Development of a Lithium-ion battery predictive model for electrified powertrains

Battery electric vehicles (BEVs), represent one of the valuable solutions to shift toward more sustainable mobility thanks to their high efficiency and to the possibility of using renewable energy for battery recharge. As a result, their market share has been growing in the last few years, with reference to the transportation of goods as well as people. Nevertheless, one of the major challenges for BEVs is still to ensure a sufficient range, allowing them to travel ever longer routes with a single recharge. In this sense, the creation of a reliable model is of primary importance to optimize the battery pack and to simulate its performance also in extreme ambient conditions. In such a framework, this master thesis, realized in collaboration with FPT Industrial S.p.A., has the purpose of creating, in the commercial software GT-AutoLion, the electrochemical model of a Li-Ion battery pack. The activity has been carried out starting from the specifications of a single cell, characterized by an elevated capacity value and suitable for automotive applications, and calibrating its model with reference to experimental data. As a second step, the model has been validated on different temperature conditions and discharge rates. Upscaling the single-cell model to the whole battery pack, an equivalent Thevenin circuit with dependency on temperature and State of Charge (SOC) has been also created. It has been integrated into the virtual test rig of a light-duty vehicle (LDV), to assess its performance and the aging phenomena. Furthermore, the electrothermal model of the cell was used to analyze the heat rejection of the pack cooling circuit.