MRFACS with nonlinear consequents by fuzzy identification of system for time delay system

We proposed two types of fuzzy MRACS (MRFACS) for time delay system, one is fuzzy controller designed with traditional fuzzy logical controller and the other is fuzzy controller design with fuzzy identification concept. GAs are applied for optimizing the rule set of a fuzzy logic control (FLC) and the coefficients of fuzzy identification system (FIS). In order to accelerate the search, we modified genetic algorithms (GA) with a eugenics policy. The conclusions we got after simulations are : (1) Modified GA (MGA) can find rule set efficiently, because the multiple-point operators do the searching more effectively. On the other hand, the design of MGA designed based on the searching history and the eugenics policy is also proved efficient. (2) MRFACS with FIS shows higher performance indices than the MRACS with FLC for the reason that the front controller designed with nonlinear equations makes the outputs ΔK more smooth than that generate by FLC. (3) The fuzzy controller is robust. (4) Although the MRFACS with FIS shows more effective than that with FLC, the coefficients searching will take longer time in MGA in case that the coefficients are all float values