Modelling the dispersion of flashing jets using CFD

Abstract

Risk assessments related to industrial environments where gas is kept in liquid form under high pressure rely on the results from predictive tools. Computational Fluid Dynamics (CFD) is one such predictive tool and it is currently used for a range of applications. One of the most challenging application areas is the simulation of multiphase flows resulting from a breach or leakage in a pressurised pipeline or a vessel containing liquefied gas. The present paper deals with the modelling of the post-flashing scenario of a jet emanating from a circular orifice. In addition to being based on the equations governing fluid flow, the models used are those related to turbulence, droplet transport, evaporation, break-up and coalescence. Some of these models are semi-empirical and based on the data from applications other than flashing. However, these are the only models that are currently available in commercial codes and that would be used by consulting engineers for the type of modelling discussed above, namely the dispersion of a flashing release. A method for calculating inlet boundary conditions after flashing is also presented and issues related to such calculations are discussed. The results from a number of CFD based studies are compared with available experimental results. The results show that whilst a number of features of the experimental results can be reproduced by the CFD model, there are also a number of important shortcomings. The shortcomings are highlighted and discussed. Finally, an optimum approach to modelling of this type is suggested and methods to overcome modelling difficulties are proposed.