Advertisement

Machine learning approaches for non-intrusive load monitoring: from qualitative to quantitative comparation

  • Christoforos Nalmpantis
  • Dimitris Vrakas
Article

Abstract

Non-intrusive load monitoring (NILM) is the prevailing method used to monitor the energy profile of a domestic building and disaggregate the total power consumption into consumption signals by appliance. Whilst the most popular disaggregation algorithms are based on Hidden Markov Model solutions based on deep neural networks have attracted interest from researchers. The objective of this paper is to provide a comprehensive overview of the NILM method and present a comparative review of modern approaches. In this effort, many obstacles are identified. The plethora of metrics, the variety of datasets and the diversity of methodologies make an objective comparison almost impossible. An extensive analysis is made in order to scrutinize these problems. Possible solutions and improvements are suggested, while future research directions are discussed.

Keywords

Non-intrusive load monitoring (NILM) Power disaggregation algorithms Hidden Markov Model Deep learning 

References

  1. Aiad M, Lee PH (2016) Unsupervised approach for load disaggregation with devices interactions. Energy Build 116:96–103.  https://doi.org/10.1016/j.enbuild.2015.12.043CrossRefGoogle Scholar
  2. Anderson K, Ocneanu AF, Benitez D, Carlson D, Rowe A, Bergés M (2012) BLUED? A fully labeled public dataset for event-based non-intrusive load monitoring research. In: Proceedings of the 2nd KDD workshop on data mining applications in sustainability (SustKDD), Oct 2011, pp 1–5Google Scholar
  3. Armel CK, Gupta A, Shrimali G, Albert A (2013) Is disaggregation the holy grail of energy efficiency? The case of electricity. Energy Policy 52:213–234.  https://doi.org/10.1016/j.enpol.2012.08.062CrossRefGoogle Scholar
  4. Baranski M, Voss J (2003) Non-intrusive appliance load monitoring based on an optical sensor. In: 2003 IEEE Bologna power tech conference proceedings, vol 4, pp 267–274. https://doi.org/10.1109/PTC.2003.1304732
  5. Batra N, Kelly J, Parson O, Dutta H, Knottenbelt W, Rogers A, Srivastava M (2014) NILMTK? An open source toolkit for non-intrusive load monitoring categories and subject descriptors. In: International conference on future energy systems (ACM E-Energy), pp 1–4. https://doi.org/10.1145/2602044.2602051
  6. Batra N, Parson O, Berges M, Singh A, Rogers A (2014) A comparison of non-intrusive load monitoring methods for commercial and residential buildings. Retrieved from arXiv:1408.6595 [Cs]
  7. Beckel C, Kleiminger W, Staake T, Santini S (2014) The ECO data set and the performance of non-intrusive load monitoring algorithms. In: Proceedings of the 1st ACM conference on embedded systems for energy-efficient buildings, pp 80–89. https://doi.org/10.1145/2674061.2674064
  8. Bonfigli R, Squartini S, Fagiani M, Piazza F (2015) Unsupervised algorithms for non-intrusive load monitoring: an up-to-date overview. In: 2015 IEEE 15th international conference on environment and electrical engineering, EEEIC 2015—conference proceedings, pp 1175–1180. https://doi.org/10.1109/EEEIC.2015.7165334
  9. Chang HH, Chien PC, Lin LS, Chen N (2011) Feature extraction of non-intrusive load-monitoring system using genetic algorithm in smart meters. In: Proceedings—2011 8th IEEE international conference on e-business engineering, ICEBE 2011, pp 299–304. https://doi.org/10.1109/ICEBE.2011.48
