【摘要】 针对汽车实际运行过程中,动力电池非满充满放工况下电池退化特征难以提取问题,以等充电压升时间、等放电压降时间为特征健康因子,构建特征健康因子与循环次数的时间序列。利用经验模态分解解耦全局退化和局部波动。同时提出将粒子滤波和多项式回归联合来预测电池剩余寿命,粒子滤波用来追踪局部波动现象,多项式回归用来拟合全局退化趋势。结果表明:本方法的预测结果与锂电池寿命实验数据误差在4%之内,说明所提出的方法容易实现且精度符合要求。更多还原

【Abstract】 For the actual operation of the car,the problem of battery degradation is difficult to extract when the power battery is not full-charged and full-discharged. The equal charging voltage rise time and the equal discharge voltage drop time are taken as the characteristic health factors and are constructed as a time series with cycle times. An empirical mode decomposition is proposed to decouple the global degradation and local fluctuation. At the same time,a combined algorithm based on particle filter and polynomial regression is proposed to predict battery cycle life. The particle filter model is used to track local fluctuations and the polynomial regression model is used to fit the global degradation trend. The experimental results show: The predicted results of the proposed method are in good agreement with the experimental data of lithium battery cycle life,and the error is within 4%. It is concluded that the method proposed in this paper is easy to implement and the accuracy also meets the requirements.更多还原