锂电池双向均衡及模型预测控制优化  

Bidirectional Balancing and Model Predictive Control Optimization for Lithium Batteries

作　　者：赵炎锋 王斌[1] 杜睿 徐世福 顾卫平 王朝晖[1] ZHAO Yanfeng;WANG Bin;DU Rui;XU Shifu;GU Weiping;WANG Chaohui(School of Mechanical Engineering,Xi’an Jiaotong University,Xi’an 710049,China;Shaanxi Dexin Intelligent Technology Co.,Ltd.,Xi’an 710075,China)

机构地区：[1]西安交通大学机械工程学院,西安710049 [2]陕西德鑫智能科技有限公司,西安710075

出　　处：《西安交通大学学报》2024年第10期24-35,共12页Journal of Xi'an Jiaotong University

基　　金：陕西省重点研发计划资助项目(2023-YBGY-376);国家自然科学基金资助项目(51907160)。

摘　　要：为了解决锂电池组内部单体不一致导致的能量利用率低、电池容量衰减等问题,提出了一种基于模型预测控制的双向均衡控制方法。首先,以锂电池单体的荷电状态(SOC)为均衡控制目标,采用带遗忘因子的最小二乘法和扩展卡尔曼滤波法联合估计电池SOC;接着,通过设计双向反激式电路,形成可升降压的均衡结构,并采用模型预测控制优化双向反激式电路的正反向均衡控制状态;最后,通过占空比控制双向反激式电路的均衡电流,实现不同锂电池单体之间的SOC均衡。研究结果表明:采用双向反激式电路和模型预测控制,不但降低了均衡电路控制的复杂程度,避免了传统均衡方法的电池组内部均衡消耗,而且有效提高了均衡可靠性和均衡速度;在动态工况下,基于模型预测控制的锂电池双向均衡控制方法,能快速实现锂电池单体之间的SOC均衡,与传统的比例均衡控制方法相比,均衡时间缩短了68.3%。To address issues such as low energy efficiency and battery capacity degradation caused by internal cell inconsistency in lithium-ion battery packs,a bidirectional balancing control method based on model predictive control(MPC)is proposed.Firstly,the state of charge(SOC)of individual lithium-ion battery cells is estimated using a combination of the least squares method with a forgetting factor and the extended Kalman filter,with the SOC of each battery cell being considered as the target for balancing control.Next,a bidirectional buck-boost circuit is designed to create a balancing structure capable of both boosting and bucking voltages.Model predictive control is then employed to optimize the balancing control states of the bidirectional circuit in both directions.Finally,by controlling the duty cycle of the bidirectional circuit,the balancing currents among different lithium-ion battery cells are adjusted to achieve SOC balancing.The research results demonstrate that the adoption of the bidirectional circuit and model predictive control offers several advantages.It reduces the complexity of balancing circuit control,avoids energy losses associated with traditional balancing methods,and improves the reliability and speed of balancing.Under dynamic conditions,the proposed bidirectional balancing control method based on model predictive control achieves rapid SOC balancing among lithium-ion battery cells.Compared with traditional proportional balancing control methods,the proposed method reduces the balancing time by 68.3%.

关 键 词：锂电池 模型预测控制 荷电状态 反激式电路 双向均衡控制 

分 类 号：TM910[电气工程—电力电子与电力传动]