2.1 Introduction to PyroSim and simulated objects
To simulate the fire development process, it is necessary to select a suitable simulation model. Through the analysis of such models, an intuitive understanding of the fire spreading process can be gained, and the temperature field distribution, smoke distribution, and visibility distribution in the fire scene can be qualitatively and quantitatively analyzed [16]. Presently, the main fire simulation software are as follows: DETACT-QS, used to calculate the power of the fire source; FDS, FLUENT, and CFAST, used to calculate fire smoke distribution, and so on. The application of FLUENT and other software to fire simulations requires a strong fluid mechanics background, which renders them unsuitable for use in this study. The FDS software is specifically developed for fire simulations and is the most widely used fire simulation software owing to its ease of use.
The FDS software was developed by the Building Fire Laboratory of The National Institute of Technology and Standards to calculate the fire dynamics. The FDS software can be used to simulate and analyze the smoke spreading process [17], carry out the geometric modeling of building structures, set the monitoring position of fire information such as the wall thermal conductivity parameters, fire type, smoke temperature, and visibility, and investigate the change law of the fire parameters during the combustion process. At the same time, the simulation results can be visually viewed by Smoke View software, including temperature, Smoke concentration, Smoke speed and other parameters of the plane distribution, Smoke spreading process. As a pre-processing software for FDS, PyroSim can be used to quickly build models, set the parameters, and adjust the monitoring point settings, in addition to other functions. PyroSim has the following characteristics and capabilities:
(1) Hydrodynamic modeling. The Neville–Stokes equation can be used to quantitatively calculate thermally driven, low-velocity fluids.
(2) Combustion modeling. Different combustion reactions can be set up and simulated by setting the combustion mixture.
(3) Radiative transport. Radiative transport equations can be used to calculate the radiative transport.
(4) Unique geometry. The software allows the setting of obstacles.
(5) Boundary conditions. The thermal boundary conditions of the materials can be defined, and the combustion characteristics can be set.
Therefore, this study selected PyroSim to numerically simulate the fire scene of an assembly building construction site.
The prefabricated building talent apartment project considered in this study is designed to facilitate education and scientific research. The land area is 13,719.29 m2, the total construction area is 45,641.49 m2, the considered construction area is 3313.74 m2, the plot ratio is 4.31, and the building density is 24.15%. The project effect is shown in Fig. 2. The structural form of the project is a frame shear wall structure, and the assembled components are a composite floor slab, composite beam, external wall hanging plate, light partition wall, precast staircase, and so on, with a precast rate of 30.53%. After the installation of the precast components, the entire post -cast concrete is combined with the frame column beam and so on.
This study considered building #F02 of the Talent apartment complex as an example for establishing a safety information model. The talent apartment building consists of two high-rise buildings, the fire protection rating is One, and each floor has two evacuation stairs. The building has 13 floors and a total height of 40.95 m. The ground floor is equipped with a duty room, public laundry room, storage room, and so on. The 2nd to 13th floors consist of two-room student dormitories that can accommodate 720 students. The room space in a standard floor of the prefabricated apartment project is narrow and restricted. Hence, a fire explosion can easily be caused when fire occurs. There is no spraying and mechanical smoke exhaust system in the building, and doors and windows are not installed; therefore, natural smoke exhaust conditions were considered in this study.
2.2 Model Building
A prefabricated apartment building with a height of 3.3 m and a total of 13 floors was considered as the object to be modeled. In this study, the Revit software was used to establish the physical model of fire in a prefabricated building, and the Revit model was then imported into PyroSim. The model renderings and standard floor plans are shown in Figs. 3 and 4. Using the ratio of 1:1, the simulation model was constructed and the model area was divided into grids to simulate a fire scene. If the grid division is too large, it may not accurately calculate and describe the change of the environmental parameters in the fire scene. If the grid division is too small, the accuracy of the simulation calculation will be higher, which may result in excessively long or infeasible calculation. Therefore, by comprehensively considering the calculation accuracy, this study adopted the uniform mesh partition method, and the mesh size was set to 1 m × 1 m × 1 m.
2.3 Fire Hazard Determination
2.3.1 Analysis of factors influencing fire at prefabricated building construction site
According to relevant statistics, most human casualties during a fire are caused by the inhalation of smoke and toxic gas coma before death. When a fire breaks out, the factors that can threaten the safe evacuation of people are the smoke visibility, toxic smoke (mainly CO), and heat radiation from the fire source [18]. The CO concentration, flue gas temperature, flue gas layer height, and flue gas visibility are the main performance parameters. Therefore, the critical values of human tolerance to these parameters should be analyzed to determine whether a fire situation has reached the dangerous state.
