First Steps in Developing a Sampling-based Approach to Incorporate Uncertainties in CFD Modeling of Involute Fuel Element Research Reactors

Abstract

To support the global non-proliferation efforts, the two European high-performance involute fuel element research reactors Forschungs-Neutronenquelle Heinz Maier-Leibnitz (FRM II) operated by Technical University of Munich (TUM) in Garching, Germany, and the R´eacteur `a Haut Flux (RHF) located at the Institut Laue-Langevin (ILL) in Grenoble, France, are working towards the conversion of their fuel elements from currently highly enriched to low enriched uranium. The design of such new fuel elements strongly benefits from Steady-State Thermal-Hydraulic (SSTH) safety parameter calculations, which take into account various sources of uncertainties like manufacturing tolerances, measurement inaccuracies and modeling assumptions. While there are well-established methods to determine the nominal SSTH safety parameters at FRM II and RHF, namely by using the 1D thermal-hydraulic code PLTEMP/ANL (PLTEMP), developed at Argonne National Laboratory, and the commercial Computational Fluid Dynamics (CFD) software Ansys CFX (CFX), the investigation of uncertainties requires the implementation of new analysis techniques. For that purpose, the Python tool TAILLEFER is currently being developed at TUM and ILL. Based on Monte-Carlo sampling, TAILLEFER allows for model-independent Sensitivity Analysis (SA) and Uncertainty Propagation (UP) studies of scalar-valued response functions with respect to parameter spaces of arbitrary shape. With the application of TAILLEFER to models implemented in Python or PLTEMP already being shown in the past, the work presented here aims at demonstrating how TAILLEFER can be used with CFX. Using a simplified CFD model of a FRM II cooling channel, the method for code-coupling is explained and verified, and a detailed interpretation of the SA and UP study results is provided.