Optimal State Feedback Controller for a Nuclear Reactor

Abstract

This article considers the design of an optimal state feedback controller for a nuclear reactor. The model of 700-MWe Indian pressurized heavy water-type reactor is described and is considered for design. The cost function is the norm of the state feedback vector, which is minimized by making use of extra degrees of freedom obtained due to nonspecific pole placement in the left half of the ${s}$ plane. The stability region is approximated by a maximal sphere in the co-efficient space of the closed-loop characteristic polynomial. The optimization problem is formulated as a quadratically constrained quadratic program and is solved using a primal-dual search method. The issues related to the choice of the center of the approximated sphere and correct step size in the search method are addressed. The norm of the state feedback vector obtained by this method is about 100 times less than those obtained using comparable methods and yet yields acceptable transient performance at various operating points of the reactor.