CFD-Based Fire Risk Assessment and Control at the Historic Dong Wind and Rain Bridges in the Western Hunan Region: The Case of Huilong Bridge

Abstract

The Dong wind and rain bridges in western Hunan are among the most extraordinary historical buildings that express the unique culture of the Dong people, and are an unparalleled display of history, culture, aesthetics and architectural skills, but they are exposed to various disasters. However, fire poses a serious challenge to historical wind and rain bridges. This study investigated the fire risk of wind and rain bridges in terms of building materials, structural forms, fire habits, and fire loads. Fire dynamics simulator software and SketchUp modeling software were used to visualize and numerically simulate fire conditions. The survey results indicated that the fire load of the Huilong bridge was 1,794,561 MJ. After a fire, the central pavilion underwent a flashover at 200 s. By 600 s, the maximum temperature of the bridge was 1200 °C, and the entire bridge was in flashover condition. Furthermore, targeted fire-mitigation strategies were proposed according to the architectural characteristics and cultural environment of the wind and rain bridges in terms of the following four components: automatic fire-extinguishing equipment, fireproof coatings, suspended ceiling, and skylights. The effectiveness of the fire-resistance performance of the strategies was also evaluated. The fireproof coating measures were the most effective, and the bridge sections delayed reaching the flashover state by 40–80 s. The reported results can help reduce the fire-related risks of wind and rain bridges and protect our historical heritage more effectively and efficiently. Furthermore, this study can serve as a reference for other historic wooden structures to develop appropriate mitigation strategies.

1. Introduction

History has shaped the world of nations’ heritage monuments [1]. Heritage buildings that have withstood the test of time represent the achievements of past eras and are an important part of the culture [2,3]. It is the preservation of historic buildings or the history that we can interpret from cultural practices connected to those historic buildings [4]. Among the various risk factors that threaten historic building sites, fire risk can have the most devastating impact [5,6,7]. Some historic buildings are lost every year due to fire damage, especially wooden-frame structures [8,9,10,11]. Wind and rain bridges are one of the most representative historical buildings of the Dong people in Hunan [12], and some have been listed as national key cultural relics (Figure 1). In addition to the traffic function, the wind and rain bridge is also a temporary shelter from the wind and rain and is the main place to hold social activities. Most of the wind and rain bridges have a wooden frame through the column and tie construction, and the timber mortise and tenon structure is used between the bridge body and the piers. The wind and rain bridge is representative of the architectural culture of the Dong minority [13]. However, 12,319 fire incidents occurred in Hunan Province during the period of 2004–2013 [14]. The extensive use of timber and the lack of scientific fire-prevention measures make the historical Dong rain and wind bridges a serious fire hazard (Figure 2). Therefore, it is crucial to assess the fire risk of Dong wind and rain bridges in Hunan Province, study fire development and spread patterns, and develop suitable fire-prevention strategies.

Numerous studies have been conducted on the issue of historical buildings facing fire risks. Currently, research on fires in historic buildings focuses on fire risk assessment [16,17,18] and disaster risk management [19]. Many researchers have analyzed and evaluated the fire risk of historic buildings through field studies. Shan et al. [4] investigated the key issues in heritage building conservation and explored the management of heritage building conservation by using China as an example. Salazar et al. [5] reviewed existing indicator-based fire risk assessment methods and proposed 22 fire vulnerability indicators related to heritage assets, categorizing them into four categories in order to assess the fire risk level of existing buildings. Yuan et al. [20] investigated the fire hazards of folk culture villages and assessed the fire safety by performing a case study on Tangjia Village in China. Marrion [8] studied the information related to why hazards develop into disasters and proposed a risk-oriented approach to address these particularly relevant hazards to fire. Romão and Bertolin [21] explored the gaps in knowledge and practice in the management of disaster risks in cultural heritage, selected topics related to disaster risk management and cultural heritage, and analyzed the objectives of risk protection in cultural heritage and historical centers for future research. In addition, the analytical hierarchy approach is an important method for assessing fire risk. Ibrahim et al. [22] employed the analytical hierarchy approach to assess the fire risk of heritage buildings in Malaysia and formulated a fire-risk assessment methodology based on the criteria and attributes of heritage buildings. Naziris et al. [23] proposed a model based on a hierarchical analysis process to assess the overall level of fire safety of a structure and validated the effectiveness of the model in optimizing fire upgrades by performing a case study at the Monastery of Mount Asos in Simonos Petra. Furthermore, the visualization of the fire process in historical buildings using Pyrosim software is an important method for numerical simulation. Huai et al. [24] analyzed the degree of fire risk in historic Chinese wooden-framed buildings using a historic temple in Beijing as a case study and verified the accuracy of the fire dynamics simulator (FDS) software for wooden-framed buildings by performing extensive tests.

