Bridge full life economic dynamic evaluation method based on engineering case knowledge transfer

Abstract

This disclosure relates to the field of bridge engineering economic evaluation and intelligent data analysis technology, specifically to a dynamic evaluation method for the full life-cycle economic performance of bridges based on knowledge transfer from engineering cases. The method includes: determining the state point corresponding to the target bridge in a preset coordinate system; mapping the full life-cycle economic data of the engineering case to the preset coordinate system; selecting a target segment from the segments to be affected corresponding to the engineering case based on selection weights to obtain a target state curve covering the remaining lifespan of the target bridge; determining the traffic loss cost of the corresponding traffic function interval based on the moment when the bridge structural damage value of the target bridge reaches each functional threshold; and determining the full life-cycle economic evaluation result of the target bridge based on the target state curve and the traffic loss cost. This allows the evaluation result to simultaneously reflect the bridge damage evolution process and traffic function loss, providing a more accurate decision-making basis for bridge maintenance decisions and traffic control.

Description

Technical Field

[0001] This disclosure relates to the field of bridge engineering economic evaluation and intelligent data analysis technology, specifically to a dynamic evaluation method for the economic performance of bridges throughout their entire life cycle based on knowledge transfer from engineering cases. Background Technology

[0002] As an important component of road traffic infrastructure, bridges are characterized by large-scale construction investment, long service life, and complex operating environment. Their entire life cycle covers multiple stages, including construction, operation and maintenance, risk event handling, and end of life. Each stage involves different types of economic costs and value changes. Therefore, the economic evaluation of the entire life cycle of bridges is of great significance for bridge maintenance and decommissioning decisions.

[0003] However, existing bridge economic evaluation methods typically employ static quota calculations and empirical assessments, focusing on cost analysis during the construction phase. They configure economic parameters for the operation period according to fixed proportions or empirical quotas, failing to dynamically adjust them according to the actual condition of the bridge structure. Furthermore, they rely on the bridge's own historical data, making it difficult to obtain accurate full-life-cycle economic evaluation results for newly commissioned bridges or those with insufficient historical data. Consequently, they cannot provide accurate decision-making basis for bridge maintenance, leading to accelerated deterioration of bridge structural performance and increased safety risks. Summary of the Invention

[0004] To overcome the problems existing in related technologies, this disclosure provides a dynamic evaluation method for the economic efficiency of bridges throughout their entire life cycle based on knowledge transfer from engineering cases, in order to address the deficiencies in related technologies.

[0005] This disclosure provides a method for dynamic evaluation of the economic performance of bridges throughout their entire life cycle based on knowledge transfer from engineering cases, including: Based on the structural state data of the target bridge, the state point corresponding to the target bridge is determined in a preset coordinate system. The preset coordinate system uses the bridge structural damage value as the horizontal axis and the bridge economic parameter as the vertical axis. The horizontal axis is divided into different traffic function intervals by the functional thresholds of the target bridge. The functional thresholds include load limit threshold, traffic restriction threshold and bridge closure threshold. The full life-cycle economic data of each engineering case are mapped to the preset coordinate system to obtain multiple state curves. The selection weight of each engineering case is determined based on preset rules. Based on the state points, the state curves and the selection weights, a target state curve covering the remaining life of the target bridge is determined. Based on the target state curve and the relationship between the bridge structure damage value of the target bridge and the service time, the time when the bridge structure damage value of the target bridge reaches each of the functional thresholds is determined, and based on the time, the passage loss cost of the corresponding passage functional section is determined. Based on the target state curve and the traffic loss cost, the life-cycle economic evaluation result of the target bridge is determined.

[0006] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects: Based on the structural state data of the target bridge, state points are determined in a preset coordinate system. The life-cycle economic data of each engineering case are then mapped to this preset coordinate system, which uses bridge structural damage values ​​as the horizontal axis and bridge economic parameters as the vertical axis. This provides a unified data foundation for the life-cycle economic evaluation of bridges, carrying comprehensive life-cycle economic information, and enables dynamic adaptation between the life-cycle economic evaluation and the structural deterioration process. Based on the state points, state curves, and selected weights, a target state curve covering the remaining lifespan of the target bridge is obtained. This fully utilizes existing bridge engineering experience and knowledge, compensates for the limitations of insufficient historical data on the bridge itself, provides a more accurate basis for bridge maintenance decisions, and reduces bridge safety risks.

