Research article Topical Sections

Low-complexity energy disaggregation using appliance load modelling

  • Received: 26 October 2015 Accepted: 23 December 2015 Published: 05 January 2016
  • Large-scale smart metering deployments and energy saving targets across the world have ignited renewed interest in residential non-intrusive appliance load monitoring (NALM), that is, disaggregating total household’s energy consumption down to individual appliances, using purely analytical tools. Despite increased research efforts, NALM techniques that can disaggregate power loads at low sampling rates are still not accurate and/or practical enough, requiring substantial customer input and long training periods. In this paper, we address these challenges via a practical low-complexity lowrate NALM, by proposing two approaches based on a combination of the following machine learning techniques: k-means clustering and Support Vector Machine, exploiting their strengths and addressing their individual weaknesses. The first proposed supervised approach is a low-complexity method that requires very short training period and is fairly accurate even in the presence of labelling errors. The second approach relies on a database of appliance signatures that we designed using publicly available datasets. The database compactly represents over 200 appliances using statistical modelling of measured active power. Experimental results on three datasets from US, Italy, Austria and UK, demonstrate the reliability and practicality.

    Citation: Hana Altrabalsi, Vladimir Stankovic, Jing Liao, Lina Stankovic. Low-complexity energy disaggregation using appliance load modelling[J]. AIMS Energy, 2016, 4(1): 1-21. doi: 10.3934/energy.2016.1.1

    Related Papers:

  • Large-scale smart metering deployments and energy saving targets across the world have ignited renewed interest in residential non-intrusive appliance load monitoring (NALM), that is, disaggregating total household’s energy consumption down to individual appliances, using purely analytical tools. Despite increased research efforts, NALM techniques that can disaggregate power loads at low sampling rates are still not accurate and/or practical enough, requiring substantial customer input and long training periods. In this paper, we address these challenges via a practical low-complexity lowrate NALM, by proposing two approaches based on a combination of the following machine learning techniques: k-means clustering and Support Vector Machine, exploiting their strengths and addressing their individual weaknesses. The first proposed supervised approach is a low-complexity method that requires very short training period and is fairly accurate even in the presence of labelling errors. The second approach relies on a database of appliance signatures that we designed using publicly available datasets. The database compactly represents over 200 appliances using statistical modelling of measured active power. Experimental results on three datasets from US, Italy, Austria and UK, demonstrate the reliability and practicality.