Fire-risk studies of historic buildings mainly utilize surveys to propose a scientific and optimal assessment framework and target fire-protection measures to mitigate fire risk [25,26,27,28,29,30]. Numerical simulations are performed mainly to reduce the impact of fire hazards on historic buildings [31,32,33]. FDS software tools are widely used in fire simulation studies and constitute one of the most prominent research methods for assessing fire risk [34,35,36,37,38,39,40]. Pyrosim software is also widely used in fire-risk studies of wooden-framed buildings; the information from the SketchUp model is directly inputted into the Pyrosim software, which allows accurate numerical simulation and visual analysis of fire combustion in buildings [41]. Therefore, in the current study, we used SketchUp software to model the Dong wind and rain bridges in Hunan Province and combined it with Pyrosim software to simulate the fire conditions of the wind and rain bridges.

Due to the unique traditions, activities, and ceremonies involving fire, the western Hunan region is the area most severely affected by fires in Hunan Province. According to the China fire protection yearbook [14], the number of fire incidents that occurred in the village area of western Hunan Province between 2003 and 2013 accounted for 25.93% of the fire incidents in Hunan Province. Wood is the predominant building material of the Dong wind and rain bridges in the western Hunan region, and wood can be easily ignited [42,43]. Furthermore, residents conduct many traditional fire-related activities on wind and rain bridges every year [44], and in the event of a fire, the surrounding mountains, forests, and villages are engulfed in flames, causing serious fire accidents. This also causes irreparable damage to the historical wind and rain bridges. Therefore, it is crucial to assess the fire risk and improve the fire performance of historical wind and rain bridges, where the risk assessment includes components of hazard (i.e., habits, rituals, ceremonies, etc.) and components of vulnerability (i.e., building materials, structural configuration, fire loads). Thus, in this study, the fire risk of Dong historical wind and rain bridges was investigated, in addition to the fire factors (smoke temperature, visibility, CO₂ concentration, CO concentration). Targeted fire mitigation strategies were developed to reduce the fire risk of the wind and rain bridges.

2. Methodology

We used a combination of a qualitative field survey and a quantitative software simulation. The study included three main elements: a field survey of fire hazards of Dong wind and rain bridges, a software simulation of fire hazards of wind and rain bridges, and the proposal of fire-mitigation strategies for wind and rain bridges. First, the structural form and architectural characteristics, the moisture content of the materials, and the local traditions, activities, and ceremonies involving fire and fire loads of the representative wind and rain bridges were investigated. The fire risk of the wind and rain bridges in the south Dong region was analyzed, and typical wind and rain bridges were selected for measurement and mapping. Second, the fire burning and spreading patterns of the wind and rain bridges were simulated and analyzed using the FDS software. Lastly, four fire-mitigation strategies, namely automatic fire-extinguishing equipment, fireproof coating, suspended ceiling, and skylights were proposed considering the architectural characteristics and cultural environment of the wind and rain bridges, and the effectiveness of the mitigation strategies was verified (Figure 3).

2.1. Field Survey

Most of the Dong wind and rain bridges were constructed by the Qing Dynasty and are unique public buildings that represent the culture of the Dong people [12]. There are approximately 330 Dong wind and rain bridges in China, and more than 30 of them are listed as key cultural relics at provincial and national levels. The Dong people in Hunan Province are mainly located in the western part of Hunan, which is divided into South Dong and North Dong based on the language [45] (Figure 4). Due to the differences in history, culture, economy, and geographical location, the Dong minority is mainly distributed in the South Dong region, and the historical wind and rain bridges in Hunan are unique historical buildings of the Dong people [46]. Therefore, in this study, the fire risk of the wind and rain bridges of the South Dong people in the western Hunan region was investigated.

