Conference paper Open Access
Hydrogen is being considered as a sustainable future energy carrier with least environmental impact in terms of combustion by-products. It has unique physical properties of very wide flammability range, between 4% to 75% by volume and high flame speeds, which are challenging factors in designing safe hydrogen installations. An accidental release in enclosures can easily result in the formation of flammable mixtures, which may upon ignition lead to fast turbulent deflagrations or even transition to detonation. Explosion venting is frequently used to mitigate explosions in industry, but it is not straightforward to design vent systems that will reduce the explosion pressure sufficiently to prevent collapse of structures and formation of projectiles. Validated predictive techniques will be of assistance to quantified analysis of possible accidental scenarios and designing effective mitigation measures such as vents. While explosion venting has been previous studied experimentally and numerically, relatively little information has been gathered about the configurations used in hydrogen energy applications and in the presence of obstacles; a viable predictive technique for such scenario is still lacking.
The use of standard 20 feet ISO shipping containers for self-contained portable hydrogen fuel cell power units is being widely considered. Fresh experiments for this configuration have been carried out by GexCon AS as part of the HySEA project supported by the Fuel Cells and Hydrogen 2 Joint Undertaking (FCH 2 JU) under the Horizon 2020 Framework Program for Research and Innovation. In the present study, numerical modelling and simulations have been conducted to aid our understanding of the vented gas explosion in these self-contained portable power units using HyFOAM, an in-house modified version of the open source Computational Fluid Dynamics (CFD) code OpenFOAM for vented hydrogen explosions. The convective and diffusive terms are discretised using Gaussian-Gamma bounded and Gaussian linear corrected numerical schemes with in OpenFOAM. The temporal terms are discretised using Euler implicit scheme making the solver second order accurate both in spatial and time coordinates.