AGV车电池SOC估算算法研究与实现

发布时间：2018-03-11 15:21

本文选题：自动引导车　切入点：电池剩余容量　出处：《湖北工业大学》2017年硕士论文　论文类型：学位论文

【摘要】：自动引导车(Automated Guided Vehicle,AGV车)是一种自动化的搬运设备,一般由蓄电池提供动力,能够按照既定的路线搬运物料,是现代工业生产线必不可少的搬运工具。AGV车工况特殊,实时准确估算其电池SOC值难度较大。常用的安时积分法在特殊工况下准确度较低,无法满足工业生产需求。本文提出一种改进扩展卡尔曼滤波算法,使用该方法估算AGV车电池SOC值可以将误差控制在5%以内。具体研究内容如下:第一:详细分析了AGV车实际运行工况,明确其充电电流大,放电电流小,充电时间短,充电频率高的工况特点。第二:建立Thevenin电池等效模型,使用扩展卡尔曼滤波法估算AGV车电池SOC值,相比安时积分法估算精度有所提高,但是传统扩展卡尔曼滤波法在AGV车特殊工况下跟踪效果差,由此带来了较大的估算误差。第三:针对传统扩展卡尔曼滤波法估算AGV车电池SOC值跟踪效果差的问题,提出改进扩展卡尔曼滤波算法,将扩展卡尔曼滤波法的滤波增益改进为动态调整滤波增益,提高扩展卡尔曼滤波法在特殊工况下的跟踪效果。第四:通过编程读取AGV车实际运行数据来模拟其工况,进而分析扩展卡尔曼滤波法估算AGV车电池SOC值的效果,验证改进扩展卡尔曼滤波算法的有效性。实验表明扩展卡尔曼滤波法相对安时积分法估算精度较高,采用动态校正的滤波增益提高了估算过程的跟踪效果,解决了AGV车特殊工况下SOC估算不准确的问题,将AGV车电池SOC值误差控制在5%以内。同时针对本算法进行系统硬件和软件设计,并进行了实验验证。鉴于AGV车与电动汽车有很多相似之处,特别是随着电动汽车的发展其工况越来越复杂,如何提高复杂工况下电池SOC值估算精度具有较大的研究意义,因此本文研究成果有一定的推广意义。
[Abstract]:Automated Guided vehicle (AGV) is a kind of automatic handling equipment, which is generally powered by batteries and can carry materials according to the established route. It is a necessary handling tool for modern industrial production line. It is difficult to estimate the SOC value of the battery in real time and accurately, and the common amp-hour integration method can not meet the demand of industrial production because of its low accuracy under special working conditions. In this paper, an improved extended Kalman filter algorithm is proposed. Using this method to estimate the SOC value of AGV vehicle battery, the error can be controlled within 5%. The specific research contents are as follows: first, the actual operating conditions of AGV vehicle are analyzed in detail, and it is clear that the charging current is large, the discharge current is small, and the charging time is short. Second, the equivalent model of Thevenin battery is established, and the SOC value of AGV vehicle battery is estimated by using extended Kalman filter method. However, the traditional extended Kalman filter method has poor tracking effect under the special working condition of AGV vehicle, which brings great estimation error. Third, aiming at the problem of poor tracking effect of traditional extended Kalman filter method in estimating SOC value of AGV vehicle battery, An improved extended Kalman filter (EKF) algorithm is proposed, in which the filter gain of the EKF is improved to dynamically adjust the filter gain. Improve the tracking effect of extended Kalman filter method under special working conditions. 4th: read the actual operation data of AGV vehicle by programming to simulate its working condition, and then analyze the effect of extended Kalman filter method to estimate the SOC value of AGV vehicle battery. The experimental results show that the extended Kalman filter method has higher estimation accuracy than the Anchorage integration method, and the tracking effect of the estimation process is improved by using the dynamic correction filter gain. The problem of inaccurate SOC estimation under special working conditions of AGV vehicle is solved, and the error of SOC value of AGV vehicle battery is controlled within 5%. At the same time, the system hardware and software are designed for this algorithm. In view of the similarities between AGV vehicle and electric vehicle, especially with the development of electric vehicle, it is of great significance to study how to improve the estimation accuracy of battery SOC value under complex operating conditions, especially with the development of electric vehicle. Therefore, the research results of this paper have certain popularization significance.
【学位授予单位】：湖北工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP23

【参考文献】