A method for assessing the fire resilience of an operating highway tunnel

Abstract

This invention discloses a method for assessing the fire resilience of operational highway tunnels. The invention constructs a five-dimensional index system encompassing prevention, resistance, resilience, fire sensitivity, and fire severity. Based on a three-level impact index under the system reliability index, a multi-state Bayesian network model is constructed. The prior probability of the root node is determined according to tunnel design specifications and measured data. Fuzzy comprehensive evaluation and the Leaky Noisy-OR model are used to determine the conditional probability of non-root nodes. System reliability is calculated through forward reasoning, the largest causal chain is determined through backward reasoning, and key influencing factors are identified through mutual information sensitivity analysis. The weights of the fire sensitivity and severity indices are determined and quantified using the analytic hierarchy process (AHP). The system reliability, fire sensitivity coefficient, and fire severity coefficient are integrated to obtain the fire resilience value and resilience level. This invention achieves a comprehensive assessment of tunnel fire resilience and precise location of weak points, providing technical support for tunnel operation safety management.

Description

Technical Field

[0001] This invention belongs to the field of tunnel operation safety technology, specifically a method for assessing the fire resilience of operating highway tunnels. Background Technology

[0002] Against the backdrop of rapid economic and social development, transportation engineering construction has been continuously improved, and a large number of highway tunnel facilities have been put into use. The semi-enclosed civil engineering structure of tunnels makes evacuation and rescue more difficult than in open road sections. Once a fire occurs in a tunnel, it can easily lead to large-scale casualties.

[0003] In response to tunnel fires, the government has invested heavily in improving the fire prevention and disaster mitigation capabilities of highway tunnels. However, due to the complexity and severity of tunnel fires, current methods for assessing the fire safety of highway tunnels are too narrow in scope and cannot fully cover all the factors influencing tunnel fires. The emergence of resilience theory provides a new direction for infrastructure disaster prevention and control.

[0004] Resilience assessment systems and methods have been applied in urban areas, hydropower networks, and major infrastructure projects. In the transportation infrastructure sector, resilience assessments have largely focused on rail transit and highway networks, with relatively few studies targeting individual engineering facilities. Highway tunnels, as a crucial means of overcoming obstacles such as mountains and rivers, play a vital role in the development of transportation networks. Therefore, developing a fire resilience assessment index system for operational highway tunnels is essential for improving their fire prevention and disaster mitigation capabilities, and has significant economic and social benefits. Summary of the Invention

[0005] This invention addresses the shortcomings of existing technologies by providing a method for assessing the fire resilience of operating highway tunnels.

[0006] The present invention provides a method for assessing the fire resilience of operating highway tunnels, comprising the following steps:

[0007] A fire resilience assessment index system for operational highway tunnels is constructed. The index system includes system reliability indicators that characterize the tunnel's fire prevention, resistance, and recovery capabilities, as well as fire sensitivity and fire severity indicators that characterize the probability and severity of tunnel fires.

[0008] A multi-state Bayesian network model is constructed based on the aforementioned index system. Each of the three-level impact indicators under the system reliability index is taken as a network node and a causal relationship is established. The prior probability of each root node is determined according to the tunnel design specifications and measured state data. The conditional probability of each non-root node is determined based on fuzzy comprehensive evaluation and uncertainty reasoning model.

[0009] Forward reasoning is performed on the Bayesian network model to calculate the reliability of the tunnel fire system, backward reasoning is performed to identify the key causal chain leading to system reliability failure, and sensitivity analysis is performed to determine the sensitivity of each third-level impact index to system reliability.

[0010] The weights of each of the three levels of indicators in the fire sensitivity index and fire severity index are determined based on the analytic hierarchy process. The actual passage parameters and design parameters of the tunnel are combined to perform quantitative scoring, and the fire sensitivity coefficient and fire severity coefficient are obtained respectively.

[0011] By integrating the system reliability, fire sensitivity coefficient, and fire severity coefficient, the fire resilience value of the operating highway tunnel is obtained.

[0012] The beneficial effects of this invention are:

[0013] This invention constructs a five-dimensional resilience assessment index system covering the entire life cycle of tunnel fires from "prevention to resistance to recovery," incorporating fire sensitivity and severity into a unified framework. This overcomes the shortcomings of existing assessment methods, which are limited in dimension and scope, and achieves a systematic quantitative characterization of the fire resilience of operating highway tunnels.

