Search

A study on the nuclear characteristics of an axially decoupled core was performed through a series of experiments using the solid moderator core in the Kyoto University Critical Assembly (KUCA). In the axially decoupled core which has an internal blanket between the upper and lower cores, the measured differential reactivity worth of a control rod moving axially was apparently asymmetric between the two core regions. Since it was considered that this phenomenon was attributed to the neutron flux tilt caused by the axial movement of control rods during the measurement of differential worth, the axial reaction rate distributions were measured by gold wires with and without a cadmium tube. It was found that the reaction rate distribution was remarkably tilted by changing the pattern of control rods inserted in the core to maintain the critical state, and that the flux distribution was tilted without energy dependency. The value of first-order eigenvalue separation was obtained using the relation between the neutron flux tilt and the first-mode eigenvalue separation based on the Explicit Higher-order Perturbation (EHP) method. This value agreed with the calculated result by the SRAC system.

第5回臨界安全性国際会議ICNC'95が1995年9月に米国アルバカーキにて開催された。参加者は17ヶ国から約300名,発表は約150件あった。今回の会議では,これまではよく知られていなかった旧ソ連の臨界実験施設,臨界安全研究のほか,臨界事故について初めて報告された。そのほか,燃焼度クレジット,動特性解析などで地道な研究の進歩が見られた。本稿では,このようなICNC'95での発表を通じて臨界安全性研究の現状を解説する。

A theoretical expression of reactivity interaction between control rods is developed on the basis of the explicit higher-order perturbation method. The expression shows that the magnitude of the interaction is inversely proportional to the λ-mode eigenvalue separation and dependence on the rod patterns is governed by the higher-harmonic eigenfunctions and the adjoint ones. Application of the formula is illustrated by numerical calculations carried out for a 1-D model of a large fast reactor core. The results indicate that the first harmonics dominantly contribute to the interaction, consequently the first-harmonic eigenvalue separation can be employed as a quantitative indication of interaction effect.

The general expression of noise coherence function including the energy dependence of neutron detection cross section and the higher harmonic contributions in multi-dimensional model is developed on the basis of the modal expansion technique. Applicability of the method is demonstrated by numerical calculations carried out for a two-dimensional model of large fast reactor assemblies ZPPR-9 and −13C. The agreement between the theory and the measurement is satisfactory, which indicates the validity of the theory and the calculational model employed. In the assemblies, the coherence for a detector pair in a specific location can be approximately described by including the fundamental and the first harmonic modes.

The improved power method in the higher-harmonic eigenvalue calculation of nuclear plant thermal-hydraulic dynamics was applied to the λ- and ωp-eigenvalue problems of thermal and fast reactor systems to obtain the two-dimensional, few-energy-group higher-harmonic eigenvalues and eigenfunctions. As a result, the efficiency of the method was confirmed in the large-scale neutron diffusion problems consisting of the variables amounting to tens of thousands, and the interesting spatial patterns of the eigenfunction were obtained.Even though a harmonic order may be lower, the discrepancy between λ- and ωp-eigen-functions, which are usually assumed to be approximately identical, is significant in a thermal system with low-absorbing region (e.g. reflector). In the higher ωp-harmonics of the system, zero loci of the eigenfunctions depend on neutron energy groups. Therefore, the particular attention should be paid to placements and energy-response of detectors in the kinetic experiments.

For the analysis of noise problems in zero-power coupled-core reactors, one-energy group multipoint Pal-Bell equations are derived and their applications are illustrated. The illustration is made for the neutron correlation experiments of loosely-coupled two-core systems. We first derive a two-point version of the Pal-Bell equations for the neutron detection probability generating functions, and then derive the formulae of Feynman-alpha and Rossi-alpha (Orndoff type) experiments for a coupled-core system, using the factorial moment and cumulant expansions. With these formulae the following coupled-core kinetic parameters can be evaluated from neutron correlation experiments: coupling delay time, coupling coefficient and neutron generation time. The simplicity in the present derivation of Pal-Bell equations is pointed out.

The mathematical equivalence of two solutions, both expressing the pulse decay in a two-point reactor, is illustrated. The first is a series superposing time eigenfunctions of two-point kinetic equations, while the second is a series summing the effects of groups of neutrons, each group having undergone a specified number of trips between the two cores after the initial pulse. The two expressions are alternative representations of a single solution. Taking a symmetric coupled-core system, numerical examples are given, and the relevant time range for the truncated solution of each is examined. The possibility of damped oscillation is investigated using the second solution. The concept of equivalence applies to the propagation of a neutronic disturbance within a large reactor core, provided the nodal model is feasible.

