Traffic safety risk assessment method for bridge, island, tunnel and underwater intercommunication cluster engineering in operation period

Abstract

The invention relates to a bridge, island, tunnel and underwater interworking cluster engineering operation period traffic safety risk assessment method, which relates to the field of traffic risk assessment and comprises four steps of risk source determination, risk analysis, risk grade assessment and risk real-time and comprehensive comment. Constructing a risk source tree diagram based on a check table method and an autonomous investigation method in combination with an engineering design drawing and traffic flow characteristics; performing qualitative and quantitative analysis on the risk source, and determining the probability consequence level of the risk source through autonomous weighting; determining a risk source level by adopting a risk matrix according to the risk event occurrence possibility level and the consequence level; a hierarchical management and control strategy is provided for each risk special item, and the risk levels of all the special items are ensured to be reduced below the level II through a series risk model. Through multi-dimensional risk identification, qualitative and quantitative analysis combination, risk grade dynamic evaluation and targeted management and control measures, systematic evaluation and control of traffic safety risks in a complex project operation period are realized.

Description

Technical Field

[0001] This invention relates to the field of traffic risk assessment, specifically to a method for assessing traffic safety risks during the operation of bridge, island, tunnel, and underwater interchange cluster projects. Background Technology

[0002] With the rapid development of cross-river / sea passage construction in my country, bridge-tunnel cluster projects (such as the Shenzhen-Zhongshan Bridge) face multi-dimensional traffic safety risks due to their complex structures, high traffic volumes, and enclosed environments, including vehicle collisions, tunnel fires, strong winds, and severe weather. Existing risk assessment methods often focus on single structures or risk types, failing to systematically address the multi-factor coupled risks of bridge-tunnel cluster projects. Furthermore, traditional methods lack quantitative analysis of truck dynamics, fire spread characteristics, and meteorological conditions, resulting in inaccurate risk assessment results and insufficiently targeted control measures. Therefore, a comprehensive assessment method that considers multiple risk factors is urgently needed to ensure the operational safety of such projects. Summary of the Invention

[0003] Purpose of the invention: To provide a method for assessing traffic safety risks during the operation of bridge, island, tunnel, and underwater interchange cluster projects, in order to solve the aforementioned problems existing in the prior art.

[0004] Technical Solution: A method for assessing traffic safety risks during the operation of a bridge, island, tunnel, and underwater interchange cluster project, comprising the following steps:

[0005] S1. Based on the checklist method and self-investigation method, combined with engineering design drawings and traffic flow characteristics, construct a risk source tree diagram that includes the impact of vehicle collisions, tunnel fires, strong winds and severe weather.

[0006] S2. Perform qualitative and quantitative analysis on the risk sources, and determine the probability and consequence level of the risk sources through autonomous weighting;

[0007] S3. Based on the probability level and consequence level of risk events, a risk matrix is ​​used to determine the risk source level, where the risk level is divided into levels I to IV, and control measures are required for level III and above.

[0008] S4. Propose tiered control strategies for each specific risk area, including measures such as limiting cargo, limiting speed, optimizing fire protection facilities, and strengthening traffic monitoring. And ensure that the risk level of all specific areas is reduced to below Level II through a series of risk models.

[0009] In a further embodiment, the analytic hierarchy process, the Delphi method, and the fuzzy comprehensive evaluation method are used to perform qualitative analysis on the risk sources.

[0010] Quantitative analysis of the aforementioned risk sources includes:

[0011] Vehicle collision risk: Based on dynamic models and Monte Carlo simulations, the probability of rear-end collisions and side collisions on typical road sections under different cargo restriction schemes is calculated.

[0012] The risk of tunnel fires is assessed by using computational fluid dynamics simulation to analyze safety indicators such as temperature, CO concentration, and smoke layer height under different types of truck fire scenarios, and the fire risk level is evaluated in conjunction with smoke extraction plans.

[0013] To mitigate the risks of strong winds and severe weather, a vehicle dynamics model under crosswind conditions was established to calculate the safe critical speed for different wind speeds, vehicle types, and meteorological conditions. Based on the classification standards, speed limit measures were developed.

[0014] In a further embodiment, in step S1, risk source identification is performed using autonomous weight calculation to reduce subjectivity and improve identification accuracy. The specific formula is as follows:

[0015]

[0016] in, Let i be the autonomous weight. For autonomous scoring of risk sources, n represents the number of participants in autonomous weight calculation.

[0017] In a further embodiment, the vehicle collision risk analysis includes:

[0018] Rear-end collision risk is assessed by calculating the average distance between vehicles and the probability of collision based on traffic flow, vehicle type ratio, and slope, and then classifying the probability into corresponding probability ratings.

[0019] Side collision risk is assessed by simulating the beta distribution of speed difference, lane change time, and acceleration, combined with a minimum safe distance model, to evaluate the probability of side collisions in the merging and diverging zones of interchanges.

[0020] In a further embodiment, the risk analysis under strong wind conditions includes:

[0021] Establish a model of the impact of crosswinds on vehicle sideslip, tilt and yaw, and calculate the safe critical speeds of container trucks, ordinary trucks and passenger cars under different wind speeds;

[0022] Speed ​​limits or traffic restrictions will be implemented based on wind speed levels.

[0023] In a further embodiment, the risk response measures include:

[0024] Vehicle collision control measures include implementing tiered restrictions on cargo in mixed passenger and freight lanes to reduce the probability of collisions.

[0025] For tunnel fire control, extra-large trucks transporting flammable materials are prohibited from passing through, and smoke extraction plans are optimized and fire-fighting facilities are installed.

[0026] In case of severe weather, speed limits and traffic flow will be dynamically adjusted based on wind speed, visibility, or rainfall intensity. Visual guidance facilities and road administration will be activated to guide traffic.

[0027] In a further embodiment, when performing vehicle collision control, a detailed analysis of the probability of rear-end collision risk is conducted, and the types of rear-end collisions are refined into three categories: speed difference type, front and rear vehicle braking type, and acceleration difference type. The established vehicle rear-end collision risk calculation formula is then used to perform vehicle rear-end collision risk analysis.

[0028] In a further embodiment, the formula for calculating the risk of a rear-end collision includes two parts: calculation of the average distance between vehicles and calculation of a rear-end collision.

[0029] The specific formula for calculating the average vehicle distance is as follows:

[0030] Based on the traffic volume, the traffic flow rate for the mixed passenger and freight lane is:

[0031]

[0032] Where Q is the total traffic volume, P1 is the proportion of different types of vehicles in the total traffic volume of the mixed passenger and freight lane, and 0.52 is the directional coefficient.

