Computational fluid dynamics simulation of gas dispersion in complex facilities using Kit Fox field experiments:Validation and statistical evaluation

Gas release and its dispersion is a major concern in chemical industries.In order to manage and mitigate the risk of gas dispersion and its consequences,it is necessary to predict gas dispersion behavior and its concentration at various locations upon emission.Therefore,models and commercial packages such as Phast and ALOHA have been developed.Computational fluid dynamics(CFD)can be a useful tool to simulate gas dispersion in complex areas and conditions.The validation of the models requires the employment of the experimental data from filed and wind tunnel experiments.It appears that the use of the experimental data to validate the CFD method that only includes certain monitor points and not the entire domain can lead to unreliable results for the intended areas of concern.In this work,some of the trials of the Kit Fox field experiment,which provided a wide-range database for gas dispersion,were simulated by CFD.Various scenarios were considered with different mesh sizes,physical conditions,and types of release.The results of the simulations were surveyed in the whole domain.The data matching each scenario was varied by the influence of the dominant displacement force(wind or diffusivity).Furthermore,the statistical parameters suggested for the heavy gas dispersion showed a dependency on the lower band of gas concentration.Therefore,they should be used with precaution.Finally,the results and computation cost of the simulation could be affected by the chosen scenario,the location of the intended points,and the release type.