[0007] Furthermore, based on the target state curve and the relationship between the bridge structural damage value and service time, the time when the bridge structural damage value reaches each functional threshold is determined, making the switching points of traffic functions such as load restriction, traffic restriction, and bridge closure predictable. Finally, based on the target state curve and the cost of traffic loss, the life-cycle economic evaluation results are determined, so that the evaluation results simultaneously reflect the bridge damage evolution process and the loss of traffic function, providing a more accurate decision-making basis for bridge maintenance decisions and traffic control measures such as load restriction, traffic restriction, and bridge closure, thereby reducing the deterioration of bridge structural performance and safety risks. Attached Figure Description

[0008] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0009] Figure 1 This is a flowchart illustrating an exemplary embodiment of the present disclosure of a dynamic evaluation method for the economic efficiency of bridges throughout their entire life cycle based on knowledge transfer from engineering cases.

[0010] Figure 2 This is a schematic diagram illustrating the structure of the bridge life-cycle economic knowledge base in the dynamic evaluation method for bridge life-cycle economic efficiency based on engineering case knowledge transfer, as shown in an exemplary embodiment of this disclosure. Detailed Implementation

[0011] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure.

[0012] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “a,” “the,” and “the” used in this disclosure are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that although the terms first, second, third, etc., may be used in this disclosure to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, first information may also be referred to as second information without departing from the scope of this disclosure, and similarly, second information may also be referred to as first information.

[0013] Existing bridge economic evaluation methods focus on cost analysis during the construction phase, configuring economic parameters for the operational period according to fixed proportions or experience-based quotas, which cannot be dynamically adjusted according to the actual condition of the bridge structure. Therefore, the evaluation results fail to reflect the actual evolution of bridge structural damage, leading to a mismatch between maintenance investment and the actual degree of structural deterioration. This results in the continuous accumulation of bridge damage, accelerating the degradation of bridge structural performance and increasing bridge safety risks. Furthermore, the reliance on long-term historical data makes it difficult to obtain reliable full-life-cycle economic evaluation results for newly commissioned bridges or those with insufficient data accumulation. Such bridges are in the early stages of damage accumulation before the defects are fully exposed; the lack of evaluation makes maintenance decisions unfounded, leading to untimely maintenance and increased structural performance degradation and safety risks.

[0014] Alternatively, existing evaluation methods focus on construction costs and direct maintenance expenses, without considering the traffic losses that occur when bridge structural damage develops to the point where traffic restrictions such as load limits and traffic restrictions are required. This leads to an underestimation of the total life-cycle cost, resulting in maintenance investment decisions based on evaluation results that tend to postpone the treatment of defects. As a result, bridge structural damage is not treated in a timely manner and continues to accumulate, which in turn accelerates the deterioration of bridge structural performance and increases bridge safety risks.

[0015] Based on this, this disclosure provides a dynamic evaluation method for the economic performance of bridges throughout their entire life cycle based on knowledge transfer from engineering cases. This method is applied to electronic devices, which include a processor, a memory, and a communication interface. The processor is communicatively connected to the memory and the communication interface. The memory stores a computer program, and the processor executes the computer program to implement the dynamic evaluation method for the economic performance of bridges throughout their entire life cycle based on knowledge transfer from engineering cases.

[0016] Reference Figure 1 The dynamic evaluation method for the economic efficiency of bridges throughout their entire life cycle based on knowledge transfer from engineering cases provided in this disclosure includes steps S101 to S104.

[0017] In step S101, based on the structural state data of the target bridge, the corresponding state point of the target bridge is determined in a preset coordinate system. The preset coordinate system uses the bridge structural damage value as the horizontal axis and the bridge economic parameters as the vertical axis. The horizontal axis is divided into different traffic function intervals by the functional thresholds of the target bridge, including load limit thresholds, traffic restriction thresholds, and bridge closure thresholds.