[0033] Under normal traffic conditions at the design speed, vehicle density is calculated using the following formula:

[0034]

[0035] Where v1 is the vehicle speed;

[0036] Based on the vehicle density per unit kilometer, the average vehicle spacing can be obtained as follows:

[0037]

[0038] Wherein, n1 to n6 represent the proportions of small cargo, medium cargo, large cargo, trailer, container, and large passenger vehicle, respectively; l1 to l6 represent the lengths of small cargo, medium cargo, large cargo, extra-large cargo, container, and large passenger vehicle, respectively, with L=1000 m.

[0039] The specific formula for calculating rear-end collisions is as follows:

[0040] Speed ​​difference type:

[0041] The time required for a collision between adjacent vehicles can be calculated using the following formula:

[0042]

[0043] in, The speed difference between the front and rear vehicles;

[0044] The probability of a rear-end collision is expressed as:

[0045]

[0046] Wherein, the driver's reaction time t1=t r +t b Average distance between trucks S1, speed difference between front and rear vehicles Both the driver's reaction time t1 and the driver's reaction time t1 are considered as random variables;

[0047] Front and rear brake type:

[0048] Parking sight distance is calculated using the following formula:

[0049]

[0050] Where v1 is the speed of the vehicle in front and behind, and t1 is the driver's reaction time;

[0051] The distance between the front and rear vehicles after braking is:

[0052]

[0053] Among them, S 前停 S is the stopping sight distance of the vehicle in front. 后停 The stopping sight distance for the vehicle behind;

[0054] when When a vehicle collision is detected, v1 is assumed to be a random variable following a beta distribution. Substituting this into the above formula and using the Monte Carlo method, the probability of a rear-end collision per hour, P0, can be calculated. The probability of a rear-end collision, P, can then be calculated using the following formula:

[0055] .

[0056] In a further embodiment, when performing vehicle collision control, a detailed analysis is conducted on the risk of side collisions when vehicles change lanes. Simulation results of speed differences and accelerations under different truck proportions are obtained, and a vehicle side collision risk calculation formula is established for vehicle side collision risk analysis.

[0057] In a further embodiment, the formula for calculating the vehicle side collision risk is as follows:

[0058] After a vehicle changes lanes in an adjacent lane, the distance between the target vehicle and the vehicle behind or in front is calculated using the following formula:

[0059]

[0060] Where v1 is the target vehicle speed, v0 is the speed of the vehicle following or in front after the lane change, v1 - v0 is the speed difference between adjacent lanes, the lane change time is t, a is the target vehicle's acceleration or deceleration, d0 is the safety distance, and the probability of a side collision between the lane-changing vehicle and the target vehicle is:

[0061]

[0062] In the formula, Ω represents the failure region; assuming v1-v2 and t are random variables following a beta distribution, and a is assumed to be a bounded asymmetric random variable, substituting these into the above formula and using the Monte Carlo method to calculate the probability P0 of a vehicle side-collision during lane changing per hour, the probability P of a vehicle side-collision during lane changing during the day or night can be calculated by the following formula:

[0063] .

[0064] Beneficial Effects: This invention relates to a method for assessing traffic safety risks during the operation of bridge, island, tunnel, and underwater interchange cluster projects. It pertains to the field of traffic risk assessment and includes four steps: risk source identification, risk analysis, risk level assessment, and real-time and comprehensive risk evaluation. Based on checklist and self-investigation methods, combined with engineering design drawings and traffic flow characteristics, a risk source tree diagram is constructed. Qualitative and quantitative analyses are performed on the risk sources, and the probability and consequence levels of each risk source are determined through self-weighting. Based on the probability and consequence levels of risk events, a risk matrix is ​​used to determine the risk source level. A tiered control strategy is proposed for each specific risk, and a cascaded risk model is used to ensure that the risk level of all specific risks is reduced to below Level II. This invention achieves systematic assessment and control of traffic safety risks during the operation of complex engineering projects through multi-dimensional risk identification, a combination of qualitative and quantitative analysis, dynamic risk level assessment, and targeted control measures. Attached Figure Description

[0065] Figure 1 This is a flowchart illustrating the traffic safety risk assessment process for the Shenzhen-Zhongshan Bridge during its operational phase, as described in this invention.

[0066] Figure 2 A tree diagram for identifying traffic safety risk sources during the operation of the Shenzhen-Zhongshan Bridge as described in this invention. Detailed Implementation

[0067] In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the invention. However, it will be apparent to those skilled in the art that the invention can be practiced without one or more of these details. In other instances, certain technical features well-known in the art have not been described in order to avoid obscuring the invention.

[0068] The traffic safety risk assessment method for bridge, island, tunnel, and underwater interchange cluster projects involved in this invention mainly includes the following steps:

[0069] Step 1: Based on the checklist method and self-investigation method, combined with engineering design drawings and traffic flow characteristics, construct a risk source tree diagram that includes the impact of vehicle collisions, tunnel fires, strong winds and severe weather.

[0070] Step 2: Conduct qualitative and quantitative analysis of the risk sources, and determine the probability and consequence level of the risk sources through autonomous weighting;

[0071] Step 3: Based on the probability level and consequence level of the risk event, use a risk matrix to determine the risk source level, where the risk level is divided into levels I to IV, and control measures are required for level III and above;

[0072] Step 4: Propose tiered control strategies for each specific risk area, including measures such as limiting cargo, limiting speed, optimizing fire protection facilities, and strengthening traffic monitoring. And ensure that the risk level of all specific areas is reduced to below Level II through a series of risk models.

[0073] The analytic hierarchy process, the Delphi method, and the fuzzy comprehensive evaluation method were used to conduct qualitative analysis of the risk sources.

[0074] Quantitative analysis of the aforementioned risk sources includes:

[0075] Vehicle collision risk: Based on dynamic models and Monte Carlo simulations, the probability of rear-end collisions and side collisions on typical road sections under different cargo restriction schemes is calculated.

[0076] The risk of tunnel fires is assessed by using computational fluid dynamics simulation to analyze safety indicators such as temperature, CO concentration, and smoke layer height under different types of truck fire scenarios, and the fire risk level is evaluated in conjunction with smoke extraction plans.

[0077] To mitigate the risks of strong winds and severe weather, a vehicle dynamics model under crosswind conditions was established to calculate the safe critical speed for different wind speeds, vehicle types, and meteorological conditions. Based on the classification standards, speed limit measures were developed.

[0078] In step S1, risk sources are assessed using autonomous weight calculations to reduce subjectivity and improve identification accuracy. The specific formula is as follows:

[0079]

[0080] in, Let i be the autonomous weight. For autonomous scoring of risk sources, n represents the number of participants in autonomous weight calculation.

[0081] Rear-end collision risk is assessed by calculating the average distance between vehicles and the probability of collision based on traffic flow, vehicle type ratio, and slope, and then classifying the probability into corresponding probability ratings.

[0082] Side collision risk is assessed by simulating the beta distribution of speed difference, lane change time, and acceleration, combined with a minimum safe distance model, to evaluate the probability of side collisions in the merging and diverging zones of interchanges.

