A novel energy management strategy for the ternary lithium batteries based on the dynamic equivalent circuit modeling and differential Kalman filtering under time-varying conditions

Shun-Li Wang, Daniel-Ioan Stroe, Carlos Fernandez, Chunmei Yu, Chuanyuan Zou, Xiaoxia Li

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

36 Citationer (Scopus)
41 Downloads (Pure)

Abstract

The dynamic model of the ternary lithium battery is a time-varying nonlinear system due to the polarization and diffusion effects inside the battery in its charge-discharge process. Based on the comprehensive analysis of the energy management methods, the state of charge is estimated by introducing the differential Kalman filtering method combined with the dynamic equivalent circuit model considering the nonlinear temperature coefficient. The model simulates the transient response with high precision which is suitable for its high current and complicated charging and discharging conditions. In order to better reflect the dynamic characteristics of the power ternary lithium battery in the step-type charging and discharging conditions, the polarization circuit of the model is differential and the improved iterate calculation model is obtained. As can be known from the experimental verifications, the maximize state of charge estimation error is only 0.022 under the time-varying complex working conditions and the output voltage is monitored simultaneously with the maximum error of 0.08 V and the average error of 0.04 V. The established model can describe the dynamic battery behavior effectively, which can estimate its state of charge value with considerably high precision, providing an effective energy management strategy for the ternary lithium batteries.
OriginalsprogEngelsk
Artikelnummer227652
TidsskriftJournal of Power Sources
Vol/bind450
Antal sider14
ISSN0378-7753
DOI
StatusUdgivet - feb. 2020

Fingeraftryk

Dyk ned i forskningsemnerne om 'A novel energy management strategy for the ternary lithium batteries based on the dynamic equivalent circuit modeling and differential Kalman filtering under time-varying conditions'. Sammen danner de et unikt fingeraftryk.

Citationsformater