04.09.2017 Aufrufe

atw 2017-09

Sie wollen auch ein ePaper? Erhöhen Sie die Reichweite Ihrer Titel.

YUMPU macht aus Druck-PDFs automatisch weboptimierte ePaper, die Google liebt.

<strong>atw</strong> Vol. 62 (<strong>2017</strong>) | Issue 8/9 ı August/September<br />

ENVIRONMENT AND SAFETY 532<br />

of the project can be summarized as<br />

follows:<br />

1) Development and validation of<br />

­advanced numerical approaches<br />

for the design and safety evaluation<br />

of advanced reactors.<br />

2) Achievement of a new or extended<br />

validation base by creation of new<br />

reference data.<br />

3) Establishment of Best Practice<br />

Guidelines, Verification & Validation<br />

methodologies, and Uncertainty<br />

Quantification methods.<br />

The project consortium consists of<br />

23 European partners as depicted in<br />

Figure 2. Early 2016, an international<br />

collaboration with the US Nuclear<br />

Energy Advanced Modelling and<br />

Simulation program (NEAMS) has<br />

been initiated allowing to combine<br />

thermal hydraulic forces on both sides<br />

of the Atlantic. The project ensures<br />

relevance and contact with the system<br />

designers through the establishment<br />

of a Senior Advisory Committee in<br />

which the main European liquid<br />

metal cooled reactor designers are<br />

represented.<br />

3 Temperature fluctuations<br />

A fundamental issue for the evaluation<br />

of liquid metal reactors is the modelling<br />

of the turbulent heat transfer over<br />

the complete range, from ­natural to<br />

mixed and to forced con­vection<br />

regimes. Current engineering tools<br />

­apply statistical turbulence ­closures<br />

and adopt the concept of the turbulent<br />

Prandtl number based on the Reynolds<br />

analogy. This analogy is valid mainly<br />

for forced convective flows with a<br />

Prandtl number of order of unity. In<br />

the particular case of liquid metal,<br />

where the Prandtl number is much<br />

smaller than 1, the turbulent Prandtl<br />

Experiment<br />

High-<br />

Fidelity<br />

Reference<br />

Simulation<br />

Flow<br />

Separation<br />

Shear / Jet<br />

number concept is not applicable, and<br />

robust engineering turbulence models<br />

are needed. Thus, a model is required<br />

which can deal with all flow regimes<br />

simultaneously in ­liquid metal flows.<br />

Roelofs et al. [2015] identified some<br />

promising routes for improvement<br />

which are tested on ­relevant available<br />

geometrically ­simple test cases. The<br />

main focus in Europe is now on the<br />

­extension of the validation base for<br />

mixed and ­natural convection regimes<br />

and for ­geometrically complex cases.<br />

To this purpose, the validation base<br />

will be extended by including mixing<br />

jets, flow separation, mixed convection,<br />

and rod bundle test cases as summarized<br />

in Table 1. The preparation<br />

of the experiments and of the high<br />

­fidelity Direct Numerical Simulations<br />

(DNS) are in progress and new<br />

­reference data sets produced by these<br />

­efforts are currently being produced<br />

and will become available in the<br />

­second half of <strong>2017</strong> and 2018. ­Figure 3<br />

shows a 3D view of the ­design of<br />

a backward facing step test section<br />

Mixed<br />

Convection<br />

| | Tab. 1.<br />

Summary of SESAME reference data for turbulence model assessment to be generated.<br />

Rod Bundle<br />

to be installed in the German KASOLA<br />

­sodium facility.<br />

In addition, a pragmatic simulation<br />

model which performs well in all<br />

flow regimes simultaneously would<br />

provide much added value. Further<br />

assessment and development of the<br />

promising approaches is foreseen.<br />

Emphasis is put on the creation of<br />

missing relevant reference data, on<br />

more robust testing of the models and<br />

on testing the models in geometrically<br />

more complex cases. Some of the<br />

­selected approaches will provide a<br />

more pragmatic computational alternative<br />

to the high fidelity but very<br />

­expensive LES. As such LES is practically<br />

not possible for the prediction of<br />

thermal fluctuations in a wide range<br />

of flows, as is required in most cases of<br />

industrial interest. This work will<br />

therefore probably also have a positive<br />

impact also on the simulation of<br />

light water reactors. The following<br />

promising models identified in Roelofs<br />

et al. [2015] will be further developed<br />

and/or implemented and validated:<br />

| | Fig. 3.<br />

KASOLA facility at KIT (left) and it’s backward facing step test section (right).<br />

Environment and Safety<br />

The SESAME Project: State of the Art Liquid Metal Thermal Hydraulics and Beyond ı F. Roelofs, A. Shams, A. Batta, V. Moreau, I. Di Piazza, A. Gerschenfeld, P. Planquart and M. Tarantino

Hurra! Ihre Datei wurde hochgeladen und ist bereit für die Veröffentlichung.

Erfolgreich gespeichert!

Leider ist etwas schief gelaufen!