Nuclear reactor power and flux distribution fitting from a diffusion theory model and experimental data

Abstract

Power and flux tracking in nuclear reactor cores are normally done using diffusion codes. These codes take into account geometric characteristics, compositions, operative conditions and eventually thermohydraulic feedbacks. However, due to operating requirements, in some cases these reactor are instrumented with in-core neutronic flux detectors. This information is usually used in order to verify instantaneous calculations, but it can also be used to fit the theoretical model and then to produce an improved solution. A scheme like this will fit into some extend with the deficiencies of the mathematical problem definition. In this work the integration between a theoretical model and experimental data is searched. It starts with a synthesis method called flux mapping and then its elements are studied. This is the minimization of a functional and selection and construction of expansion functions. For the first expansion function the solution of a diffusion code is chosen, while the other ones are built from Helmholtz equation solutions for a reactor having the same dimensions and boundary conditions. Flux mapping is then tested in a real reactor, a CANDU-600, for several reactivity device configurations. The set of measurement data available even from different physics principles made it suitable to perform the verifications. Comparison with experimental data in the zone at which flux detectors were located shows, for flux mapping, agreements of 2.4%, but 3.2% for the diffusion code used as standard for this reactor. Comparisons throughout the core shows agreements of 3.4 and 5.0%, respectively.