Band 73, Heft 5-6

Pressure fluctuations are generated in pipeline systems, when a fluid is rapidly accelerated or decelerated by fast closing or opening a valve, pump trip or start, breaking of pipes etc. TUEV NORD SysTec GmbH & Co. KG performs calculations of pressure surges in power plants and especially in nuclear power plants for many years. For this reason TUEV NORD has developed and qualified the pressure surge computer code DYVRO. This contribution describes in detail the actual code version DYVRO mod. 3 that has been modified with regard to the system of partial differential equations and the numerical scheme. Also the validation against representative experiments like the Simpson's experiment and an experiment at the Cold Water Hammer Test Facility CWHTF is shown. Results of DYVRO calculations are afterwards compared with results of own and external simulations with system codes (ATHLET and RELAP5) and with the pressure surge code WAHA. For this purpose the outcome of the EU project WAHALOADS that was supported within the 5 th framework programme has been used.

A computer code “TURDYN” has been developed for prediction of high pressure and low pressure turbine torque under thermodynamic transient conditions. The model is based on the conservation laws of mass and energy. All the important components of turbine systems, e. g. high pressure turbine, low pressure turbine, feed heaters, reheater, moisture separator have been considered. The dynamic equations are solved simultaneously to obtain the stage pressure at various load conditions. The details of the mathematical formulation of the model and open loop responses for specific disturbances are presented.

In this paper, we present the steady state and stability behaviour of a boiling two-phase natural circulation loop with a small concentration of Al 2 O 3 nanofluids. For this, experiments were conducted in a natural circulation loop at different operating powers, pressures, and water levels in the steam drum. The test results revealed that while the steady state natural circulation behaviour is similar, however, the buoyancy induced flow rates are found to be relatively higher with nanofluids than with water alone. In addition, with nanofluids, the amplitudes of the boiling induced Type I instabilities are found to be significantly reduced, However, the Type II instabilities are found to appear at relatively lower power with addition of nanofluids as compared to that with water alone. The most interesting result is that with addition of nanofluids, the amplitudes of Type II instabilities are found to get suppressed with increase in power similar to that of low quality Type I instability.

So far, only individual absorbing materials have been studied for neutron absorption in the pool-type reactor spent fuel storages. In this paper, the effect of combining various materials, which are already known as good neutron absorbers by themselves, on lowering the multiplication factor of spent fuel pool, is described. The results of Monte Carlo computer code MCNP calculations for the spent fuel pool of the TRIGA research reactor in Ljubljana, Slovenia, are specified. Due to property completion of the chosen absorber materials, some of their combinations partially decrease the multiplication factor and thus contribute to the enhancement of the safety conditions of the pool.

Several series of experiments with a “FDR” reactor (advanced pressurized light water reactor) were performed in 1972 in the Geesthacht critical facility ANEX. The experiments were performed to test the core prior to its usage for the propulsion of the first German nuclear merchant ship “Otto-Hahn”. In the present paper a calculational re-evaluation of the experiments is described with the use of the up-to date computer codes (Monte-Carlo code MCNP5) and nuclear data (ENDF/B-VI release 6). It is focused on the determination of uncertainties in the benchmark model of the experimental set-up, originating mainly from the limited set of information still available about the experiments. Effects of the identified uncertainties on the multiplication factor were studied. The sensitivity studies include parametric variation of material composition and geometry. The combined total uncertainty being found 0.0050 in k eff , the experiments are qualified as criticality safety benchmark experiments.

Renewal of nuclear power programs in countries with modest electricity consumptions and weak electrical grid interconnections has raised the question of optimal nuclear power plants sizes for such countries. The same question would be also valid for isolated or weakly connected regions within a large country. Building large size nuclear power plant could be prevented by technical or financial limits. Research programs have been initiated in the International Atomic Energy Agency and in the USA (within the framework of the Global Nuclear Energy Partnership (GNEP) program) with the aim to inspect under which circumstances small and medium reactors could be the preferred option compared to large nuclear plants. The economy of scale is a clear advantage of large plants. This paper compares, by using probabilistic methods, the net cash flow of large and medium size plants, taking as example a large nuclear plant (around 1200 MW) and four sequentially built smaller plants (300 MW). Potential advantages and disadvantageous of both options have been considered. Main advantages of the sequential construction of several identical small units could be the reduced investor risk and reduced investment costs due to the learning effect. This analysis is a part of studies for the Croatian power generating system development.

Interim storage is one major program of fuel cycle. Because there are some individuals staying or living very close to the ISFSI site that is being constructed, this paper examines the potential radiological consequences of hypothesized events for storage casks. This study uses ORIGEN, HOTSPOT and RESRAD software to calculate the source terms, acute dose and long-term dose, separately. The results show that explosive-induced buoyancy would disperse the source much farther than the site boundary in a sabotage accident. Nevertheless, the amount of radioactivity released to offsite would be very low, and the maximum dose received by any individual staying or living in the vicinity of ISFSI, even under the most conservative scenario, would be very much less than the local public relocation dose limit.

Ultraspherical polynomial P (λ) n approximations are used in reflected slab criticality calculations for strongly anisotropic scattering. A unique formulation is developed for reflected slab criticality conditions whose special cases are spherical harmonics approximations, Chebyshev polynomial approximations of first and second kind. It is shown in the calculations that all ultraspherical polynomial approximations are equivalent regarding the convergence of the solution of one-speed neutron transport equation for various degrees of anisotropy and cross section parameters. Although some ultraspherical polynomials approximations converge in the mean as the order of approximation is increased and some converge uniformly, they all converge to the same critical thickness which implies their equi-convergence. It is also demonstrated that the first order approximation P (λ) 1 would give the exact critical thickness for a certain value of the parameter λ.

The ultraspherical polynomial approximation which unifies all classical polynomial sequences in a unique form is used to calculate the albedo for isotropic scattering in a homogeneous spherical medium. This is the most general polynomial approach in the sense that it includes all classical polynomial methods to solve the transport equation such as and methods. For the first time an antisymmetric polynomial (ultraspherical polynomial) solution to the corresponding pseudo-slab problem is proposed. Very accurate and consistent albedo values are obtained for a variety of methods when compared to the literature. It is also shown that various approximations differ only in convergency characteristics; some converge monotonically, some in the mean.

In this work, the F N method with a different numerical approximation is used to solve both, half-space and slab albedo problems for a linearly anisotropic scattering kernel. The numerical results are compared with data available in the literature and the accuracy of the method is verified.

The effect of anisotropic scattering on the variation of the critical thickness for one-speed neutrons in a uniform finite slab has been studied using Marshak boundary conditions. Chebyshev polynomials of the second kind are used in the angular part of the neutron flux. The critical slab thicknesses are calculated for various values of collision and anisotropy parameters. Numerical results obtained by the present method for the critical thickness are presented in the tables together with the results obtained by Legendre polynomials approximation and the ones available in the literature for comparison.