The feedback control systems where control loops are closed through real-time communication networks are called the networked control systems (NCSs). It is a new interdiscipline combined with control, communication and computer technologies. In this thesis, the design and implementation of NCSs is studied, based on the networked predictive control (NPC) method. The predictive controller and network delay compensator are designed for linear systems and nonlinear systems respectively. The methods are implemented and realized on actual IP network based control systems. The validity of the proposed methods is demonstrated by experimental results. For the investigation of the NCSs, a networked control system experiment platform is constructed which involves of three typical network structures, and the data transmission performance and characteristics of each structure are tested and evaluated. The network induced delay is modeled as a Markov chain, and the computation of the Markov transition matrix is analyzed. According to linear systems, a mathematical model of the networked predictive control scheme is established. Based on the model, a networked predictive stochastic optimal control method is presented, and the formulae of the predictive control sequence generator for both state feedback controller and output feedback controller are derived. For applications in NCSs with rapid sample time and comparatively long network induced delay, a simplified approach is proposed, based on a one-step jump Markov delay model. The DC motor control experiments are carried out via wireless local area networks and wireless metropolitan area networks. The performance of the proposed method is compared with the performance of the stochastic optimal control and networked predictive control. According to nonlinear systems, two types of nonlinear predictive control approaches are presented. For the first of systems—the relative degree of the systems is 1, the algorithm of a strong tracking filter along with the generic model controller is adopted, and a nonlinear adaptive networked predictive control scheme is designed. The validity of the proposed approach is illustrated by the networked dual-tank control experiment and networked DC motor control experiment. For the second type of systems with ill-defined relative degree, a networked nonlinear model predictive control scheme is presented, which can be used in the control of ball-beam systems. The emulational experiments ar...
修改评论