The transfer under dynamic conditions of volatile species from a liquid pool to the surrounding air is gaining interest in the engineering community.
In particular, increasingly stringent regulations and standards apply to all types of flammable substances. This is especially the case in stationary gas turbine applications, where the vaporization of accidentally occurring pools of liquid fuels attracts increasing attention.
Since the flame-to-explosion transition cases are insufficiently controlled, the current, only practicable approach to assess the explosion risks arising from a fuel pool in an enclosure consists in quantifying the amount of vapor that leaves the pool and minimizing, by means of a proper dilution strategy, the potential damages entailed by the ignition of the resulting cloud. This approach requires two steps:
(1) The accurate assessment of vaporization rates under given ventilation conditions, a task that calls for skills in thermal and mass transfers.
(2) The reliable prediction of the transport of the vapors in the ventilation stream, a task specifically focused on fluid dynamics.
The three teams involved in this paper have joined their efforts to achieve this multidisciplinary objective. As a first task, the LRGP team (Laboratoire des Reactions et Génie des Procédés) and GE Energy have experimentally validated a vaporization model initially devised for water pools. This work has been reported in a recent paper. Concurrently, EURO/CFD and GE Energy have developed a CFD approach devoted to the mixing/dilution processes in defined enclosure geometries and under specified ventilation conditions. Finally both approaches havebeen coupled by EURO/CFD to produce predictive isopleth pictures of the vapor clouds generated under given temperature and velocity conditions.
The present paper covers the integration of the liquid pool vaporization model in thecommercial CFD software ANSYS Fluent and sets out the results obtained.
This dual, concerted approach is a first of the kind to the authors’ knowledge and proves fruitful for the prediction of the spatial distributions of the volatile species developed when a volatile pool vaporizes in a ventilated enclosure. It fills a gap in the analysis of safety scenarios arising from spillages of liquid fuels and provides a rational tool in zone classification studies.