Capacity Estimation of Lithium-Ion Batteries Based on Multiple Small Voltage Sections and BP Neural Networks

Yong Tian, Qianyuan Dong, Jindong Tian and Xiaoyu Li ()
Additional contact information
Yong Tian: College of Physics and Optoelectronic Engineering, Shenzhen University, Shenzhen 518060, China
Qianyuan Dong: College of Physics and Optoelectronic Engineering, Shenzhen University, Shenzhen 518060, China
Jindong Tian: College of Physics and Optoelectronic Engineering, Shenzhen University, Shenzhen 518060, China
Xiaoyu Li: College of Physics and Optoelectronic Engineering, Shenzhen University, Shenzhen 518060, China

Energies, 2023, vol. 16, issue 2, 1-18

Abstract: Accurate capacity estimation of onboard lithium-ion batteries is crucial to the performance and safety of electric vehicles. In recent years, data-driven methods based on partial charging curve have been widely studied due to their low requirement of battery knowledge and easy implementation. However, existing data-driven methods are usually based on a fixed voltage segment or state of charge, which would be failed if the charging process does not cover the predetermined segment due to the user’s free charging behavior. This paper proposes a capacity estimation method using multiple small voltage sections and back propagation neural networks. It is intended to reduce the requirement of the length of voltage segment for estimating the complete battery capacity in an incomplete charging cycle. Firstly, the voltage segment most possibly covered is selected and divided into a number of small sections. Then, sectional capacity and skewness of the voltage curve are extracted from these small voltage sections, and severed as health factors. Secondly, the Box–Cox transformation is adopted to enhance the correlation between health factors and the capacity. Thirdly, multiple back propagation neural networks are constructed to achieve capacity estimation based on each voltage section, and their weighted average is taken as the final result. Finally, two public datasets are employed to verify the accuracy and generalization of the proposed method. Results show that the root mean square error of the fusion estimation is lower than 4.5%.

Keywords: capacity estimation; lithium-ion batteries; multiple voltage sections; back propagation neural network; Box–Cox transformation (search for similar items in EconPapers)
JEL-codes: Q Q0 Q4 Q40 Q41 Q42 Q43 Q47 Q48 Q49 (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/1996-1073/16/2/674/pdf (application/pdf)
https://www.mdpi.com/1996-1073/16/2/674/ (text/html)

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:gam:jeners:v:16:y:2023:i:2:p:674-:d:1027060

Access Statistics for this article

Energies is currently edited by Ms. Agatha Cao

More articles in Energies from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().