SIA OpenIR  > 装备制造技术研究室
面向水下机器人耐压型电池管理系统设计研究
其他题名Research on Design of Pressure-Resistant Battery Management System for Unmanned Underwater Vechicle
李晓鹏1,2
导师袁学庆
分类号TM912
关键词考虑温度影响 电池管理系统 耐压型 状态参数估计 算法对比研究
索取号TM912/L35/2018
页数83页
学位专业控制工程
学位名称硕士
2018-05-17
学位授予单位中国科学院沈阳自动化研究所
学位授予地点沈阳
作者部门装备制造技术研究室
摘要

本文的内容可分为两大部分,第一部分内容为在对锂电池考虑温度的前提下,建立了较为精确的锂电池等效电路模型,并阐述了三种匹配锂电池等效电路模型参数辨识方法,通过对比三种辨识方法发现,三者中卡尔曼(Kalman Fliter KF)辨识算法在初值给定不合理时会出现收敛速度慢的情况,而渐消记忆最小二乘法(FMRLS Fading Memory Recursive Least Squares FMRLS)对噪声敏感,所辨识出的结果抖动幅度较大,并且这两种算法都需要将所建立的模型转化成离散的差分方程,推导过程繁琐,因此,文章最终采用了随机性和确定性混型状态子空间辨识(Combined deterministic-Stochastic Subspace Identification CD-SSI)法,成功的确定出了不同温度下的锂电池各种模型的系统参数,在此基础上,本文进一步利用了安时(Ampere Hour AH)积分、扩展卡尔曼滤波(Extended Kalman Filter EKF)、中心差分卡尔曼滤波( Central Difference Kalman Filter CDKF)三种算法对不同模型在不同温度下实现了荷电状态(State Of charge SOC)的对比预测,得出在使用同一种模型的前提下,CDKF算法综合性能最高,估计效果在可接受的范围之内。在较为准确估计SOC的前提下,本文进一步实现对锂电池功率边界(State Of Power SOP)的估计,为UUV提供了重要的数据支撑,除此之外,本文还研究了EKF和CDKF的时间算法复杂度,结果表明在同样的条件下CDKF的浮点算法时间复杂度大约是EKF的两倍左右,这个结论对于CDKF在嵌入式系统中的应用研究有着重要的参考意义。文章的第二部分内容为PRBMS硬件系统设计研究,该部分是本文另一大重点内容,首先依据计算机仿真分析的结果,确定水压敏感性器件和非水压敏感性器件;其次,对水压敏感性器件采用有限元仿真分析,确定最大受力概况,着重对水压敏感型电子元器件进行了封装处理和计算机仿真优化设计;最后,设计系统硬件并进行综合打压实验,加压试验结果表明,BMS在105MPa 打压测试后运行良好,系统通讯、检测、均衡等功能均正常。上述结论、仿真及试验方法对PRBMS在深海中的应用具有重要的参考意义。

其他摘要

The content of this paper can be divided into two parts. The first part is to establish a more accurate lithium battery equivalent circuit model under the premise of considering the temperature of lithium battery, And elaborated three matching lithium battery equivalent circuit model parameter identification method. By comparing the three identification methods, it is found that the kalman Filter identification algorithm will slow down if the initial value is given unreasonably. Fading Memory Recursive Least Squares is sensitive to noise, and the identified results have large jitter amplitudes. Both of these algorithms need to convert the established model into discrete ones. The difference equation and the derivation process are cumbersome. Therefore, the paper finally adopted a combined deterministic-stochastic subspace identification method to successfully determine the system of various models of lithium batteries under different temperatures. On this basis, the paper further uses Ampere Hour , Extended Kalman Filter EKF and Central Difference Kalman Filter three different algorithms achieves state of charge estimate at different temperatures. The research results show that the CDKF algorithm has the highest overall performance and the estimated effect is within the acceptable range. Under the premise of accurately estimating the SOC, this paper realizes estimation the state of power of lithium. The SOP provides important data support for UUV. In addition, the time complexity of the EKF and CDKF is also studied in this paper.The result show in the case of same condition, the time complexity of the CDKF algorithm is about twice that of EKF. This result has important reference significance for the application of CDKF in embedded systems. The second part of the article is about the PRBMS hardware system design and research. This part is another major content of this article. First, based on the results of computer simulation analysis, determine the water pressure sensitive device and non-water pressure sensitive device; secondly, the water pressure Sensitive devices use finite element simulation analysis to determine the maximum force profile, focusing on water pressure sensitive electronic components packaged and computer simulation optimization design, and finally design system hardware, comprehensive pressure test, pressurization test results show that The BMS worked well after the 105MPa pressure test, and the system communication, detection, and balance functions were all normal. The above conclusions, simulations and test methods have important reference significance for the application of PRBMS in deep sea.

语种中文
产权排序1
文献类型学位论文
条目标识符http://ir.sia.cn/handle/173321/21782
专题装备制造技术研究室
作者单位1.中国科学院沈阳自动化研究所
2.中国科学院大学
推荐引用方式
GB/T 7714
李晓鹏. 面向水下机器人耐压型电池管理系统设计研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2018.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
面向水下机器人耐压型电池管理系统设计研究(4800KB)学位论文 开放获取CC BY-NC-SA请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[李晓鹏]的文章
百度学术
百度学术中相似的文章
[李晓鹏]的文章
必应学术
必应学术中相似的文章
[李晓鹏]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。