Utilization of battery analysis methodologies for parametrization and enhancement of an electrochemically approximated simulation model approach for thermal management battery system tests

Abstract

The exact investigation of the thermal management of battery electric vehicles requires the precise simulation of the processes inside the cell. However, there is a broad trade-off between the complexity and manageability of the simulative methodology on the one hand and the necessary spatial accuracy of the simulation result on the other, which is why complex models that describe the cell down to local current densities and diffusion processes are not suitable for this overriding purpose. An equivalent circuit model proves to be the most suitable for the desired objectives. This model reproduces the electrochemical processes by means of an equivalent circuit and simulates the thermal processes through the cell layers to the surface by means of a thermal network model. Such an approach was pursued in the previous work, using various literature references and empirically obtained parameters (Lorbeck and Fruehwirth in Autom Engine Technol 10:2 https://doi.org/10.1007/s41104-024-00146-2, 2025). This approach proved to be pragmatic, but it quickly became clear that a more precise method to parameterize the equivalent circuit diagram was required for this specific application. The improved parametrization includes the measurement of several cells in terms of electrochemical impedance spectroscopy and open-circuit voltage, which are carried out and processed on site. This method makes it possible to achieve an adapted parametrization relatively quickly by measuring other cells. After carrying out these measurements and integrating the new parameters into the simulation methodology, a validation against measured cells and against the previously used simulation methodology (literature-based, partly empirically determined parameters) could be carried out. Despite the additional effort involved in measuring the cells, the validation speaks in favor of introducing the new simulation methodology, as the accuracy of the predictions and in particular the simulation of transient parts within the cycles has improved. This means that in future, it will be possible to better illuminate and understand various areas of thermal and electrical cell stress that are of interest for the observations. A further validation, which compares different real driving measurements and testbed cycles on cell level with the according simulation data, is planned for 2025 and is currently being analyzed as part of ongoing investigations.