Analysis And Design Of Quantized Networked Control Systems With Composite Lyapunov Function

Networked Control Systems(NCSs) have received increasing attention in recent years due to their great advantages, for instance, lower installation and maintenance costs, increased flexibility, rapid installation and diagnostics, and higher reliability.Netwoked control system is a closed-loop control system where the plant, sensors,controllers and actuators are combined by a shared network. However, there exist undesirable phenomena including packet dropout, induced time delays and quantization,which may brings tremendous negative impacts on the system. This thesis concerns with a new method(composite Lyapunov function method) to achieve better system performance.First, the quantization density and H? performance for networked control systems with logarithmic quantization are studied. Different from the traditional quadratic Lyapunov method, we utilize the composite Lyapunov functional to get smaller quantization density and better H? performance.Second, the regional stabilization of constrained discrete-time NCSs is addressed.The packet dropout phenomenon is modeled as a stochastic variable satisfying Bernoulli process. The initial states set and convergence domain are derived such that all the trajectories starting from the initial set will eventually converge to the domain of convergence in the mean square stable sense.Finally, the regional stabilization of constrained continuous-time NCSs is addressed. The packet dropout phenomenon is modeled as system's induced time delay.Meanwhile, the initial set and convergence domain can be computed by the maximal Lyapunov-Krasovskii functional method.