Quantitative Evaluation of Available Capacity of Series Battery Pack Based on Monte Carlo Simulation

To obtain high-power and large-capacity, power battery system for EVs or energy storage battery system for power grid usually takes the pack form that connects hundreds or even thousands of cells in a series-parallel structure. Therefore, the inconsistency between cells becomes a key factor that affects the performance of the battery pack. To evaluate this impact, this paper propose a quantitative evaluation method from the statistical analysis by considering the impact of the cells’ inconsistency on the battery pack performance as a stochastic process. Firstly, the available capacity is taken as the indicator of battery pack’s performance and its probability distribution features is derived based on the probability distribution of cells’ capacity, SOC and internal resistance. Then, the probability distribution function of available capacity and its numerical characteristics including expectation and variance are calculated based on Monte Carlo simulation in the case that the cells’ parameters follow the normal distribution. The results of the theoretical analysis and the simulating calculation show that the available capacity of the battery pack follows Burr distribution with significant skewness and the expectation of pack’s available capacity loss is in direct proportion to the variance of cells’ parameters and is proportional to the power function of cells’ number when cells’ parameters are normally distributed.