LSTM Network-Based Method for Flexibility Prediction of Aggregated Electric Vehicles in Smart Grid

Huayanran Zhou, Yihong Zhou, Haijing Zhang, Junjie Hu*, Lars Nordströmd, Guangya Yang

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

The flexibility of Demand Response (DR) resources in grid operations has become a valuable solution to respond to the several problems brought about by the growth of intermittent renewable generation. However, the flexibility prediction of the DR resources has not yet been fully addressed in the available literature. This paper trained a long short-term memory (LSTM) recurrent neural network to predict the aggregated flexibility of electric vehicles (EVs). The prediction is based on the historical charging behavior of EVs and the DR signal (DS) which is proposed to facilitate prediction and DR management. Both the size and the maintaining time of the aggregated flexibility can be obtained from the prediction results. The accuracy of the flexibility prediction is verified through the simulation of case studies. The simulation results reveal that the size of flexibility changes under different maintaining time. The proposed flexibility prediction method may be of great assistance for DR management as well as the reserve of the gird.

Original languageEnglish
Title of host publicationProceedings of 2020 International Top-Level Forum on Engineering Science and Technology Development Strategy and The 5th PURPLE MOUNTAIN FORUM, PMF2020
EditorsYusheng Xue, Yuping Zheng, Anjan Bose
PublisherSpringer
Publication date2021
Pages962-974
ISBN (Print)9789811597459
DOIs
Publication statusPublished - 2021
EventInternational Top-Level Forum on Engineering Science and Technology Development Strategy and 5th PURPLE MOUNTAIN FORUM on Smart Grid Protection and Control, PMF2020 - Nanjing, China
Duration: 15 Aug 202016 Aug 2020

Conference

ConferenceInternational Top-Level Forum on Engineering Science and Technology Development Strategy and 5th PURPLE MOUNTAIN FORUM on Smart Grid Protection and Control, PMF2020
CountryChina
CityNanjing
Period15/08/202016/08/2020
SeriesLecture Notes in Electrical Engineering
Volume718
ISSN1876-1100

Bibliographical note

Funding Information:
Funding Information.. This work was supported by the State Key Laboratory of smart grid protection and operation control open project (SGNR0000KJJS1907535).

Publisher Copyright:
© 2021, Springer Nature Singapore Pte Ltd.

Keywords

  • Deep learning (DL)
  • Demand response (DR)
  • Electric vehicles (EVs)
  • Load flexibility
  • Long short-term memory (LSTM)

Fingerprint Dive into the research topics of 'LSTM Network-Based Method for Flexibility Prediction of Aggregated Electric Vehicles in Smart Grid'. Together they form a unique fingerprint.

Cite this