Model-based fast charging of lithium-ion batteries: Impact of thermal gradients on the degradation of battery systems with cells connected in parallel and series
Year: 2026
Type: article
Abstract: Fast-charging times of less than 15 min are a key enabler for the widespread adoption of battery electric vehicles. Various fast-charging strategies have been developed in recent years to ensure low charging times without shortening the battery lifetime. While these strategies were primarily investigated at the cell level, transferring them to the system level is challenging. This study investigated the aging behavior of a battery module with four cells connected in parallel and four strands connected in series (4s4p topology) using passive cell balancing. To achieve charging times of approximately 15 min while mitigating lithium plating, anode-potential control was employed as a charging strategy. Aging experiments revealed that thermal gradients between the serially connected strands led to diverging degradation. The strand with the highest temperature exhibited increased capacity loss and resistance growth. As a self-reinforcing effect, this strand was exposed to the highest depth of discharge due to its reduced capacity, which exacerbated the degradation. Differential voltage analyses revealed loss of lithium inventory as the dominant aging mode, presumably caused by solid electrolyte interface growth. The presented approach reveals the underlying self-reinforcing degradation loop mechanisms in module-level fast charging, which demonstrates that a holistic view of the thermal management with cells connected in parallel and series is crucial to maximize battery lifetime. • Model-based fast charging profile on system level. • Experimental aging of a battery system until a state of health of 70%. • Investigation of the thermal gradients and cell balancing within the module. • Degradation analysis using differential voltage analysis (DVA) and ECM fitting.
Source: eTransportation
Authors: Christian Allgäuer, Tobias Brehler, Yijun Liu, Claudius Diez, Kareem Abo Gamra, Michael Kick, Raphael Urban, Cristina Grosu, Markus Lienkamp
Institution: Technical University of Munich
Language: English
Subfield: Automotive Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,580
OpenAlex ID: W7127510425
