Fusion of state feedback, prediction, and fuzzy logic for power regulation in a research reactor

A novel control scheme for power regulation in a research nuclear reactor has been developed. The scheme combines different techniques such as state variable feedback, first order numerical integration for estimation, and fuzzy logic to attain a stable power regulation in the reactor. The proposed control strategy attains, in short time and without overshoot, different values of steady-state power departing from a wide range of different values of initial power. The algorithm was subjected to exhaustive simulation tests and the results obtained are highly satisfactory.