基于粒子群优化的超级电容器模型结构与参数辨识

Structure and Parameter Identification of Supercapacitors Based on Particle Swarm Optimization

为了分析超级电容器的动态特性,并准确估计荷电状态、健康状态等信息,需要建立超级电容器模型。提出以系统辨识方法作为建模手段,为克服广义误差准则不能保证模型输出误差最小的缺点,采用输出误差准则并推导出对应的非线性目标函数。应用粒子群优化算法对该目标函数进行优化求解,并获得模型参数。针对结构辨识问题,采用最终输出误差准则计算结构判别指标,通过比较该指标值确定模型最佳阶次。实验和仿真结果表明,所提模型能精确描述超级电容器的动态特性,建模方法可行。

In order to analyze the dynamical characteristics of supercapacitor and predict its state of charge （SOC） and state of health （SOH） correctly, the model of supercapacitor should be built at first. On the basis of analyzing the characteristics of different modeling methods, a system identification method was presented to model the supercapacitor. Because the general error criterion cannot insure the minimal output error, the output error criterion was adopted and a corresponding nonlinear objective function was deduced. Then the particle swarm optimization algorithm was used to solve the problem, and then the model＇s parameters were obtained. With respect to the structure identification, the final output error （FOE） criterion was introduced. By comparing the value of FOE, the best order of the model was determined. The results of experiments and simulations show that the model presented in the paper can describe the supercapacitor＇s dynamic characteristics precisely and the modeling method is feasible and valid.