtt.org/.
[8] Vega, A., et al., Modeling for home electric energy management: A review. 2015. 52: p. 948-959.
[9] Agyeman, K.A., S. Han, and S.J.E. Han, Real-time recognition non-intrusive electrical appliance monitoring algorithm for a residential building energy management system. 2015. 8(9): p. 9029-9048.
[10] Serra, H., et al. Domestic power consumption measurement and automatic home appliance detection. in Intelligent Signal Processing, 2005 IEEE International Workshop on. 2005. IEEE.
[11] Najmeddine, H., et al. State of art on load monitoring methods. in Power and Energy Conference, 2008. PECon 2008. IEEE 2nd International. 2008. IEEE.
[12] Lee, S.-c., et al. Appliance Recognition and Unattended Appliance Detection for Energy Conservation. in Plan, Activity, and Intent Recognition. 2010.
[13] Meehan, P., et al. Temporal and frequency analysis of power signatures for common household appliances. in ICT and Energy Efficiency and Workshop on Information Theory and Security (CIICT 2012), Symposium on. 2012. IET.
[14] Patel, S.N., et al. At the flick of a switch: Detecting and classifying unique electrical events on the residential power line (nominated for the best paper award). in International Conference on Ubiquitous Computing. 2007. Springer.
[15] Zoha, A., et al., Non-intrusive load monitoring approaches for disaggregated energy sensing: a survey. Sensors (Basel), 2012. 12(12): p. 16838-66.
[16] Abeykoon, V., et al. Real Time Identification of Electrical Devices through Power Consumption Pattern Detection. in Proceedings of the International Conference on Micro and Nano Technologies, Modelling and Simulation, Kuala Lumpur, Malaysia. 2016.
[17] Ruzzelli, A.G., et al. Real-time recognition and profiling of appliances through a single electricity sensor. in Sensor Mesh and Ad Hoc Communications and Networks (SECON), 2010 7th Annual IEEE Communications Society Conference on. 2010. IEEE.
[18] De Baets, L., et al., Appliance classification using VI trajectories and convolutional neural networks. 2018. 158: p. 32-36.
[19] Hoogsteen, G., et al. Non-intrusive appliance recognition. in Innovative Smart Grid Technologies (ISGT Europe), 2012 3rd IEEE PES International Conference and Exhibition on. 2012. IEEE.
[20] Liang, J., et al., Load signature study—Part I: Basic concept, structure, and methodology. 2010. 25(2): p. 551-560.
[21] Ridi, A., C. Gisler, and J. Hennebert. A survey on intrusive load monitoring for appliance recognition. in Pattern Recognition (ICPR), 2014 22nd International Conference on. 2014. IEEE.
[22] Blanco-Novoa, Ó., et al., An Electricity Price-Aware Open-Source Smart Socket for the Internet of Energy. 2017. 17(3): p. 643.
[23] Weiss, M., et al. Leveraging smart meter data to recognize home appliances. in Pervasive Computing and Communications (PerCom), 2012 IEEE International Conference on. 2012. IEEE.
[24] Jiang, L., S. Luo, and J. Li. An approach of household power appliance monitoring based on machine learning. in Intelligent Computation Technology and Automation (ICICTA), 2012 Fifth International Conference on. 2012. IEEE.
[25] Lukaszewski, R., K. Liszewski, and W. Winiecki. Methods of electrical appliances identification in systems monitoring electrical energy consumption. in Intelligent Data Acquisition and Advanced Computing Systems (IDAACS), 2013 IEEE 7th International Conference on. 2013. IEEE.
[26] Pezeshkfar, O., Low cost framework for non-intrusive load monitoring (NILM) to monitor human behavioral pattern. 2013, University of Missouri--Columbia.
[27] Su, S., et al. Home energy management system using smart socket with appliance recognition. in Information and Automation (ICIA), 2016 IEEE International Conference on. 2016. IEEE.
[28] Tsai, K.-L., F.-Y. Leu, and I.J.I.A. You, Residence energy control system based on wireless smart socket and IoT. 2016. 4: p. 2885-2894.
[29] Cortes, C. and V.J.M.l. Vapnik, Support-vector networks. 1995. 20(3): p. 273-297.
[30] Breiman, L.J.M.l., Random forests. 2001. 45(1): p. 5-32.

論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2019-02-18起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2019-02-18起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw