Road and bridge safety assessment method and system based on big data analysis

Abstract

The invention discloses a road and bridge safety assessment method and system based on big data analysis, and relates to the technical field of road and bridge safety assessment, and the method comprises the steps: collecting and aligning road and bridge structure monitoring, traffic flow behavior, environmental load and operation and maintenance event data, forming an alignment data set, building a digital twinborn model based on the alignment data set, extracting scene features, and carrying out the construction of a digital twinborn model; the method comprises the following steps of: aligning a data set with scene features, performing traffic flow and structure response coupling analysis by utilizing the aligned data set and the scene features, generating an operation risk index, and calculating a safety assessment result according to the operation risk index, and combining technologies of multi-source data fusion, digital twin modeling and traffic flow and structure coupling analysis. A comprehensive, dynamic and intelligent solution is provided for road and bridge safety management, the operation safety and traffic management efficiency of the road and bridge are greatly improved by evaluating the safety state of the road and bridge in real time, predicting the future safety risk and automatically generating and optimizing the safety regulation and control strategy, and the method has important application value and wide popularization prospects.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of road and bridge safety assessment, and in particular to a road and bridge safety assessment method and system based on big data analysis. BACKGROUND

[0002] In modern transportation infrastructure, roads and bridges serve as important transportation hubs, carrying a large amount of traffic flow. With the acceleration of urbanization, the service life and safety of road and bridge structures are facing increasing challenges. Especially in complex scenarios such as high traffic density and extreme weather, the operation safety and structural health of roads and bridges are particularly prominent. Traditional road and bridge safety monitoring methods mainly rely on fixed structural health monitoring systems and periodic inspections, but these methods often have the shortcomings of data isolation, lack of real-time and predictive nature, and limitations in safety assessment.

[0003] At present, the Chinese invention patent with application number 202311068313.6 discloses a multi-level safety assessment method for large vehicles and ordinary vehicles on parallel highway bridges. This method fully considers the bridge load composition when large transport vehicles pass through highway bridges, provides a clear combination form of mixed vehicle load of large transport vehicles and ordinary vehicles and a calculation process, and according to the internal force response calculation results, gives a multi-level evaluation conclusion. Compared with traditional safety assessment methods, this application has higher approval efficiency, greatly reduces the safety risks of bridges caused by ignoring parallel vehicles during the evaluation of large transport vehicles, provides a scientific basis for the formulation of relevant management measures and the evaluation of large transport vehicles, solves the safety risk problem caused by ignoring parallel ordinary vehicle load during the evaluation of large transport vehicles, ensures the safe operation of large transport vehicles and bridges along the line, and improves the approval efficiency of large transport vehicles.

[0004] The above-mentioned technology cannot generate and optimize comprehensive safety assessment and control strategies in real time through multi-source data fusion under dynamic changes in traffic flow, environmental load and structural state, and cannot effectively improve the safety management level of roads and bridges. SUMMARY

[0005] The technical problem solved by the present application is that the prior art cannot generate and optimize comprehensive safety assessment and control strategies in real time through multi-source data fusion under dynamic changes in traffic flow, environmental load and structural state, and cannot effectively improve the safety management level of roads and bridges.

[0006] To solve the above technical problems, the present application provides the following technical solutions: A road and bridge safety assessment method based on big data analysis, comprising the following steps: Step S1: Collect road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data, and perform time synchronization and spatial mapping to form a road and bridge operation aligned dataset; Step S2: Based on the road and bridge operation alignment dataset and road and bridge geometric parameters and monitoring location parameters, construct a digital twin topology model. In the digital twin topology model, establish a corresponding spatiotemporal unit index for each road segment number and bridge component number, and record traffic behavior data and structural response data corresponding to each spatiotemporal unit index. Group the spatiotemporal unit indexes according to weather conditions, traffic flow level and time period to form a set of operation scenario features. Step S3: In the digital twin topology model, based on the road and bridge operation alignment dataset and the operation scenario feature set, perform traffic flow and structural response coupling analysis. Under each spatiotemporal unit index, extract traffic flow behavior features and structural response features simultaneously, calculate the mapping relationship and residual index between traffic flow behavior features and structural response features, and generate operation risk indicators. Step S4: Calculate the safety assessment results based on the operational risk indicators; Step S5: Generate a security control strategy based on the security assessment results and update the alignment dataset.

[0007] Preferably, step S1 includes the following sub-steps: Step S101: Collect road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data; add timestamps, collection location coordinates, road segment numbers, and bridge component numbers to the road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data. Step S102: Time synchronization of road and bridge structure monitoring data, traffic behavior data, environmental load data and operation and maintenance event data based on a unified time reference, and resampling of data with different sampling frequencies to a unified time step. Step S103: Based on the road centerline and bridge geometric parameters, project the synchronized road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data onto a unified spatial coordinate system. Aggregate the road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data in each time window and spatial unit to form a road and bridge operation aligned dataset with spatiotemporal labels.

[0008] Preferably, the construction logic of the aligned dataset is as follows: Based on the timestamps of each road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data in the aligned dataset, a fixed-length time window is divided and a time window index is established. Based on the geometric parameters of the road centerline and bridge components, vehicle trajectory data and accident log data are mapped to the corresponding road segment index and bridge component index to establish spatial units; For each time window and spatial unit combination, statistical data coverage and consistency indicators are combined. When the coverage indicator is lower than the preset coverage threshold or the consistency indicator is higher than the preset conflict threshold, the corresponding spatial unit combination is marked as a low-quality data unit and the marking result is recorded in the aligned dataset.

