The main objectives of the meeting are to:
IMPORTANT: The Call for Papers is now open. Contributions selected for Oral presentation may now submit full papers as specified in the Event Information Sheet. Contributions selected for Poster presentations may now submit extended abstracts (2-3 pages) using the same full paper template.
Banner image reference: INTERNATIONAL ATOMIC ENERGY AGENCY, Benchmark Analysis of EBR-II Shutdown Heat Removal Tests, IAEA-TECDOC-1819, IAEA, Vienna (2017).
The Versatile Test Reactor (VTR) is currently under development by the US Department of Energy. This reactor will rely on fast neutrons to enable novel and wide-ranging experiments to support development of various advanced reactor technologies. With the high flux achievable, accelerated testing of fuel and materials will be made possible. To support VTR design efforts, an experimental facility has been designed and constructed at Argonne National Laboratory to create the hydraulic flow conditions within the VTR’s nuclear core region. This facility, the Pressure drop Experimental Loop for Investigations of Core Assemblies in advanced Nuclear reactors, PELICAN, measures the pressure drop across a full-scale fuel assembly containing prototypic axial reflectors, fuel, and plena components. The PELICAN facility was designed and built to offer maximum flexibility, allowing testing from short sub-sections all the way to the full-length VTR assemblies. Subcooled water at elevated temperatures and pressures is used to match the thermophysical properties of liquid sodium. A 50-HP centrifugal pump, capable of providing flow rates up to 40 kg/s across the test section, is used to match the full-scale Reynolds numbers and flow velocities expected within the prototypic VTR design. The hexagonal test section reflects the true dimensions of the fuel assembly ducts planned for the VTR core and is expandable up to 4.0-m in length. In this work, we describe the design and construction of PELICAN, planned test articles, and recent results from the testing of these articles over a broad range of flow rates and temperatures that cover the expected operational conditions of the VTR. The as-built facility is actively generating empirical data for verification and benchmarking of computational models and is well poised to provide continuous support of the VTR program as the reactor design continues to mature.
Turbulent heat transfer is a complex phenomenon, which has become the focus of turbulence modelling research in recent years. The closure of turbulent heat flux has conventionally been approached by the so-called eddy diffusivity approach and its most trivial version, the Reynolds analogy. While this approach provides a simple and efficient closure, it lacks accuracy when the similarity hypothesis between thermal and momentum fields are less justified, i.e. in presence of low Prandtl number fluids, such as liquid metals, for which reference data are scarce. The present paper discusses the recent advancements in heat flux modelling approaches, including the closures of local turbulent Prandtl number models, explicit and implicit algebraic models with special attention to low-Prandtl cases.
Although these recently developed models provide a better alternative to the conventional approach, they also suffer from limitations of their own. The present paper provides a critical review of these shortcomings, including isotropic nature, wall-resolving low-Reynolds number approach, and the need for a priori knowledge of flow and heat transfer regimes. Another major criteria to rank such models is their applicability to an ‘integral setting’ where multiple flow regimes exist in a single flow domain. Under the framework of the collaborative European PATRICIA project, it is planned to further develop and enhance the current heat flux modelling approaches to overcome these shortcomings. After rigorous testing and validation, the models are planned to be applied to prediction of heat transfer in an integral flow case over complex geometry.
This work reports post-processed data of the experimental campaign carried out in the HLM-operated NACIE-UP facility in the framework of the HORIZON2020 SESAME project. NACIE-UP is a rectangular loop cooled by lead-bismuth eutectic. A prototypical wire-spaced fuel pin bundle simulator (FPS) is installed in the bottom part of the riser, while a shell and tubes heat exchanger (HX) is placed in the upper part of the right descending vertical branch. The difference in height between the heat source (FPS) and heat sink (HX) is about 5.5m and allows the establishment of the natural circulation regime inside the loop. The mass flow rate is measured by a prototypical thermal flow meter. The forced circulation is realized by a gas lift pumping in the riser. Several thermocouples measure temperatures along the loop while the FPS is instrumented with 67 N-type thermocouples.
A PLOFA test is presented in the paper with a power transition from 100 kW to 10 kW and the stop of the pumping gas-lift. Temperature trends showed a coherent behavior with a sharp decrease due to the power decrease followed by local maximum due to the gas-lift stop. The time trend of the main thermal-hydraulic parameters during the transient are illustrated in details. From the experimental data, it is proved that the thermal field develops along the FPS with larger radial thermal differences in top monitored section with respect to the bottom one.
Nusselt numbers in the fully developed top section were computed and exhibited values close to the Kazimi-Carelli correlation. On the initial and final steady states, a statistical analysis was carried out to determine average overall and local values, and the associated uncertainties. The error propagation theory was applied for the derived quantities.
The formation of velocity and temperature fields in the fuel assemblies of fast reactor core occurs under the influence a lot of factors. It is shown, that the most important factors are the deformations during the campaign under the influence of temperature irregularities and radiation effects. The results of studies of the velocity fields and shear stress, turbulence microstructure in the central and peripheral areas of fuel assemblies, as well as in case of the rod lattice deformation are presented and analyzed. An intensification of turbulent momentum transfer in channels in azimuthal and radial directions in the area of gaps between the rods is demonstrated. The performed analysis is indicated that there are a significant difference between the experimental dependences for turbulent momentum transfer coefficients in the radial and azimuthal directions and calculated within the framework of semiempirical models of turbulent momentum transfer as well as the anisotropy coefficients of the turbulent momentum transfer in rod bundles. The results of benchmark on the thermohydraulics of fuel assemblies showed that the common commercial codes describe experimental data only approximately. It is shown, that the intensification of turbulent momentum transfer in the channels of rod assemblies is due to the appearance of large-scale turbulent momentum transfer (secondary currents). The contribution of large-scale turbulent momentum transfer to the turbulent momentum transfer coefficients in the channels of rod assemblies is calculated. The dependence for coefficient of inter-channel turbulent exchange of momentum is obtained, and the intensification of inter-channel turbulent exchange in close rod lattices is explained. A dependence for the dissimilarity coefficients of forced inter-channel convective exchange of momentum and mass, as well as energy and mass in rod bundles spaced by wire winding is obtained. The calculation methods and the numerical modeling results for temperature regime of fuel assemblies with randomly distributed initial parameters by the Monte Carlo method are presented as well as the thermomechanical analysis of the temperature field in the fuel assemblies during the campaign. An idea about the equilibrium configuration of a fuel rod bundle in a hexahedral cladding during irradiation, the stress-strain state of an individual fuel rod and a fuel assembly cladding is obtained. The tasks of further investigations are formulated and discussed.
The Steam Generator Tube Rupture postulated event in a pool type Gen IV Heavy Liquid Metal cooled Fast Reactor system is investigated from the point of view of the possible hazardous consequences affecting the structural integrity of internals. The selection of the limiting SGTR initiating event is based on the assumption that the effects of a single break envelopes all effects from realistic tube leak to the multi-tube rupture propagation mechanisms. The justification of this assumption requires demonstration that the adjacent tubes are not subjected to excessive mechanical loading and therefore the scenario of multiple tube ruptures cannot occur. LIFUS5/Mod2 facility is a separate effect test facility aimed at investigating the heavy liquid metal-water interaction. It has been designed and constructed to withstands high pressure and temperature (i.e. up to 200bar and up to 500°C) and to record the fast pressure transients triggered by the water flashing in heavy liquid metal melt. This type of transient is typical of a Steam Generator Tube Rupture event in a pool type Gen IV Heavy Liquid Metal cooled Fast Reactor system. The experimental campaign B-series, presented in this paper, was executed using a test section based on a vertical tube bundle of 188 tubes. The dummy central tube simulates the rupture injecting water at about 180 bar and 270°C into Lead Bismuth Eutectic at 400°C. Seven experimental tests were performed with injection orifices equal to 10%, 50%, and 100% of the reference steam generator tube flow area. The experiments are aimed at investigating and evaluating the mechanical effects on tubes and shell surrounding the injector. The experimental pressure, temperature, strain and mass flow rate time trends provide the characterization of heavy liquid metal- water interaction phenomena and a detailed database of data for codes’ validation. The experimental campaign B-series showed limited strain consequences on the test section tubes and shell during the transients.
Over the last 20 years, research and development on fast reactors and in particular on Generation IV concepts cooled by heavy liquid metals has been greatly supported through several national and international funds. Looking at the European community, the technology benefitted from approximately 145 M€ of investments, increasing the technology readiness level through various areas including safety and resilience, neutronics and thermal hydraulics. One of the projects currently pursued in Europe and supported by numerous research centers, private companies and member states is ALFRED (Advanced Lead Cooled Fast Reactor European Demonstrator), born around 10 years ago within the LEADER project. Following the FP7 project, Romania declared its support in ALFRED development by proposing itself as hosting country for the demonstrator and at the same time the FALCON consortium (Fostering ALfred CONstruction) was set out, consisting of Ansaldo Nucleare, ENEA and RATEN-ICN. In recent years, the consortium has further revised the reactor concept to solve some thermal hydraulic issues with an increase of technology know-how. The changes were supported by a new deployment strategy based on different power stages, appointing ALFRED as the prototype of an SMR-type LFR, adding to the formal role as the demonstrator. To facilitate this process, several experimental campaigns and numerical analyses carried out by the consortium members and supporting organizations have been of extreme importance. The areas that benefited mostly from this effort were the thermal hydraulics of the reactor pool, of the core and of the safety systems, thanks to numerical studies made with system codes, computational fluid dynamic tools and, above all, to the technological infrastructure owned by ENEA. In this article we present the new plant concept, reporting some of the most important results that have made it possible to solve complex problems such as thermal striping, thermal stratification and freezing of lead by overcooling. Lastly, this article aims to highlight the next steps planned for the development of ALFRED, such as the construction of various experimental facilities that will develop the test campaigns for the safety justification and licensing of the plant.
The results of experimental investigations in the field of hydrodynamics and heat exchange of fast reactors and accelerator-driven system with liquid metal coolants are presented, and the problems and issues for further investigations are formulated. Physical phenomena, effects, laws and process characteristics occurring in reactors are considered and analyzed, including a flow path, a core, a steam generator, etc. Special attention is paid to the results of hydrodynamics and heat exchange studies in channels and structural elements of fast reactors: velocity and temperature fields, structure and characteristics of turbulent transfer of momentum and energy, hydraulic resistance of channels and fuel rod assemblies, hydrodynamics of collector systems, vibro-acoustics, heat transfer in channels and fuel rod assemblies cooling with liquid metals, contact thermal resistance, inter-channel exchange, simulation of mixing flows with different temperatures.
The data of experimental studies are presented for a single-tube model of a large-module steam generator, and for a fragmentary thermohydraulic model of a steam generator of a reactor with twisted steam-generating tubes operating at subcritical and supercritical water pressure.
The results of investigations of temperature and velocity fields on a small-scale water model of a fast reactor vessel with an integral layout in nominal, transient and emergency operating regimes are demonstrated. It is shown that the effect of thermogravitational forces leads to temperature stratification with stagnant and recirculating formations, restructuring of the flow and temperature regime.
The data of liquid metal boiling in a large volume and in fuel rod bundles, liquid metal condensation are analyzed. It is shown that the liquid metal boiling process in channels and fuel rod bundles is formed under the influence of various factors, has a complex structure, is characterized by both stable (bubble, annular- dispersed) and pulsation (slug) regimes with significant fluctuations of technological parameters (flow rate, pressure, temperature), which can last for tens of seconds and cause a crisis of heat transfer. Heat transfer was studied, a cartogram of two-phase flow regimes was obtained for liquid metal boiling in fuel rod bundles, the effect of the surface roughness of fuel rods on liquid metal heat transfer and boiling regimes in fuel rod bundles was found. The principle possibility of long-term stable cooling of the core during sodium boiling was shown by using of a new technical solution – "sodium cavity" above the reactor core.
As a result of studies of the degradation of the simulated fuel assembly in fast reactor core during the thermal interaction of uranium-containing fuel simulators (high-temperature destruction of fuel rods) with static sodium, the thermal interaction of the corium simulators with sodium, the kinetic and mechanical characteristics of the process and their dependence on temperature, hydrodynamic parameters and design system were determined.
Information about the key problems of thermo-physical investigations in relation to the development of innovative nuclear energy technologies such as a high-temperature fast reactor with a sodium coolant, reactors with a fast neutron spectrum is given.
In the last years, Japan Atomic Energy Agency proceeded R&Ds to enhance the safety of a sodium cooled fast reactor. The decay heat removal (DHR) including situations under post-accident heat removal (PAHR) is a central topic in the R&Ds, and two experimental research programs, named PHEASANT and PLANDTL-2, have been performed to provide the knowledge on the thermal hydraulics on the DHR and to accumulate experimental data for V&V of numerical codes.
A water experiment, PHEASANT, has a 1m class of cylindrical vessel with three types of DHR systems i.e., a dipped type of direct heat exchanger (D-DHX) and a penetrated type of direct heat exchanger and simulated reactor vessel wall cooling system. The simulated core is modeled by concentric three circular layers, and inner two layers have electric heaters to simulate the decay heat. PHEASANT also has electric heaters both in the upper and lower plena to simulate the decay heat from the debris for PAHR experiments. Thus, PHEASANT enables to examine the overall thermal hydraulics inside the vessel and thermal hydraulic interactions between different two types of DHR systems under DHR operating conditions including PAHR. Another sodium experiment, PLANDTL-2, have a 1m class of simulated core with 1MW of electric heater output and a 2m class of upper plenum with a D-DHX. The simulated core is formed by 55 hexagonal-shaped wrapper tube channels, with the inter-wrapper gap. Therefore, PLANDTL-2 can provide knowledge on core cooling behavior under D-DHX operating conditions, such as the penetration of cold sodium from the D-DHX into the inter-wrapper gap and wrapper tube channels.
This paper describes major outcomes from PHEASANT and PLANDTL-2. In PHEASANT, PAHR experiments were performed and overall flow paths were grasped with detailed temperature and velocity distributions. In PLANDTL-2, experimental data under D-DHX operating conditions were expanded including temperature distributions in the inter-wrapper gap.
MYRRHA is a flexible fast-spectrum pool-type research reactor cooled by lead bismuth eutectic (LBE), under development at the Belgian Nuclear Research Centre (SCK CEN). The research and development program (R&D) supports both the design and the safety assessment of the reactor in a context of pre-licensing. At this stage, the R&D activities aim to bridge the gap in knowledge in several disciplines like fuel behaviour, LBE chemistry and materials corrosion, and thermal-hydraulics, needed to increase the confidence in the prediction of the reactor performance in different scenarios.
In this context, thermal-hydraulic experiments are necessary at different scales. Separate-effect tests allow the investigation of basic phenomena, such as turbulent heat transfer in liquid metals. Prototypical tests can provide a direct confirmation of the thermal-hydraulic behaviour of a given component. Integral tests represent the dynamic evolution of the system accounting for the interaction between components. As a consequence, the complete study requires many experimental facilities focusing on specific features and with specific requirements in terms of instrumentation, supported by numerical simulations. In addition to their own experimental and simulation activities, SCK CEN also relies strongly on collaboration with partner research institutions worldwide.
This article covers the main achievements from thermal-hydraulic experiments in recent years and it outlines the further needs identified for the near future. Examples of completed experiments related to fuel assembly and pool thermal-hydraulics are presented. Some ongoing activities are related to key components such as the primary heat exchanger and primary pump.
This paper presents the results of subsystem efficiency calculating for monitoring small leaks of a reverse-type steam generator (RSG) of the MBIR.
This subsystem is a part of the automatic protection reverse steam generator system (APS RSG) and is designed to detect small leaks of water into sodium and generate a signal to turn off the steam generator.
Small leaks of water into sodium don’t cause noticeable hydrodynamic effects in the second sodium loop. Therefore, to detect small leaks, special devices are used to detect the products of the sodium with water reaction. At the MBIR, sensors for monitoring dissolved hydrogen in sodium (EHDV-N), gaseous hydrogen in a sodium flow (IRIS/Taran) and gaseous hydrogen in a cover gas (EHDV-G) are used.
