Regional non-intrusive electric vehicle monitoring based on graph signal processing

Jiahang Li, Ran Li, Shuangyuan Wang, Yue Xiang, Yunjie Gu

Research output: Contribution to journalArticlepeer-review

18 Downloads (Pure)

Abstract

Electricity network is leading to a low carbon future with high penetration of plug-in electric vehicles (EVs). However, it is extraordinarily difficult to acquire detailed information on regional EV electrification with an incomplete monitoring system for network operators. In this study, a flexible graph signal processing (GSP)-based non-intrusive monitoring on aggregated EVs is proposed to enhance the EVs visibility for operating power system safely and cost-efficiently. It can deduce the individual EV charging status with the highest possibility iteratively from the limited dataset using a GSP-based possibility calculation after processing a daytime EV characteristic charging patterns. The experiment is developed with realistic EV charging datasets collected in London, and the results show the daily EVs number in a specific region of 500 EVs daily aggregation can be estimated efficiently with an around 4.77% value of relative mean absolute deviation applying the proposed method.

Original languageEnglish
Pages (from-to)6512-6517
Number of pages6
JournalIET Generation, Transmission and Distribution
Volume14
Issue number26
Early online date17 Feb 2021
DOIs
Publication statusPublished - 17 Feb 2021

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Regional non-intrusive electric vehicle monitoring based on graph signal processing'. Together they form a unique fingerprint.

Cite this