Enhancing energy efficiency in the residential sector with smart meter data analytics
- 425 Downloads
Tailored energy efficiency campaigns that make use of household-specific information can trigger substantial energy savings in the residential sector. The information required for such campaigns, however, is often missing. We show that utility companies can extract that information from smart meter data using machine learning. We derive 133 features from smart meter and weather data and use the Random Forest classifier that allows us to recognize 19 household classes related to 11 household characteristics (e.g., electric heating, size of dwelling) with an accuracy of up to 95% (69% on average). The results indicate that even datasets with an hourly or daily resolution are sufficient to impute key household characteristics with decent accuracy and that data from different yearly seasons does not considerably influence the classification performance. Furthermore, we demonstrate that a small training data set consisting of only 200 households already reaches a good performance. Our work may serve as benchmark for upcoming, similar research on smart meter data and provide guidance for practitioners for estimating the efforts of implementing such analytics solutions.
KeywordsGreen information systems Decision support systems Data analytics Energy efficiency Sustainability Classification
JEL classificationC80 D10 M310 Q20 R20
We thank Ilya Kozlovskiy for his contribution to the data analysis in this study. We kindly acknowledge financial support from the Swiss Federal Office of Energy (Grant numbers SI/501053-01, SI/501202-01) and want to thank Michael Moser and Roland Brüniger for the very helpful comments during the research project.
- Beckel, C., Sadamori, L., & Santini, S. (2012). Towards automatic classification of private households using electricity consumption data. In G. J. Pappas (Ed.), Proceedings of the fourth ACM workshop on embedded sensing Systems for Energy-Efficiency in buildings (pp. 169–176). Toronto: ACM.CrossRefGoogle Scholar
- Chapman, P., Clinton, J., Kerber, R., Khabaza, T., Reinartz, T., Shearer, C., & Wirth, R. (2000). CRISP-DM 1.0. SPSS. Retrieved from ftp://ftp.software.ibm.com/software/analytics/spss/support/Modeler/Documentation/14/UserManual/CRISP-DM.pdf.
- Cramer, H. (1946). Mathematical methods of statistics. Princeton: Princeton University Press.Google Scholar
- Darby, S. (2006). The effectiveness of feedback on energy consumption. University of Oxford. Retrieved from http://www.usclcorp.com/news/DEFRA-report-with-appendix.pdf.
- Ecoplan. (2015). Smart Metering Roll Out – Kosten und Nutzen: Aktualisierung des Smart Metering Impact Assessments 2012 (Final Report). Bern: Bundesamt für Energie Retrieved from http://www.bfe.admin.ch/php/modules/publikationen/stream.php?extlang=de&name=de_678554277.pdf&endung=Smart%20Metering%20Roll%20Out%20%96%20Kosten%20und%20Nutzen.Google Scholar
- European Commission. (2012). Commission recommendation of 9 march 2012 on preparations for the roll-out of smart metering systems. Official Journal of the European Union. Retrieved from http://eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX:32012H0148.
- European Commission. (2014). COMMISSION STAFF WORKING DOCUMENT Cost-benefit analyses & state of play of smart metering deployment in the EU-27 Accompanying the document Report from the Commission Benchmarking smart metering deployment in the EU-27 with a focus on electricity (COMMISSION STAFF WORKING DOCUMENT no. SWD/2014/0189). Brussels: European Commission.Google Scholar
- Fei, H., Kim, Y., Sahu, S., Naphade, M., Mamidipalli, S. K., & Hutchinson, J. (2013). Heat pump detection from coarse grained smart meter data with positive and unlabeled learning. In Proceedings of the 19th ACM SIGKDD international conference on knowledge discovery and data mining (pp. 1330–1338). New York: ACM. https://doi.org/10.1145/2487575.2488203.CrossRefGoogle Scholar
- Fernández-Delgado, M., Cernadas, E., Barro, S., & Amorim, D. (2014). Do we need hundreds of classifiers to solve real world classification problems? Journal of Machine Learning Research, 15(1), 3133–3181.Google Scholar
- Graml, T., Loock, C.-M., Baeriswyl, M., & Staake, T. (2011). Improving Residential Energy Consumption at Large Using Persuasive Systems. Presented at European Conference on Information Systems (ECIS). In: ECIS 2011 Proceedings. Helsinki, Finland: AIS electronic library. http://aisel.aisnet.org/ecis2011/184/.
