Applying a numerical optimization to determine the entropic heat coefficient of a battery

The quality of modeling the thermal behavior of a battery also depends on the accuracy of the entropic heat coefficient (EHC) applied. The EHC has so far been experimentally determined using time-consuming measuring methods. This paper shows the application of a method determining the EHC based on experiments and numerical optimization. The applied practical steps to determine the EHC are presented and discussed. This method provides sufficiently accurate results for practical applications within a relatively short time (within six days). The simulations and measurements for the validation were performed simultaneously on a Hardware-in-the-Loop test bench. At the beginning of each experiment, the characteristic maps for the battery module's electrical simulation were unknown and then determined simultaneously online. The simulation results compared to the simultaneously recorded measurement data of different dynamic test scenarios (about one hour each) show a temperature difference of at most 0.4 K, with an overall increase of 3.9 K. Furthermore, the results show that the influence of the EHC contributes significantly to the resulting temperature curve and should not be neglected in thermal analyses of a battery: Deviations of up to 38% in the temperature values have been shown when the EHC is not considered.