阶数位于1到2之间的分数阶神经网络同步控制  

Fractional Order Neural Network Synchronization Control with Orders Between 1 and 2

作　　者：蒋文芳 田焕欢 刘恒 JIANG Wenfang;TIAN Huanhuan;LIU Heng(School of Mathematics and Physics,Guangxi Minzu University,Nanning 530006,China;School of Mathematics and Statistics,Anhui Normal University,Wuhu Anhui 241000,China)

机构地区：[1]广西民族大学数学与物理学院,广西南宁530006 [2]安徽师范大学数学与统计学院,安徽芜湖241000

出　　处：《佳木斯大学学报（自然科学版）》2024年第5期176-180,共5页Journal of Jiamusi University：Natural Science Edition

基　　金：国家自然科学基金(12001012,61967001)。

摘　　要：近年来,神经网络同步问题吸引了系统控制、数学和计算机科学等多个领域学者的关注。针对阶数位于1到2之间的分数阶神经网络同步控制问题,通过变量替换法,将阶数在(1,2)之间的分数阶神经网络同步控制问题转化为阶数在(0,1)之间的分数阶神经网络同步控制问题。由于分数阶神经网络中含有未知非线性函数项,因此使用模糊逻辑系统进行逼近。之后,为了使从神经网络跟踪主神经网络,设计了合适的控制器,并且,对其同步误差系统进行稳定性分析。最后,通过仿真实例验证了所提出方法的有效性。Synchronization of neural networks has drawn considerable attention from many fields such as systems control,Mathematics and computer science etc.,during the past years.This paper mainly focuses on the synchronization control problem of fractional order neural networks with orders between and.In the paper,the variable replacement method is used to transform the fractional order neural network synchronization control problem with orders between into a fractional order neural network synchronization control problem with orders between.Because the fractional neural network contains unknown nonlinear function terms,fuzzy logic systems are used for approximation.Afterwards,in order to enable the slave neural network to track the main neural network,a suitable controller is designed.And stability analysis of synchronization error system.Finally,the effectiveness of the proposed method is verified through simulation examples.

关 键 词：模糊逻辑系统 分数阶神经网络 同步控制 

分 类 号：O231.1[理学—运筹学与控制论]