2.1.1. Features and Structures of a Dong Wind and Rain Bridge

In this study, Huilong bridge, located in Pingtan village, Tongdao county, Hunan Province, China [13] (Figure 5), a representative wind and rain bridge of South Dong in the Hunan region, was selected as the research object. The Huilong Bridge is part of China’s sixth batch of key protected cultural relics. It was built in 1761 over the Pingtan River and is an arc-shaped bridge 63.01-m long and 3.86-m wide. The bridge is equipped with three pavilions, a three-story roof with heavy eaves, and hexagonal roofs. The larger pavilion in the middle has a shrine underneath for Guan Yu, considered the god of fire by the local Dong residents. In this study, the characteristics and structure of the Huilong bridge were investigated.

2.1.2. Moisture Content of Wind and Rain Bridge Materials

Wood is the most commonly used building material in ancient wind and rain bridges. Many of the Dong wind and rain bridges were constructed 100 years ago and have been subjected to wind and rain for a long time. The moisture content of the materials is very different from that of common construction wood. The moisture content of building materials can affect fire combustion. In this study, a Biaozhi high-precision wood moisture tester was used to analyze the materials of the wind and rain bridges of the South Dong people in the Hunan region, including columns, purlins, beams, bridge deck planks, and railings. The measurement points selected for determining the wood moisture content are depicted in Figure 6.

2.1.3. Traditions, Activities, and Ceremonies Involving Fire

In addition to fires caused by natural causes such as lightning strikes and spontaneous combustion, approximately 80% of fires in villages with traditional wooden-framed structures are caused by human activities [14]. The inhabitants of the south Dong region in Hunan Province have unique fire practices and cultures that are different from those of other regions in China. In addition to fires caused by unextinguished cigarettes, the daily fire habits of residents and fire-related rituals during festivals pose significant fire risks and often lead to fire incidents [48]. In this study, the unique fire cultural practices of the Dong people were investigated, including the fire practices carried out on the wind and rain bridges and the daily and cultural fire practices carried out on the houses to mitigate the risk of fire.

2.1.4. Fire Loads of Wind and Rain Bridges

The wind and rain bridge is built with a large amount of wood. At the same time, there are large structural voids in the form of wood members (Figure 7), which will cause serious fires if the wood frame is ignited under well-ventilated conditions. In order to formulate targeted mitigation measures based on the fire burning of the wind and rain bridge, the Hui Long bridge was selected for fire load statistics. Fire loads consist of fixed and mobile fire loads. In this study, the fire load was based on the measurement results of field surveys, the density of fir wood was 440 kg/m³, the calorific value of wood combustion was 18.4 MJ/kg [41], and the fire load density calculation formula was based on the code for fire safety of steel structures in buildings [49].

2.2. Software Simulations

2.2.1. Simulation Model

Pyrosim software was used to model the fire situation of the storm bridges, and the Huilong bridge was selected as a case study. First, a model of the Huilong bridge was created using SketchUp software, and the output of DXF modeling of SketchUp software was then used as the input of Pyrosim software. Finally, the burning of Huilong bridge was simulated by setting the parameters on Pyrosim (Figure 8). Based on the actual measurements of the Huilong bridge, the dimensions of the simulated calculation area were established as 64.55 m × 9.75 m × 8.63 m, and the volume of the calculation area was 5431.4 m³. According to the user manual of the Pyrosim software, the grid size was set as 0.25. The grid size and the number of grids were set as 0.3 m × 0.3 m × 0.3 m and 215 × 33 × 29, respectively, with a total of 205,755 grids.

Large-eddy simulation was performed because the simulation process was low-Mach [50]. The values of Smagorinsky coefficient Cs, turbulent Prandtl number Prt, and turbulent Schmidt number Sct were 0.2, 0.5, and 0.5, respectively [51]. The ratio of the diameter of the fire source characteristic to the mesh size was 6.5658 according to the equations of the meshing method in the Pyrosim software user manual, and the simulation results were consistent with the results of the mesh independence test.

To simplify the data-collection process, the following model setup and assumptions were employed [24,52]:

• Cylindrical columns were replaced with square columns of the same volume.
• All combustible materials were assumed to be involved in the burning and were completely burned.
• For determining the heat release rate of wood combustion, the moisture content of the wood was considered, but heat loss due to vaporization was not considered.

2.2.2. Simulation Parameters

1.

Point of origin and size of the fire

Based on the cause of the fire in the wind and rain bridge ascertained from the field survey, the fire source was set under the display table of Guan Sheng Di hall in the Huilong bridge. The fire source was selected as type t², which represents a fast-spreading fire. The heat release rate of the fire source was 8 MW [53], and the fire area was 1 m². The boundary conditions were consistent with the external environment, and the ambient temperature was set as 20 °C. Considering the worst-case scenario, all doors and windows of the Huilong bridge were kept open throughout the simulation. The heat release rate of firewood was 160 kW/m², and the ignition point was 260 °C [53]. The fire simulation time was set as 800 s.

2.

Slice and measurement point parameters

The simulation slice settings employed for Huilong bridge are illustrated in Figure 9. The slices were placed on the main axes (x and y axes) to observe the spread of smoke and the variations in temperature in the upper space inside the Huilong bridge. The lateral slice height was set as z = 1.6 m, which is the same as the characteristic height of an adult’s eye from the ground. Longitudinal slices were set at a depth of entry x = 5.5 m, 17 m, 29 m, 42.5 m, 56 m, i.e., and the slices were set at the central part of the three bridge pavilions and at the middle of the pavilions. Gas-phase equipment detectors were installed in the center of each slice to better visualize the dynamic variations in the gas temperature, CO concentration, and CO₂ concentration. The detectors were located at the center of the longitudinal slices and the bridge crossing surfaces. The locations of the measurement points are shown in Figure 8b.

2.2.3. Conditions for Flashover

Flashover is an important parameter for determining the combustion process of a fire accident and marks the sudden increase in the fire temperature; and if there is sufficient air, the entire building will become one large burning mass. The critical conditions for flashover differ for different fire situations. Based on the fire studies conducted by Cheng [54], Chow [55], and Huai et al. [24], in the current study, the critical conditions for flashover ignition included three aspects: the heat flux on the ground of Huilong bridge was subjected to 20 kW/m², the temperature below the ceiling was 600 °C, and the burning rate exceeded 40 g/s. Satisfying any one of these conditions can be considered a flashover accident.

2.3. Mitigation Strategies

In this study, tailored fire mitigation strategies were formulated to address the fire risk and combustion patterns of wind and rain bridges in terms of four components: automatic fire-extinguishing equipment, fireproof coating, suspended ceiling, and skylights. Automatic fire extinguishing equipment is an important way to prevent disasters in modern buildings. The wind and rain bridge is close to the river, so the wet fire extinguishing method was chosen. The fire extinguishing nozzles are placed in the center of the ceiling of the bridge pavilion and in the middle of the ceiling between the two pavilions, for a total of five (Figure 10a). In addition, fireproof coating is an important way for materials to be flame retardant. The fireproof coatings of organophosphorus ring flame retardants or phosphorus-nitrogen flame retardants are proposed, and they are intumescent transparent fireproofing coatings with enhanced fire resistance without seriously affecting the traditional characteristics of the bridge. The roof structure of wind and rain bridges has a large amount of wood and is the most characteristic building component. Based on the fire-protection measures employed for the roof structure of the Chinese Palace Museum, the installation of a glass ceiling was proposed to reduce the damage to the bridge by fire. The ceiling material was selected as Class A fireproof glass with a fire resistance limit of 3 h, which not only can reduce the fire damage to the wooden frame of the bridge roof, but can also protect the visitors’ sight lines from being affected. The location of the ceiling was 200 mm below the roof timber frame, covering the entire bridge roof timber frame (Figure 10b). In addition, high-temperature smoke gathers on the roof. As such, the installation of skylights was proposed to accelerate the diffusion of smoke to the exterior of the bridge and improve visibility. Only a 1.2 m × 1.2 m hole in the middle roof of the bridge pavilion was designed to increase smoke dispersion (Figure 10c). The parameters of the materials recommended for fire control are presented in

Table 1. Parameters of the materials recommended for fire control.

Material	Density kg/m3	Specific Heat Capacity kJ/(kg·K)	Thermal Conductivity W/(m·K)
Fireproof glass	2500	0.84	0.75
Fireproof board	1086	0.8	0.109
Fireproof coating	634	0.9	0.167

.