The SLEAK code is used to calculate the small leak control subsystem efficiency. The SLEAK code allows to calculate the concentration of dissolved hydrogen and oxygen in sodium, the volumetric content of hydrogen gas in the sodium flow and in the cover gas; the time of reaching the emergency setpoint by the leak control sensors; the time of self-destruction (propogation) of the initial leak.
It’s shown that the MBIR small leak control subsystem provides 100% efficiency of detection of inter-circuit leakage (before the appearance of the secondary defects) in the range of initial leaks from 0.025 g/s to 10 g/s. And, with a leakage rate more than 0.2 g/s, the instruments readings for monitoring gaseous hydrogen in the sodium flow make it possible to determine the failure RSG module.
Prototype Fast Breeder Reactor (PFBR) is a 500 MWe pool type liquid sodium cooled nuclear reactor presently under commissioning at Kalpakkam, India. The design for next generation higher capacity Fast Breeder Reactors 1&2 (FBR1&2) has been commenced. The intermediate heat exchangers (IHX) of FBR1&2 are typical shell and tube type counter flow heat exchanger used for transferring heat from primary sodium system to secondary sodium system. Primary sodium enters the shell side of IHX through an inlet window at the top and exits through an outlet window located at the bottom. The secondary sodium enters IHX from top into a down comer at the centre and flows downward to an inlet plenum also called as bottom header and then flows upwards through the tubes to outlet header. There are 3900 tubes arranged in circular pitch surrounding the central down comer in 25 rows. The heat exchanged by various tubes of IHX is not the same due to the following reasons: (a) cross flow of primary sodium at the inlet and outlet window regions because of which inner rows sees lower temperature sodium, and (b) primary sodium flow near the inner rows in less when compared to outer rows. Consequently, temperature of secondary sodium at the outlet of various tubes is not the same resulting in thermal loading of tube sheet and other structures of IHX. Since the secondary sodium flowing in the outer rows receive significantly large heat compared to the inner rows, temperature of secondary sodium at the outlet of various tubes can be made more uniform by admitting more flow through the outer rows. This is possible by increasing the hydraulic resistance of inner rows compared to the outer rows. A desirable option is to introduce a flow distribution device (FDD) in the bottom header to accomplish the required flow zoning. Based on a simplified 1-D network model it is recommended that outer 7 rows of tubes in IHX should have 30 % more flow rate compared to the 18 inner rows.
Three dimensional CFD study of the bottom header of IHX of FBRs has been carried out to explore the ways to simplify the design of FDD. Parametric studies have been carried out to achieve the desired flow distribution. The effect of conical diffuser and vertical baffle inside bottom header is quantified. A vertical baffle of 225 mm height (located 12 mm below the tube sheet) provided after the 18th row of tubes rendered a flow distribution very close to the desired one. With this arrangement, the average absolute deviation between the flow distribution achieved and the desired one is ~ 10 %.
Full paper will discuss in detail about the modelling strategy, solution technique and the results of the parametric studies.
Subchannel analysis is the common method in thermal-hydraulics (T/H) analysis of Fast Reactors (FR). Due to the limitation of computation time and machine storage, traditional subchannel codes generally adopt the simplified or rough geometric model for core calculation, such as calculating only for a single assembly or treating an assembly as a sub-channel. This paper focuses on development and application of a new subchannel code to mini- and whole-core FR calculations on the pin-cell resolved level. FR is the typical hexagonal-assembly reactor in which there is little interaction between assemblies and each assembly can be calculated independently. The computation of each assembly can be assigned to different computation processes simultaneously. To equalizes the pressure loss, an outer pressure iteration loop is applied to adjust the inlet flow rate over all assemblies. Subsequent analysis results imply that the code can be used to model and perform pin-level core safety analysis with acceptable computational efficiency. The work in this paper is significant for the parallel simulation for thermal-hydraulics of virtual reactors.
While testing materials in research reactors, heat is generated in the test chamber due to neutron/gamma heating. It is required to remove this heat to maintain the specimen at required temperature and also to maintain the integrity of structures. Hence, cooling of the chamber becomes essential. The amount of heat generated is huge and the space available for cooling arrangement is usually limited. Liquid sodium becomes a coolant of choice for this purpose because of its good heat transfer characteristics. However, sodium inventory in the cooling systems is limited from safety considerations. A cooling arrangement with an oscillating sodium column which in turn is cooled by another circulating fluid has been proposed in order to overcome these difficulties. In this, the specimen chamber at the bottom is connected to two limbs/legs half-filled with liquid sodium. Argon cover gas is present above sodium to prevent its contact with air. The level of sodium in the limbs is varied continuously with the help of an oscillator to cool the specimen chamber. Outer tubes are provided around the limbs and form the annular cooling jackets through which helium fluid is passed to remove the heat from sodium in the limbs. A similar cooling arrangement has been proposed for testing materials in Fast Breeder Test Reactor (FBTR). RISHI (Research facility for Irradiation studies in Sodium at HIgh temperature) loop has been set up in IGCAR, Kalpakkam to test the functionality of this arrangement.
Mathematical modeling of the cooling arrangement is essential for its efficient design. The oscillating nature of sodium makes the prediction of heat transfer among various components in the arrangement a challenging task. Modeling of the arrangement has been carried out using general purpose in-house developed system dynamics code PINET. One dimensional mass, momentum, and energy equation for fluid and two-dimensional energy equation for solid are solved in a coupled way using finite difference method in this code. The code has been validated against various thermal fluid phenomenon such as pressure transients, conjugate heat transfer, natural circulation etc. For the current problem, sodium columns are modeled using pipe (with interface) component, the structures are modeled with heat slab components and the helium jackets are modeled with pipe (without interface) component. Conduction heat transfer in the structures and convection heat transfer between structure and fluid are considered in the study.
Simulations were carried out firstly for a reference case and then parametric studies with different parameters were carried out to understand their influences on heat removal rate. For the reference case with helium inlet temperature as 313 K and flow rate as 40 CMH (at STP conditions), the average specimen chamber temperature was predicted as ~445 K and its range of variation was predicted as 1.1 K. The total heat to be removed by cooling helium was ~1.7 kW. Parametric studies were carried out with different helium inlet temperatures and mass flow rates. Chamber temperature increases with increase in helium inlet temperature and decreases with increase in helium flow rate.
The helical coil heat exchanger is a tube and shell heat exchanger, used in multiple fast reactor designs such as the MONJU fast breeder reactor and KAERI’s KALIMER -600 liquid metal reactor, with concentric tube bundles that coil around an axis as compared to conventional straight tubes. The complex tube bundle geometry increases the turbulence of the shell-side fluid. While this increases the heat transfer efficiency of the system, flow phenomenon such as vortex shedding and bulk flow anomalies increases the potential forces acting on the tubes. Previous studies have focused on typical tube and shell tube configurations as well as allowing a single tube within the bundle to vibrate. An experimental study was conducted to determine the effect the vibration of adjacent tubes have on the flow-induced vibration of each other. A model helical coil heat exchanger was constructed where tubes within the center of the bundle were allowed two-dimensional self-excited vibration at Re = 7,500. High speed-camera images taken at 1,000 fps captured the motion of the vibrating tubes where springs were used to represent structural vibration damping mechanisms. Experiments showed that the addition of a vibrating upstream or downstream tube increased the frequency of the vibration from 10.3 Hz to 11.6 Hz and 8.6 Hz to 11.6 Hz, respectively. The amplitude of the vibration also changes from fluctuating to structured when either an upstream or downstream tube is allowed to vibrate. Results support previous studies that showed upstream vibrating tube influence but introduce the influence a downstream tube affects the vibration of an adjacent tube. Frequency and amplitude changes of the vibration also suggest a redistribution and stabilization of energy when two adjacent tubes are affected by flow-induced vibration compared to a single tube affected.
Upper plenum temperature distribution in advanced nuclear reactors as liquid metal, molten salts and gas cooled reactors is a vital aspect of the performance of FAST reactors. The mixing and circulation of fluid in the upper plenum plays an important role in performance and thermal efficiency. High fidelity experimental data is needed for validation of a jet supplying liquid into a hemispherical upper plenum with a cooling jacket around the outlet channel. In order to study the temperature distribution in the upper plenum, a rack of thermocouples was placed in a manner that would not disrupt the flow upstream of the thermocouples and would allow for measurement in both the radial and axial dimensions. The experiment was conducted over a twenty-four-hour period beginning with all water at room temperature. The data was recorded for two hours to establish a baseline condition and then the heated pipe was turned on. After approximately seventeen hours of heating, a steady state condition was reached. At this point, the thermocouple rack, comprised of eight radially distributed thermocouples, was displaced axially at various elevations for one hour per location. As a result of this process, the transient temperature distribution at the lowest elevation was measured over eighteen hours and the steady state distribution in the lower half of the plenum was measured for one hour at each location. The sampling rate of the thermocouples was four Hz, matching the time constant of the bare welded thermocouple wire. The analysis of the temperature distribution includes the average and transient temperature distributions with respect to radial, axial, and total distance from the inlet of the plenum and a one-way analysis of variance was performed to determine the individual effects of each element of displacement. These results were compared with previous experimentation utilizing linearly distributed optical fiber measurements and demonstrate a similarly flat distribution with respect to the radial temperature distribution outside of the jet core. The primary advantage of thermocouple measurement is the relative thermal isolation between measurement points allowing for more accurate measurement of the temperature in the core without conducting heat along the fiber sheathing radially. This results in a measurement demonstrating a higher temperature gradient from jet centerline to shear layer, and from the shear layer to the outer radial positions while matching the fiber measured gradient fairly well outside of the core and shear layer.
The paper is centered around the design and modeling of the bayonet tube steam generator, part of the lead-cooled reactor (LFR). Such a model of steam generator is expected to be installed in a 300 MW (thermal power) plant called Advanced Lead Fast Reactor European Demonstrator (ALFRED) in the near future. The model was built using finite element analysis software with multiphysical capabilities and the results were compared with similar models. The paper presents calculation elements regarding the heat transfer for a single bayonet tube, the final heat exchanger having in its component over 500 tubes.
Trouble free operation of steam generator is a key factor for the plant availability of sodium cooled fast reactors. Hence to validate the design of 157 MW t steam generator (SG) module for PFBR, a model of SG was tested in Steam Generator Test Facility in IGCAR. Thermal hydraulic studies were carried out with the 19 tube once through sodium heated steam generator model to characterize the heat transfer and stability behavior of once through SG used in PFBR. The model SG is with same tube size and tube material as in the PFBR SG. It is designed to operate with the same process conditions but with a nominal heat transfer rating of only 5.5MWt. Nominal sodium inlet temperature to the steam generator is 525°C and it is intended to produce steam at 493°C temperature and 17.2Mpa pressure. Heat transfer capability of the steam generator was assessed and confirmed by experiments. The testing revealed the adequacy of heat transfer capability of the steam generator to transfer the intended power. From the experimental data it is estimated that the steam generator has 8.3% more tube surface area than the required to produce steam at nominal conditions. The model steam generator was subjected to different design basis simulated plant transients and the thermal loading on the thick tube sheets were evaluated. Experiments were conducted to map the stable operating regimes of steam generator during normal operation and plant transients.
For demonstrating the functionality of the Safety Grade Decay Heat Removal System of PFBR and characterising the same, a test facility named SADHANA with 1:22 scaled down in nominal power was realized. Scaling of the system was performed based on the philosophy of Richardson number similitude. Heat transport capacity of system was experimentally demonstrated. From the experiments, it was found that, heat transported by the system was 19.4% higher than the nominal heat transport capacity. Thus the effectiveness of heat exchangers and the system was proven. Stability of sodium flow and rate of heat transport were characterized during different design basis plant transients by simulating these events in SADHANA facility. Further the experimental facility was utilised to study the response time for setting up heat removal followed by SCRAM and Station Blackout situations.
This paper brings out the details of the experimental facility and heat transfer experiments carried out for normal heat transport path and safety grade heat transport path of PFBR.
Prototype Fast Breeder Reactor (PFBR)  is a 500 MWe pool type liquid sodium cooled nuclear reactor presently under commissioning at Kalpakkam, India. The design for next generation higher capacity Fast Breeder Reactors 1&2 (FBR1&2)  has been commenced. The experience gained from the design and construction of PFBR has been utilized in the optimized design of FBR1&2 with enhanced safety and improved economy as the main targets. FBR1&2 is a pool type 500 MWe fast breeder reactor adopting twin-unit concept. The design changes envisaged for FBR1&2 includes, (i) four primary pipe per sodium pump, (ii) inner vessel with single torus lower shell, (iii) reduced main vessel diameter with narrow gap cooling baffles, and (iv) safety vessel integrated with reactor vault. This paper discussed about the 3D CFD thermal hydraulics studies carried out for the FBR1&2 in the following topics:
(a) Studies on optimization of the location & number of anti-gas entrainment baffles towards reducing the gas entrainment from the sodium free surface,
(b) Prediction of velocity and temperature distribution in the inlet & outlet window of intermediate heat exchanger (IHX) for providing the input to FIV studies of tube bundle and also to estimate the temperature variation in the outlet plenum of the IHX,
(c) Assessing the adequacy of the decay heat exchanger (DHX) toward removing the decay heat,
(d) Transient response of hot pool to reactor thermal hydraulic transients for estimating the transient thermal load on the hot pool components,
(e) Steady & transient study of the integrated hot pool – cold pool for estimating the heat transfer through inner vessel and also the transient thermal load on the cold pool and hot pool components,
(f) Flow distribution in inlet bottom header of IHX towards optimizing the flow distribution device, and
(g) Optimization of flow distribution device inside spherical header for reducing the pressure drop.
Full paper discusses about the methodology, solution and the findings of the parametric studies carried out.
 Chetal, S. C., Balasubramaniyan, V., Chellapandi, P., Mohanakrishnan, P., Puthiyavinayagam, P., Pillai, C. P., Raghupathy, S., Shanmugham, T. K., Sivathanu Pillai, C., “The design of the prototype fast breeder reactor”, Nucl. Eng. Design, 236, pp. 852–860 (2006).
 Puthiya Vinayagam, P., and Chellapandi, P., “Sustainable Energy Security from Fast Breeder Reactors”, 6th Nuclear Energy Conclave, New Delhi, October (2014).
Obtaining a good representation of the phenomena involved in liquid metals thermal hydraulics still represents one of the key issues to be solved for the development of the upcoming GEN IV Liquid Metal Fast Reactors (LMFRs). During the last decade the European Union funded numerous projects with the aim to pave the way for such a technology and, consequently, several experimental campaigns were performed. The University of Pisa joined the common effort providing numerical analyses in support of the experimental campaigns both in the pre-test and post-test phases. The recent activation of the new EU project PATRICIA, to which the University of Pisa participates as well, provided room for a further investigation of the involved phenomena by means of both experimental and numerical analyses.
In the frame of the PATRICIA project, the ENEA Brasimone research centre foresees to carry-out an experimental campaign involving the well-known CIRCE facility. The facility will be updated substituting one of its key features, the steam generator, which will now take into account an helicoidal steam generator (THETIS): the experimental campaign will involve the analysis of steady-state and postulated accidental scenarios. The University of Pisa will provide numerical support both in the pre-test and post-test phases to assist the design of the experiment, e.g., suggesting a possible thermocouple layouts that may help in better measuring the occurring thermal phenomena and further validating the capabilities of the adopted numerical models.
This work reports about the preliminary results obtained by numerical simulations of the CIRCE-THETIS facility. The analyses were performed both using CFD codes (ANSYS Fluent) and STH codes (RELAP5-3D). The analyses mainly focused on temperature and velocity distributions inside the CIRCE pool during the postulated nominal steady-state conditions; three transient analyses were also performed investigating the behaviour of the addressed facility in case of a PLOFA scenario. The limits and capabilities of both the approaches were observed and are discussed in the present work trying to provide guidelines for a correct application of the adopted codes.
The obtained results provide interesting suggestions for the experimentalists and represent a valuable supportin better setting the experimental conditions and measurements tools layout. In the frame of future works, further analyses will be performed also trying to develop coupled STH/CFD application which, trying to overcome the intrinsic limits of both STH and CFD approaches, proved to be an interesting candidate for obtaining improved predictions of systems involving liquid metals.
The thermal striping phenomenon came to the attention of nuclear scientists during early 1990.
