摘要:

AbstractA model for two-phase (water-vapor) flow in two-dimensional Cartesian or cylindrical coordinates is described that is implemented in the code DRIX-2D. The model includes slip between the phases and accounts for thermodynamic nonequilibrium. The code was designed as a“best-estimate”model for simulation of loss-of-coolant accidents (LOCAs) in light water reactor safety analysis. In this paper results of DRIX-2D applications are reported that can be used to assess the validity of simplified LOCA models.The main results are that both Cartesian and axisymmetric coordinates in two dimensions show considerable disadvantages as far as the pressure history in the downcomer is concerned. Yet, both models yield acceptable results concerning the gross blowdown behavior. Due to a 90-deg change in flow direction, considerable radial profiles in mass flow rate, velocity, and void fraction establish in the blowdown pipe. Nevertheless, a minor difference in the averaged mass flow rate exists only between a one- and two-dimensionally modeled blowdown pipe.A nonequilibrium state establishes at the pipe inlet in the case of subcooled vessel conditions and is maintained up to the orifice at least for pipe lengths<5m. However, the increase in mass flow rate caused by this nonequilibrium state is generally small enough for typical reactor conditions, so that an equilibrium assumption in the blowdown pipe should be appropriate for LOCA calculations.

ISSN:0029-5639    

DOI:10.13182/NSE80-A19443

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

AbstractSince it is desirable for power reactors to operate at steady power at a temperature of∼300°C, the main purpose of this study is to relate the dispersion and spheroidization of zirconium hydrides to operating procedures. Accordingly, we stress the hydride attack and supersaturation of hydrogen solubilities in the pre- and post-irradiated Zircaloys. Through this study we could find a way to improve the dispersed spheroidization process. The hydrogen diffusion coefficient of post-irradiated Zircaloy-4 with a neutron fluence of 1.64×1019n/cm2is 5 to 50% higher than that of the pre-irradiated Zircaloy-4. We considered there is a workable way to spheroidize hydrides with a temperature lower than the eutectoid temperature for irradiated Zircaloy, 547°C. Therefore, we propose to adapt the peritectoid reaction temperature, 255°C, to spheroidize zirconium hydrides. In the next section, we have studied the creep and corrosion behavior of annealed, hydrided, and spheroidized pre-irradiated Zircaloy-4 specimens following the proposed process. An annealed Zircaloy-4 specimen has the lowest minimum creep rate and the highest ductility and loading strain. A hydrided Zircaloy-4 specimen has the smallest loading strain and the lowest ductility. The spheroidized Zircaloy-4 specimen following the proposed process has a higher minimum creep rate than that of a hydrided one; however, the ductility of the specimen with sperhoidized hydrides is recovered to∼90% of the annealed one at 500°C The spheroidization treatment can improve the corrosion resistance of the hydrided specimen effectively in the temperature range of 200 to 400°C with the hydrogen concentration of the specimen up to 1000 ppm, although at 500°C the effect of spheroidization treatment on the hydride is decreased. We conclude that the proposed process with pre-irradiated Zircaloy and partially complete spheroidization can still improve the mechanical properties and corrosion behavior of the Zircaloy.

ISSN:0029-5639    

DOI:10.13182/NSE80-A19444

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

AbstractReaction rate and dose rate distributions due to streaming of neutrons and gamma rays were measured in the nearly cubical cavity and in the ducts leading to the cavity at the fast reactor YAYOI. The experimental configuration was arranged in a way that streaming through the ducts occurred only after scattering in the cavity. This was done by making the duct axes perpendicular to the source radiation from the reactor core. The spectrum of the source was modified from fast to thermal by putting moderators in the beam hole of the core assembly. The spatial distribution of the neutron reaction rates and gamma-ray dose rates in the cavity is almost flat except for a small decrease at the wall. The cadmium ratios at the duct inlets and the attenuation profiles in the ducts are almost identical irrespective of the spectrum of neutrons entering the cavity. The two-dimensional discrete ordinates transport calculation by TWOTRAN-II has shown that although the ray effect is observed in the fast neutron flux distribution in the cavity, the calculated results agree rather well with the experimental distributions in the cavity and with the streaming in the ducts with regard to the results measured with the l/v type neutron detectors. Also, for the gamma rays the agreement is rather good between the calculation and the experiment.

ISSN:0029-5639    

DOI:10.13182/NSE80-A19445

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

AbstractThe mass region around A = 90 was chosen for examination of the validity of nuclear models and input-parameter determination techniques often used to meet nuclear data requirements where no experimental data exist. Consistent sets of input parameters, determined through analysis of independent data available in this mass region, were applied to the calculation of all major neutron reactions on89Y and90Zr occurring between 0.05 and 20 MeV. These parameters were then tested under even more stringent conditions through calculation and comparison to experimental data on unstable target nuclei available for neutron energies of 14 to 15 MeV. These calculations, both on stable and unstable nuclei, serve to indicate that reliable cross-section predictions can be obtained from nuclear models that use carefully determined parameters verified in concurrent comparisons to available experimental data.