The fire spread of Huilong bridge under different mitigation strategies was simulated for a quantitative visual evaluation of the mitigation strategies. The effectiveness of the mitigation strategies was also verified by comparing the time to flashover ignition for the main activity area (the bridge pavilion, the bridge head, the bridge tail) and for the entire Huilong bridge with and without the mitigation strategies by delaying the flashover ignition time with a fire optimization enhancement factor P, which can be defined as follows:

$$\mathrm{P}=\frac{t_i-t_0}{t_0}$$

(1)

where $t_i$ is the time for the building with the i-th fire protection measure to reach the flashover state, and $t_0$ is the time for the original building to reach the flashover state.

3. Results and Discussion

3.1. Survey Result

3.1.1. Features and Structure of Dong Wind and Rain Bridges

The support structure of the Huilong bridge includes a cantilevered beam extension structure and a stacked beam extension structure, and the bridge deck building has a pierced wooden frame. The bridge deck has 22 wooden-framed units in total, and each wooden-framed unit comprises four columns and three spans. The roofing material is small green tiles, and the bridge pier is built of stone. The parameters of the construction materials of Huilong bridge and the materials recommended for fire control are presented in

Table 2. Parameters of the construction materials commonly used for fire-risk mitigation.

Material	Density kg/m3	Specific Heat Capacity kJ/(kg·K)	Thermal Conductivity W/(m·K)
Fir wood	500	2.52	0.108
Small green tile	2800	0.92	0.76
Fireproof curtain	38.3	3.528	0.125
Stone	2000	0.5	1.5
Paper	300	1.9	0.1

.

3.1.2. Moisture Content of the Construction Materials Used for Wind and Rain Bridges

The results of the moisture content measurement for each part of the Dong wind and rain bridges are presented in

Table 3. Moisture content of each part of typical wind and rain bridges.

Number	Wind and Rain Bridge	Test Point 1	Test Point 2	Test Point 3	Test Point 4	Test Point 5	Test Point 6	Test Point 7	Test Point 8
1	Huilong	11	10.5	10	10	9.5	9	9	9.5
2	Yongfu	11.5	11.2	9.5	8.7	8.2	7	7.5	6.5
3	Yongding	9	9.5	9	8.5	8.5	8.3	7.8	7.5
4	Huifu	9	9	10	9.5	10	10.5	9	9.5

. The average moisture content of the wood used in the bridges was 11% (range: 5–12%), which is in the range of the moisture content of “all-dry wood”. The moisture content of freshly cut lumber is generally around 60%. Due to outdoor ventilation and solar radiation, the moisture content of wood gradually decreases to 12–18%. Dong wind and rain bridges have been exposed to years of wind and sun; thus, the moisture content of the wood is much lower than that of wood in typical modern wooden-framed buildings, and the ignition point is much lower and therefore more prone to fire accidents.

3.1.3. Traditions, Activities, and Ceremonies Involving Fire

1.

Fire cultural practices performed on wind and rain bridges

The fire cultural practices performed on wind and rain bridges are shown in Figure 11a. Wind and rain bridges are used not only for transportation but also as a place of sacrifice, feast, and entertainment. Wind and rain bridges play the role of a temple and a bridge in one; large wind and rain bridges have at least one site set up for animal sacrifice. In the Dong wind and rain bridges constructed by the Qing Dynasty, each pavilion can be used as an independent place of sacrifice; the larger ones are called temples, and the smaller ones are shrines. Dong wind and rain bridges also show the influence of Han culture, and the objects of sacrifice are mostly gods in Han culture, such as Guandi, Kui Xing, and Wen Chang Jun. The sacrificial firepit at the shrine is constantly burning throughout the year, which can easily cause a fire accident.

2.

Fire cultural practices of the Dong people

The fire cultural practices of the Dong people are shown in Figure 11b,c. The firepit is the most important room of the Dong people; the central area of the room is used for daily public activities such as family gatherings, meetings, and rituals. In ancient times, the fire in the firepit room was extinguished only once a year during the Spring Festival. Residents use the fire pit for warmth and cooking. The main building material of Dong dwellings is wood, and the building houses numerous flammable objects, including wooden floors, wooden furniture, and wooden walls. As such, the fire of the fire pit can easily ignite the objects inside and cause a major fire accident.