The coolant (sodium) attains different temperatures when it is made to pass through the fuel and blanket zones of the Liquid Metal cooled Fast Breeder Reactor (LMFBR) core. The temperature difference between the hot jet and cold jet can be as high as 150◦C. The turbulent oscillating jets interact with each other, giving rise to high magnitudes of temperature fluctuations. Highly conducting sodium makes it easy to transfer the temperature fluctuations to the adjacent solid structures without any loss due to boundary-layer attenuation. This results in the thermal fatigue of the solid structures and thereby their failure by generation of cracks in the structures. This phenomenon is referred to as thermal striping.
A three-dimensional numerical analysis has been carried out to study the phenomenon of thermal striping in fast breeder reactors using multi-jet water models that represents a row of the reactor core consisting of fuel and blanket zones. A commercial CFD code was employed for the analysis. As the phenomena is highly dependent on the time step and mesh size, detailed grid and time independent studies were carried out to arrive at suitable time step and mesh size. The simulations were carried out for different velocity ratios and temperature differences between the hot and cold jet coming out of the fuel and blanket zones respectively. The simulations were also performed using Reynolds stress model and Large Eddy Simulations (LES) in order to understand the capabilities of different turbulence models and their ability to capture turbulence characteristics of jet mixing phenomena. The main objective of the study is to identify the effect of velocity ratios and temperature differences on the thermal striping phenomena. The paper describes the locations prone to thermal striping phenomena. The study also identifies and thermal hydraulic conditions of velocity and temperatures that can lead to thermal striping damages to the structure.
The temperature distribution on the surface of components of the argon space in the main vessel of the demonstration fast reactor has a great influence on the safety of the reactor. In order to obtain the temperature distribution and flow field, the paper studies the argon space of the demonstration fast reactor under normal operation conditions by numerical simulation. Based on the existing research results, the low Reynolds number turbulence model and DO radiation model are used. The present work presents flow field of the argon space and analyzes the temperature field and the velocity distribution of the surface of the pump support, the intermediate heat exchanger support and dump heat exchanger support. The calculation results of this work will provide the basis for the design of argon space of the demonstration fast reactor and the reference for the corresponding experimental pre research.
Spiral spring is often used to seal the pins of fast reactor fuel assembly to control the core leakage flow and prevent excessive core leakage flow which resulting in insufficient core cooling. Although the spiral spring sealing effect is good, under normal circumstances, there will still be a amount of leakage flow in the pin, and whether the leakage flow can meet the design requirements needs to be considered. In order to provide theoretical calculation basis for determining the pin leakage flow rate of lead bismuth fast reactor assembly, this paper uses CFD method to conduct numerical simulation analysis on the assembly pin, and obtains the pin leakage flow characteristics of the assembly pin. By comparing the calculation results with the leakage flow experimental results, it is found that the leakage flow calculation results are relatively consistent with the experimental measurement results.
In fast reactor the high temperature jet flowing from the outlet of the throttle has a strong thermal shock to support structures, which may cause thermal fatigue and affect its strength and service life. How to quickly simulate related thermal-hydraulic phenomena and provide support data for evaluating strength of support structures is an important engineering problem. A special calculation and analysis code was developed in this paper which integrates modeling, meshing, thermal calculation, and post-processing functions to achieve rapid simulation and analysis for the problem. Based on a fast reactor design scheme, the code was used to simulate the thermal shock of high temperature jet on support structures, and the results were found to be reasonable and in line with expectations, which demonstrated the correctness of the code's physical model and the availability of various functions. The code can simplify the functions that required multiple steps and different professional codes to complete through one code. With this code we can conduct the calculation and analysis of the thermal shock problem of the support structure and conduct sensitivity analysis and fast optimization calculation for important parameters such as the size and layout of the throttle. In the field of sodium-cooled fast reactor engineering application this kind of thermal shock problems are of widespread existence, thus the code has a broad application prospect.
Horizontal-type lead-based fast reactor (LFR) is a novel conceptual design, which is smaller and more suitable for extreme environments. In this study, a sub-channel code for horizontal-type LFR was developed, and the influence of gravity on transverse flow was fully considered. The developed sub-channel code was validated by the CFD method, and the deviations between the predicted coolant temperature and simulated values are within ±10%. Based on the sub-channel code, the thermal-hydraulic characteristics of the horizontal-type assembly were conducted. The results show that along the gravity direction, the coolant flow of the assembly outlet increases, and the temperature first increases and then decreases. Compared with the vertical-type reactor core, the peak coolant temperature in the horizontal-type reactor is higher, which means the assessment of its safety performance is more important. The sensitivity analysis was carried out as well. The impact of the lateral resistance coefficient, conduction shape factor, turbulent mixing factor, heat transfer correlation, and friction resistance model on the coolant flow and temperature distribution was evaluated. This work could provide a reference for the subsequent design and development of the LFR.
Due to the unique design of wire wrapped spacers and inter-wrapper flow (IWF), the sodium-cooled fast reactor (SFR) has special thermal-hydraulic characteristics compared to thermal reactors. Therefore, obtaining accurate thermal-hydraulic performance is of great importance for the design and safety assessment of SFRs. A subchannel code SACOS-NA is developed specifically for the calculation of SFR in both normal and transient condition. The SIMPLE algorithm which is suitable for problems with low or reverse flows is adopted to solve the issues of backflow especially under natural circulation conditions. In this paper, the steady-state sodium experiments of 19-pin rod bundle performed by ORNL and 37-pin rod bundle performed by Toshiba, as well as the transient experiment of EBR-II SHRT-17 XX09 subassembly performed by ANL are select as benchmark for single-assembly verification. Code to code comparisons are also demonstrated by the results of MATRA-LMR, COBRA-IV-I and so on. In the case of multi-assemblies, the experiments of PLANDTL-DHX and CCTL-CFR conducted by JNC are used to verify the heat transfer capabilities between assemblies. The calculated results of the sodium temperature distribution are in good agreement with the experimental values. In order to study heat removal characteristics of SFR, SACOS-NA is coupled with the system analysis code THACS (developed by XJTU), and the natural circulation characteristics of CEFR in the event of a station blackout accident (SBO) has been calculated and analyzed.
Sodium-cooled fast reactors use secondary sodium systems to detach contact of the primary system with final heat sink to minimize the chances of sodium water/ sodium air chemical reactions. Sodium to sodium Heat Exchangers (HEs)are used for transferring the heat from primary system. These HEs are sized using the available correlations on the heat transfer and pressure drop. As the HEs are to be designed for the requirements of the reactor vis-à-vis heat removal during normal operation, decay heat removal, etc, details of heat transfer rate, temperature, and pressure drops are required to be estimated for all the anticipated events. A mathematical model would help in checking the acceptability of HEs for their intended service and estimation of process parameters in off-normal conditions. Computational Fluid Dynamics (CFD) is an effective technique for the above-mentioned purpose. However, the heat exchangers are generally large, and doing a detailed study with CFD necessitates higher computing resources.
Sodium HE from reported literature is taken to check the effectiveness of standard CFD methodology for simulating the behavior of HEs.The HE is a vertical counter-current unit installed in a test loop consisting of a heater vessel for heating sodium, sodium to air HE for heat rejection, and an electromagnetic pump for the circulation of sodium. The pump circulates Sodium on both the shell and the tube sides. At rated 3MW capacity, 14.31 kg/s sodium at 811 K flows into the shell side of the HE, gets cooled to 644 K by cold sodium flowing inside the tubes. Cold sodium (14.31 kg/s) at 616 K enters the tubes flows upwards getting heated to 783 K. The sodium flow rate is measured by an electromagnetic flow meter. Thermocouples fixed in thermo-wells at the inlets and outlets of the HE measure temperatures. Experiments were performed up to a flow rate of 10.86 kg/s due to insufficient heat rejection capacity of the sodium to air HE to maintain stable temperatures in the loop.
A 90° sector of the HE is used for CFD assessment using the symmetry boundary conditions. Both shell side and tube side fluids are considered for the study.3 – D conservation equations for mass, momentum, and energy equations are solved using finite volume and steady conjugate heat transfer analysis is carried using the model. Turbulence is modeled according to the k – ɛ Realizable model. The inlet temperatures for the shell side (811 K) & tube side (639 K) and discharge of 10.86 kg/s (both sides) are given as input parameters and the other process parameters are evaluated. There is a fair agreement between test and CFD results with CFD predicting a 3.8% more heat removal rate. The predicted outlet temperatures are within the 0.8 % error band and the convergence of the flow is ~10E-6. It is proposed that the standard CFD methodology can be used for simulating steady-state heat transfer of sodium to sodium HEs.
Coolant flow in Sodium Cooled Fast Reactor (SFR) is of significant importance to reactor safety because of SFR’s high power density. Loss of coolant flow sometimes appears in a single fuel bundle because of the hexagonal duct wall. Coolant flow reduction in a subassembly may cause coolant boiling and even dry out in worst cases. Therefore, prediction of the sodium mass flow and cladding temperature is essential in SFR safety analysis. In this work, the subchannel analysis method is used to simulate sodium boiling. Hexagonal subassembly cross section is divided into three kinds of subchannels with different area and perimeter. The effect of wire wrap is also considered to fit actual situation. Thermal-hydraulic models of Mikityuk heat transfer model and Kaiser friction model are used. Validation calculations are performed on ORNL 19-pin bundle experiment, which shows that predictions made with subchannel analysis method are reasonable. Subchannel analysis of sodium boiling in one bundle is then studied for the Sodium Cooled Fast Reactor.
The Demonstration Fast Reactor(CFR600) takes the design of putting decay heat exchanger into the lower plenum for the first time. Compared with the upper plenum DHX layout, this innovative design brings unique problems in the inlet and outlet runner design and the heat exchange of the lower and upper plenum. Therefore, the numerical study of the flow field and temperature field under its standby and operating conditions is of great significance. The fluent program is used to conduct a three-dimensional steady-state simulation of the DHX watershed of the lower plenum of the Demonstration Fast Reactor. The results show that there is still room for optimization of DHX cooling power and guide tube width under standby conditions. The calculated temperature data can be used as the external boundary conditions for the thermal calculation of the secondary side of the cold pool DHX. The internal flow field and temperature field distribution characteristics of the cold pool DHX can provide a reference for the demonstration fast reactor lower plenum DHX optimization design, and can provide reference for the DHX guide tube design.
Flow blockages in liquid metal rod bundles can have significant consequences that affect flow behavior and consequently heat transfer.
As part of this work we examine the flow in a 61 pin wire-wrapped rod bundle subjected a large blockage. The conditions mimic recent experiments conducted at Texas A&M with Particle Image Velocimetry.
Simulations are conducted with large eddy simulation in the open source spectral element code Nek5000/NekRS and compared against experimental results. the flow structures induced by the blockage are investigated in detail.
The numerical results serve both as a validation of Nek5000/NekRS and as an examination of the flow structures presented in the wake of blockages.
Activities have been currently under way across the world to build heavy liquid metal cooled reactors. Thus, for example, the ALFRED and BREST-OD-300 reactor designs use lead coolant, while the MYRRHA and SVBR-100 reactors use lead-bismuth coolant. High corrosive and erosive activity of the coolant requires observing the oxygen content (1÷4Е-6 Wt%) and coolant rate (0.5÷2.5 m/s) regulations. Due to the complex geometry of the reactor circulation circuit flow paths, numerical simulation methods are used extensively to justify scheduled modes of operation for liquid metal coolants. Correct justification of the complex oxygen transport processes in liquid metals requires the respective physicochemical computation model, which takes into account the main reactions of oxygen with the coolant and the structural materials.
This paper presents a physicochemical model, which includes the following processes: erosion, growth and dissolution of the two-layer oxide film, coagulation and dissolution of metal oxides in the circuit with subsequent deposition on filtering elements, and inflow from mass transfer apparatuses. STAR-CCM+, a commercial CFD code, was used as the tool in this study. The physicochemical model was implemented using models of passive impurities, which are used to simulate oxygen transport in the circulation circuit.
Capabilities of the presented model were demonstrated based on the results of investigating thermohydraulic and physicochemical processes obtained at the KALLA laboratory experimental facility. The duration of the simulation was 1000 hours.
The distributions of impurity concentrations, increased erosive activity areas, as well as the total amount of oxides deposited on filters and the amount of oxygen entering the circuit have been obtained as a result of the calculations. The surface distribution of the oxide film thickness on the test facility surfaces contacting the coolant has also been calculated.
Simulation of thermocouples, as well as taking into account the surface layer of the oxide film, have made it possible to improve the accuracy of calculating the flow’s thermohydraulic characteristics as compared to earlier studies. This allows one to hope that the approach used to simulate the experiments at the KALLA facility will make it possible to improve accuracy of the process simulations in HLMC reactor circuits as well.
Recently, the Light Water Reactor core simulator DYN3D was extended to perform steady-state and transient calculations of Sodium cooled Fast Reactors (SFR) on reactor core level. The essential supplementary methods included, among others, time-dependent axial and radial core expansions models.
Scaling up the simulation capabilities to system level requires the coupling of DYN3D with thermal-hydraulics (TH) system code capable of sodium flow modeling. In this study, we describe the adaptation of existing coupling of DYN3D with the TH system code ATHLET to transient analyses of SFRs. The approach to the modeling of out-of-core thermal expansions is presented. The extended DYN3D/ATHLET is validated with the help of selected tests performed during the start-up of the Superphenix reactor. The results of validation are presented.
In this work, the rotating cage setup, a well-known flow-accelerated corrosion testing system, was optimized for lead-cooled nuclear reactor applications. The rotating cage setup comprises a fixed cylindrical vessel filled with the corrosive liquid of interest and a rotating cage where testing samples, manufactured from the material of interest, are located. During operation, the relative motion between the samples and the liquid induces friction on the samples’ surfaces, thereby reproducing the shear-flow conditions found in actual applications such as pipes, flow channels and impellers. The samples normally used in the rotating cage setup have a blunt shape with rectangular cross section. Using Computational Fluid Dynamics (CFD) simulations, we show that the complex flow that develops around blunt samples causes a large form drag force on the samples and, correspondingly, a large power required to spin the cage. The power requirement becomes prohibitively large when lead is the working fluid because of its high density, particularly when high sample speeds are targeted. Furthermore, the unsteady massive flow separations from the surfaces of the samples make the interpretation of observed corrosion patterns particularly challenging. We show that these issues can be circumvented by using a new, streamlined sample design, conceptually similar to a classic airfoil but simpler and easier to manufacture. This new sample design reduces the flow resistance and associated power requirement to a manageable size. Indeed, for a sample Reynolds number of about 10^6, the torque contribution of the samples decreased from 75% to 17% of the total, and a power rating reduction of 57% was achieved. Despite the reduction in the total torque (which is due to both pressure and friction forces), the new design induces more shear on the samples, thereby enhancing the flow-accelerated corrosion phenomenon. The variation of the wall shear stress acting on the lateral surfaces of the streamlined testing samples is also less pronounced, which allows a clear association of the observed corrosion rates with the shear stresses.
The multi-year research program carried out by NRG and funded by the Dutch ministry of economic affairs is called ‘Program for Innovation and cOmpetence development for NuclEar infrastructurE and Research’ (PIONEER). The current PIONEER program runs from 2021 to 2024. The program comprises seven themes, i.e. long term operation, nuclear modelling and simulation, nuclear safety and compliance, fuels & materials, radioactive waste management, radiation protection, and innovative nuclear systems. One of the pillars in the theme of innovative nuclear systems is fast reactor research, particularly in the field of thermal hydraulics. This paper provides an overview of all fast reactor thermal hydraulics activities in the program, covering development and validation of system thermal hydraulics (STH) and 3D (engineering as well as high-fidelity) Computational Fluid Dynamics (CFD) codes and simulation approaches. Applications range from fundamental turbulent heat transport to core, pool and system thermal hydraulics. With the recent improvements in computational infrastructure and power, also further developments of multi-scale and multi-physics computational approaches are being integrated in the PIONEER program. A generic coupling tool ‘myMuscle’ is under development which is introduced in this paper. Recent results and current developments are presented together with an outlook for the results to be expected at the end of the current multi-year program and beyond.
