Model-Based Virtual Thermal Sensors for Lithium-Ion Battery in EV Applications

Continuous monitoring of temperature distribution in lithium-ion (Li-ion) batteries is critical in preventing rapid degradation, mismatch in cell capacity, and potentially thermal runaway. A model based on virtual thermal sensor (VTS) for automotive grade Li-ion batteries is presented in this paper. This model, using a small number of physical sensors, is able to estimate temperature distribution throughout the battery in real time. First, the thermal model of the battery is developed and the characteristic parameters of the battery are tuned using the prediction error minimization method. Then, the tuned model is combined with a Kalman filter to estimate the temperature distribution of the battery under unknown initial values and model uncertainty. The proposed model-based VTS has been experimentally validated on an automotive grade 70-Ah lithium iron phosphate (LiFePO₄) battery.