This content is not included in your SAE MOBILUS subscription, or you are not logged in.
Modelling and Analysis of a Weak Cell in Different String Configurations
ISSN: 1946-391X, e-ISSN: 1946-3928
Published February 26, 2021 by SAE International in United States
Citation: Rolt, R., Douglas, R., Nockemann, P., and Best, R., "Modelling and Analysis of a Weak Cell in Different String Configurations," SAE Int. J. Commer. Veh. 14(2):2021, https://doi.org/10.4271/02-14-02-0016.
As electric vehicles (EVs) begin to increase their market share in the transport sector, the efficiency of battery packs becomes critical to their performance. Within large battery packs, cell variations occur due to manufacturing processes but can also become prominent during operation due to ineffective thermal management and accelerated degradation of some cells. A battery management system (BMS) will generally account for variations in state of charge (SOC) for cells in series through balancing, but conventional BMSs do not tend to consider the imbalances of cells in parallel as their SOCs should eventually converge themselves. This can, however, lead to cells experiencing higher currents and therefore increased degradation compared to other cells within the pack.
This work looks at natural cell balancing within an EV pack both with and without load and applies the methodology to two different system configurations to analyze their performance when a “weak” cell is present. Equations are derived for the current dynamics between cells in parallel and are written in the general form for any number of parallel cells. The simulation results of the two systems show that with the baseline system configuration (parallel strings of cells in series), there is a noticeable difference between the performance of the weak cell and the performance of the rest of the pack as its SOC varies significantly. In comparison, in the alternative system configuration (series branches containing parallel cells), the weak cell converges significantly quicker to the SOC of the rest of the pack, reducing the overall cell SOC variation. The overall system SOC remains identical for both system configurations, as the mean cell SOC across the pack is unchanged; however, with a greater variance in SOC in the pack, this could potentially cause overcharging/overdischarging for some cells leading to accelerated degradation.