Transfer learning in demand response: A review of algorithms for data-efficient modelling and control

A number of decarbonization scenarios for the energy sector are built on simultaneous electrification of energy demand, and decarbonization of electricity generation through renewable energy sources. However, increased electricity demand due to heat and transport electrification and the variability...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Thijs Peirelinck, Hussain Kazmi, Brida V. Mbuwir, Chris Hermans, Fred Spiessens, Johan Suykens, Geert Deconinck
Formato:	article
Lenguaje:	EN
Publicado:	Elsevier 2022
Materias:	Demand response Transfer learning Reinforcement learning Review Smart grid Electrical engineering. Electronics. Nuclear engineering TK1-9971 Computer software QA76.75-76.765
Acceso en línea:	https://doaj.org/article/d313e56a574c4f5a9f42632ea63e8ffa
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:d313e56a574c4f5a9f42632ea63e8ffa
record_format	dspace
spelling	oai:doaj.org-article:d313e56a574c4f5a9f42632ea63e8ffa2021-11-28T04:38:49ZTransfer learning in demand response: A review of algorithms for data-efficient modelling and control2666-546810.1016/j.egyai.2021.100126https://doaj.org/article/d313e56a574c4f5a9f42632ea63e8ffa2022-01-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S2666546821000732https://doaj.org/toc/2666-5468A number of decarbonization scenarios for the energy sector are built on simultaneous electrification of energy demand, and decarbonization of electricity generation through renewable energy sources. However, increased electricity demand due to heat and transport electrification and the variability associated with renewables have the potential to disrupt stable electric grid operation. To address these issues using demand response, researchers and practitioners have increasingly turned towards automated decision support tools which utilize machine learning and optimization algorithms. However, when applied naively, these algorithms suffer from high sample complexity, which means that it is often impractical to fit sufficiently complex models because of a lack of observed data. Recent advances have shown that techniques such as transfer learning can address this problem and improve their performance considerably — both in supervised and reinforcement learning contexts. Such formulations allow models to leverage existing domain knowledge and human expertise in addition to sparse observational data. More formally, transfer learning embodies all techniques where one aims to increase (learning) performance in a target domain or task, by using knowledge gained in a source domain or task. This paper provides a detailed overview of state-of-the-art techniques on applying transfer learning in demand response, showing improvements that can exceed 30% in a variety of tasks. We observe that most research to date has focused on transfer learning in the context of electricity demand prediction, although reinforcement learning based controllers have also seen increasing attention. However, a number of limitations remain in these studies, including a lack of benchmarks, systematic performance improvement tracking, and consensus on techniques that can help avoid negative transfer.Thijs PeirelinckHussain KazmiBrida V. MbuwirChris HermansFred SpiessensJohan SuykensGeert DeconinckElsevierarticleDemand responseTransfer learningReinforcement learningReviewSmart gridElectrical engineering. Electronics. Nuclear engineeringTK1-9971Computer softwareQA76.75-76.765ENEnergy and AI, Vol 7, Iss , Pp 100126- (2022)
institution	DOAJ
collection	DOAJ
language	EN
topic	Demand response Transfer learning Reinforcement learning Review Smart grid Electrical engineering. Electronics. Nuclear engineering TK1-9971 Computer software QA76.75-76.765
spellingShingle	Demand response Transfer learning Reinforcement learning Review Smart grid Electrical engineering. Electronics. Nuclear engineering TK1-9971 Computer software QA76.75-76.765 Thijs Peirelinck Hussain Kazmi Brida V. Mbuwir Chris Hermans Fred Spiessens Johan Suykens Geert Deconinck Transfer learning in demand response: A review of algorithms for data-efficient modelling and control
description	A number of decarbonization scenarios for the energy sector are built on simultaneous electrification of energy demand, and decarbonization of electricity generation through renewable energy sources. However, increased electricity demand due to heat and transport electrification and the variability associated with renewables have the potential to disrupt stable electric grid operation. To address these issues using demand response, researchers and practitioners have increasingly turned towards automated decision support tools which utilize machine learning and optimization algorithms. However, when applied naively, these algorithms suffer from high sample complexity, which means that it is often impractical to fit sufficiently complex models because of a lack of observed data. Recent advances have shown that techniques such as transfer learning can address this problem and improve their performance considerably — both in supervised and reinforcement learning contexts. Such formulations allow models to leverage existing domain knowledge and human expertise in addition to sparse observational data. More formally, transfer learning embodies all techniques where one aims to increase (learning) performance in a target domain or task, by using knowledge gained in a source domain or task. This paper provides a detailed overview of state-of-the-art techniques on applying transfer learning in demand response, showing improvements that can exceed 30% in a variety of tasks. We observe that most research to date has focused on transfer learning in the context of electricity demand prediction, although reinforcement learning based controllers have also seen increasing attention. However, a number of limitations remain in these studies, including a lack of benchmarks, systematic performance improvement tracking, and consensus on techniques that can help avoid negative transfer.
format	article
author	Thijs Peirelinck Hussain Kazmi Brida V. Mbuwir Chris Hermans Fred Spiessens Johan Suykens Geert Deconinck
author_facet	Thijs Peirelinck Hussain Kazmi Brida V. Mbuwir Chris Hermans Fred Spiessens Johan Suykens Geert Deconinck
author_sort	Thijs Peirelinck
title	Transfer learning in demand response: A review of algorithms for data-efficient modelling and control
title_short	Transfer learning in demand response: A review of algorithms for data-efficient modelling and control
title_full	Transfer learning in demand response: A review of algorithms for data-efficient modelling and control
title_fullStr	Transfer learning in demand response: A review of algorithms for data-efficient modelling and control
title_full_unstemmed	Transfer learning in demand response: A review of algorithms for data-efficient modelling and control
title_sort	transfer learning in demand response: a review of algorithms for data-efficient modelling and control
publisher	Elsevier
publishDate	2022
url	https://doaj.org/article/d313e56a574c4f5a9f42632ea63e8ffa
work_keys_str_mv	AT thijspeirelinck transferlearningindemandresponseareviewofalgorithmsfordataefficientmodellingandcontrol AT hussainkazmi transferlearningindemandresponseareviewofalgorithmsfordataefficientmodellingandcontrol AT bridavmbuwir transferlearningindemandresponseareviewofalgorithmsfordataefficientmodellingandcontrol AT chrishermans transferlearningindemandresponseareviewofalgorithmsfordataefficientmodellingandcontrol AT fredspiessens transferlearningindemandresponseareviewofalgorithmsfordataefficientmodellingandcontrol AT johansuykens transferlearningindemandresponseareviewofalgorithmsfordataefficientmodellingandcontrol AT geertdeconinck transferlearningindemandresponseareviewofalgorithmsfordataefficientmodellingandcontrol
_version_	1718408285411868672

Transfer learning in demand response: A review of algorithms for data-efficient modelling and control

Ejemplares similares