[0014] A multi-state Bayesian network model was used to establish causal relationships among 14 third-level influence indicators. The conditional probabilities of nodes were calculated by combining fuzzy comprehensive evaluation and Leaky Noisy-OR model. This effectively addressed the problem of expert cognitive uncertainty and nonlinear coupling between factors, making probabilistic reasoning both mathematically rigorous and engineering robust.

[0015] By using forward propagation to calculate system reliability, backward propagation to determine the largest cause chain, and mutual information sensitivity analysis to rank key factors, a closed-loop diagnostic mechanism of "macro rating - reverse tracing - priority intervention" is formed. This mechanism can provide an overall resilience level and accurately locate weak links, significantly improving the engineering guidance value of the assessment results.

[0016] The system innovatively integrates system reliability with fire sensitivity coefficient and fire severity coefficient, so that the resilience value can simultaneously reflect the inherent disaster resistance capacity of the disaster-bearing body and the intensity of external disturbances in a specific fire scenario. This enables adaptive assessment of different tunnels and different fire scenarios, avoiding the limitations of the traditional static assessment which is "one-size-fits-all". Attached Figure Description

[0017] Figure 1: Fire resilience profile of operating highway tunnels;

[0018] Figure 2: Fire resilience index system for operating highway tunnels;

[0019] Figure 3: Flowchart of fire resilience assessment method for operating highway tunnels;

[0020] Figure 4: Reliability BN model of fire toughness system for highway tunnels;

[0021] Figure 5: Forward propagation analysis diagram;

[0022] Figure 6 Backpropagation analysis diagram;

[0023] Figure 7 Sensitivity analysis chart. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to a case study of an operating highway tunnel project.

[0025] This application focuses on the fire prevention and disaster mitigation capabilities of operational highway tunnels, taking the entire fire process as its object, including prevention before a fire occurs, resistance during a fire, and recovery capabilities after a fire. Through a resilience management approach of "prevention-resistance-recovery" in three stages, and combining sensitive factors such as the tunnel's own design parameters and traffic parameters as well as the severity of the fire, it proposes an assessment index system and method for the fire resilience of operational highway tunnels, providing more comprehensive and systematic technical support for the operation and management of highway tunnels.

[0026] This invention provides a method for assessing the fire resilience of operating highway tunnels, such as... Figure 3 As shown, the method includes the following steps:

[0027] First, an evaluation index system for fire resilience of operating highway tunnels is constructed. The index system includes system reliability indicators that characterize the tunnel's fire prevention, resistance and recovery capabilities, as well as fire sensitivity indicators and fire severity indicators that characterize the probability and severity of tunnel fires.

[0028] Secondly, a multi-state Bayesian network model is constructed based on the aforementioned index system. Each of the three-level impact indicators under the system reliability index is taken as a network node and a causal relationship is established. The prior probability of each root node is determined based on the tunnel design specifications and measured state data. The conditional probability of each non-root node is determined based on fuzzy comprehensive evaluation and uncertainty reasoning model.

[0029] Subsequently, forward reasoning is performed on the Bayesian network model to calculate the reliability of the tunnel fire system, and backward reasoning is performed to identify the key cause chain leading to system reliability failure. Sensitivity analysis is then performed to determine the sensitivity of each tertiary impact index to system reliability. Furthermore, the weights of each tertiary index in the fire sensitivity index and fire severity index are determined based on the analytic hierarchy process (AHP). Quantitative scoring is then performed by combining the actual tunnel traffic parameters with the design parameters to obtain the fire sensitivity coefficient and fire severity coefficient, respectively.

[0030] Finally, the system reliability, fire sensitivity coefficient, and fire severity coefficient are integrated to obtain the fire toughness value of the operating highway tunnel, and the toughness level is determined based on the fire toughness value, so as to take targeted management and maintenance measures.

