Band 81, Heft 4

A burn-up calculation of large systems by Monte-Carlo code (MCU) is complex process and it requires large computational costs. Previously prepared isotopic compositions are proposed to be used for the Monte-Carlo code calculations of different system states with burnt fuel. Isotopic compositions are calculated by an approximation method. The approximation method is based on usage of a spectral functionality and reference isotopic compositions, that are calculated by the engineering codes (TVS-M, BIPR-7A and PERMAK-A). The multiplication factors and power distributions of FAs from a 3-D reactor core are calculated in this work by the Monte-Carlo code MCU using earlier prepared isotopic compositions. The separate conditions of the burnt core are observed. The results of MCU calculations were compared with those that were obtained by engineering codes.

One of the goals of neutronic calculations of large cores may be self-consistent distribution of equilibrium xenon through the reactor core. In deterministic calculations such self consistency is relatively simply achieved with the help of additional outer iterations by xenon, which can increase several times solution run time. But in stochastic calculation of large cores such increase is utterly undesirable, since even without these outer iterations it demands modeling of billion of histories, which in case of complicated large core may take about a day of 100 processors work. In addition the unavoidable statistical uncertainty here plays role of transient process, which excites xenon oscillations. In this work the rise of such oscillations and the way of their overcoming with the help of hybrid stochastic/deterministic calculation is studied. It is proposed to make at first single static Monte Carlo calculation of given core and to receive multi-group mesh cell characteristics for future use in operative code. This one will evaluate xenon distribution through the core, which will be equilibrium for deterministic solution and substantially close to equilibrium Monte Carlo solution, paid with enormous computing cost.

The error detection in core loading in many cases happens in conditions of a significant asymmetry in the distribution of power density, which is caused by thermal mechanical deformations of reactor core, and temperature differences in the cold legs of coolant. The asymmetry of power distribution essentially decreases the effectiveness of algorithms used to detect errors in the core loading using ICIS data. The paper proposes the ways for solving this problem by means of special filtration algorithms.

Nowadays the installation of a new modernized VERONA core monitoring system (version V7.0) is in process at the NPP Paks. The most important steps of the current improvements are as follows: complete replacement of the hardware and the local area network; application of a new operating system and “virtual machine” (VM) technology; implementation of a new human-system interface; and last but not least, introduction of improved reactor physics calculations. Basic novelty of the modernized core analysis is the application of general purpose graphical processing units (GPGPU) in the on-line core-follow module. This new technology has allowed of performing the real-time node-wise core analysis by standard Paks NPP core design codes HELIOS/C-PORCA. The present paper gives a brief overview of the system version (V7.0), focusing to the models of reactor physics and results of validation. Main characteristics of new approaches of the modified on-line reactor physics calculations are also described.

At WWER-1000 NPPs, as well as at PWR NPPs, there is a problem of determining the correct weighted mean coolant temperature in the primary circuit hot legs based on the measuring channels information. The problem is caused by the coolant temperature stratification. The technical documentation for engineering support and maintenance of I&C systems does not provide any regulatory guidelines to consider this effect. Therefore, it is very important to represent a new methodology for determining the weighted mean coolant temperature in the primary circuit hot legs of the WWER-1000 reactor plants. The given paper presents the basic preconditions and approaches applied during the methodology development. They were worked out on the basis of the executed numerical and experimental research taking into account the analysis of the extensive material obtained by the authors from full-scale tests during the commissioning of WWER-1000 power units, as well as operational data obtained from several power units with different fuel loadings.

This work deals with TRAP-KS code verification. TRAP-KS is used for coupled neutron and thermo-hydraulic process calculations of VVER reactors. The three-dimensional neutron kinetics model enables consideration of space effects, which are produced by energy field and feedback parameters variations. This feature has to be investigated especially for asymmetrical multiplying variations of core properties, power fluctuations and strong local perturbation insertion. The presented work consists of three test definitions. First, an asymmetrical control rod (CR) ejection during power operation is defined. This process leads to fast reactivity insertion with short-time power spike. As second task xenon oscillations are considered. Here, small negative reactivity insertion leads to power decreasing and induces space oscillations of xenon concentration. In the late phase, these oscillations are suppressed by external actions. As last test, an international code comparison for a hypothetical main steam line break (V1000CT-2, task 2) was performed. This scenario is interesting for asymmetrical positive reactivity insertion by decreasing coolant temperature in the affected loop.

At present time, Ukraine faces the problem of small margins of acceptance criteria in connection with the implementation of a conservative approach for safety evaluations. The problem is particularly topical conducting feasibility analysis of power up-rating for Ukrainian nuclear power plants. Such situation requires the implementation of a best-estimate approach on the basis of an uncertainty analysis. For some kind of accidents, such as loss-of-coolant accident (LOCA), the best estimate approach is, more or less, developed and established. However, for reactivity initiated accident (RIA) analysis an application of best estimate method could be problematical. A regulatory document in Ukraine defines a nomenclature of neutronics calculations and so called “generic safety parameters” which should be used as boundary conditions for all VVER-1000 (V-320) reactors in RIA analysis. In this paper the ideas of uncertainty evaluations of generic safety parameters in RIA analysis in connection with the use of the 3D neutron kinetic code DYN3D and the GRS SUSA approach are presented.