[0009] Preferably, step S2 includes the following sub-steps: Step S201: Construct a digital twin model based on the aligned dataset and road and bridge geometric information, map the road and bridge structure monitoring data to the corresponding bridge component nodes, map the traffic behavior data to the corresponding lane units, map the environmental load data to the corresponding exposed area units, and map the operation and maintenance event data to the corresponding components and time windows. Step S202: Based on weather conditions, traffic flow levels, vehicle type composition ratios, and time periods, the aligned dataset is divided into multiple road and bridge operation scenarios, and scenario labels are assigned to each time window and spatial unit combination to form scenario segmentation results. Step S203: In each scenario, statistical characteristics of bridge strain data, vibration acceleration data, deflection data, vehicle speed data, vehicle trajectory data, and environmental load data are collected to form a scenario feature dataset. The scenario feature dataset is then associated with the corresponding components and lane units in the digital twin model.

[0010] Preferably, step S3 includes the following sub-steps: Step S301: In the digital twin model, a traffic flow behavior and structural response coupling model is constructed based on the aligned dataset. Traffic flow data, vehicle type composition data, vehicle speed data, and vehicle trajectory data are used as traffic flow behavior features, and bridge strain data, vibration acceleration data, and deflection data are used as structural response features. The mapping relationship between traffic flow behavior features and structural response features is calculated in each scenario to generate a traffic flow behavior and structural response coupling result set. Step S302: In the set of coupling results between traffic behavior and structural response, calculate the residual sequence between the predicted structural response and the measured structural response for each time window and spatial unit, associate the residual sequence with the scene feature dataset, and statistically analyze the residual amplitude and residual duration to obtain the abnormal mode score. Step S303: Mark the time window and spatial unit with the abnormal mode score higher than the preset abnormal threshold as abnormal mode unit. According to the corresponding sudden deceleration event, sudden lane change event, sudden shortening of vehicle distance event and accident log in the abnormal mode unit, count the number and distribution of near miss accident events. Generate operation risk indicators based on abnormal mode score and number of near miss accident events, and establish a correspondence between operation risk indicators and the corresponding road segment number and bridge component number in the aligned dataset.

[0011] Preferably, step S4 includes the following sub-steps: Step S401: Based on the operational risk indicators and the bridge strain data, vibration acceleration data, deflection data, component temperature data and maintenance event data in the aligned dataset, calculate the structural safety index of each bridge component; Step S402: Based on the traffic flow data, vehicle type composition data, vehicle speed data and vehicle trajectory data in the operational risk indicators and aligned dataset, calculate the operational safety index for each road segment and lane unit; Step S403: Calculate the environmental resilience index for each bridge component and road segment based on environmental load data and operation and maintenance event data.

[0012] Preferably, step S5 includes the following sub-steps: Step S501: Based on the safety assessment results, candidate safety control strategies are generated for each bridge component and road segment. The safety control strategies include maintaining existing traffic conditions, adjusting speed limits, adjusting lane function allocation, restricting the passage of specific vehicle types, and triggering preventive maintenance operations. Step S502: Input the candidate safety control strategy into the digital twin model, simulate traffic behavior data and structural response data after implementing the safety control strategy based on the aligned dataset, and calculate the changes in operational risk indicators and safety assessment results before and after implementation. Step S503: Select the target safety control strategy based on the changes in the safety assessment results before and after implementation, and send the speed limit scheme, lane function allocation scheme and maintenance operation suggestions corresponding to the target safety control strategy to the control terminal. Record the target safety control strategy and the corresponding implementation effect as feedback data and update the alignment dataset.

[0013] Preferably, the remaining life prediction logic in the safety assessment results is as follows: Based on bridge strain data, vibration acceleration data, deflection data, component temperature data and operation and maintenance event data in the aligned dataset, the stress amplitude distribution and disease development rate of each bridge component under different scenarios are statistically analyzed, and the cumulative damage index is calculated. By combining the cumulative damage index with the structural safety index, a remaining life prediction model is constructed. The remaining life prediction model uses historical failure records and maintenance operation records as samples to output the remaining life range of each bridge component in the current scenario, and writes the remaining life range into the safety assessment results.

[0014] Preferably, the selection logic of the security control strategy is as follows: If the structural safety index, operational safety index, and environmental resilience index are all higher than the first safety threshold, the safety assessment result is determined to be a safe state, and the safety control strategy is set to maintain the existing passage conditions. If at least one of the structural safety index or the operational safety index is lower than the first safety threshold but higher than the second safety threshold, the safety assessment result is determined to be a warning state, and the safety control strategy is set to adjust the speed limit or adjust the lane function allocation. If at least one of the structural safety index or operational safety index is lower than the second safety threshold, the safety assessment result is determined to be a high-risk state, and the safety control strategy is set to restrict the passage of specific vehicle types and trigger preventive maintenance operations.

[0015] A road and bridge safety assessment system based on big data analysis includes a data acquisition and alignment module, a model building and analysis module, a traffic flow coupling analysis module, a risk assessment calculation module, and a strategy generation and feedback module. The data acquisition and alignment module is used to collect road and bridge structure monitoring data, traffic behavior data, environmental load data and operation and maintenance event data, and perform time synchronization and spatial mapping to form a road and bridge operation alignment dataset. The model building and analysis module is used to build a digital twin topology model based on the road and bridge operation alignment dataset and road and bridge geometric parameters and monitoring location parameters. In the digital twin topology model, a corresponding spatiotemporal unit index is established for each road segment number and bridge component number, and traffic behavior data and structural response data corresponding to each spatiotemporal unit index are recorded. The spatiotemporal unit indexes are grouped according to weather conditions, traffic flow level and time period to form a set of operation scenario features. The traffic flow coupling analysis module is used to perform traffic flow and structural response coupling analysis in the digital twin topology model based on the road and bridge operation alignment dataset and the operation scenario feature set. Under each spatiotemporal unit index, it simultaneously extracts traffic flow behavior features and structural response features, calculates the mapping relationship and residual index between traffic flow behavior features and structural response features, and generates operation risk indicators. The risk assessment calculation module is used to calculate the safety assessment results based on operational risk indicators; The strategy generation feedback module is used to generate security control strategies based on security assessment results and update the alignment dataset.