[0031] Furthermore, in the process of constructing the aforementioned indicator system, such as Figure 1 As shown, the SARSI method was used to characterize the fire resilience of operational highway tunnels from five aspects: scientific validity (S), availability (A), representativeness (R), stability (S), and independence (I). This was combined with literature, case studies, expert surveys, and the 4R characteristics of resilience. Three key capabilities were identified: prevention, resistance, and recovery, based on robustness, redundancy, resource availability, and timeliness. The ultimate effect is reflected in controlling operational performance loss, reducing destructive consequences, and shortening recovery time. Furthermore, when analyzing tunnel fire resilience, tunnel traffic design parameters and actual traffic conditions must be considered, treating them as unified indicators of tunnel fire sensitivity that influence fire resilience. On the other hand, the severity and destructiveness of fire accidents must also be taken into account, treating them as indicators of tunnel fire severity that influence fire resilience. Based on literature, case studies, and expert survey data, 21 factors influencing the fire resilience of operational highway tunnels were identified.

[0032] The system reliability indicators include preventive capability, resistance capability, and resilience capability. Among them, the preventive capability characterizes the tunnel fire monitoring, early warning, and pre-emptive control capabilities, including three levels of impact indicators: monitoring and early warning facilities, traffic control measures for hazardous chemical vehicles, and tunnel flow restriction measures.

[0033] The resistance characterizes the ability of engineering facilities to withstand and ensure the evacuation of personnel during a tunnel fire, including seven Level 3 impact indicators: ventilation and smoke extraction facilities, fire protection facilities, lighting facilities, communication facilities, power supply and distribution facilities, traffic control and guidance facilities, civil engineering structure conditions, and fire protection design.

[0034] The resilience characterizes the emergency rescue and post-disaster recovery capabilities after a tunnel fire, including four three-level impact indicators: rescue and escape facilities, emergency response time, emergency management measures, and rectification measures.

[0035] In addition, the fire sensitivity indicators include five third-level influencing indicators: tunnel length, number of lanes, maximum tunnel gradient, average traffic flow, and proportion of large vehicles. These indicators mainly consider objective factors that affect the probability of fire and the degree of fire damage from the perspective of the tunnel itself.

[0036] The fire severity index includes two three-level impact indicators: fire heat release rate and fire duration, which take into account the destructiveness and persistence of the fire from an environmental perspective.

[0037] Therefore, a resilience index comprising three levels is established, see... Figure 2The first level of indicators is the fire resilience of operating highway tunnels; the second level of indicators are prevention capability, resistance capability, resilience capability, fire sensitivity, and fire severity; the third level of indicators are the 21 key influencing factors selected above.

[0038] Among these, prevention, resistance, and resilience indicators primarily relate to the tunnel operation and maintenance management system under human intervention conditions. They are related to the tunnel's resilience and stability, and are used to assess the tunnel's ability to prevent and resist fires and recover from disasters, reflecting system reliability. Fire sensitivity mainly considers objective factors affecting the probability and severity of fires from the tunnel's own perspective, including tunnel traffic design parameters and actual traffic conditions. Fire severity considers the destructiveness and persistence of fires from an environmental perspective. Fire sensitivity and fire severity affect the tunnel's system reliability, and conversely, tunnel system reliability also affects fire sensitivity and fire severity.

[0039] When constructing a multi-state Bayesian network model, based on Bayesian network theory, through literature review and expert investigation, the preventive force, resistance, and resilience of the system reliability indicators are integrated into a Bayesian model (BN) with a clear causal relationship. For example... Figure 4 As shown, the model contains 18 nodes, each with two state parameters: "0" and "1". "0" indicates that the node has not failed (not in failure), and "1" indicates that the node has failed (in failure).

[0040] The 18 nodes specifically include: one primary evaluation indicator node, representing system reliability (D). f The three secondary assessment indicator nodes represent preventive capacity (P) and preventive capacity (P). f ), resistance (I) f ) and resilience (R f The 14 level-3 evaluation indicator nodes represent 14 influencing factors on system reliability, covering types such as facilities and equipment, organization and management, and technology.

[0041] When obtaining the prior probabilities of the root nodes, a scoring method for 14 tertiary indicators encompassing prevention, resistance, and resilience was established based on the Highway Tunnel Design Specifications (JTG3370.1-2018, JTGD70 / 2-2014) and relevant fire risk assessment guidelines (JTT52 / 13-2023). On-site surveys were conducted on the tunnel to be assessed to obtain detailed tunnel information; relevant data that were unavailable or difficult to obtain were assessed subjectively through expert scoring. The 14 tertiary indicators of the tunnel were scored using the above methods, and the scores were divided by 100 to convert them into probabilistic form. This probabilistic form represents the probability of non-failure of each root node (tertiary indicator) in the Bayesian network model during a fire, i.e., the prior probability value of each root node. Taking an operating highway tunnel as an example, the scoring methods, scoring standards, and scoring results for each tertiary indicator are shown in Tables 1-1 and 1-2.

