双碳背景下，光伏和风力等可再生能源存在间歇性和波动性， 其电力并网过程需要储能系统进行能量与功率调节。磷酸铁锂电池具有长循环寿命和高安全性， 被广泛应用在储能系统中。 然而， 磷酸铁锂电池的开路电压滞回特性导致储能电池电压的精准建模困难， 储能电池系统复杂串联并联构型的状态评估及不一致性控制效果较差， 成为储能系统安全高效运行的关键问题。 研究以一款商用 120Ah 磷酸铁锂电池串并联组成的 1MW/2.5MWh 储能电站系统为对象，分析了储能工况下磷酸铁锂电池的滞回特性， 建立了精准的电压模型；构建了面向于复杂串并联电池系统的健康状态和荷电状态估计方法；探索了并联环流在单循环和长循环间的演变规律， 提出了基于内阻均一的热管理方法， 抑制了系统环流和循环容量损失。 首先，针对于调频、削峰填谷和平抑发电功率波动三种储能工况，搭建了描述磷酸铁锂电池电压特征的二阶 RC 等效电路模型、一态滞回模型、滞回电压重构模型和神经网络模型，分析了不同储能工况下磷酸铁锂电池的滞回特性，及其对端电压仿真和荷电状态估计反馈的影响规律，对比了不同模型在端电压误差、 计算时间、模型复杂度和荷电状态估计准确度的性能表现， 获取了面向储能工况的磷酸铁锂电池优选电压模型。 其次，针对于储能电池串并联复杂构型的健康状态和荷电状态估计问题，利用两点法和压容率曲线变换法准确评估了储能电池系统内电池单元的容量。 选取电池簇内最小放电电量单元， 建立了电池滞回电压重构模型和基于扩展卡尔曼滤波的荷电状态估计方法， 解析了串并联电池组内电池单体差异对系统充放电容量的影响规律，准确评估了储能电池系统的荷电状态。 最后，针对于电池簇并联环流问题，搭建了电池簇并联系统的热-电-老化-环流模型，分析了磷酸铁锂电池内阻和容量差异影响的电池并联电流的分配特征， 挖掘了参数差异导致单循环内可逆容量损失以及长循环不可逆容量损失的影响机理，提出了以内阻均一为指标的并联电池簇热管理策略，实现了末端环流的抑制， 有效抑制了电池簇可逆与不可逆容量的损失。
In the context of dual carbon, renewable energy sources such as photovoltaic and wind power exhibit intermittency and volatility. The process of grid connection for these sources necessitates energy storage systems to regulate energy and power. Lithium iron phosphate batteries, known for their long cycle life and high safety, are widely employed in energy storage systems. However, the hysteresis characteristic of the open-circuit voltage of lithium iron phosphate batteries complicates precise modeling of storage battery voltage. Additionally, the state evaluation and inconsistent control effectiveness of complex series-parallel configurations in energy storage battery systems are poor, posing critical challenges to the safe and efficient operation of energy storage systems. The study focuses on a 1MW/2.5MWh energy storage station system composed of commercially 120Ah lithium iron phosphate batteries arranged in series-parallel configuration. It analyzes the hysteresis characteristics of lithium iron phosphate batteries under storage conditions and establishes an accurate voltage model.Furthermore, it constructs methods for estimating the health and state of charge of complex series-parallel battery systems. The study explores the evolution patterns of parallel circulation between single and long cycles and proposes a thermal management approach based on uniform internal resistance to suppress system circulation and cycle capacity losses. Firstly, focusing on three energy storage scenarios: frequency regulation, peak shaving, and power smoothing, we construct second-order RC equivalent circuit models, single-state hysteresis models, hysteresis voltage reconstruction models, and neural network models to describe the voltage characteristics of lithium iron phosphate batteries. We analyze the hysteresis characteristics of lithium iron phosphate batteries under different energy storage scenarios and their impact on terminal voltage simulation and state of charge estimation feedback. Comparative analysis of different models is conducted in terms of terminal voltage error, computation time, model complexity, and state of charge estimation accuracy. Finally, an optimized voltage model for lithium iron phosphate batteries tailored to energy storage scenarios is derived. Secondly, addressing the health and state of charge estimation challenges in complex series-parallel configurations of energy storage batteries, we utilize the two-point method and pressure-capacity curve transformation method to accurately assess the capacity of battery units within the energy storage battery system. By selecting the battery unit with the minimum discharge capacity within the battery cluster, we establish a hysteresis voltage reconstruction model and a state of charge estimation method based on extended Kalman filtering. We analyze the impact of individual battery differences within the series-parallel battery group on the system's capacity, thereby accurately evaluating the state of charge of the energy storage battery system. Finally, addressing the issue of parallel circulation within battery clusters, we have constructed a thermal-electric-aging-circulation model for the battery cluster parallel system. We analyze the distribution characteristics of the parallel currents of lithium iron phosphate batteries influenced by internal resistance and capacity differences. Furthermore, we delve into the mechanisms of reversible capacity loss within a single cycle and irreversible capacity loss over long cycles induced by parameter differences. We propose a thermal management strategy for parallel battery clusters based on uniform internal resistance, achieving the suppression of terminal circulation and effectively mitigating both reversible and irreversible capacity losses within the battery clusters.
