The battery is one of the most important parts of electric vehicles (EVs), and the establishment of an accurate battery state estimation model is of great significance to improve the management strategy of EVs. However, the battery degrades with the operation of EVs, which brings great difficulties for the battery modeling issue. This paper proposes a novel aging phenomenon considered vehicle battery modeling method by utilizing the cloud battery data. First of all, based on the Rain-flow cycle counting (RCC) algorithm, a battery aging trajectory extraction method is developed to quantify the battery degradation phenomenon and generate the aging index for the cloud battery data. Then, the deep learning algorithm is employed to mine the aging features of the battery, and based on the mined aging features, an aging phenomenon considered battery model is established. The actual operation data of electric buses in Zhengzhou is used to validate the practical performance of the proposed methodologies. The results show that the proposed modeling method can simulate the characteristic of the battery accurately. The terminal voltage and SoC estimation error can be limited within 2.17% and 1.08%, respectively.

电池是电动汽车最重要的组成部分之一，建立准确的电池状态估计模型对改善电动汽车的管理策略具有重要意义。然而，电池随着电动汽车的运行而退化，这给电池建模问题带来了很大的困难。本文提出了一种利用云电池数据考虑车辆电池建模方法的新型老化现象。首先，基于雨流循环计数（RCC）算法，开发了一种电池老化轨迹提取方法，以量化电池退化现象并生成云电池数据的老化指标。然后，采用深度学习算法来挖掘电池的老化特征，并基于提取的老化特征，建立了考虑电池模型的老化现象。郑州电动公交车的实际运行数据被用来验证所提出方法的实际性能。结果表明，所提出的建模方法可以准确地模拟电池的特性。端子电压和SoC估计误差可以分别限制在2.17％和1.08％之内。