Requirements for nuclear safety has been tightened after accidents to make up weak points of the safety. In case of Fukushima, passive safety was emphasized to secure safety of the reactor. Passive safety has advantages in terms of operator action and emergency power. Combination of the passive safety system could lead to fully passive reactor. Regard to long-term cooling, working fluid should be supplied passively. In case of water-cooling system, it should be designed as closed loop with heat exchanger. Otherwise, water should be supplied from the outside. Most of the safety systems in PWR employs water latent heat as their main heat removal method. In this kind of design, we should be continuously supplied, and it requires external aid for supplying new water except for marine application. It degrades level of passiveness of the safety system. Different to water, air always exists just beside the reactor. Just opening valve is enough to supply fresh air into the system. Therefore, cooling system with air does not have to designed as closed loop. In terms of passive safety, air has many advantages for working fluid of long-term cooling system.
However, air has inherently low heat transfer capability. Therefore, the air-cooling system could be applied only long-term cooling when decay heat is sufficiently decrease. It is better to be combined with short-term cooling system with high heat removal capacity but not passively sustainable. If sufficient heat transfer area was provided, and thermal inertia of the system is sufficiently large to endure energy storage during insufficient cooling, air cooling system has potential to be applied solely. Limitation of the air-cooling system could be relived large heat transfer area and large thermal inertia. Although these limitations were not solved, it is true that air-cooling system could be adopted as long-term cooling system for all type of the reactor.
Air-cooling system was already developed for SFR. Whose coolant sodium is very reactive with water. The most popular air-cooling system is RVACS. RVACS is abbreviation of the reactor vessel axillary cooling system. It cools reactor vessel using natural circulation of the water. It also employs natural circulation inside of the reactor pool to transport decay heat from the core to the reactor vessel. Therefore, it is fully passive safety system, and there are two main research topics for the RVACS; one is internal coolant natural circulation, and the other is external air natural circulation.
I have been researched natural circulation of the reactor pool in UNIST. It is hard to conduct experiment with liquid metal, similarity law was adopted. Natural circulation similarity for the temperature distribution between water and liquid metal was experimentally validated. Water could predict liquid metal temperature with an error of 27 % in maximum. Based on the similarity law, simulating experiments for 2-D and 3-D were conducted. Temperature distribution inside of the reactor pool was analyzed. Now we are designing integrated experiments with outside air natural circulation.
Experimental thermal hydraulic studies in sodium cooled fast reactor (SFR) systems include various testing and simulation in both sodium and water. Thermal response of the sodium systems is obtained by experiments in sodium and the hydraulic responses of the sodium can be precisely predicted by testing in water with suitable similarity criteria. Argon cover gas entrainment in sodium can cause reactivity fluctuations. The phenomenon was studied by hydraulic studies in water simulating Froud number and Weber number in a 5/8th scale model of SFR. Various devices to mitigate the gas entrainment in SFR systems by reducing the free surface velocity of sodium were evolved based on the testing. These devices were also effective in reducing the free level fluctuations of the sodium and hence the reduction in the high cycle fatigue of the SFR components such as inner vessel. Gas entrainment studies in the secondary sodium system components such as surge tank were also carried out in a 5/8 scale model. These studies resulted in improvements in the exiting design and significant reduction in the secondary sodium pump duty. The pressure transient arise during sodium water reaction were also studied experimentally using a dedicated scale down model of the secondary circuit. These experiments aimed at transmission and attenuation of the pressure pulse and helped in optimization of the surge tank design. While testing of SFR components such as control & safety rod drive mechanisms (CSRDM) in high temperature sodium, the sodium pool in the test setup is at 547°C and the bulk cover gas is at 300°C. This can result in cellular convection in the annular spaces of the test setup and its deflection. Experimental and 3-D CFD analysis to investigate the cellular convection in the annular spaces of high temperature sodium test vessels was carried out. There is a close agreement between the analytical and experimental results. This work gave good insight into the cellular convection in the annular paces of test vessels and methodology to limit the same.
The Karlsruhe liquid metal laboratory KALLA at the Karlsruhe Institute of Technology KIT offers a wide range of experimental facilities for thermal hydraulic experiments in support of safety studies for heavy liquid metal cooled fast reactor systems.
For the reliable operation of a fuel assembly in the reactor core, the knowledge of the heat transfer to the coolant is essential. Moreover, during the lifecycle of the assembly its geometry can be deformed by swelling, creeping and mechanical defects or blocked by debris. As a consequence, locally reduced cooling and hot spots are expected.
A large variety of experiments has been conducted in the last years on the heat transfer in rod bundles This includes experiments in lead bismuth eutectic LBE and water of blocked and unblocked rod bundles as well as the formation of blockages in a number of projects of the European Commission in the framework of Horizon 2020.
For the new project PATRICIA the effect of deformation in a wire spaced 7-pin fuel bundle mock up and the influence of a well-defined porous blockage in a wire spaced 19-pin rod bundle will be studied and is shown in this paper. Detailed instrumentation is needed in order to capture hot spots as well as recirculation patterns to have the potential to compare and validate the experimental results with numerical models.
The 61-Pin hexagonal test bundle in operation at Texas A&M University is a replica of a typical wire wrapped fuel assembly adopted in Liquid Metal Fast Reactors. The test facility has produced unique experimental datasets of pressure and flow fields to further understand the thermal-hydraulic behavior of these types of fuel assemblies, under operating conditions and hypothetical accident scenarios.
A detailed characterization of the axial and transverse pressure drops has been conducted within a wide range of Reynolds numbers spanning the laminar, transition, and turbulent regimes. Experimental subchannel and bundle averaged friction factors have been compared with available correlations.
High-resolution flow velocity fields have been obtained at different locations in the bundle using laser-based flow visualization and measurement techniques. Structure and characteristics of the flow within interior and exterior (near-wall) subchannels are described through the experiments conducted.
The effect of localized blockages of different sizes and configurations have been studied through a series of dedicated measurements to study the effects on pressure and flow.
The experimental data produced have supported the validation of advanced computer codes and the improvements of existing correlations.
The CEA is involved in the development of Sodium Fast Reactor since the 60s. In the purpose of the design of operating SFR reactors, which fulfill the 4th generation standard, the CEA is developing codes, which must be validated from experimental data. Since experiments with sodium are complicated, mainly because of its high reactivity with water and its opacity, part of the studies is performed on small scale mock-ups using water thanks to the dimensional analysis. In this purpose, the PLATEAU hydraulic loop has been designed and built to provide hydraulic conditions to those mock-ups. This facility has been operated during five years with numerous models characterizing different parts of the reactor and specific issues. The first mock-up, MICAS, was representative of the ASTRID upper plenum at a scale 1/6th.The experimental campaigns provided numerous results about the thermal hydraulics behavior in the vessel, which were compared to the numerical calculations. The velocities are in good agreement but regarding the gas entrainment study, the experimental and numerical results do not correlate. The second mock-up at scale 1, DANAH, aims at studying the flow in a sodium gas exchanger in the purpose of validating CFD calculation and optimizing the design of the sodium side. The velocity was measured by PIV and LDV with different geometries and compared to numerical results. The good agreement allowed further CFD studies for the optimization of the design. The third mock-up was about sodium fire in case of a pipe breach. The aim was to study the droplets induced by a jet for creating a model in a code. The droplet size were measured using the shadowscopy technic for different sizes and orientations of the nozzle. The last mock-up was dedicated to study the cavitation in the pump-diagrid pipes. Fast pressure sensor and accelerator gauges were installed on the pipe at different locations along the pipe. The measurements showed the occurrence of the cavitation from a threshold. Afterthought are in progress to transpose this result to the sodium case. Since, most of these results were obtained on reduced scale mock-up, investigations are in progress to assess their transposition to higher scales, especially the reactor one.
Studies of the liquid metal boiling show that compared with water boiling, the boiling process of liquid metals has essential features. There is only limited data on the sodium boiling in fuel assemblies (ULOF). A series of sodium-potassium alloy boiling experiments conducted at the IPPE with using the models of single fuel assembly and system of parallel fuel assembly with natural coolant circulation in order to study heat transfer and circulation stability are presented taking into account the various factors that influence on the boiling process.
The results of experiments show:
– the stable nucleate boiling in the fuel assembly model is observed only in a restricted region of heat fluxes, its transition to unstable pulsation slug boiling is determined by various factors;
– the boundaries of the transition from the nucleate regime to the slug, annular-dispersed and dispersed flow regimes of liquid metal two-phase flow in the fuel rod bundles are approximated by simple dependencies;
– the occurrence of an oscillatory process during coolant boiling in one of the parallel fuel assemblies leads to an antiphase oscillatory process in another fuel assembly;
– hydrodynamic interaction of the loops causes a significant increase in the amplitude of coolant flow rate oscillations ("resonance") and possible "choking" or inversion of the coolant flow rate in the loops, an increase in the temperature of the coolant and the fuel rod cladding; and to the heat transfer crisis;
– the liquid metal boiling heat transfer coefficients of fuel rod simulators in models in single loops and in the case of their parallel operation are in agreement, and are in the same range as the data on liquid metal boiling heat transfer in the tubes and pool boiling.
The effect of the influence of the surface roughness of fuel elements on heat transfer and flow regimes during boiling of liquid metal in bundles is demonstrated:
– in the assembly with low surface roughness of the fuel rod simulators, evolution of an unstable (slug) regime with sharp coolant flow rate oscillations and overheating of the simulator wall can result in a heat transfer crisis, in fact, there is no margin before the crisis;
– for the fuel rod simulators with industrially-manufactured surface roughness due to the appearance of a liquid film on the surface of the simulators, a transition from an unstable slug regime to the stable annular-dispersed one has been observed.
The experimental study results of sodium boiling heat transfer under natural and forced convection in a fuel assembly model with a "sodium cavity" located above the reactor core which is designed for compensation of the positive sodium void reactivity effect in accident situations with sodium boiling are also presented. It is shown, to provide continuous sodium cooling of fuel rod simulators in fuel assemblies under these conditions is possible. The data on of liquid metal boiling heat transfer in bundles were generalized and a cartogram of the flow regimes for a liquid metals two-phase flow in bundles are presented.
The Westinghouse Lead Fast Reactor (LFR) is a 950 MWt (~460 MWe) lead-cooled, fast neutron spectrum, pool-type reactor being developed by Westinghouse in collaboration with domestic and international organizations. The reactor features a pool-type configuration with hybrid microchannel-type primary heat exchangers (PHEs) directly immersed in the primary coolant. The reactor vessel (RV) is surrounded by a guard vessel (GV) in order to contain the lead coolant in the unlikely event of reactor vessel failure. The emergency decay heat removal is performed by the passive heat removal system (PHRS), which consists of a water pool system surrounding the GV filled with enough water to remove decay heat for the first seven days, and a number of stacks to circulate air and remove decay heat indefinitely after the water has boiled off.
The SAS4A/SASSYS-1 system code was coupled with the GOTHIC code to perform safety analysis of the Westinghouse LFR. In addition to in-vessel thermal-hydraulics, SAS4A/SASSYS-1 simulates neutronics with reactivity feedback, thermal and mechanical responses of the fuel and core, fuel pin failure, and the potential for fuel pin failure propagation. It also has primary heat exchanger and reactor coolant pump models. GOTHIC tracks fluid flow and heat transfer outside of the RV, i.e. in the PHRS, including the heat transfer between the RV and GV wall.
In this paper several accident scenarios are selected and the response of the reactor system to each scenario is described as examples of safety analysis of the Westinghouse LFR. Station Blackout is initiated by a loss of off-site power. Consequently, all active systems including the reactor coolant pumps, primary heat exchangers, and normal decay heat removal system become unavailable. The normal reactor shutdown system is assumed to fail. Hence, the fission power is dumped to the coolant briefly until a passive shutdown system is actuated by high coolant temperature. Subsequent heat up and gradual cooling of the fuel, fuel cladding, coolant, and RV wall; decay heat removal by PHRS; and successful water to air cooling transition in PHRS are demonstrated. Other scenarios analyzed include Loss of Flow (LOF), Loss of Heat Sink (LOHS), and Transient Overpower (TOP). The safety performance of the Westinghouse LFR is evaluated by examining its response to each scenario.
One of the key requirements to innovative liquid metal cooled fast reactors is the necessity of ensuring, in the event of an accident, the removal of heat from the reactor core using passive heat removal systems. This makes it possible to increase greatly the reactor safety and to reduce the risk of the accident reaching a more severe phase due to failure of active safety systems.
The process of the reactor facility changeover from normal operation to natural circulation can be accompanied by a number of negative aspects, which can be identified at the reactor design stage only using computational fluid dynamics tools.
This paper presents the results of a CFD simulation for the process of the BREST-OD-300 reactor changeover from the NO mode to the NC mode during the RCPS emergency trip with the simultaneous failure of two ECCS loops. To this end, a 3D CFD model of the reactor circuit flow path was developed, including all of the reactor’s key circuit components (steam generators, RCPS, ECCS cooldown heat exchangers, filters, mass transfer apparatuses). Cover gas was taken into account in the CFD model for the correct description of the changeover process. The duration of the changeover process simulation was ≈8500 sec, which corresponds to the core thermal power change from 100% (rated value) to 1% (initial stage of the NC development).
The transient process characteristics of the NC development in the BREST-OD-300 reactor circuit have been obtained as a result of the study. The CFD simulation results have shown that for the first two hours of the accident the heat is securely removed from the reactor core without the ECCS involvement, while the maximum fuel cladding temperature did not exceed 600 °C in the considered time interval.
This work reports the experimental and numerical activities performed to investigate the heat transfer in the lead-bismuth (LBE) cooled fuel pin bundle simulator (FPS) of the CIRCE facility. The FPS is a wrapped, grid spaced pin bundle composed of 37 electrical pins placed on a hexagonal lattice with a pitch to diameter ratio of 1.8. Each pin has an outer diameter of 8.2 mm an active length of 1000 mm. The linear power is 25 kW/m with a maximum wall heat flux of 1 MW/m2 and a total electrical power of 925 kW. The bundle is hosted in the CIRCE large pool facility and represent the hot source of the test section.
Several thermocouples (0.5 mm O.D.) are installed in the test section to measure both clad and subchannel LBE temperature at different ranks and at two different sections. Experimental data are used to evaluate the Nu number in the Pe range 500-3000 and obtained data are compared with Ushakov and Mikityuk correlations having a validity range containing the experimental p/d ratio and for the selected Pe range.
In parallel, Protected Loss of Heat Sink (PHLOS) and Protected Loss of Flow (PLOFA) experiments were run in CIRCE facility. The PLOHS-PLOFA experiments were aimed at investigating the thermal-hydraulics phenomena related to safety issues occurring in a Heavy Liquid Metal-cooled fast reactor in response to hypothetical accidental scenarios. A series of four experimental tests were carried out simulating the total loss of the secondary circuit and the coolant pump trip with the subsequent simulation of the reactor scram (reduction of the electric power supplied to the fuel pin simulator) and activation of DHR system to remove the decay heat power (~5% of the nominal value).
Tests differ from each other by the applied boundary conditions such as the electrical power supplied to the Fuel Pin Simulator, the duration of the test, the power removed by the HX etc., while test #4 also differs for the forced circulation maintained after the simulation of the accidental transient.
This document describes the achieved experimental results for the heat transfer in fuel bundle geometry, for the full power steady state condition (normal operation conditions) and for the transient phase (transition to decay heat removal conditions). Experimental data relating to the obtained Nu number and the thermal hydraulic phenomena such as the modification of the thermal stratification in “pool type” configuration, coolant mass flow rate modification in the test section occurring during the simulation of the designed accidental conditions and the capability to cool the FPS under natural circulation conditions are here reported and described. Finally, the experimental database was also used for the validation of the geometrical and numerical model of the FPS adopted for 3D CFD calculations. In the performed simulations a sensitivity analysis was carried out for the turbulent Prandtl number in the range between 1-3 and a comparison between the first order turbulence model k-omega SST and the higher order Reynolds Stress Model was accomplished.
