ABSTRACT
One of the most frequent fire protection strategies used in conventional road tunnels is to employ jet fans to achieve longitudinal ventilation, aiming to eliminate the hazards associated with smoke back-layering. In this context, it is necessary to determine the “critical ventilation velocity” that prevents the development of toxic smoke flow upstream of the fire. Towards this end, a range of empirical correlations, aimed at estimating the critical ventilation velocity, has been proposed. This study aims to comparatively assess several such empirical correlations, using Computational Fluid Dynamics (CFD) simulations as a benchmark. Initially, the Fire Dynamics Simulator (FDS) CFD code was validated against experimental measurements obtained in a mechanically ventilated, 854 m long tunnel. Subsequently, the FDS code was used to estimate the same tunnel’s critical velocity in a series of parametric simulations, by varying the exhaust volume flow rates delivered by the jet fans, as well as the maximum Heat Release Rate (HRR) of the fire. It was found that the critical ventilation velocity generally increases with increasing fire size, until a certain HRR value, after which it becomes independent of the fire power. Predictions of nine different engineering correlations used to determine the critical ventilation velocity were then compared against the obtained CFD numerical results; the obtained discrepancies varied significantly, ranging from 1% to 47%, thus suggesting that the investigated correlations exhibit notably different levels of prediction quality.
