Multi-Model Based Transfer Learning for Battery Thermal Process

Xingchen Zhang; Han-Xiong Li

doi:10.1109/ICICIP67436.2026.11417503

Multi-Model Based Transfer Learning for Battery Thermal Process

Department of Systems Engineering

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

Abstract

This paper proposed a multi-model based transfer learning (TL) approach for battery thermal process. The method begins with K-means clustering to achieve domain division, enabling the identification of distinct operating regions. Within the Karhunen-Loève framework, a radial basis function neural network (RBFNN) is used to capture dynamics of the source domain. TL is then applied to facilitate parameter transfer of the RBFNN model to the target domain, enabling the rapid construction of multiple spatiotemporal models. During online operation, model adaptability is performed by comparing the Euclidean distance between sensor data and cluster centers. The experimental validation on cylindrical batteries demonstrates its effectiveness. © 2026 IEEE.

Original language	English
Title of host publication	2026 14th International Conference on Intelligent Control and Information Processing (ICICIP)
Publisher	IEEE
Pages	266-270
Number of pages	5
ISBN (Electronic)	979-8-3315-9520-3
DOIs	https://doi.org/10.1109/ICICIP67436.2026.11417503
Publication status	Published - 2026
Event	14th International Conference on Intelligent Control and Information Processing (ICICIP 2026) - Chiang Mai, Thailand Duration: 21 Feb 2026 → 24 Feb 2026 https://conference.cs.cityu.edu.hk/icicip/ICICIP2026/index.html

Publication series

Name	International Conference on Intelligent Control and Information Processing, ICICIP

Conference

Conference	14th International Conference on Intelligent Control and Information Processing (ICICIP 2026)
Place	Thailand
City	Chiang Mai
Period	21/02/26 → 24/02/26
Internet address	https://conference.cs.cityu.edu.hk/icicip/ICICIP2026/index.html

Funding

The work in this paper is supported by the General Research Fund project from the Research Grants Council of Hong Kong (CityU: 11206623).

Research Keywords

Battery Thermal Process
Multi-model
Transfer Learning

RGC Funding Information

RGC-funded

Access Output

10.1109/ICICIP67436.2026.11417503

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{e6c870e2320147a1ace370c1379927ee,

title = "Multi-Model Based Transfer Learning for Battery Thermal Process",

abstract = "This paper proposed a multi-model based transfer learning (TL) approach for battery thermal process. The method begins with K-means clustering to achieve domain division, enabling the identification of distinct operating regions. Within the Karhunen-Lo{\`e}ve framework, a radial basis function neural network (RBFNN) is used to capture dynamics of the source domain. TL is then applied to facilitate parameter transfer of the RBFNN model to the target domain, enabling the rapid construction of multiple spatiotemporal models. During online operation, model adaptability is performed by comparing the Euclidean distance between sensor data and cluster centers. The experimental validation on cylindrical batteries demonstrates its effectiveness. {\textcopyright} 2026 IEEE.",

keywords = "Battery Thermal Process, Multi-model, Transfer Learning",

author = "Xingchen Zhang and Han-Xiong Li",

year = "2026",

doi = "10.1109/ICICIP67436.2026.11417503",

language = "English",

series = "International Conference on Intelligent Control and Information Processing, ICICIP",

publisher = "IEEE",

pages = "266--270",

booktitle = "2026 14th International Conference on Intelligent Control and Information Processing (ICICIP)",

address = "United States",

note = "14th International Conference on Intelligent Control and Information Processing (ICICIP 2026) ; Conference date: 21-02-2026 Through 24-02-2026",

url = "https://conference.cs.cityu.edu.hk/icicip/ICICIP2026/index.html",

}

Zhang, X & Li, H-X 2026, Multi-Model Based Transfer Learning for Battery Thermal Process. in 2026 14th International Conference on Intelligent Control and Information Processing (ICICIP). International Conference on Intelligent Control and Information Processing, ICICIP, IEEE, pp. 266-270, 14th International Conference on Intelligent Control and Information Processing (ICICIP 2026), Chiang Mai, Thailand, 21/02/26. https://doi.org/10.1109/ICICIP67436.2026.11417503

Multi-Model Based Transfer Learning for Battery Thermal Process. / Zhang, Xingchen ; Li, Han-Xiong.
2026 14th International Conference on Intelligent Control and Information Processing (ICICIP). IEEE, 2026. p. 266-270 (International Conference on Intelligent Control and Information Processing, ICICIP).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Multi-Model Based Transfer Learning for Battery Thermal Process

AU - Zhang, Xingchen

AU - Li, Han-Xiong

PY - 2026

Y1 - 2026

N2 - This paper proposed a multi-model based transfer learning (TL) approach for battery thermal process. The method begins with K-means clustering to achieve domain division, enabling the identification of distinct operating regions. Within the Karhunen-Loève framework, a radial basis function neural network (RBFNN) is used to capture dynamics of the source domain. TL is then applied to facilitate parameter transfer of the RBFNN model to the target domain, enabling the rapid construction of multiple spatiotemporal models. During online operation, model adaptability is performed by comparing the Euclidean distance between sensor data and cluster centers. The experimental validation on cylindrical batteries demonstrates its effectiveness. © 2026 IEEE.

AB - This paper proposed a multi-model based transfer learning (TL) approach for battery thermal process. The method begins with K-means clustering to achieve domain division, enabling the identification of distinct operating regions. Within the Karhunen-Loève framework, a radial basis function neural network (RBFNN) is used to capture dynamics of the source domain. TL is then applied to facilitate parameter transfer of the RBFNN model to the target domain, enabling the rapid construction of multiple spatiotemporal models. During online operation, model adaptability is performed by comparing the Euclidean distance between sensor data and cluster centers. The experimental validation on cylindrical batteries demonstrates its effectiveness. © 2026 IEEE.

KW - Battery Thermal Process

KW - Multi-model

KW - Transfer Learning

UR - http://www.scopus.com/inward/record.url?scp=105035827126&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-105035827126&origin=recordpage

U2 - 10.1109/ICICIP67436.2026.11417503

DO - 10.1109/ICICIP67436.2026.11417503

M3 - RGC 32 - Refereed conference paper (with host publication)

T3 - International Conference on Intelligent Control and Information Processing, ICICIP

SP - 266

EP - 270

BT - 2026 14th International Conference on Intelligent Control and Information Processing (ICICIP)

PB - IEEE

T2 - 14th International Conference on Intelligent Control and Information Processing (ICICIP 2026)

Y2 - 21 February 2026 through 24 February 2026

ER -

Multi-Model Based Transfer Learning for Battery Thermal Process

Abstract

Publication series

Conference

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

GRF: Dual-scale Spatiotemporal Learning Based Multiscale Detection for BMS under Edge Sensor Network

Cite this

Multi-Model Based Transfer Learning for Battery Thermal Process

Abstract

Publication series

Conference

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Projects

GRF: Dual-scale Spatiotemporal Learning Based Multiscale Detection for BMS under Edge Sensor Network

Cite this