![]() | Only 14 pages are availabe for public view |
Abstract A neuro-fuzzy control system with a learning algorithm is proposed for Egypt second research reactor, ETRR-2. The proposed algorithm is derived from the neuro¬fuzzy model NEFCON (NEuro-Fuzzy CONtrol) to learn and optimize the fuzzy rules of a Mamdani fuzzy controller online using a fuzzy error measure. The neuro-fuzzy control system is’ applied to control ETRR-2 simulation model in operational transients under MA TLAB/SIMULINK. The results are compared with those of the accepted PD controller used in ETRR-2 real-time environment. As a result, it is shown that the neuro-fuzzy control algorithm is superior to the conventional PD controller and can be implemented for real-time control implementations. The neuro¬fuzzy control can optimize the fuzzy rules online. Therefore, it is more suitable and showing a better performance than fuzzy control when the fuzzy control rules are inappropriate |