分数阶细胞神经网络自适应同步控制设计及电路仿真  

Fractional-order cellular neural networks adaptive synchronization control circuit design and simulation

作　　者：张小红[1] 俞梁华 

机构地区：[1]江西理工大学信息工程学院,江西赣州341000

出　　处：《控制理论与应用》2016年第6期753-762,共10页Control Theory & Applications

基　　金：国家自然科学基金项目(61363076);江西省自然科学基金项目(20142BAB207020);江西省研究生创新专项基金(YC2014-S368)资助~~

摘　　要：构建一个新的分数阶细胞神经网络系统,设计驱动系统非线性参数已知而响应系统非线性参数值未知的驱动–响应系统,运用自适应同步控制器及参数自适应调整律实现该驱动–响应系统同步.数值仿真和动力学分析结果表明新的分数阶细胞神经网络系统具有混沌特性.结合分数阶电路理论设计出新的分数阶细胞神经网络系统同步控制的电路原理图.本方案实际可实现4096种多元组合电路,为简洁起见,选取分数阶qi(i=1,2,3)相同值(即q1=q2=q3=0.95)的组合电路进行电路仿真.仿真结果表明,多元电路仿真和数值仿真实验结果具有很高的吻合度.从而证实了该自适应同步控制方法在物理上的可实现性,在工程领域中具有现实的应用价值.On a new fractional order cellular neural networks system, building a driver-response system which the nonlinear in the driver system is known, while in response is not. Use the adaptive synchronization controller and adaptive adjustment law to make the driver-response system to be synchronized. Numerical simulation and dynamic analysis results show that the new system of fractional order cellular neural networks has chaotic characteristics and the correctness of the adaptive synchronization control method. Combining the theory of fractional order circuit designed the synchronous control circuit schematic diagram of the new fractional order cellular neural networks system. In fact, this scheme can realize 4096 kinds of multi-combination circuits. For the sake of brevity, selected the same fractional-order values（namely q1 = q2 = q3 = 0.95） for circuit simulation. The simulation results showed that multiple circuit simulation and numerical simulation results had high inosculation. Thus it is proved that the adaptive synchronization control method can be realized in physics and has practical application value in engineering field.

关 键 词：同步控制 自适应 分数阶 细胞神经网络 混沌 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP273[自动化与计算机技术—控制科学与工程]