1. 6th International Conference of Hydrogen Safety
Development of a model evaluation protocol for CFD analysis of hydrogen safety issues the SUSANA project
D. Baraldi (1), D. Melideo. (1), A. Kotchourko (2), K. Ren (2), J. Yanex (2), O. Jedicke (2), S.G. Giannissi (3), I.C. Tolias (3), A.G. Venetsanos (3), J. Keenan (4), D. Makarov (4), V. Molkov (4), S. Slater (5), F. Verbecke (6), A. Duclos (6)
1 - European Commission, Joint Research Centre (JRC),
2 - Karlsruhe Institute for Technology (KIT),
3 - Environmental Research Laboratory, National Center for Scientific research Demokritos,
4 - HySAFER Centre, University of Ulster,
5 - Element Energy Limited
6 - Areva Stockage d'Energie SAS
6th International Conference of Hydrogen Safety
18-21 October – Yokohama, Japan

2. Background
Hydrogen safety issues is a crucial aspect for a wide spread deployment and use of hydrogen and fuel cell technologies
Hydrogen technologies must have the same or lower level of hazard and associated risk compared to conventional fossil fuel technologies
European Commission | ICHS2015

3. Context
Computational Fluid Dynamics (CFD) is increasingly used to perform safety analysis of potential accident scenarios (production, storage, distribution of hydrogen and its use in fuel cells)
CFD is a powerful numerical tool that can provide useful data and insights but it also requires a high level of competence and knowledge in order to be used in a meaningful way
To apply CFD with a high level of confidence on the accuracy of the simulation results, two main issues have to be addressed:
the capability of the CFD models to accurately describe the relevant physical phenomena
the capability of the CFD users to follow the correct modelling strategy.
The reliability/accuracy of the CFD results remains a significant concern.
European Commission | ICHS2015

4. Gaps in CFD modelling
Workshop with recognised experts in the field of hydrogen safety
Identification of gaps in CFD modelling and simulation of hydrogen release and combustion
European Commission | ICHS2015

5. Gaps in CFD modelling
One of the main gaps: the lack of a Model Evaluation Protocol (MEP) for hydrogen technologies safety
HyMEP:
The first MEP for hydrogen technologies safety
To be beneficial for all the CFD developers (academia and research institutes) and users (like industry and consultancy companies) but also for regulatory/certifying bodies
To evaluate the accuracy of the CFD models
To assess user capability of correctly using the codes
European Commission | ICHS2015

6. The SUSANA Project
SUpport to SAfety ANalysis of Hydrogen and Fuel Cell Technologies (
The SUSANA project (co-funded by the Fuel Cell and Hydrogen Joint Undertaking): producing a Model Evaluation Protocol for hydrogen technologies safety (HyMEP)
European Commission | ICHS2015

7. Existing Protocols in other fields
Model Evaluation Group (MEG) established by the EC in 1994 for consequence models
Evaluation protocol in specific areas like heavy gas dispersion (HGD) and gas explosions (MEGGE, 1996)
Scientific Model Evaluation Techniques for Dense Gas Dispersion Models in Complex Situations (SMEDIS) (Daish et al., 2000)
Evaluation protocol for LNG dispersion models (Ivings et al., 2007)
HyMEP is the first protocol for hydrogen technologies safety
HyMEP is the first protocol including all relevant phenomena (release, dispersion, ignition, deflagrations, detonations and fires)
European Commission | ICHS2015

8. HyMEP Structure Stage 1: Scientific Assessment
Initial critical analysis of the model based on available knowledge in the field
Critical review: physical, mathematical and numerical model basis
Identify the known and/or expected weakness and strengths from available literature and knowledge
Scientific content: Assumptions/simplifications/applicability range
European Commission | ICHS2015

9. HyMEP Structure Stage 2: Verification
Verification is used to ensure that a mathematical model has been correctly implemented in software i.e. the equations are correctly solved
Verification Database
European Commission | ICHS2015

10. HyMEP Structure Stage 3: Validation
Model outputs are compared with measurements of physical parameters to demonstrate that the model captures “real world” behaviour across its intended range of applicability.
Quantitative comparison of experimental observations vs. model predictions
Validation Database
European Commission | ICHS2015

11. HyMEP Structure Stage 4: Sensitivity study Grid independency Time-step
CFL sensitivity
Numerical scheme
Boundary conditions
Domain size
European Commission | ICHS2015

12. HyMEP Structure Stage 5: Quantitative Assessment Criteria
Identification of target variables for each phenomenon under consideration
Statistical analysis:
Performance parameters
Methodology
Quantitative criteria
Sensitivity and uncertainty
European Commission | ICHS2015

13. HyMEP Structure Stage 6: Assessment Report
Analysis of information supplied by model developer/expert user.
Detailed model description
Scientific assessment
Verification and validation
Sensitivity study
Statistical analysis
Conclusions
European Commission | ICHS2015

14. HyMEP supporting documents
D2.1: State of the art in physical and mathematical modelling of safety phenomena relevant to FCH technologies.
D2.2: Critical analysis and requirements to physical and mathematical models.
D3.2: Guide to best practices in numerical simulations
European Commission | ICHS2015

15. The Model Validation Database
First version ( including ~30 experiments
Type of experiments:
release and dispersion (indoors and outdoors, small enclosures and garage facilities, and vented configurations): performed with gaseous hydrogen (or helium) and only one with liquid hydrogen
ignition: investigation of the self-ignition of gaseous hydrogen in a pressurized tube at different pressures with a T shaped pressure relief device
deflagrations (different hydrogen concentrations, an open environment, a closed or vented box, and the presence of obstacles)
detonations (detonations of lean hydrogen-air mixtures (20%, 25%) in a closed large scale facility)
DDT: first set with hydrogen in straight pipes of three different diameters and with different gas concentrations; second set explosions in an obstructed 12 m long tube with a 15% hydrogen-air mixture
Fires (in the next version of the database)
European Commission | ICHS2015

