Conference Agenda

Due to the advantages of efficiency and flexibility, floating nuclear power plants have become a focal point for research and development across various countries worldwide. In marine conditions, the movement of vessels alters the Critical Heat Flux (CHF) characteristics of nuclear reactors, which is essential to be reconsidered. In this paper, the CHF experiment, operated with R134a in the pressure range of 1.6-2.7 MPa and the mass flux range of 1000–3000 kg·m-2·s-1, has been conducted in both vertical and inclined conditions. The test section consists of a movable tube with an inner diameter of 8 mm and a heated length of 0.8 m or 1.6 m. The experimental results show that as the critical quality increases, the effect of inclination on CHF changes from reduction to no effect. Computational Fluid Dynamics (CFD) method was employed to simulate the experiment with inclination angles of 0° and 25°. The results indicate that the inclination causes a shift in the symmetrical distribution of the flow field, with a particularly significant impact observed on void fraction distribution. Bubbles tend to migrate towards the upper part of the inclined tube, leading to the accumulation of bubbles. Meanwhile, the liquid also supplements the upper wall. It may be the combined effect between the two that influences the reduction or invariability of the CHF in the inclined tube.

Accurate prediction of rod-bundle critical heat flux (CHF) remains a great challenge in evaluating the thermal safety margin of a nuclear reactor due to the lack of realistic models and experimental databases for the complex CHF phenomenon. This study examines the occurrence of CHF in a rod bundle using the MARS-KS subchannel analysis to verify CHF models and to provide useful supplements to CHF modeling. The examination uses the Wisconsin 2x2 rod bundle CHF experimental data. The CHF is detected by a rapid rise of the rod surface temperature with a step flow reduction or a step power increase, and the detected CHF values are compared with the measured values and with potential CHF mechanistic models (e.g., bubble crowding and sublayer dryout). The influence of radial/axial power distribution, space grid, cold wall, and bundle size on CHF is evaluated. Especially, CHF under the low flow low pressure is emphasized. The study is expected to provide a realistic methodology for evaluating CHF models based on more actual flow conditions and to broaden understanding of important factors affecting rod-bundle CHF.

Subchannel analysis plays an important role in nuclear reactor safety analysis, enabling better core thermal hydraulic predictions of parameters like the critical heat flux (CHF). This research focuses on developing a benchmark exercise for the COBRA-TF (CTF) computational code by performing subchannel analysis of the Electric Power Research Institute (EPRI) CHF database. The EPRI database comprises over eleven thousand experimental data points from rod bundles with diverse geometries, with uniform and non-uniform axial heat flux distributions. Operating conditions range widely, with pressures from 1 MPa to 17 MPa and mass fluxes from 50 kg/m²s to 6000 kg/m²s, providing a robust foundation for assessing CHF models. This benchmark aims to systematically compare the performance of widely used CHF correlations, including the Look-up table, Biasi, and W-3 correlations, under varying flow regimes and geometrical configurations. We aim to assess CTF flow regime and heat transfer models against CHF predictability. This work is expected to enhance the fidelity of CTF predictions and improve safety margin and performance evaluations in nuclear reactor design and operation. Furthermore, the benchmark will be a valuable resource for validating and refining CHF models.

The US nuclear industry is exploring an option to improve operational economics for the current fleet by seeking a cladding-performance based safety criteria as opposed to the current limit requiring complete avoidance of critical heat flux (CHF). Past experience has shown that not all events leading to a dryout are severe enough to cause fuel performance degradation. Allowing temporary dryout of the fuel, known as time-at-temperature (TaT), could allow for economic improvements to current plants without compromising fuel integrity. To support this effort, a comprehensive program is being executed by the US Department of Energy that includes generating cladding material data under TaT conditions, development of new mechanistic models, and demonstration of modeling and simulation capabilities for transients of interest. This paper presents current work done to assess the thermal-hydraulic subchannel code, CTF, which is a package used in the VERA core simulator, which will ultimately be used for TaT analysis. CTF will provide the T/H boundary conditions that will be needed for fuel performance analysis in the Bison code and it will therefore be necessary to quantify both the accuracy and uncertainty of post-CHF models. This paper presents the results of using the BFBT and Harwell tests for CTF validation, which both experience dryout conditions. This work also led to the implementation of an alternate post-CHF heat transfer modeling package that leads to improved agreement with experimental data. Agreement with experimental data for tube-geometry is generally good, but biases were detected for rod-bundle geometry that will require future improvements.

The CTF subchannel code which is used for the Thermal Hydraulic (T/H) solution in the Virtual Environment for Reactor Applications (VERA) is supporting the light water reactor application area in the Nuclear Energy Advanced Modeling and Simulation (NEAMS) program. Under this program, the ability of CTF is sought to be improved to model dry-out and re-wetting behavior in BWRs, which impacts its ability to model anticipated operational occurrence (AOO) transients using the time-at-temperature (TAT) approach which aims to demonstrate that the fuel rod’s integrity is not challenged during a mild elevated temperature transient. These models must be validated so that the thermal-hydraulic behavior can be used as boundary conditions in fuel performance codes. In order to improve CTF’s dryout location prediction and the post-dryout behavior prediction, the primary goal of this study is to perform sensitivity analysis based on SOBOL indices and other sensitivity analysis methods to identify the physical models which most affect the Figure of merit (FOM) in the flow regimes of interest. A multitude of tests will be used for model improvement such as the harwell tests, the BFBT turbine trip test, the FEBA tests etc., which are all part of the CTF V&V assessment suite, as well as expanding the test suite with the IFA613 tests. The second goal of the study is to perform model calibration using a Bayesian approach, which will also provide model uncertainty that will be used in a future uncertainty quantification study.

OpenSTREAM is a new open-source, one-dimensional, flexible computational environment designed to simulate boiling two-phase flows in single straight channels using various multi-field solvers ranging from the homogeneous equilibrium model to an advanced four-field model of annular two-phase flow. This paper applies OpenSTREAM’s three-field model to simulate a series of tests conducted at the Karlstein Thermal Hydraulic (KATHY) Test Loop under Boiling Water Reactor (BWR) conditions, including core instabilities. The 10´10 rod bundle geometry is represented in the code as a three-wall channel, accounting for (1) the adiabatic fuel shroud and central water channel, (2) the fuel rod with the highest radial power peaking factor, and (3) the remaining fuel rods. Initial simulations of single- and two-phase pressure drop tests are performed to calibrate the pressure loss coefficients of the spacer grids. A feature to account for enhanced droplet deposition downstream of the spacer grids is implemented in OpenSTREAM and calibrated against critical power tests. This feature enables accurate prediction of critical power and its associated elevation, determined by iterating the power until complete liquid film dryout is achieved anywhere on the hot rod. The simulation results show consistent agreement with the experimental data for the steady-state critical power across the range of tested boundary conditions. Preliminary transient simulations show that OpenSTREAM can predict dryout and rewet with time delays from inlet conditions representative of density waves.