基于改进最小二乘法的超级电容器特征参数辨识  

Characteristic Parameters Identification of Supercapacitor Based on Improved Least Square Method

作　　者：赵洋[1,2] 孙家南[2] 姜鸣[1] 

机构地区：[1]东莞理工学院电子工程学院,广东东莞523808 [2]同济大学电子与信息工程学院,上海201804

出　　处：《高压电器》2016年第5期199-204,共6页High Voltage Apparatus

基　　金：国家自然科学基金项目资助(51407031);东莞市社会科技发展项目资助(2013108101007)~~

摘　　要：准确辨识超级电容器特征参数值是分析超级电容器电气特性和估计其健康状态的重要研究基础。为克服采用阻抗谱分析法辨识特征参数所需激励信号幅值过小,且不能反映研究对象真实工况以及难以实现在线辨识特征参数的缺点,文中采用递推最小二乘法解决此问题。基于实验分析结果发现直接采用该算法辨识出的参数值与真实值差异较大,因此在深入分析辨识算法的基础上,对最小二乘法的辨识结构加以改进。实验和仿真分析结果表明,基于改进后的最小二乘法辨识得到的超级电容器特征参数值更加准确和可信。Identifying the super capacitor＇s characteristic parameters correctly is the key research basis for analyzing the super capacitor＇s electrical performance and estimating its state of health. In order to overcome the impedance spectroscopy method＇s disadvantages, such as its stimulating signal amplitude is too low to depict the research object＇s real working condition, and it is hard to realize online parameter identification, the recursive least square method is adopted in the paper. Based on the test results, large differences between the identified parameters＇ values and its real values can be observed. So the identification structure of the method on the basis of deep analysis of the least square method is imoroved. The experimental and simulation results show that,based on the improved recursive least square method,the characteristic parameter identification values are more correct and credible.

关 键 词：超级电容器 改进最小二乘法 特征参数 等效电路模型 参数辨识 

分 类 号：TM53[电气工程—电器]