Investigation of Field Experiments and Mathematical Models for Heavy Gas Dispersion

Release and dispersion of toxic and flammable gases in atmosphere is one of the most important incident in safety of the processes. Risk analysis with the aim of prevention from harm and damage usually carries out by software packages, which are based on the field experiments and mathematical models.

In order to derive dispersion models and evaluate existing models, some different experiments are done. Experiments of the gas release and gas dispersion are in two categories, experiments which took place in wind tunnels and which are field experiments. Kit Fox, Thorney Island and coyote are some of the most famous field experiments. PERP group and EMU tests are major experiments in wind tunnels.  In many of studies, gas dispersion was investigated in the open places in absence or presence of obstacles because most of the industrial accident happens in open places. Others are also taking place in indoors and large buildings. Early, simple models such as box models, steady state plume and integral models were proposed. Thereafter, group models like Lagrangian models and Lagrangian- Gaussian models were evinced. One of the other approach is using more complex and computational methods. Fluid dynamics methods are designed and developed for this purpose. The models of the heavy gas dispersion can be categorized to four major group. The first is simple and experimental models. intermediate and integral or shallow layer models include box models, steady state / general steady state plume models, one dimensional integral models is located in the next. The third group is advanced and lagrange models. The last and latest models are computational fluid dynamic models: RANS, LES and DNS

Different gases, distinct release scenarios are studied in researches. As another effective parameter on the path of gas dispersion, topology of release location can be mentioned which is investigated in field experiments and simulations.

in order to use and refer the field   experiments and models as an evaluation, desired scenarios’ condition should be as close as possible to the models or experiments’ condition. These conditions could be such as type of gas, terrain and topology of release path, puff or plume release and other environment and physical conditions.