摘要
在大量充放电模拟试验和随车试验数据采集的基础上,构建了基于数据信号处理器(DSP)芯片TMS320C2812的电池管理系统,实现了数据监测、荷电状态(SOC)估计、控制局域网(CAN)通信及USB存储等功能.在SOC估计算法上,根据电池所处状态进行了分类分析,并对估算难度最大的电池动态放电状态的算法进行了仿真实验.实验结果表明,该算法对镍氢电池的SOC能进行准确预测,并具有较高的精度.
According to the data obtained from charge/discharge experiments and on-site data from a running Electric Vehicle（EV）, a battery management system has been developed on the basis of 32-bits DSP controller TMS320C2812. The functions of the system include data monitoring, State of Charge（SOC） estimation, Controller Area Network（CAN） communications, USB-based data storage and so on. This paper also proposed an improved SOC estimation method, which divides into three algorithms according to the state of batteries. In the algorithm of the dynamic discharge state, Ampere/hours method is integrated with RBF neural network, and its accuracy and practicality are validated via Matlab and Advisor simulation.
出处
《湖南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2009年第5期33-36,共4页
Journal of Hunan University:Natural Sciences
基金
湖南大学“211工程”基金项目资助
关键词
电动汽车
电池管理系统
数据信号处理器
荷电状态
electric vehicles
battery management systems
Digital Signal Processor（DSP）
SOC
