The analytic nodal diffusion solver ANDES in multigroups for 3D rectangular geometry: Development and performance analysis

Abstract

In this work we address the development and implementation of the analytic coarse-mesh finite-difference (ACMFD) method in a nodal neutron diffusion solver called ANDES. The first version of the solver is implemented in any number of neutron energy groups, and in 3D Cartesian geometries; thus it mainly addresses PWR and BWR core simulations.The details about the generalization to multigroups and 3D, as well as the implementation of the method are given. The transverse integration procedure is the scheme chosen to extend the ACMFD formulation to multidimensional problems. The role of the transverse leakage treatment in the accuracy of the nodal solutions is analyzed in detail: the involved assumptions, the limitations of the method in terms of nodal width, the alternative approaches to implement the transverse leakage terms in nodal methods -implicit or explicit -, and the error assessment due to transverse integration. A new approach for solving the control rod "cusping" problem, based on the direct application of the ACMFD method, is also developed and implemented in ANDES.The solver architecture turns ANDES into an user-friendly, modular and easily linkable tool, as required to be integrated into common software platforms for multi-scale and multi-physics simulations. ANDES can be used either as a stand-alone nodal code or as a solver to accelerate the convergence of whole core pin-by-pin code systems. The verification and performance of the solver are demonstrated using both proof-of-principle test cases and well-referenced international benchmarks.

Extension of the analytic nodal diffusion solver ANDES to triangular-Z geometry and coupling with COBRA-IIIc for hexagonal core analysis

“…During the last years, the reactor physics group at UPM has been working in the development of the in-house nodal neutrón diffusion code ANDES (Lozano et al, 2007(Lozano et al, , 2008a. It is based on the ACMFD method developed by Y.A.…”

Generación de librerías optimizadas con cuantificación de incertidumbres para cálculos realistas (BEPU) de reactores nucleares de agua ligera

“…[92] A la hora de aplicar métodos nodales a núcleos LWR de forma fiable y satisfactoria se requiere, además de códigos de difusión precisos a nivel nodal, metodologías adecuadas para producir adecuadamente parámetros "equivalentes" (secciones eficaces y factores de discontinuidad) usados como input en los códigos. Ello se corresponde con la primera fase de la aproximación estándar que se vio en el apartado 2.3.…”

“…Como extensión de los modelos presentes en SEANAP, y fruto de la participación en los proyectos de los Programa Marco de la Unión Europea, nace el código COBAYA3, incluyendo el código nodal ANDES [92,91], el código de difusión en malla fina COBAYA3k [53], y la posibilidad de acoplamiento con los códigos termohidráulicos COBRAIIIc/MIT-2, COBRA-TF, FLICA-4 y SIMULATH mediante la descomposición en subdominios y utilizando la técnica de disecciones alternadas [67].…”

“…Las diferencias en la concentración de boro crítico son inferiores a 5 ppm. En [92] se encuentra una comparación más detallada incluyendo diferentes nodalizaciones, estructuras de grupos de energía y comparaciones en los factores de pico. En cualquier caso, se encuentra una diferencia no despreciable entre la concentración de boro crítico calculada por los códigos de difusión empleando la librería del benchmark y la solución proporcionada por los códigos BARS y DeCART.…”

Approximation of the Neutron Diffusion Equation on Hexagonal Geometries

“…To avoid the rod cusping effect on nodal methods, different strategies have been developed [50]. As, for example, a flux weighting method [54], moving meshes method [55], an equivalence method [54], and a three nodes homogeneization method [56]. These methods have to solve a small one-dimensional eigenvalue problem for each one of the partially rodded nodes.…”