    加载中
    [1] Smart metering equipment technical specifications: Second version: Part 2. Department of Energy & Climate Change UK, Dec. 2013.
    [2] Armel KC, Gupta A, Shrimali G, et al. (2013) Is disaggregation the holy grail of energy efficiency? The case of electricity. Energy Policy 52: 213–234. doi: 10.1016/j.enpol.2012.08.062
    [3] Hart G, Nonintrusive Appliance Load Data Acquisition Method, MIT Energy Laboratory Technical Report, Sept. 1984.
    [4] Zeifman M, Roth K (2011) Nonintrusive appliance load monitoring: Review and outlook. IEEE Trans Consumer Electronics 57: 76–84. doi: 10.1109/TCE.2011.5735484
    [5] Zoha A, Gluhak A, Imran MA, et al. (2012) Non-intrusive load monitoring approaches for disaggregated energy sensing: A survey. Sensors 12: 16838–16866. doi: 10.3390/s121216838
    [6] Perez KX, Cole WJ, Baldea M, et al. (2014) Meters to models: Using smart meter data to predict home energy use. in Process. ACEEE Summer Study on Energy Efficiency in Buildings, Pacific Grove, CA.
    [7] Murray D, Liao J, Stankovic L, et al., A data management platform for personalised real-time energy feedback. in Proc EEDAL-2015 8th Int Conf Energy Efficiency in Domestic Appliances and Lighting, Lucerne-Horw, Switzerland, Aug. 2015.
    [8] Liao J, Elafoudi G, Stankovic L, et al. Power disaggregation for low-sampling rate data. 2nd Int. Non-intrusive Appliance Load Monitoring Workshop, Austin, TX, June 2014.
    [9] Marchiori A, Hakkarinen D, Han Q, et al. (2011) Circuit-level load monitoring for household energy management, IEEE Pervas Comput 10: 40-48.
    [10] Berges M, Goldman E, Matthews HS, et al. (2011) User-centered non-intrusive electricity load monitoring for residential buildings. J Comput Civil Eng 25: 471-480. doi: 10.1061/(ASCE)CP.1943-5487.0000108
    [11] Kim H, Marwah M, Arlitt M, et al., Unsupervised disaggregation of low frequency power measurements, in Proc 11th SIAM Int Conf Data Mining, Mesa, AZ, April 2011.
    [12] Parson O, Ghosh S, Weal M, et al. (2012) Non-intrusive load monitoring using prior models of general appliance types. in Proc. the 26th Conf. Artificial Intelligence (AAAI-12), Toronto, CA, pp. 356–362.
    [13] Kolter J, Jaakkola T (2012) Approximate inference in additive factorial HMMs with application to energy disaggregation. in J Machine Learning 22: 1472–1482.
    [14] Johnson MJ, Willsky AS (2013) Bayesian nonparametric Hidden Semi-Markov Models. J Machine Learning Research 14: 673–701.
    [15] Kolter J, Batra S, Ng AY, Energy Disaggregation via Discriminative Sparse Coding. in Proc Advances in Neural Inform Processing Sys 23 (NIPS 2010).
    [16] Shao H, Marwah M, Ramakrishnan NA, Temporal motif mining approach to unsupervised energy disaggregation. in Proc. the 1st Int Workshop Non-Intrusive Load Monitoring, Pittsburgh, PA, May 2012.
    [17] Elafoudi G, Stankovic L, Stankovic V, Power disaggregation of domestic smart meter readings using Dynamic Time Warping. ISCCSP-2014 IEEE Intl Symp Communications, Control, and Signal Processing, Athens, Greece, May 2014.
    [18] Altrabalsi H, Liao J, Stankovic L, et al., A low-complexity energy disaggregation method: Performance and robustness. SSCI-2014 IEEE Symp Comput Intelligence Applications in Smart Grid, Orlando, FL, Dec. 2014.
    [19] Xia XL, Lyu MR, Lok LM,et al., Methods of decreasing the number of support vectors via k-mean clustering. in Proc ICIC 2005, LNCS 3644, pp. 717–726, Spinger-Verlag Berlin Heidelberg, 2005.
    [20] Wang j, Wu x, Zhang C (2005) Support vector machines based on K-means clustering for realtime business intelligence systems. Int J Business Intelligence and Data Mining 1: 54–64.
    [21] Yao Y, Liu Y, Yu Y, et al. (2013) K-SVM: An effective SVM algorithm based on k-means clustering. J Computers 8: 2632–2639.
    [22] Gu Q, Jan J, Clustered Support Vector Machines. in Proc AISTATS-2013 16th Int Conf Artificial Intelligence and Statistics, Scottsdale, AZ, 2013.
    [23] Kolter J, Johnson M. REDD: A public data set for energy disaggregation research. in Workshop on Data Mining Applications in Sustainability (SIGKDD), San Diego, CA, 2011.
    [24] Monacchi A, Egarter D, Elmenreich W, et al. GREEND: An Energy Consumption Dataset of Households in Italy and Austria. in Proc IEEE SmartGridComm, Venice, Italy, Nov. 2014.
    [25] Gao J, Giri S, Kara EC, et al. ( 2014) PLAID: a public dataset of high-resolution electrical appliance measurements for load identification research: demo abstract. in Proc the 1st ACM Conference on Embedded Systems for Energy-Efficient Buildings, 198-199.
    [26] Ruzzelli AG, Nicolas C, Schoofs A, et al. (2010) Real-time recognition and profiling of appliances through a single electricity sensor. in Proc IEEE SECON-2010 7th Annual Conf Sensor Mesh and Ad Hoc Communications and Networks, 1–9.
    [27] Laughman C, Lee K,Cox R, et al. (2003) Power signature analysis. IEEE Power and Energy Magazine 1: 56–63.
    [28] Liang J, Ng SKK, Kendall G, et al. (2010) Load signature study part I: Basic concept, structure, and methodology. IEEE Trans Power Delivery 25: 551–560.
    [29] Berges M, Goldman E, Matthews HS, et al., Learning systems for electric consumption of buildings. in Proc 2009 ASCE Int Workshop Computing in Civil Engineering, Austin, TX, 2009.
    [30] Barker s, Kalra s, Irwin D, et al., NILM redux: The case for emphasizing applications over accuracy. NILM-2014 Workshop, Austin, TX, June 2014.
    [31] Markonin S, Bajic IV, Popowich F, Efficient sparse metric processing for nonintrusive load monitoring. 2nd Int Non-intrusive Appliance Load Monitoring Workshop, Austin, TX, June 2014.
    [32] Al-Harbi SH, Rayward-Smith VJ (2006) Adapting k-means for supervised clustering. Appl Intell 24: 219–226. doi: 10.1007/s10489-006-8513-8
  • Reader Comments
  • © 2016 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(7270) PDF downloads(1453) Cited by(43)

Article outline

Figures and Tables

Figures(8)  /  Tables(8)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog