Research Of Networked Control System Based On GPC And Error Correction Using Fuzzy Inference

In the control system, traditionally,we applied the point-to-point connectionsbetween the various devices and equipments,and this way successfully is used severaldecades in industrial system. with the development of the science and technology andthe development of computer technology,because of the physical equipment inindustry and the expansion of the system, the traditional connection mode can not meetthe new performance requirements, and then we introduce into the network. In theclosed-loop control system we connect controller, actuators and sensors with thenetwork which we called network control system(NCS). The network control systemhas the advantages of the completesystem structure, the complete distributed operationmode, relatively independent and it is good at internet communication mode, and somuch saving of the wiring and reliability of the signals, etc.However, owing to theintroduction of the network, meanwhile it also brought some inevitable problems. Theexistence of the delay of the network will greatly reduce the performance of the controlsystem, narrow the scope of stability of the system, and even seriously influence thestability of the system. This paper mainly comes up with the suggested solutions innetwork delay.The main works and outcomes of this thesis are as follows:(1)First of all,the background and the status of the networked control system arebriefly introduced;the basic problems brought by the introduction of communicationnetwork are illustrated;the time delay problem is mostly discussed,including the causesof the network delay, the main composition of network delay and the Time delay toinfluence on the performance of the network control system.(2)Then, the background and the status of the predictive control and fuzzy controlare introduced; the three basic principle of the predictive control forecast areillustrated,which include prediction model, rolling optimization and feedbackcorrection. Meanwhile the basic principle of fuzzy control and the design method of fuzzy controller are briefly introduced.(3) In this, Error prediction in building the system model based on fuzzy inferenceis proposed to solve the time delay of the networked control system. In the forwardchannel, generalized predictive controller is designed to compensate the forward delayand the the prediction error is corrected by the fuzzy controller, namely the controlvariable is composed of two parts:generalized predictive control variable and fuzzycontrol variable.(4) At last, the paper introduces the TrueTime Toolbox, a networked controlsystem platform is established based on TrueTime; traditional generalized predictivecontrol algorithm and the proposed algorithm are simulated respectively; thesimulation results show that the the proposed method is effective and better thantraditional generalized predictive control algorithm.