异时滞分数阶细胞神经网络的电路设计与仿真  

Circuit Design and Simulation of Fractional Order Cellular Neural Networks with Different Time-Delay

作　　者：张梓涛 袁晓[1] ZHANG Zi-tao;YUAN Xiao(College of Electronics&Information Engineering,Sichuan University,Chengdu Sichuan 610064,China)

机构地区：[1]四川大学电子信息学院,四川成都610064

出　　处：《计算机仿真》2021年第3期222-227,共6页Computer Simulation

摘　　要：针对目前细胞神经网络的电路模型中鲜有人考虑系统延时的问题,提出一种异时滞分数阶细胞神经网络模型.用Matlab数值仿真结果表明理论可行,再借助Multisim软件分别构造输出函数单元电路、分数阶单元电路,用带LCL低通滤波的有源电路在一定频率附近产生固定时滞值τ,研究不同时滞组合下系统的混沌特性.电路仿真结果表明,改变系统某参数的值可使异时滞系统产生与Matlab数值仿真相似的混沌相图,从而证实异时滞分数阶细胞神经网络的物理可实现性和系统鲁棒性,相比于无时滞系统更具现实意义和工程价值.A few people in the current cellular neural network circuit model consider the problem of system delay.This paper presents a fractional order cellular neural network model with different time-delay.The results of numerical simulation based on Matlab show the theoretical feasibility,then the Multisim software was used to construct the output function unit circuit and the fractional order unit circuit respectively.The active circuit with LCL low-pass filtering generated fixed time-delay values near a certain frequency,and then the chaotic characteristics of systems with different time-delay combinations was studied.The circuit simulation results show that changing the value of a certain parameter of the system can make the chaotic phase diagram similar to Matlab numerical simulation of the time-delay system,thus confirm the physical achievability and system robustness of the fractional-order cellular neural network with different time-delay.The system without time-delay has more practical significance and engineering value.

关 键 词：时滞 分数阶 细胞神经网络 混沌 电路仿真 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]