[0042] Table 1-1 Scores of each of the three levels of system reliability indicators in fire resilience (items 1-8)

[0043] Table 1-2 Scores of each of the three levels of system reliability indicators in fire resilience (items 9-14)

[0044] When obtaining the conditional probabilities of non-root nodes, Wickens' seven-level evaluation theory was used to determine the fuzzy language and fuzzy numbers for each evaluation level. Based on the Noisy-OR model, which describes the intrinsic relationship between variables (parent nodes) and their influencing variables (child nodes), an expert questionnaire survey was conducted. According to the survey results, the dependent uncertainty ordered weighted average operator was used, and the similarity between the fuzzy evaluation number of each expert and the average fuzzy evaluation number of all experts was calculated using the fuzzy comprehensive evaluation method (FCEM) to determine the expert weights. The comprehensive expert fuzzy evaluation number was then calculated. The mean area method was used for defuzzification to obtain the conditional probabilities of relevant non-root nodes.

[0045] Based on this, using the Leaky Noisy-OR model, a default node is introduced to represent all other accidental, unimportant, or unconsidered factors. The conditional probability values ​​of each non-root node for different state combinations of each influencing factor are calculated, forming a complete conditional probability table. The conditional probabilities of the non-root nodes for preventative force, resistance, resilience, and system reliability are shown in Tables 2 to 5, respectively.

[0046] Table 2. Non-root node conditional probability of fire resilience prevention capability

[0047]

[0048] Note: 0 indicates that the node is not invalid, and 1 indicates that the node is invalid.

[0049] Table 3-1 Conditional Probability Table of Fire Resilience Resistance (Non-Root Node) (Data Set 1) Table 3-2 Conditional Probability Table of Fire Resilience Resistance for Non-Root Nodes (Data Set 2)

[0050] Table 4. Non-root node conditional probability table of fire resilience recovery.

[0051]

[0052] Table 5. Non-root node conditional probability table of fire resilience system reliability

[0053]

[0054] Using the constructed Bayesian network model and node probability table, forward propagation analysis was performed using Netica software to calculate the reliability (D) of the fire resilience system of the operating highway tunnel. f According to calculations, the reliability (D) of a certain operating highway tunnel system f The value is 90.1.

[0055] Based on the analysis of the positive propagation results (such as...) Figure 5 As shown in the figure, the factors with the highest probability of system reliability failure in the highway tunnel relying on this project under fire conditions were identified, mainly monitoring and early warning facilities, traffic control measures for hazardous chemical vehicles, and emergency response time, with failure probabilities of 35.0%, 40.0%, and 76.5%, respectively. Backpropagation analysis was then performed (e.g., Figure 6 As shown in the figure, the system reliability node is set to the failure state, and the posterior probability of each parent node is calculated by backpropagation. The cause path with the highest posterior probability is determined as the largest cause chain, namely "monitoring and early warning facilities → prevention force → system reliability under fire conditions".

[0056] The "Sensitivity to findings" function in the "Network" toolbar of Netica software was used to obtain the mutual information between various influencing factors and the reliability of the fire system. This information was used to determine the sensitivity of each influencing factor to the system reliability, and the sensitivity of each level-three influencing indicator was ranked accordingly to determine the optimization priority of each influencing factor. Based on 14 nodes (see...),... Figure 7 The sensitivity analysis results of the system reliability under tunnel fire conditions show that the five influencing factors, namely monitoring and early warning facilities, fire protection facilities, emergency response time, ventilation and smoke exhaust facilities, and emergency management measures, have a high degree of influence on the system reliability under tunnel fire conditions. This indicates that targeted intervention and optimization of these five influencing factors can effectively improve the system reliability under tunnel fire conditions.