Among the goals of the core thermal-hydraulic design of Lead-cooled Fast Reactors (LFRs) exploiting the closed sub-assembly (SA) option, cold by-passes must be avoided and excessive thermal gradients among opposite faces of the assembly ducts prevented. To achieve these goals, a suitable coolant flow outside the assemblies themselves must be guaranteed, compatibly with the inter-wrapper gap, which is established by the core thermo-mechanical design. Moreover, for wrapped assemblies, the possibility of gagging arises, giving an extra degree of freedom to the designer for leveling thermal gradients at the assemblies’ outlet. Therefore, the design process requires knowledge of the axial and radial coolant temperature profiles in the inter-wrapper gaps throughout the whole core (i.e., including all core SAs), as well as the axial and perimetrical wrapper temperature profiles, and notably the (possibly) different values of each side of the wrapper itself which could induce SA bowing.
In view of the above-mentioned requirements, a Design-Oriented Code (DOC), TIFONE, was developed and verified in compliance with the ENEA software quality assurance requirements. The sub-channel approach was chosen, since it allows to achieve a sufficient level of spatial resolution while retaining the key features of a DOC, namely equilibrium, a low computational time and a clear application domain. The current version of TIFONE solves, for an LFR exploiting the closed SA option in hexagonal geometry, the inter-assembly coolant mass, energy and momentum equations, as well as the convection equations between the coolant and the wrapper. The calculation domain extends radially over the inter-wrapper region of the entire core, and axially between the dividing and the merging points of the inter- and intra-SA coolant flows. Among the perspective applications of TIFONE is the coupling with codes for the thermal-hydraulic analysis of the single SA, so to allow for a full-core simulation.
The code has been preliminarily validated against experimental data from the KALLA inter-wrapper flow and heat transfer experiment, as well as against data from the PLANDTL facility, showing satisfactory agreement. These first applications of TIFONE confirmed its ability in reproducing the measured data in its anticipated validity domain.
A computational campaign has been carried out at the Department of Astronautical, Electrical and Energy Engineering (DIAEE) of “Sapienza” University of Rome, aiming at the assessment of RELAP5-3D© capabilities for subchannel analysis. More specifically, the investigation has involved LBE-cooled wire-spaced fuel pin bundle, by comparing simulation outcomes with experimental data from the NACIE-UP facility, hosted at ENEA Brasimone Research Center. The thermal-hydraulic nodalization of the facility has been developed with a detailed subchannel modelling of the Fuel Pin Simulator (FPS). Four different methodologies for the subchannel simulation have been investigated, increasing step by step the complexity of the thermal-hydraulic model. In the simplest approach, the subchannels have been modelled one by one, neglecting heat and mass transfer between them. As expected, this model has shown noticeable discrepancies with the experiment and thus it has been improved with multiple cross junctions, realizing the hydraulic connection between adjacent subchannels. In this case, mass transfer depends on pressure gradient and hydraulic resistance only, ignoring the turbulent mixing promoted by the wire wrapped. Simulation results have been not satisfactory, and a further improvement has been introduced in the third approach. In this case, several control variables calculate at each time step the energy transfer between adjacent control volumes associated to the turbulent mixing, induced by the wires. This energy is transferred using “ad hoc” heat structures, where the boundary conditions are calculated by the control variables. The present model has highlighted good capabilities in the prediction of the radial temperature distribution within the FPS, considerably reducing disagreement with experimental data. Moreover, the influence of the radial conduction within the fluid domain has been assessed, introducing further heat structures. Although this most complex model provides the best estimation of the experimental acquisition, the improvements given by the radial conduction has not been so relevant to justify an increase of computational costs.
Subchannel codes remain a critical tool for fast reactor design because they can generate intermediate-fidelity thermal-hydraulics (TH) results with relatively inexpensive calculations. The conceptual design of the Versatile Test Reactor (VTR) relied on subchannel TH analyses for calculation of the peak coolant, clad, and fuel temperatures, as well as for optimization of the coolant orificing strategy. Follow-up efforts aimed at understanding the impact of experiments on core thermal hydraulics will additional flexibility in subchannel analysis capabilities, and later stages of VTR analysis will require a subchannel code that is verified and validated.
To that end, the Ducted Assembly Steady-State Heat transfer software (DASSH) is being developed at Argonne National Laboratory (ANL) to supersede the legacy subchannel TH software SE2-ANL. SE2-ANL is a modified version of SUPERENERGY-2 that obtains power distributions from the Argonne Reactor Computational suite and is routinely used as part of the advanced reactor design workflow. Like SE2-ANL, DASSH performs steady-state TH calculations to determine coolant flow and temperature distributions for hexagonal reactor cores with ducted assemblies. In DASSH, each assembly contains a hexagonal bundle of wire-wrapped pins and is divided into subchannels. The energy balances are applied to each subchannel to determine the coolant and duct temperatures. Assemblies are coupled via radial heat transfer through the duct wall and inter-assembly gap. DASSH bypasses solving coupled energy and momentum equations through the use of correlations.
This paper introduces the methodology used in DASSH as well as the code capabilities. Although DASSH is built upon on the methodology used in SE2-ANL, it features several key added capabilities and general improvements. DASSH is coupled to the current neutronics software DIF3D-VARIANT, whereas SE2-ANL is coupled to the legacy software DIF3D-FD. Additionally, DASSH offers simplified models for non-rod bundle regions, temperature-dependent material properties, an overhauled keyword-value input structure, and built-in visualization capabilities. DASSH is written in Python and is being released through an open-source software license on GitHub. A full-core case study on VTR will be used to highlight some of the DASSH features and make comparisons with SE2-ANL.
DYNASTY (DYnamics of NAtural circulation for molten SalT internallY heated) is an experimental facility designed and built to investigate natural convection under distributed heat generation. Molten salt reactors are characterised by liquid fuel, a unique condition amongst all nuclear reactor concepts, and therefore the decay heat source is no longer localised within the core, but rather is distributed along the entire primary circuit. As such, during cooling instability conditions may arise, with unwanted oscillations in the flow regime and even inversion because of the interactions of several forces, such as buoyancy ones and pressure losses. The DYNASTY facility aims at studying these phenomena by simulating a distributed heat source, to better understand the various phenomena that may occur within the molten salt reactor during cooling. The modelling of the experimental facility has been carried out using the object-oriented programming language MODELICA, focusing both on a sensitivity analysis of the available numerical algorithms and facility parameters and on the validation of the model. To this end, experiments have been done using the DYNASTY facility to identify the peculiarities of natural circulation heating and cooling compared to the forced one and to validate the model. Results show a very good agreement between model and experiments, whilst identifying some interesting phenomena such as flow inversion during cooling and fluid oscillations due to Welander wave packets.
As a part of the India’s three stage nuclear power program, Fast Reactors play a significant role towards maximizing utilization of limited resources of uranium available in the country for the long term and sustainable energy demands. The operation of Fast Breeder Test Reactor (FBTR) in Kalpakkam has successfully demonstrated the fast breeder technology. With the experience gained from the operation of FBTR, the design for a 500MWe pool type prototype fast breeder reactor (PFBR) was taken up. The PFBR is currently in the initial stages of commissioning. Sodium cooled reactors with pool type design have a large volume of primary sodium and its thermal capacity and high conductivity favourably attenuate thermal transients in the primary circuit. The prediction of temperature evolution in the core during steady-state, transient and shutdown heat removal regimes play an important role in establishing design limits for core, coolant and structures. Several unique thermal hydraulics phenomena that are encountered during design are thermal stripping, cover gas entrainment, free level fluctuations, buoyancy effects within the hot and cold pools, flow distribution in the subassemblies, flow recirculation etc which needs detailed understanding and improved codes for accurate prediction.
The reactor design of PFBR has undergone extensive safety review by the regulatory body based on the Safety criteria for Sodium cooled fast reactors. Further the improvements in regulation have been brought about by considering operating experience feedback, assessment of changes in national and international regulation etc. A Safety Code on design of sodium cooled fast reactor based nuclear power plants has been developed by the Atomic Energy Regulatory Body (AERB).
This paper address the evolution of safety requirements of Fast Breeder reactors with specific emphasis on core thermal hydraulics, challenges in the review and acceptance of unique thermal hydraulic characteristics of the sodium cooled core such as thermal stratification, cellular convection, thermal stripping, gas entrainment, natural circulation under decay heat removal etc.
Keywords: Thermal stripping, gas entrainment, natural circulation, stratification
The SIMMER code (SIMMER-III and SIMMER IV) includes advanced fluid-dynamics/multiphase-flow and neutronics models. We apply the code for simulation of severe accidents in sodium fast reactors (SFRs) and other systems. An accident initiation phase (IP) of a severe accident in SFR, before can-wall melting onset, can usually be simulated with a different code; this may facilitate IP analyses but may introduce uncertainties related to coupling of SIMMER with this different code at the end of IP. We have developed several new models for SIMMER recently in order to facilitate its application to IP. The following thermal hydraulic models have been established and tested in KIT:
• Inter-wrapper gap model
• Subchannel-scale mesh model
• Heat exchange model
• Gas-Expansion Module (GEM)
Moreover, the following neutronic models have been developed and tested in KIT:
• Thermal expansion model in both axial and radial direction with serval options
• Control rod driveline (CRDL) feedback model
As an example of application of some of abovementioned models, we show our recent simulation results of loss of flow without scram (LOFWOS) tests in the Fast Flux Test Facility (FFTF) with emphasis of Gas Expansion Module (GEM) direct simulation. The benchmark was organized as an IAEA collaborative research project (CRP), including a blind phase and a second phase, during which the models can be improved using experimental results. The GEM and Doppler feedback effects are two dominant ones, which are negative and positive during the transient, respectively. The flow rate, the net reactivity and the power are simulated quite accurately with the improved GEM model. We present calculations of the sodium level in the GEM and related reactivity feedbacks. One may conclude that the SIMMER code has a large potential to application to initiation phase analyses.
In the context of GEN-IV heavy liquid metal-cooled reactors safety studies, the flow blockage in a fuel assembly (FA) is considered one of the main issues to be addressed and the most important accident for LFR fuel assembly. The flow blockage in a fuel assembly (FA) consists of a partial or total occlusion of the flow passage area.
The present paper is intended to provide a analysis of such phenomena and a modelling of the unprotected blockage of the hottest fuel assembly for ALFRED LFR DEMO was being performed with CATHARE (Code for Analysis of Thermalhydraulics during an Accident of Reactor and Safety Evaluation). At this stage, a conservative analysis has been carried out based on the current main geometrical and physical features. Reactivity feedback, as well as axial power profile, were not included in this analysis. The blockage was placed at the foot of the fuel assembly and was appearing instantly at 10 sec transient time. The flow area was progressive decrease in the hottest channel using a control valve.
Results indicate that the fuel melting is not expected to be a safety issue for the ALFRED as fuel melting temperatures was not reached (2700 °C), but clad failures (creep rupture) should be expected, critical conditions with clad temperatures around ∼1000 °C, was reached.
Kew words: ALFRED, THERMAL-HYDRAULIC, TRANSIENT, CATHARE, FUEL ASSEMBLY
In order to enhance the safety of sodium-cooled fast reactor (SFR), installation of diverse decay heat removal systems is necessarily demanded to improve the reliability of SFRs by ensuring the decay heat removal capability. The importance of the decay heat removal during a long-term station blackout (SBO) was reconfirmed by the accident of Fukushima Daiichi nuclear power plant. When SBO occurs in SFR, the natural circulation which do not need power supply can be expected in the heat transport system, and the decay heat in the core can be removed by auxiliary cooling system (ACS) utilizing the natural circulation. In the design of advanced SFR (A-SFR) as a GEN-IV power plant, the plant dynamics analyses must be conducted to confirm the core coolability by ACS during plant transient behavior not only in the force circulation condition but also in the natural circulation condition. A plant dynamics analysis code named Super-COPD has been developed in JAEA to implement evaluation of design margin and safety evaluation for the SFRs in Japan. Up to this day, Super-COPD has been used for the plant dynamics analysis of the experimental SFR, JOYO, the prototype SFR, MONJU, and a conceptual A-SFR, JSFR (Japan Sodium-cooled Fast Reactor). Through the studies in the evaluation of SFRs, knowledge regarding the decay heat removal by natural circulation has been accumulated in JAEA.
In this paper, the outline of Super-COPD and the recent status of development of numerical models to analyze the important phenomena under the natural circulation condition for the plant design and the safety evaluation of the A-SFR which is under conceptual investigation in Japan are described with summarization of the experiences of validation studies. In view point of the necessary models for the plant dynamics analysis during the decay heat removal by natural circulation, important physical phenomena are the radial heat transfer among subassemblies, the flow re-distribution in the core, the pressure loss and natural convection head in the heat transport system, and the thermal stratification in the upper plenum of the reactor. The radial heat transfer among subassemblies and the flow re-distribution in the core affect the temperature distribution in the core. Therefore, all subassemblies and radial heat transfer through the gap were modeled to evaluate the natural convection head and the pressure loss of subassemblies. The thermal stratification in the upper plenum of the reactor vessel appearing in the beginning of the plant transient can significantly affect the natural convection head in the heat transport system. The multi-volume mixing model in which the plenum is divided into several regions based on the results of computational fluid dynamics analysis has been developed to predict these transient phenomena in the upper plenum. The validation studies of Super-COPD have been performed through the benchmark analyses of JOYO, MONJU, EBR-II, FFTF, and several experiment facilities. Based on these validations, applicability of Super-COPD with the models for decay heat removal under natural circulation condition is confirmed and the requirements to be modified for the evaluation of A-SFR are extracted.
The 42-channel model representing 1/6th of the European Sodium Fast Reactor core was developed in frame of the Euratom ESFR-SMART project using the TRACE system code featuring a new sodium boiling modeling functionality developed at PSI. The model of a channel includes a 1D pipe component coupled to 1D heat structure components representing fuel rods, hexcan wall and upper reflector/shielding structures. The coast down curve for flowrate and constant pressure were specified as boundary conditions at the core inlet and outlet, respectively. Point-reactor kinetics with pre-computed kinetic parameters, reactivity coefficient and power distribution was used. First, the sodium boiling evolution in ULOF was analysed for the reference design of the assembly. Then, the design modification aiming at the sodium boiling regime stabilization was proposed and supported by the calculations. Finally, a number of sensitivity studies were conducted to demonstrate the robustness of the proposed design modification.
A multi-scale platform integrating system analysis module THACS, sub-channel module SACOS-Na and 3-D computational fluid dynamics module was developed for conducting thermal-hydraulic analysis to sodium-cooled fast reactor (SFR). Flexible coupling strategy holds out possibility of versatile combinations for system components that each of them can be modeled by 1-D methodology or high-fidelity choice. Each module has been validated independently and comparisons between stand-alone and coupled simulation have been performed to entail limitations of independent application. Three cases of coupled simulation for CEFR (China Experiment Fast Reactor) were reported in this paper. In first case, the reactor core was built with 3-D model and other parts in system and were simplified with system module. Secondly, the reactor core was replaced with sub-channel module. Finally, the whole primary loop was constructed in detail and the DARCS system out of the reactor vessel was simulated with 1-D model. Analysis of these results helped identify critical safety-related phenomena that can’t be resolved by existing tools.
One major safety feature in China Fast Reactor (CFR600) is the adoption of passive residual heat removal system. The natural circulation capability during accident is one of the main tasks in reactor design and the effectiveness and reliability should be validated. The RELAP5 code is used for the analysis of CFR600 primary natural circulation capability validation experiment. The residual heat removal transient process after station blackout accident is simulated and the flow and temperature distribution in the core and pools are obtained. The results show that the code simulates the natural circulation flow paths properly; and the main thermal-hydraulic phenomenon such as core flow re-distribution, main vessel cooling system flow inversion, and hot pool temperature stratification are reproduced; the key parameters such as R-type assembly outlet temperature, the primary flow rate and inlet/outlet temperature of different heat exchangers agree well with the test results. The RELAP5 code could predict the natural circulation transient phenomenon during SBO accident.
One of the main causes of failure of steam generators (SG), as well as various heat exchangers of nuclear power plants, is the flow induced vibrations (FIV) of tube bundles, and potentially may lead to structural damage or component lifetime shortening in the contact “tube – spacer grid”. According to statistics, due to increased vibration of the tube bundles occurs about 30% of the shutdowns of various purposes power equipment in the world.