[0083] The risk analysis under strong wind conditions includes:

[0084] Establish a model of the impact of crosswinds on vehicle sideslip, tilt and yaw, and calculate the safe critical speeds of container trucks, ordinary trucks and passenger cars under different wind speeds;

[0085] Speed ​​limits or traffic restrictions will be implemented based on wind speed levels.

[0086] The risk response measures include:

[0087] Vehicle collision control measures include implementing tiered restrictions on cargo in mixed passenger and freight lanes to reduce the probability of collisions.

[0088] For tunnel fire control, extra-large trucks transporting flammable materials are prohibited from passing through, and smoke extraction plans are optimized and fire-fighting facilities are installed.

[0089] In case of severe weather, speed limits and traffic flow will be dynamically adjusted based on wind speed, visibility, or rainfall intensity. Visual guidance facilities and road administration will be activated to guide traffic.

[0090] When conducting vehicle collision control, a detailed analysis of the probability of rear-end collision risks is performed, and the types of rear-end collisions are categorized into three types: speed difference type, front and rear vehicle braking type, and acceleration difference type. The established vehicle rear-end collision risk calculation formula is then used to analyze the vehicle rear-end collision risks.

[0091] The formula for calculating the risk of a rear-end collision includes two parts: the calculation of the average distance between vehicles and the calculation of a rear-end collision.

[0092] The specific formula for calculating the average vehicle distance is as follows:

[0093] Based on the traffic volume, the traffic flow rate for the mixed passenger and freight lane is:

[0094]

[0095] Where Q is the total traffic volume, P1 is the proportion of different types of vehicles in the total traffic volume of the mixed passenger and freight lane, and 0.52 is the directional coefficient.

[0096] Under normal traffic conditions at the design speed, vehicle density is calculated using the following formula:

[0097]

[0098] Where v1 is the vehicle speed;

[0099] Based on the vehicle density per unit kilometer, the average vehicle spacing can be obtained as follows:

[0100]

[0101] Wherein, n1 to n6 represent the proportions of small cargo, medium cargo, large cargo, trailer, container, and large passenger vehicle, respectively; l1 to l6 represent the lengths of small cargo, medium cargo, large cargo, extra-large cargo, container, and large passenger vehicle, respectively, with L=1000 m.

[0102] The specific formula for calculating rear-end collisions is as follows:

[0103] Speed ​​difference type:

[0104] The time required for a collision between adjacent vehicles can be calculated using the following formula:

[0105]

[0106] in, The speed difference between the front and rear vehicles;

[0107] The probability of a rear-end collision is expressed as:

[0108]

[0109] Wherein, the driver's reaction time t1=t r +t b Average distance between trucks S1, speed difference between front and rear vehicles Both the driver's reaction time t1 and the driver's reaction time t1 are considered as random variables;

[0110] Front and rear brake type:

[0111] Parking sight distance is calculated using the following formula:

[0112]

[0113] Where v1 is the speed of the vehicle in front and behind, and t1 is the driver's reaction time;

[0114] The distance between the front and rear vehicles after braking is:

[0115]

[0116] Among them, S 前停 S is the stopping sight distance of the vehicle in front. 后停 The stopping sight distance for the vehicle behind;

[0117] when When a vehicle collision is detected, v1 is assumed to be a random variable following a beta distribution. Substituting this into the above formula and using the Monte Carlo method, the probability of a rear-end collision per hour, P0, can be calculated. The probability of a rear-end collision, P, can then be calculated using the following formula:

[0118] .

[0119] When conducting vehicle collision control, a detailed analysis of the risk of side collisions during lane changes is performed. Simulation results of speed differences and accelerations under different truck proportions are obtained, and a vehicle side collision risk calculation formula is established for vehicle side collision risk analysis.

[0120] The specific formula for calculating vehicle side collision risk is as follows:

[0121] After a vehicle changes lanes in an adjacent lane, the distance between the target vehicle and the vehicle behind or in front is calculated using the following formula:

[0122]

[0123] Where v1 is the target vehicle speed, v0 is the speed of the vehicle following or in front after the lane change, v1 - v0 is the speed difference between adjacent lanes, the lane change time is t, a is the target vehicle's acceleration or deceleration, d0 is the safety distance, and the probability of a side collision between the lane-changing vehicle and the target vehicle is:

[0124]

[0125] In the formula, Ω represents the failure region; assuming v1-v2 and t are random variables following a beta distribution, and a is assumed to be a bounded asymmetric random variable, substituting these into the above formula and using the Monte Carlo method to calculate the probability P0 of a vehicle side-collision during lane changing per hour, the probability P of a vehicle side-collision during lane changing during the day or night can be calculated by the following formula:

[0126] .

[0127] This invention addresses traffic safety during the operation of the Shenzhen-Zhongshan Bridge by identifying potential risk sources that could lead to traffic accidents, analyzing vehicle collision risks in interchanges and tunnels, vehicle fire risks in tunnels, and driving risks on the main bridge under adverse weather conditions. It also conducts a traffic safety risk assessment of the Shenzhen-Zhongshan Bridge based on the project design, traffic volume, traffic composition, vehicle power performance, and relevant regulations of the Ministry of Transport.

[0128] In a further preferred embodiment, the basic steps include: risk identification, risk assessment, risk analysis, risk control, and risk reassessment, as shown in the accompanying drawings. Figure 1 As shown.

[0129] This invention is based on a self-investigation method, employs the analytic hierarchy process (AHP), and utilizes the Delphi method and fuzzy mathematics to identify risk sources. Combined with specific risk assessments, it conducts quantitative risk analysis based on traffic flow, traffic composition, vehicle dynamic performance, and road driving simulation data. Furthermore, through quantitative risk probability analysis and qualitative risk loss assessment, and in accordance with traffic safety risk evaluation criteria, it confirms the traffic safety risk level.

[0130] According to the draft of the "Technical Specification for Safety Risk Assessment of Highway Tunnel Operation" (JTG / T 2024), the risk level of highway traffic safety can be determined based on both the probability level of a risk event and the severity level of the risk event. Different risk control measures are adopted for different risk levels.

[0131] Based on the probability of risk events and the severity of their consequences, a risk assessment matrix is ​​determined, as shown in the table below.

[0132] Table 1 Risk Matrix Classification

[0133]

[0134] The acceptance criteria for different risk levels are shown in Table 2. The specific risk assessment level is Level III, which means that the risk can be conditionally accepted or not accepted.

[0135] Table 2 Risk Level Acceptance Criteria

[0136]

[0137] The risk level assessment of traffic safety risk sources mainly adopts the self-investigation method, assigning autonomous weights. The level of each risk source is then determined through aggregation and weight calculation. To reduce the subjectivity of autonomous experience estimation and to expand the risk source survey sample as much as possible, all invited personnel are required to have extensive experience in engineering construction and traffic engineering, and to be familiar with or understand the design of the Shenzhen-Zhongshan Bridge.

