分数阶M-CNN系统的混沌特性分析及同步控制  

Chaotic Characteristic Analysis and Synchronization Control of Fractionalorder Memristor Cell Neural Network

作　　者：王仁明[1] 冀同康 张赟宁[1] WANG Ren-ming;JI Tong-kang;ZHANG Yun-ning(College of Electrical Engineering&New Energy,China Three Gorges University,Yichang 443002,China)

机构地区：[1]三峡大学电气与新能源学院,湖北宜昌443002

出　　处：《控制工程》2020年第8期1412-1418,共7页Control Engineering of China

基　　金：国家自然科学基金资助项目(61603212)。

摘　　要：将忆阻器作为细胞间的突触连接引入细胞神经网络(Cellular Neural Network,CNN)模型,构建了一个四维分数阶忆阻细胞神经网络(Memristor Cellular Neural Network,分数阶M-CNN)模型。通过数值仿真分析了该模型随忆阻器参数和系统阶次变化时的动力学特性,如混沌吸引子,分岔图和李雅普诺夫指数谱。结果显示随着系统参数的改变,该系统出现了周期极限环,单涡旋混沌吸引子和双涡旋混沌吸引子现象。接下来,讨论了该系统的混沌同步问题,基于李雅普诺夫第二方法设计了滑模控制器和自适应更新律,实现了2个同结构分数阶M-CNN模型、分数阶M-CNN系统模型和异结构的分数阶Chen系统的混沌同步。MATLAB数值仿真验证了设计方法的可行性和有效性。This paper constructs an 4-D fractional-order memristive cellular neural network model by introducing a memristor used as a synaptic connection between two cells.As the memristor parameters and the fractionalorder of the model are changed,the dynamic characteristics of the considered model,such as chaotic attractors,bifurcation diagrams and Lyapunov exponents are analyzed based on MATLAB simulation.The results show that periodic limit cycles,single-scroll chaotic attractors and double-scroll chaotic attractors appear with the changing of system parameters.Next,the chaos synchronization of the considered system is investigated.By means of the Lyapunov second method,the sliding mode controller and the adaptive updating law are designed and the chaotic synchronization between two identical 4-D fractional-order memristor cell neural network models and between two non-identical chaotic systems,namely,4-D fractional-order memristor cell neural network model and 3-D fractional-order Chen system are realized.The numerical simulation verifies the feasibility and effectiveness of the proposed method.

关 键 词：忆阻器 分数阶细胞神经网络 混沌同步 自适应滑模控制 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]