CFD gas release model performance evaluation through wind tunnel experiments. (February 2022)

• Record Type: Journal Article
• Authors: Moscatello, Alberto; Gerboni, Raffaella; Ledda, Gianmario; Uggenti, Anna Chiara; Piselli, Arianna; Carpignano, Andrea
• Is Part Of: Journal of loss prevention in the process industries. Volume 75(2022)
• Publication Date: 2022-02
• Subjects: Oil & gas -- CFD -- ANSYS Fluent -- Accidental gas release -- Experimental test -- Wind tunnel; Chemical industries -- Safety measures -- Periodicals; 660.2804
• DOI: 10.1016/j.jlp.2021.104715
• Languages: English
• ISSNs: 0950-4230

Abstract

Abstract: In this paper, the experimental validation of an innovative CFD approach, called SBAM ("Source Box Accidental Model"), is described. SBAM was developed in ANSYS Fluent and is aimed at a more efficient characterisation of accidental high-pressure gas releases in congested plants. The experimental setup, methodology and a preliminary CFD-experimental data comparison are described. The validation campaign has been carried out in the SEASTAR-WT wind tunnel, realized at the Environment Park in Turin (Italy). A 1:10 scaled Oil & Gas platform mockup, equipped with flow and gas concentration sensors, was built and installed inside the wind tunnel, in order to reproduce an accidental gaseous release in dynamic similarity conditions with the real cases. A set of gas releases were performed, and the predicted concentrations were compared to the observed ones in order to validate the CFD model. Several statistical measures with their range of acceptability allowed to compare the two sets of data. Results showed that, in most of the cases, acceptance criteria were met and a good consistency between experimental and numerical values is found. Furthermore, an overestimating tendency of SBAM is observed, suggesting that it is a conservative tool for consequences estimation. Highlights: A novel CFD approach for gas dispersion is validated against experimental data. A Wind Tunnel and an offshore platform mockup are employed. The CFD and experimental data are compared through statistical parameters. The CFD approach resulted conservative and in good agreement with the experiments.