Recovering large-scale battery aging dataset with machine learning

Xiaopeng TANG; Kailong LIU; Kang LI; Widanalage Dhammika WIDANAGE; Emma KENDRICK; Furong GAO

doi:10.1016/j.patter.2021.100302

Recovering large-scale battery aging dataset with machine learning

Xiaopeng TANG, Kailong LIU^*, Kang LI, Widanalage Dhammika WIDANAGE, Emma KENDRICK, Furong GAO

^*Corresponding author for this work

Research output: Journal Publications › Journal Article (refereed) › peer-review

93 Citations (Scopus)

Abstract

Batteries are crucial for building a clean and sustainable society, and their performance is highly affected by aging status. Reliable battery health assessment, however, is currently restrained by limited access to sufficient aging data, resulting from not only complicated battery operations but also long aging experimental time (several months or even years). Refining industrial datasets (e.g., the field data from electric vehicle batteries) unlocks opportunities to acquire large-volume aging data with low experimental efforts. We introduce the potential of combining industrial data with accelerated aging tests to recover high-quality battery aging datasets, through a migration-based machine learning. A comprehensive dataset containing 8,947 aging cycles with 15 operational modes is collected for evaluation. While saving up to 90% experimental time, aging data can be recovered with ultra-low error within 1%. It provides an alternative solution to significantly improve data shortage issues for assessment of battery and energy storage aging.

Original language	English
Article number	100302
Journal	Patterns
Volume	2
Issue number	8
Early online date	30 Jun 2021
DOIs	https://doi.org/10.1016/j.patter.2021.100302
Publication status	Published - 13 Aug 2021
Externally published	Yes

Bibliographical note

Publisher Copyright:
© 2021 The Authors

Funding

This paper was financially supported, in part, by the Ministry of Science and Technology of the People's Republic of China ( SQ2019YFB170029 ), the Foshan-HKUST Project ( FSUST19-FYTRI01 ), Guangzhou Science and Technology Project ( 202002030323 ), the Faraday Institution Multi-scale Modelling programme (faraday.ac.uk; EP/S003053/1 and FIRG003 ), and EPSRC grants EP/R030243/1 and EP/P004636/1.

Keywords

accelerated battery aging experiments
battery aging assessment
battery aging dataset generation
DSML 1: Concept: Basic principles of a new data science output observed and reported
incremental capacity analysis
lithium-ion battery management
machine learning
model migration

UN SDGs

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

Access to Document

10.1016/j.patter.2021.100302Licence: CC BY-NC-ND

Cite this

@article{700d8e75de4f4cdc9f2deee03e0ac4ea,

title = "Recovering large-scale battery aging dataset with machine learning",

abstract = "Batteries are crucial for building a clean and sustainable society, and their performance is highly affected by aging status. Reliable battery health assessment, however, is currently restrained by limited access to sufficient aging data, resulting from not only complicated battery operations but also long aging experimental time (several months or even years). Refining industrial datasets (e.g., the field data from electric vehicle batteries) unlocks opportunities to acquire large-volume aging data with low experimental efforts. We introduce the potential of combining industrial data with accelerated aging tests to recover high-quality battery aging datasets, through a migration-based machine learning. A comprehensive dataset containing 8,947 aging cycles with 15 operational modes is collected for evaluation. While saving up to 90% experimental time, aging data can be recovered with ultra-low error within 1%. It provides an alternative solution to significantly improve data shortage issues for assessment of battery and energy storage aging.",

keywords = "accelerated battery aging experiments, battery aging assessment, battery aging dataset generation, DSML 1: Concept: Basic principles of a new data science output observed and reported, incremental capacity analysis, lithium-ion battery management, machine learning, model migration",

author = "Xiaopeng TANG and Kailong LIU and Kang LI and WIDANAGE, {Widanalage Dhammika} and Emma KENDRICK and Furong GAO",

note = "Publisher Copyright: {\textcopyright} 2021 The Authors",

year = "2021",

month = aug,

day = "13",

doi = "10.1016/j.patter.2021.100302",

language = "English",

volume = "2",

journal = "Patterns",

issn = "2666-3899",

publisher = "Cell Press",

number = "8",

}

TY - JOUR

T1 - Recovering large-scale battery aging dataset with machine learning

AU - TANG, Xiaopeng

AU - LIU, Kailong

AU - LI, Kang

AU - WIDANAGE, Widanalage Dhammika

AU - KENDRICK, Emma

AU - GAO, Furong

PY - 2021/8/13

Y1 - 2021/8/13

N2 - Batteries are crucial for building a clean and sustainable society, and their performance is highly affected by aging status. Reliable battery health assessment, however, is currently restrained by limited access to sufficient aging data, resulting from not only complicated battery operations but also long aging experimental time (several months or even years). Refining industrial datasets (e.g., the field data from electric vehicle batteries) unlocks opportunities to acquire large-volume aging data with low experimental efforts. We introduce the potential of combining industrial data with accelerated aging tests to recover high-quality battery aging datasets, through a migration-based machine learning. A comprehensive dataset containing 8,947 aging cycles with 15 operational modes is collected for evaluation. While saving up to 90% experimental time, aging data can be recovered with ultra-low error within 1%. It provides an alternative solution to significantly improve data shortage issues for assessment of battery and energy storage aging.

AB - Batteries are crucial for building a clean and sustainable society, and their performance is highly affected by aging status. Reliable battery health assessment, however, is currently restrained by limited access to sufficient aging data, resulting from not only complicated battery operations but also long aging experimental time (several months or even years). Refining industrial datasets (e.g., the field data from electric vehicle batteries) unlocks opportunities to acquire large-volume aging data with low experimental efforts. We introduce the potential of combining industrial data with accelerated aging tests to recover high-quality battery aging datasets, through a migration-based machine learning. A comprehensive dataset containing 8,947 aging cycles with 15 operational modes is collected for evaluation. While saving up to 90% experimental time, aging data can be recovered with ultra-low error within 1%. It provides an alternative solution to significantly improve data shortage issues for assessment of battery and energy storage aging.

KW - accelerated battery aging experiments

KW - battery aging assessment

KW - battery aging dataset generation

KW - DSML 1: Concept: Basic principles of a new data science output observed and reported

KW - incremental capacity analysis

KW - lithium-ion battery management

KW - machine learning

KW - model migration

UR - http://www.scopus.com/inward/record.url?scp=85112251675&partnerID=8YFLogxK

U2 - 10.1016/j.patter.2021.100302

DO - 10.1016/j.patter.2021.100302

M3 - Journal Article (refereed)

AN - SCOPUS:85112251675

SN - 2666-3899

VL - 2

JO - Patterns

JF - Patterns

IS - 8

M1 - 100302

ER -

Recovering large-scale battery aging dataset with machine learning

Abstract

Bibliographical note

Funding

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this