Equivalent circuit model based voltage and surface temperature prediction for a lithium-ion pouch cell under different ambient temperatures

The growing demand for fast and accurate modelling of lithium-ion cells is driven by their increasing use as energy storage devices in electric vehicles and other related applications. This study aims to develop a coupled electro-thermal model based on an equivalent circuit architecture, capable of predicting both voltage and multi-point temperature data for a 20Ah pouch cell discharging under varying ambient temperature conditions. Discharge experiments are conducted under ambient temperature conditions of 10, 25 and 40 °C, using an in-house developed experimental setup. The electrical model is first trained using constant C-rate pulse discharge data, with model parameters estimated as a function of cell temperature and state of charge. A thermal model is then developed by assigning a thermal mass, connected by thermal resistances, to each temperature measurement point and the two tabs. Six temperature measurements are used to train and validate the thermal model. The validation of the electrical model, along with the parameter identification and testing of the independent thermal model, is performed using variable C-rate discharge experiments. The combined model, formed by integrating the electrical and thermal models, is tested with experimental data from the US06 driving cycle. The results show that the voltage estimation error of the combined model is within ± 100 mV and the temperature accuracy is within 1 °C. The final model successfully captures both single-cycle behaviour and cycle-to-cycle variations in cell voltage and maximum cell temperature. Additionally, the combined model effectively reflects the variation of these electro-thermal parameters across the three different ambient temperature conditions.