基于图计算的可重构电池网络能效提升与热安全管控  

Energy efficiency improvement and thermal safety control of reconfigurable battery networks based on graph computing

作　　者：党建[1] 罗永春 张旭[1] 肖逗 贾嵘[1] DANG Jian;LUO Yongchun;ZHANG Xu;XIAO Dou;JIA Rong(School of Electrical Engineering,Xi’an University of Technology,Xi’an 710054,China)

机构地区：[1]西安理工大学电气工程学院,陕西西安710054

出　　处：《实验技术与管理》2025年第1期82-89,共8页Experimental Technology and Management

基　　金：国家自然科学基金项目(52009106);陕西省重点研发计划项目(2024GX-YBXM-443)。

摘　　要：可重构电池系统凭借其能量信息耦合优势成为解决电化学储能“木桶效应”的一个有效手段。但其可用路径规划量会随着电池数目的增多而成倍增长,使得在路径选择时需要考虑不同拓扑结构和负载需求约束,从而使重构策略难以快速制定和应用。该文利用可重构电池网络与图的一致性,将电池网络动态控制问题转化为图的遍历寻优问题,提出了基于记忆化搜索的图深度优先遍历的重构电池网络能效提升及热安全管控策略,从而获取了满足拓扑约束条件的最优开关配置,并可排除温度异常的故障电池模组。最后,通过搭建可重构电池网络实验平台对该方法进行了验证。该研究结果能够为提高电池网络系统能效及安全性提供一定借鉴。[Objective]The“short-plate effect”of battery storage systems with fixed series and parallel connection topologies can lead to substantial reductions in the effective capacity and cycle life of such systems,as well as safety issues,such as localized overheating.A reconfigurable battery system has become one of the effective means to eliminate this effect of electrochemical energy storage by virtue of its advantage in energy–information coupling.However,the amount of available path planning grows exponentially with the number of batteries,resulting in the need to consider different topologies and load demand constraints in path selection,making it difficult to quickly develop and apply reconfiguration strategies.In this paper,using the consistency of reconfigurable battery networks with graphs,the dynamic control problem of battery networks is transformed into a graph traversal optimization search problem and a graph-computing-based energy efficiency improvement and thermal safety control strategy for reconfigurable battery networks is proposed.[Methods]This paper first constructs a reconfigurable battery network graph model with the battery as the vertex set,the switch as the edge set,and the battery charge state and temperature as the weighting set.Then,a reconfigured battery network configuration optimization and dynamic policy formulation method is proposed to obtain the optimal on–off mode and the battery network topology under the corresponding switching state using a graph depth-first traversal strategy based on memristive search.The graph depth-first traversal algorithm evaluates the state of charge value of each battery cell while taking into account their connection mode and order.It uses memristive search to avoid repeated calculations so as to plan a path that meets the current demand and maximizes the battery performance,improving the overall usable capacity and consistency of the battery pack.In addition,by dynamically reconfiguring the connections between battery cells to manage and disperse

关 键 词：可重构电池网络 路径规划 图模型 能效提升 热安全管控 

分 类 号：TM912[电气工程—电力电子与电力传动]