Study on longitudinal temperature distribution of fire-induced ceiling flow in tunnels with different sectional coefficients

• A sectional coefficient is introduced to describe the tunnel cross section.
• A dimensionless relationship between the smoke temperature and HRR is proposed.
• The prediction models for smoke temperature distribution is proposed.
• The CFD simulation result is verified by the 1/10 small-scale experiments.

This paper proposes two prediction models for the maximum smoke temperature rise and the temperature distribution in tunnel fires, in which sectional coefficient ζζ was introduced to describe geometrical characteristic of the tunnel section. At first, the theoretical analysis was conducted. The dimensionless maximum smoke temperature rise was deduced by applying the dimensional analysis method while the smoke temperature exponential decay law was proposed based on the one-dimensional theory. Then, CFD simulations were conducted in nine tunnels with different cross sectional shapes by Fire Dynamics Simulator, version 5.5. With the ‘numerical experiments’, two prediction models for the maximum smoke temperature and the smoke temperature distribution were obtained. Meanwhile, complementary experiments were conducted in a 1/10 scale tunnel in order to provide a verification. The experiment results show a good agreement with the numerical simulations. Moreover, the proposed prediction models were compared with the prediction models proposed by Kurioka model and Li model.