This fire accident underscored the importance to enhance
smoke control system in order to keep a safe evacuation path free
of smoke. An appropriate fire safety design should ensure occupant
safety at first as building fire occurs (Chu and Sun, 2008).
Conventional measures against smoke in subway station largely
depend on smoke-ventilation equipment. Many scholars have
studied the air distribution during a non-accidental period.
Yuan and You (2007) made use of AIRPAK software to optimize
the ventilation for subway side-platform. To obtain the velocity
and temperature field of subway station and the optimal
ventilation mode, a three dimensional model of station was built
and the optimization design was conducted. At last, an optimal
ventilation mode of subway side-platform station was put forward.
However, the optimal mode is appropriate for non-accidental period
and the efficacy is not clear when the mode is applied during a
fire accident. For the fire accident, Yan et al. (2011) studied the air
distribution for a smoke exhaust system in operation. They found
that the wind velocity of smoke control system in a subway station
is not universal and the air could not be distributed equally.
Mathematical simulation was done under three different smoke
extract system modes. In a non-uniform smoke extract system,
the temperature was the biggest near the fire source and the uniform
smoke exhaust system was best to control smoke movement.
However, the settings of the exhaust system are not proposed.
Park et al. (2006) studied the smoke movement in a subway station
during a fire accident. To improve reliability to the results of
the simulation, the measured velocity was as the boundary condition
in numerical analysis in order to predict smoke behavior and
the measured data agreed reasonably well with simulation results.
Their study indicated that the capacity of exhausts installed had a
significant influence on the movement of heat and smoke in case of
a fire and a larger capacity was attributable to more effective control
of smoke. Rie et al. (2006) studied the vent mode for the smoke
control of subway station fire and FDS software was used to simulate
the soot concentration distributions for each smoke vent cases.
During their study, three different cases were set and the settings