- Hopf, K., Sodenkamp, M., Kozlovskiy, I., & Staake, T. (2014). Feature extraction and filtering for household classification based on smart electricity meter data. Computer Science-Research and Development, 31(3), 141–148. Zürich: Springer Berlin Heidelberg. https://doi.org/10.1007/s00450-014-0294-4.CrossRefGoogle Scholar
- Hopf, K., Sodenkamp, M., & Kozlovskiy, I. (2016). Energy data analytics for improved residential service quality and energy efficiency. Presented at 24. European Conference on Information Systems (ECIS), Istanbul: Turkey, June 12-15, In: ECIS 2016 Proceedings, AIS electronic library. http://aisel.aisnet.org/ecis2016_rip/73/.
- Hopf, K., Riechel, S., Sodenkamp, M., & Staake, T. (2017). Predictive customer data analytics – the value of public statistical data and the geographic model transferability. Presented at 38. International Conference on Information Systems (ICIS), Seoul: South Korea 2017, Dec 10-13. In: ICIS 2017 Proceedings, AIS electronic library. http://aisel.aisnet.org/icis2017/DataScience/Presentations/9/.
- Kim, H., Marwah, M., Arlitt, M., Lyon, G., & Han, J. (2011). Unsupervised disaggregation of low frequency power measurements. In: Proceedings of the 2011 SIAM International Conference on Data Mining (Vols. 1–0, pp. 747–758). Society for Industrial and Applied Mathematics. https://doi.org/10.1137/1.9781611972818.64.CrossRefGoogle Scholar
- Kozlovskiy, I., Sodenkamp, M., Hopf, K., & Staake, T. (2016). Energy informatics for environmental, economic and social sustainability: A case of the large-scale detection of households with old heating systems. Presented at 24. European Conference on Information Systems (ECIS), Istanbul: Turkey, June 12-15, In: ECIS 2016 Proceedings, AIS electronic library. https://aisel.aisnet.org/ecis2016_rp/37.
- Kwac, J., Tan, C.-W., Sintov, N., Flora, J., & Rajagopal, R. (2013). Utility customer segmentation based on smart meter data: Empirical study. In: Smart Grid Communications (SmartGridComm), 2013 I.E. International Conference on (pp. 720–725).Google Scholar
- Liaw, A., & Wiener, M. (2015). randomForest: Breiman and Cutler’s Random Forests for Classification and Regression (Version 4.6–12). Retrieved from https://cran.r-project.org/web/packages/randomForest/index.html.
- Romanski, P., & Kotthoff, L. (2014). FSelector: Selecting attributes. Retrieved from http://CRAN.R-project.org/package=FSelector.
- Sodenkamp, M., Kozlovskiy, I., Hopf, K., & Staake, T. (2017). Smart Meter Data Analytics for Enhanced Energy Efficiency in the Residential Sector. In: Wirtschaftsinformatik 2017 Proceedings. St. Gallen: AIS electronic library.Google Scholar
- Swiss Federal Statistical Office. (2017). Sustainable development, regional and international disparities / Statistical basis and overviews (dataset no. FSO: Je-d-21.03.01). Retrieved from https://www.bfs.admin.ch/bfs/en/home/statistics/regional-statistics/regional-portraits-key-figures/communes.assetdetail.2422865.html.
- U.S. Energy Information Administration. (2017). How many smart meters are installed in the United States, and who has them? Retrieved January 18, 2018, from https://www.eia.gov/tools/faqs/faq.php?id=108&t=3.
- U.S. National Centers for Environmental Information. (2016). Climate Data Online. Retrieved January 2, 2016, from http://www.ncdc.noaa.gov/cdo-web/.
- Watson, R. T., Howells, J., & Boudreau, M.-C. (2012). Energy informatics: Initial thoughts on data and process management. In J. vom Brocke, S. Seidel, & J. Recker (Eds.), Green business process management (pp. 147–159). Berlin Heidelberg: Springer. https://doi.org/10.1007/978-3-642-27488-6_9.CrossRefGoogle Scholar