[0138] Based on the independent assessment of the probability of occurrence and the severity of consequences of risk events for each risk source, the weights are calculated according to the following formula:

[0139]

[0140] —Weighted average of risk source levels;

[0141] —The weight value corresponding to the i-th autonomous position;

[0142] —The risk source level given autonomously by the i-th element;

[0143] —Number of participants in the autonomous weight calculation.

[0144] The "weighted average of the probability levels of risk events" is calculated using the above weights. ")" and "weighted average of risk event consequence levels ()" Each risk event is assigned a probability level and consequences according to the following evaluation criteria.

[0145] Risk event occurrence probability level:

[0146] Level 1: 0 ≤ V < 1.5

[0147] Level 2: 1.5 ≤ V < 2.5

[0148] Level 3: 2.5 ≤ V < 3.5

[0149] Level 4: 3.5 ≤ V < 4.5

[0150] Level 5: 4.5 ≤ V < 5

[0151] Risk event consequence level:

[0152] Generally: 0 ≤ V < 1.5

[0153] Larger: 1.5 ≤ V < 2.5

[0154] Critical: 2.5 ≤ V < 3.5

[0155] Particularly critical: 3.5 ≤ V < 4

[0156] For example, when V=1, the probability level of the risk event is level 1, and the consequence level of the risk event is moderate.

[0157] After calculating the probability and consequences of risk events, the risk source level is determined according to Table 2.

[0158] Based on existing standards, guidelines, and research findings related to traffic safety risk source identification, and in accordance with the construction drawings and change documents of the Shenzhen-Zhongshan Bridge project, a traffic safety risk identification tree diagram for the Shenzhen-Zhongshan Bridge during its operation period is constructed (see appendix). Figure 2 The study identified the main risk sources for vehicle collisions, truck fires in tunnels, strong winds on the main bridge, and heavy fog and rainstorms on the main bridge.

[0159] In a further preferred embodiment, the probability analysis of the risk of rear-end collisions is as follows:

[0160] Formula for calculating the risk of a rear-end collision:

[0161] (1) Calculation of average vehicle distance

[0162] The Shenzhen-Zhongshan Bridge is a two-way eight-lane bridge. The outer two lanes of each lane allow mixed passenger and freight traffic. Based on traffic volume, the traffic flow of the mixed passenger and freight lanes is:

[0163]

[0164] Where Q is the total traffic volume (veh / h), P1 is the proportion of different types of vehicles in the total traffic volume of the mixed passenger and freight lane, and 0.52 is the directional coefficient.

[0165] Under normal traffic conditions at the design speed, the vehicle density (veh / km) can be calculated using the following formula:

[0166]

[0167] Where v1 is the vehicle speed (km / h).

[0168] Based on the vehicle density per unit kilometer, the average vehicle spacing (distinguishing between small trucks, medium trucks, large trucks, trailers, containers, and large buses) (m) is as follows:

[0169]

[0170] Wherein, n1 to n6 represent the proportions of small trucks, medium trucks, large trucks, trailers, containers, and large buses, respectively. l1 to l6 represent the lengths (in meters) of small trucks, medium trucks, large trucks, extra-large trucks, containers, and large buses, respectively, with L=1000 m. The proportions and lengths of the vehicles are shown in Table 3.

[0171] Table 3. Percentage and Length of Different Vehicle Types

[0172]

[0173] (2) Rear-end collision calculation

[0174] Vehicle rear-end collision type 1: This type only considers the speed difference between the front and rear vehicles.

[0175] The time required for a collision between adjacent vehicles can be calculated using the following formula:

[0176]

[0177] in, The difference in speed between the front and rear vehicles (m / s).

[0178] The probability of a rear-end collision of type 1 can be expressed as:

[0179]

[0180] Wherein, the driver's reaction time t1=t r +t b Driver's judgment time t r Take 1.5s as an example, and the driver's action time t. b Take 1.0s. Average distance between trucks S1, speed difference between trucks in front and behind. Both the driver's reaction time t1 and the driver's reaction time t1 are considered as random variables.

[0181] (2) Vehicle rear-end collision type two: This type considers the braking situation of the front and rear vehicles.

[0182] The stopping sight distance (m) considering the braking of vehicles in front and behind can be calculated using the following formula:

[0183]

[0184] Where v1 represents the speed (km / h) of the vehicles in front and behind. t1 represents the driver's reaction time, typically 2.5s for the vehicle in front and 4s for the vehicle behind. The latter vehicle has 1.5s more reaction time because it needs to observe the movement of the vehicle in front before it can react. f represents the coefficient of friction. g represents the gradient, which is between 0 and 1.

[0185] The distance (m) between the front and rear vehicles after braking is:

[0186]

[0187] Among them, S 前停 S is the stopping sight distance of the vehicle in front. 后停 This refers to the stopping sight distance of the vehicle behind.

[0188] In the second type of rear-end collision, when When the time is right, it is determined to be a vehicle collision. Assuming v1 is a random variable that follows a beta distribution, and substituting it into the above formula, the Monte Carlo method can be used to calculate the probability of a rear-end collision P0 per hour. Then, the probability of a rear-end collision P during the day (12 hours) or at night (12 hours) can be calculated by the following formula.

[0189]

[0190] In a further preferred embodiment, the probability of a rear-end collision is calculated as follows:

[0191] According to the "Technical Requirements and Test Methods for Braking Systems of Commercial Vehicles and Trailers" (GB12676-2014), the coefficient of friction for concrete / asphalt pavement is taken as 0.75. Due to the complex road structure, the following typical road sections are selected for rear-end collision risk analysis: the airport interchange branching and merging points; the west artificial island interchange branching and merging points; the main tunnel section and the tunnel's uphill and downhill sections.

[0192] Two scenarios are considered for calculating the probability of rear-end collisions. Scenario 1: Inner lanes 1 and 2 for passenger cars; outer lanes 3 and 4 for mixed traffic of trucks and buses. Scenario 2: Inner lane 1 for passenger cars; inner lane 2 for buses; outer lanes 3 and 4 for mixed traffic of trucks and buses. The calculations are divided into two time periods: daytime (7:00 AM to 7:00 PM) and nighttime (7:00 PM to 7:00 AM). Considering scenarios such as no restrictions on freight, restrictions on trailers and containers, and restrictions on large trucks and medium-sized trucks, the probability of rear-end collisions on different typical road sections during the day and night in 2025 is calculated.

[0193] In 2025, Scheme 1 will restrict medium-sized freight vehicles during the day, with P1 set at 0.23 and Q at 4008 veh / h. Under this scenario, the probability of rear-end collisions on all road sections will approach 0%, as shown in the table below:

[0194] Table 4. Probability of rear-end collisions under different typical road conditions in Scheme 1

[0195]

[0196] In 2025, Option 2 will restrict medium-sized freight vehicles during the day, with P1 set at 0.19 and Q at 4008 veh / h. Under this condition, the probability of rear-end collisions on each road segment will approach 0%, as shown in the table below.

