Conference Agenda

International Atomic Energy Agency conducts Coordinated Research Project (CRP) on “Transition from Forced to Natural Circulation Experiment with Heavy Liquid Metal Loop (NACIE)”. The project is a benchmark simulation of three experiments from NACIE lead-bismuth eutectic (LBE) experimental loop located at the ENEA Brasimone Research Center, in Italy, which includes simulation of transition from forced circulation by gas lift-off pumps to natural circulation. The experimental data from the tests, along with other benchmark specifications, was provided to the CRP participants for analysis with computational codes. The project work is organized in the five work packages (WP), of which WP1 is the system thermal hydraulics. In this work package, participants do calculations and submit the result of the test steady-state and transient simulations with system level codes. The codes used by the CRP participants for the WP1 simulations include: FRTAC, LOCUST, THACS, TRACE, RELAP5, RELAP5-3D, GAMMA+, SPECTRA, THOR, HYDRA-IBRAE/LM, and SAS4A/SASSYS-1.

This paper presents the collective results from all WP1 participants for the benchmark tests, as well as comparison with the experimental data. The results of interest for this package include the void fraction in the gas lift-off region from gas bubble, pressures and temperatures along the loop, and the LBE flow rate. The comparison is presented for the steady-state result at the beginning and end of the test, as well as for the transient results. Several conclusions are drawn from the collective comparison, mostly in terms of where the particular models are different from other codes or the test data.

The Rod Bundle Heat Transfer (RBHT) facility provides advanced and detailed experimental data on coolant flow and heat transfer in a 7x7 fuel bundle model under reflood conditions. Tractebel participated in the first phase of the project (RBHT-I) from 2019 to 2022, using the subchannel code CTF to model reflood conditions at low pressure, with varying flow rates, average power, and subcooled core inlet temperatures. This investigation revealed deficiencies in the CTF reflood model, highlighting the need for improvements, particularly in the flow regime map and the entrainment model. RBHT experiments offer a valuable database for code validation and are utilized in the current investigation to assess enhancements in the latest version of CTF’s reflood models. System codes such as CESAR are also evaluated. The newly implemented models in CTF show improved agreement with experimental data in some cases, especially regarding quenching time. However, they still tend to overestimate the Peak Cladding Temperature and predict a delayed quenching front. CESAR calculations, when coupled with a thermo-mechanics module in the DRACCAR Multiphysics platform (MP), demonstrate high sensitivity to initial conditions, such as the initial rod temperature.

The advanced system analysis tool, System Analysis Module (SAM), and subchannel code MOOSE-Subchannel are both developed under the U. S. DOE-NE’s Nuclear Energy Advanced Modeling and Simulation (NEAMS) program. To improve the prediction accuracy in the reactor fuel assemblies, SAM and MOOSE-Subchannel are coupled with domain overlapping coupling approach. In the coupled simulation, SAM system model provides the overall system behavior, while MOOSE-Subchannel model provides more detailed solution within an assembly. Natural Circulation Experiment (NACIE) facility was a lead-bismuth eutectic (LBE) experimental loop located at the ENEA Brasimone Research Center in Italy, for study of the thermal-hydraulics behavior of LBE in rod bundle configurations. The NACIE loop includes a fuel pin simulator (FPS), which consists of 19 wire-wrapped electrically heated pins. The instrumentations of NACIE can provide temperature at different locations, mass flow rate, pressure during transient tests. In this study, a coupled SAM and MOOSE-Subchannel model of the NACIE loop is developed and benchmarked against experimental measurements. The temperatures from the coupled calculation agree well with the experimental data from thermocouples at various locations, including different points of the loop and local subchannel positions in FPS. Furthermore, the coupled SAM and MOOSE-Subchannel simulation results are compared with SAM standalone results, in which the FPS region is modeled using single-channel or multi-channel approaches.

Core reflooding, the injection of water into the reactor core during a Loss-of-Coolant Accident (LOCA), is a critical Accident Management strategy for water-cooled reactors. As part of the PERFROI project, the COAL experiments were designed by IRSN (now ASNR since January 2025) to study the coolability of intact and partially deformed fuel assemblies under reflooding conditions. Following this program, OECD/NEA/CSNI launched the International Standard Problem (ISP-53) based on the IRSN reflooding COAL experiments, which started in February 2024. This benchmark aims to evaluate the predictive capabilities of computational codes against COAL experimental data, focusing on undeformed and deformed fuel rods. ENEA is contributing to the international benchmark with the two codes DRACCAR and ASTEC. While ASTEC employs a simplified 2D core geometry designed to simulate a comprehensive Severe Accident scenario, DRACCAR features a more detailed 3D core representation with detailed thermo-mechanical modelling of fuel rods. Both codes share CESAR thermal-hydraulic module and include comparable models for reflooding and thermal-hydraulic related phenomena. This paper presents simulation results from both codes for two COAL experiments involving reflooding of a 7x7 undeformed fuel bundle. A detailed comparison against experimental data, along with key thermal-hydraulic parameters analysis, highlights performances and predictive capabilities of the two codes.

When a break occurs in a nuclear reactor, a fast cooldown has to be down to prevent the melting of the core. This is done by the injection of cold water at 7°C, in a pressurized vessel at 295°C. This is a Pressurized Thermal Shock. To improve the safety of the nuclear reactor, the EDF R&D team built an experimental facility in order to analyse the mix of hot and cold water in the downcomer of a 1300 MWe French Pressurized-Water-Reactor. This is the HYBISCUS-II experiment. Salt water at 45°C was injected into stagnant pure water at around 15°C to represent the injection of cold water in hot water. A scaling has been established to link the experiment to the reactor case. Here, we present a numical simulation of the HYBISCUS-II facility, made within the CATHARE code. We use a second scaling, developped for the BORA4-PTS experiment, in order to compare the numerical and the experimental results. The numerical simulations gives results that shows less than 1°C of difference with the experimental ones. With this experiment, we demonstrate the excellent capacity of CATHARE for the modelisation and simulation of a downcomer in a Pressurized Thermal Shock situation.

Within the frame of the European project PASTELS, which aimed to improve understanding of passive safety systems for PWR applications, several experiments were carried out and analyzed numerically. These experiments studied passive systems such as the Safety Condenser (SACO) or the Containment Wall Condenser (CWC). The new databases acquired during the project were simulated by the various project partners using simulation tools at system scale, lumped parameter codes for severe accidents or CFD.

The article will focus on the modeling of the PASI experiments representing a CWC in an open loop configuration. The experimental facility consists of a thermosyphon loop connected to a water pool at ambient pressure located in the upper part, and heated in the lower part through a tubular heat exchanger placed in a vessel which acts as the reactor containment.

A series of ten tests was interpreted by seven different organizations using four system codes (CATHARE, ATHLET, RELAP, TRACE) and two severe accident codes (MELCOR, ASTEC).

The article will briefly present the experimental set-up and the tests carried out. The modeling challenges will then be detailed: the physical phenomena such as condensation, flashing, dynamic instability, thermal stratification, and the different geometric domains to be represented (pipe, heat exchanger, volumes). Finally, the article will present different strategies to model these interrelated phenomena, and will discuss the main results, with a particular focus on two phase flow instabilities.