Gas Detectors Allocation in Process Plants Using Computational Fluid Dynamics

Gas detector placement is one of the most important issues in the gas and oil industries since fast detection and subsequently fast control of the toxic and flammable gas releases could prevent major accidents or disasters. In this paper, dispersion of Hydrogen Sulfide at the one of the units of Sirri Gas Compression Facility of Iranian Offshore Oil Company has been simulated with computational fluid dynamics methods. In this simulation, different scenarios according to the different wind directions and speeds, leakage hole diameters, direction of leakages, operating fluid types and topology of the environment have been studied. Therefore, 4 wind directions, 6 wind speed steps, 11 sources of leakage or their directions, and 2 equivalent holes were considered, and the results were evaluated in 550 scenarios which each scenario has its own frequency of occurrence. The main range of gas dispersion discretized to 2391 points on the three parallel surface with 0.5 m intervals from ground surface. Probability of each scenario was calculated from meteorological data, and discharge probability data were gathered from references. Probability of presence of more than 10 ppm of toxic gas after 20 seconds in each scenario was computed. Finally, by summing up the probabilities of the all scenarios in every point, the optimized locations for placement of detectors was determined, and the results of these simulations were compared with the existing detectors placement. Finally, by considering the results, it is found out  that the current location of two detectors is not correct, and better locations can be selected.