Fire plays an essential role in many sacrifice activities of the Dong people, such as the Huapao Festival, Dragon Lantern Dance, Sacrifice Festival, and HeLong Feast. The dwellings of the Southern Dong people are made of wood, and the act of using fire during festivals can easily cause a fire accident.

3.1.4. Fire Loads of Wind and Rain Bridges

The fixed and movable fire load survey results of Huilong bridge are presented in Table 1. The survey results revealed that the fixed and movable fire loads are 1,794,561 and 62,762 MJ, respectively. According to the fire load density calculation formula [56], the total fire load density is 8011.43 MJ/m² (

Table 4. Fire load value of Huilong bridge.

Components	Volume (m3)	Quality (kg)	Heat of Combustion (MJ)
Wooden columns	11.8	5900	105,905
Purlin	10.0	5000	89,750
Beam	18.2	9100	163,345
Wooden board under the bridge	29.5	14,750	164,763
Wooden partition on the bridge	17.9	8950	160,653
Wooden deck of the bridge	29.9	14,950	268,353
Wooden railing	0.6	300	5385
Shrine	6.2	3100	55,645
Pyramidal roof	10.4	5200	93,340
Wooden extension beam	69.6	34,800	624,660
Wood shelf	0.968	410	7124
Display desk	7.56	410	55,638
Total	212.628	102,870	1,794,561

). The fire load density of the designs assumed by the fire design verification method of the New Zealand Code [57], the design value of the Eurocode standard [58], and modern office buildings [59] are 400, 780, and 420 MJ/m², respectively, which is 20.03, 10.27, and 19.07 times that of the Huilong bridge.

3.2. Simulation Results

3.2.1. Spread of Fire

The fire-spread situation of Huilong bridge at different moments is depicted in Figure 12. After the fire incident occurred, the smoke quickly spread from the central pavilion to both ends of the bridge. Smoke filled the central pavilion at 200 s. At 400 s, the fire spread to both ends of the bridge, and at 600 s, the entire bridge was engulfed in flames. Smoke collected mainly above the bridge and spread in all directions.

Variations in parameters with time are shown in Figure 13. Within 200 s after the fire started, the central bridge pavilion was the main area where the temperature exceeded 600 °C. At 300–500 s, the temperature of the parts on both sides of the central pavilion gradually increased. At 800 s, the temperature of all parts of Huilong bridge exceeded 600 °C. The trends of CO, CO₂, and visibility over time were consistent with the temperature variations. Influenced by the south wind, the CO concentration and CO₂ concentration on the north side of Huilong bridge were higher than those on the south side, and the visibility on the north side was lower.

3.2.2. Variations in the Parameters at Each Measurement Point

The variations in the parameters with time at each measurement point are depicted in Figure 14. Temperature variations with time results revealed that temperature at measurement points C and B increased the fastest, and the highest temperature exceeded 1000 °C. The temperature at measurement points A and D initially increased gradually and then started to increase rapidly at 600 s and reached 300 °C at 800 s. The results of visibility over time revealed that the visibility at measurement point C decreased considerably at 200 s and was 0 at 500 s. The visibility at measurement point B decreased rapidly after 320 s and reached 0 at 500 s. The visibility of measurement point D fluctuated from 320 to 600 s and decreased rapidly to 0 at 600 s. The trends of CO and CO₂ concentrations over time were consistent for all the measurement points. The trend of CO and CO₂ concentrations at each measurement point was consistent, and the trend and magnitude of the increase in the CO and CO₂ concentrations at measurement points A and D were substantially greater than those at measurement points A and D. The CO and CO₂ concentrations at measurement points A and D increased rapidly from 400 to 760 s and reached the maximum value at 760 s.

4. Mitigation Strategies

The wind and rain bridge studied here is the most representative historical building in the south Dong area and has the architectural feature of being narrow and long. The wind and rain bridge was constructed using a large amount of firewood, which has a high fire load density, and in case of fire, the fire lasts for a long time and causes serious damage. Once a fire occurs, because the wind and rain bridge is far from the village, it is difficult for residents to find it in time for rescue attempts, which can cause serious or even irreparable damage to historic buildings. In addition, the fire is very likely to spread from the wind and rain bridge to the surrounding mountains, which can result in serious fire accidents. Therefore, formulating a scientific and reasonable fire-mitigation strategy for wind and rain bridges is crucial.

