3. sbep-v2 fh-ict balloon test - HySafe

SIXTH FRAMEWORK PROGRAMME
Sustainable Energy Systems NETWORK OF EXCELLENCE [pic]
Contract No SES6-CT-2004-502630
Safety of Hydrogen as an Energy Carrier First Status report on code validation and applicability based on the
results of Standard Benchmark Exercise Problems (SBEPs) V1 and V2 Deliverable No. 23 Lead Participant: UPM (E. Gallego, J. García, E. Migoya, J.M. Martín-
Valdepeñas, A. Crespo)
FzK (A. Kotchourko, T. Jordan, J. Travis, J. Yañez)
NCSRD (A. Venetsanos, E. Papanikolau)
Partners Contributing: BRE (S. Kumar, S. Miles)
CEA (H. Paillère, A. Beccantini, E. Studer)
DNV (T. Elvehøy)
EC-JRC (D. Baraldi, H. Wilkening)
Fh-ICT (H. Schneider)
FZJ (K. Verfondern)
GexCon (O.R. Hansen)
HSE/HSL (S. Ledin)
INERIS (Y. Dagba)
NH (S. Høiset)
TNO (M.M. van der Voort)
UU (V. Molkov, D. Makarov)
WUT (A. Teodorczyk, J. Piechna)
Dissemination Level: PU (Public)
Document Version: Draft 0.1
Date of submission: 10.06.2005
Due date of delivery: 31.05.2005
Co-funded by the European Commission within the Sixth Framework Programme
(2002-2006)
Executive Summary This paper presents a compilation and discussion of the results supplied by
HySafe partners participating in two Standard Benchmark Exercise Problems
(SBEPs): SBEP-V1, which is based on an experiment on hydrogen release,
mixing and distribution inside a vessel (Shebeko et al., 1988), and SBEP-
V2, which is based on an experiment on hydrogen combustion (Schneider H.
and Pförtner, 1983). Each partner has his own point of view of the problems
and used a different approach to the solutions. The main characteristics of
the models employed for the calculations are compared in a very succinct
way by using tables. For each SBEP, the comparison between results,
together with the experimental data when available, is made through a
series of graphs. Explanations and interpretations of the results are
presented, together with some useful conclusions for future work.
The main conclusions derived from each exercise are summarised below.
SBEP-V1: An intercomparison exercise on the capabilities of CFD models to
predict distribution and mixing of H2 in a closed vessel
Different approaches have been used to simulate the experiment. It is
difficult to compare the combined effects on simulation results of
turbulence model (LES RNG, RANS k-e standard), grid (structured vs.
unstructured), size of the grid, the time steps... A first conclusion is
that comparison between numerical results and experimental data should only
be performed once a grid-convergence study has been made, because it is
necessary to demonstrate that the computed results are driven by physical
phenomena and not by numerical diffusion or inadequate grid resolution.
In general, the simulations have a good agreement with the measurement, but
many models have underpredicted the transport of the hydrogen to the bottom
region at the beginning, and improved their results at the end. The
simulations are better improved using a different Prandtl turbulent numbers
during the diffusion. But the final reasons for hydrogen transport down to
the bottom of the vessel remain a gap of knowledge. To improve our
understanding of slow hydrogen movement in a closed vessel, further
research on flow decay during long periods of time is needed.
A recommendation for future works is checking conservation of hydrogen
(mass) and numerical loss of hydrogen at points of poor convergence.
Shorter time steps and stricter convergence criteria could probably
guarantee the mass conservation. WUT partner suggested that the simulation
should be performed with two different turbulence models, one before and
other after the end of the release. In all cases, an appropriate choice of
turbulence model must be selected because, in a closed vessel after the end
of the release, turbulent flow becomes laminar in a relative short time.
Appropriate models should be chosen to simulate hydrogen transport in the
last stages, when turbulence velocities are very low.
Some remarks were also made with respect to the ideal conditions needed for
an experiment in order to be fully useful for code validation. Certainly,
the Russian-2 test presented several weak aspects, between others:
reproducibility was not reported; temperatures at the release exit and at
the walls were not monitored; uncertainty ranges of the measured were not
provided.
An important open issue, in order to quantify the convection mixing due to
gas heating transport, is the effect of the non-adiabatic walls.
