基于电池组多维一致性演变机制与估计方法的均衡构型和算法研究

Large-scale battery energy storage systems require battery packs with hundreds of cells connected in series. A distinguishable feature between the battery packs and single cells is the cell consistency. This feature becomes more significant for the long-term used and second life battery packs. In view of the degradation problems of battery power and energy density, as well as the shortened lifespan caused by cell consistency, the project intends to develop multidimensional coordinated equalization algorithms for battery packs through the multidimensional consistency evolution mechanism and estimation method of battery packs with the novel equalization configuration. Firstly, we try to establish a three-dimensional consistency of the capacity, power and internal resistance evolution model for series connected battery packs to reveal the dominant factors of the three-dimensional consistency evolution. Secondly, an online state estimation method for three-dimensional consistency of the battery pack based on multi-model error features is expected to be proposed, which would support the development of equalization algorithm. Thirdly, a novel equalization configuration by increasing the degree of freedom that can be used for three-dimensional equalization is to be developed. Finally, the fuzzy equalization algorithm is under consideration with the error estimation of the polytope of matrices under the condition of the imprecise state estimation for the three-dimensional states. We plan to use the additional degree of freedom for the energy management with the novel equalization topology to change the surface constraints under the combined effect of battery energy and power density, and to realize a novel equalization algorithm optimized for energy and power density. The findings are expected to rich the theory of battery pack consistency and enhance the equalization technology.

大规模电池储能需要上百个单体串联组成电池组，其区别于单体的重要特征是一致性问题，而长期使用的电池组和退役电池在一致性问题上更为显著。针对一致性导致的电池组功率和能量密度下降、寿命缩短的问题，本项目拟通过电池组多维一致性演变机制和估计方法，结合新型均衡构型，开展电池组多维协同均衡算法的研究。首先，拟通过建立串联电池组容量、电量和内阻三维一致性演变模型，揭示三维一致性演变的主导因素；其次，拟提出基于多模型误差特征的电池组三维一致性状态在线融合估计方法，支持电池组均衡算法的开发；第三，拟开发可用于三维均衡的增能量管理自由度新型均衡构型；最后，研究非精准估计状态下基于三维状态误差多胞型矩阵估计的模糊均衡算法，拟利用新型电池均衡系统的额外能量管理自由度，改变电池组能量和功率密度共同作用下的曲面约束，实现以能量和功率密度最优化的新型均衡算法。研究结果将丰富电池组一致性理论，提升电池组均衡技术。

限于单体锂电池的电压和容量，电动汽车必须将成百个的单体电池串并联形成电池组，由于制造工艺的不一致和使用过程环境的不一致，单体间总是存在无法消除的不一致性。单体在成组后，电池组的能量密度、耐久性和安全性等关键性能都会因为单体间的不一致而下降。本项目拟通过电池组多维一致性演变机制和估计方法，结合新型均衡构型，开展电池组多维协同均衡算法的研究。本研究有助于丰富电池组多维一致性的理论认识，同时可从提高电池组寿命的角度指导电池组的均衡设计。.经过项目预研和四年持续研究，本项目建立了基于多参数不一致性的电池组演变模型，揭示了多维一致性演变的主导因素，从单体容量、电量和内阻三维一致性演变的新角度探索电池组能量和功率密度下降的原因。其次项目进一步提出三维演变过程中电池组能量和功率密度共同约束下的电池单体状态空间表达新方法，为电池能量存储的一致性状态表达提供新方法。再次，项目结合云端大数据进一步研究提出不同容量估计方法的误差特征，结合寿命经验模型容量估计误差特征的连续性和累积性，充电曲线特征模型容量估计误差特征的间歇性和精确性，以及电池组分频模型容量估计误差特征的波动性，提出基于误差特征的容量在线融合估计新方法，并给出了基于云端充电数据的电池组一致性评价方法。项目研究了模组协同均衡的软硬件设计，实现了电池模组间的协同均衡，为电池组的一致性提升提供了重要保障。.在本项目研究过程中，共发表以标注有项目号的论文52篇，其中SCI论文43篇，国内期刊论文9篇，其中第一标注号的论文43篇。项目资助期间申请国家发明专利获得授权37项。项目负责人多次出席国际国内会议。在本项目的支持下，培养毕业硕士10人。

内容获取失败，请点击重试