  10. Department of Energy and Climate Change (2008) Climate change act, Tech rep, UK Retrieved from http://www.legislation.gov.uk/ukpga/2008/27/pdfs/ukpga_20080027_en.pdf
  11. Darby S (2006) The effectiveness of feedback on energy consumption a review for defra of the literature on metering, billing and direct displays. Retrieved from http://www.eci.ox.ac.uk/research/energy/downloads/smart-metering-report.pdf
  12. Dong R, Ratliff L, Ohlsson H, Sastry SS (2013) Fundamental limits of nonintrusive load monitoring. In: 3rd ACM international conference on high confidence networked systems (HiCoNS), pp 11–18. https://doi.org/10.1145/2566468.2576849
  13. Egarter D, Bhuvana VP, Elmenreich W (2015) PALDi: online load disaggregation via particle filtering. IEEE Trans Instrum Meas 64(2):467–477.  https://doi.org/10.1109/TIM.2014.2344373CrossRefGoogle Scholar
  14. Egarter D, Pöchacker M, Elmenreich W (2015) Complexity of power draws for load disaggregation 1–26. https://arxiv.org/abs/1501.02954Google Scholar
  15. Eia (2017) Annual Energy Outlook 2017 with projections to 2050. Retrieved from https://www.eia.gov/outlooks/aeo/pdf/0383(2017).pdf
  16. Froehlich J, Larson E, Gupta S, Cohn G, Reynolds M, Patel S (2011) Disaggregated end-use energy sensing for the smart grid. IEEE Perv Comput 10(1):28–39.  https://doi.org/10.1109/MPRV.2010.74CrossRefGoogle Scholar
  17. Greveler U, Justus B, Loehr D (2012) Forensic content detection through power consumption. In: IEEE international conference on communications, pp 6759–6763. https://doi.org/10.1109/ICC.2012.6364822
  18. Hart GW (1992) Nonintrusive appliance load monitoring. Proc IEEE 80(12):1870–1891CrossRefGoogle Scholar
  19. Inagaki S, Egami T, Suzuki T, Nakamura H, Ito K (2011) Nonintrusive appliance load monitoring based on integer programming. Electr Eng Jpn (English Translation of Denki Gakkai Ronbunshi) 174(2):1386–1392.  https://doi.org/10.1002/eej.21040Google Scholar
  20. Johnson MJ, Willsky AS (2013) Bayesian nonparametric hidden semi-Markov models. 14:673–701. arXiv Preprint Retrieved fromarxiv:1203.1365
  21. Kamat PV (2007) Meeting the clean energy demand: nanostructure architectures for solar energy conversion. https://doi.org/10.1021/jp066952u
  22. Kato T, Cho HS, Lee D, Toyomura T, Yamazaki T (2009) Appliance recognition from electric current signals for information-energy integrated network in home environments. In: Lecture notes in computer science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol 5597, LNCS, pp 150–157. https://doi.org/10.1007/978-3-642-02868-7_19
  23. Katos V, Vrakas D, Katsaros P (2011) A framework for access control with inference constraints. In: Proceedings—international computer software and applications conference, pp 289–297. https://doi.org/10.1109/COMPSAC.2011.45
  24. Kelly J, Knottenbelt W (2015a) Neural NILM: deep neural networks applied to energy disaggregation. In: Proceedings of the 2nd ACM international conference on embedded systems for energy-efficient built environments. ACM, pp 55–64Google Scholar
  25. Kelly J, Knottenbelt W (2015b) The UK-DALE dataset, domestic appliance-level electricity demand and whole-house demand from five UK homes. Sci Data 2:150007.  https://doi.org/10.1038/sdata.2015.7CrossRefGoogle Scholar
  26. Kim H, Marwah M, Arlitt MF, Lyon G, Han J (2011) Unsupervised disaggregation of low frequency power measurements. In: Proceedings of the 11th SIAM international conference on data mining, pp 747–758. https://doi.org/10.1137/1.9781611972818.64