[0018] For example, structural condition data includes periodic inspection data and health monitoring data. Periodic inspection data includes bridge technical condition assessment results, crack width, concrete carbonation depth, and steel reinforcement corrosion potential. Health monitoring data includes deflection, strain, and vibration frequency.

[0019] For example, the defect scale of each bridge component can be assessed according to the bridge technical condition assessment standard, and the scores of each component can be aggregated step by step. Then, the bridge technical condition score is obtained by weighted summation according to the component weights. The bridge structural damage value is then converted into the difference between 1 and the ratio of the bridge technical condition score to 100. For example, if the bridge technical condition score is 68 after weighting the scores of each component, the bridge structural damage value is 1 - 68 / 100 = 0.32. The bridge economic parameters are calculated from the economic data such as the current year's maintenance costs, repair costs, and inspection and monitoring costs of the target bridge. The two constitute the coordinates of the state point in the preset coordinate system. The preset coordinate system places the bridge structural damage value and the bridge economic parameters in the same reference system, so that the bridge structural state and economic performance correspond one-to-one, providing a coordinate benchmark for subsequent case mapping and economic evaluation.

[0020] For example, the structural damage value of the target bridge during the current evaluation period is D=0.32. Simultaneously, the annual maintenance and repair costs of the target bridge are converted to obtain the current year's economic parameter E=850,000 yuan / year. A pre-defined coordinate system uses the bridge structural damage value as the horizontal axis, ranging from [0,1], and the bridge economic parameter as the vertical axis, with values ​​in ten thousand yuan / year. Based on the traffic management requirements of the target bridge, the functional thresholds include a load limit threshold D1=0.55, a traffic restriction threshold D2=0.75, and a bridge closure threshold D3=0.90. The horizontal axis is divided into a normal traffic zone [0,D1), a load limit functional zone [D1,D2), a traffic restriction functional zone [D2,D3), and a bridge closure functional zone [D3,1]. Thus, each traffic functional zone corresponds to an actual traffic management state, providing a structural basis for the subsequent measurement of traffic loss costs.

[0021] In step S102, the full life-cycle economic data of each engineering case is mapped to a preset coordinate system to obtain multiple state curves. The selection weight of each engineering case is determined based on preset rules. Based on the state points, state curves and selection weights, the target state curve covering the remaining life of the target bridge is determined.

[0022] It should be understood that the life-cycle economic data of bridges is characterized by a long accumulation period and significant individual differences. The historical economic data of the target bridge itself is often insufficient to support economic predictions for its remaining lifespan. However, the existing bridges in the engineering case library have completed part or even all of their life-cycle processes, and their life-cycle economic data contains patterns in the evolution of economic parameters with structural damage. Therefore, the life-cycle economic data of each engineering case is mapped to a preset coordinate system, and the economic data is organized using the bridge structural damage value as the index, rather than the service time. This eliminates the alignment difficulties caused by differences in the service years of different bridges, making the economic patterns of the cases comparable on a unified damage axis. Furthermore, using the state point of the target bridge as a boundary, along the positive direction of the horizontal axis (i.e., the direction of increasing damage value), the portion of the case curve to the left of the state point represents the process the target bridge has already experienced, i.e., the completed segment; the portion to the right of the state point represents the process the target bridge has not yet experienced, i.e., the segment to be encountered.

[0023] For example, weights are selected to characterize the degree of matching between the potential damage segment and the target bridge condition in each engineering case. The higher the weight, the more likely the future economic trend of the case is to be replicated in the target bridge. Based on the selected weights, target damage segments are selected and sequentially spliced ​​along the damage axis to obtain a target state curve covering the remaining lifespan of the target bridge.