16. Benchmarking activities (ongoing)
Some experiments have been selected from the Validation database for benchmarking activities
To test the stages of the HyMEP
Indication of the accuracy
Range of applicability of each modelling approach
Assessment or the performance of each model for each kind of phenomena
To suggest values for the statistical performance measures
European Commission | ICHS2015

17. Benchmarking activities (ongoing)
Release and Dispersion
GAMELAN
JRC, NCSRD, UU
SBEP_21
HSL, JRC
Ignition
PRD (Pressure Relief Device) UU
Deflagrations
HyIndoor_WP3
KIT
Open atmosphere deflagration
NCSRD, UU
Detonations
KI_RUT_hyd05
KI_RUT_hyd09
Release and Dispersion (HELIUM)
He is often used in release experiments for safety reasons
He is the most similar element to H2 in terms of buoyancy
See presentation on Tuesday: "Comparisons of helium and hydrogen releases in 1 m3 and 2 m3 two vents enclosures: concentration measurements at different flow rates and for two diameters of injection nozzle" (Gilles Bernard-Michel, Deborah Houssin)
European Commission | ICHS2015

18. Release and dispersion: GAMELAN
Example of statistical analysis
European Commission | ICHS2015

19. Release and dispersion: GAMELAN
CEA-GAMELAN facility with size 0.93x0.93x1.26 m
Helium source centred horizontally is located at 21 cm from floor
Release rate 180 NL/min (nozzle = 5 mm)
The vent (180 x 180 mm) located in the middle of the wall and 20 mm below the ceiling
He concentration (v/v) was measured at various heights from floor
European Commission | ICHS2015

20. Release and dispersion: GAMELAN
CFD RESULTS - NCSRD
ADREA-HF CFD code
The turbulence model is the standard k-ε including the buoyancy terms
Steady 400 s
European Commission | ICHS2015

21. Release and dispersion: GAMELAN
CFD RESULTS – Statistical analysis
ADREA-HF CFD code
The turbulence model is the standard k-ε including the buoyancy terms
"Good" model (in terms of atmospheric dispersion):
FB abs value < 0.3
0.7 < MG < 1.3
FB negative
MG < 1
model overall over-predicts the He conc. at steady state
European Commission | ICHS2015

22. Release and dispersion: SBEP21
Example of model sensitivity
European Commision | ICHS2015

23. Release and dispersion: SBEP21
CEA-GARAGE facility representing a realistic single vehicle private garage
Release phase (18 L/min for 3740 s) + diffusion phase
European Commission | ICHS2015

24. Release and dispersion: SBEP21
CFD RESULTS - JRC
European Commission | ICHS2015

25. Release and dispersion: SBEP21
CFD RESULTS – JRC (effect of the mesh type)
Tetra is not suitable for diffusion phase
European Commission | ICHS2015

26. Spontaneous ignition Example of validation
Investigation of the effect of a relevant parameter
European Commission | ICHS2015

27. Spontaneous ignition
University of Ulster has simulated the experiments carried by Golub et al. [1]
H2 released from a high pressure system into a channel ending in a T-shaped nozzle mimicking a Pressure Relief Device (PRD)
Initial H2 pressures of 1.5 MPa and 2.9 MPa
[1] V.V. Golub, V.V. Volodin, D.I. Baklanov, S.V.Golovastov, D.A. Lenkevich, Experimental investigation of hydrogen ignition at the discharge into channel filled with air. In: Physics of extreme states of matter, ISBN ; 2010, Chernogolovka.
High Pressure tube
High Pressure tube
PRD
Burst disk
European Commission | ICHS2015

28. Spontaneous ignition 1.5MPa Initial P 2.9MPa Initial P Temperature
No combustion chemical reaction
1.5MPa Initial P
hydroxyl concentration
Temperature
2.9MPa Initial P
hydroxyl concentration
European Commission | ICHS2015

29. Example of geometry model sensitivity
Deflagration
Example of geometry model sensitivity
European Commission | ICHS2015

30. Deflagration The experiment HyIndoor_WP3 performed by KIT
Reproduced numerically with the code COM3D by KIT
The KIT facility is a chamber similar to a garage box with glass walls filled with a 18% hydrogen-air mixture
The ignition was triggered at the centre of the wall opposite to a 0.5mx0.5m vent
European Commission | ICHS2015

31. Deflagration Initial simple CFD model Intermediate complex CFD model
European Commission | ICHS2015

32. Conclusions Intermediate results from the collaborative SUSANA project
Aim of the project is to develop a Model Evaluation Protocol (HyMEP) for CFD models for safety analyses of hydrogen and fuel cell technologies
HyMEP draft document is almost ready
The HyMEP structure with the main stages have been shown
A verification database and a validation database under development
Some examples of benchmarking exercises have been shown
European Commission | ICHS2015

33. Next steps The SUSANA project will finish in August 2016
Would you like to participate to the review of the HyMEP draft document in the first months of 2016?
Would you like to participate to the final dissemination workshop probably in June 2016 (Exact date and place to be defined)?
Would you like to receive the final version of the document?
If interested in any of the above possibilities, please get in contact with
ACKNOWLEDGEMENTS
The authors would like to thank the Fuel Cell and Hydrogen Joint Undertaking for the co-funding of the SUSANA project (Grant-Agreement FCH-JU )
European Commission | ICHS2015