Experiments with accurate flow field measurements under adiabatic
temperature conditions could shed some light into this problem. Open
questions often remained unanswered due to uncontrolled boundary
conditions, in particular the configuration of the exit mouth for hydrogen
release. A better control of the boundary conditions is a necessary aspect
in order to produce experimental data for benchmark exercises. This has to
be a requirement for further SBEP exercises.
However, the performed SBEP simulations provided very useful comparison of
performance of different models, which could hardly be possible to conduct
by any single partner alone.
SBEP-V2: An intercomparison exercise on the capabilities of CFD models to
predict deflagration of a large-scale H2-air mixture in open atmosphere
The results show quite good agreement with the experimental data. Most of
the calculations reproduced quite well the flame velocity, an important
parameter for safety purposes. The pressure dynamics obtained numerically
are in good agreement with the experiments, for the positive values. The
negative pressures are more sensitive to far field boundary condition and,
as a consequence, to the size of the computational domain. Therefore, the
numerical values obtained present more dispersion. This can be avoided
using larger domains and finer grids. Nevertheless, taking into account the
possible errors in some measured pressures, the agreement cannot be
considered bad for the negative pressures.
Lessons learnt from this exercise will be useful for improving our models
and codes that will be tested soon against new SBEPs. Depending on the
numerical implementation of the same combustion model CREBCOM (Efimenko and
Dorofeev, 1998), numerical oscillations appeared in CAST3M and not in the
COM3D code (Kotchourko and Breitung, 2000). A future modification of the
combustion criterion is expected to eliminate these oscillations and to
allow using a second-order reconstruction and then to provide more accurate
results. The AutoReaGas (Berg et al., 2000) model will be properly
calibrated for H2 and will make use of larger domain to avoid
underprediction of the flame velocity. The code used by NH was and old
version of FLACS (Hansen et al., 2005) that, together with the assumption
of non stationary initial velocities inside the balloon, lead to an
inaccurate flame front propagation. However, with the newest version of
FLACS, GexCon obtained results considerably improved. The adaptive meshing
used by the REACFLOW code (Wilkening and Huld, 1999) is a peculiar
characteristic that seems to contribute to improve the accuracy of the
pressure wave propagation both in the reaction zone and beyond it into the
far field region. The LES combustion model used by UU is based on the use
of the progress variable equation and the gradient method to reproduce
flame front propagation with proper mass burning rate. This approach helps
to decouple physics and numerics of the simulated process and make
simulations less grid dependent.
References
Berg A.C. van den, Robertson, N.J., and Maslin, C.M. AutoReaGas, for gas
explosion and blast analyses. In Course on Explosion Prediction and
Mitigation, University of Leeds, 7-10 November, 2000. CAST3M: http://www-cast3m.cea.fr/cast3m/index.jsp Efimenko, A.A. and Dorofeev, S.B. CREBCOM code system for description of
gaseous combustion. Technical report, Russian Research Kurchatov
Institute, Moscow, 1998. Hansen, O.R., Renoult, J., Tieszen, S.R. and Sherman, M., Validation of
FLACS-HYDROGEN CFD Consequence Model Against Large-Scale H2 Explosion
Experiments in the FLAME Facility. Paper to be presented at the
International Conference on Hydrogen Safety, Pisa 8-10 September, 2005. Kotchourko, A., Breitung, W., Numerische Simulation der turbulenten
Verbrennung von vorgemischten Gasen in komplexen 3D-Geometrien. FZK-
Nachrichten, Forschungszentrum Karlsruhe, 2000. Schneider H., Pförtner H. PNP-Sichcrheitssofortprogramm,
Prozebgasfreisetzung-Explosion in der gasfabrik und auswirkungen von
Druckwellen auf das Containment. Dezember 1983. Shebeko, Y.N., Keller, V.D., Yeremenko, O.Y., Smolin, I.M., Serkin, M.A.,
Korolchenko A.Y., Regularities of formation and combustion of local
hydrogen-air mixtures in a large volume, Chemical Industry, 21, 1988, pp.
24(728)-27(731) (in Russian). Wilkening H. and Huld T. An adaptive 3-D solver for modelling explosions on
large industrial environment scales. Combustion Science and