Duck Curve Aware Dynamic Pricing and Battery Scheduling Strategy Using Reinforcement Learning

Watari, Daichi; Taniguchi, Ittetsu; Onoye, Takao

doi:https://doi.org/10.1109/TSG.2023.3288355

Duck Curve Aware Dynamic Pricing and Battery Scheduling Strategy Using Reinforcement Learning

Watari, Daichi; Taniguchi, Ittetsu; Onoye, Takao

IEEE Transactions on Smart Grid, 2023, 15(1), 457-471

アクセス数:321件（2025-08-14 16:02 集計）

固定URL: https://hdl.handle.net/11094/92478

閲覧可能ファイル

ファイル	フォーマット	利用条件	サイズ	閲覧回数	利用開始日	説明	information
IEEETransSmartGrid	pdf	なし	2.01 MB	147	2023.09.04

Creative Commons : 表示

論文情報

ファイル出力 EndNote Basic出力 Mendeley出力

タイトル	Duck Curve Aware Dynamic Pricing and Battery Scheduling Strategy Using Reinforcement Learning
著者	著者名 Watari, Daichi
	NRID 1000040551453 1000040551453 ORCID 0000-0002-7843-5907 0000-0002-7843-5907 researchmap permalink ittetsu ittetsu Scopus Author ID 23013657500 23013657500 著者名 Taniguchi, Ittetsu
	NRID 1000060252590 1000060252590 researchmap permalink read0096872 read0096872 著者名 Onoye, Takao
キーワード等	Batteries
	Battery scheduling
	Costs
	Deep reinforcement learning
	Demand response
	Duck curve
	Dynamic pricing
	Heuristic algorithms
	Power generation
	Pricing
	Prosumer
	State of charge
	Vehicle dynamics
内容	Watari D., Taniguchi I., Onoye T.. Duck Curve Aware Dynamic Pricing and Battery Scheduling Strategy Using Reinforcement Learning. IEEE Transactions on Smart Grid , (2023); https://doi.org/10.1109/TSG.2023.3288355.
抄録	The duck curve is becoming a global problem in energy technology due to the rapid increase in solar power adoption and the rise of prosumers. To address this issue, a resource aggregator (RA) has emerged to provide flexible solutions through aggregating the prosumers and demand response such as dynamic pricing. This paper proposes an optimal strategy for the RA that dispatches dynamic pricing to the prosumers and leverages the battery system at both RA and prosumer levels. The proposed method is based on a model-free deep reinforcement learning (DRL) algorithm to optimize each prosumer’s retail prices and schedule usage of the RA’s battery power station. An objective reward function is used to maximize the RA’s profit, minimize the prosumer’s cost, and maximize the improvement of the duck curve. The performance of the proposed DRL-based strategy was demonstrated by simulation experiments using actual wholesale price, demand, and PV generation data. The results show that the proposed strategy can improve the standard deviation and peak-to-average ratio of net load by up to 57.1% and 23%, respectively.
出版者	Institute of Electrical and Electronics Engineers Inc.
収録物名	IEEE Transactions on Smart Grid
巻(号)	15(1)
ページ	457-471
刊行年月	2023-06-21
言語	英語
ハンドルURL	https://hdl.handle.net/11094/92478
PISSN	19493053
eISSN	19493061
NCID	AA12479464
関連情報 (isIdenticalTo)	DOI (出版社版) https://doi.org/10.1109/TSG.2023.3288355
アクセス権	open access
権利情報	This article is licensed under a Creative Commons Attribution 4.0 International License.
出版タイプ	Version of Record
カテゴリ	学術雑誌論文 Journal Article

資源タイプ	学術雑誌論文
ローカル資源タイプ	学術雑誌論文
DCMI資源タイプ	text
DCTERMS.bibliographicCitation	IEEE Transactions on Smart Grid. 2023, 15(1), p. 457-471
DC.title	Duck Curve Aware Dynamic Pricing and Battery Scheduling Strategy Using Reinforcement Learning
DC.creator	Watari, Daichi
	Taniguchi, Ittetsu
	Onoye, Takao
DC.publisher	Institute of Electrical and Electronics Engineers Inc.
DC.language' scheme='DCTERMS.RFC1766	英語
DC.type' scheme='DCTERMS.DCMIType	text
DCTERMS.issued' scheme='DCTERMS.W3CDTF	2023-06-21
DC.identifier	https://doi.org/10.1109/TSG.2023.3288355
DC.subject	Batteries
	Battery scheduling
	Costs
	Deep reinforcement learning
	Demand response
	Duck curve
	Dynamic pricing
	Heuristic algorithms
	Power generation
	Pricing
	Prosumer
	State of charge
	Vehicle dynamics
DC.description	Watari D., Taniguchi I., Onoye T.. Duck Curve Aware Dynamic Pricing and Battery Scheduling Strategy Using Reinforcement Learning. IEEE Transactions on Smart Grid , (2023); https://doi.org/10.1109/TSG.2023.3288355.
DCTERMS.abstract	The duck curve is becoming a global problem in energy technology due to the rapid increase in solar power adoption and the rise of prosumers. To address this issue, a resource aggregator (RA) has emerged to provide flexible solutions through aggregating the prosumers and demand response such as dynamic pricing. This paper proposes an optimal strategy for the RA that dispatches dynamic pricing to the prosumers and leverages the battery system at both RA and prosumer levels. The proposed method is based on a model-free deep reinforcement learning (DRL) algorithm to optimize each prosumer’s retail prices and schedule usage of the RA’s battery power station. An objective reward function is used to maximize the RA’s profit, minimize the prosumer’s cost, and maximize the improvement of the duck curve. The performance of the proposed DRL-based strategy was demonstrated by simulation experiments using actual wholesale price, demand, and PV generation data. The results show that the proposed strategy can improve the standard deviation and peak-to-average ratio of net load by up to 57.1% and 23%, respectively.
DC.rights	This article is licensed under a Creative Commons Attribution 4.0 International License.
DC.format	92478
citation_title	Duck Curve Aware Dynamic Pricing and Battery Scheduling Strategy Using Reinforcement Learning
citation_author	Watari, Daichi
	Taniguchi, Ittetsu
	Onoye, Takao
citation_publisher	Institute of Electrical and Electronics Engineers Inc.
citation_language	英語
citation_date	2023-06-21
citation_journal_title	IEEE Transactions on Smart Grid
citation_volume	15
citation_issue	1
citation_firstpage	457
citation_lastpage	471
citation_doi	https://doi.org/10.1109/TSG.2023.3288355
citation_public_url	https://hdl.handle.net/11094/92478

Duck Curve Aware Dynamic Pricing and Battery Scheduling Strategy Using Reinforcement Learning

閲覧可能ファイル

論文情報

論文詳細を表示