[0016] The beneficial effects of this invention are as follows: This invention combines multi-source data fusion, digital twin modeling, and traffic flow and structure coupling analysis technologies to provide a comprehensive, dynamic, and intelligent solution for road and bridge safety management. By assessing the safety status of roads and bridges in real time, predicting future safety risks, and automatically generating and optimizing safety control strategies, it greatly improves the operational safety of roads and bridges and the efficiency of traffic management. It has significant application value and broad prospects for promotion. Attached Figure Description

[0017] Figure 1 A flowchart illustrating the steps of a road and bridge safety assessment method based on big data analysis, provided as an embodiment of the present invention; Figure 2 This is a basic flowchart of a road and bridge safety assessment system based on big data analysis, provided as an embodiment of the present invention. Detailed Implementation

[0018] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0019] Example 1, referring to Figure 1 This paper provides a road and bridge safety assessment method based on big data analysis, which includes the following steps: Step S1: Collect road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data, and perform time synchronization and spatial mapping to form a road and bridge operation aligned dataset.

[0020] Step S2: Construct a digital twin topology model based on the road and bridge operation alignment dataset and road and bridge geometric parameters and monitoring location parameters. In the digital twin topology model, establish a corresponding spatiotemporal unit index for each road segment number and bridge component number, and record traffic behavior data and structural response data corresponding to each spatiotemporal unit index. Group the spatiotemporal unit indexes according to weather conditions, traffic flow level and time period to form an operation scenario feature set.

[0021] Step S3: In the digital twin topology model, traffic flow and structural response coupling analysis is performed based on the road and bridge operation alignment dataset and the operation scenario feature set. Traffic flow behavior features and structural response features are extracted simultaneously under each spatiotemporal unit index. The mapping relationship and residual index between traffic flow behavior features and structural response features are calculated to generate operation risk indicators.

[0022] Step S4: Calculate the safety assessment results based on the operational risk indicators.

[0023] Step S5: Generate a security control strategy based on the security assessment results and update the alignment dataset.

[0024] This invention combines multi-source data fusion, digital twin modeling, and traffic flow and structure coupling analysis technologies to provide a comprehensive, dynamic, and intelligent solution for road and bridge safety management. By assessing the safety status of roads and bridges in real time, predicting future safety risks, and automatically generating and optimizing safety control strategies, it greatly improves the operational safety of roads and bridges and the efficiency of traffic management, and has significant application value and broad prospects for promotion.

[0025] Step S1 includes the following sub-steps: Step S101: Collect road and bridge structure monitoring data, traffic behavior data, environmental load data, and maintenance event data. Add timestamps, collection location coordinates, road segment numbers, and bridge component numbers to the road and bridge structure monitoring data, traffic behavior data, environmental load data, and maintenance event data. The road and bridge structure monitoring data includes bridge strain data, vibration acceleration data, deflection data, and component temperature data. The traffic behavior data includes traffic flow data, vehicle type composition data, vehicle speed data, and vehicle trajectory data. The environmental load data includes meteorological data, water level data, wind speed data, and rainfall intensity data. The maintenance event data includes inspection report data, maintenance operation record data, and accident log data.

[0026] Step S101 involves collecting relevant data from different sources and classifying and labeling it. This process ensures that the collected data covers all dimensions required for road and bridge safety assessment, providing a comprehensive information foundation for subsequent data processing.

[0027] Step S102: Time synchronization is performed on road and bridge structure monitoring data, traffic behavior data, environmental load data and operation and maintenance event data based on a unified time reference, and data with different sampling frequencies are resampled to a unified time step.

[0028] Step S102 performs time and space consistency processing on all collected data. By unifying the time base and aligning the spatial coordinate system, it ensures accurate matching between different data sources, avoiding data inconsistencies caused by time misalignment or spatial differences, thus providing an accurate data foundation for subsequent analysis.

[0029] Step S103: Based on the road centerline and bridge geometric parameters, project the synchronized road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data onto a unified spatial coordinate system. Aggregate the road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data in each time window and spatial unit to form a road and bridge operation aligned dataset with spatiotemporal labels.

[0030] The logic for constructing the aligned dataset is as follows: Based on the timestamps of each road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data in the aligned dataset, a fixed-length time window is divided and a time window index is established.

[0031] Based on the geometric parameters of the road centerline and bridge components, vehicle trajectory data and accident log data are mapped to the corresponding road segment index and bridge component index to establish spatial units.

[0032] For each time window and spatial unit combination, statistical data coverage and consistency indicators are combined. When the coverage indicator is lower than the preset coverage threshold or the consistency indicator is higher than the preset conflict threshold, the corresponding spatial unit combination is marked as a low-quality data unit and the marking result is recorded in the aligned dataset.

[0033] Step S103 involves data cleaning, which removes outliers and fills in missing data, ensuring the integrity and reliability of the dataset. This process ensures that the final dataset meets quality standards, reduces the interference of erroneous data on subsequent analysis and evaluation, and enhances the accuracy and reliability of the system.

[0034] The road and bridge operation alignment dataset described in this invention refers to a dataset with spatiotemporal labels obtained by aggregating road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data according to time windows and spatial units after time synchronization and spatial mapping under a unified time reference and a unified spatial coordinate system.

[0035] Step S1 ensures the uniformity and consistency of multi-source data by comprehensively collecting road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data. Through time synchronization and spatial alignment, discrepancies between data are eliminated, ensuring high-quality and complete data input for subsequent analysis and safety assessments, thereby improving data usability and accuracy.

[0036] The road and bridge operation alignment dataset described in this invention is a set of spatiotemporally labeled data obtained by synchronizing and spatially mapping road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data under a unified time reference and a unified spatial coordinate system. Through this dataset, traffic behavior features and structural response features can be accessed simultaneously under any combination of road segment number, bridge component number, and time window.

