摘要:

AbstractDuring a severe nuclear plant accident, molten fuel can contact water in the core region, the lower plenum, or in the cavity below the reactor vessel. The interactions that take place can vary from benign boiling to explosive vaporization. If the fuel contains a metallic component, rapid oxidation of the metal can occur during the fuel/coolant interaction (FCI). The hydrogen generated from this reaction can increase the threat to containment integrity.Experiments have been conducted with 10 to 20 kg of two kinds of thermite-generated molten fuel simulants: corium and iron-aluminum oxide. Both saturated and subcooled water were employed as coolants. Explosive and nonexplosive FCIs were observed. Up to 30% of the metal was reacted in some cases. For some of the tests, the extent of reaction appeared to depend more on the water subcooling than on the degree of fragmentation as measured by posttest sieving. Models of hydrogen generation are proposed and compared to a broad range of experiments. Predictions agree qualitatively with many of the experimental observations. A more accurate model of hydrogen generation would have to be coupled to a dynamic thermal-hydraulic calculation of the important phases of an FCI: coarse mixing, fine fragmentation, explosion propagation, and vapor expansion.

ISSN:0029-5639    

DOI:10.13182/NSE88-A23521

出版商:Taylor&Francis    

年代:1988

数据来源: Taylor

摘要:

AbstractA model for small-scale single droplet fuel/coolant interactions (FCIs) is proposed, which considers the growth of a coolant vapor/liquid interfacial disturbance into a coolant liquid jet during the collapse of the vapor film surrounding the fuel. This results in the encapsulation of the jet as coolant drops beneath the fuel surface and leads to fragmentation of the fuel. In this model, the FCI process is divided into four stages: film boiling around a molten fuel droplet in an infinite coolant pool, film collapse and coolant jet formation, coolant jet penetration and entrapment in the fuel, and rapid evaporation of entrained coolant and fragmentation of the fuel. The process repeats itself cyclically from the second stage. For the single-droplet experiments performed previously, the model predicts the qualitative trends of steam bubble growth and collapse, the final size of fuel fragments, and time scale for the fuel fragmentation.

ISSN:0029-5639    

DOI:10.13182/NSE88-A23522

出版商:Taylor&Francis    

年代:1988

数据来源: Taylor

摘要:

AbstractVery high computational efficiencies have been achieved recently by introducing higher order approximations to nodal formalisms for the discrete ordinates, neutron transport equation. However, the difficulty of the nodal formalism, its final discrete variable equations, and the solution algorithms have limited the usefulness and applicability of nodal methods in spite of their extremely high accuracy.A general order, general dimensionality nodal transport method cast in a simple, compact, singleweight, weighted diamond-difference form is derived. The new form is a consistently formulated nodal method, which can be solved using either the discrete nodal-transport method or the nodal-equivalent finite difference algorithms without any approximations. The final discrete variable equations for the two-dimensional case are implemented in a computer code to solve monoenergetic, isotropic scattering, external source problems to any given order, i.e., C-C, L-L, Q-Q, etc. A simple test problem with large homogeneous regions is solved using this code, on meshes ranging from 2×2 to 128×128, and orders ranging from zero to nine. The results show that, for this problem, the CPU time and the storage size required to achieve a given accuracy decrease monotonically up to order five. Hence, very high order methods may be more computationally efficient in solving practical problems with large homogeneous regions.

ISSN:0029-5639    

DOI:10.13182/NSE88-6

出版商:Taylor&Francis    

年代:1988

数据来源: Taylor

摘要:

AbstractA new neutron transport method, called discrete elements (LN), is derived and compared to discrete ordinates methods, theoretically and by numerical experimentation.The discrete elements method is based on discretizing the Boltzmann equation over a set of elements of angle. The zeroth and first angular moments of the directional flux, over each element, are estimated by numerical quadrature and yield a flux-weighted average streaming direction for the element. Data for this estimation are fluxes infixed directions calculated as in SN. The spatial quadrature then propagates the element flux in this“steered”direction, strongly ameliorating ray effect. The discrete elements method is shown to be more cost-effective than discrete ordinates, in terms of accuracy versus execution time and storage, for the cases tested. In a two-dimensional test case, a vacuum duct in a shield, the LNmethod is more consistently convergent toward a Monte Carlo benchmark solution.

ISSN:0029-5639    

DOI:10.13182/NSE98-01-41

出版商:Taylor&Francis    

年代:1988

数据来源: Taylor

摘要:

AbstractA new direction biasing approach to a target point and to finite detectors for Monte Carlo simulation is developed, presented, and tested. It properly accounts for the weight adjustments that must be made for the combined choice of a particular scattering (polar) and rotational (azimuthal) angle to obtain a given biasing angle about either a target point or a finite detector. Sample Monte Carlo simulations for a neutron transport problem with isotropic center-of-mass scattering and a gamma-ray transport problem with Klein-Nishina scattering have been done by both the analog and new direction biasing methods. The results indicate that the direction biasing approach is valid and will be very efficient for deep-penetration problems of these two types.

ISSN:0029-5639    

DOI:10.13182/NSE88-A23525

出版商:Taylor&Francis    

年代:1988

数据来源: Taylor

摘要:

AbstractA reliability analysis methodology, GO-FLOW, is presented. Detailed explanations and two examples of GO-FLOW analysis are given.The GO-FLOW is a success-oriented system analysis technique. The modeling technique produces the GO-FLOW chart, which is composed of operators and signal lines and represents a function of the system. A signal does not represent a“change of condition”but some physical quantity or information. The intensity of a signal represents the probability of actual or potential existence of a physical quantity, the probability that some information exists, or a time interval between two successive time points.The examples of analysis show the applicability of the GO-FLOW method to a phased mission problem (a boiling water reactor emergency core cooling system) and to a time-dependent unavailability analysis (a pressurized water reactor auxiliary feedwater system).The GO-FLOW has proved to be a valuable and useful tool for system reliability analysis.

ISSN:0029-5639    

DOI:10.13182/NSE88-A23526

出版商:Taylor&Francis    

年代:1988

数据来源: Taylor

摘要:

AbstractA Bayesian analysis is made of the probability of core melt (defined as“core on the floor”) for U.S. reactors, using methods that are an extension of those used in an earlier work, and that can therefore be used to answer different questions. The essential new point is that at any time the estimate of the probability of core melt varies continuously as melt-free experience accumulates, with a substantial effect on the estimates. The most interesting result is that the probability of no-melt-yet stays large somewhat longer than one might have guessed, largely because melt-free experience feeds on itself in reducing the probability of subsequent core melt, and this process continues until the first melt.

ISSN:0029-5639    

DOI:10.13182/NSE88-A23527

出版商:Taylor&Francis    

年代:1988

数据来源: Taylor

摘要:

AbstractA new uncertainty analysis method based on the generalized bias operator method is presented. The new method can be used for those cases in which the other uncertainty analysis methods cannot be applied, due to the lack of or the incomplete knowledge about the input parameter's uncertainties, or due to an inadequate model. The method presented can be applied in those cases in which at least one of the system's responses is measured and the accuracy of the measurements is better than the deviation between the measured and the calculated value obtained. The theory of the method is presented, as well as methods to obtain the special sensitivity matrices associated with the method.

ISSN:0029-5639    

DOI:10.13182/NSE88-A23528

出版商:Taylor&Francis    

年代:1988

数据来源: Taylor