Interval approach for stability analysis of a nuclear reactor with appropriate thermal-hydraulic model

Abstract

A nuclear reactor exhibits changing behavior with variation in reactor power. Further, uncertainties in the measurement of the actual power produced by the reactor, the heat transfer from the fuel to the coolant and the reactivity change due to the change in the fuel and coolant temperatures and condition of the core, make it necessary to adopt a robust control approach for design of control systems for such plants. In this dissertation, a 540MWe Indian PHWR consisting of 14 zones has been modeled using inexact models incorporating bounded parametric variations in power, heat transfer from the fuel to the coolant and temperature coefficients of reactivity of fuel and coolant. The stability analysis of the inexact models suggests that though the system is marginally stable in nature, the system is controllable. Therefore for stable operation, these inexact models are then used to obtain a robust and optimal PID controller gains for controlling the reactor power with specified time response under parametric variations, using an interval approach. The methodology is established with credible real-time and offline simulations as and where applicable. Enhancement of available power output from a 540MWe Pressurized Heavy Water Reactor (PHWR) with limited coolant voiding has been attempted in the recent past. Coolant voiding in such a core usually occurs in the high power regime and introduce positive reactivity in the core which makes a reactor increasingly unstable. This dissertation presents a scheme for a 700MWe PHWR, which is capable of controlling the reactor from 60%to 93.2% of its Full Power (FP) using the Liquid Zone Control System (LZCS) and the again from 93.2% to 100%FP in conjunction with a pressurizer which is activated to change the pressure in the coolant channel by a small amount, thus keeping the void fraction constraint. The dissertation presents a dynamic model for estimation of void fraction in the 700MWe PHWR core as a function of reactor power in transient time and then proposes an optimal controller for pressurizer control. The pressurizer pressure controller has been designed using Genetic Algorithm (GA) to keep void fraction limited. The controller performance has been validated using credible Hardware-in-Loop (HiL) simulation.