Hydrogen gas concentrations and jet velocities were measured downstream by a high response speed flame ionization detector and PIV in order to investigate the characteristics of dispersion and ignitability for 40 to 82 MPa high-pressurized hydrogen jet which was discharged from the nozzle with 0.2mm diameter,. The lights emitted from both OH radical and water vapor species yielded from hydrogen combustion ignited by an electric spark were recorded by two high speed cameras. From the results, the empirical formula concerning the relations among time-averaged concentrations, concentration fluctuations and ignition probability were obtained to suggest that the relations would be expressed independent of hydrogen discharge pressure.

The highly combustible nature of hydrogen poses a great hazard, creating a number of problems with its safety and handling. As a part of safety studies related to the use of hydrogen in a confined environment, it is extremely important to have a good knowledge of the dispersion mechanism. The release of hydrogen may result in the formation of an explosive atmosphere, which can explode and cause serious damage. The present work investigates the concentration field and flammability envelope from a small scale leak. The hydrogen is released into a 0.47 m x 0.33 m x 0.20 m enclosure designed as a 1/15 – scale model of a room in a nuclear facility. The performed tests evaluates the influence of the initials conditions at the leakage source on the dispersion and mixing characteristics in a confined environment. The role of the leak location and the presence of obstacles, are also analyzed. Throughout the test, during the release and the subsequent diffusion phase, temporal profiles of hydrogen concentration are measured using the thermal conductivity gauges within the enclosure. In addition, the BOS (Background Oriented Schlieren) technique is used to visualise the cloud evolution inside the enclosure. These instruments allow the observation and quantification of the stratification effects.

In SUSANA E.U. project a rather broad CFD benchmarking exercise was performed encompassing a number of CFD codes, a diversity of turbulence models... It is concluded that the global agreement is good. But in this particular situation, the experimental data to compare with were known to the modelers. In performing, this exercise, the present authors explored the influence of some modeling choices which may have a significant impact on the results (apart from the traditional convergence testing and mass conservation) especially in the situation where little relevant data are available. The configuration investigated is geometrically simple: a vertical round hydrogen jet in a square box. Nevertheless, modeling aspects like the representation of the source and of the boundary conditions have a rather strong influence on the final results as illustrated in this communication. In other words, the difficulties may not be so much in the intrinsic capabilities of the code (which SUSANA tends to show) but more in the physical representation the modelers have. Even in the specific situation addressed in this communication, although looking simple, it may not be so obvious to grasp correctly the leading physical processes.

To define a strategy of mitigation for containerized hydrogen systems (fuel cells for example) against explosion, the main characteristics of flammable atmosphere (size, concentration, turbulence…) shall be well-known. This article presents an experimental study on accidental hydrogen releases and dispersion into an enclosure of 4 m3 (2 m x 2 m x 1 m). Different release points are studied: two circular releases of 1 and 3 mm and a system to create ring-shaped releases. The releases are operated with a pressure between 10 and 40 bars in order to be close to the process conditions. Different positions of the release inside the enclosure i.e. centred on the floor or along a wall are also studied. A specific effort is made to characterize the turbulence in the enclosure during the releases. The objectives of the experimental study are to understand and quantify the mechanisms of formation of the explosive atmosphere taking into account the geometry and position of the release point and the confinement. Those experimental data are analyzed and compared with existing models and could bring some new elements to improve them.

With the development of the hydrogen economy, it requires a better understanding of the potential for fires and explosions associated with the unintended release of hydrogen within a partially confined space. In order to mitigate the hydrogen fire and explosion risks effectively, accurate predictions of the hydrogen transport and mixing processes are crucial. It is well known that turbulence modelling is one of the key elements for a successful simulation of gas mixing and transport. GASFLOW-MPI is a scalable CFD software solution used to predict fluid dynamics, conjugate heat and mass transfer, chemical kinetics, aerosol transportation and other related phenomena. In order to capture more turbulence information, the Large Eddy Simulation (LES) model and LES/RANS hybrid model Detached Eddy Simulation (DES) have been implemented and validated in 3-D CFD code GASFLOW-MPI. The standard Smagorisky SGS model is utilized in LES turbulence model. And the k-epsilon based DES model is employed. This paper assesses the capability of algebraic, k-epsilon, DES and LES turbulence model to simulate the mixing and transport behavior of highly buoyant gases in a partially confined geometry. Simulation results agree well with the overall trend measured in experiments conducted in a reduced scale enclosure with idealized leaks, which shows that all these four turbulent models are validated and suitable for the simulation of light gas behavior. Furthermore, the numerical results also indicate that the LES and DES model could be used to analysis the turbulence behavior in the hydrogen safety problems.

