基于中心差分卡尔曼滤波的动力电池SOC估算研究

发布时间：2018-05-11 14:19

本文选题：磷酸铁锂电池 + 等效电路模型　；参考：《吉林大学》2014年硕士论文

【摘要】：进入20世纪后，能源危机与环境污染等问题在全球范围内表现的愈加突出，在此形势下，新能源汽车尤其是电动汽车，得到了世界各国主要汽车厂商的高度关注。电池管理系统（BMS）作为电动汽车的重要组成部分，可以对动力电池进行有效管理和控制，保障电池高效使用及行车安全。电池管理技术仍处于发展期，很多技术并不成熟，而其中研究的重点和难点便是如何提高电池荷电状态（SOC）的估算精度。本文的研究对象选定为磷酸铁锂电池，从电池模型和基于模型的SOC估算方法这两个方面进行深入研究。首先，介绍了锂电池的基本原理和主要参数，通过对电池一系列的试验分析了电池的开路电压与荷电状态关系、欧姆内阻及容量等基本特性。以此为基础，结合常用电路模型的优缺点对比，提出考虑电池容量时变性的二阶RC等效电路模型，并采用指数拟合方法，在Matlab软件中得到模型参数初值。为了更好的反应电池特性，本文利用Matlab/Simulink对参数初值进行了修正，并在各SOC点处估算出电池模型参数。实验结果表明修正参数后的等效电路模型提高了跟踪电池电压变化的精度。其次，分析了传统SOC估算方法，结合磷酸铁锂电池的非线性特性和二阶RC等效电路模型确立了磷酸铁锂电池的状态空间方程。在扩展卡尔曼滤波算法对非线性状态方程估算精度有限基础上，提出了中心差分卡尔曼滤波算法。仿真结果表明中心差分卡尔曼滤波算法在同条件下对SOC的估算精度优于拓展卡尔曼滤波算法。最后，利用AVL Cruise软件搭建电动汽车整车模型，并在模拟城市道路工况下进行仿真实验，，得到了电池在工况下的仿真数据。通过Cruise软件与Matlab的接口，将动力电池组的仿真数据输入到估计模型中，利用中心差分卡尔曼滤波算法对SOC进行估算，并与拓展卡尔曼滤波算法对比，结果表明基于中心差分卡尔曼滤波算法对整车SOC估算具有抗干扰性、收敛性与更高估算精度。
[Abstract]:After entering the 20th century, the problems of energy crisis and environmental pollution have become more and more prominent in the world. Under this situation, new energy vehicles, especially electric vehicles, have been highly concerned by the major automobile manufacturers in the world. As an important part of electric vehicle, Battery Management system (BMS) can effectively manage and control the power battery and ensure the efficient use of the battery and the safety of driving. The battery management technology is still in the developing stage, and many technologies are not mature, and the emphasis and difficulty of the research is how to improve the estimation accuracy of SOC. The research object of this paper is lithium iron phosphate battery, which is studied from two aspects: battery model and SOC estimation method based on model. Firstly, the basic principle and main parameters of lithium battery are introduced. The relationship between open circuit voltage and charge state, ohmic internal resistance and capacity are analyzed by a series of experiments. Based on this, combined with the comparison of the advantages and disadvantages of the common circuit models, a second-order RC equivalent circuit model considering the time-varying capacity of the battery is proposed, and the initial values of the model parameters are obtained by using the exponential fitting method in the Matlab software. In order to improve the characteristics of the reaction cell, the initial parameters of the cell were modified by Matlab/Simulink and the parameters of the model were estimated at each SOC point. The experimental results show that the precision of tracking the voltage change of the battery is improved by the equivalent circuit model. Secondly, the traditional SOC estimation method is analyzed, and the state space equation of the lithium iron phosphate battery is established by combining the nonlinear characteristics of the lithium iron phosphate battery and the second-order RC equivalent circuit model. Based on the limited estimation accuracy of the extended Kalman filtering algorithm for nonlinear state equations, a central differential Kalman filter algorithm is proposed. The simulation results show that the SOC estimation accuracy of the central differential Kalman filter algorithm is better than that of the extended Kalman filter algorithm under the same conditions. Finally, the AVL Cruise software is used to build the whole vehicle model of electric vehicle, and the simulation experiment is carried out under the simulation of the urban road condition, and the simulation data of the battery under the working condition are obtained. Through the interface between Cruise software and Matlab, the simulation data of power battery pack is input into the estimation model, and the SOC is estimated by using the central differential Kalman filter algorithm, and compared with the extended Kalman filter algorithm. The results show that the central differential Kalman filter algorithm has the advantages of anti-jamming, convergence and higher estimation accuracy for vehicle SOC estimation.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM912;U469.72

【参考文献】