The invention provides a lithium ion battery pack dynamic inconsistency and health state evaluation method which comprises the following steps: processing acquired abnormal data, and segmenting continuously acquired data into segments to obtain an average voltage curve; after obtaining the segmented monomer voltage data and average voltage data, evaluating the similarity between the monomer voltage and the average voltage by using DTW (Dynamic Time Warning) to obtain the similarity between each monomer and the average voltage of the segment; after obtaining the similarity, estimating probability density distribution from the similarity of each monomer of each fragment, and performing probability density estimation by using a kernel function to obtain a probability density function; and after a probability density function is obtained, an inconsistency index is solved for each fragment, data fitting is carried out after abnormal values are removed by applying DBSCAN, and a final SOH change curve is obtained. The method is small in external interference, low in sampling precision requirement, high in robustness to abnormal data and better in adaptability to real vehicle data and cloud data.
本发明提供了一种锂离子电池组动态不一致性与健康状态评估方法,包括如下步骤:对采集的异常数据进行处理,将连续采集的数据切分为片段获得平均电压曲线;在获得切分后的单体电压数据与平均电压数据后,使用DTW对单体电压与平均电压间的相似度进行评估,获得每个单体与该片段平均电压的相似度;获得相似度后,从每个片段各单体的相似度中估计概率密度分布,使用核函数进行概率密度估计,获得概率密度函数;获得概率密度函数后,对每个片段求取不一致性指标,应用DBSCAN去除异常值后对数据进行拟合,获得最终的SOH变化曲线。本方法受外界干扰小、对采样精度要求低、对异常数据的鲁棒性强,对实车数据与云端数据的均具有更好的适应性。
Lithium ion battery pack dynamic inconsistency and health state evaluation method
一种锂离子电池组动态不一致性与健康状态评估方法
2024-01-30
Patent
Elektronische Ressource
Chinesisch
SOC Estimation of Battery Pack Considering Cell Inconsistency
SAE Technical Papers | 2019
|Parameter Identification of Battery Pack Considering Cell Inconsistency
SAE Technical Papers | 2017
|