[0057] Fire sensitivity and fire severity significantly influence system reliability under tunnel fire conditions, thereby affecting tunnel fire resilience. To address fire sensitivity and severity, an expert questionnaire was first developed to assess the weights of fire sensitivity and severity indicators. The questionnaire results were then analyzed, and the Analytic Hierarchy Process (AHP) was used to determine the weights of each of the three levels of indicators in both fire sensitivity and severity. Subsequently, based on relevant standards or literature, each of the three levels of indicators in fire sensitivity and severity was classified. Finally, based on the classification of each level of indicator, quantitative values ​​were assigned to each level of indicator under different classification standards. The weights, classification standards, and quantitative assignment results of the fire sensitivity and fire severity indicators are shown in Tables 6 and 7, respectively.

[0058] Table 6. Weights and Scores of Fire Sensitivity Indicators

[0059]

[0060] Table 7. Weights and Scores of Fire Severity Indicators

[0061]

[0062] According to the survey, a certain operating highway tunnel is 3616m long, with four lanes in both directions, a maximum gradient of 2.5%, and an average daily traffic flow of about 400-500 pcu / (h·ln), with large vehicles accounting for about 40%-50%. The fire occurred during the evening rush hour, with a traffic flow greater than 500 pcu / (h·ln). The source of the fire was a heavy semi-trailer carrying leather, and the fire lasted for about 3 hours. Based on the above data, the scoring results of each of the three levels of indicators included in fire sensitivity and fire severity are shown in Tables 6 and 7. Using equations (1) and (2), the fire sensitivity coefficient (S) of the tunnel is obtained respectively. f ) and fire severity coefficient (G) f ):

[0063] (1)

[0064] In the formula: S f R is the fire sensitivity coefficient; n is the number of indicators; fi The actual scores for the three-level indicators (i = 1, 2, 3, ...); ω fi For each level of indicator, a single weight is assigned (i=1, 2, 3, ...).

[0065] (2)

[0066] In the formula: G f R is the fire severity coefficient; n is the number of indicators; tiThe actual scores for the three-level indicators (i = 1, 2, 3, ...); ω ti For each level of indicator, a single weight is assigned (i=1, 2, 3, ...).

[0067] Calculations show that the fire sensitivity coefficient of the tunnel supporting the project is 0.8925, and the fire severity coefficient is 0.85.

[0068] Furthermore, using equation (3), the system reliability value (D) is... f ) and fire sensitivity coefficient (S) f ) and fire severity coefficient (G) f Multiply by , and obtain the fire toughness value (R) of the operating highway tunnel:

[0069] (3)

[0070] In the formula, R is the fire toughness value of the operating highway tunnel, and S f G f D f These represent the tunnel fire sensitivity index, fire severity coefficient, and system reliability, respectively.

[0071] The calculated fire toughness value (R) of the highway tunnel under this project is 68.35.

[0072] Based on R, the toughness level is determined. According to the fire toughness values ​​[0,50], (50,70], (70,85], and (85,100], it is divided into four categories: extremely low, low, medium, and high toughness, as shown in Table 8.

[0073] Table 8 Classification of Fire Resilience Levels

[0074]

[0075] According to the fire toughness classification standard in Table 8, the fire toughness of this operating highway tunnel is classified as low, indicating that the tunnel's ability to resist fire disturbance is generally limited, and it will require a relatively long time to recover after being damaged by fire. The repair process is complex and requires reinforcement or renovation. This result is caused by the high severity of the fire in the tunnel, which caused extensive damage to the tunnel structure and electromechanical facilities, significantly affecting the tunnel's ability to resist fire.

[0076] Based on the aforementioned fire resilience values ​​and their classification results, combined with the results of forward and reverse propagation analysis and sensitivity analysis, targeted measures will be taken to manage and maintain the tunnel in order to improve its fire resilience. For example, regarding the maximum causal chain identified by the reverse propagation analysis ("monitoring and early warning facilities → prevention capabilities → system reliability"), and the highly sensitive influencing factors identified by the sensitivity analysis, priority will be given to strengthening the maintenance and upgrading of monitoring and early warning facilities, improving the configuration of fire protection facilities, shortening emergency response time, optimizing the ventilation and smoke extraction system, and strengthening emergency management measures. This will scientifically and rationally improve the system reliability under tunnel fire conditions, thereby enhancing the overall fire resilience level.

