Hydrogen release and atmospheric dispersion: Experimental studies and comparison with parametric simulations

In our society the use of hydrogen is continually growing and there will be a widespread installation of plants with high capacity storages in several towns as automotive refueling stations. For this reason, it is necessary to make accurate studies on the safety of these kinds of plants to protect our town inhabitants. Generally several simulation models, whether or not concerned with fluid dynamics, used in safety and risk studies are validated in the past for hydrogen use. However there is a very important need of experimental data covering a broad range of system configurations for strict validation of CFD simulations of hydrogen. This aspect may imply that the results of validation studies can be too accurate and realistic. This paper introduces an experimental activity which was performed by the Department of Energetics of Politecnico of Torino with the collaboration of the University of Pisa. Accidental hydrogen release and dispersion were studied in order to acquire a set of experimental data to validate simulation models for such studies. A pilot plant called Hydrogen Pipe Break Test was built. During the experimental activity, data was acquired regarding hydrogen concentration as a function of distance from the release hole, also lengthwise and vertically. In this paper some of the experimental data acquired during the activity have been compared with the integral models, Effects and Phast.