[0037] Step S2 includes the following sub-steps: Step S201: Based on the road centerline, bridge component parameters and monitoring location parameters in the road and bridge operation alignment dataset, establish nodes for each road segment number and bridge component number in the digital twin topology model, and divide each node into multiple time windows to form a spatiotemporal unit index composed of node number and time window number.

[0038] Step S202: Under each spatiotemporal unit index, extract the traffic flow behavior data subset and structural response data subset from the road and bridge operation alignment dataset, and establish a one-to-one correspondence between the extracted subsets and the corresponding spatiotemporal unit index in the digital twin topology model.

[0039] Step S203: Group the spatiotemporal unit index according to weather conditions, traffic flow range, vehicle type composition ratio and time period. Within each group, statistically aggregate the subset of traffic flow behavior data and the subset of structural response data to generate a set of operating scenario features.

[0040] The digital twin topology model described in this invention uses road segment numbers and bridge component numbers as nodes, and monitoring location parameters and road connection relationships as edges. Under each spatiotemporal unit index, it records the corresponding subset of traffic flow behavior data and subset of structural response data, providing an index structure and geometric constraints for coupled analysis of traffic flow and structural response.

[0041] The operational scenario feature set described in this invention is a feature view obtained by grouping and statistically analyzing weather conditions, traffic flow intervals, and time periods based on the road and bridge operation alignment dataset. It is a different aggregation form of the same data source as the road and bridge operation alignment dataset.

[0042] Step S2 involves constructing a digital twin model of the road and bridge based on the aligned dataset, and extracting relevant scene features according to different scenarios. This process can simulate and predict the operating status and structural response of the road and bridge under different scenarios, providing accurate model support for subsequent safety assessments and control strategies.

[0043] Step S3 includes the following sub-steps: Step S301: In the digital twin topology model, based on the spatiotemporal unit index with the same road segment number, bridge component number, and time window number, traffic flow behavior feature vector and structural response feature vector are extracted from the road and bridge operation alignment dataset. The traffic flow behavior feature vector includes traffic volume, average vehicle speed, vehicle type composition ratio, and headway distribution parameters. The structural response feature vector includes representative values ​​of strain, root mean square values ​​of vibration acceleration, and representative values ​​of deflection. Based on regression analysis or correlation analysis, the mapping relationship between the traffic flow behavior feature vector and the structural response feature vector is calculated, and the coupling result set of traffic flow and structural response is output.

[0044] Step S301 uses data such as traffic flow, vehicle speed, and vehicle type composition to perform coupled analysis with structural response data such as bridge strain, vibration, and deflection. By establishing the relationship between traffic flow behavior and structural response, the dynamic impact of traffic flow on the bridge structure can be predicted, and potential risks such as structural overload or instability under specific traffic flow conditions can be identified.

[0045] Step S302: In the set of coupling results between traffic behavior and structural response, calculate the residual sequence between the predicted structural response and the measured structural response for each time window and spatial unit, associate the residual sequence with the scene feature dataset, and statistically analyze the residual amplitude and residual duration to obtain the abnormal mode score.

[0046] Step S302 analyzes the coupling results of traffic flow and structural response, calculates the residuals for each time window and spatial unit, identifies abnormal patterns, and generates corresponding operational risk indicators. If the difference between the structural response caused by traffic flow behavior and the expected standard exceeds a certain threshold, the system will mark it as an abnormal pattern, thereby generating potential risk points. These risk indicators can reflect the safety status of roads and bridges under different conditions and provide quantitative basis for subsequent safety assessments.

[0047] Step S303: Mark the time window and spatial unit with the abnormal mode score higher than the preset abnormal threshold as abnormal mode unit. According to the corresponding sudden deceleration event, sudden lane change event, sudden shortening of vehicle distance event and accident log in the abnormal mode unit, count the number and distribution of near miss accident events. Generate operation risk indicators based on abnormal mode score and number of near miss accident events, and establish a correspondence between operation risk indicators and the corresponding road segment number and bridge component number in the aligned dataset.

[0048] Step S303 further assesses the impact of traffic behavior on road and bridge safety by identifying and statistically analyzing near-miss incidents on road segments. By analyzing the frequency and distribution of these incidents, high-risk areas in certain road segments or lanes can be identified, thereby helping to provide early warnings of potential accidents and offering a reference for the formulation of control strategies.

[0049] Step S3 calculates operational risk indicators through coupled analysis of traffic behavior and structural response, helping to assess the safety status of roads and bridges under different traffic and environmental conditions. By analyzing the impact of traffic flow on the bridge structure and combining it with real-time monitoring data, potential risks and anomalies can be accurately identified, thus providing a strong basis for subsequent safety assessments and control strategies.

[0050] Step S4 includes the following sub-steps: Step S401: Based on the operational risk indicators and the bridge strain data, vibration acceleration data, deflection data, component temperature data, and maintenance event data in the aligned dataset, calculate the structural safety index of each bridge component.

[0051] With the first A bridge component, in the scene Below, time window set For example: In the scene Next, calculate the representative value for each physical quantity: The mathematical expression for the absolute average strain is: ; in, The absolute average value of strain, For the scene The number of sampling points for this component in the statistics. To align the component in the dataset at time [time] Strain data.

[0052] The mathematical expression for the root mean square value of acceleration is: ; in, Let be the root mean square value of acceleration, and let be the value of the component at time . Vibration acceleration data.

[0053] The mathematical expression for the absolute average deflection is: ; in, The absolute average value of deflection. For this component at time Deflection data.

[0054] The absolute average of the relative temperature deviation (relative to a certain reference temperature) The mathematical expression for ) is: ; in, The absolute average of the relative temperature deviation is given by the value of the component at time [time value missing]. Component temperature data;

[0055] Dimensionlessizing the above quantities and allowable values ​​yields the ratios of various demands: , , , ; in, , , and These are the corresponding design allowable values.

[0056] All four quantities mentioned above are dimensionless.

