Citation

Woodburn PJ, Britter RE. Fire Safety J. 1996; 26(1): 63-90.

Copyright

(Copyright © 1996, Elsevier Publishing)

DOI

unavailable

PMID

unavailable

Abstract

The sensitivity of results of simulations of the area downstream of a fire in a tunnel to several factors was quantified. The factors were natural convective heat transfer, radiative heat transfer, wall roughness, downstream boundary position, slope and turbulence model. The results were most sensitive to natural convection, radiation and wall roughness. The results were not sensitive to which of the two turbulence models were used, nor to the position of the downstream boundary position. Overall uncertainties between the simulations were up to 60%, and temperatures were overpredicted in the base case by up to 75 K. The simulation carried out using a combination of factors gave maximum temperatures at the downstream end of the tunnel within 25 K of the experiments.