For design of new installations, it is necessary to consider different design variants. This requires the development of a new approach for calculating flow induced vibration of SG tubes based on a combination of engineering methods and the results of CFD modeling, which allows quick analysis and take into account the peculiarities of the flow in the current design.
In the past, the occurrence and the stability of FIV were investigated with the help of empiric correlations. Today, to allow an accurate fluid structure interaction (FSI) analysis of FIV uses 3D CFD programs are coupled with CSM tools, but this method is very expensive concerning CPU time. To evaluate the intensity flow induced vibrations of the tube bundle of direct-flow SG in transverse water flow, a new approach was suggested and validated against experimental data. This approach combines a CFD model of the spatial fluid flow in tube bundle domain and analytical model of flow induced vibrations based on semi-empirical correlations. This calculation method allows us to obtain a conservative estimate of the vibrations intensity in the transverse coolant flow. An algorithm based on the developed approach was developed for calculating the parameters of tube bundles oscillations in direct-flow SG.
The approach was applied to calculate the intensity of FIV of the tube simulators of the experimental 61-pipe model of direct-flow SG. The vibration characteristics of the tube simulators of the 61-pipe direct-flow steam generator test facility were determined. The average relative error of the RMS acceleration values didn’t exceed 15%.
The main advantage of this approach is the ability to optimize the geometry of the tube bundle at the inlet part of the SG for lot of design variants without carrying out large amount of experimental research.
For a reliable assessment of the consequences of an extremely unlikely reactor accident resulting in core meltdown key questions arise: how to remove the decay heat from the reactor system and how to retain the radioactive core debris within the containment.
This study aims to analyze the debris bed coolability of an innovative 4th generation sodium cooled fast reactor during the severe accident that follows an unprotected loss of flow. Due to the topological characteristics of the domain (pool-type reactor) and the complexity of the flow driven by natural convection, a CFD tool is selected to perform this work. The findings of previous studies on the post-accident material relocation phase are used as initial conditions for the thermal-hydraulics coupled codes Saturne and Syrthes to assess the post-accident heat removal phase.
The capacity of two decay heat removal systems (in- and ex-vessel) available for this mitigation scenario are evaluated, paying special attention to the reactor vessel cooling system. Temperatures and heat fluxes in several locations of the collectors and core catcher are calculated to verify the safety criteria and assess the risk of a scenario disruption to a non-desired highly energetic event.
In this paper, a thermal-hydraulic transient analysis code for lead-cooled fast reactor, LETHAC is developed. The mathematical models and modules are presented in detailed. Two experimental facility, NACIE-UP and CIRCS were modeled in the code, and calculated three experimental cases, GFT and PLOFA tests for NACIE-UP and Test-1 for CIRCS using the present code. The calculated data has been compared with the experimental data, the comparing results shows that the calculated results using the present code are in good agreement with the experimental result, and the maximum relative error is 2% for GFT test,10% for PLOFA test, and 7% for TEST-1 of CIRCS. This indicated that the present code is suitable for predicting the transient behavior of lead cooled system. Possible reasons for the error were analyzed in the paper.
For the purpose of designing advanced nuclear reactors as a safe, economic, and sustainable carbon-free energy source, Japan Atomic Energy Agency has begun development of "Advanced Reactor Knowledge- and AI-aided Design Integration Approach through the whole plant lifecycle (ARKADIA)". The development will be done in two phases: ARKADIA-Design and ARKADIA-Safety which respectively deal with design and safety will be constructed independently in the first phase till the end of FY2023 and they will be merged into one ARKADIA in the second phase till the end of FY2028. In this paper, outline and recent achievements in ARKADIA-Design are described. ARKADIA-Design offers functions to support design optimization both in normal operating conditions and design basis events mainly during the conceptual design stage in the fields of core design, plant structure design including thermal-hydraulics analysis, and maintenance plan design.
It is necessary for optimization of the design to conduct various numerical analyses using one-dimensional plant dynamics analysis (1D) code which performs the efficient evaluation of various design options and multi-dimensional analysis code which evaluates local phenomena in detail including multi-physics. In the conventional design procedure, for instance, in order to find the core specifications that satisfy the feasible conditions and achieve the maximum core performance, physical phenomena are analyzed individually by different scales and fields where the boundary conditions of each analysis are determined on safe side. Hence, the core specifications as the results tend to be conservative. ARKADIA-Design, therefore, performs a whole plant analysis based on the multi-level simulation (MLS) technique in which the analysis codes are coupled to simulate the phenomena in intended degree of resolution. For MLS technique, at present, three coupling analysis methods with 1D code as the base module focusing on the physical phenomena related to the core performance have been developed: (1) the coupling method with 1D and computational fluid dynamics analysis to predict the effect of multi-dimensional thermal-hydraulics phenomena in a core upper plenum on the whole plant dynamics, (2) the coupling method with the core thermal-hydraulics analysis in 1D code, neutronics calculation, and core structural mechanics analysis to evaluate core deformation reactivity feedback, and (3) the coupling method with 1D and subchannel thermal-hydraulics analysis to evaluate detailed temperature distribution in a subassembly with thermal interaction between adjacent subassemblies and to offer the detail wrapper tube temperature distribution for accurate prediction of core deformation.
These three coupling methods were applied to the numerical simulation of the experimental fast breeder reactor EBR-II tests. Through the numerical analysis of EBR-II tests, applicability of the coupling methods was confirmed, which suggests ARKADIA-Design will allow for the optimal performance core design with reasonable conservativeness.
Development of Advanced Reactor Knowledge- and Artificial Intelligence (AI)-aided Design Integration Approach through the whole plant lifecycle, ARKADIA has been started in Japan Atomic Energy Agency. ARKADIA can automatically provide possible solutions of design, safety measures, and a maintenance program to optimize the lifecycle performance of advanced reactors by using the state-of-the-art numerical simulation technologies. In the first phase of this project, ARKADIA-Safety is developed for the purpose of automatic optimization of the severe accident (SA) management and its feedback to the plant design.
This paper describes the outline, development items, and example problem of ARKADIA-Safety. ARKADIA-Safety performs the in- and ex-vessel integrated numerical analysis during the SA, the statistical safety evaluation, and the dynamic probabilistic risk assessment (PRA). This evaluation is achieved by the AI technology and the knowledge-base constructed from the data obtained through the previous fast reactor development programs such as Monju or the future research and development. The prime target of ARKADIA-Safety at this point is sodium-cooled fast reactors (SFR).
The principal simulation system in ARKADIA-Safety is the SPECTRA (Severe-accident PhEnomenological computational Code for TRansient Assessment) code. This code has been developed for integrated analysis of the in- and ex-vessel phenomena during the SA of SFRs. The in- and ex-vessel analyses exchanges their boundary parameters at every time step. As one example, the amount of the sodium which leaks from the failed primary pipe is computed from the pressure difference between the inside and outside of the pipe. Some models for elemental physical phenomena are integrated into the thermal hydraulics models. The in-vessel basic module employs a multi-dimensional compressible two-phase flow model to simulate thermal-hydraulics of the sodium coolant coupling with the molten fuel behavior. The in-vessel fuel behavior is simulated by the individual modules for the core disruption and the fuel relocation. The ex-vessel thermal-hydraulic basic module employs a lumped mass model to simulate inter-room heat and mass transfer. Their basic equations include the source terms due to ex-vessel phenomena, such as sodium fire and sodium-debris-concrete interaction. These are modeled in the individual physical modules.
Fundamental capability of SPECTRA has already been demonstrated through the application to the loss of reactor level event. This analysis simulates lowering of the coolant level in the reactor vessel due to sodium leakage and increase in temperature and pressure due to sodium-debris-concrete interaction and sodium fire in the ex-vessel compartment. Each module in SPECTRA will be advanced in the future work. Also, the statistical safety evaluation and the dynamic PRA methods will be developed and incorporated into SPECTRA. In addition, in order to confirm the capability of ARKADIA-Safety for optimization problems, we plan an application to an example optimization of the containment vessel (CV) design considering the SA phenomena, such as sodium fire and leakage of fission products. The size of the CV and mitigation measures for the SA is optimized by ARKADIA-Safety employing the SA evaluation of SPECTRA.
The Advanced Lead-cooled Fast Reactor European Demonstrator (ALFRED) is a 300 MWth pool-type reactor aimed at demonstrating the safe and economic competitiveness of the Generation IV LFR technology. The ALFRED design, currently being developed by ANSALDO NUCLEARE and ENEA in the frame of the FALCON Consortium, is based on prototypical solutions intended to be used to boost the DEMO-LFR development.
In the frame of the research activities devoted to ALFRED development, the flow blockage in a fuel sub-assembly is considered one of the main issues to be addressed.
This work reports the experimental results and post-test analysis carried out in the prototypical test section of the lead-bismuth eutectic (LBE) -operated NACIE-UP facility. NACIE-UP is a rectangular loop, where the two vertical pipes, which work as riser and downcomer, are 8 m long and the two horizontal pipes are 2.4 m long. A prototypical 19-pins Blockage Fuel Pin Simulator test section is installed in the bottom part of the riser, whereas a shell and tubes heat exchanger is placed in the upper part of the downcomer. Several degrees of internal blockage were tested in the facility from 10% to 33 % flow area blockage.
The peak temperature value in the most severe case is around 45 °C in the experimental conditions. Lots of data were produced in different configurations by varying blockage degree and mass flow rate.
A CFD numerical post-test validation activity is carried out on a limited number of cases. The CFD numerical model reproduces the geometry of the test section in a detailed way. The model is described accurately both in terms of geometry and of meshing technique. For the single sector blockage numerical and experimental results are compared in detail. Results show a maximum in the temperature field just behind the blockage and this feature is also evidenced by experimental tests.
After 2011 Fukushima accident more and more regulatory body from around the world had increased the safety requirements for new built plant but also for existing fleet. In Romanian case, CNCAN required a stress test for Cernavoda NPP Unit 1 and 2. In this test a new design based accident was consider namely Station Black Out. Cernavoda owner improved electrical station and electric back-up diesel generators. On the other hand, this paper we will present efforts made in RATEN-CITON in order to implement a new passive safety system capable to successful cooldown the reactor core and transport of all residual heat generation for at least three days. The solution adopted it was similar with passive safety system designed for ALFRED LFR demonstrator reactor, a Passive Isolation Condenser with or without non-condensable gases connection. In order to design this system we calculated total energy from decay heat released in reactor core, calculate the water volume required to evacuate the decay heat, looking for a solution to layout it on the current situation of Unit 2 Cernavoda NPP. Due to a high energy to be evacuated we considered a water tower located outside of containment just in its proximity. In order to operate the new system will have to be operated only by passive condition, only one operation of isolating valves would be credited. After system is put in function the operator will have no action required for 72 hours, time sufficient enough to find a new power supply in order to start active safety systems or to find another heat sink for reactor core.
In this paper we will present current status in implementing a complete new passive safety system to an active safety system design plan and its major implication in nuclear safety procedures because current procedures in case of loss of cooling agent implies depressurisation of steam generator by releasing steam in atmosphere in order to have a long term heat sink comparing with isolation condenser system that implies preserving all secondary circuit inventories and water circulating using natural circulation due to buoyancy effect. All heat transfer calculation, natural circulation and operation of this system applies also to LFR ALFRED passive system with isolated condenser. The main difference between these two branches is made by implement or not of a non-condensable subsystem designed for self-regulate heat transfer to isolated condenser heat exchanger. The noncondensable gases are a must for ALFRED LFR reactor due to high freezing point the lead, in CANDU the effect of non-condensable is to reduce temperature gradients for long time cooling of reactor core in order to reduce stress and aging of power plant components.
Historically, the reference procedure for simulation of severe accidents in sodium fast reactors (SFRs) foresees the use of different codes for different phases of the transients. The mechanistic code SIMMER, in its 2D version SIMMER-III, and in its 3D version SIMMER IV, is one of reference codes for severe accident simulations in SFR, in particular for the transition phase, including massive core melting that starts after failures of hex-cans of fuel sub-assemblies.
Relevant efforts were made at KIT in order to extend the applicability of the SIMMER code to the earlier stages of the transients. However, SIMMER still needs an external input regarding the geometry and properties of fuel pins. In order to take into account the influence of irradiation on fuel pin properties at near nominal conditions, and consequently to perform transient analyses, an interface between SIMMER and fuel performance codes was developed.
An example of application of this interface is presented in this paper, including transient simulations for the Fast Flux Test Facility (FFTF).
The simulations are performed at KIT in the framework of the ongoing IAEA collaborative research project (CRP) benchmark, which includes comparisons to experimental results. The differences between the regular SIMMER approach with user-defined fuel properties and a more advanced approached, i.e. using the mentioned interface with a fuel performance code, shows the impact of a more detailed fuel pin treatment on the transient evolution.
The application of the interface - that provides detailed on fuel properties at the beginning of the transient - may reduce the related uncertainties in results of SIMMER simulations for the initiation phase of a severe accident.
The multiphysics modelling approach has extented the framework of conventional reactor analysis towards the most innovative next generation reactor concepts. In this perspective, the study of the Molten Salt Fast Reactor (MSFR), given its tight coupling between thermal-hydraulics, neutronics and fuel chemistry, is one of the most prominent examples.
The transport of non-soluble fission products is a major design aspect of fluid-fueled reactors such as the MSFR, where they are carried by the fuel flow and can deposit on reactor boundaries in the form of solid precipitates. The aim of the present study is a preliminary investigation of appropriate strategies for the extension of state-of-the-art multiphysics MSFR codes to fission product transport simulation, which is currently being addressed in the context of the SAMOSAFER H2020-Euratom project. A Eulerian single-phase transport model is developed and integrated in a previously developed multiphysics solver based on the OpenFOAM CFD library. We discuss the problem of the modelling of particle-wall interaction and surface deposition, with reference to classical turbulent particle transport and deposition approaches.
The resulting model is tested on a well-known 2D MSFR benchmark case, showing the combined effect of complex flow patterns and distributed production on particle transport and deposition. Analytical formulations for simplified reference problems are also employed to highlight the role of physical parameters affecting deposition rates and concentration gradients close to walls, giving rise to potential numerical issues in more complex geometries. Finally, we discuss the inclusion of inertial transport mechanisms for particles of non-negligible size as a further extension of the methodology.
The fuel assembly of the sodium-cooled fast reactor generally adopts a hexagonal structure, and the fuel rod bundles are wrapped by a hexagonal tube. Considering the effect of void swelling, there is a certain gap between the outermost fuel rods and the inner wall of hexagonal tube. Therefore, the rod bundle arrangement is random, and its tightness is difficult to determine. In order to study the thermal-hydraulic characteristics of the rod bundles with different degrees of arrangement tightness, three-dimensional simulation of fuel assemblies with 169 rods was performed with FLUENT code. The results show that when bundles are arranged in the center of the hexagonal tube, the maximum temperature difference between the fuel rods in a completely loose state and that in a completely compact state can reach 30 K, and when bundles are in a completely compact state, deflection of the rod bundle has little effect on the maximum temperature even if all rod bundles are squeezed to one side of the hexagonal tube. This article can provide a reference for the thermal hydraulic design of the fast reactor core.
Knowledge of the rate and total amount of fuel discharged from a failed fuel pin during postulated low probability accidents such as unprotected transient overpower event in sodium and lead cooled fast reactors is crucial for the prediction of fuel dispersion behavior, potential for coolant channel blockage, and fuel pin failure propagation. The latter phenomenon is the motivation for this study.
Both solid and molten fuel ejection from a failed fuel pin are analyzed. The models are developed for a design adopting annular fuel pellets. Outside the fuel pin the fission gas/fuel release takes the form of a high-momentum fission gas jet submerged in the metal coolant stream that flows upward through the reactor core. The jet contains the ejected fuel in the form of solid grains or molten drops. The jet entrains the surrounding coolant so that, in addition to fuel particles, the jet carries metal coolant drops downstream where it impinges upon a neighboring fuel pin. The strength of the fuel drop/coolant drop interaction within the jet affects the intensity of jet impingement heat transfer upon the neighboring fuel pin, which is the essential crux for fuel pin failure propagation.
