本工作以大容量磷酸铁锂电池储能电站为研究对象,立足于储能系统中处于工程场景的电池组日常运行数据。首先,根据电池运行数据分析表征锂离子电池电压、温度的一致性关键参量;其次,提取能够有效反映电池组一致性的评估特征;最后,基于此类特征,将储能电站的一致性分析分为两个层次,提出了针对站内大规模电池组的一致性分析算法以及基于DBSCAN (density-based spatial clustering of applications with noise) 聚类的电池组内异常单体电池筛选算法。结果表明,所提算法能够高效地获取储能电站运行数据中与电池组一致性相关的关键电气特征量,准确判断储能系统内电池组的一致性情况以及定位可能出现故障的单体电池。本研究有助于判断在工程场景中大规模电池组的一致性状态,并能及时准确地筛选出异常单体电池,保障储能电站电池组的安全性。
关键词:储能电站
;
锂离子电池
;
DBSCAN聚类算法
;
一致性评估
Abstract
This study takes a large-capacity power station of lithium iron phosphate battery energy storage as the research object, based on the daily operation data of battery packs in the engineering scene of energy storage systems. First, the key parameters characterizing the voltage and temperature consistency of Li-ion batteries were analyzed according to the operating data of the battery. Second, the evaluation features that can effectively reflect the battery pack consistency were extracted. Finally, based on such characteristics, the consistency analysis of the energy storage power station was divided into two levels, and the consistency analysis algorithm was proposed for large-scale battery packs in the station. Furthermore, a screening algorithm was proposed for abnormal cells in battery packs based on density-based spatial clustering of applications with noise (DBSCAN) clustering. The results showed that the proposed algorithm could efficiently obtain the key electrical characteristics related to the battery pack consistency in the operation data of the energy storage power station. Moreover, it could accurately judge the battery pack consistency in the energy storage system and locate the single battery that may fail. This study is helpful in judging the consistent state of large-scale battery packs in engineering scenarios. It can also timely and accurately screen out abnormal single batteries to ensure the battery packs' safety in energy storage power stations.
Keywords:energy storage power station
;
lithium-ion batteries
;
DBSCAN clustering algorithm
;
consistency evaluation
GAO Xin. Consistency evaluation method of battery pack in energy storage power station based on running data[J]. Energy Storage Science and Technology, 2023, 12(9): 2937-2945
DBSCAN (density-based spatial clustering of applications with noise)为具有噪声的基于密度的聚类方法,是一种基于密度的空间聚类算法[20]。该算法无须设置聚类个数,通过邻域半径和最少点数目的设置对聚类点的簇进行划分,将簇定义为密度相连点的最大集合。此方法的优点是可以在具有离群点的空间数据集中发现任意形状的簇。算法设置过程中需要设置两个参数,分别为邻域半径eps和最少点数目minpoints,算法流程如下:
WU Z Y, ZHOU M, WANG J X, et al. Review on market mechanism to enhance the flexibility of power system under the dual-carbon target[J]. Proceedings of the Chinese Society for Electrical Engineering, 2022, 42(21): 7746-7763.
WANG N, CHEN Z Q, LIU M Y, et al. Health status assessment of lithium-ion battery based on fuzzy comprehensive evaluation[J]. Power Generation Technology, 2022, 43(5): 784-791.
WANG L S. Consistency analysis model of lithium-ion power battery and simulation analysis of its influencing factors[D]. Beijing: General Research Institute for Nonferrous Metals, 2020.
SEQUINO L, VAGLIECO B M. Potential of infrared temperature measurements for the online estimation of the state-of-charge of a Li-polymer battery[J]. Journal of Energy Storage, 2021, 44: 103532.
YANG S J, LUO B Y, WANG J, et al. State of health estimation for lithium-ion batteries based on peak region feature parameters of incremental capacity curve[J]. Transactions of China Electrotechnical Society, 2021, 36(11): 2277-2287.
LIU Y L, MENG J H, PENG Q, et al. NSGA-Ⅱ genetic algorithm-based optimization of the lithium battery equalization index[J]. Energy Storage Science and Technology, 2023, 12(6): 1946-1956.
HUANG B S, ZHANG W. Influence of cell consistency on performance of power lithium battery[J]. Chinese Journal of Power Sources, 2018, 42(9): 1310-1311, 1320.
TIAN J Q, WANG Y J, LIU C, et al. Consistency evaluation and cluster analysis for lithium-ion battery pack in electric vehicles[J]. Energy, 2020, 194: 116944.
JIN W T, LI Y Y, JIA X C, et al. Research on comprehensive evaluation method of battery consistency in energy storage power station[J]. Electrical & Energy Management Technology, 2022(5): 23-28.
WANG S, YIN Z D, ZHENG Z, et al. Representation of influence factors for battery module consistency on discharge voltage curves[J]. Transactions of China Electrotechnical Society, 2020, 35(8): 1836-1847.
PAN Y, HAN X B, OUYANG M G, et al. Research on the detection algorithm for internal short circuits in lithium-ion batteries and its application to real operating data[J]. Energy Storage Science and Technology, 2023, 12(1): 198-208.
ZENG J B, ZHANG Y Y, ZHANG Z, et al. Identification of power battery voltage inconsistency faults in electric vehicles based on K-means++ clustering with dynamic k-values[J]. Scientia Sinica (Technologica), 2023, 53(1): 28-40.
CHENG Y Y. An improved K-means clustering algorithm based screening and recombination method of retired battery pack[J]. Energy Engineering, 2022, 42(4): 36-42, 55.
JIA J, HU X S, DENG Z W, et al. Data-driven comprehensive evaluation of lithium-ion battery state of health and abnormal battery screening[J]. Journal of Mechanical Engineering, 2021, 57(14): 141-149, 159.
LIN D, LI Z H, FENG Y B, et al. Research on inconsistency identification of lithium-ion battery pack based on operational data[C]//2021 4th IEEE International Conference on Industrial Cyber-Physical Systems (ICPS). May 10-12, 2021, Victoria, BC, Canada. IEEE, 2021: 498-502.
SANDER J, ESTER M, KRIEGEL H P, et al. Density-based clustering in spatial databases: The algorithm GDBSCAN and its applications[J]. Data Mining and Knowledge Discovery, 1998, 2(2): 169-194.
... DBSCAN (density-based spatial clustering of applications with noise)为具有噪声的基于密度的聚类方法,是一种基于密度的空间聚类算法[20].该算法无须设置聚类个数,通过邻域半径和最少点数目的设置对聚类点的簇进行划分,将簇定义为密度相连点的最大集合.此方法的优点是可以在具有离群点的空间数据集中发现任意形状的簇.算法设置过程中需要设置两个参数,分别为邻域半径eps和最少点数目minpoints,算法流程如下: ...