[0197] Table 5. Probability of rear-end collisions under different typical road conditions in Scheme 2

[0198]

[0199] In a further preferred embodiment, the formula for calculating the probability of a vehicle side collision during lane changing is as follows:

[0200] After a vehicle changes lanes in an adjacent lane, the distance between the target vehicle and the vehicle behind or in front can be calculated using the following formula:

[0201]

[0202] Where v1 is the target vehicle's speed (m / s), v0 is the speed of the vehicle following or preceding it after the lane change (m / s), and v1 - v0 is the speed difference between adjacent lanes. The lane change time is t. a is the target vehicle's acceleration or deceleration. d0 is the safety distance, taken as 5m. The probability of a side collision between the lane-changing vehicle and the target vehicle is:

[0203]

[0204] In the formula, Ω represents the failure region. Assuming v1-v2 and t are random variables following a beta distribution, and a is assumed to be a bounded asymmetric random variable, substituting them into the above formula and using the Monte Carlo method to calculate the probability P0 of a vehicle side collision when changing lanes per hour, the probability P of a vehicle side collision when changing lanes during the day or at night can be calculated by the following formula.

[0205] .

[0206] In a further preferred embodiment, the probability of a lane-changing side collision is calculated in conjunction with the corresponding scheme, as follows:

[0207] 1. Risk of side collisions when vehicles change lanes at airport interchanges

[0208] Considering daytime conditions, based on the predicted freight traffic volume of the Shenzhen-Zhongshan Bridge, the speed difference between the deceleration lane and the outermost lane in the diversion area of ​​the G ramp on the Shenzhen direction of the Airport Interchange ranges from (15, 20) km / h, following the mean. Standard deviation The beta distribution; the lane change time t follows the mean. Standard deviation The beta distribution is such that the lane change time t ranges from (1, 2.5) s; considering the large traffic volume at this location, the safety distance d0 is taken as 5m for this purpose.

[0209] (1) Option 1

[0210] It is calculated by dividing it into two time periods: daytime and nighttime.

[0211] During the day, restrictions are divided into three categories: those on containers and trailers, and those on large trucks and medium-sized trucks. At night, restrictions are divided into two categories: no restrictions on cargo and those on containers and trailers.

[0212] Table 6 Probability of Side Collision for Option 1

[0213]

[0214] As can be seen from Table 6, due to the higher traffic volume than the previous year, the probability of side collisions under the daytime restrictions on containers and trailers increased from close to 0 to 0.6983, while the probability of collisions under the nighttime restrictions on cargo was 0.3712, and the probability of collisions under the restrictions on containers and trailers approached 0.

[0215] (2) Option 2

[0216] It is calculated by dividing it into two time periods: daytime and nighttime.

[0217] During the day, restrictions are divided into three categories: those on containers and trailers, and those on large trucks and medium-sized trucks. At night, restrictions are divided into two categories: no restrictions on cargo and those on containers and trailers.

[0218] Table 7 Probability of Side Collision in Scheme 2

[0219]

[0220] As shown in Table 7, the probability of a side collision at the airport interchange during the day when there are no restrictions on cargo is 0.9440, while the probability at night is 0.2117. During the day when there are restrictions on containers and trailers, the probability of a side collision is 0.2179, while the probability at night approaches 0.

[0221] 2. Risk of side collisions when vehicles change lanes on the roads leading to and from the West Artificial Island

[0222] Considering daytime conditions, based on the predicted truck traffic volume ratio of the Shenzhen-Zhongshan Bridge, the speed difference between the deceleration lane and the outermost lane in the merging zone of the ramps ranges from (10, 15) km / h, following the mean. Standard deviation The beta distribution, the lane change time t follows the mean Standard deviation The beta distribution is such that the lane change time t ranges from (1, 2.5) s. Considering the high traffic volume at this location, the safety distance d0 is set to 5m.

[0223] (1) Option 1

[0224] It is calculated by dividing it into two time periods: daytime and nighttime.

[0225] During the day, restrictions can be divided into three scenarios: restrictions on trailers and containers, restrictions on large trucks, and restrictions on medium-sized trucks. At night, restrictions can be divided into two scenarios: no restrictions on cargo and restrictions on containers and trailers.

[0226] Table 8 Probability of Side Collision in Scheme 1

[0227]

[0228] As can be seen from Table 8, even when the traffic volume is greater than that of the previous year, under Scheme 1, the probability of a side collision when vehicles merge into the West Artificial Island is still close to 0.

[0229] (2) Option 2

[0230] It is calculated by dividing it into two time periods: daytime and nighttime.

[0231] During the day, restrictions can be divided into two categories: those on trailers and containers, and those on large trucks. At night, restrictions can be divided into two categories: no restrictions on cargo, and restrictions on containers and trailers.

[0232] Table 9 Probability of Side Collision in Scheme 2

[0233]

[0234] As shown in Table 9, under Scheme 2, the probability of a side collision when a vehicle merges into the West Artificial Island in 2025 under the condition of no cargo restrictions is only 0.0013.

[0235] By calculating the probabilities of rear-end collisions and side collisions, quantitative analysis results of the risks of rear-end collisions and side collisions under different cargo restriction conditions were obtained for two scenarios. Further, combining the loss levels of traffic safety accidents occurring on different typical road sections, a risk level assessment of rear-end collisions and lane-changing side collisions was conducted. Based on practical experience and the connotation of the probability of rear-end collisions and side collisions as defined in this paper, the probability levels of vehicle collisions are defined in Table 10. On this basis, combined with the risk level assessment criteria, the risk levels of rear-end collisions and side collisions on typical road sections under different cargo restriction conditions were evaluated under Scenario 1, and the evaluation results are shown in Tables 11 and 12. Under Scenario 2, the risk levels of rear-end collisions and side collisions on typical road sections under different cargo restriction conditions were evaluated, and the evaluation content and results are shown in Tables 13 and 14.

[0236] Table 10 Risk Event Probability Rating

[0237]

[0238] Table 11 Risk Assessment of Rear-End Collisions under Scheme 1 under the Condition of Restricting Medium and Large Freight Vehicles

[0239]

[0240] As can be seen from Table 11, after restricting medium and heavy goods, the risk level of rear-end collisions on typical daytime road sections has been reduced to Level I.

[0241] Table 12 Risk Assessment of Side Collision During Lane Changing in Scheme 1 under Limited Container and Trailer Vehicle Conditions

[0242]

[0243] As can be seen from Table 12, under the condition of restricting containers and trailers, the risk level of side collisions when vehicles change lanes on the airport interchange section during the day is Level II, and the risk level of the West Artificial Island is Level I.

