| With the rapid development of computer technology and communication technology, the control system gets larger and widely distributed. The network as the backbone of networked control systems (NCSs) which is composed of sensors, controllers and actuators exists uncertain problems:network induced delay and data packet dropout and so on. Such problems can lead to the system instability or slowdowns in performance. Thus, the research fault tolerant control of NCSs has theory significance and practical significance.The main work is as follows:Firstly, this paper analyzes the current research status about the method of parameter estimation on the state observer, and proposes a new approach to design the parameters of ESO. According to the new method, we design the linear extended state observer (LESO) and non-linear extended state observer (NESO) separately. The experiments testify the validity of this new method.Secondly, we set up the model for linear controlled object with network induced delay, data packet loss and faults, and construct an ESO. Then, based on the new parameter estimation method, we design the LESO which can trace estimation the states of control system. The new state can compensate the system’s unmeasured state, improves the performance of the system. Using the Lyapunov function to analysis the stability of the controlled system is provided, and finally testified the method is better than the old ones by experiments.Finally, in view of the nonlinear controlled object, set up the model of control system with the network induced delay, packet dropout and faults. Because of the complexity modeling of nonlinear object. Using the ESO to solve the problem of that is difficult to model. The state feedback controller is designed to track targets stability, This paper constructs a Lyapunov function to analyze the stability of NESO. The simulation experiment proves the validity and stability of this method that using the NESO dealing with the nonlinear control system. |
PDF Full Text Request