电动汽车用电池SOC估算方法研究

发布时间：2018-02-15 11:24

本文关键词： 电动车电池荷电状态 Thevenin模型参数识别无迹卡尔曼滤波　出处：《辽宁工业大学》2016年硕士论文　论文类型：学位论文

【摘要】：随着全球环境污染的日益严重以及世界各国不可再生能源的日益短缺,目前恶劣的环境问题给各个国家都来了不少的麻烦,由此引发世界各国对电动汽车的研究热潮。而电动汽车最重要的部件是电池,电池的核心技术是电池荷电状态(State of Charge,SOC)的估算;电池SOC估算的精准性越高,驾驶员可以对电池剩余状态有一个更好的了解,同时也可以提高电池自身的性能与寿命,对电动汽车的未来有着重大的意义与研究价值。因此,本文以电池SOC的估算为研究对象,主要工作如下:首先,介绍了电动汽车的发展与现状,引出动力电池是电动汽车的心脏,同时电池SOC估算是动力电池的核心技术,简单的介绍了电池SOC的估算现状与方法。其次,分析及对比目前市场上几种常用的动力电池,最终选取锂电池为本文的研究对象,介绍了锂电池的一些基本特性与工作原理,同时还介绍了对锂电池SOC影响的几种因素,如温度、老化、充放电倍率等;接着介绍了几种常用的动力电池SOC估算方法,诸如安时积分法、开路电压法、卡尔曼滤波法等,最终选取无迹卡尔曼滤波法为电池SOC的估算方法。再次,介绍了几种常用的电池模型,如Rint、Thevenin、PNGV等模型,分析这几种模型的优缺点,最后选取以Thevenin为基础,串联一个RC等效电路的二阶RC电池模型,通过实验确定了开路电压与电池SOC的对应关系,同时对改进后的模型进行参数识别与验证,确保改进后的模型可以较准确的模拟实际电池的状态。最后,介绍卡尔曼滤波法的原理与无迹卡尔曼滤波的理论,无迹卡尔曼滤波算法主要适用于非线性系统,恰好动力电池在实际使用过程中也是非线性的;它的主要原理是提出预估计值然后进行不断的循环更新及选取最优的估计结果,同时对误差有修正作用,更接近于理论值;将实验数据导入由MATLAB搭建的无迹卡尔曼滤波算法模型,对电池的SOC进行仿真与分析,将仿真结果与理论数值进行比较,得出无迹卡尔曼滤波算法对动力电池的估算有较高的准确性。
[Abstract]:With the increasing pollution of the global environment and the increasing shortage of non-renewable energy sources in various countries of the world, at present, the bad environmental problems have brought a lot of trouble to all countries. The most important component of an electric vehicle is a battery, the core of which is the estimation of the state of charge of the battery. The more accurate the battery SOC is, the more accurate the battery SOC is. The driver can have a better understanding of the remaining state of the battery, but also can improve the performance and life of the battery itself, which has great significance and research value for the future of electric vehicles. The main work of this paper is as follows: firstly, the development and present situation of electric vehicle are introduced, and the power battery is the heart of electric vehicle, and the SOC estimation of battery is the core technology of power battery. This paper briefly introduces the current situation and methods of estimating battery SOC. Secondly, it analyzes and compares several kinds of power batteries commonly used in the market at present. Finally, lithium battery is selected as the research object of this paper, and some basic characteristics and working principles of lithium battery are introduced. At the same time, several factors affecting the SOC of lithium battery are introduced, such as temperature, aging, charge / discharge rate, and so on, and then introduces several commonly used SOC estimation methods of power battery, such as an hour integration method, open circuit voltage method, Kalman filter method and so on. Finally, the unscented Kalman filter method is selected as the estimation method of battery SOC. Thirdly, several commonly used battery models, such as RintTevenin PNGV and so on, are introduced, the advantages and disadvantages of these models are analyzed, and finally, based on Thevenin, several battery models are selected. The second order RC battery model is connected with a RC equivalent circuit. The relationship between the open circuit voltage and the SOC of the battery is determined by experiments, and the parameters of the improved model are identified and verified. Finally, the principle of Kalman filter and the theory of unscented Kalman filter are introduced. The unscented Kalman filter algorithm is mainly suitable for nonlinear systems. The main principle of the battery is to put forward the pre-estimated value and then carry on the continuous cycle renewal and select the best estimation result, at the same time it has the correction function to the error, which is closer to the theoretical value. The experimental data are imported into the unscented Kalman filter algorithm model built by MATLAB, and the SOC of the battery is simulated and analyzed. The simulation results are compared with the theoretical values. It is concluded that the unscented Kalman filter algorithm has a high accuracy in estimating the power cell.
【学位授予单位】：辽宁工业大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U469.72

【相似文献】