[0244] Table 13 Risk Assessment of Rear-End Collisions under Scheme Two (Limited to Large Trucks and Above)

[0245]

[0246] As can be seen from Table 13, after the restriction on large trucks, the risk level of rear-end collisions on various road sections during the day has been reduced to Level II.

[0247] Table 14 Risk Assessment of Vehicle Lane Changing Side Collision under Option 2 for Container and Trailer Vehicle Restrictions

[0248]

[0249] As can be seen from Table 14, under the condition of limiting containers and trailers, the risk level of side collision when changing lanes is Level I.

[0250] To ensure traffic safety on the Shenzhen-Zhongshan Bridge under normal traffic conditions, this report proposes the following risk mitigation measures for Schemes 1 and 2, based on the risk assessment of rear-end collisions and side impacts, as shown in Table 15. The results indicate that different levels of cargo restriction measures are needed to reduce the risk of vehicle collisions at different times.

[0251] Table 15 Vehicle Collision Risk Mitigation Measures (Option 1)

[0252] In a further preferred embodiment, the risk assessment of truck fires inside the tunnel is as follows:

[0253] Based on a survey of parameters of common tanker trucks and heavy-duty trucks, two truck sizes were selected for analysis. Condition 1 involved an extra-large truck with a fire source area of ​​8 m × 2.5 m, and was divided into four stages. Condition 2 involved a large truck with a fire source area of ​​6 m × 2.5 m, and was divided into three stages. The detailed safety indicators for fire identification are shown in the table below.

[0254] Table 16 Fire Safety Indicators

[0255]

[0256] The longitudinal length of the fire smoke exhaust system model in this paper is 500 m, and the cross-sectional dimensions of the computational domain are taken as the actual dimensions of the tunnel. Fluent meshing is used for mesh generation. The fire numerical simulation study of the Shenzhen-Zhongshan Bridge is carried out using the fluid dynamics software Fluent. The safety of three indicators obtained based on the fire hazard judgment criteria—temperature at a height of 1.8 meters above the ground, CO volume concentration at a height of 1.8 meters above the ground, and smoke layer height—is shown in the table below.

[0257] Table 17 Safety Indicators for Temperature, Toxic Gases, and Smoke Layer Height

[0258]

[0259] The total time for detection, alarm, and pre-emptive action by personnel is 180 seconds. From the start of evacuation to the complete evacuation to a safe area, 120 seconds are required. Including the detection, alarm, and pre-emptive action time, the total evacuation time is 300 seconds. Based on the analysis in this paper, assuming no restrictions on vehicle types, the safe time for fires in extra-large trucks and large trucks does not meet the required evacuation time. Therefore, considering the resulting casualties, economic losses, and environmental losses, the risk level is assessed as Level 3, which is high risk. The risk level is conditionally acceptable, and the transport of certain hazardous flammable goods should be prohibited. Effective control measures should be implemented for permissible hazardous flammable goods transport.

[0260] Table 18 Risk Assessment of Truck Fires under the Most Unfavorable Conditions

[0261]

[0262] When the Shenzhen-Zhongshan Bridge tunnel adopts the maximum smoke extraction scheme, the probability level of flammable goods transportation risk in the tunnel is determined to be Level II. Considering the resulting casualties, economic losses, and environmental damage, the risk level is assessed as Level II, i.e., medium risk. This risk level is conditionally acceptable, but further preventative measures and strengthened tunnel safety management are required. The fire risk assessment for extra-large and large trucks using the maximum smoke extraction scheme is shown in the table below.

[0263] Table 19 Fire Risk Assessment of Trucks under Maximum Smoke Exhaust Scheme

[0264]

[0265] When a fire breaks out in a large truck transporting flammable materials inside a tunnel, and the Shenzhen-Zhongshan Bridge is using its maximum smoke extraction system, the control measures should be as follows:

[0266] Even with the tunnel's smoke extraction system operating at maximum capacity, the safe time for a fire is only 273 seconds, which is insufficient for evacuation. Therefore, the fire must be controlled within 4.55 minutes using the tunnel's fire suppression systems; otherwise, additional fire suppression equipment must be deployed to prevent the rapid spread of the fire. When extra-large trucks catch fire, they can cause not only casualties but also structural damage. Given that the safe time for evacuation is insufficient when extra-large trucks transporting flammable materials catch fire, restrictions must be imposed on their use.

[0267] When a fire breaks out in a large truck transporting flammable materials inside a tunnel, and the Shenzhen-Zhongshan Bridge is using its maximum smoke extraction scheme, the control measures should be as follows:

[0268] Since the tunnel's smoke extraction system is activated at maximum capacity, the safe time for fire evacuation is 336 seconds, which meets the requirements for personnel evacuation. Therefore, the fire only needs to be controlled within 5 to 6 minutes using the tunnel's fire-fighting facilities; otherwise, additional fire-fighting facilities must be added to prevent the fire from spreading rapidly.

[0269] In a further preferred embodiment, the risk assessment of vehicle traffic on the main bridge under severe weather conditions is as follows:

[0270] (1) Risk assessment and countermeasures for vehicle traffic on bridge surface under strong winds

[0271] Comprehensive analysis of the calculation results shows that when the wind at the bridge site reaches level 9, the calculated critical safe speed for passenger cars is 80 km / h. Standard ordinary cargo trucks are less affected by wind safety in strong winds. However, when the modified truck's sideboard height is 1.95m, it is significantly affected by level 10 winds, with a calculated critical safe speed of 62 km / h at the main bridge of the Shenzhen-Zhongshan Bridge under level 10 wind conditions. Large container trucks are less affected by strong winds when fully loaded, but their safety is significantly reduced when empty. Empty container trucks have a large frontal area and a light load, making them highly susceptible to skidding and other accidents in rainy weather. Based on quantitative analysis and a self-assessed qualitative risk assessment of risk losses, the safety risk levels of containers under strong winds are shown in Table 20.

[0272] Table 20 Safety Risk Levels for Driving Under Strong Winds

[0273]

[0274] As can be seen from Table 20, the risk level of driving on the waterway bridge surface under different wind speeds has reached Level III or above, and speed limit measures need to be taken. For details of the specific speed limit measures, please refer to Table 21. After taking the above measures, it can be seen that the safety risk level of container trucks driving on the waterway bridge surface under strong winds is less than Level III, and the safety risk level is acceptable.

[0275] Table 21 Speed ​​Limits (km / h) for Container Trucks (Category 6) on the Navigation Channel Bridge Deck During Strong Winds

[0276]

[0277] (2) Risk assessment of truck traffic on the main bridge under fog and rainstorm weather

[0278] By analyzing speed control standards and severe weather conditions under heavy fog and rainstorms, vehicle restriction measures for the Shenzhen-Zhongshan Bridge under heavy fog and rainstorm conditions were formulated. The classification criteria for typical fog and rainstorm weather conditions in the Shenzhen-Zhongshan Bridge area are listed in the table below.

