Approach to determine the entropy coefficient of a battery by numerical optimization

Until now, the entropy coefficient of a battery has usually been determined experimentally using time-consuming potentiometric or calorimetric measuring methods. This paper aims to present an approach for the experimental determination of the entropy coefficient with numerical optimization using a thermal 0-dimensional simulation. The method was developed and validated by simulations of a typical battery module. A thermally validated simulation model of a real battery served as the basis for the simulations performed. The simulation results confirm that it is possible to determine the entropy coefficient of a battery module by numerical optimization. Results in different test scenarios (dynamic load cycles: Artemis cycle, US06 cycle) show a maximum absolute temperature difference of 0.1 K when comparing the results of a thermal higher-dimensional simulation (in this paper, a quasi-2-dimensional simulation is used) with the results of the thermal 0-dimensional simulation. The new method can be implemented on a real battery system on a test bench. Previous time-consuming measurement procedures can thus be shortened. [Display omitted] •Determine the entropy coefficient of a battery based on numerical optimization.•Reduction of the time needed to identify the entropy coefficient of a battery.•Entropy coefficient as an essential term describing the thermal behavior of a battery.•Design, parametrization, and application of a universal 0-D thermal model.