[0057] Definition of the first Components in the scene The comprehensive demand indicators are as follows: ; in, As a comprehensive demand indicator, The weights corresponding to the strain data, The weights corresponding to the vibration acceleration data. The weights corresponding to the deflection data. The weights corresponding to the component temperature data.

[0058] Then map it to a structural safety index : ; value range .

[0059] Note: When the comprehensive structural demand index is equal to 0, the structural safety index is equal to 1, and the safety margin is at its maximum. When the comprehensive structural demand index is greater than or equal to 1, the structural safety index is less than or equal to 0 and is truncated to 0, indicating that the safety margin is exhausted.

[0060] Step S401 calculates the structural safety index of each bridge component under the current scenario based on structural health monitoring data. By combining design standards and actual monitoring data, the safety margin of bridge components can be quantified, reflecting their safety status under current traffic flow and environmental loads. This calculation result provides a crucial structural safety reference for subsequent risk assessment.

[0061] Step S402: Based on the traffic flow data, vehicle type composition data, vehicle speed data, and vehicle trajectory data in the operational risk indicators and aligned dataset, calculate the operational safety index for each road segment and lane unit.

[0062] With the first Individual road segment or lane unit, in the scenario Here is an example.

[0063] Calculate the near-loss accident density (normalized by length and time): ; in, Near-miss accident density, For road section In the scene The number of near-miss incidents For the first Road segment length, For the scene The total duration of the corresponding time window.

[0064] Calculate accident density: ; in, Accident density, For road section In the scene The actual number of accidents.

[0065] Calculate the standard deviation of the speed: ; in: ; in, For the speed standard deviation, For road section At any moment The speed of vehicles passing through, In the scene Lower section The number of samples used in the speed statistics.

[0066] Calculate the proportion of violations related to vehicle frontage clearance: ; in This is an indicator function.

[0067] in, The proportion of violations related to the distance between the front of the vehicle. In the scene Lower section The number of samples used in the statistics of vehicle frontage distance For road section At any moment The distance between the front of the car, For road section The design minimizes the safe front-end spacing.

[0068] By setting reference values ​​and normalizing them, a dimensionless operational risk factor is obtained, and a comprehensive operational risk index is calculated. Its calculation logic is the same as that of the demand ratio and comprehensive demand index in step S401.

[0069] The operational safety index is defined as: ; illustrate: The larger the value, the safer the operation. When the total risk factor is 0, the operation safety index is equal to 1. When the sum of risk factors exceeds 1, the operation safety index is reduced to 0.

[0070] Step S402 calculates the operational safety index for each road segment or lane based on traffic flow behavior data. By combining factors such as traffic density and speed fluctuations, the potential risks of traffic flow to road and bridge safety are assessed. This index can reflect the impact of traffic flow changes on road and bridge safety and identify potential areas of traffic congestion or high-risk accidents.

[0071] Step S403: Calculate the environmental resilience index for each bridge component and road segment based on environmental load data and operation and maintenance event data.

[0072] Similarly, with the first Individual bridge components or road sections, in the scene Here is an example.

[0073] Define the relative intensity factor of water level: ; in, The relative intensity factor of water level. For the scene The corresponding water level data is below. To design a high water level, This is the highest water level.

[0074] Relative intensity factor of wind speed The calculation logic is the same as that of the water level relative intensity factor.

[0075] Resilience factor (the faster the recovery, the higher the resilience; this can be considered negative in terms of risk): First, define the dimensionless recovery time: ; in, For dimensionless recovery time, For components The average recovery time required to recover from performance degradation to normal levels in similar past environmental events. This serves as a reference recovery time threshold.

[0076] Redefining the Comprehensive Index of Environmental Risk Intensity The calculation logic is the same as that of the comprehensive demand index.

[0077] Mapping the above environmental risk intensity to an environmental resilience index Take the same : ; Note: The closer the environmental load is to or the greater it is than the design limit, the larger the value becomes, and the lower the value becomes.

[0078] The longer the recovery time and the larger the size, the lower the environmental resilience index will be.

[0079] In step S4, the safety assessment results should comprehensively consider the structural safety index, operational safety index, and environmental resilience index.

[0080] No. Components in the scene The comprehensive safety assessment index is a weighted sum of the structural safety index, operational safety index, and environmental resilience index. Step S403 calculates the environmental resilience index for each bridge component or road segment based on meteorological and environmental load data. By assessing the bridge's resilience under extreme weather and environmental conditions, the impact of environmental factors on road and bridge structures can be quantified, and vulnerable points susceptible to extreme weather can be identified. This index helps assess the bridge's durability and its ability to cope with sudden environmental changes.

[0081] Step S4, based on the calculated operational risk indicators and structural safety index, combined with environmental resilience data, calculates the comprehensive safety assessment result for the road and bridge. This step provides a comprehensive safety status assessment for each road and bridge component or section by quantifying three major indicators: structural safety, operational safety, and environmental resilience. This helps decision-makers identify potential high-risk areas and provides precise support for subsequent safety control strategies.

[0082] Step S5 includes the following sub-steps: Step S501: Based on the safety assessment results, candidate safety control strategies are generated for each bridge component and road segment. The safety control strategies include maintaining existing traffic conditions, adjusting speed limits, adjusting lane function allocation, restricting the passage of specific vehicle types, and triggering preventive maintenance operations.

[0083] Step S501 selects a suitable safety control strategy based on the safety assessment results. If the assessment results indicate that the road or bridge is in a high-risk or warning state, the system will automatically recommend measures such as adjusting speed limits, restricting specific vehicle types, reallocating lanes, or initiating maintenance work. This sub-step ensures that the control strategy can dynamically respond to different safety states and adjust traffic flow or management measures in a timely manner.

[0084] Step S502: Input the candidate safety control strategy into the digital twin model, simulate traffic behavior data and structural response data after implementing the safety control strategy based on the aligned dataset, and calculate the changes in operational risk indicators and safety assessment results before and after implementation.