[0279] Table 22 Criteria for Classifying Severe Weather Conditions

[0280]

[0281] Based on this classification criterion, recommendations are made for control measures for the bridge sections of the Shenzhen-Zhongshan Bridge under typical weather conditions.

[0282] ① Traffic control measures in conjunction with heavy fog

[0283] Under low visibility (dense fog) conditions, different traffic control measures will be implemented in the controlled area and the affected area according to the degree or level of impact, and different information content will be released, as shown in the table below.

[0284] Table 23 Recommended Traffic Control Measures for Low Visibility (Dense Fog)

[0285] ② Traffic control measures in conjunction with rainfall

[0286] Under conditions of heavy rainfall, different traffic control measures will be implemented in the controlled area and the affected area according to the degree or level of impact, and different information content will be released, as shown in the table below.

[0287] Table 24. Traffic Control Measures Recommended for Heavy Rainfall

[0288]

[0289] In a further preferred embodiment, the overall risks of the four specific risks—vehicle collision, fire safety, traffic safety under strong winds, and rain / fog weather—are grouped into a series system. This means that the occurrence of any one of these specific risks would be unacceptable for the traffic safety of the Shenzhen-Zhongshan Bridge. Based on traffic volume and composition forecasts and the current traffic design status, without risk mitigation measures, the traffic safety risk levels corresponding to the four specific risks can be described by the following four tables (Tables 25-28). The risk level assessment in Table 28 references the evaluation method for traffic safety risk levels under strong winds in Table 27.

[0290] Table 25 Vehicle Collision Risk Levels in Schemes 1 and 2

[0291]

[0292] Table 26 Fire Risk Levels in Tunnels for Trucks Carrying Flammable Goods

[0293]

[0294] Table 27 Safety Risk Levels for Driving Under Strong Winds

[0295]

[0296] Table 28 Driving Safety Risk Levels in Rainy and Foggy Weather

[0297] Table 25-28 shows that, under different circumstances, traffic safety risk levels of III or higher occurred in all four risk categories. Considering the interconnectedness of the different risk categories, every risk situation of level III or higher is unacceptable for the traffic safety of the Shenzhen-Zhongshan Bridge. Without any measures, the overall assessment of the traffic safety risk level of the Shenzhen-Zhongshan Bridge is Level III. Therefore, risk mitigation measures are needed to reduce its risk level.

[0298] Based on the research on the response measures in the four risk-specific projects, control measures were implemented for risk events of Level III and above, as detailed below:

[0299] (1) Regarding vehicle collision risk, a special study indicates that mixed passenger and freight traffic in freight lanes leads to a rapid increase in collision risk, resulting in a significant decrease in freight capacity on the Shenzhen-Zhongshan Bridge. This report proposes two suggested solutions. Referring to the countermeasures in the special study, the following measures can reduce the risk of vehicle collisions, as shown in Table 29.

[0300] Table 29 Collision Risk Levels of Vehicles After Implementing Countermeasures under Option 1

[0301]

[0302] (2) Regarding the fire safety risks in tunnels, a special study indicates that the combustion of extra-large trucks carrying flammable materials can lead to unsafe factors such as a rapid rise in temperature inside the tunnel, putting pressure on the fire-fighting capabilities of fire protection facilities. Referring to the countermeasures proposed in the special study, measures can be taken to reduce the fire safety risks, as shown in Table 30.

[0303] Table 30 Truck Fire Risk Levels under Maximum Smoke Exhaust Scheme

[0304]

[0305] (3) Regarding the safety risks of train operation under strong winds, a special study indicates that empty containers are at risk of sideslip and tilting under strong winds, necessitating speed limits to reduce the safety risk level. Referring to the countermeasures in the special study, the following measures can reduce the risk of train operation, as shown in Table 31.

[0306] Table 31 Safety Risk Levels of Vehicles Under Strong Winds After Measures Are Taken

[0307] (4) Regarding the driving safety risks in rainy and foggy weather, a special study shows that under different severe weather conditions, the level of safety risk can be reduced by taking measures such as speed limits. The risk levels of driving safety in rainy and foggy weather after taking these measures are shown in Table 32.

[0308] Table 32 Measures to be Taken for Driving Safety Risk Levels in Rainy and Foggy Weather

[0309] Based on the above assessment, it can be concluded that by implementing corresponding risk response measures for the four specific risk scenarios, the overall risk level of traffic safety for the Shenzhen-Zhongshan Bridge can be reduced from Level III to Level II, thus meeting traffic safety requirements. It is important to emphasize that the risk control measures for different specific risk scenarios are parallel and need to be implemented simultaneously.

[0310] This paper conducts a risk analysis and research on traffic safety during the operation of the Shenzhen-Zhongshan Bridge. Considering the characteristics of the construction project, the paper identifies major traffic safety risk sources during the operation of the Shenzhen-Zhongshan Bridge through investigation and identification. Quantitative analysis methods such as dynamic analysis, probabilistic analysis, and fluid numerical simulation analysis, as well as qualitative analysis methods such as self-investigation, are used to analyze and evaluate the risks to traffic safety during the operation of the Shenzhen-Zhongshan Bridge, and risk control measures are proposed.

[0311] Based on the above traffic safety risk mitigation measures and the actual effectiveness of the control measures adopted during the initial operation phase of the Shenzhen-Zhongshan Bridge, it is recommended that the Shenzhen-Zhongshan Bridge currently implement traffic control measures prohibiting Class IV and above trucks (or large trucks and above). The implementation of this measure can effectively address and significantly reduce traffic safety risks such as vehicle collisions, tunnel fires, and driving on bridge surfaces in strong winds, which is of great significance for ensuring the safe operation of the Shenzhen-Zhongshan Bridge.

[0312] As described above, although the invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the invention itself. Various changes in form and detail may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A method for assessing traffic safety risks during the operational phase of a bridge, island, tunnel, and underwater interchange cluster project, characterized by: Includes the following steps: S1. Based on the checklist method and self-investigation method, combined with engineering design drawings and traffic flow characteristics, construct a risk source tree diagram that includes the impact of vehicle collisions, tunnel fires, strong winds and severe weather. S2. Perform qualitative and quantitative analysis on the risk sources, and determine the probability and consequence level of the risk sources through autonomous weighting; S3. Based on the probability level and consequence level of risk events, a risk matrix is ​​used to determine the risk source level, where the risk level is divided into levels I to IV, and control measures are required for level III and above. S4. Propose tiered control strategies for each specific risk area, including measures such as limiting cargo, limiting speed, optimizing fire protection facilities, and strengthening traffic monitoring. And ensure that the risk level of all specific areas is reduced to below Level II through a series of risk models.