[0077] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method of operating a highway tunnel fire resilience assessment, characterized by, Includes the following steps: A fire resilience assessment index system for operational highway tunnels is constructed. The index system includes system reliability indicators that characterize the tunnel's fire prevention, resistance, and recovery capabilities, as well as fire sensitivity and fire severity indicators that characterize the probability and severity of tunnel fires. A multi-state Bayesian network model is constructed based on the aforementioned index system. Each of the three-level impact indicators under the system reliability index is taken as a network node and a causal relationship is established. The prior probability of each root node is determined according to the tunnel design specifications and measured state data. The conditional probability of each non-root node is determined based on fuzzy comprehensive evaluation and uncertainty reasoning model. Forward reasoning is performed on the Bayesian network model to calculate the reliability of the tunnel fire system, backward reasoning is performed to identify the key causal chain leading to system reliability failure, and sensitivity analysis is performed to determine the sensitivity of each third-level impact index to system reliability. The weights of each of the three levels of indicators in the fire sensitivity index and fire severity index are determined based on the analytic hierarchy process. The actual passage parameters and design parameters of the tunnel are combined to perform quantitative scoring, and the fire sensitivity coefficient and fire severity coefficient are obtained respectively. By integrating the system reliability, fire sensitivity coefficient, and fire severity coefficient, the fire resilience value of the operating highway tunnel is obtained.

2. The method of claim 1, wherein, The system reliability indicators include prevention capability, resistance capability, and resilience capability. The prevention capability represents the ability to monitor, warn, and control tunnel fires in advance. The resistance capability represents the ability of engineering facilities to withstand and evacuate personnel when a tunnel fire occurs. The resilience capability represents the ability to provide emergency rescue and post-disaster repair after a tunnel fire.

3. The method of claim 2, wherein, The preventive force includes at least one Level 3 impact indicator from monitoring and early warning facilities, traffic control measures for hazardous chemical vehicles, and tunnel flow restriction measures; the resistance force includes at least one Level 3 impact indicator from ventilation and smoke extraction facilities, fire protection facilities, lighting facilities, communication facilities, power supply and distribution facilities, traffic control and guidance facilities, civil engineering structural conditions, and fire protection design; the resilience force includes at least one Level 3 impact indicator from rescue and escape facilities, emergency response time, emergency management measures, and rectification measures.

4. The method according to claim 2 or 3, characterized in that, The determination of the prior probability of each root node based on tunnel design specifications and measured state data includes: scoring each of the three-level impact indicators under the system reliability index according to highway tunnel design specifications and fire risk assessment guidelines, converting the score values ​​into probability form, and using them as the prior probability of the corresponding root node.

5. The method of claim 1, wherein, The determination of the conditional probability of each non-root node based on fuzzy comprehensive evaluation and uncertainty reasoning model includes: constructing an expert survey questionnaire based on the Noisy-OR model, processing the questionnaire results using the fuzzy comprehensive evaluation method to obtain the initial conditional probability, and introducing the Leaky Noisy-OR model to calculate the conditional probability value of each non-root node under different combinations of parent node states.

6. The method of claim 1, wherein, The process of performing reverse reasoning to determine the key causal chain leading to system reliability failure includes: setting system reliability nodes as failure states, backpropagating to calculate the posterior probability of each parent node, and determining the causal path with the highest posterior probability as the largest causal chain.

7. The method according to claim 1 or 6, characterized in that, The sensitivity analysis to determine the sensitivity of each tertiary impact indicator to system reliability includes: calculating the mutual information between each tertiary impact indicator and system reliability, and ranking the sensitivity of each tertiary impact indicator based on the mutual information magnitude.

8. The method of claim 1, wherein, The three-level indicators of the fire sensitivity index include at least one of the following: tunnel length, number of lanes, maximum tunnel gradient, average traffic flow, and proportion of large vehicles; the three-level indicators of the fire severity index include at least one of the following: fire heat release rate and fire duration.

9. The method of claim 1, wherein, The integration of system reliability, fire sensitivity coefficient, and fire severity coefficient includes calculating the fire toughness value R of the operating highway tunnel according to the following formula: where S f , G f , D f represent fire sensitivity coefficient, fire severity coefficient and system reliability, respectively.

10. The method according to claim 1 or 9, characterized in that, It also includes determining the toughness level based on the fire toughness value: the fire toughness value is divided into four levels, namely, extremely low toughness, low toughness, medium toughness and high toughness, according to the range of [0,50], (50,70], (70,85] and (85,100], respectively. The corresponding toughness level is determined according to the range to which the calculated fire toughness value belongs.

Images