4.1. Flashover Ignition Time and Fire Optimization Enhancement Factor P

The fire simulations of Huilong bridge under different mitigation strategies are shown in Figure A1, Figure A2, Figure A3, Figure A4, Figure A5, Figure A6, Figure A7 and Figure A8 in the Appendix A. The statistics of the fire-spread parameters of Huilong bridge with different mitigation strategies are presented in

Table 5. Statistics of the fire spread parameters under different mitigation strategies.

Num	Mitigation Strategies	Time to Reach the Flashover of Main Activity Sites (s)				CO2 Concentration of Main Activity Sites (ppm)
		Bridge Pavilion	Bridge Head	Bridge Tail	Overall	Bridge Pavilion	Bridgehead	Bridge Tail	Overall
1	Original	540	780	820	820	80,000	80,000	80,000	80,000
2	Fireproof coating	580	830	900	900	40,000	50,000	40,000	40,000
3	Automatic fire extinguishing equipment	580	850	880	880	80,000	80,000	80,000	80,000
4	Skylight	540	800	800	800	40,000	40,000	40,000	40,000
5	Ceiling	420	650	650	650	80,000	90,000	80,000	80,000

. The bridge with automatic fire-extinguishing equipment reached the flashover state at 880 s, a delay of 60 s compared to the time for the original bridge. The central pavilion reached the flashover state at 580 s, with a delay of 40 s compared to the time for the original bridge, and the bridgehead reached flashover at 850 s, 180 s earlier than that for the original bridge. The central pavilion reached the flashover state at 580 s, a delay of 40 s compared with the time for the original bridge; the bridgehead reached the flashover state at 830 s, a delay of 50 s compared with the time of the original bridge. When the ceiling was set in the pavilion part, the dragon bridge reached the flashover state at 650 s, 170 s earlier than the original time; the central pavilion reached the flashover state at 420 s, 120 s earlier than the original time; and the bridgehead reached the flashover state at 600 s. When the skylight was installed on the roof, the bridge did not reach flashover at the end of the bridge in 850 s; the central pavilion reached flashover in 540 s, 0 s later than the original time; and the bridge reached flashover in 800 s, 20 s later than the original time.

The improvement in the fire-mitigation performance of the Huilong bridge under different mitigation strategies is shown in

Table 6. Fire optimization enhancement factor P under different mitigation strategies.

Num	Mitigation Strategies	Bridge Pavilion	Bridgehead	Bridge Tail	Overall
2	Fireproof coating	0.07	0.05	>0.1	>0.1
3	Automatic fire extinguishing equipment	0.07	0.09	0.07	0.07
4	Skylight	0	0.03	>0.04	>0.04
5	Ceiling	−0.22	−0.15	−0.22	−0.21

. The p-values for Huilong bridge with automatic fire-extinguishing equipment, fireproof coating, suspended ceiling, and skylights were 0.07, >0.1, −0.21, and >0.04, respectively. The fireproofing performance of the bridge improved the most when fireproof coating measures were implemented, and the time taken for each part of the bridge to reach the flashover state was delayed by 40–80 s. The p-value of each part of the bridge with suspended ceiling measures was negative, and the suspended ceiling measures would reduce the fire performance of the bridge. The p-values of each part of the bridge with automatic fire extinguishing equipment and skylight measures were positive, and these two measures could mitigate the fire, but the effect was not as good as that of fireproof coating.

Implementing ceiling measures aggravated the burning of the bridge. The time taken for each part of the bridge with a suspended ceiling to reach the flashover state decreased because the ceiling prevented the high-temperature flue gas from diffusing to the outside of the bridge, thereby increasing the concentration of high-temperature flue gas and intensifying the combustion. The measures of automatic fire extinguishing equipment and skylights could also mitigate the fire, but the effect is not as good as that of fireproof coating. Therefore, we recommend fireproof coating as the preferred fire protection measure for wind and rain bridges.

