基于无迹卡尔曼滤波算法的动力电池剩余电量估算

发布时间：2018-11-09 07:39

【摘要】：近几年,电动汽车凭借其清洁、高效、无污染等优点成为城市交通实现低排放甚至零排放的理想交通工具。电动汽车所搭载的电池组性能的好坏是影响整车的续航里程、加速性能和制动能量回收的效率等性能的直接因素。锂电池荷电状态(State of Charge,SoC)估算是电池管理系统(Battery Management System,BMS)的关键功能,是电池的使用可靠性以及安全性的关键。本文研究内容主要包括以下几个方面:首先,本文介绍了锂离子电池的结构和工作原理,阐述了荷电状态(SoC)的定义,比较分析了几种常用的估算方法,然后给出了本文采用的估算方法——卡尔曼滤波法。然后,为了提高SoC的估算精度,并且准确的对电池的特性以及工作过程中的状态及行为做出仿真和模拟。在便于硬件实现的前提下,建立了Thevenin模型和二阶RC电路相结合的电池模型,并且采用电池HPPC实验数据对模型参数进行辨识,并且通过搭建的电池模型对于参数辨识结果进行了验证,采用了遗忘因子递推最小二乘法对于模型参数进行在线辨识,结合离线辨识结果对照分析可知,此方法可有效地提高模型参数的准确性。最后,采用无迹卡尔曼滤波算法以及自适应匹配的噪声对电池的荷电状态(SoC)进行估算,并且将仿真估算结果与实际的测量结果进行比较分析,在恒流放电工况下验证了本文估算方法的有效性。
[Abstract]:In recent years, electric vehicles (EVs) have become the ideal vehicle to realize low emission and zero emission due to their advantages of cleanliness, high efficiency, no pollution and so on. The performance of battery pack on electric vehicle is a direct factor affecting the performance of the whole vehicle, such as the range of the vehicle, the acceleration performance and the efficiency of braking energy recovery. The estimation of charge state (State of Charge,SoC is the key function of the battery management system (Battery Management System,BMS) and the key to the reliability and safety of the battery. The main contents of this paper are as follows: firstly, the structure and working principle of lithium ion battery are introduced, the definition of charged state (SoC) is expounded, and several commonly used estimation methods are compared and analyzed. Then the Kalman filter method, which is used in this paper, is given. Then, in order to improve the accuracy of SoC estimation, and to accurately simulate and simulate the characteristics of the battery and the state and behavior in the working process. On the premise of easy hardware implementation, the battery model combining Thevenin model with second-order RC circuit is established, and the parameters of the model are identified by using the battery HPPC experimental data. The battery model is used to verify the parameter identification results, and the forgetting factor recursive least square method is used to identify the model parameters online. The results of off-line identification are compared and analyzed. This method can effectively improve the accuracy of model parameters. Finally, the unscented Kalman filter algorithm and adaptive matching noise are used to estimate the charge state (SoC) of the battery, and the simulation results are compared with the actual measurement results. The effectiveness of the proposed method is verified under constant current discharge conditions.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U469.72

【参考文献】