Transient Reactor Analysis Code Research Articles

Whole-core analysis of Watts bar benchmark with three-dimensional MOC code STREAM3D

This paper presents a high-fidelity simulation of the Organization for Economic Co-operation and Development (OECD) Nuclear Energy Agency (NEA) 3D whole-core Watts Bar benchmark using the UNIST in-house STREAM3D (Steady State and Transient Reactor Analysis code with Method of Characteristics) neutronic code. The benchmark encompasses various whole-core exercises, including single physics problems, multi-physics simulations, and depletion problems. When comparing parameters during the zero-power physics tests, including ITC, DBW, CRW, and criticality tests, STREAM3D results indicate a strong agreement with the measured data and KENO-VI. The comparison with the MC21/CTF code in 3D HFP BOC condition demonstrated strong agreement, with only a 0.42% difference in the normalized radial power distribution, a 0.38 K difference in the RMS of the assembly coolant exit temperature, and a mere 4 ppm difference in CBC.

High-Fidelity Transient Analysis with STREAM Based on the Predictor-Corrector Quasi-Static Method

The predictor-corrector quasi-static method (PCQM) is used to solve the transient problem in the STREAM code, a steady-state and transient reactor analysis code with the method of characteristics. In PCQM, the angular neutron flux undergoes a factorized split to form the product of shape and amplitude functions. The time-dependent neutron transport equation is solved to obtain the shape function whereas the amplitude function is obtained by resolving the exact point kinetics equations (EPKEs). A two-level coarse mesh finite difference technique is implemented to reduce the transient running time of the transport solution. Moreover, high-order polynomial interpolation is applied to the kinetics parameters utilized in EPKEs to reduce the error when the reactivity insertion is nonlinear. Several numerical benchmarks are solved to justify the application of the procedure, proving that the method maintains solution accuracy.

Depletion chain optimization of lattice code STREAM for LWR fuel assembly burnup analysis

A depletion chain simplification method is applied to UNIST lattice code STREAM (Steady state and Transient REactor Analysis code with Method of Characteristics) in this paper to alleviate the computational burden of depletion calculation associated with a large depletion matrix. A simplified burnup matrix (burnup chain) of 464 nuclides and 10,638 transitions is thus created from a large detailed burnup matrix containing 3,837 nuclides and 43,416 transitions from ENDF/B-VII.0 nuclear decay data. The simplified burnup matrix is optimized for the purpose of calculating effective neutron multiplication factors (keff) and power distributions with reduced computation time and memory usage. The nuclide selection method relies on the Generalized Perturbation Theory (GPT): by exploiting the adjoint function of nuclide number densities as derived with GPT, a set of nuclides which must be included in the simplified chain (nuclides with important contribution to reactivity) is determined. Numerical verification of the simplified burnup chain is conducted using the deterministic neutron transport analysis code STREAM, developed to perform whole light water reactor (LWR) core calculations with the direct transport analysis method and the two-step method. The simplified burnup chain is tested on the 16 LWR fuel assembly depletion problems from the Virtual Environment for Reactor Application (VERA) benchmarks, including burnable poison (BP) pin cells commonly used in nuclear reactor design such as gadolinia, Pyrex, AIC, B4C and IFBA, and one OPR-1000 fuel assembly with gadolinium bearing fuel depletion problem. For burnup up to 80 MWd/kg and 235U enrichment ranging from 2.1 w/o to 4.6 w/o, the calculations with the simplified burnup chain predict the keff variations within 10 pcm and identical power distributions, with a speed-up factor from 20 to 40 in depletion calculation and a reduction factor by 3–4 of the total simulation time compared to the ones with the detailed burnup chain.

