Distributed Optimization With Complex Variables And Synchronization Of Delayed Competitive Neural Networks

Biological neural system is a super large information system including molecules,cells,clusters,brain regions and other different levels.It is the most complex nonlinear network system found.Neural information activities are mainly realized through electrophysiological processes.Therefore,people need to deeply explore the complex discharge activities of brain neural network system from the perspective of dynamics,Based on the abundant data obtained from modern neuroscience research,neurodynamics,as an important part of the theoretical basis of neuroscience,was born in good time,We should also actively explore its application in the field of complex networks and artificial intelligence,and play a theoretical guiding role in the research of neural function regulation,neural diseases and brain like intelligence.In this paper,the application of neural dynamic model in real and complex domains is studied.Firstly,the problem of exponential synchronization of competitive neural networks with time delay under intermittent control is discussed.By using the method of differential inequality and Lyapunov function,the sufficient conditions for global exponential synchronization of competitive neural networks with time delay are given,and the corresponding criteria are established,On the basis of not selecting penalty parameters,a neural network model of differential inclusion description is established to solve complex variable optimization problems.The algorithm does not need to design penalty parameters,which is convenient for practical application.Under appropriate conditions,some convergence theorems of the model are given.Finally,some numerical examples are given to verify the correctness and effectiveness of the model.