Return to search

Modelling vaporizing fluid flow through porous media with applications to liquefied natural gas

The problem of vaporizing flow of liquefied natural gas (LNG) through porous or penetrable media has received very little attention despite its importance in assessing the performance and risk-based safety of large membrane tank LNG ships under barrier leakages. In this work, a fluid flow model is proposed and used to analyse the vaporizing flow behaviour of LNG through soil and glass wool porous materials. Furthermore, a modified vaporizing liquid pool model is implemented and used to examine the problem of vaporizing LNG pool on non-penetrable solid substrates. We employed an explicit, finite difference and a fourth-order Runge-Kutta algorithms coded in FORTRAN to respectively solve the flow and pool models. Both models were successfully verified and validated by comparisons to experimental data, analytical solutions, and to predictions of a commercial software (TOUGH2). Results from the vaporizing flow and pool analyses demonstrate that, for some of the applications considered, the liquid is expected to reach considered threshold depths, seep through the porous layer and contact, contaminate and/or embrittle surrounding natural or engineered systems. For the specific application to LNG cargo containment systems (or cargo tanks), this work has shown that there are safety risks associated with LNG leakage, which are ultra-low temperature of the inner hull, cryogenic damage and subsequent failure of the cargo containment system. Thus, for any LNG membrane cargo containment system to continue to be safe and secure, the various structural members of the insulation system should be designed and equipped with new and improved materials that possesses the necessary mechanical and thermophysical properties to maintain and/or improve the critical temperature standard and low-temperature performance of these systems. Further work should consider additional experimental evidence in order to fully validate and establish that solution predictions by the proposed models are describing the actual physical effects.

Identiferoai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:576051
Date January 2013
CreatorsOkafor, Emeka Joachin
ContributorsVesovic, Velisa
PublisherImperial College London
Source SetsEthos UK
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation
Sourcehttp://hdl.handle.net/10044/1/11675

Page generated in 0.0017 seconds