Estimation And Synchronization Of Switched Complex Networks Based On PDT Switching Mechanism

Switched complex networks have attracted the attention of a large number of researchers because they can model most dynamic systems with switching characteristics.For the above-mentioned networks,when part of the state information cannot be directly obtained,it is of great practical significance to reconstruct the estimator to track the internal state information,and then to achieve the subsequent control goals.At the same time,synchronization,as a basic dynamic characteristic of complex networks,also has certain research value.Up to now,the related theoretical research scope is very wide and the research results are quite abundant,but considering the influence of time-varying delay,multi-time scale phenomenon,randomly occurring gain fluctuation,nonlinearity and network-induced phenomena on the system performance,the issue of estimation/synchronization of persistent dwell-time(PDT)switched complex networks still needs to be solved urgently.Therefore,this paper aims to explore the stability,performance analysis and estimator/controller co-design of discrete-time switched complex networks based on PDT switching rule.The main content is briefly described as below:1.Under the adjustment of the PDT switching rule,the estimation/synchronization issues of two types of switched inertial neural network models with different structures are studied,and the effects of time-varying delay and gain uncertainty on the two kinds of network models are considered respectively.By using Lyapunov stability theory,matrix analysis method,stochastic analysis theory and PDT switching rule concept,several sufficient criteria are derived to make the estimation/synchronization error system satisfy globally uniform exponential stability/exponentially mean-square stability and the specified3_?(9)/l₂-l_?performance requirements.Eventually,the gain of mode-dependent state estimator/synchronization controller is obtained by using the common decoupling method.The rationality of the proposed scheme and the effectiveness of the designed synchronization controller are verified by comparing the state responses of the open-loop and closed-loop systems in the numerical simulation examples.2.Firstly,the state estimation problem of a group of singularly perturbed switched complex dynamic networks constrained by PDT switching rule is considered.Then another group of singularly perturbed coupled neural networks is considered,and the PDT switching rule is used to adjust the changes of the transition probability matrix in the Markov chain,which solves the non-fragile synchronization problem of the network.By selecting the appropriate Lyapunov function and combining with the theory of switched systems,singular perturbation theory and improved matrix decoupling method,some sufficient conditions are derived to ensure that the estimation/synchronization error system satisfy the definition of globally uniform exponential stability/exponentially mean-square stability and the specified generalized dissipation/)93(_?performance requirements,and to ensure the existence of feasible solutions for the designed estimator/controller.Then,the gain of the fast and slow parts of the estimator/controller is obtained by using convex optimization technique.Finally,a numerical example is used to verify the rationality of the proposed scheme and the effectiveness of the designed estimator/controller.3.Consider a group of distributed sensor networks with switching characteristics and use PDT switching rule to arrange the switching between system parameters.For a common target device,a sensor network formed by multiple sensors distributed in space is used to track the state information of the target device that cannot be directly measured,so as to achieve subsequent control requirements.Additionally,some sufficient criteria are derived for the estimation error system to achieve exponentially mean-square stability with the specified)93(_?performance.Eventually,a calculation example shows that the proposed scheme is effective.