A model of a submerged high-momentum fission gas/fuel jet is constructed. The model can predict the structure of the jet; namely, the jet velocity, jet diameter, fuel volume fraction, entrained metal coolant volume fraction, and jet density as a function of downstream distance. A theory for metal coolant droplet deposition and heat transfer to the cladding of a neighboring (target) fuel pin is proposed. In the theory the solid fuel particles or initially molten fuel drops, which are crust covered by the time they arrive at the target pin, do not exchange heat with the target pin cladding during the short periods of fuel particle/cladding contacts. The entrained and fuel-heated metal coolant drops exchange heat with the target pin cladding, but only after splashing off the cladding surface and then returning to the surface in the form of smaller secondary drops at a rate dictated by fission gas flow turbulent velocity fluctuations.
Target pin impingement heat transfer by a pure fission gas jet released into liquid sodium was studied experimentally in 1970s and it was found that the jet insulates (blankets) the target pin, but the presence of entrained liquid sodium drops in the jet limits the target pin's surface temperature rise to less than about 200 C. It is shown here that the addition of hot escaped fuel particles to the fission gas jet can lead to much higher target pin surface temperature and possible clad failure by melting. The impingement heat transfer model developed here is shown to be capable of accounting for many observed features of the experiments. The fuel ejection and jet impingement heat transfer models developed in this study have been incorporated into the SAS4A/SASSYS-1 code currently used by Westinghouse for safety analysis of the Westinghouse Lead Fast Reactor.
Lead and lead-bismuth eutectic are promising reactor coolants due to a series of advantages, but there are also some challenges as corrosion and erosion of structural materials, impurity formation and their deposition as well as coolant reaction with air which can cause a decrease of the heat transfer performance.
A numerical model has been developed in order to simulate the thermal-hydraulic parameters of the steam generator for the ALFRED reactor (Advanced Lead Fast Reactor Demonstrator) which has a thermal power of 300 MW. The model has been verified in the case of normal operating conditions and shows good agreement with previous RELAP 5 results. This model has been applied in order to simulate the following accident condition: air ingress in the primary system which causes lead oxide deposition on the steam generator tubes, which shows a large decrease of the extracted thermal power of the steam generators.
Assessing these effects are important in order to recognize that this type of event has happened and to develop mitigation techniques, although these conditions might be hard to mitigate or control because of various feedback effects (e.g. deposition rate increases as the temperature of the lead decreases).
In the frame of the Italian activities focused on the deployment of Lead Fast Reactor (LFR) technology, in the last few years Politecnico di Torino has been developing a code for the multiphysics analysis of liquid-metal cooled cores, named FRENETIC (Fast REactor NEutronics/Thermal-hydraulICs). The code aims at the time-dependent simulation of the neutronics (NE) and thermal-hydraulics (TH) of the reactor core, composed of closed hexagonal fuel assemblies (FAs). The NE module adopts a multigroup diffusion model for neutrons, discretized with a coarse-mesh nodal method at the assembly level. The TH module solves the transient 1D mass, momentum and energy conservation equations along the axial length of each closed FA. Neighbouring FAs are (weakly) thermally coupled in the radial direction by taking into account inter-assembly heat transfer. The two modules are coupled passing the power density distribution (computed by the NE module) to the TH module, which computes then the temperature distribution in both the fuel and the coolant. This temperature map is sent back to the NE module, that uses it to update the cross-sections adopted by the NE module in the neutron flux calculations. The code has been validated against some experimental data sets, including the international benchmark on the Shutdown heat removal tests on the EBR-II proposed by the IAEA in a Coordinate Research Project (CRP).
In this work, FRENETIC is applied to study some safety-relevant transients for the Advanced Lead Fast Reactor European Demonstrator (ALFRED), with specific focus on TH initiating events, such as an unprotected loss-of-flow accident. Notwithstanding the simplified models implemented in FRENETIC, the computational burden associated to extensive parametric studies of the transient, coupled NE/TH behaviour of the full core remains significant. This motivates the use of a Parametric Non-Intrusive Reduced Order Model (PNIROM), trained by means of FRENETIC results. This approach, based on a combination of Proper Orthogonal Decomposition and Radial Basis Functions techniques, allows to build a meta-model that approximates the full-order model solution with a limited computational effort. The off-line FRENETIC calculations are relatively computationally demanding, as they require to compute the solution snapshots for an exhaustive set of input parameters, e. g. the mass flow reduction factor, but they are performed only once, to train the model. When the model approximation error becomes acceptable, the PNIROM can be adopted to estimate the code outcome also for cases not used for the training. The PNIROM will be used to study the influence of different loss-of-flow scenarios on the reactivity evolution, as well as on the core thermal power density behaviour.
The sodium-cooled fast reactor is one of the major designs in the fourth-generation nuclear reactor. Currently, research efforts have been intensified on sodium-cooled fast reactors. The China Experimental Fast Reactor (CEFR) has been commissioned, and the 600MW Demonstration Fast Reactor (CFR600) is currently under construction. For the pool-type sodium-cooled fast reactor, the thermal-hydraulic analysis of the primary cooling system is one of the important tasks in its design and development. However, most of the developed sodium-cooled fast reactor programs are dedicated programs for specific reactors. Their application to other pool-type sodium-cooled fast reactors require significantly large modification and could result in a system with insufficient flexibility and operability. Considering the specific structure and operating characteristics of the primary cooling system of fast reactors, this paper optimizes the original one-pool sodium cooling system code by adopting a modular modeling approach, tabular reactor parameters, parameterized control bodies to establish a mathematical model of the system.
The focus of this paper is mainly on the following two aspects: One aspect is the use of a reactor design parameter table (EXCEL) for a specific reactor type. The structural geometric parameters and physical parameters of the different structures of the reactor are modified, and the database required parameters are read in from the EXCEL table. This aids the visualization of the parameters, facilitates faster and simpler modification of the reactor parameters, and results in an easily adaptable application for different reactors. The second aspect is to model based on lumped parameters. With the aid of the control body, the idea is to number the modules of the primary loop, and to parameterize the number of control bodies, depending on the specific structure and operating characteristics of the primary loop of the pool-type sodium-cooled fast reactor. The novelty in this work is the complexity of the physical problems solved, which include how to reasonably divide and select the control body of the research object, integrated the abov parameter table, read in the number of control bodies in the relevant part. To ensure computing efficiency and reduce the computation cost of the program, the accuracy and speed of the operation are improved. To evaluate the effectiveness and verify the robustness of the program, the computation results from the program is compared with the steady-state thermal-hydraulic data of the primary circuit system of the China Demonstration Fast Reactor at full power and the temperature distribution of the primary cooling system. The accuracy of the control body division, the intermediate heat exchanger (IHX) under accident conditions, and the sensitivity analysis of the temperature difference between the inlet and outlet of IHX is also verified. The results show that this program is capable of providing a complete, robust and flexible computation. Also, the program provides a simplified modification for different reactor parameters to accurately simulate different configurations of the pool-type sodium-cooled fast reactor. Moreover, the number of control bodies can also be realistically selected to accurately perform specific analysis and to achieve the sensitivity requirements of the physical parameters.
The capability to perform reliable numerical analyses of large thermal-hydraulics system is one of the key features for the development of the new GEN IV of nuclear power plants. The analysis of such large and complicated systems often requires several simplifying assumptions which may help to cope with the required computational efforts: System Thermal Hydraulics (STH) codes were thus developed as a useful tool allowing to obtain sufficiently reliable predictions in reasonable calculation times. STH codes usually assume that the addressed thermal-hydraulics system may be simulated adopting 1D analyses: this of course reduces the computational cost of the calculations but introduces assumptions that may not be consistent with the addressed problem. In particular, the presence of large 3D environments, for which the one-dimensional approach is no more suitably applicable, represent one of the intrinsic limits of STH codes.
In several of the proposed Liquid Metal Fast Reactors (LMFRs ) designs, the reactor core is located inside a large pool which contains the whole primary loop: the reactor core, the steam generators and primary pumps represent some of the most relevant internal component of the reactor vessel. It is clear that, in such a complicated configuration, which involves large 3D environments, the numerical 1D approach of STH codes is no more suitable. As a consequence, CFD, which instead proved to have very good capabilities in dealing with complicated 3D geometries, may thus be considered for the analysis of such environments. Nevertheless, on the other hand, CFD requires large computational efforts and computational time; in addition its actual capabilities in dealing with two-phase flow conditions is still questionable. Consequently, none of the cited approaches seems to be able to provide a reliable analysis of the addressed thermal-hydraulic problem.
During the last years, the University of Pisa has been involved in several EU projects aiming at providing a better understanding of the thermal-hydraulics aspects of LMFRs. In the frame of these studies a coupled STH/CFD approach has been developed trying to overcome the observed limits of both the codes in stand alone calculations. STH codes are adopted for the simulation of the thermal-hydraulic system, nevertheless CFD intervenes when a more detailed analysis of complicated 3D environments is required. This way the capabilities of both the codes are maintained, trying to avoid the drawbacks.
The present paper reports on the recent works performed at the University of Pisa trying to highlight both the advantages and limits of the adopted STH/CFD coupled applications. The performed works also helped in defining useful guidelines for a suitable use of the addressed approach providing a sound basis for a more extensive adoption of STH/CFD calculations in the frame of future works.
To support the licensing of a nuclear installation any software tool used in substantiating performance-related claims, with particular regard for the safety ones, must be qualified. Nuclear Regulatory Authorities indeed demand that for software used in the safety demonstration, the uncertainties affecting target quantities, in a given application domain, are known with traceable confidence.
In the perspective of approaching the licensing of the Advanced Lead-cooled Fast Reactor European Demonstrator (ALFRED), the need for qualified software tools involves ANTEO+, a sub-channel code used for the steady-state thermal-hydraulic analysis of the fuel assemblies in support to the design of the core.
The path to qualification of any software is composed of several steps, the most relevant of which are validation and the subsequent uncertainty quantification. The former, in particular, is also a milestone of the software development process and is here presented in relation to the ANTEO+ code, with a scope extended to the steady-state thermal-hydraulic analysis of fuel assemblies of liquid metal-cooled reactors in general.
Based on the guidelines and best practices in place in the European context, the validation effort has been devised as composed of a number of steps with the intent of unambiguously define the target quantities (in the ANTEO+ case namely, the coolant temperature field, the clad temperature and the pressure drops through the pin bundle) and the domain over which the validation claim can be supported. Via a cascade of physical dependencies of the quantities of interest to elementary phenomena, and by connection of the latter with basic operational and geometrical parameters defining a given configuration, it is indeed possible to establish a bounding validation domain.
Inside such domain both a separate-effects (i.e., for each main phenomenon independently) and an integral (i.e., for each target quantity) validation are comprehensively performed so to retrieve the uncertainty to be associated to the target quantities for a given confidence level, feeding the successive uncertainty quantification step for completing the qualification path.
This rigorous approach, once fine-tuned via a dialogue with the pertinent regulatory body can be applied in the future, together with the standard practices of software quality assurance, to all other computational tools envisaged for the given safety demonstration.
The present paper is focused on the thermal fluid dynamic simulation approach for fuel bundle in liquid lead environment. In the development program for LFR, reliable tools are needed to predict and investigate the main TH phenomena in a fuel pin bundle. CFD codes are considered effective instruments to address the design and safety assessment in the nuclear field, however large domains with high fidelity will require enormous computational efforts, especially if transient conditions are studied. In this paper different approaches are investigated to reduce the computational cost of the analysis by mean of the ANSYS CFX code, validating the result against experimental database on CIRCE, a lead-bismuth pool type facility located at the ENEA research center in Brasimone (Italy). Firstly, the resolution of the boundary layer has been performed by comparing the SST k-ω and Standard k-ε two equation turbulence model with RMS turbulence model in the evaluation of global quantities. Porous media setting parameters are extracted from the most reliable turbulence model at different mass flow rate. Eventually, three different computational cost reduction methods are applied: a high y+ approach with the application of wall function (30<y+<300) for the boundary layer analysis, a porous media model in the entire bundle region and an hybrid porous media model where the interaction wrapper-coolant is preserved outside the porous media domain. The advantages and disadvantages of the three strategies are summarized in the conclusion section.
Today the International Atomic Energy Agency (IAEA) fosters an international exchange of information on the advances in reactor technology, including for Molten Salt Reactors (MSRs).
This paper mainly considers the MOlten Salt Actinide Recycler & Transmuter (MOSART) system with homogeneous core without U-Th support fueled with different compositions of transuranic elements from VVER 1000/1200 used nuclear fuel. Last developments concerned single fluid MOSART design addresses advanced large power unit with main design objectives to close nuclear fuel cycle for all actinides (An), including Np, Pu, Am and Cm. The optimum spectrum for Li,Be,An/F MOSART is fast spectrum of homogeneous core without graphite moderator. The effective flux of such system is near 1x1015 n cm-2 s-1. Single fluid 2.4 GWt MOSART unit can utilize more than 250 kg of minor actinides per year from VVER 1000/1200 UNF.
The main attractive features of MOSART system deals with the use of (1) simple configuration of the homogeneous core (no solid moderator or construction materials under high flux irradiation); (2) proliferation resistant multiple recycling of actinides (separation coefficients between transuranic (TRU) and lanthanide groups are very high, but within the TRU group are very low); (3) the proven container materials (high nickel alloys) and system components (pump, heat exchanger etc.) operating in the fuel circuit at temperatures below 1023K, (4) core inherent safety due to large negative temperature reactivity coefficient (-3.7 pcm/K), (5) long periods for soluble fission products removal (1-3 yrs). The fuel salt clean up flowsheet for the Li,Be,An/F MOSART system is based on reductive extraction in to liquid bismuth.
The need for the experimental 10 MWt Li,Be,An/F MOSART test unit to demonstrate the control of the reactor and fuel salt management with different minor actinides loadings for start up, transition to equilibrium, drain-out, shut down etc. with its volatile and fission products, is also discussed.
The main design choices and thermal hydraulics characteristics for large power and test Li,Be,An/F MOSART units are explained.
The paper has the main objective of presenting the thermal and hydraulic peculiarities of the 2400 MWt and 10 MWt MOSART units while accounting technical constrains and experimental data on fuel Li,Be,An/F salt. In this paper, results of the thermal hydraulic simulation made with ANSYS Fluent code were used to improve core and fuel circuit configuration.
As the result of the calculation optimization, homogeneous cores of 2400 MWt and 10 MWt MOSART units satisfy the most important requirements: (1) no recirculation or stagnation zones of fuel salt stream in the homogeneous core and (2) maximum temperature of solid nickel reflectors is low enough (1027K) for a long time operation, (3) fuel inventory outside core is minimized.
Discharge of sodium coolant into containment from a sodium fast reactor (SFR) can occur in the event of a pipe leak or break. In this situation, some of the liquid sodium droplets discharged will react with oxygen in the air before reaching to the containment. This phase of the event is normally termed the sodium spray fire phase. Unreacted sodium droplets pool on the containment floor where continued reaction with containment atmosphere oxygen occurs. This phase of the event is
normally termed the sodium pool fire phase. Both phases of these sodium-oxygen reactions (or fires) are important to model because the heat addition and aerosol generation that occur. Any fission products trapped in the sodium coolant may also be released during this progression of events, which if released from containment could pose a health risk to workers and public. This paper describes progress of an international collaborative research in the area of SFR sodium fire modeling between the United States and Japan under the framework of the Civil Nuclear Energy Research and Development Working Group (CNWG). In this collaboration between Sandia National Laboratories (SNL) and Japan Atomic Energy Agency (JAEA), the validation basis for and modeling capabilities of sodium spray and pool fires in MELCOR of SNL and SPHINCS of JAEA are being enhanced.
In this paper, we document MELCOR and SPHINCS sodium pool fire model validation exercises against the JAEA’s sodium pool fire experiments, F7-1 and F7-2. We also describe our proposed enhancement of the sodium pool fire models through addition of thermal hydraulic and sodium spreading models that enable a better representation of experimental results. These enhancements establish a refined means to characterize key phenomena observed in the sodium pool fire experiments. With these enhancements, both MELCOR and SPHINCS are able to capture the F7- 1 and F7-2 experimental data well. In addition, to the assessment of the sodium pool fire dynamics,
additional analysis sodium fire aerosol generation is reported in this validation study. Despite
limited experimental data being available, the relevant sodium aerosol generation trends are
characterized to develop insights of relevance to the design of future experimental campaigns.
