Data driven discovery of an analytic formula for the life prediction of Lithium-ion batteries

Jie Xiong; Tong Xing Lei; Da Meng Fu; Jun Wei Wu; Tong Yi Zhang

doi:10.1016/j.pnsc.2022.12.002

Data driven discovery of an analytic formula for the life prediction of Lithium-ion batteries

Jie Xiong
, Tong Xing Lei
, Da Meng Fu
, Jun Wei Wu^*
, Tong Yi Zhang^*

^*Corresponding author for this work

Harbin Institute of Technology
Hong Kong University of Science and Technology

Research output: Contribution to journal › Article › peer-review

11 Scopus citations

Abstract

Predicting the cycle life of Lithium-Ion Batteries (LIBs) remains a great challenge due to their complicated degradation mechanisms. The present work employs an interpretative machine learning of symbolic regression (SR) to discover an analytic formula for LIB life prediction with newly defined features. The novel features are based on the discharging energies under the constant-current (CC) and constant-voltage (CV) modes at every five cycles alternately. The cycle life is affected by the CC-discharging energy at the 15th cycle (E_15−CCD) and the difference between the CC-discharging energies at the 45th cycle and 95th cycle (Δ₄₅₋₉₅). The cycle life highly correlates with a simple indicator (E_15−CCD−3)/Δ₄₅₋₉₅ with a Pearson correlation coefficient of 0.957. The machine learning tools provide a rapid and accurate prediction of cycle life at the early stage.

Original language	English
Pages (from-to)	793-799
Number of pages	7
Journal	Progress in Natural Science: Materials International
Volume	32
Issue number	6
DOIs	https://doi.org/10.1016/j.pnsc.2022.12.002
State	Published - Dec 2022
Externally published	Yes

UN SDGs

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

SDG 7 Affordable and Clean Energy

Keywords

Cycle life prediction
Lithium-ion batteries
Machine learning
Symbolic regression

Access to Document

10.1016/j.pnsc.2022.12.002

Cite this

@article{b8d3d9b92f944149a4c0e4c95af909f5,

title = "Data driven discovery of an analytic formula for the life prediction of Lithium-ion batteries",

abstract = "Predicting the cycle life of Lithium-Ion Batteries (LIBs) remains a great challenge due to their complicated degradation mechanisms. The present work employs an interpretative machine learning of symbolic regression (SR) to discover an analytic formula for LIB life prediction with newly defined features. The novel features are based on the discharging energies under the constant-current (CC) and constant-voltage (CV) modes at every five cycles alternately. The cycle life is affected by the CC-discharging energy at the 15th cycle (E15−CCD) and the difference between the CC-discharging energies at the 45th cycle and 95th cycle (Δ45−95). The cycle life highly correlates with a simple indicator (E15−CCD−3)/Δ45−95 with a Pearson correlation coefficient of 0.957. The machine learning tools provide a rapid and accurate prediction of cycle life at the early stage.",

keywords = "Cycle life prediction, Lithium-ion batteries, Machine learning, Symbolic regression",

author = "Jie Xiong and Lei, \{Tong Xing\} and Fu, \{Da Meng\} and Wu, \{Jun Wei\} and Zhang, \{Tong Yi\}",

note = "Publisher Copyright: {\textcopyright} 2022 Chinese Materials Research Society",

year = "2022",

month = dec,

doi = "10.1016/j.pnsc.2022.12.002",

language = "英语",

volume = "32",

pages = "793--799",

journal = "Progress in Natural Science: Materials International",

issn = "1002-0071",

publisher = "Elsevier B.V.",

number = "6",

}

TY - JOUR

T1 - Data driven discovery of an analytic formula for the life prediction of Lithium-ion batteries

AU - Xiong, Jie

AU - Lei, Tong Xing

AU - Fu, Da Meng

AU - Wu, Jun Wei

AU - Zhang, Tong Yi

PY - 2022/12

Y1 - 2022/12

N2 - Predicting the cycle life of Lithium-Ion Batteries (LIBs) remains a great challenge due to their complicated degradation mechanisms. The present work employs an interpretative machine learning of symbolic regression (SR) to discover an analytic formula for LIB life prediction with newly defined features. The novel features are based on the discharging energies under the constant-current (CC) and constant-voltage (CV) modes at every five cycles alternately. The cycle life is affected by the CC-discharging energy at the 15th cycle (E15−CCD) and the difference between the CC-discharging energies at the 45th cycle and 95th cycle (Δ45−95). The cycle life highly correlates with a simple indicator (E15−CCD−3)/Δ45−95 with a Pearson correlation coefficient of 0.957. The machine learning tools provide a rapid and accurate prediction of cycle life at the early stage.

AB - Predicting the cycle life of Lithium-Ion Batteries (LIBs) remains a great challenge due to their complicated degradation mechanisms. The present work employs an interpretative machine learning of symbolic regression (SR) to discover an analytic formula for LIB life prediction with newly defined features. The novel features are based on the discharging energies under the constant-current (CC) and constant-voltage (CV) modes at every five cycles alternately. The cycle life is affected by the CC-discharging energy at the 15th cycle (E15−CCD) and the difference between the CC-discharging energies at the 45th cycle and 95th cycle (Δ45−95). The cycle life highly correlates with a simple indicator (E15−CCD−3)/Δ45−95 with a Pearson correlation coefficient of 0.957. The machine learning tools provide a rapid and accurate prediction of cycle life at the early stage.

KW - Cycle life prediction

KW - Lithium-ion batteries

KW - Machine learning

KW - Symbolic regression

UR - https://www.scopus.com/pages/publications/85144788284

U2 - 10.1016/j.pnsc.2022.12.002

DO - 10.1016/j.pnsc.2022.12.002

M3 - 文章

AN - SCOPUS:85144788284

SN - 1002-0071

VL - 32

SP - 793

EP - 799

JO - Progress in Natural Science: Materials International

JF - Progress in Natural Science: Materials International

IS - 6

ER -

Data driven discovery of an analytic formula for the life prediction of Lithium-ion batteries

Abstract

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this