Framework for Digital Twin Real-Time Battery System for Model-in-the-loop and Hardware-in-the-loop Simulations

The complexity of Battery systems for electric vehicles (EV) is increasing as EVs become more popular. This corresponds to an increased cost in testing physical battery packs. In such scenarios, moving towards simulation is a fast and less expensive approach. 1D simulation is needed to consider the physical dynamics of systems in an EV, such as the battery pack and the thermal management system, to accurately predict how the EV will perform in the real world. Lithium-ion battery pack modeling can be a challenging task due to its latency and capturing cell-to-cell heterogeneity. Moreover, the thermal management system plays a vital role in ensuring the battery pack is always within its operable range. Here, we use a gaussian distribution to estimate the variations of battery modules in the pack and a 1D thermal model to capture the temperature distribution amongst the cells. To develop a more accurate battery management system (BMS), it is necessary to use high fidelity models to capture as much physics as possible. The complex architecture of electric vehicles (EVs) makes it computationally expensive to perform Model-in-the-Loop (MIL) and Hardware-in-the-Loop (HIL) simulations which have hard Real-Time requirements. In the following work we present an efficient HIL framework for a battery pack that takes into consideration module-to-module variation within a lithium-ion battery pack along with simple thermal model.