In case of a Hypothetical Core Disruptive Accident (HCDA) in Sodium-cooled Fast Reactors (SFRs) with vessel failure or pipe rupture in the primary system, large amounts of contaminated sodium with suspended or solved fuel particles and fission products are expected to be released into the containment. For source term considerations, the investigation of the release of volatile species from evaporating pools into a gas atmosphere is of main importance to determine the sodium pool retention capability.
In the past, theoretical mechanistic models for the prediction of sodium and volatile fission products release into an inert gas atmosphere have been developed. In these models, the evaporative release of the volatile FPs and sodium is governed by diffusive and convective transport processes. Based on a mass transfer formulation, the retention factor (defined as the ratio of pool concentration to released concentrations) of the species of interest is calculated.
This work synthesizes the results of the work developed by CIEMAT in the frame of the ESFR-SMART project. A critical review of the theoretical mass transfer model applied to the FPs release in hot sodium pools has been done through the comparison of the model results against experimental data from NALA program. The comparison reveals the high sensitivity of the model on the gas transport coefficient calculation, i.e., the Sherwood correlation choice with differences in the Na mass flux up to one order of magnitude. In the model, the possible effect in the species release of Stefan flow and condensation within the thermal boundary is taken into account through two correction factors. Although no major impact has been observed for the condensation effect, the Stefan flow effect must be taken cautiously because of the dependence on fitted variables in the correction factor equation. As future work, a two-film model will be proposed in which the volatile species within the sump of sodium will be transferred to the atmosphere by diffusion due to concentration gradients but taking into account the enhancement due to dragging by the sodium vapour flow from the liquid gas interface.
With the increase in computational power and capacity, and the advancements being made in numerical modeling, it has become possible to model the various physical phenomena that take place in nuclear reactors with more and more detail and accuracy. This includes phenomena related to structural mechanics, fluid dynamics and reactor physics amongst others. Additionally, there has been an increased interest to simulate these combined, interacting phenomena simultaneously by coupling various numerical tools. This coupling of different codes is currently a topic high on the research and development agenda of the international nuclear community. It is also a focus point within the research done at NRG in the national PIONEER research program funded by the Dutch ministry of economic affairs.
This focus has resulted in two branches of research at NRG: multi-scale modeling of the complete primary system of a nuclear reactor by coupling a 3D Computational Fluid Dynamics (CFD) code with a 1D System Thermal Hydraulics (STH) code, and multi-physics modelling through the coupling of a CFD code to a Computational Structural Mechanics (CSM) code in order to perform Fluid-Structure Interaction simulations. The relation between these two fields lies in the efficient and correct coupling of data between the two codes, a crucial aspect in order to get a converged and accurate solution. This paper presents simulation results of both branches applied to fast reactors. First, multi-scale simulations of the primary system of various fast reactors are presented, both in steady-state and transient conditions. Secondly, FSI simulation results with liquid metal as coolant are presented. Finally, as both fields of research require the coupling of codes, it has led to the creation of an independent, external, Fortran-based coupling tool named myMUSCLE (MultiphYsics MUltiscale Simulation CoupLing Environment) that arranges the efficient and robust coupling of the different codes. This paper presents the proof-of-principle and first validation of the myMUSCLE tool under development.
The improved understanding of the safety, technical gaps, and major uncertainties of advanced Fast Reactors (FRs) will result in designing their safe and economical operation. This paper focuses on applying results of high-fidelity thermal-hydraulic simulations to inform the improved use of lower-order models within fast-running design and safety analysis tools to predict improved estimates of local safety parameters for efficient evaluation of realistic safety margins for FRs. Due to their complexity, the high-fidelity calculations are computationally expensive. This motivates the use of low-fidelity models, which are less comprehensive but provide numerical efficiency required for practical applications in design and safety evaluations. To address the integration of high-fidelity and low-fidelity codes to predict quantities of interest in an efficient manner, a framework of High-to-Low (Hi2Lo) model informing procedures is usually utilized for hydraulics and fuel simulations in a fast reactor core and then combined in overall thermal-hydraulics calculations. The advanced sub-channel thermal-hydraulic code CTF and its fuel rod solver CTFFuel capabilities have been extended to model FRs. These improvements included adding sodium material property correlations for pressure drop in the hexagonal fuel bundles, a flow mixing correlation for wire-wrapped rod bundles, and the addition of a heat conduction model across sub-channel gaps. Further work was conducted to implement additional friction factor correlations and a correlation from a scoping analysis for fast test reactor cores for irradiated fuel thermal conductivity. Modifications to CTF/CTFFuel were made to enable parallel full-core modeling of fast reactors. To verify and validate the above-described developments, benchmarks to publicly available FR experimental data have been performed and code-to-code comparisons have been conducted. The obtained verification and validation results demonstrated that CTF/CTFFuel has the capability to simulate wire-wrapped FR fuel and to perform the parallel full-core modeling of FRs. This paper focuses on further enhancements of CTF/CTFFuel by applying results of high-fidelity simulations to inform the improved use of lower-order models within CTF/CTFFuel. The high-fidelity computational fluid dynamics code Nek5000 is used to inform CTF for improved modeling of pressure losses and inter-channel mixing in wire-wrap fuel bundles. The use of metallic fuel requires understanding of the behavior of the uranium alloys. The doping of the uranium metal with zirconium affects its properties, such as thermal conductivity and heat capacity, as well as its performance. The presence of the sodium between fuel and cladding as a bonding element is another difference between the mixed oxide or uranium dioxide fuels, and the uranium metal fuel. The bonding sodium can infiltrate and diffuse into the fuel due to porosity interconnection, further modifying the fuel temperature profile and effective thermal conductivity. We use as a high-fidelity tool (VASP) for ab initio Molecular Dynamics methods to study the interaction of the metallic fuel and the infiltrating sodium to develop a correlation for the thermal conductivity as a function of zirconium content and burnup. The above-described Hi2Lo model information improvements are being verified and validated using benchmarks such as the OECD/NRC Liquid Metal Fast Reactor Core Thermal-Hydraulic Benchmark and code-to-code comparisons.
Safety calculations of decay heat removal (DHR) in sodium fast reactors involve
complex thermalhydraulics phenomena. In such conditions, the pump is stopped
and a heat exchanger is operating directly in the hot pool. As a consequence,
the flow is governed by natural convection, and a significant amount of the
power is extracted through the side of the subassemblies thanks to the
inter-wrapper flow. The numerical simulation of such transient are challenging,
and require specific validation data. PLANDTL-2 is a large sodium loop
installed in JAEA Oarai, with a test section including 55 subassemblies, among
which 30 are electrically heated, and a hot pool with a dipped heat exchanger.
In this paper, numerical simulations using a hybrid sub-channel/CFD approach
for the test section coupled to a system code for the heat exchanger are
compared to experimental results of DHR tests performed in Plandtl2.
Development of Advanced Reactor Knowledge- and Artificial Intelligence (AI)-aided Design Integration Approach through the whole plant lifecycle, ARKADIA was started in Japan Atomic Energy Agency. ARKADIA will automatically provide possible solutions of design, safety measures, maintenance program to optimize the lifecycle of advanced reactors by using the state-of-the-art numerical simulation technologies. In the first phase of this project, ARKADIA-Design and -Safety will be constructed individually for different applications. This paper describes a development concept, basic structure, functions of each system which comprises ARKADIA, core technologies of ARKADIA-Design and -Safety.
ARKADIA integrates state-of-the-art numerical simulation technology, accumulated knowledge, and AI technology. It provides automatic optimization of plant design and various conditions such as safety, risk response, economics, and environmental compatibility. To realize optimization of plant lifecycle, ARKADIA consists of a Virtual plant Life System (VLS), a Knowledge Management System (KMS), an Enhanced and AI-aided optimization System (EAS), and a platform controlling the three systems. EAS constructs the objective function according to user’s requirements at the beginning of evaluation. And then, EAS obtains data required for numerical simulation from KMS and selects appropriate evaluation condition. VLS can evaluate all possible events during the plant lifecycle by numerical simulation. After evaluation by VLS, EAS calculates the objective function from the analytical result. If necessary, EAS changes the evaluation conditions to find an optimum solution iteratively. AI technology will be incorporated for highly efficient optimum solution retrieval.
ARKADIA-Design offers functions to support design optimization both in normal operating conditions and design basis events mainly during the conceptual design stage in the fields of core design, plant structure design including thermal-hydraulics analysis, and maintenance plan design. ARKADIA-Design performs a multi-level and multi-physics simulation such as neutronics−core deformation−thermal hydraulics coupled analysis for core design. A one- and three-dimensional coupling methods and a multi-physics code-to-code synchronization script were developed in this study. ARKADIA-Safety is developed for the purpose of automatic optimization of the severe accident (SA) management and its feedback to the plant design. ARKADIA-Safety performs in- and ex-vessel integrated numerical analysis of the SA, the statistical safety evaluation, and the dynamic PRA. ARKADIA-Safety includes SPECTRA (Severe-accident PhEnomenological computational Code for TRansient Assessment) as a core simulation code. SPECTRA code consists of in- and ex-vessel modules which have a thermal hydraulics model and individual models for physical phenomena appearing in SA. The in- and ex-vessel modules are coupled by exchanging the amount of leaked sodium and debris at every time step. A loss of reactor level event which is one of the representative SA scenarios was successfully simulated by SPECTRA. The technical data obtained from previous fast reactor development programs are being stored into the KMS. The KMS will provide knowledge data which is required for numerical simulation in ARKADIA-Design and -Safety. The second phase of this project will be completed within ten years to provide a system that fully integrates ARKADIA-Design and -Safety.
The heat removal in the refrigeration pools of GEN-IV reactors is done by passive natural convection. The assessment of its flow pattern, safety operation, as well as the analysis of possible accident scenarios could be done using CFD computations.
Nevertheless, CDF calculations have a certain credibility deficit. That means that practitioners should carry out costly experiments. This can be partially remediated by providing a thorough analysis of the calculations uncertainty. However, that task is not simple and requires the usage of a specialized fast methodology.
We propose to use a surrogate model built utilizing Proper Orthogonal Decomposition and Galerkin projection. The methodology is based on multiple snapshots of an initial value problem computed with a CFD solver and post-processing. This results in a reduced base. Subsequently, the governing equations are rewritten into that base. Thus, a model soundly derived from first principles and without further assumptions is deduced.
The results found with such methodology, already presented in previous meetings, have been encouraging. We now focus on the latest improvements. This concerns two topics: a) The extension of our methodology to include turbulence modeling, specifically the standard k-ε model; b) The inclusion of non-homogeneous, complex and arbitrary boundary conditions. Those two topics have required exhaustive analysis and in particular the creation of a complex formulation due to the usage of a Sobolev inner product whose elaborated and detailed treatment will be discussed in detail.
The SAS4A/SASSYS-1 system level computer code is used by Westinghouse to perform safety analysis of the Westinghouse LFR. The SAS4A/SASSYS-1 code is capable to simulate anticipated operational occurrences, design basis accidents and beyond design basis accidents of liquid metal fast reactors. The major strengths of SAS4A/SASSYS-1 include a complete reactivity feedback model, a comprehensive fuel rod performance model, and extensive applications to liquid metal reactors demonstrated as part of multiple domestic and international programs.
Over the past few years, the SAS4A/SASSYS-1 code was further improved as part of various DOE programs to extend its applicability to LFR systems. The improvements include the capability of mechanistic source term analysis through coupling with the FATE code developed by Fauske & Associates, oxide fuel modeling enhancements, the capability to simulate the passive heat removal system through coupling with the GOTHIC code, the capability to simulate in-vessel primary heat exchangers, and the enhancement of computer code verification and validation base specific to LFR systems, such as verification database, separate effects tests, and integral effects tests. The applicability of SAS4A/SASSYS-1 to the LFR system is summarized in the paper.
One of the fuel assembly designs considered for sodium-cooled fast reactors utilizes wires helically wrapped around each fuel rod as spacers. The wires keep the fuel pins separated, enhancing the turbulent mixing and heat transfer, but also increasing the pressure drop. This study investigates the effects of this geometrical feature on pressure drop and velocity distributions. Pressure and velocity fields within the assembly are calculated using the Reynolds-Averaged Navier-Stokes (RANS) and Large-Eddy-Simulation approaches within the context of Computational Fluid Dynamics (CFD) framework. The configuration under study is based on the 61-pin wire wrapped hexagonal test bundle at Texas A&M University, that has produced high-fidelity experimental data of the flow velocities and pressure drops at different locations in the bundle, and under a wide range of flow rates spanning the laminar, transition and turbulent regimes. Effects of localized blockages have been analyzed. Results of friction factor and velocity distribution are compared with the experimental data as well as with predictions of the Upgraded Cheng and Todreas Detailed (UCTD) friction factor correlation, confirming that RANS and LES predictions are in a reasonable agreement with the experimental data. Available experimental data was used to validate the simulation of the blocked scenario simulated using unsteady RANS models. Results show an increase in the pressure drop for the blocked case as well as a change of the flow field around the blockage in comparison with the base case (no blockage). It is observed that the accuracy of unsteady RANS models is reduced when they are used to simulate flows in presence of blockages, indicating that a scale-resolving simulation would be more suitable for these conditions.
Loss of flow and other thermal transient scenarios in sodium-cooled fast reactors (SFRs) are often accompanied with thermal stratification in the hot or upper plena. As a result, adjoining structural walls are subjected to high temperature gradients which can result in complex multi-physical problems. Thermal stratification in the plenum can cause differential thermal expansion, introduce non-linear reactivity changes and impact natural circulation driven passive safety features. Within a range of flow rates, the thermally stratified interface can experience flow and buoyancy induced fluctuations which can cause sudden temperature changes in the solid structures leading to accelerated fatigue. SFR system scale models lack the fidelity to capture thermal stratification accurately and advanced computational fluid dynamics (CFD) codes need validation. Kansas State University has developed an experimental facility to study integral effects in the scaled reactor plena and rest of the system using Gallium as heat transfer fluid which acts as a surrogate for Sodium. The use of liquid Gallium simplified the design and operation of experimental facility and allowed the deployment of validation grade measurement techniques. Several experiments were conducted to emulate cold shock transients and loss of flow transients with wide range of parameters. Temperature data was captured using Optical Fiber Domain Reflectometry and velocity data was captured using Ultrasonic Doppler Velocimetry. Experimental data was used to qualify system-scale models and CFD models.
Global interest in fast reactors has been growing since their inception in 1960 because they can provide efficient, safe, and sustainable energy. Their closed fuel cycle can support long-term nuclear power development as part of the world’s future energy mix and decrease the burden of nuclear waste. In addition to current fast reactors construction projects, several countries are engaged in intense R&D and innovation programs for the development of innovative, or Generation IV, fast reactor concepts. Within this framework, NINE is very actively participating in various Coordinated Research Projects (CRPs) organized by the IAEA, aimed at improving Member States’ fast reactor analytical simulation capabilities and international qualification through code-to-code comparison, as well as experimental validation on mock-up experiment results of codes currently employed in the field of fast reactors. The first CRP was focused on the benchmark analysis of Experimental Breeder Reactor II (EBR-II) Shutdown Heat Removal Test (SHRT-17), protected loss-of-flow transient, which ended in the 2017 with the publication of the IAEA-TECDOC-1819. In the framework of this project, the NINE Validation Process– developed in the framework of NEMM (NINE Evaluation Model Methodology) – has been proposed and adopted by most of the organizations to support the interpretation of the results calculated by the CRP participants and the understanding of the reasons for differences between the participants’ simulation results and the experimental data. A second project regards the CRP focused on benchmark analysis of one of the unprotected passive safety demonstration tests performed at the Fast Flux Test Facility (FFTF), the Loss of Flow Without Scram (LOFWOS) Test #13, started in 2018. A detailed nodalization has been developed by NINE following its nodalization techniques and the NINE validation procedure has been adopted to validate the Simulation Model (SM) against the experimental data of the selected test. The present paper intends to summarize the results achieved using the codes currently employed in the field of fast reactor in the framework of international projects and benchmarks in which NINE was involved and emphasize how the application of developed procedures allows to validate the SM results and validate the computer codes against experimental data.