汽车工程

汽车工程 ›› 2018, Vol. 40 ›› Issue (12): 1418-1425.doi: 10.19562/j.chinasae.qcgc.2018.012.007

• • 上一篇    下一篇

基于模拟退火算法的锂电池模型参数辨识*

罗勇1,2,3, 祁朋伟2, 阚英哲2, 李沛然1, 刘莉2, 崔环宇2   

  1. 1.中国汽车工程研究院股份有限公司,汽车噪声振动和安全技术国家重点实验室,重庆 400054;
    2.重庆理工大学,汽车零部件先进制造技术教育部重点实验室,重庆 400054;
    3.重庆青山工业有限责任公司技术中心,重庆 400054
  • 收稿日期:2017-11-13 出版日期:2018-12-25 发布日期:2018-12-25
  • 通讯作者: 祁朋伟,硕士研究生,E-mail:736386623@qq.com
  • 基金资助:
    国家自然科学基金(51305475)、汽车噪声振动和安全技术国家重点实验室2017年度开放基金(NVHSKL-201702)、重庆市教委科学技术研究项目(KJQN201801143)、重庆市重点产业共性关键技术创新专项(cstc2015zdcy-ztzx60013)和重庆市技术创新与应用示范专项产业类重点研发项目(cstc2018jszx-cyzdx0069)资助。

Parameter Identification of Lithium Battery Model Based on Simulated Annealing Algorithm

Luo Yong1,2,3, Qi Pengwei2, Kan Yingzhe2, Li Peiran1, Liu Li2, Cui Huanyu2   

  1. 1.China Automotive Engineering Research Institute Co., Ltd., State Key Laboratory of Vehicle NVH and Safety Technology, Chongqing 400054;
    2.Chongqing University of Technology, Key Laboratory of Advanced Manufacturing Technology for Automobile Parts, Chongqing 400054;
    3.Department of Technology, Chongqing Tsingshan Industrial Co., Ltd., Chongqing 400054
  • Received:2017-11-13 Online:2018-12-25 Published:2018-12-25

PDF (PC)

568

赞

摘要： 为建立准确的电池模型,以精确估算电池荷电状态,本文中通过试验获取锂电池在不同充放电电流和SOC下的充放电特性,进而对不同充放电电流和SOC下的电池R及C参数进行辨识。针对该参数辨识过程中参数初始值未知、数据处理量大、易陷入局部最优点的特点,采用无需初始参数、收敛速度快、能获得全局最优解的模拟退火算法对电池R及C参数进行辨识。仿真和试验结果表明,采用以上方法建立的电池模型具有较高精度。

关键词: 电动汽车, 电池管理系统, 参数辨识, 模拟退火算法

Abstract: In order to set up an accurate battery model for precisely estimating its state of charge (SOC), the charge and discharge characteristics of lithium battery under different currents and SOCs are obtained by test, and the parameters R and C of battery under the same conditions are identified in this paper. In view of the features of parameter identification, i.e. unknown initial values, massive data processing and being prone to fall into local optimum, simulated annealing algorithm, which can rapidly converged to the global optimum without initial parameters, is adopted to identify the parameters R and C of battery. Simulation and test results show that the battery model established by the above method has relatively high accuracy.

Key words: electric vehicle, battery management system, parameter identification, simulated annealing algorithm