Delay System Control Method

In the industrial production process, the controled object with time-delay characteristics is very common, but it is difficult to control, especially the long time-delay control is always a difficult problem. This paper introduces the current some advanced control methods in dealing with the long time-delay system, especially analyzes the general theory methods to overcome the long time-delay field of the predictive control and fuzzy predictive control theory.In solving the problem of complex system control, the fuzzy identification and control theory show enormous potential. For the control problem of complex systems, this paper combines fuzzy control and predictive control, and presents a fuzzy prediction identification-free algorithm for adaptive control method by absorbing their respective advantages. The research and simulation results show that the algorithm above can obtain good control effect.The fuzzy identification algorithm based on T_S model is a good fuzzy control algorithm. An improved on-line identification method based on recursive fuzzy clustering is presented, which can adaptive adjust the model on line. The improved algorithm reduces the computation of parameter identification and structural identification significantly, and guarantees the requirements of online identification.The simulations obtain a better tracking result. The improved on-line identification method presents generalized predictive control based on the identification of fuzzy model, and amends the GPC recursive algorithm, then obtains a better control effect, which has robustness, anti-interference and the ability to overcome the long time-delay.