*2.1. Model Tunnel*

The applicability of FDS (6.7) in tunnel fire calculation has been widely verified [24–27]. In this paper, the full-scale model tunnel is 600 m long, 13 m wide, and 6.5 m high. The exhaust duct is 1.7 m high and 13 m wide. The area of the exhaust vent is 14.93 m<sup>2</sup> and the distance between smoke vents is 60 m. The smoke exhaust position consists of two smoke vents with a size of 1 × 3.5 m, and there are six groups of smoke vents, as shown in Figure 1. The two smoke vents in each smoke exhaust position are always in the same state (open or closed). The vault and side walls of the tunnel are made of concrete. During the whole simulation process, the environmental temperature is 20 ◦C and the mesh boundary condition is "OPEN" [28]. The mass flow of gas and CO mass fraction at the smoke vent should also be measured. In order to monitor the temperature distribution of the tunnel vault, the thermocouple is arranged at 0.20 m below the tunnel ceiling and at intervals of 0.25 m.

**Figure 1.** Schematic diagram of model tunnel. **Figure 1.** Schematic diagram of model tunnel.

at intervals of 0.25 m.

Note that the HRR is closely related to the combustibles. The HRR caused by a car is 5–7 MW and that of medium-sized truck or bus is 20–30 MW. The HRR in a tunnel fire may be higher, such as when an oil tank truck is involved [21]. However, tunnel ventilation design must take into account the heat release rate according to the traffic flow and vehicle type after tunnel operation. Therefore, only one heat release rate is generally considered in tunnel fire ventilation design, and it is often much greater than the estimated value. In addition, China's national regulations (Guidelines for Design of Ventilation of Highway Tunnel, JTG/TD70/2-02-2014) clearly stipulate that the heat release rate of most highway tunnels in ventilation design should be 20 or 30 MW. Therefore, the power of the fire source is 30 MW, and the exhaust volume is 200–280 m3/s in this paper. The fire source (gasoline) is located at the center line of the tunnel floor, and the power of gasoline is controlled by defining the mass loss rate (MLR) in FDS. The mass loss rate of gasoline is 0.055 kg/(m²·s), and the HRR can reach 30 MW when the area of fire source is 12.6 m² [29]. The soot and CO yield are set to 0.1 and 0.05, respectively [30]. A detailed calculation scheme is shown in Table 2. Note that the HRR is closely related to the combustibles. The HRR caused by a car is 5–7 MW and that of medium-sized truck or bus is 20–30 MW. The HRR in a tunnel fire may be higher, such as when an oil tank truck is involved [21]. However, tunnel ventilation design must take into account the heat release rate according to the traffic flow and vehicle type after tunnel operation. Therefore, only one heat release rate is generally considered in tunnel fire ventilation design, and it is often much greater than the estimated value. In addition, China's national regulations (Guidelines for Design of Ventilation of Highway Tunnel, JTG/TD70/2-02-2014) clearly stipulate that the heat release rate of most highway tunnels in ventilation design should be 20 or 30 MW. Therefore, the power of the fire source is 30 MW, and the exhaust volume is 200–280 m3/s in this paper. The fire source (gasoline) is located at the center line of the tunnel floor, and the power of gasoline is controlled by defining the mass loss rate (MLR) in FDS. The mass loss rate of gasoline is 0.055 kg/(m<sup>2</sup> ·s), and the HRR can reach 30 MW when the area of fire source is 12.6 m<sup>2</sup> [29]. The soot and CO yield are set to 0.1 and 0.05, respectively [30]. A detailed calculation scheme is shown in Table 2.

same state (open or closed). The vault and side walls of the tunnel are made of concrete. During the whole simulation process, the environmental temperature is 20 °C and the mesh boundary condition is "OPEN" [28]. The mass flow of gas and CO mass fraction at the smoke vent should also be measured. In order to monitor the temperature distribution of the tunnel vault, the thermocouple is arranged at 0.20 m below the tunnel ceiling and



26–30 6 30 200, 220, 240, 260, 280 Note that a calculation model of temperature decay was also studied in this paper. If only one fire source heat release rate is considered, the temperature calculation model is not rigorous. Therefore, we studied the temperature attenuation beneath the tunnel ceil-Note that a calculation model of temperature decay was also studied in this paper. If only one fire source heat release rate is considered, the temperature calculation model is not rigorous. Therefore, we studied the temperature attenuation beneath the tunnel ceiling with HRR of 10, 20, and 30 MW.

