Research On Fuzzy PID Control Methods Based On Smith Predictor

Time delay is a common phenomenon in many industrial processes. Because of the time delay, the disturbance can't be observed in time and the system output can be affected after a long time and the parameters of controller can not be selected easily. Because of insufficiency of the parameters selected, the control performance of the system will be deteriorated , and then the system will not be steady. When the parameters of complex system are variable by time, it is more difficult to design the control system.In this paper, research history and actuality of time-delay processes is introduced first. Different kinds of control methods for time-delay process in recent years are analyzed here. And the research indicates that the control of time-delay process is still a research hotspot of academe and engineering field and its intelligent control technique is still on primary phase. Theoretic and practical significance of research on intelligent control is illuminated in this paper.Based on the above theories, a fuzzy intelligent control method is proposed for time -delay processes in this paper. In this control system, fuzzy controller possessing intelligent integration is used as a controller. Having an integrator serially connected to the output of PD fuzzy controller, the improved PID type fuzzy controller actually behaves like a nonlinear PID controller. In additional, the rule based of the two-dimension PID type fuzzy controller is easier to be established. In this paper, a self-tuning methodology is implemented based on fuzzy self-corrector for factor β which can adjust the controller's tache of proportion andintegral. A rule self-adjustable fuzzy controller is presented based on intelligent weight function which use absolute value of error and error change and adjust fuzzy control rules indirectly. So this algorithm has good adaptability and robustness. The effectiveness of the proposed PID type fuzzy controller is verified by the simulation results. However, it can't control the large time-delay system efficiently because of time-delay parameter. So the improved fuzzy controller and Smith predictor combined to a kind of intelligent controller. Smith predictor possessing one-order inertia tache is used as a predictor which can adjust itself with the change of the object under the control of the self-adaptive organ to keep the system in optimal state. By simulation it change the object parameter including gain, time constant and delay time, and obtained regulation rules about parameters of inertia tachechanging. A self-adaptive organ to adjust inertia tache parameter on real time is designed and realized based on these rules.Since controller and predictor in this self-adaptive control system can overcome the change of model parameters, precision of object model required by control system can be reduced, and system has better robust and control performance. Aiming at some kinds of typical control arithmetic and making use of MATLAB language, it analyzes and contrasts the influence of various arithmetic that caused by the change of system parameters. Simulation results have shown that the proposed method has better control performance and reveal good application potential for the large time-delay processes.