2. The method for assessing traffic safety risks during the operation of a bridge, island, tunnel, and underwater interchange cluster project according to claim 1, characterized in that: The analytic hierarchy process, the Delphi method, and the fuzzy comprehensive evaluation method were used to conduct a qualitative analysis of the risk sources. Quantitative analysis of the aforementioned risk sources includes: Vehicle collision risk: Based on dynamic models and Monte Carlo simulations, the probability of rear-end collisions and side collisions on typical road sections under different cargo restriction schemes is calculated. The risk of tunnel fires is assessed by using computational fluid dynamics simulation to analyze safety indicators such as temperature, CO concentration, and smoke layer height under different types of truck fire scenarios, and the fire risk level is evaluated in conjunction with smoke extraction plans. To mitigate the risks of strong winds and severe weather, a vehicle dynamics model under crosswind conditions was established to calculate the safe critical speed for different wind speeds, vehicle types, and meteorological conditions. Based on the classification standards, speed limit measures were developed.

3. The method for assessing traffic safety risks during the operation of a bridge, island, tunnel, and underwater interchange cluster project according to claim 1, characterized in that: In step S1, risk sources are assessed using autonomous weight calculations to reduce subjectivity and improve identification accuracy. The specific formula is as follows: in, Let i be the autonomous weight. For autonomous scoring of risk sources, n represents the number of participants in autonomous weight calculation.

4. The method for assessing traffic safety risks during the operation of a bridge, island, tunnel, and underwater interchange cluster project according to claim 1, characterized in that, The vehicle collision risk analysis includes: Rear-end collision risk is assessed by calculating the average distance between vehicles and the probability of collision based on traffic flow, vehicle type ratio, and slope, and then classifying the probability into corresponding probability ratings. Side collision risk is assessed by simulating the beta distribution of speed difference, lane change time, and acceleration, combined with a minimum safe distance model, to evaluate the probability of side collisions in the merging and diverging zones of interchanges.

5. The method for assessing traffic safety risks during the operation of a bridge, island, tunnel, and underwater interchange cluster project according to claim 1, characterized in that, The risk analysis under strong wind conditions includes: Establish a model of the impact of crosswinds on vehicle sideslip, tilt and yaw, and calculate the safe critical speeds of container trucks, ordinary trucks and passenger cars under different wind speeds; Speed ​​limits or traffic restrictions will be implemented based on wind speed levels.

6. The method for assessing traffic safety risks during the operation of a bridge, island, tunnel, and underwater interchange cluster project according to claim 1, characterized in that, The risk response measures include: Vehicle collision control measures include implementing tiered restrictions on cargo in mixed passenger and freight lanes to reduce the probability of collisions. For tunnel fire control, extra-large trucks transporting flammable materials are prohibited from passing through, and smoke extraction plans are optimized and fire-fighting facilities are installed. In case of severe weather, speed limits and traffic flow will be dynamically adjusted based on wind speed, visibility, or rainfall intensity. Visual guidance facilities and road administration will be activated to guide traffic.

7. The method for assessing traffic safety risks during the operation of a bridge, island, tunnel, and underwater interchange cluster project according to claim 6, characterized in that: When conducting vehicle collision control, a detailed analysis of the probability of rear-end collision risks is performed, and the types of rear-end collisions are categorized into three types: speed difference type, front and rear vehicle braking type, and acceleration difference type. The established vehicle rear-end collision risk calculation formula is then used to analyze the vehicle rear-end collision risks.

8. The method for assessing traffic safety risks during the operation of a bridge, island, tunnel, and underwater interchange cluster project according to claim 7, characterized in that: The formula for calculating the risk of a rear-end collision includes two parts: the calculation of the average distance between vehicles and the calculation of a rear-end collision. The specific formula for calculating the average vehicle distance is as follows: Based on the traffic volume, the traffic flow rate for the mixed passenger and freight lane is: Where Q is the total traffic volume, P1 is the proportion of different types of vehicles in the total traffic volume of the mixed passenger and freight lane, and 0.52 is the directional coefficient. Under normal traffic conditions at the design speed, vehicle density is calculated using the following formula: Where v1 is the vehicle speed; Based on the vehicle density per unit kilometer, the average vehicle spacing can be obtained as follows: Wherein, n1 to n6 represent the proportions of small cargo, medium cargo, large cargo, trailer, container, and large passenger vehicle, respectively; l1 to l6 represent the lengths of small cargo, medium cargo, large cargo, extra-large cargo, container, and large passenger vehicle, respectively, with L=1000 m. The specific formula for calculating rear-end collisions is as follows: Speed ​​difference type: The time required for a collision between adjacent vehicles can be calculated using the following formula: in, The speed difference between the front and rear vehicles; The probability of a rear-end collision is expressed as: Wherein, the driver's reaction time t1=t r +t b Average distance between trucks S1, speed difference between front and rear vehicles Both the driver's reaction time t1 and the driver's reaction time t1 are considered as random variables; Front and rear brake type: Parking sight distance is calculated using the following formula: Where v1 is the speed of the vehicle in front and behind, and t1 is the driver's reaction time; The distance between the front and rear vehicles after braking is: Among them, S 前停 S is the stopping sight distance of the vehicle in front. 后停 The stopping sight distance for the vehicle behind; when When a vehicle collision is detected, v1 is assumed to be a random variable following a beta distribution. Substituting this into the above formula and using the Monte Carlo method, the probability of a rear-end collision per hour, P0, can be calculated. The probability of a rear-end collision, P, can then be calculated using the following formula: 。 9. The method for assessing traffic safety risks during the operation of a bridge, island, tunnel, and underwater interchange cluster project according to claim 6, characterized in that: When conducting vehicle collision control, a detailed analysis of the risk of side collisions during lane changes is performed. Simulation results of speed differences and accelerations under different truck proportions are obtained, and a vehicle side collision risk calculation formula is established for vehicle side collision risk analysis.

10. The method for assessing traffic safety risks during the operation of a bridge, island, tunnel, and underwater interchange cluster project according to claim 9, characterized in that, The specific formula for calculating the vehicle side collision risk is as follows: After a vehicle changes lanes in an adjacent lane, the distance between the target vehicle and the vehicle behind or in front is calculated using the following formula: Where v1 is the target vehicle speed, v0 is the speed of the vehicle following or in front after the lane change, v1 - v0 is the speed difference between adjacent lanes, the lane change time is t, a is the target vehicle's acceleration or deceleration, d0 is the safety distance, and the probability of a side collision between the lane-changing vehicle and the target vehicle is: In the formula, Ω represents the failure region; assuming v1-v2 and t are random variables following a beta distribution, and a is assumed to be a bounded asymmetric random variable, substituting these into the above formula and using the Monte Carlo method to calculate the probability P0 of a vehicle side-collision during lane changing per hour, the probability P of a vehicle side-collision during lane changing during the day or night can be calculated by the following formula: 。

Images