Influence of Wind Direction on Gas Detector Allocation in Offshore Production Unit

One of the most successful mitigation measures against hydrocarbon leaks in offshore platforms is the installation of gas detector systems. A probabilistic analysis of gas dispersion by means of fluid dynamical simulations and input statistics is a very effective way to aid allocating gas detectors. The main objective of this study is to evaluate to what extent the assessment of several wind directions is needed to obtain accurate optimal detector mapping.

The software FLACS was used for CFD dispersion simulations of 360 scenarios inside a Gas Injection Module of a FPSO, comprehending ten leak positions in three different group of equipment, four leak directions and nine wind directions, including no wind condition. The gas concentration patterns were post-processed with developed software that, based on leak frequency and weather statistics, generates a three-dimensional map of gas occurrence probability. The allocation of gas detectors is determined by a chain-like algorithm that places the devices in positions with the highest gas probabilities, removes the detected clouds from the mapping, and places the next devices in sequence until the entire domain is covered.

Three wind directions were determined as most frequent, contributing to 60% of the wind rose. The detectors’ allocation was compared when calculated from all 360 simulations and only 120 from the frequent wind directions. The results show great similarity between the two approaches for the gas occurence mapping. As for detection allocation, 20 devices were proven necessary to cover all leaks when considering only the three most frequent winds. The study performed for all wind directions resulted in 23 detectors, but with 99.52% coverage of leak scenarios when the first 20 devices are placed. This leads to the conclusion that fewer CFD simulations are necessary for developing an optimal detector layout.