Numerical and experimental study of coolant water flow in ITER divertor outer vertical target

ITER divertor outer vertical target (OVT) consists of 22 plasma facing units (PFUs) and 2 stainless support structures (SSSs) including a complex internal structure for coolant water. Pressure drop and flow rate in each PFU are identified as important parameters for acceptable range in terms of the pumping power of cooling water system and performance of heat removal. In this study, computational fluid dynamics (CFD) and a transient reactor analysis code (TRAC-PF1) which is simpler simulation than the CFD have been performed to estimate the pressure drop and flow distribution in 11 PFUs mounted on an OVT SSS (half of the OVT). Results of both prediction methods showed the proper pressure drop and the homogeneous flow distribution within acceptable range of design requirement. To verify these numerical simulations, the results were compared to experimentally measured values in the flow test. This study demonstrated that the TRAC-PF1 as well as the CFD are applicable tools for further optimization of the divertor OVT cooling structure in view of manufacturing.

Improvements to the modelling of two-phase flow and heat transfer in a transient nuclear reactor analysis code

Accurate two-phase flow heat transfer prediction is of great importance in the analysis of reactor safety and TRAC (transient reactor analysis code) is a best estimate (BE) system code developed for such analysis. The work described here forms part of a research project that aims to evaluate and improve the TRAC code behavior by comparing code predictions with a range of “single effect” experiments. It has been shown that the necessary friction factor can be obtained from a voidage correlation and the drift flux parameters and the wall friction coefficient can be further derived from the annular flow model of Owen and Hewitt. These modifications have been combined to give what is regarded as an optimum code for vertical pipe flows. The work has been extended to make further improvements to the heat transfer correlations of the code. The performance of both the original and the improved codes has been tested against experiments on both evaporating and condensing flows. It is found that the improved code, combining the strongest parts of the best available correlations, gives better predictions in every case.

Consistent Comparison of the Codes RELAP5/PARCS and TRAC-M/PARCS for the OECD MSLB Coupled Code Benchmark

A generalized interface module was developed for coupling any thermal-hydraulic code to any spatial kinetic code. In the design used here the thermal-hydraulic and spatial kinetic codes function as independent processes and communicate using the Parallel Virtual Machine software. This approach helps maximize flexibility while minimizing modifications to the respective codes. Using this interface, the U.S. Nuclear Regulatory Commission (NRC) three-dimensional neutron kinetic code, Purdue Advanced Reactor Core Simulator (PARCS), has been coupled to the NRC system analysis codes RELAP5 and Modernized Transient Reactor Analysis Code (TRAC-M). Consistent comparison of code results for the Organization for Economic Cooperation and Development/Nuclear Energy Agency main steam line break benchmark problem using RELAP5/PARCS and TRAC-M/PARCS was made to assess code performance.

Three-dimensional analysis of water–vapor void fraction in a fusion experimental reactor under water ingress

An integrated Ingress-of-Coolant Event (ICE) test facility was constructed to demonstrate that the International Thermonuclear Experimental Reactor (ITER) safety design approach and design parameters for the ICE are adequate. The integrated ICE test facility simulates the actual ITER components with a scaling factor of 1/1600. Before the integrated ICE experiments the water–vapor two-phase flow characteristics inside the test facility during the ICE were predicted by the three-dimensional numerical simulations using the modified Transient Reactor Analysis Code (TRAC). From the present study it was clarified numerically that the actual ITER safety approach during the ICE is adequate and the present numerical method is very effective to predict the water–vapor void fraction during the ICE.

Analysis and experimental results on ingress of coolant event in vacuum vessel

Experiments on the ingress of coolant event (ICE) in the vacuum vessel of a fusion reactor have been carried out in the Japan Atomic Energy Research Institute (JAERI) as one of the research and development tasks for the International Thermonuclear Experimental Reactor (ITER) to obtain the thermofluid data for validation of safety analysis codes. The ICE experiment is numerically analyzed using the transient reactor analysis code (TRAC) which is one of the codes preparing for the safety analysis of the ITER. The TRAC has been modified so as to analyze the ICE phenomena in the vacuum vessel of a fusion reactor. Several ICE experiments have been carried out as benchmark tests for the safety analysis codes. We have analyzed those experiments by using the TRAC, and considered the difference between the analysis and experimental results. Analysis results of the temperature in the vacuum vessel show a tendency completely different from the experimental result. It is clarified that the present TRAC has not been verified on the scattering behavior of water droplets.