[0085] In the digital twin model, the traffic behavior data, structural response data, and environmental load data of the current scenario are updated first.

[0086] The updated data is adjusted according to security control strategies, for example: If the policy is speed limit, update the vehicle speed data and adjust the vehicle speed according to the speed limit value set by the policy. If the strategy is lane adjustment, then update the distribution of traffic density data in different lanes; If the strategy is to increase inspections or reduce traffic flow, then update the distribution of traffic flow data across different road segments or time periods.

[0087] In digital twin models, the impact of vehicle behavior on the bridge structure's response is simulated based on updated data. Using input data, a vehicle-bridge coupled simulation model is used to predict the bridge's dynamic response. For example: Calculate bridge strain under different traffic flow densities and speeds, and predict the impact of traffic flow changes on bridge vibration acceleration, deflection, temperature changes, and other responses.

[0088] The prediction process is based on a vehicle-bridge coupled mechanics model, in which the impact of traffic flow on the structure is determined by factors such as vehicle speed, vehicle weight, and road conditions, while the structural response is determined by the dynamic characteristics of the bridge.

[0089] Before the simulation, the current operational risk indicators were calculated based on the current traffic flow data and structural response data. After the implementation of the safety control strategy, the new traffic flow data and structural response data were updated, and the operational risk indicators were recalculated using the updated data. The changes in risk indicators before and after implementation were calculated. Before implementing the strategy, a comprehensive safety assessment result was calculated based on the current operational risk indicators and structural safety index. After implementing the strategy, a comprehensive safety assessment result was calculated based on the updated operational risk indicators, structural safety index, and environmental resilience index. Finally, the changes in the comprehensive safety assessment results before and after implementation were calculated. If the change in the safety assessment result is greater than 0, it indicates that the safety has improved after implementing the safety control strategy; if the change in the safety assessment result is less than 0, it indicates that the safety has decreased after implementing the safety control strategy; if the change in the safety assessment result is equal to 0, it indicates that the implementation of the safety control strategy has not significantly changed the safety.

[0090] Step S502 inputs the selected safety control strategy into the digital twin model to simulate traffic behavior and structural response data after implementing the strategy. The actual effectiveness of the selected strategy is evaluated by comparing the changes in operational risk indicators and safety assessment results before and after implementation. If the safety assessment results improve after implementation, the strategy is considered effective; otherwise, adjustments or optimizations may be necessary.

[0091] Step S503: Select the target safety control strategy based on the changes in the safety assessment results before and after implementation, and send the speed limit scheme, lane function allocation scheme and maintenance operation suggestions corresponding to the target safety control strategy to the control terminal. Record the target safety control strategy and the corresponding implementation effect as feedback data and update the alignment dataset.

[0092] The remaining life prediction logic in the safety assessment results is as follows: Based on bridge strain data, vibration acceleration data, deflection data, component temperature data, and maintenance event data from the aligned dataset, the stress amplitude distribution and damage development rate of each bridge component under different scenarios are statistically analyzed, and the cumulative damage index is calculated.

[0093] By combining the cumulative damage index with the structural safety index, a remaining life prediction model is constructed. The remaining life prediction model uses historical failure records and maintenance operation records as samples to output the remaining life range of each bridge component in the current scenario, and writes the remaining life range into the safety assessment results.

[0094] The selection logic for security control strategies is as follows: If the structural safety index, operational safety index, and environmental resilience index are all higher than the first safety threshold, the safety assessment result is determined to be in a safe state, and the safety control strategy is set to maintain the existing passage conditions.

[0095] If at least one of the structural safety index or the operational safety index is lower than the first safety threshold but higher than the second safety threshold, the safety assessment result is determined to be in a warning state, and the safety control strategy is set to adjust the speed limit or adjust the lane function allocation.

[0096] If at least one of the structural safety index or operational safety index is lower than the second safety threshold, the safety assessment result is determined to be a high-risk state, and the safety control strategy is set to restrict the passage of specific vehicle types and trigger preventive maintenance operations.

[0097] Step S503 adjusts and optimizes the control strategy based on the simulation and post-implementation safety assessment results. If the post-implementation safety assessment results show a reduction in risk, the strategy will be maintained or further strengthened; if the post-implementation effect is unsatisfactory, the strategy will be adjusted based on feedback data to ensure maximum improvement in road and bridge safety. This process enhances adaptability and dynamic adjustment capabilities through real-time feedback and self-optimization.

[0098] Step S5 generates and adjusts control strategies adapted to different safety conditions based on the safety assessment results obtained in Step S4, and optimizes the strategies by simulating the effects after implementation. This step ensures that implementing reasonable safety control measures, such as speed limits, lane adjustments, load limits, or maintenance work, can effectively reduce potential risks and improve the safety of roads and bridges. Through a feedback mechanism, the control strategies are continuously optimized to improve the efficiency of overall safety management.

[0099] The operational scenario feature set described in this invention is a feature view obtained by grouping and statistically analyzing weather conditions, traffic flow intervals, and time periods based on the road and bridge operation alignment dataset. It is a different aggregation form of the same data source as the road and bridge operation alignment dataset.

[0100] Example 2, refer to Figure 2 This paper presents a road and bridge safety assessment system based on big data analysis, which includes a data acquisition and alignment module, a model building and analysis module, a traffic flow coupling analysis module, a risk assessment calculation module, and a strategy generation and feedback module.

[0101] The data acquisition and alignment module is used to collect road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data, and to perform time synchronization and spatial mapping to form a road and bridge operation alignment dataset.

[0102] The model building and analysis module is used to build a digital twin topology model based on the road and bridge operation alignment dataset and road and bridge geometric parameters and monitoring location parameters. In the digital twin topology model, a corresponding spatiotemporal unit index is established for each road segment number and bridge component number, and traffic behavior data and structural response data corresponding to each spatiotemporal unit index are recorded. The spatiotemporal unit indexes are grouped according to weather conditions, traffic volume levels and time periods to form a set of operation scenario features.

