Analysis And Synthesis Of Hybrid Systems With Quantized Signals

The hybrid system is widespread in the robotics, automotive industry, aerospace,communication engineering, chemical processes, the nervous system and the bio-engineeringfield. It has been successfully applied in many practical engineering systems. With thedevelopment of the computer technology, especially for the case that modern control sys?tems are becoming increasingly dependent on digital devices, the quantization technologyof signals is applied to the control ifeld, and thus the modeling, analysis and control ofhybrid systems with quantized effects become the forefront of the system control and itsapplied disciplines. The dissertation studies the analysis and synthesis of hybrid systemswith quantized signals, thus has theoretical significance and application values. The maincontributions are listed as follows:1.By means of partial contraction theory and matrix measure concept, the syn?chronization of delayed hybrid dynamical networks with quantized impulsive effects andintermittent complex dynamical networks with quantized signals is studied, respectively.New criteria for synchronization of the hybrid dynamical networks are proposed. Numer?ical simulations are presented to illustrate the effectiveness of the proposed results.2.With the help of the free-weighting matrix method and the converse Lyapunov the?orem, the stabilities of two classes of hybrid systems are discussed, respectively. We firstaddress the stochastic discrete-time genetic regulatory networks with Markovian jump?ing parameters, and present the criteria for the global robust exponential stability in themean square sense by means of the probability distributions of the time-varying delays.Second, we analyze the characterization of its input/output and describe the equiva?lency of the uniformly input-output-to-state stability (UIOSS) Lyapunov function andthe exponential-decay UIOSS-Lyapunov function. Finally, the suiffcient condition of theUIOSS is obtained through the state-norm estimators.3.On the basis of the algebraic graph theory, the consensus of switched multia-gent networks with communication constraints is achieve. We first construct the hybridmodel of switched multiagent systems under quantized feedback based on a hysteresisquantizer, further analyze the finite-time convergence, and present the lower bound oftime to consensus convergence. Second, we discuss the quantized average consensus of aswitched weighted multi-agent network. Then the relationship between the algebra con?nected graph and the performance of consensus agreement is explored. Furthermore, theconvergent region of the average consensus error is derived.4.Based on the converse Lyapunov theorem, the robustness of the hybrid control fornonlinear systems with encoded state feedback or quantized state feedback is discussed,respectively. Meanwhile, the hybrid models of closed-loop systems for both cases areconstructed, respectively. Moreover, the criteria for the semi-global practical asymptoticalstability of the closed-loop hybrid systems are presented. Finally, the simulation resultsillustrate the benefit of the proposed control strategy.Finally, some conclusions of the thesis and prospects to be further studied are pre?sented.