Some computational challenges of developing efficient parallel algorithms for data-dependent computations in thermal-hydraulics supercomputer applications

The Transient Reactor Analysis Code (TRAC), which features a two-fluid treatment of thermal-hydraulics, is designed to model transients in water reactors and related facilities. One of the major computational costs associated with TRAC and similar codes is calculating constitutive coefficients. Although the formulations for these coefficients are local, the costs are flow-regime- or data-dependent; i.e., the computations needed for a given spatial node often vary widely as a function of time. Consequently, a fixed, uniform assignment of nodes to parallel processors will result in degraded computational efficiency due to the poor load balancing. A standard method for treating data-dependent models on vector architectures has been to use gather operations (or indirect addressing) to sort the nodes into subsets that (temporarily) share a common computational model. However, this method is not effective on distributed memory data parallel architectures, where indirect addressing involves expensive communication overhead. Another serious problem with this method involves software engineering challenges in the areas of maintainability and extensibility. For example, an implementation that was hand-tuned to achieve good computational efficiency would have to be rewritten whenever the decision tree governing the sorting was modified. Using an example based on the calculation of the wall-to-liquid and wall-to-vapor heat-transfer coefficients for three nonboiling flow regimes, we describe how the use of the Fortran 90 WHERE construct and automatic inlining of functions can be used to ameliorate this problem while improving both efficiency and software engineering. Unfortunately, a general automatic solution to the load-balancing problem associated with data-dependent computations is not yet available for massively parallel architectures. We discuss why developers should either wait for such solutions or consider alternative numerical algorithms, such as a neural network representation, that do not exhibit load-balancing problems.

A phenomenological model of the thermal hydraulics of convective boiling during the quenching of hot rod bundles. Part III: Model assessment using Winfrith steady-state, post-CHF, void-fraction and heat-transfer measurements and Berkeley transient-reflood test data

Alter completion and assessment of the thermal-hydraulic model developed in companion papers, we performed further independent assessment calculations using recently available Winfrith steady-state, post-CHF, void-fraction and heat-transfer data, and Berkeley transient-reflood test data. This paper discusses the results of those calculations. The thermal-hydraulic model in the Transient Reactor Analysis Code, Modification 2 (TRAC/PF1-MOD2, or simply TRAC), was used to predict the axial variation of void fraction as measured in Winfrith post-CHF tests. The predictions for reflood calculations were reasonable. The model correctly predicted the trends in void fraction due to the effect of pressure and power, with the effect of pressure being more apparent than that of heat flux. The effects of pressure, test-section power, and flow rate on the axial variation of tube wall temperature are predicted reasonably well for a large range of operating parameters. The predicted precursory cooling rates in Berkeley transient-reflood tests were in reasonably good agreement with the measured data. For high-test section input powers, oscillations associated with the void-fraction predictions observed. The predicted quench-front velocities (rewetting velocities) and their variations along the test tube were also found to be in reasonably good agreement with measured data.

An HMS/RRAC Analysis of A High-Level Radioactive Waste Tank Farm

It has been observed that a high-level radioactive waste tank generated quantities of hydrogen and nitrous oxide mixture that are potentially well within flammability limits. These gases are produced from chemical and nuclear decay reaction in a slurry of radioactive waste materials. The slurry is covered by a thick crust and containing sodium nitrate and nitrite salts. Significant amounts of combustible and reactant gases accumulate under the crust over a 110- to 120-day period. These gases cause the crust to rupture, allowing the gases to vent into the air cover-gas space above the crust. We postulated an ignition source for the hydrogen/nitrousoxide/air mixture after it has vented into the cover gas, and we calculated the resulting pressure and temperature loading on the double-walled waste tank using the three-dimensional, time-dependent fluid dynamics coupled with the chemical kinetic computer code HMS (Hydrogen Mixing Studies).The waste tank farm has a ventilation system designed to maintain a slight negative gage pressure during normal operation. We modeled the ventilation system with TRAC (Transient Reactor Analysis Code), and we coupled these two best-estimate accident analysis tools to model the ventilation system response to pressures and temperatures generated by the hydrogen combustion. Significant pressure are produced by this event, which could possibly threaten the tank's integrity.

