Ghent University Academic Bibliography

Advanced

On the Bayesian optimization and robustness of event detection methods in NILM

Leen De Baets UGent, Joeri Ruyssinck UGent, Chris Develder UGent, Tom Dhaene UGent and Dirk Deschrijver UGent (2017) ENERGY AND BUILDINGS. 145. p.57-66
abstract
A basic but crucial step to increase efficiency and save energy in residential settings is to have an accurate view of energy consumption. To monitor residential energy consumption cost-effectively, i.e., without relying on per-device monitoring equipment, non-intrusive load monitoring (NILM) provides an elegant solution. The aim of NILM is to disaggregate the total power consumption (as measured, e.g., by smart meters at the grid connection point of the household) into individual devices' power consumption, using machine learning techniques. An essential building block of NILM is event detection: detecting when appliances are switched on or off. Current state-of-the-art methods face two open issues. First, they are typically not robust to differences in base load power consumption and secondly, they require extensive parameter optimization. In this paper, both problems are addressed. First two novel and robust algorithms are proposed: a modified version of the chi-squared goodness-of-fit (x(2) GOF) test and an event detection method based on cepstrum smoothing. Then, a workflow using surrogate-based optimization (SBO) to efficiently tune these methods is introduced. Benchmarking on the BLUED dataset shows that both suggested algorithms outperform the standard x2 GOF test for traces with a higher base load and that they can be optimized efficiently using SBO. (C) 2017 Elsevier B.V. All rights reserved.
Please use this url to cite or link to this publication:
author
organization
year
type
journalArticle (original)
publication status
published
keyword
IBCN
journal title
ENERGY AND BUILDINGS
volume
145
pages
57 - 66
Web of Science type
Article
Web of Science id
000401593900005
ISSN
0378-7788
1872-6178
DOI
10.1016/j.enbuild.2017.03.061
language
English
UGent publication?
yes
classification
A1
id
8524157
handle
http://hdl.handle.net/1854/LU-8524157
date created
2017-06-19 08:08:04
date last changed
2017-06-21 13:26:14
@article{8524157,
  abstract     = {A basic but crucial step to increase efficiency and save energy in residential settings is to have an accurate view of energy consumption. To monitor residential energy consumption cost-effectively, i.e., without relying on per-device monitoring equipment, non-intrusive load monitoring (NILM) provides an elegant solution. The aim of NILM is to disaggregate the total power consumption (as measured, e.g., by smart meters at the grid connection point of the household) into individual devices' power consumption, using machine learning techniques. An essential building block of NILM is event detection: detecting when appliances are switched on or off. Current state-of-the-art methods face two open issues. First, they are typically not robust to differences in base load power consumption and secondly, they require extensive parameter optimization. In this paper, both problems are addressed. First two novel and robust algorithms are proposed: a modified version of the chi-squared goodness-of-fit (x(2) GOF) test and an event detection method based on cepstrum smoothing. Then, a workflow using surrogate-based optimization (SBO) to efficiently tune these methods is introduced. Benchmarking on the BLUED dataset shows that both suggested algorithms outperform the standard x2 GOF test for traces with a higher base load and that they can be optimized efficiently using SBO. (C) 2017 Elsevier B.V. All rights reserved.},
  author       = {De Baets, Leen and Ruyssinck, Joeri and Develder, Chris and Dhaene, Tom and Deschrijver, Dirk},
  issn         = {0378-7788},
  journal      = {ENERGY AND BUILDINGS},
  keyword      = {IBCN},
  language     = {eng},
  pages        = {57--66},
  title        = {On the Bayesian optimization and robustness of event detection methods in NILM},
  url          = {http://dx.doi.org/10.1016/j.enbuild.2017.03.061},
  volume       = {145},
  year         = {2017},
}

Chicago
De Baets, Leen, Joeri Ruyssinck, Chris Develder, Tom Dhaene, and Dirk Deschrijver. 2017. “On the Bayesian Optimization and Robustness of Event Detection Methods in NILM.” Energy and Buildings 145: 57–66.
APA
De Baets, L., Ruyssinck, J., Develder, C., Dhaene, T., & Deschrijver, D. (2017). On the Bayesian optimization and robustness of event detection methods in NILM. ENERGY AND BUILDINGS, 145, 57–66.
Vancouver
1.
De Baets L, Ruyssinck J, Develder C, Dhaene T, Deschrijver D. On the Bayesian optimization and robustness of event detection methods in NILM. ENERGY AND BUILDINGS. 2017;145:57–66.
MLA
De Baets, Leen, Joeri Ruyssinck, Chris Develder, et al. “On the Bayesian Optimization and Robustness of Event Detection Methods in NILM.” ENERGY AND BUILDINGS 145 (2017): 57–66. Print.