  27. Kolter JZ, Johnson MJ (2011) REDD? A public data set for energy disaggregation research. SustKDD Workshop, no (1), pp 1–6. Retrieved from http://users.cis.fiu.edu/~lzhen001/activities/KDD2011Program/workshops/WKS10/doc/SustKDD3.pdf
  28. Kolter Z, Jaakkola T (2012) Approximate inference in additive factorial HMMs with application to energy disaggregation. In: Proceedings of the international conference on artificial intelligence and statistics, vol XX, pp 1472–1482. Retrieved from http://people.csail.mit.edu/kolter/lib/exe/fetch.php?media=pubs:kolter-aistats12.pdf
  29. Krizhevsky A, Sutskever I, Geoffrey EH (2012) ImageNet classification with deep convolutional neural networks. In: Advances in neural information processing systems 25 (NIPS2012), pp 1–9. https://doi.org/10.1109/5.726791
  30. Lai YX, Lai CF, Huang YM, Chao HC (2013) Multi-appliance recognition system with hybrid SVM/GMM classifier in ubiquitous smart home. Inf Sci 230:39–55.  https://doi.org/10.1016/j.ins.2012.10.002CrossRefGoogle Scholar
  31. Lange H, Bergés M (2016) The neural energy decoder: energy disaggregation by combining binary subcomponents. In: NILM2016 3rd international workshop on non-intrusive load monitoring. Retrieved from nilmworkshop.orgGoogle Scholar
  32. Laughman C, Lee K, Cox R, Shaw S, Leeb S, Norford L, Armstrong P (2003) Power signature analysis. IEEE Power Energy Mag 1(2):56–63.  https://doi.org/10.1109/MPAE.2003.1192027CrossRefGoogle Scholar
  33. Liang J, Ng SKK, Kendall G, Cheng JWM (2010) Load signature studypart I: basic concept, structure, and methodology. IEEE Trans Power Deliv 25(2):551–560.  https://doi.org/10.1109/TPWRD.2009.2033799CrossRefGoogle Scholar
  34. Lin GY, Lee SC, Hsu JYJ, Jih WR (2010) Applying power meters for appliance recognition on the electric panel. In: Proceedings of the 2010 5th IEEE conference on industrial electronics and applications, ICIEA 2010, pp 2254–2259. https://doi.org/10.1109/ICIEA.2010.5515385
  35. Makonin S, Popowich F (2015) Nonintrusive load monitoring (NILM) performance evaluation: a unified approach for accuracy reporting. Energy Effic 8(4):809–814.  https://doi.org/10.1007/s12053-014-9306-2CrossRefGoogle Scholar
  36. Makonin S, Popowich F, Bajic IV, Gill B, Bartram L (2015) Exploiting HMM sparsity to perform online real-time nonintrusive load monitoring. IEEE Trans Smart Grid.  https://doi.org/10.1109/TSG.2015.2494592Google Scholar
  37. Makonin S, Popowich F, Bartram L, Gill B, Bajic IV (2013) AMPds: a public dataset for load disaggregation and eco-feedback research. In: 2013 IEEE electrical power and energy conference, EPEC 2013. https://doi.org/10.1109/EPEC.2013.6802949
  38. Marchiori A, Hakkarinen D, Han Q, Earle L (2011) Circuit-level load monitoring for household energy management. IEEE Perv Comput 10(1):40–48.  https://doi.org/10.1109/MPRV.2010.72CrossRefGoogle Scholar
  39. Mauch L, Yang B (2015) A new approach for supervised power disaggregation by using a deep recurrent LSTM network. In proceedings of the 3 rd IEEE global conference on signal and information processing (GlobalSIP), pp 63–67Google Scholar
  40. Paradiso F, Paganelli F, Giuli D, Capobianco S (2016) Context-based energy disaggregation in smart homes. Future Internet 8(1):4. Retrieved from http://www.mdpi.com/1999-5903/8/1/4
  41. Parson O, Ghosh S, Weal M, Rogers A (2011) Using hidden Markov models for iterative non-intrusive appliance monitoring. Electronics and Computer Science, University of Southampton, Hampshire, UK, pp 1–4. Retrieved from http://eprints.soton.ac.uk/272990/