An experimental study of bottom rewetting in a rod bundle geometry

The effect of various flow parameters on the rewetting velocity in a bottom rewetting phenomena has been experimentally studied in a 3×3 rod bundle geometry. The rod bundle geometry consisted of nine test rods surrounded by a radiatively heated square shroud. The rewetting tests were conducted at atmospheric pressures for low mass and heat flow ranges. Typical rewetting velocities measured in this study were in order to 1 to 5 mm/s. For a given test condition, an increasing inlet subcooling was accompanied by an increasing rewetting velocity. The rewetting velocity was first found to be decreased with increasing vapor quality until a critical value, for a given mass flux. For the vapor qualities higher than the critical value, the rewetting velocity was found to be increased with increasing vapor quality. For a given mass flux, the critical value of the vapor quality corresponded to the flooding limit of the flow. A Transient Reactor Analysis Code (TRAC) post-test calculation was performed for a subcooled inlet reflood test. The TRAC model considered only the test bundle and consisted of four components including ten fluid cells. The calculated and measured wall and vapor temperatures were the principle parameters for comparison. The precursory cooling rate above the quench front is found to be overpredicted by TRAC, resulting in overprediction of the rewetting velocity.

Analysis results from the Los Alamos 2D/3D program

Los Alamos National Laboratory is a participant in the 2D/3D program. Activities conducted at Los Alamos National Laboratory in support of 2D/3D program goals include analysis support of facility design, construction, and operation; provision of boundary and initial conditions for test facility operations based on analysis of pressurized water reactors; performance of pretest and posttest predictions and analyses; and use of experimental results to validate and assess the single- and multidimensional nonequilibrium features in the Transient Reactor Analysis Code (TRAC). During Fiscal Year 1986, Los Alamos conducted analytical assessment activities using data from the Cylindrical Core Test Facility and the Slab Core Test Facility. Los Alamos also continued to provide support analysis for the planning of Upper Plenum Test Facility experiments. Finally, Los Alamos either completed or is currently working on three areas of TRAC modeling improvement. In this paper, Los Alamos activities during Fiscal Year 1986 are summarized; several significant accomplishments are described in more detail to illustrate the work activities at Los Alamos.

An Assessment of TRAC-PD2 Refill Calculations Based on Creare Countercurrent Flow Tests

An important step in computer code development is the assessment of code capabilities through comparison of calculated results with experimental data A number of Creare countercurrent flow tests were simulated with the Transient Reactor Analysis Code (TRAC)-PD2 code to assess the emergency core coolant (ECC) lower plenum penetration and refill predictive capabilities. The tests examined in this study indicate a prediction of complete bypass and delivery at countercurrent steam flows where these phenomena occurred experimentally. Steam flows leading to partial delivery experimentally did not always lead to partial delivery in the calculations, however. A number of parameters can potentially effect TRAC refill calculations. Sensitivity studies indicate the TRAC results are most sensitive to droplet Weber number variations that affect interfacial shear and heat transfer rates. The condensation model also affects calculations with subcooled ECC liquid.

TRAC Analysis of Loss-of-Fluid Test Non-Nuclear Test L1-4

The Transient Reactor Analysis Code (TRAC) is being developed at the Los Alamos Scientific Laboratory as an advanced best-estimate computer program for analyses of postulated accidents in light water reactors. As a part of the TRAC verification effort, a posttest analysis of loss-of-fluid test (LOFT) non-nuclear test L1-4 has been conducted. The results of this analysis show that TRAC accurately predicts the thermal-hydraulic response of the entire LOFT system, including the delayed effects from emergency core coolant injection into the cold leg.