A generic data driven approach for low sampling load disaggregation

Kaustav Basu (1) , Vincent Debusschere (2) , Seddik Bacha (3, 2) , Ahmad Hably (1) , Danny van Delft (4) , Geert Jan Dirven (4)

1 GIPSA-SYSCO - GIPSA - Systèmes non linéaires et complexité
2 G2ELab - Laboratoire de Génie Electrique de Grenoble
3 UGA [2016-2019] - Université Grenoble Alpes [2016-2019]
4 Greeniant B.V

Non-intrusive load monitoring (Nilm) deals with the disaggregation of individual appliances from the total reading at the power meter. This work proposes an industrial scale solution which uses a specific modeling technique for appliance detection, trained and tested on two distinct databases extracted from actual customers readings. The proposed method is tested for different household categories to address its robustness. The validation of the implemented solution is done over a period of one month with a sampling rate of 10 seconds. The results indicate that high energy consuming appliance can be correctly detected (>80 % of accuracy). In addition, general cases of errors are analyzed, paving the way of the next step in the development of a commercial application of the proposed method.

Keywords: Non-intrusive load monitoring; load disaggregation; data mining; smart meters; energy efficiency; smart grids

Domains: Automatic

Kaustav Basu, Vincent Debusschere, Seddik Bacha, Ahmad Hably, Danny van Delft, et al.. A generic data driven approach for low sampling load disaggregation. Sustainable Energy, Grids and Networks, 2017, 9, pp.118 - 127. ⟨10.1016/j.segan.2016.12.006⟩. ⟨hal-01514076⟩