Band 71, Heft 3

There is international consensus that for the various nuclear Advanced Reactor Concepts (ARCs), which are currently under development, design and operational safety goals as well as modern review frameworks for risk assessment required for licensing related decision-making shall be developed. The objective of this paper is to screen information available from different countries and international organisations on such efforts as far as they consider the use of Probabilistic Safety Assessment (PSA) or specific probabilistic evaluations. The screening of available information is done for many different types of ARCs (so-called Generation III, III+ and IV), including either active, passive or both types of systems. Main conclusion is that very little information is available for full probabilistic assessments, only recently some activities started on partial probabilistic evaluations, such as on the reliability of certain types of passive systems. Therefore, published statements on significantly lower risk levels achieved by certain ARCs are at least not verifiable.

Modelling of pressure surges in power plant pipe systems means to carefully judge and classify the relevant variables. A decision must be made, whether the numerous effects that can occur during a transient event, should be modelled. One main point is to determine if fluid structure interaction (FSI) is important for the special situation/scenario. Also unsteady friction and degassing of dissolved gases can occur. But with each model more sources of errors are introduced into the mathematical system, such as uncertainty about bearing clearance, determination of fixing points/stiffness/spring constants, determination of friction coefficients and damping parameters, uncertainty about the amount of non-condensables in the liquid, dynamics of de-aeration processes and unsteady friction coefficients. Another complicated problem is the definition of boundaries, like specifications of pumps, valves, fittings, etc. It is obvious that a combination of all uncertainties may lead to quite uncertain predictions and must therefore be very carefully taken into account. Within this contribution, valve-closing experiments of different initial temperature and system pressure at Fraunhofer test rig PPP are presented. Two scenarios are modelled using different assumptions and model depths and the simulation results are compared with experimental data.

For the investigation of air/water slug flow, a horizontal channel with rectangular cross-section was built at Forschungszentrum Rossendorf. The channel allows the investigation of air/water co- and counter-current flows at atmospheric pressure, especially the slug behaviour. Optical measurements were performed with a high-speed video camera, and were complemented by simultaneous dynamic pressure measurements. Moreover velocity-fields were measured using Particle Image Velocimetry (PIV). A CFD simulation of the stratified co-current flow was performed using the code CFX-5, applying the Euler-Euler two fluid model with the free surface option. The grid contains 410 5 control volumes. The turbulence was modelled separately for each phase using the k- based shear stress transport (SST) turbulence model. To achieve wave generation in such a short channel, the inlet water level had to be varied in time. For this purpose, the water level history was taken from a recorded image sequence and set as time-dependent boundary condition at the model inlet. The results show a wave formation up to slug development with closure of the whole channel cross-section and consequently an increase of the pressure level behind the slug. Despite unsteady conditions at the inlet of the test channel and simplified initial conditions in the model, the slug simulation with CFX is in good qualitative agreement with the experiment, while the slug length increases during its progression, witch was not observed in reality.

The best-estimate accident analysis approach is widely used for studying safety issues. The introduction of nominal values implies the evaluation of solution uncertainty, therefore an uncertainty analysis is required. New results of an uncertainty and sensitivity analysis of the Control and Protection System Cooling Circuit (CPS-CC) voiding accident in an RBMK-1500 reactor core are presented here. A typical reactor core state of Ignalina NPP Unit2 before the implementation of the independent diverse shut-down system was chosen for the study. The analyzed reactivity initiated accident is one of the most severe ones in RBMK-1500 reactors causing a positive reactivity insertion of about 2 in the case of a full CPS-CC voiding. The simulation of the CPS-CC dynamic voiding process was conducted without taking into account an additional failure of the reactor equipment. The analysis was performed by employing the GRS uncertainty and sensitivity method based on the SUSA software package in combination with the 3D core model QUABOX/CUBBOX-HYCA, adjusted for reactors of RBMK type. The results of the analysis give indications for which parameters the uncertainty of knowledge should be reduced in order to reduce the uncertainty of results most effectively.

The irradiation by a high intensity proton beam of the spallation target in Accelerator Driven Systems (ADS) gives rise to structural damages that must be considered in the design and handling of such materials. In this work, some of these damages are evaluated in the case of the irradiation of a tantalum target. Tantalum was recently proposed as solid target in the case of low power systems. Unfortunately, not enough experimental data of spallation reactions in this material are available. Therefore, our calculations were made using MCNPX and the stand-alone version of the spallation codes INCL4+ABLA. Results of this work are the concentrations of spallation residues produced in the target after one year of irradiation and calculations of the activity derived. Also displacements per atom in the target material have been evaluated. Different components (elastic scattering and non-elastic reactions) of the displacement cross-sections were calculated with two different methods.

In this article, multigroup discrete ordinates interface conditions for neutron reflection and leakage were derived for axial models of nuclear reactors with anisotropic scattering. These conditions are completely free from spatial truncation error. The key to the derivation is a suitable reformulation of exact discretized equations – the multigroup spectral nodal equations – defined in the nonactive boundary regions at top and bottom. The resulting interface conditions can be used to replace the nonactive boundary regions of a nuclear reactor core in axial reactor computations. In order to illustrate the improvements that can be experienced by most one-dimensional multigroup discrete ordinates codes with our multigroup interface conditions, we perform a numerical experiment with a representative multigroup code and an axial core model with anisotropic scattering.

An analysis of the performance of the Hot Water Layer System, HWLS, and the Interconnection System, IS, at the ETRR-2 research reactor is presented. The behavior of the HWLS during the formation period of the hot water layer has been studied and is presented together with the pool-top measured radiation level associated with that behavior. Two different designs of the HWLS have been experimentally evaluated. For these two different designs, the radiation level at the ETRR-2 pool-top were characterized and discussed. The effect of adding a water purification system to the HWLS is demonstrated as well as the effect of the IS on the pool-top radiation level considering different operational conditions of the HWLS. The study shows the importance of these systems from the radiological protection point of view.

The incorporation monitoring during decommissioning of nuclear reactors is faced with a complex situation arising from a mixture of radionuclides. The development of an interpretation procedure to enable the application of an optimised incorporation monitoring on conditions of decommissioning of nuclear reactors has been carried out combining results from radionuclide analyses of contaminated or activated materials and the proven methods of incorporation monitoring in the case of the intake of single radionuclides. In this study the radiological requirements, the mathematical elements of the interpretation procedure are presented as well as the application to an activated concrete of biological shielding.

The thermo-hydraulics of the fixed bed nuclear reactor (FBNR), possessing spherical HTGR type fuel, is described. Its main characteristic is the variable fixed bed core in suspense and its “cold” fuel in case of a LOCA (< 400°C). A 40 MW el reactor has a pressure loss across the fixed bed of 9.5 bar and needs a pump power of 3.4 MW el – more than in a classical PWR but being the price for its excellent inherent safety characteristics.

The H N method is used to calculate the critical slab thickness for a parallel media surrounded bay a reflector for linearly anisotropic scattering. It is shown that the method leads to concise equations and accurate numerical results even in the lowest order of approximation.