Implementation and validation of an environmental feedback pool fire model based on oxygen depletion and radiative feedback in FDS. (October 2016)

• Record Type: Journal Article
• Authors: Wahlqvist, Jonathan; van Hees, Patrick
• Is Part Of: Fire safety journal. Volume 85(2016:Oct.)
• Page Start: 35
• Page End: 49
• Publication Date: 2016-10
• Subjects: Fire -- CFD -- FDS -- Pool fire -- Oxygen depletion -- Radiation feedback -- Mechanical ventilation -- HVAC -- Validation; Fire prevention -- Periodicals; Incendies -- Prévention -- Recherche -- Périodiques; Fire prevention -- Research; Periodicals; 628.92205
• DOI: 10.1016/j.firesaf.2016.08.003
• Languages: English
• ISSNs: 0379-7112

Abstract

Abstract: This paper has been aimed at implementing and validating a simplified environmental feedback fire model in FDS as a complement to more advanced pyrolysis models. The two main means of environmental feedback have been identified as the oxygen concentration close to the fuel base and the radiative feedback from the surrounding obstructions and smoke layer. The oxygen concentration at the fuel base has previously been identified to linearly influence the normalized burning rate compared to the free burning behavior; this correlation has been implemented in FDS as a simple way to compensate for the reduced radiative feedback the fuel surface receives when the oxygen concentration is lowered and the flame is cooled down, prolonged or detached from the fuel base. In large pool fires it is often considered that the net radiative heat flux to the fuel surface is the dominant factor compared to convection when determining the total mass loss rate, and in enclosed spaces the additional radiative heat flux feedback from other sources than the flame itself might be significant. To predict these environmental effects another simplified model has been implemented; the external radiative heat flux (radiative heat flux not sourced from the flames) divided by the fuel heat of vaporization is assumed to be directly proportional to the additional mass loss rate. This model is mostly needed in very hot enclosures and its effect has been limited in the cases used for validation in this paper. Overall the model produces accurate predictions of the mass loss rate as long as the overall flow is reasonably resolved by the model. The grid dependency has been observed to be relatively small even at coarse meshes which can be a benefit compared to more detailed models. Highlights: Non-prescribed fire sources were modeled using a modified version of FDS 6. Interaction with the fire enclosure was the main focus in the comparison. Enclosures consisted of both naturally and mechanically ventilated configurations. Resulting mass loss rates for pool fires were predicted with satisfactory accuracy.