Expert system driven fuzzy control application to power reactors

The authors describe a fuzzy logic application to nuclear reactor control, namely, the automated start-up control of the Experimental Breeder Reactor-II. A rule-based expert system supervises the fuzzy controller (which also uses a rule-based representation). The overall system successfully emulates the behavior of the actual power plant operator. The operator control strategy is based on linguistic statements which translate into fuzzy productions describing heuristic control rules. The developed methodology is verified through computer simulations using a valid nonlinear model of the reactor. The necessary heuristic decisions, which are vitally important for the implementation of fuzzy control in the actual plant, were identified