Research On Self-adaptive Intelligence Control Technology And Its Application

This subject puts emphasis on the research of some shortcomings existing in general intelligence control and presents some solutions to those shortcomings.First of all, based on the shortcoming of conventional fuzzy logic controller, neural network controller and general controller, a self-adaptive compound intelligence controller is put forward through modifying normal controller, which includes some advantages of those intelligence control arithmetic. Then a kind of fuzzy-Smith control strategies for a large dead-time system is studied and improved. The simulation results show that the fuzzy-smith control method has some excellent performances, which is suitable for systems with deadtime-varying parameters.Considering that the conventional fuzzy has a shortcoming of invariable fuzzy rules, an algorithm of adjustable fuzzy rules based on a single neural cell is proposed in this paper, which is used for modifying the fuzzy control query table on-line. The simulation results show that the improved control method has some excellent features, such as high accuracy and robustness.In the end, a single neural cell self-adaptive controller is studied; furthermore, a control system of SRD based on the single neural cell self-adaptive controller is designed by using 80C196KB microprocessor.