Submeter based Training of Multi-class Support Vector Machines for Appliance Recognition in Home Electricity Consumption Data

Marco Mittelsdorf, Andreas Hüwel, Thole Klingenberg, Michael Sonnenschein

Abstract

In this paper we employ smart meter and support vector machines (SVM) for the problem of recognizing household appliances’ load patterns in measured load time series, which is an important step for various applications in energy consulting, process recognition or health care applications. We present an automated data collection and preprocessing approach that intrinsically avoids many privacy (and security) issues by keeping the whole process local to the household. In the experimental part we investigate multi-class SVMs in the problem domain of automatically recognizing appliances in load profiles of smart meters. For the learning phase, we use low intrusive submeters to automatically and locally generate household specific test data for the supervised training and validation of the SVMs. We analyze classifiers w.r.t. various training sets and feature spaces. Comparing data from household simulator and real household data, we find that excellent recognition rates can be achieved even with low resolution data and rather unsophisticated feature space.

References

  1. Baranski, M. (2006). Energie-Monitoring im privaten Haushalt. PhD thesis, University of Paderborn.
  2. Hart, G. (1992). Nonintrusive appliance load monitoring. Proceedings of the IEEE, 80(12):1870 -1891.
  3. Jiang, L., Luo, S., and Li, J. (2012). An Approach of Household Power Appliance Monitoring Based on Machine Learning. 2012 Fifth International Conference on Intelligent Computation Technology and Automation, pages 577-580.
  4. Klingenberg, T. (2010). Smart Submetering - Effizienter Einsatz von Submetern zur Aktivitätsbestimmung in Privathaushalten mit Hilfe adaptiver Lernverfahren. Master's thesis, University of Oldenburg.
  5. Kramer, O., Wilken, O., Beenken, P., Hein, A., Hüwel, A., Klingenberg, T., Meinecke, C., Raabe, T., and Sonnenschein, M. (2012). On ensemble classifiers for nonintrusive appliance load monitoring. 7th International Conference on Hybrid Artificial Intelligence Systems (HAIS), pages 322-331.
  6. Leeb, S. B., Shaw, S. R., and Kirtley, J. L. (1995). Transient Event Detection in Spectral Envelope Estimates. IEEE Transactions on Power Delivery, 10(3):1200-1210.
  7. Lin, Y.-h. and Tsai, M.-S. (2011). Applications of hierarchical support vector machines for identifying load operation in nonintrusive load monitoring systems. 2011 9th World Congress on Intelligent Control and Automation, pages 688-693.
  8. Mattern, F., Staake, T., and Weiss, M. (2010). ICT for green: how computers can help us to conserve energy. International Conference on Energy-Efficient Computing and Networking.
  9. Matthews, H. S., Soibelman, L., Berges, M., and Goldman, E. (2008). Automatically disaggregating the total electrical load in residential buildings: a profile of the required solution. In International Workshop on Intelligent Computing in Engineering, page 381389.
  10. Mittelsdorf, M. (2012). Evaluierung des Einflusses von Merkmalsauswahl und Abtastrate auf die Geräteerkennung mit Support-Vektor-Maschinen. Master's thesis, University of Oldenburg.
  11. Onoda, T., Murata, H., and Ratsch, G. (2002). Experimental analysis of support vector machines with different kernels based on non-intrusive monitoring data. Neural Networks, 2002., pages 2186-2191.
  12. Patel, S., Robertson, T., Kientz, J., Reynolds, M., and Abowd, G. (2007). At the flick of a switch: Detecting and classifying unique electrical events on the residential power line (nominated for the best paper award). In Krumm, J., Abowd, G., Seneviratne, A., and Strang, T., editors, UbiComp 2007: Ubiquitous Computing, volume 4717 of Lecture Notes in Computer Science, pages 271-288. Springer Berlin Heidelberg.
  13. Pihala, H. (1998). Non-intrusive appliance load monitoring system based on a modern kWh-meter. Technical Research Centre of Finland.
  14. Raabe, T., Sonnenschein, M., Beenken, P., Hüwel, A., and Meinecke, C. (2012). Energieberatung in haushalten auf basis des smartmetering. O kologisches Wirtschaften, 1:46-50.
  15. Weiss, M., Staake, T., Mattern, F., and Fleisch, E. (2012). Powerpedia: changing energy usage with the help of a community-based smartphone application. Personal and Ubiquitous Computing, 16(6):655-664.
  16. Zeifman, M., Akers, C., and Roth, K. (2011). Nonintrusive appliance load monitoring (nialm) for energy control in residential buildings. International Conference on Energy Efficiency in Domestic Appliances and Lighting (EEDAL), Copenhagen.
  17. Zeifman, M. and Roth, K. (2011). Nonintrusive appliance load monitoring: Review and outlook. IEEE Transactions on Consumer Electronics, 57(1):76-84.
Download


Paper Citation


in Harvard Style

Mittelsdorf M., Hüwel A., Klingenberg T. and Sonnenschein M. (2013). Submeter based Training of Multi-class Support Vector Machines for Appliance Recognition in Home Electricity Consumption Data . In Proceedings of the 2nd International Conference on Smart Grids and Green IT Systems - Volume 1: SMARTGREENS, ISBN 978-989-8565-55-6, pages 151-158. DOI: 10.5220/0004380001510158


in Bibtex Style

@conference{smartgreens13,
author={Marco Mittelsdorf and Andreas Hüwel and Thole Klingenberg and Michael Sonnenschein},
title={Submeter based Training of Multi-class Support Vector Machines for Appliance Recognition in Home Electricity Consumption Data},
booktitle={Proceedings of the 2nd International Conference on Smart Grids and Green IT Systems - Volume 1: SMARTGREENS,},
year={2013},
pages={151-158},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004380001510158},
isbn={978-989-8565-55-6},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 2nd International Conference on Smart Grids and Green IT Systems - Volume 1: SMARTGREENS,
TI - Submeter based Training of Multi-class Support Vector Machines for Appliance Recognition in Home Electricity Consumption Data
SN - 978-989-8565-55-6
AU - Mittelsdorf M.
AU - Hüwel A.
AU - Klingenberg T.
AU - Sonnenschein M.
PY - 2013
SP - 151
EP - 158
DO - 10.5220/0004380001510158