[0103] The traffic flow coupling analysis module is used to perform traffic flow and structural response coupling analysis in the digital twin topology model based on the road and bridge operation alignment dataset and the operation scenario feature set. Under each spatiotemporal unit index, it simultaneously extracts traffic flow behavior features and structural response features, calculates the mapping relationship and residual index between traffic flow behavior features and structural response features, and generates operation risk indicators.

[0104] The risk assessment calculation module is used to calculate the safety assessment results based on operational risk indicators.

[0105] The strategy generation feedback module is used to generate security control strategies based on security assessment results and update the alignment dataset.

[0106] This invention integrates multi-source data from road and bridge structure monitoring, traffic behavior, environmental load, and maintenance events to construct a digital twin model, which can comprehensively reflect the safety status of roads and bridges under different scenarios. By considering the combined effects of multiple factors such as traffic flow, vehicle type, weather conditions, and structural health, this method can provide more accurate safety assessment results.

[0107] By collecting and analyzing traffic flow and structural response data in real time, the safety status of roads and bridges can be monitored. In extreme weather or high traffic density conditions, potential risks can be identified and early warnings issued, providing traffic management personnel with the possibility of early intervention. Furthermore, predictive models based on historical and real-time data can predict the safety status of roads and bridges for the next few hours or days, allowing for proactive measures to prevent accidents.

[0108] After obtaining the security assessment results, corresponding security control strategies will be generated based on different security states. By comparing the security assessment results before and after implementation, the control strategies can be optimized, enabling self-learning and adjustment, and ensuring the best security management effect in different scenarios.

[0109] This invention offers high flexibility in practical applications, adapting to various combinations of road sections, bridges, weather conditions, and traffic flows. For example, in extreme scenarios such as morning rush hour rain or typhoons with strong winds and speed limits, the method can automatically identify risks and adjust assessment parameters, ensuring accurate safety assessments and effective control strategies in a variety of complex environments.

[0110] Automated safety assessments and strategy generation reduce the time and errors associated with human intervention, improving the efficiency of emergency response. Traffic management departments can obtain real-time information on the safety status and potential risks of roads and bridges, allowing for timely adjustments to traffic control measures or the scheduling of maintenance work, effectively reducing the risk of accidents and ensuring traffic safety.

[0111] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0112] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A road and bridge safety assessment method based on big data analysis, characterized in that, Includes the following steps: Step S1: Collect road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data, and perform time synchronization and spatial mapping to form a road and bridge operation aligned dataset; Step S2: Based on the road and bridge operation alignment dataset and road and bridge geometric parameters and monitoring location parameters, construct a digital twin topology model. In the digital twin topology model, establish a corresponding spatiotemporal unit index for each road segment number and bridge component number, and record traffic behavior data and structural response data corresponding to each spatiotemporal unit index. Group the spatiotemporal unit indexes according to weather conditions, traffic flow level and time period to form a set of operation scenario features. Step S3: In the digital twin topology model, based on the road and bridge operation alignment dataset and the operation scenario feature set, perform traffic flow and structural response coupling analysis. Under each spatiotemporal unit index, extract traffic flow behavior features and structural response features simultaneously, calculate the mapping relationship and residual index between traffic flow behavior features and structural response features, and generate operation risk indicators. Step S4: Calculate the safety assessment results based on the operational risk indicators; Step S5: Generate a security control strategy based on the security assessment results and update the alignment dataset.

2. The road and bridge safety assessment method based on big data analysis as described in claim 1, characterized in that, Step S1 includes the following sub-steps: Step S101: Collect road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data; add timestamps, collection location coordinates, road segment numbers, and bridge component numbers to the road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data. Step S102: Time synchronization of road and bridge structure monitoring data, traffic behavior data, environmental load data and operation and maintenance event data based on a unified time reference, and resampling of data with different sampling frequencies to a unified time step. Step S103: Based on the road centerline and bridge geometric parameters, project the synchronized road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data onto a unified spatial coordinate system. Aggregate the road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data in each time window and spatial unit to form a road and bridge operation aligned dataset with spatiotemporal labels.

3. The road and bridge safety assessment method based on big data analysis as described in claim 2, characterized in that, The logic for constructing the alignment dataset is as follows: Based on the timestamps of each road and bridge structure monitoring data, traffic behavior data, environmental load data, and operation and maintenance event data in the aligned dataset, a fixed-length time window is divided and a time window index is established. Based on the geometric parameters of the road centerline and bridge components, vehicle trajectory data and accident log data are mapped to the corresponding road segment index and bridge component index to establish spatial units; For each time window and spatial unit combination, statistical data coverage and consistency indicators are combined. When the coverage indicator is lower than the preset coverage threshold or the consistency indicator is higher than the preset conflict threshold, the corresponding spatial unit combination is marked as a low-quality data unit and the marking result is recorded in the aligned dataset.

4. The road and bridge safety assessment method based on big data analysis as described in claim 3, characterized in that, Step S2 includes the following sub-steps: Step S201: Based on the road centerline, bridge component parameters and monitoring location parameters in the road and bridge operation alignment dataset, establish nodes for each road segment number and bridge component number in the digital twin topology model, and divide each node into multiple time windows to form a spatiotemporal unit index composed of node number and time window number. Step S202: Under each spatiotemporal unit index, extract the traffic flow behavior data subset and structural response data subset from the road and bridge operation alignment dataset, and establish a one-to-one correspondence between the extracted subsets and the corresponding spatiotemporal unit index in the digital twin topology model; Step S203: Group the spatiotemporal unit index according to weather conditions, traffic flow range, vehicle type composition ratio and time period. Within each group, statistically aggregate the subset of traffic flow behavior data and the subset of structural response data to generate a set of operating scenario features.

