SUMMARY The thermal behavior of a Li‐ion battery module that belongs to the battery system of an actual electric vehicle prototype was numerically investigated. Realistic driving loads and passive cooling conditions were considered. A combination of a vehicle dynamics model, an equivalent electric circuit battery model, and a 3D finite‐element thermal model was used in the analysis. Temperature and electric potential measurements, performed at the cell and module levels, were first used for model calibration. Electric currents, associated with the ARTEMIS driving cycles, were then calculated and applied in the battery model to predict the heat sources for the thermal model. It was found that the temperature increase corresponding to urban transportation requirements in European countries is tolerable. Nevertheless, road and highway applications would result in a temperature increase that accelerates cell ageing, and an active cooling strategy is required. Copyright © 2012 John Wiley & Sons, Ltd.

One of the challenges faced when using Li-ion batteries in electric vehicles is to keep the cell temperatures below a given threshold. Mathematical modeling would indeed be an efficient tool to test virtually this requirement and accelerate the battery product lifecycle. Moreover, temperature predicting models could potentially be used on-board to decrease the limitations associated with sensor based temperature feedbacks. Accordingly, we present a complete modeling procedure which was used to calculate the cell temperatures during a given electric vehicle trip. The procedure includes a simple vehicle dynamics model, an equivalent circuit battery model, and a 3D finite element thermal model. Model parameters were identified from measurements taken during constant current and pulse current discharge tests. The cell temperatures corresponding to an actual electric vehicle trip were calculated and compared with measured values. The resulting accuracy was high enough (max error 1.07 K) and suggests that designers could rely largely on similar numerical thermal simulations during the design of Li-ion battery systems for electric vehicles. Additionally, the thermal model could be used on-board in a battery management system control strategy to keep the cell temperatures within a safe window. A model reduction procedure is nevertheless needed to scale down the computational effort to the on-board capabilities.