Incomplete Information Model Based Control And State Estimation For Complex Networks System

In this thesis, the control and state estimation problem are discussed for several classes of complex networks based on the randomly occurring incomplete information model. The incomplete information to be considered here mainly refers to the useful signals usually can not be recovered completely through the transmission channel due to the inaccurate of system modeling or/and the physical constraints of the practical components and devices, e.g. data fully/partially missing, such as missing measurements, time delay, sensor saturation, signal quantified and so on. Moreover, by using the statistical mean which is more close to the background of the practical applications to characterize the incomplete information phenomena occurring in a random way, an objective and applicable system and measurement models are established. The main content of this thesis is totally comprised of two parts. In the first part, attention is focused on the design of a series of desired state estimators which are satisfied with the stability and H∞ performance for a class of discrete complex network system with randomly occurring coupling delay and missing measurements, and some sufficient conditions are obtained for the existence of desired estimators in terms of LMI(Linear Matrix Inequality). Then the synchronization control problem is investigated for a class of complex network with randomly varying nonlinearities in the second part. Some novel notions such as randomly occurring coupling delay are first put forward in the above parts and some efforts are devoted to extending the existing incomplete information models to the complex networks. Concretely, the skeleton and description of the thesis are given as follows:In Chapter 1, the research meaning、background and motivation are clarified, the outline and contribution of the thesis are introduced, and the research problems to be addressed in each individual chapter are also highlighted.In Chapter 2, a state estimation problem is investigated for a class of discrete complex network system with randomly occurring coupling delay and missing measurements. Some sufficient conditions are obtained for the existence of desired estimators in terms of LMI(Linear Matrix Inequality), furthermore, the design problem for the estimators are also considered according to the derived conditions, where the basic theory of complex networks、LMI method、Kronecker product and the Lyapunov stability theory are intensively employed.In Chapter 3, a synchronization control problem is addressed for a class of complex dynamic networks with randomly varying nonlinearities. The random occurring phenomenon of time varying nonlinearities is first extended to the complex networks and the corresponding controllers are designed.In Chapter 4, we summarize the results of the thesis and outline the future works which may be further studied.