Protecting privacy in microgrids using federated learning and deep reinforcement learning

W. Chen; H. Sun; J. Jiang; M. You; P. William J.S.

doi:10.1049/icp.2023.0100

Protecting privacy in microgrids using federated learning and deep reinforcement learning

W. Chen, H. Sun^*, J. Jiang, M. You, P. William J.S.

^*Corresponding author for this work

Mathematics, Physics and Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

22 Downloads (Pure)

Abstract

This paper aims to improve the energy management efficiency of home microgrids while preserving privacy. The proposed microgrid model includes energy storage systems, PV panels, loads, and the connection to the main grid. A federated multi-objective deep reinforcement learning architecture with Pareto fronts is proposed for total carbon emission and electricity bills optimization. The privacy of data is protected by federated learning, by which the original data will not be uploaded to the server. Numerical results show that compared with the traditional single Deep-Q network, using the proposed method the accumulated carbon emission decreased by 3% and the electricity bills decreased by 21%.

Original language	English
Title of host publication	12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM 2022)
Place of Publication	Stevenage
Publisher	IET
Pages	205-210
Number of pages	6
ISBN (Print)	9781839538513
DOIs	https://doi.org/10.1049/icp.2023.0100
Publication status	Published - 7 Nov 2022
Event	12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM): Enabling high performance & resilience power grid: decarbonization, digitalization, automation, and beyond - Hyatt Regency Tsim Sha Tsui, Hong Kong, Hong Kong Duration: 7 Nov 2022 → 9 Nov 2022 https://www.apscom.org/

Conference

Conference	12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM)
Abbreviated title	APSCOM 2022
Country/Territory	Hong Kong
City	Hong Kong
Period	7/11/22 → 9/11/22
Internet address	https://www.apscom.org/

Keywords

Microgrids
Privacy
Deep learning
Multi-objective

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1049/icp.2023.0100

APSCOM2022_paper_59Accepted author manuscript, 352 KB

Cite this

@inproceedings{3f814e236b0443aba44ff50c43b192fa,

title = "Protecting privacy in microgrids using federated learning and deep reinforcement learning",

abstract = "This paper aims to improve the energy management efficiency of home microgrids while preserving privacy. The proposed microgrid model includes energy storage systems, PV panels, loads, and the connection to the main grid. A federated multi-objective deep reinforcement learning architecture with Pareto fronts is proposed for total carbon emission and electricity bills optimization. The privacy of data is protected by federated learning, by which the original data will not be uploaded to the server. Numerical results show that compared with the traditional single Deep-Q network, using the proposed method the accumulated carbon emission decreased by 3% and the electricity bills decreased by 21%.",

keywords = "Microgrids, Privacy, Deep learning, Multi-objective",

author = "W. Chen and H. Sun and J. Jiang and M. You and J.S., {P. William}",

year = "2022",

month = nov,

day = "7",

doi = "10.1049/icp.2023.0100",

language = "English",

isbn = "9781839538513",

pages = "205--210",

booktitle = "12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM 2022)",

publisher = "IET",

note = "12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM) : Enabling high performance & resilience power grid: decarbonization, digitalization, automation, and beyond, APSCOM 2022 ; Conference date: 07-11-2022 Through 09-11-2022",

url = "https://www.apscom.org/",

}

Chen, W, Sun, H, Jiang, J, You, M & J.S., PW 2022, Protecting privacy in microgrids using federated learning and deep reinforcement learning. in 12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM 2022). IET, Stevenage, pp. 205-210, 12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM), Hong Kong, Hong Kong, 7/11/22. https://doi.org/10.1049/icp.2023.0100

Protecting privacy in microgrids using federated learning and deep reinforcement learning. / Chen, W.; Sun, H.; Jiang, J. et al.
12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM 2022). Stevenage: IET, 2022. p. 205-210.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Protecting privacy in microgrids using federated learning and deep reinforcement learning

AU - Chen, W.

AU - Sun, H.

AU - Jiang, J.

AU - You, M.

AU - J.S., P. William

PY - 2022/11/7

Y1 - 2022/11/7

N2 - This paper aims to improve the energy management efficiency of home microgrids while preserving privacy. The proposed microgrid model includes energy storage systems, PV panels, loads, and the connection to the main grid. A federated multi-objective deep reinforcement learning architecture with Pareto fronts is proposed for total carbon emission and electricity bills optimization. The privacy of data is protected by federated learning, by which the original data will not be uploaded to the server. Numerical results show that compared with the traditional single Deep-Q network, using the proposed method the accumulated carbon emission decreased by 3% and the electricity bills decreased by 21%.

AB - This paper aims to improve the energy management efficiency of home microgrids while preserving privacy. The proposed microgrid model includes energy storage systems, PV panels, loads, and the connection to the main grid. A federated multi-objective deep reinforcement learning architecture with Pareto fronts is proposed for total carbon emission and electricity bills optimization. The privacy of data is protected by federated learning, by which the original data will not be uploaded to the server. Numerical results show that compared with the traditional single Deep-Q network, using the proposed method the accumulated carbon emission decreased by 3% and the electricity bills decreased by 21%.

KW - Microgrids

KW - Privacy

KW - Deep learning

KW - Multi-objective

U2 - 10.1049/icp.2023.0100

DO - 10.1049/icp.2023.0100

M3 - Conference contribution

SN - 9781839538513

SP - 205

EP - 210

BT - 12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM 2022)

PB - IET

CY - Stevenage

T2 - 12th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM)

Y2 - 7 November 2022 through 9 November 2022

ER -

Protecting privacy in microgrids using federated learning and deep reinforcement learning

Abstract

Conference

Keywords

UN SDGs

Access to Document

Fingerprint

Cite this