  42. Parson O, Ghosh S, Weal M, Rogers A (2012) Non-intrusive load monitoring using prior models of general appliance types. In: Proceedings of the 26th AAAI conference on artificial intelligence, pp 356–362Google Scholar
  43. Parson O, Ghosh S, Weal M, Rogers A (2014) An unsupervised training method for non-intrusive appliance load monitoring. Artif Intell 217:1–19.  https://doi.org/10.1016/j.artint.2014.07.010CrossRefzbMATHGoogle Scholar
  44. Polat H, Du W, Renckes S, Oysal Y, Polat H, Renckes S, Du W (2010) Private predictions on hidden Markov models. Artif Intell Rev 34:53–72.  https://doi.org/10.1007/s10462-010-9161-2CrossRefGoogle Scholar
  45. Ruzzelli AG, Nicolas C, Schoofs A, O’Hare GMP (2010) Real-time recognition and profiling of appliances through a single electricity sensor. In: SECON 2010—2010 7th Annual IEEE communications society conference on sensor, mesh and ad hoc communications and networks. https://doi.org/10.1109/SECON.2010.5508244
  46. Sankar L, Raj Rajagopalan S, Mohajer S, Vincent Poor H (2013) Smart meter privacy: a theoretical framework. IEEE Trans Smart Grid 4(2):837–846.  https://doi.org/10.1109/TSG.2012.2211046CrossRefGoogle Scholar
  47. Srinivasan D, Ng WS, Liew AC (2006) Neural-network-based signature recognition for harmonic source identification. IEEE Trans Power Deliv 21(1):398–405CrossRefGoogle Scholar
  48. Wytock M, Kolter JZ (2013) Contextually supervised source separation with application to energy disaggregation. In: Twenty-eighth AAAI conference on artificial intelligence, pp 1–10. Retrieved from arXiv:1312.5023
  49. Zachary CL (2015) A critical review of recurrent neural networks for sequence learning. arXiv Preprint, pp 1–35. https://doi.org/10.1145/2647868.2654889
  50. Zeifman M (2012) Disaggregation of home energy display data using probabilistic approach. IEEE Trans Consum Electron 58(1):23–31.  https://doi.org/10.1109/TCE.2012.6170051MathSciNetCrossRefGoogle Scholar
  51. Zeifman M, Roth K (2011) Nonintrusive appliance load monitoring: review and outlook. IEEE Trans Consum Electron 57(1):76–84.  https://doi.org/10.1109/TCE.2011.5735484CrossRefGoogle Scholar
  52. Zhang C, Zhong M, Wang Z, Goddard N, Sutton C (2017) Sequence-to-point learning with neural networks for non-intrusive load monitoring. Retrieved from https://pdfs.semanticscholar.org/b519/ffec16d6256b872c6c108023a64d02646293.pdf
  53. Zhong M, Goddard N, Sutton C (2015) Latent Bayesian melding for integrating individual and population models. In: Advances in neural information processing systems, pp 3618–3626. http://papers.nips.cc/paper/5756-latent-bayesianmelding-for-integrating-individual-and-population-models.pdf
  54. Zia T, Bruckner D, Zaidi A (2011) A hidden Markov model based procedure for identifying household electric loads. In: IECON proceedings (industrial electronics conference), pp 3218–3223. https://doi.org/10.1109/IECON.2011.6119826
  55. Zoha A, Gluhak A, Imran MA, Rajasegarar S (2012) Non-intrusive load monitoring approaches for disaggregated energy sensing: a survey. Sensors (Switzerland) 12(12):16838–16866.  https://doi.org/10.3390/s121216838CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media B.V., part of Springer Nature 2018

Authors and Affiliations

  1. 1.School of InformaticsAristotle University of ThessalonikiThessalonikiGreece

Personalised recommendations