4.2. Combustion Parameters for Different Mitigation Strategies

The comparison analysis of the combustion parameters at the measurement points of Huilong bridge under different fire-mitigation strategies is shown in Figure 15. The difference between the temperature at each measurement point and the original one was not significant for the ceiling measures. The temperature at each measurement point with fireproof coating was significantly lower than that of the original one. Moreover, the visibility at each fireproof coating measurement point was the highest. In addition, the CO₂ and CO concentrations at each measurement point under different mitigation measures were consistent in magnitude. The CO₂ and CO concentration fluctuations were the highest, and the mean values were the worst for the ceiling measures. The CO₂ and CO concentrations at each measurement point when fireproof coating and automatic fire-extinguishing measures were implemented were significantly lower than the original ones.

Taking into account all parameters before and after the implementation of fire prevention strategies, the order of effectiveness of the mitigation strategy was as follows: fireproof coating > automatic fire-extinguishing equipment > suspended ceiling > skylights. The implementation of suspended ceilings can aggravate the fire because the installation of suspended ceilings reduces the internal space of the bridge, resulting in a large collection of smoke inside of it. Further verification of the ceiling treatment is required. Fireproof paint was found to be the most effective fire-prevention measure for wind and rain bridges.

5. Conclusions

In this study, a computational fluid dynamics-based fire-risk assessment of historic Dong wind and rain bridges was conducted. Important potential fire-hazard factors were identified with respect to the structural form and materials of the Dong wind and rain bridge and the fire habits of local residents. A visual and quantitative fire-simulation analysis of the wind and rain bridges was conducted using the FDS and SketchUp modeling software. Four fire-mitigation strategies were proposed and validated for the fire risks encountered in the wind and rain bridges. The Dong historical wind and rain bridges pose a serious fire risk, and the measures of using fireproof paint, automatic fire-extinguishing equipment, and skylights can reduce the fire risk. The conclusions are summarized as follows:

• The results of the survey revealed that historic Dong wind and rain bridges are at risk of serious fires. Wood is the main building material used in wind and rain bridges, and the wood has been exposed to the wind and sun for a long time and has an average moisture content of 11%, which makes it highly prone to fire accidents. The fire habits of the Dong people (i.e., fire pits, sacrificial activities, and other fire-related customs held at the wind and rain bridges) increase the fire risk. The typical fire load of a wind and rain bridge is 1,794,561 MJ, and in case of fire, the surrounding forests and wooden dwellings are likely to be engulfed in fire, resulting in a serious fire accident.
• The simulation results showed that the historic Huilong bridge in the Hunan region is vulnerable to fire damage. After a fire, the central pavilion underwent a flashover at 200 s. By 600 s, the maximum temperature of the bridge was 1200 °C, and the entire bridge was in flashover condition. The high-temperature smoke spread upward, and the wooden-framed roof of the bridge was found to be highly vulnerable to damage by fire. Because the wind and rain bridge is far from the village, it is difficult for local residents to find it in time for rescue, which will cause serious or even irreparable damage to the historical bridge.
• Four fire-mitigation strategies were proposed by considering the characteristics of the storm bridges: automatic fire-extinguishing equipment, fireproof coating, skylights, and suspended ceilings. The mitigation strategies were verified by simulation. The order of effectiveness of the mitigation strategies was as follows: fireproof coating > automatic fire-extinguishing equipment > suspended ceiling > skylight. We recommend prioritizing the use of a fireproof coating to enhance the fire resistance of the wind and rain bridges without causing damage to the traditional features. Automatic fire-extinguishing equipment and skylight measures can also reduce the fire risk of the storm bridges; however, the effect is not as good as that of fireproof coating. Ceiling measures are not recommended because they increase the fire risk of the wind and rain bridges.

There are some shortcomings in this study. The Huilong bridge in the Hunan region was selected for the study as the representative bridge for fire-risk simulation. However, we only conducted a comparative study of a single fire control strategy and did not explore a comprehensive strategy. In addition, wind and rain bridges are also present in Guangxi Province, China, and differ from those in the Hunan region in terms of structure, form, local culture, and habits. The fire-simulation analysis conducted in this study may not be exactly the same for cases where the fire types differ greatly, but the fire-risk assessment methods and fire-prevention strategies are applicable. In addition, the specific construction and parameters of the fire-protection measures must be analyzed experimentally. These shortcomings will be investigated in-depth in future studies.