5. The road and bridge safety assessment method based on big data analysis as described in claim 4, characterized in that, Step S3 includes the following sub-steps: Step S301: In the digital twin topology model, based on the spatiotemporal unit index with the same road segment number, bridge component number, and time window number, traffic flow behavior feature vector and structural response feature vector are extracted from the road and bridge operation alignment dataset. The traffic flow behavior feature vector includes traffic volume, average vehicle speed, vehicle type composition ratio, and vehicle headway distribution parameters. The structural response feature vector includes representative values ​​of strain, root mean square values ​​of vibration acceleration, and representative values ​​of deflection. Based on regression analysis or correlation analysis, the mapping relationship between the traffic flow behavior feature vector and the structural response feature vector is calculated, and the coupling result set of traffic flow and structural response is output. Step S302: In the set of coupling results between traffic behavior and structural response, calculate the residual sequence between the predicted structural response and the measured structural response for each time window and spatial unit, associate the residual sequence with the scene feature dataset, and statistically analyze the residual amplitude and residual duration to obtain the abnormal mode score. Step S303: Mark the time window and spatial unit with the abnormal mode score higher than the preset abnormal threshold as abnormal mode unit. According to the corresponding sudden deceleration event, sudden lane change event, sudden shortening of vehicle distance event and accident log in the abnormal mode unit, count the number and distribution of near miss accident events. Generate operation risk indicators based on abnormal mode score and number of near miss accident events, and establish a correspondence between operation risk indicators and the corresponding road segment number and bridge component number in the aligned dataset.

6. The road and bridge safety assessment method based on big data analysis as described in claim 5, characterized in that, Step S4 includes the following sub-steps: Step S401: Based on the operational risk indicators and the bridge strain data, vibration acceleration data, deflection data, component temperature data and maintenance event data in the aligned dataset, calculate the structural safety index of each bridge component; Step S402: Based on the traffic flow data, vehicle type composition data, vehicle speed data and vehicle trajectory data in the operational risk indicators and aligned dataset, calculate the operational safety index for each road segment and lane unit; Step S403: Calculate the environmental resilience index for each bridge component and road segment based on environmental load data and operation and maintenance event data.

7. The road and bridge safety assessment method based on big data analysis as described in claim 6, characterized in that, Step S5 includes the following sub-steps: Step S501: Based on the safety assessment results, candidate safety control strategies are generated for each bridge component and road segment. The safety control strategies include maintaining existing traffic conditions, adjusting speed limits, adjusting lane function allocation, restricting the passage of specific vehicle types, and triggering preventive maintenance operations. Step S502: Input the candidate safety control strategy into the digital twin model, simulate traffic behavior data and structural response data after implementing the safety control strategy based on the aligned dataset, and calculate the changes in operational risk indicators and safety assessment results before and after implementation. Step S503: Select the target safety control strategy based on the changes in the safety assessment results before and after implementation, and send the speed limit scheme, lane function allocation scheme and maintenance operation suggestions corresponding to the target safety control strategy to the control terminal. Record the target safety control strategy and the corresponding implementation effect as feedback data and update the alignment dataset.

8. The road and bridge safety assessment method based on big data analysis as described in claim 6, characterized in that, The remaining life prediction logic in the safety assessment results is as follows: Based on bridge strain data, vibration acceleration data, deflection data, component temperature data and operation and maintenance event data in the aligned dataset, the stress amplitude distribution and disease development rate of each bridge component under different scenarios are statistically analyzed, and the cumulative damage index is calculated. By combining the cumulative damage index with the structural safety index, a remaining life prediction model is constructed. The remaining life prediction model uses historical failure records and maintenance operation records as samples to output the remaining life range of each bridge component in the current scenario, and writes the remaining life range into the safety assessment results.

9. The road and bridge safety assessment method based on big data analysis as described in claim 7, characterized in that, The selection logic for the security control strategy is as follows: If the structural safety index, operational safety index, and environmental resilience index are all higher than the first safety threshold, the safety assessment result is determined to be a safe state, and the safety control strategy is set to maintain the existing passage conditions. If at least one of the structural safety index or the operational safety index is lower than the first safety threshold but higher than the second safety threshold, the safety assessment result is determined to be a warning state, and the safety control strategy is set to adjust the speed limit or adjust the lane function allocation. If at least one of the structural safety index or operational safety index is lower than the second safety threshold, the safety assessment result is determined to be a high-risk state, and the safety control strategy is set to restrict the passage of specific vehicle types and trigger preventive maintenance operations.

10. A road and bridge safety assessment system based on big data analysis, applied in a road and bridge safety assessment method based on big data analysis as described in any one of claims 1-9, characterized in that, It includes a data acquisition and alignment module, a model building and analysis module, a traffic flow coupling analysis module, a risk assessment and calculation module, and a strategy generation and feedback module; The data acquisition and alignment module is used to collect road and bridge structure monitoring data, traffic behavior data, environmental load data and operation and maintenance event data, and perform time synchronization and spatial mapping to form a road and bridge operation alignment dataset. The model building and analysis module is used to build a digital twin topology model based on the road and bridge operation alignment dataset and road and bridge geometric parameters and monitoring location parameters. In the digital twin topology model, a corresponding spatiotemporal unit index is established for each road segment number and bridge component number, and traffic behavior data and structural response data corresponding to each spatiotemporal unit index are recorded. The spatiotemporal unit indexes are grouped according to weather conditions, traffic flow level and time period to form a set of operation scenario features. The traffic flow coupling analysis module is used to perform traffic flow and structural response coupling analysis in the digital twin topology model based on the road and bridge operation alignment dataset and the operation scenario feature set. Under each spatiotemporal unit index, it simultaneously extracts traffic flow behavior features and structural response features, calculates the mapping relationship and residual index between traffic flow behavior features and structural response features, and generates operation risk indicators. The risk assessment calculation module is used to calculate the safety assessment results based on operational risk indicators; The strategy generation feedback module is used to generate security control strategies based on security assessment results and update the alignment dataset.

Images