ISSN:0029-5639    

DOI:10.13182/NSE80-A19446

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

AbstractThe neutron capture and subthreshold fission cross section of241Am was measured in the energy range from 10 to 250 keV, using197Au and235U as the respective standards. Neutrons were produced via the7Li(p,n) and the T(p,n) reaction with the Karlsruhe 3-MV pulsed Van de Graaff accelerator. Capture events were detected by a Moxon-Rae detector and fission events by a NE-213 liquid scintillator with pulse-shape discriminator equipment. Flight paths as short as 50 to 66 mm were used to obtain optimum signal-to-background ratio. The capture cross section could be determined with a total statistical and systematic uncertainty of 4 to 10% while the respective values are 13 to 20% for the fission cross section. The results are compared with recent data of other authors, which in some cases are severely discrepant.

ISSN:0029-5639    

DOI:10.13182/NSE80-A19447

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

AbstractExperimental and analytical techniques have been developed for delayed neutron (DN) signal analysis and characterization that can provide diagnostic information to augment data from cover-gas analyses in the detection and identification of breached elements in a liquid-metal fast breeder reactor. Eleven flow reduction tests have been run in Experimental Breeder Reactor II to provide base data support for predicting DN signal characteristics during exposed fuel operation. Results from the tests demonstrate the feasibility and practicability of response-analysis techniques for determining (a) the transit time, Ttr, for DN emitters traveling from the core to the detector, and (b) the isotopic holdup time, Th, of DN precursors in the fuel element. The value Ttrvaries with the relative grid location of the DN source, and This affected by the form of fuel exposed to the coolant as well as the condition of the breach site. These parameters are incorporated into a mathematical formulism that enables one to compute for any exposed-fuel test an“equivalent recoil area.”This concept provides a basis for comparison of different run-beyond-cladding-breach tests in fast reactors.

ISSN:0029-5639    

DOI:10.13182/NSE80-A19448

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

AbstractMultigroup k andαeigenvalue problems for plane asymmetric cell lattices with superimposed small buckling are considered. Expressions for k andαeigenvalues and corresponding fundamental modes are given. A modified diffusion equation is introduced and some formulas for homogenized cross sections and diffusion coefficients are derived.

ISSN:0029-5639    

DOI:10.13182/NSE80-A19449

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

AbstractIn this paper we treat the problem of resonance absorption in isolated Breit-Wigner resonances of an absorber in an infinite homogeneous mixture of the absorber and moderator with an explicit treatment of the moderator collision integral. It is shown that Fourier transform techniques can profitably be used to treat this problem. However, the treatment calls for certain ideas from the theory of distributions similar to those used by Case in singular eigenfunction theory. The formulation leads to Fredholm integral equations in the transform variable whose solution gives the integral parameter of interest, namely, the effective resonance integral directly. In the limit of zero temperature, we obtain a second-order differential equation in the transform variable and formulate an accurate and fast converging iterative scheme to extract the resonance integral from its solution. Explicit formulas are derived for the resonance integral including the effect of resonance potential interference scattering. The analysis also provides an analytical expression for the asymptotic flux distribution well below the resonance energy. Numerical results are presented to demonstrate the accuracy of the method.

ISSN:0029-5639    

DOI:10.13182/NSE80-A19450

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

AbstractRecently, a new Monte Carlo method called the contributon Monte Carlo method was developed. The method is based on the theory of“contributons”and utilizes a new recipe for estimating target responses using a volume integral over the contributon current. The analog features of the new method were discussed in previous publications. We examine the application of some biasing methods to the new contributon scheme. A theoretical model is developed that enables an analytic prediction of the benefit to be expected when these biasing schemes are applied to both the contributon method and regular Monte Carlo. This model is verified by a variety of numerical experiments and is shown to yield satisfying results, especially for deep-penetration problems. Other considerations regarding the efficient use of the new method are also discussed and remarks are made as to the application of other biasing methods.

ISSN:0029-5639    

DOI:10.13182/NSE80-A19451

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor

摘要:

AbstractA nodal method for the solution of the multidimensional neutron diffusion equation is developed and evaluated. The method is based on the linear form of the nodal balance equation written in terms of the average partial currents across the surfaces of the node. Green’s functions for one-dimensional in-group diffusion-removal operators are used to generate a coupled set of one-dimensional integral equations defined over a subdomain or node. These integral equations represent an exact (local) solution to the coupled set of one-dimensional differential equations obtained by spatially integrating the multidimensional diffusion equation over directions transverse to each coordinate direction. The integral equations are approximated using a weighted residual procedure applied within each node. The resulting matrix equations, when solved in conjunction with the linear form of the nodal balance equation, provide the necessary additional relationships between the interface partial currents and the flux within the node.The nodal method is applied to several two- and three-dimensional light water reactor benchmark problems and to a four-group liquid-metal fast breeder reactor problem. These results demonstrate the capability of the method to yield very accurate steady-state and transient results in significantly smaller computing times than those required by standard finite difference methods.

ISSN:0029-5639    

DOI:10.13182/NSE80-A19452

出版商:Taylor&Francis    

年代:1980

数据来源: Taylor