超级电容器的老化趋势分析

Aging tendency Investigation of supercapacitor

超级电容器的老化状态直接影响着储能系统的寿命。通过预测超级电容器性能参数的老化趋势,为系统的控制和管理提供预测性维护信息,将有利于提高系统的可靠性。首先在不同工况下对超级电容器进行了老化实验,并通过阻抗谱分析对其容值和等效串联电阻的变化趋势进行了分析。为了利用有限的实验数据预测超级电容器的老化趋势,提出一种基于支持向量机(SVM)的超级电容器老化预测方法,进一步采用粒子群算法(PSO)优化SVM预测模型的参数,从而完善超级电容器老化趋势的预测模型。结果表明预测值与实验值基本吻合,预测误差小于0.046,验证了基于PSO-SVM的超级电容器老化趋势预测方法的有效性。

The aging state of the supercapacitor directly affects the life of the energy storage system. By predicting the aging tendency of supercapacitor performance,it provides predictive maintenance information for the control and the management of the system,which will improve the reliability of the system. Frist,the aging experiments of supercapacitor under different working conditions are carried out,and the change of the capacitance and the equivalent series resistance of supercapacitor is quantified by the impedance spectrum analysis. In order to predict the aging tendency of supercapacitor by using limited experimental data,a new method of predicting aging tendency of supercapacitor based on support vector machine( SVM) is proposed in this paper. The particle swarm optimization( PSO) is further used to optimize the parameters of the SVM prediction model,so as to improve the prediction model of the aging tendency of supercapacitor. The results show that the predicted values are basically identical with the experimental values,and the prediction error is less than 0.046,so the validity of the PSO-SVM method for predicting the aging tendency of supercapacitors is verified.