Hydrogen leakage is a key safety issue for hydrogen energy application. For hydrogen leakage, hydrogen releases with low momentum, hence the development of the leakage jet is dominated by both initial momentum and buoyancy. It is important for a computational code to capture the flow characteristics transiting from momentum-dominated jet to buoyancy dominated plume during leakage. GASFLOW-MPI is a parallel computational fluid dynamics (CFD) code, which is well validated and widely used for hydrogen safety analysis. In this paper, its capability for small scale hydrogen leakage is validated with unintended hydrogen release experiment. In the experiment, pure hydrogen is released into surrounding stagnant air through a jet tube on a honeycomb plate with various Froude numbers (Fr). The flow can be fully momentum-dominated at the beginning, while the influence of buoyancy increases with the Fr decreases along the streamline. Several quantities of interest including velocity along the centerline, radial profiles of the time-averaged H2 mass fraction are obtained to compare with experimental data. The good agreement between the numerical results and the experimental data indicates that GASFLOW-MPI can successfully simulate hydrogen turbulent dispersion driven by both momentum and buoyant force. Different turbulent models i.e. k-ε, LES, and DES model are analyzed for code performance, the result shows that all these three models are adequate for hydrogen leakage simulation, k-ε simulation is sufficient for industrial applications, while, LES model can be adopted for detail analysis for a jet/plume study like entrainment. The DES model possesses both characters of the former two model, only the performance of its result depends on the grid refinement.

Preparing for the arrival of the hydrogen society, it is necessary to develop suitable sensors to use hydrogen safely. There are many methods to know the hydrogen concentration by using conventional sensors, but it is difficult to know the behavior of hydrogen gas from long distance. This study measured hydrogen dispersion by using Raman scattering light. Generally, some delays occur when using conventional sensors, but there are almost no delays by using the new Raman sensor. In the experiments, 6mm & 1mm diameter holes are used as a spout nozzle to change initial velocities. To ensure the result, a special sheets are used which turns transparent when it detected hydrogen, and visualized the hydrogen behaviour. As a result, the behaviour of the hydrogen gas in the small container was observed. In addition, measurable distance is increased by the improvement of the device.

The article deals with LES simulations of an air-helium buoyant jet in a two vented enclosure and their validation against particle image velocimetry experiments. The main objective is to test the ability of LES models to simulate such scenarios. These types of scenarios are of first interest considering safety studies for new hydrogen systems. Three main challenges are identified. The two first are the ability of the LES model to account for a rapid laminar-to-turbulence transition, mainly due to the buoyancy accelerations, and the Rayleigh-Taylor instabilities that can develop due to sharp density gradients. The third one is the outlet boundary conditions to be imposed on the vent surfaces. The influence of the classical pressure boundary condition is studied by comparing the simulations results when an exterior region is added in the simulations. The comparisons against particle image velocimetry experiments show that the use of an exterior domain gives more accurate results than the classical pressure boundary condition. This result and the description of the phenomena involved are the main outlets of the article.

The present work proposes improvments on a model developped by Linden to predict the concentration distribution in a 2 vented cavities. Recent developments on non constant entrainment coefficient from Carazzo et al as well as a non constant pressure distribution at the vents - the vents being vertical - are included in the Linden approach. This model is compared with experimental results from a parametric study on the influence of the height of the release source on the helium dispersion regimes inside a naturally ventilated 2 vents enclosure. The varying parameters of the study were mainly the height of the release, the releasing flow rate and the geometry of the vents. At last, Large Eddy Similations of the flow and Particle Image Velocimetry measurements performed on a small 2 vented cavity are presented. The objective is to have a better understanding of the flow structure which is at the origin of the 2 layers concentration distribution described by Linden.

This paper summarises the results from a blind-prediction study for models developed for estimating the consequences of vented hydrogen deflagrations. The work is part of the project Improving hydrogen safety for energy applications through pre-normative research on vented deflagrations (HySEA). The scenarios selected for the blind-prediction entailed vented explosions with homogeneous hydrogen-air mixtures in a 20-foot ISO container. The test program included two configurations and six experiments, i.e. three repeated tests for each scenario. The comparison between experimental results and model predictions reveals reasonable agreement for some of the models, and significant discrepancies for others. It is foreseen that the first blind-prediction study in the HySEA project will motivate developers to improve their models, and to update guidelines for users of the models.