Design Of Adaptive Sampled-data Control For High-Order Nonlinear Systems

With the rapid development of computer technology and digital control technology,the sampled-data control system composed of continuous time control system and discrete time controller has been widely used in practical industrial engineering.In this paper,for several types of typical nonlinear systems,the stability of the system under sampled data control is studied under the condition that both system state information and nonlinear information are unknown.The main research results are given as follows:1)A type of sampled-data control strategy based on system observer is proposed with respect to a class of nonlinear systems whose nonlinear function satisfy the local Lipschitz condition.Firstly,the sampled-data state observer is designed to estimate the unknown state of the system.Secondly,the sampled-data output feedback controller is obtained by using the time-delay dependent control method.Finally,the conservativeness of the designed control strategy is reduced by constructing piecewise Lyapunov functions,and the stability of the system is analyzed utilizing linear matrix inequalities,and the effectiveness of the proposed sampled-data control strategy is verified by MATLAB simulation example.2)For a class of nonlinear systems with unknown nonlinear information and state information,an observer based radial basis function(RBF)neural network adaptive sampleddata control strategy is proposed.Firstly,for the nonlinear system with unknown state information,the system state is estimated by designing a state observer.Secondly,the unknown nonlinearity of the system is approximated by using the estimated system state and the excellent function approximation characteristic of RBF neural network.Afterwards,combining the observer design and RBF neural network,the neural network adaptive sampled-data controller is designed by introducing Backstepping method,which guarantees the semi-global uniformly ultimately bounded of the closed-loop system.Finally,the conditions for system stability are obtained based on Lyapunov stability theory,and the effectiveness of the proposed neural network adaptive sampled-data control strategy is illustrated through MATLAB simulation example.3)A predictor-based global feedback sampled-data control strategy is proposed for a class of nonlinear uncertain systems with more general nonlinear constraints,and which the exact knowledge of the system nonlinearity as well as the full state information are both unknown.Firstly,by using the prediction technique,the dynamic output values of the system during the sampling period are obtained to compensate for the adverse effects on the system due to sampling.Secondly,the output predictor is used to build a system state observer to estimate the unknown state of the nonlinear system.Finally,the sampled-data controller is designed based on the above information,the hybrid control system is constructed and analyzed for system stability by utilizing the Lyapunov function method,and the effectiveness of the proposed sampled-data control strategy is illustrated by MATLAB simulation of the single-link robotic arm system.