A resistance-based electro-thermal coupled model for an air-cooled battery pack that considers branch current variation

The large, complex batteries that are increasingly used in applications such as electric vehicles generate heat. As such, they require thermal management systems that can predict this heat generation. In this study, an electric-thermal coupled model was established to predict the temperature evolution of an air-cooled battery pack comprising three parallel branches with four cells in each branch. The model considers the influences of cell temperature and state of charge on ohmic and polarization resistances, and the interactions of cell temperature, resistance and current distribution. It can accurately predict temperature increases in cells in a pack unevenly cooled by air under different conditions. Experiments were conducted to verify the model's precision at different discharge rates and ambient temperatures. The majority of average relative errors (RE_ave) between the estimated and measured values were <3%, with a maximum of only 4.02%. The maximum RE_ave values were 24.58% for the T-relevant resistance model and 16.52% for the constant resistance model. This demonstrates that the proposed electro-thermal model can predict the temperature evolution of a battery pack much more precisely than existing methods.