Inconsistency modeling of lithium-ion battery pack based on variational auto-encoder considering multi-parameter correlation

Xitian He, Bingxiang Sun, Weige Zhang, Xiaojia Su, Shichang Ma, Hao Li and Haijun Ruan

Energy, 2023, vol. 277, issue C

Abstract: —The accurate battery pack model is of great significance for the strategy development and functional verification of battery management system with the advantages of the high repeatability, fast state switching, and high safety. In this paper, a total of 49 dimensions of battery pack parameters are obtained through systematic experiments, including equivalent circuit model parameters and thermal characteristic parameters. Since the variational auto-encoder (VAE) can well preserve the correlation between parameters by training the neural network with small samples (95 samples), a novel battery pack inconsistency method is proposed based on VAE. Moreover, the parameter correlation and similarity are further considered based on the original loss function, which enables parameters generation for any amount of samples. The simulation results of the 95 and 500 generated samples illustrated that the proposed method can well maintain the similarity with the original parameters in both parameters distribution and parameters correlation, compared with the Copula-based and Metropolis-Hastings-based method. The average temperature error of the VAE-based method (0.02%) is much smaller than that of the Copula-based method (0.80%), and the temperature standard deviation error of the VAE-based method can be reduced to 4.53%, while the Copula-based method can reach 105.38%.

Keywords: Lithium-ion battery pack; Inconsistency modeling; Variational auto-encoder; Multi-parameter correlation (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544223008034
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:277:y:2023:i:c:s0360544223008034

DOI: 10.1016/j.energy.2023.127409

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().