The 7th AER dynamic benchmark is a continuation of the efforts to validate the codes systematically for the estimation of the transient behavior of VVER type nuclear power plants. The main part of the benchmark is the simulation of the re-connection of an isolated circulation loop with low temperature in a VVER-440 plant. This benchmark was calculated by the National Research Centre “Kurchatov Institute” (with the code ATHLET/BIPR-VVER), ÚJV Řež (with the code RELAP5-3D © ) and HZDR (with the code DYN3D/ATHLET). The paper gives an overview of the behavior of the main thermal hydraulic and neutron kinetic parameters in the provided solutions.

The best estimate simulation of three-dimensional phenomena in nuclear reactor cores requires the use of coupled neutron physics and thermal-hydraulics calculations. However these analyses should be supplemented by the survey of the corresponding uncertainties. In this paper the uncertainties of the coupled KIKO3D-ATHLET calculations are presented for a VVER-1000 type core using the OECD NEA Kalinin-3 (Phase II) benchmark data, although only the neutronic uncertainties are considered and further simplifications are applied and discussed. Additionally, this study has been performed in the conjunction with the OECD NEA UAM benchmark, as well. In the first part of the paper, the uncertainties of the effective multiplication factor, the assembly-wise radial power distribution, the axial power distribution, the rod worth, etc. are presented at steady-state. After that some uncertainties of the transient calculations are discussed for the considered switch-off of one Main Circulation Pump (MCP) type transient.

An uncertainty and sensitivity analysis is performed for the OECD/NEA coolant transient Benchmark (K-3) on measured data at Kalinin-3 Nuclear Power Plant (NPP). A switch off of one main coolant pump (MCP) at nominal reactor power is calculated using a coupled thermohydraulic and neutron-kinetic ATHLET-PARCS code. The objectives are to study uncertainty of total reactor power and to identify the main sources of reactor power uncertainty. The GRS uncertainty and sensitivity software package XSUSA is applied to propagate uncertainties in nuclear data libraries to the full core coupled transient calculations. A set of most important thermal-hydraulic parameters of the primary circuit is identified and a total of 23 thermohydraulic parameters are statistically varied using GRS code SUSA. The ATHLET model contains also a balance-of-plant (BOP) model which is simulated using ATHLET GCSM module. In particular the operation of the main steam generator regulators is modelled in detail. A set of 200 varied coupled ATHLET-PARCS calculations is analyzed. The results obtained show a clustering effect in the behavior of global reactor parameters. It is found that the GCSM system together with varied input parameters strongly influence the overall nuclear power plant behavior and can even lead to a new scenario. Possible reasons of the clustering effect are discussed in the paper. This work is a step forward in establishing a “best-estimate calculations in combination with performing uncertainty analysis” methodology for coupled full core calculations.

This article describes differences between two approaches of understanding the main goal of nuclear energy system: effective plutonium consumption versus effective natural uranium and thorium consumption. Several options of sodium cooled fast reactor are considered to provide increased breeding and thorium involving into nuclear energy system.

The basic strategy of Russian Nuclear Energy development is the formation of the closed fuel cycle based on fast breeder and thermal reactors, as well as the solution of problems of spent nuclear fuel accumulation and availability of resources. Three options of multiple Pu and U recycling in VVER reactors are considered in this work. Comparison of MOX and REMIX fuel recycling approaches for the closed fuel cycle involving thermal reactors is presented. REMIX fuel is supposed to be fabricated from non-separated mixture of uranium and plutonium obtained in spent fuel reprocessing with further makeup by enriched U. These options make it possible to recycle several times the total amount of Pu and U obtained from spent fuel. The main difference is the full or partial fuel loading of the core by assemblies with recycled Pu. The third option presents the concept of heterogeneous arrangement of fuel pins made of enriched uranium and MOX in one fuel assembly. It should be noted that fabrication of all fuel assemblies with Pu requires the use of expensive manufacturing technology. These three options of core loading can be balanced with respect to maximum Pu and U involvement in the fuel cycle. Various physical and economical aspects of Pu and U multiple recycling for selected options are considered in this work.

Experimental reactor ALLEGRO is a gas cooled fast reactor in the design stage. The current design of its reactivity control system is based on control rods filled with boron carbide as the absorber. Because of disadvantages connected to high boron enrichment a possibility of using other absorbent materials was explored to lower the boron enrichment and increase the worth of the control rods. The results of neutronic Monte-Carlo analyses in a computational supercell are presented in this paper. Three absorbent materials most suitable for a use in reactor ALLEGRO (B 4 C, EuB 6 and ReB 2 ) have been analysed also in a full core model. A possible benefit of a neutron trap concept is explored as well but materials with satisfactory neutronic properties proved to be not suitable for expected high temperatures in the reactor.

In the paper measurement results from the experimental modelling of a molten salt reactor concept will be presented along with detailed uncertainty analysis of the experimental system. Non-intrusive flow measurements are carried out on the scaled and segmented mock-up of a homogeneous, single region molten salt fast reactor concept. Uncertainty assessment of the particle image velocimetry (PIV) measurement system applied with the scaled and segmented model is presented in detail. The analysis covers the error sources of the measurement system (laser, recording camera, etc.) and the specific conditions (de-warping of measurement planes) originating in the geometry of the investigated domain. Effect of sample size in the ensemble averaged PIV measurements is discussed as well. An additional two-loop-operation mode is also presented and the analysis of the measurement results confirm that without enhancement nominal and other operation conditions will lead to strong unfavourable separation in the core flow. It implies that use of internal flow distribution structures will be necessary for the optimisation of the core coolant flow. Preliminary CFD calculations are presented to help the design of a perforated plate located above the inlet region. The purpose of the perforated plate is to reduce recirculation near the cylindrical wall and enhance the uniformity of the core flow distribution.