### ing with HRR of 10, 20, and 30 MW. *2.2. Mesh Size*

Generally, mesh size is the most important aspect for numerical simulation because it determines the reliability of the numerical results. When the grid is less than 0.1*D*∗ , the numerical results are acceptable to guarantee the reliable operation of FDS [31]. The characteristic diameter *D*∗ can be calculated by:

$$D^\* = \left(\frac{Q}{\rho\_a c\_p T\_a g^{1/2}}\right)^{2/5} \tag{1}$$

where *T<sup>a</sup>* is the ambient air-temperature (K), *ρ<sup>a</sup>* is the ambient air density (kg/m<sup>3</sup> ), *c<sup>p</sup>* is the specific heat capacity of air at constant pressure (kJ/kg·K), *g* is the gravity acceleration

(m2/s), and *Q* is the heat release rate of fire source (kW). *D*<sup>∗</sup> is calculated to be 3.74 m when the HRR of fire is 30 MW, thus 0.1*D*∗ is approximately 0.374 m. In the previous reports on the use of FDS to study tunnel fires, the selection of mesh size was described in detail, and the mesh size was verified, as shown in Table 3. At present, when studying the smoke movement and temperature profile based on FDS, the commonly used grid sizes are 0.1667 and 0.20 m [30,32–36]. Since the tunnel length in this paper is 600 m, we set the mesh size as 0.20 m. Note that the HRR is the main factor affecting the characteristic diameter of the fire source. Since the research scenario in this paper is the same as the existing tunnel fire research based on FDS, we do not repeat the grid sensitivity analysis. reports on the use of FDS to study tunnel fires, the selection of mesh size was described in detail, and the mesh size was verified, as shown in Table 3. At present, when studying the smoke movement and temperature profile based on FDS, the commonly used grid sizes are 0.1667 and 0.20 m [30,32–36]. Since the tunnel length in this paper is 600 m, we set the mesh size as 0.20 m. Note that the HRR is the main factor affecting the characteristic diameter of the fire source. Since the research scenario in this paper is the same as the existing tunnel fire research based on FDS, we do not repeat the grid sensitivity analysis. **Table 3.** Details of previous model tunnel grids.

Generally, mesh size is the most important aspect for numerical simulation because

2/5

is the ambient air density (kg/m3),

is approximately 0.374 m. In the previous

**HRR** 

, the

(1)

c*<sup>p</sup>* is

is calculated to be 3.74 m

it determines the reliability of the numerical results. When the grid is less than 0.1*D*\*

numerical results are acceptable to guarantee the reliable operation of FDS [31]. The char-

<sup>=</sup> <sup>ቆ</sup> *<sup>Q</sup> ρa* c*pTag1/2*<sup>ቇ</sup>

the specific heat capacity of air at constant pressure (kJ/kg K), *g* is the gravity acceleration


**Table 3.** Details of previous model tunnel grids. **References Dimension (m** × **m)** 

where *Ta* is the ambient air-temperature (K), *ρ<sup>a</sup>*

when the HRR of fire is 30 MW, thus 0.1*D*\*

*Fire* **2022**, *5*, x FOR PEER REVIEW 4 of 17

can be calculated by:

(m2/s), and *Q* is the heat release rate of fire source (kW). *D*\*

*D*\*

*2.2. Mesh Size* 

acteristic diameter *D*\*
