摘要
针对超级电容模型多参数辨识问题以及传统辨识算法收敛精度差,收敛速度慢问题,提出基于动态自学习粒子群算法的超级电容参数辨识方法。根据超级电容等效电路模型,采用双线性变换进行离散化获得辨识模型,使用动态自学习粒子群算法辨识各分支的参数。仿真结果分析表明,与基本粒子群、自适应惯性权重粒子群对比分析,基于动态自学习粒子群算法的超级电容参数辨识方法收敛速度快、收敛精度高、全局寻优能力强,可以更准确地反映出超级电容的动态特性。
In order to solve the problem of multi-parameter identification of super capacitor model and the problems of poor convergence accuracy and slow convergence speed of traditional identification algorithm,a parameter identification method based on dynamic self-learning particle swarm optimization(PSO)is proposed in this paper.According to the three-branch equivalent circuit model of super capacitor,the identification model is discretized by bilinear transformation,and the parameters of each branch are identified by dynamic self-learning particle swarm optimization.The simulation results show that,compared with the basic particle swarm and the adaptive inertial weighted particle swarm,the parameter identification method based on dynamic self-learning particle swarm algorithm has fast convergence speed,high convergence accuracy and strong global optimization ability,which can more accurately reflect the dynamic characteristics of super capacitors.
作者
刘继超
王维庆
王海云
Liu Jichao;Wang Weiqing;Wang Haiyun(Renewable Energy Power Generation and Grid Technology,Engineering Research Center of Ministry of Education,Xinjiang University,Urumqi 830047,China)
出处
《电测与仪表》
北大核心
2021年第6期33-38,共6页
Electrical Measurement & Instrumentation
基金
国家自然科学基金资助项目(51667020)
新疆自治区万人计划后备人选项目—"天山雪松"人才培养计划(2017XS02)
新疆自治区重点实验室开放课题(2018D04005)
教育部创新团队项目(IRT_16R63)。
关键词
超级电容
三分之等效电路模型
双线性变换
动态自学习粒子群
super capacitor
three-branch equivalent circuit model
bilinear transformation
dynamic self-learning particle swarm algorithm
作者简介
刘继超(1992-),男,河南人,硕士研究生,从事混合储能的应用研究。Email:2428228052@qq.com;王维庆(1959-),男,新疆乌鲁木齐人,博士生导师,从事风力发电及并网技术的研究。Email:3064486275@qq.com;王海云(1973-),女,新疆乌鲁木齐人,硕士生导师,教授,从事风电并网技术和高压直流输电与电力系统稳定性等方面的研究。Email:327028229@qq.com。
