A Method of Cell-to-cell Variation Evaluation for Battery Packs in Electric Vehicles with Charging Cloud Data

Yifan Lu,Kai Li,Xuebing Han,Xuning Feng,Zhengyu Chu,Languang Lu,Peifeng Huang,Zhi Zhang,Yongsheng Zhang,Fuqiang Yin,Xiao Wang,Feng Dai,Minggao Ouyang,Yuejiu Zheng

DOI: https://doi.org/10.1016/j.etran.2020.100077

IF: 11.9

2020-01-01

eTransportation

Abstract:During the use of electric vehicles (EVs), especially with the decay of battery, the cell-to-cell variations in Lithium-ion batteries increase. The cell-to-cell variations of power battery may lead to battery failure and cause safety problems. With the fast development of cloud data, EV data can be monitored on the cloud to evaluate the safety and cell-to-cell variations of EVs. Although the data are sampled with a high frequency in EVs, the recording frequency of the cloud data is relatively low. Thus, the charging data are more valuable and suitable to evaluate cell-to-cell variations of EVs. In this paper, a method based on charging cloud data is proposed to evaluate the cell-to-cell variations of lithium-ion batteries. 5 indicators, including variations of the voltage, temperature, internal resistance, capacity and electric quantity, are analyzed and evaluated by the original signals. Cell capacities and the electric quantities are achieved with the estimation of the remaining charging/discharging capacities by the charging voltage curves transformation. To comprehensively score the cell-to-cell variations of the battery pack, a weighted score mechanism is proposed. The weight factors are decided by the analytic hierarchy process based on the judgment matrix. Finally, experiments are carried out, and 3 packs are evaluated using the proposed scoring system. The results show the proposed method can effectively evaluate the cell-to-cell variations of battery packs.