Research Of Modeling And Simulation To Ncs With Random Time-delay Characteristic

The recent research status of Networked Control System (NCS) is introduced in this thesis, and then the construction, characteristics and principle of network delay of NCS are discussed.The simulation model is established for the different time delay in the Matlab toolbox under Truetime, and the conclusion for the NCS system stability, convergence will be significantly lower with the network delay increases is verified .The accurate discrete mathematical model for the deterministic time delay and continuous mathematical model for the random short time delay is established. The reliability and stability of the NCS under certain conditions is tested with Matlab.Discussion and research the discrete mathematical modeling of the NCS for the short delay and long delay conditions is the main work of this thesis.Under the condition of assumption that the sensors are time-driven, controllers and actuators are things-driven, a discrete NCS model for the random short time delay is established. The accuracy of the derivation about K and the effectiveness of the discrete NCS model for the random short time delay are proved by an actual example of the model is simulated in the Matlab toolbox under Truetime.Under the condition of assumption that the sensors are time-driven, controllers and actuators are things-driven, a discrete T-S fuzzy NCS model for the random long time delay is putted. The theory of model combines the fuzzy control theory and discrete mathematical models manner. A state observer that similar to the NCS systems is design, and settled in the controller side to compensate the network delay. The simulation system model is built by examples in the Matlab environment using Truetime toolbox. The simulation about NCS system model of traditional,no state observer of the TS NCS model and TS NCS model with state is studied, and the input and output curves of the NCS models are achieved. Superiority of the discrete T-S fuzzy control theory model over the conventional models and the effectiveness of the state observer are proved.Finally, a summary of this thesis is given, and look ahead the direction of future research and development trend in the future.