(1) Smoke layer height
The height of the smoke layer determines whether the smoke will affect the human body. The high-temperature smoke produced in a fire will rise to the upper space of the building under the action of thermal buoyancy. When the smoke encounters the roof of the building, it accumulates and gradually thickens to form a smoke layer. As the smoke continues to rise and the smoke layer continues to thicken, the height of the smoke layer gradually decreases; when the height drops to the height of the fire, harm is caused to the personnel and the evacuation process is greatly affected. Therefore, the smoke layer should be kept above a certain height to ensure the safety of personnel and rapid evacuation to a greater extent. One of the quantitative assessment criteria is that the height of the smoke layer should satisfy the following formula in the process of personnel evacuation.
where HS is the clear height (m); HC is the critical height (m); Hp is the average height (m) and typically amounts to 1.6 m; HB is the height inside the building (m).
The average height of the investigated prefabricated apartment project was approximately 3.3 m. According to Equation 1, the critical height HC = 1.6 + 0.1 × 3.3 = 1.93 (m). Therefore, the height of 2.00 m above the ground was considered as the critical height of the smoke layer.
(2) Flue gas temperature
When a fire breaks out, hot smoke is typically produced, and people breathing the overheated air may suffer heat stroke and skin burns. Because thermal radiation data cannot be obtained directly, it is necessary to determine the critical temperature that the human body can withstand by analyzing the fire hazard conditions. The tolerance times of the human body under different conditions are listed in Table 1. The critical value of the selected temperature is 60°C.
Table 1
Limit of human body tolerance to hot air [19].
Temperature and Humidity
|
<60°C, Water saturation
|
60°C, Moisture content <1%
|
100°C, Moisture content<1%
|
Tolerance time (min)
|
>30
|
12
|
1
|
(3) CO concentration
A large amount of high-temperature smoke is produced during a fire, and the many types of combustibles in buildings, including various toxic and harmful gases and solid particles, among which the CO content is the largest and most harmful to the human body, result in complex fire smoke composition. After CO gas poisoning, carbonyl hemoglobin is produced in the blood and affects the human respiration and nerve reaction. The degrees of harm of different CO concentrations to the human body are listed in Table 2. According to the hazard assessment, the CO concentration of 500 ppm is determined as the critical value for the emergence of danger.
Table 2
Degree of harm to human body of different CO concentrations [19].
CO concentration of flue gas (ppm)
|
Time (min)
|
Degree of hazard
|
200
|
120–180
|
Mild headache and fatigue
|
400
|
60–120
|
Secondary headaches
|
800
|
45
|
Unconsciousness, vomiting
|
1600
|
20
|
Headache, dizziness, nausea
|
3200
|
10–15
|
Death
|
(4) Smoke visibility
The visibility of flue gas determines people’s capability of assessing the surrounding environment and recognizing the evacuation passage and safety exit, the perception and decision-making process of trapped personnel, the escape time, and other factors affecting the efficiency of an emergency evacuation. Low visibility results from the blocking of light by the solid particles in flue gas. In a low-visibility environment, the suspended solid particles stimulate the eyes, make it difficult for trapped people to accurately assess the surrounding environment, and reduce people’s ability to recognize the evacuation channels and safety exits, which results in perceptual decision-making mistakes and missing the optimal escape time. The Australian “Fire Engineer Guide” specifies the critical value of visibility for spaces with different size. The critical value of visibility is 10 m in a large space and 5 m in a small space [20]. The investigated prefabricated apartment project has small space; therefore, the smoke visibility index is 5 m.
2.3.2 Fire Hazard Determination Conditions
Through the analysis of the above-mentioned factors influencing the safety of personnel evacuation, the temperature, CO concentration, and visibility conditions in the fire scene should be set to ensure the safe evacuation of construction personnel. This study set the fire risk assessment conditions of the prefabricated apartment project as follows:
(1) Temperature assessment conditions: a smoke layer temperature exceeding 60°C at a height of 2.00 m above the ground is considered as a fire danger state.
(2) CO concentration assessment conditions: a CO concentration greater than 500 ppm at a height of 2.00 m above the ground is considered as a fire danger state;
(3) Visibility assessment conditions: if the visibility of the smoke layer at 2.00 m above the ground is less than 5 m, it is considered that the fire has reached the dangerous state.
2.4 Fire Scene Setup
Initial simulation environment: the flow field in the room is static, the temperature is 20°C, and the pressure is standard atmospheric pressure. A building fire has the characteristics of fast spread, difficult management, and high rescue difficulty. Because the prefabricated building is under construction, there is no automatic fire extinguishing system, smoke exhaust system, or other similar equipment, and the construction situation of each part must be analyzed simultaneously. Two ignition sources were set in the numerical simulation study. The first fire was located in the left corridor on the 2nd floor of the lower level, and the second fire was located in the left corridor on the 11th floor of the higher level. To monitor the spread of office building fire and the visibility and temperature change characteristics of the escape routes, several temperature sensors, CO concentration detectors, smoke layer height detectors, and visibility detectors were set in the fire layer and upper and lower adjacent layers. The height is 2m above the ground of this layer, as shown in Figure 5.
