Identifying household EV models via weighted power recurrence graphs

Wang, H; Ma, J; Zhu, J

Identifying household EV models via weighted power recurrence graphs

Wang, H Ma, J Zhu, J

Permalink

Publisher:: ELSEVIER SCIENCE SA
Publication Type:: Journal Article
Citation:: Electric Power Systems Research, 2023, 217
Issue Date:: 2023-04-01

Closed Access

	Filename	Description	Size
	1-s2.0-S037877962300010X-main.pdf	Published version	2.62 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Wang, H
dc.contributor.author	Ma, J
dc.contributor.author	Zhu, J https://orcid.org/0000-0002-9763-4047
dc.date.accessioned	2024-04-22T05:12:00Z
dc.date.available	2024-04-22T05:12:00Z
dc.date.issued	2023-04-01
dc.identifier.citation	Electric Power Systems Research, 2023, 217
dc.identifier.issn	0378-7796
dc.identifier.issn	1873-2046
dc.identifier.uri	http://hdl.handle.net/10453/178195
dc.description.abstract	Electric vehicles (EVs) are becoming the mainstream transport means in the near future and will play an important role in smart grids with the benefit of energy storage and power factor improvement. Identifying the EV models connected to the power grid for accurate balance in Vehicle-to-Grid and Vehicle-for-Grid operations is crucial. This paper proposes a novel non-intrusive load monitoring (NILM) method based on Weighted Power Recurrence Graphs (WPRG) at a 16.5 kHz sampling rate to accomplish EV model identification irrespective of brand and state of charge (SOC). WPRG can consider the relationship between voltages and currents and thus increase the variance of original electrical current values by using the extracted current and voltage waveforms. The proposed method is tested in a synthetic dataset composed of simulated EV charging loads combined with SOC information and a real-world household appliance dataset. The proposed WPRG method is compared with two weighted-based methods in the NILM background. The proposed WPRG method shows superior performance in identifying EV models and other household appliances than the other two weighted-based methods, with an overall macro F1-score of 94.3% in identifying the whole charging loads in households.
dc.language	English
dc.publisher	ELSEVIER SCIENCE SA
dc.relation.ispartof	Electric Power Systems Research
dc.relation.isbasedon	10.1016/j.epsr.2023.109121
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	0906 Electrical and Electronic Engineering
dc.subject.classification	Energy
dc.subject.classification	4008 Electrical engineering
dc.title	Identifying household EV models via weighted power recurrence graphs
dc.type	Journal Article
utslib.citation.volume	217
utslib.for	0906 Electrical and Electronic Engineering
pubs.organisational-group	University of Technology Sydney
pubs.organisational-group	University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	University of Technology Sydney/Faculty of Engineering and Information Technology/School of Electrical and Data Engineering
utslib.copyright.status	closed_access	*
dc.date.updated	2024-04-22T05:11:59Z
pubs.publication-status	Published
pubs.volume	217

Abstract:

Electric vehicles (EVs) are becoming the mainstream transport means in the near future and will play an important role in smart grids with the benefit of energy storage and power factor improvement. Identifying the EV models connected to the power grid for accurate balance in Vehicle-to-Grid and Vehicle-for-Grid operations is crucial. This paper proposes a novel non-intrusive load monitoring (NILM) method based on Weighted Power Recurrence Graphs (WPRG) at a 16.5 kHz sampling rate to accomplish EV model identification irrespective of brand and state of charge (SOC). WPRG can consider the relationship between voltages and currents and thus increase the variance of original electrical current values by using the extracted current and voltage waveforms. The proposed method is tested in a synthetic dataset composed of simulated EV charging loads combined with SOC information and a real-world household appliance dataset. The proposed WPRG method is compared with two weighted-based methods in the NILM background. The proposed WPRG method shows superior performance in identifying EV models and other household appliances than the other two weighted-based methods, with an overall macro F1-score of 94.3% in identifying the whole charging loads in households.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/178195