Output feedback Model Reference Adaptive Control of nuclear reactor

Abstract

This paper presents an output feedback Model Reference Adaptive Controller (MRAC) for controlling the total power of a nuclear reactor. It ensures set point tracking and estimation of unmeasurable states in the presence of unknown external disturbances and parametric uncertainties. A mathematical model of reactor core based on normalized point kinetic equations, with six groups of delayed neutron precursors, is considered along with reactivity feedbacks due to lumped fuel and coolant temperature. Parametric variations in thermal feedback coefficients are taken as uncertainties, reactivity disturbance is considered as a matched external disturbance and change in inlet coolant temperature is considered as an unmatched disturbance. The desired setpoints for reactor power control are generated by choosing a stable reference model based on LQR theory. An observer is designed based on the asymptotic properties of LQG/LTR for the estimation of unmeasurable states as well as external disturbances. Adaptive laws based on output error are developed to compensate for the effect of matched uncertainties and external disturbances. Further, a feedforward controller is augmented to the adaptive laws to compensate for unmatched disturbances. Overall stability and ultimate boundedness of the closed loop signals are analyzed with Lyapunov theory. Performance of the proposed controller is demonstrated through dynamic nonlinear simulations.