(104at) A CFD-Based Parametric Study on the Effect of Small Obstructions on the Dispersion of Flammable Jet Releases
AIChE Spring Meeting and Global Congress on Process Safety
2012
2012 Spring Meeting & 8th Global Congress on Process Safety
Global Congress on Process Safety
Poster Session
Tuesday, April 3, 2012 - 5:00pm to 6:30pm
For any given petrochemical facility, high-momentum jet releases (in the form of a flashing liquid or a pure vapor) from piping connections, valves, etc. are among the most likely scenarios to be included in a risk analysis and, as such, have a potentially large impact on the overall risk. Therefore, an accurate prediction of the consequences from such releases – particularly, the hazard area that may be exposed to flammable gas concentrations and the size of the flammable cloud in a congested area – is critical to the reliability of any risk assessment.
In petrochemical facilities, the density of piping, vessels and other equipment is often such that most releases would interact with one or more obstacles as they are dispersed. However, the effect of obstructions on the dispersion of high-momentum jet releases is frequently neglected in gas dispersion hazard analyses, based on the erroneous assumption that jet impingement upon obstacles increase turbulent mixing and therefore reduces the downwind dispersion distance of the gas cloud as well as the size of the flammable cloud within a congested region. In fact, the jet impingement results in increased turbulent mixing but also in a loss of momentum of the jet, which reduces air entrainment and the associated dilution of the gas concentration. Therefore, jet impingement has two competing effects, whose net result will depend upon several parameters, such as: the size of the obstruction and its distance from the source of the jet; the elevation of the release; the density of the stream; wind speed and direction; etc.
Even though some analytical models allow “impinging jets” to be simulated, the effects of impingement are very simplistically incorporated by reducing the jet release velocity at the source, and then simulating the dispersion of an unobstructed jet; this approach has no scientific basis nor any guidance on how much the jet source velocity should be reduced as a function of the obstruction characteristics. In order to provide some understanding of the effect of the parameters described above, this paper will present the results from a sensitivity study on the dispersion of impinged jets, performed using the CFD tool FLACS.