Experimental studies on fire-induced buoyant smoke temperature distribution along tunnel ceiling: Fid-Bau Portal

Experimental studies on fire-induced buoyant smoke temperature distribution along tunnel ceiling

Hu, L.H. / Huo, R. / Wang, H.B. / Li, Y.Z. / Yang, R.X.

AbstractTwelve tests were conducted to study the distribution of smoke temperature along the tunnel ceiling in the one-dimensional spreading phase, two tests in a large-scale tunnel and the other ten in full scale vehicular tunnels. The fire size and the height above the floor, the tunnel section geometry and longitudinal ventilation velocity varied in these tests. Experimental results showed that when the fire size was larger, the smoke temperature below the ceiling was higher, but it decayed faster while traveling down the tunnel. The longitudinal ventilation velocity seemed to take much influence on the smoke temperature decay speed downstream. A “barrier effect” was shown for the smoke temperature distribution of the upstream back layering. The smoke temperatures measured were higher upstream than downstream before the “barrier”, and were much lower and decreased faster along the tunnel ceiling after the “barrier”. The temperature and the traveling velocity of the upstream smoke flow decreased largely when the longitudinal ventilation velocity increased a bit. The dimensionless excess smoke temperature distributions along the tunnel ceiling in all tests fell into good exponential decay. But the decay speed along the tunnel seemed to be much larger in the large-scale tunnel than that in full-scale tunnels. The measured data on ceiling jet temperature decay along the tunnel was compared with predictions of Delichatsios's model, a model built based on small-scale tests, with hydraulic diameter introduced. Results showed that Delichatsisos’ model over estimated the decay speed of ceiling jet temperature for the downstream flow. However, good agreement was achieved between the measured data and the model predictions for the upstream back layering. All the experimental data presented in this paper can be further applied for verification of numerical models, bench-scale results and building new models on ceiling jet temperature distribution.