With the aid of residue evaluation, the timedependent reflection response G /SUB gr/ (t) of Shinkawa et al., which was recently used in a coupled-core stability analysis, is verified. The expansion series of the G /SUB gr/ (t) is summed with the Watson method, and its consistency with neutron conservation is checked. The result justifies the previous addition of -delta(t) to a reflection term, which was made in the stability analysis. The meaning of the response flux eigenfunction expansion for a moderator region is discussed.

A two-group, one-dimensional formulation of a coupled-core system is proposed as a revision of the one-group response function method by Shinkawa et al. The coupling coefficient of the Kyoto University Critical Assembly symmetric coupled-core loading is revised.

For single-core reflected neutronic systems, generalized neutron generation time is derived and given physical interpretations in terms of importance. A system kinetic equation containing the moderator region response function previously introduced is reduced by a slow-variation approximation to the form of a conventional one-point kinetic equation, in which a parameter can be identified as generalized neutron generation time by analogy with a bare system. In such a mathematical expression for the parameter, one can further identify the amount of increase due to reflection over the bare system generation time. This amount is found to be the reflection time multiplied by the number of migrations that neutrons undergo between reflector and core in one generation. The theoretical generation time of the SHE assembly, a thermal-energy, graphite-moderated critical assembly, calculated by such a formation with cylindrical geometry, agreed well with that from pulsed neutron experiments.

A number of workers in Japan have undertaken neutronic kinetic experiments in reactor physics during the past decade, and the theoretical background for these experiments has been established. Many of the important reports related to these studies have appeared in the Journal of Nuclear Science & Technology.The present survey briefly reviews the work covered by outstanding papers published in this Journal, as well as some particularly noteworthy reports that have appeared in foreign journals in the same period.Pulsed neutron experiments in multiplying and non-multiplying systems are discussed separately for thermal reactor physics and fast reactor physics. Statistical correlation methods are covered in such forms as Rossi-α, Feynman-α, zero-count probability, polarity correlation and reactivity determination problems. Progress in neutron wave propagation studies is reviewed in respect of both theoretical and experimental aspects.

A stability analysis using a one-group model is presented for a coupled-core system. Positive prompt feedback of a ..gamma..p /SUB j/ form is assumed, where p /SUB j/ is the fractional power variation of core j. Prompt power variations over a range of a few milliseconds after a disturbance are analyzed. The analysis combines Lapunov's method, prompt jump approximation, and the eigenfunction expansion of coupling region response flux. The last is treated as a pseudo-delayed neutron precursor. An asymptotic stability region is found for p /SUB j/. For an asymmetric flux variation over a system of two coupled cores, either p /SUB I/ or p /SUB II/ can slightly exceed, by virtue of the coupling effect, the critical value (..beta../..gamma..-1) of a single-core case. Such a stability region is increased by additional inclusion of the coupling region fundamental mode in the treatment. The coupling region contributes to stability through its delayed response and coupling. An optimum core separation distance for stability is found.

A method is proposed to derive three kinetic parameters of a zero-power coupled-core system, namely coupling reactivity Δ, mean neutron transit time , and neutron generation time Λ of respective co...

One-dimensional, one-energy-group diffusion theory is applied to a coupled-core slab reactor to derive kinetic equations for the system, with different modes of formulation taken for moderator regions and for core regions. For the former, the diffusion equation is exactly solved to obtain the time-dependent neutron currents from moderator to core (the moderator response function) in response to the neutron incident current in the form of a unit impulse on the boundary. For the core regions, the neutron flux psi(x,t) is written as a product of a shape function, phi(x,t), and a time function, P(t), as suggested by Henry, with P(t) chosen to represent the time variation of total importance over the respective core. The boundary terms that arise in the equations for P(t) are combined with incoming neutron currents at the boundaries, which in turn are expressed in terms of the moderator region reponse functions above. The equations for P(t) derived by such procedures include the coupling effect between the two cores, without a need for the conventional, a priori assumption of coupling coefficient. For the Argonaut two-slab core, the transfer functions are obtained and compared with existing values. The value of the conventional coupling coefficient is also inferred bymore » reducing the present form of coupling terms by approximation. From the approximation needed in the procedure, the limitation of the coupling coefficient approach is discussed.« less

By cell calculation with the SRAC code system, void reactivity is evaluated for a high conversion light water reactor tight lattice, with an emphasis on the breakdown of the void effect into compon...