基于改进遗传算法的Volterra核辨识研究  

Volterra Series Identification Method Based on Improved Genetic Algorithm

作　　者：王鑫超 张宾 WANG Xinchao;ZHANG Bin(School of Mechanical Engineering,North China University of Water Resources and Electric Power,Zhengzhou 450045,China)

机构地区：[1]华北水利水电大学机械工程学院,河南郑州450045

出　　处：《河南科技》2022年第5期24-27,共4页Henan Science and Technology

摘　　要：本研究提出了一种基于改进遗传算法辨识Volterra级数模型的方法。该方法根据Volterra核与系统输出的相关程度来调整模型结构,利用重启策略与自适应搜索范围解决进化停滞与算法早熟收敛等问题。通过仿真试验将改进遗传算法与标准遗传算法、量子粒子群算法进行比较。结果表明,该方法在辨识精度、收敛速度及抗噪性能等方面明显优于其他方法。In this paper,a method for identifying Volterra series model based on improved genetic algorithm(IGA)is proposed.This method adjusts the model structure according to the correlation between Volterra kernel and system output,uses the restart strategy and adaptive search range to solve the problems of evolutionary stagnation and premature convergence of the algorithm.Through simulation experiments,the IGA method is compared with standard genetic algorithm(GA)and quantum particle swarm optimization(QPSO)algorithm.The analysis results indicate that the IGA method is superior to other methods in identification accuracy,convergence speed and anti-noise performance.

关 键 词：VOLTERRA级数 改进遗传算法 非线性系统 系统辨识 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]