Probabilistic design optimization of battery pack in considering the effect of external pressure with uncertainty

Abstract

Lithium-ion batteries have attracted wide attention as automotive applications with their advantages of high energy density and performance. The pouch type cell has been a preferred candidate based on their light weight, cost effectiveness, and design flexibility. However, due to the low mechanical stability, their characteristics are strongly influenced by environmental conditions. Especially, external pressure on cell surface directly affects the swelling phenomenon which is closely related to performance, life cycle and structural safety of the battery pack. In this paper, a novel framework for design optimization of battery pack is proposed to apply appropriate pressure on pouch cells. The effect of external pressure is investigated through cell cycling tests while thickness, pressure and capacity changes are measured. This investigation shows a recognizable correlation between pressure and cell degradation and also indicates the needs of certain pressure level which should be ensured by pack structures. The mechanical relation between cell and structural components is demonstrated in a free body diagram and utilized for deterministic analysis. To consider uncertainties in the external pressure formulation, the system is hierarchically decomposed and the uncertainty for each sub-component is analyzed. Then, the uncertainty propagation is conducted using Monte-Carlo Simulation to predict the distribution of external pressure in pack level. Based on the results, probabilistic design optimization is performed to minimize the weight of battery pack structure ensuring the external pressure range. The results of deterministic and probabilistic design optimization are compared. The deterministic analysis includes the safety factor based method and the arithmetic worst case based method. The probabilistic analysis is formulated for ensuring the minimum required pressure with the confidence level of 99.7%. After the pressure distribution analysis, the design of module and pack structure is modified to generate the appropriate pressure on the cell surface. The improved module and pack designs are verified by numerical simulations and tests. By adopting the probabilistic design optimization, it is expected that the life cycle reliability and the performance of battery pack can be improved. In addition, more than 17% of weight reduction can be achieved compared to conventional deterministic based design optimization. A novel framework for probabilistic design optimization of battery pack is proposed in considering the effect of external pressure on pouch cell. Various uncertainty factors in formulating external pressure are analyzed by using probabilistic method and compared with the deterministic method. This proposed design technique can be utilized for developing compatible automotive battery packs and improving reliability under uncertainty.