Correlations of lithium-ion battery parameter variations and connected configurations on pack statistics

Abstract

Battery cell-to-cell parameter variations and connected configurations jointly affect pack performance. Knowledge of the quantitative correlations of lithium-ion battery parameter variations and connected configurations on pack statistics is crucial for understanding and improving the pack performance in the automotive industry. From a statistical perspective, this study theoretically developed analytical correlations to analyze the impact of the statistics (mean and standard deviation) of cell-level parameters (capacity and resistance) and connected configurations on module/pack statistical performance under a random sampling scenario. These correlations are applied to various cell-to-cell parameter variations and module/pack configurations and are verified by Monte Carlo simulations. The results demonstrate that cell-to-cell variations have a negative effect on the statistical performance of modules, especially series mean capacity. Meanwhile, parallel and series configurations can dramatically reduce the variation in module-level parameters, and the variations in parallel or series modules with four cells are 50% of those of cell-to-cell. Therefore, parallel configurations can effectively compensate for the loss of mean capacity caused by series configuration so that the pack with parallel modules connected in series has the capacity advantages compared to that of the pack with series modules connected in parallel, but their resistance statistics are the same. The correlations developed in this study provide important guidance for battery screening and the selection of pack configuration and equalization topology, thereby giving critical information for optimizing pack performance from the cell level to the pack level.