Scaling procedure for the design of a validation experiment on an accidental gas release

Oil&Gas platforms are complex structures characterized by the presence of flammable and pressurized fluids and by a unique working environment, with limited spaces and congested presence of processing equipment. Risk assessment on these structures is mandatory according to international directives and it involves the evaluation of frequencies and consequences of possible accidental scenarios. The Italian Ministry of Economic Development is supporting the SEADOG (Safety and Environmental Analysis Division for Oil&Gas) laboratory at Politecnico di Torino to develop research projects on Oil&Gas offshore safety. At SEADOG, we have developed an innovative CFD hybrid approach to simulate the consequences of accidental gas releases. This approach simulates two separate and consequent steps that happen in an accidental gas jet release: the first one is the initial supersonic behavior modeled into a portion of the total platform domain; the second step is the following subsonic dispersion of the released gas into the total domain. The supersonic characteristics appear within the first tens of centimeters from the release point, where the compressibility effects drive the phenomena evolution. The subsonic dispersion domain may corresponds to entire platform (tens of meters), where the fluid can be treated as incompressible. The two steps are computationally evaluated and optimized separately, while the coupling follows. A key boundary condition for the phenomenon simulation is the wind speed. This real scale speed has to be defined according with the typical velocity distribution of the atmospheric wind present in the location (Adriatic Sea) where the platform is positioned. To validate this theoretical approach, a physical experiment is designed for the investigation in a wind tunnel of a scaled (1:10) mockup of a real platform will be equipped with a gas release supply line and a sensors’ network. While the mock up is scaled according to a geometric ratio, a finer assessment is needed to apply the correct scaling approach to define the reference speed in the wind tunnel and the mass flow release. The scaling of the wind velocity has followed the scaling theory for wind tunnels, respecting the logarithmic evolution of the wind profile in proximity to the sea surface. The scaling of the mass flow release is inspired by the approach proposed by Hall and Walker (1997) with some adjustments to fit our specific case. In particular, a dimensionless group is chosen to link pressure release, nozzle diameter, scaling factor and gas properties: the dimensionless discharge momentum flux. Furthermore, this group contains information about both the real and the scaled wind velocity. The approach proposed in this paper aims at scaling supersonic jet releases, maintaining their critical behavior for the test. A selection of case studies will be tested in the experimental setup to be built by the end of year 2019 under the supervision of the SEADOG research team.