2.5 Analysis Of Fire Simulation Results
2.5.1 Fire smoke spread analysis
Figure 6 shows the process of simulating the spread and filling of smoke in an office building after a fire. As can be seen from the simulation results presented in Fig. 6 (a)–(f), As can be seen from the simulation results in FIG. 6 (a) ~ (f), smoke mainly collects around the fire source in the initial stage of fire combustion when the fire occurs 30 seconds. Sixty seconds after the fire has broken out, the hall on the left side of the 2nd floor and 11th floor is filled with smoke spreading rapidly upward through the stairwell and elevator shaft. At 110 s after the onset of fire, the smoke from the fire source on the 11th floor spreads to the top floor, and the smoke from the fire source on the second floor spreads to the talent apartment through the west window. At this time, the smoke does not fill the entire corridor on the second floor. As the fire develops, the smoke continues to spread, and the smoke concentration gradually increases. At t = 160 s, the east, middle, and west areas of the 2nd and 11th floors are completely filled with smoke. At 200 s after the fire occurs, the floors adjacent to the fire floor are completely filled with smoke. Three hundred seconds after the fire has broken out, the smoke continues to spread and all floors except the first floor are filled with smoke. Approximately 75% of the entire office building is filled with smoke.
2.5.2 Temperature Change Analysis
The temperature change curve and the temperature change slice diagram is drawn in Figs. 7 and 8. According to the temperature change data obtained by the temperature sensor. By analyzing the temperature change curve, the following conclusions are obtained: (1) the fire location exhibits approximately the same change trend, which reveals that, after the initial stage of fire combustion, the temperatures soars to the maximum temperature fluctuation and then declines. With the spread of the fire, the temperature began to rise after reaching the lowest point, and the temperature tended to be stable with the development of time. (2) When fire occurs, the temperature at the fire source linearly increases, which can easily cause severe threats to the life and safety of personnel close to the fire source. Hence, it is very important for people to escape the danger immediately and stay away from the fire source. (3) The temperature change trend at the exit of the stairs is the same and the temperature change is relatively slow; as the distance to the fire decreases, the temperature change becomes faster. ④ According to the above discussion, the ambient temperature limit of the human body is approximately 60°C, and the data of six sensors exceed 60°C within 350 s.
2.5.3 Co Concentration Change Analysis
Under the action of plume, the buoyancy of the flue gas increases and the flue gas rapidly rises to the ceiling. The produced flue gas jets to the ceiling and rapidly spreads to both sides of the corridor along the ceiling. The carbon monoxide concentration is maximum at the detection point farthest from the fire source, where a large amount of flue gas has accumulated. Under the positive chimney effect, hot flue gas is gradually injected into each floor over time. The flue gas in the stairwell is affected by the positive chimney effect, and the temperature of the flue gas increases; therefore, the flue gas produces greater buoyancy. Under these conditions, a large number of high temperature flue gas moves to the upper layer. However, during the movement, the flue gas is hampered by the staircase structure, and a large amount of high temperature flue gas accumulates at the stairs, which leads to a further increase in the flue gas temperature. Hence, the flue gas produces stronger buoyancy, and the temperature flue gas action is high until the gas moves to the top of the stairs. Additionally, at 11 layer detection points, the carbon monoxide concentration is higher relative to other floors, and thus more dangerous, as shown in Figs. 9 and 10. When the height of the smoke layer is smaller than the height of the human eye, the carbon monoxide concentration at all detection points does not reach the critical value during the simulation time, and the carbon monoxide is not harmful to humans.
2.5.4 Visibility Change Analysis
In the process of combustion, solid and liquid particles are produced in large numbers, which greatly affects the visibility of the fire site and produces great resistance for the emergency evacuation of personnel and the smooth extinguishing of the fire. Therefore, this study installed visibility sensors at the same positions as the temperature sensors. The visibility curve was drawn according to the sensor data. As shown in Figs. 11 and 12, in the early stage of the fire, the visibility of the floor was approximately 30 m and decreased exponentially as smoke filled the office building with the spread of fire. Owing to the impact of the fire source on the 11th floor, the visibility of the sensor decreased very early. The visibility dropped to 5 m in 50 s and reached the danger point for casualties. At approximately 400 s, the data collected by all visibility sensors reveal that the visibility at each exit was less than 5 m. The optimal escape time elapses after 400 s.