[0024] For example, refer to Figure 2 The bridge life-cycle economic knowledge base consists of multiple engineering case knowledge units. Historical engineering cases are presented in a structured and standardized manner, forming a data foundation for knowledge transfer and similarity calculation. The life-cycle economic data for each engineering case includes: construction investment data, such as construction and installation costs, survey and design fees, land acquisition and demolition costs, and construction period interest; annual maintenance data, such as annual minor repair costs, medium repair costs, major repair costs, routine maintenance costs, testing and monitoring costs, and management fees; event-related maintenance data, such as the cost of a single repair or reinforcement event; traffic operation data, such as toll revenue and traffic loss costs during load restrictions, traffic restrictions, and bridge closures; risk management data, such as the cost of handling and repairing structural damage accidents and emergencies; and end-of-life data, such as demolition costs and residual value recovery. The economic parameters of each engineering case are converted according to a unified benchmark to eliminate differences in economic data caliber across different periods. The life-cycle economic data is organized by economic data record, with each record including the bridge's service time and economic parameters.

[0025] For example, using the target bridge's state point P(0.32, 85) as the boundary, each state curve is divided into completed segments and segments to be generated. The selection weights of each engineering case are determined based on preset rules; this process is detailed below. If the selection weights of engineering cases A, B, and C are 0.45, 0.32, and 0.23 respectively, then the segment to be generated from engineering case A, with the highest weight, is selected as the target segment to be generated. Furthermore, if the segment to be generated from engineering case A does not cover the remaining lifespan of the target bridge, the segments to be generated from other engineering cases are selected sequentially according to their selection weights and spliced ​​together until the spliced ​​curve covers the remaining lifespan of the target bridge.

[0026] Therefore, by using case knowledge transfer, the life-cycle economic laws of existing bridges can be transferred to target bridges, making up for the limitations of insufficient historical economic data of the target bridges. This allows newly commissioned bridges or bridges with insufficient data accumulation to obtain economic parameter curves covering the entire remaining life, improving the accuracy and applicability of the evaluation, and thus providing more accurate decisions for bridge maintenance investment.

[0027] In step S103, based on the target state curve and the relationship between the bridge structure damage value of the target bridge and the service time, the time when the bridge structure damage value of the target bridge reaches each functional threshold is determined, and based on the time, the passage loss cost of the corresponding passage functional section is determined.

[0028] It should be understood that the target state curve represents the correspondence between the bridge's economic parameters and the changes in the bridge's structural damage value, which belongs to the state dimension of information. The relationship between the bridge's structural damage value and its service time represents the evolution of damage over time, which belongs to the time dimension of information. Combining the two allows us to extend the economic laws of the state dimension to the time dimension, thereby determining the moments when the target bridge's structural damage value reaches the load limit threshold, traffic restriction threshold, and bridge closure threshold. This makes the switching points of traffic function states such as load limit, traffic restriction, and bridge closure predictable.

[0029] For example, based on the periodic inspection data of the target bridge over the years, the relationship between the bridge structural damage value and service time is fitted to obtain the damage evolution model as: D(t) = 0.32 + 0.028t + 0.0006t 2Where t represents the remaining service time from the current moment, in years. Continuing with the example above, substituting D1=0.55, D2=0.75, and D3=0.90 into the damage evolution model, we obtain the following timeframes for the target bridge's structural damage: t1 = year 7.2, t2 = year 12.2, and t3 = year 15.5, when the bridge structural damage reaches the load limit threshold; t2 = year 12.2 when the damage reaches the traffic restriction threshold; and t3 = year 15.5 when the damage reaches the bridge closure threshold. Therefore, the duration of the load-limited functional zone is from year 7.2 to year 12.2, or 5.0 years; the duration of the traffic restriction functional zone is from year 12.2 to year 15.5, or 3.3 years; and the bridge closure functional zone begins from year 15.5. Then, based on the traffic volume data and transportation cost data of the road section where the target bridge is located, the annual traffic loss cost caused by the decrease in transportation efficiency due to the load restriction measures in the load-restricted functional area is determined to be L1=1.2 million yuan / year, the annual traffic loss cost in the restricted functional area is L2=2.6 million yuan / year, and the annual traffic loss cost in the bridge closure functional area is L3=4.8 million yuan / year. The traffic loss cost of each traffic functional area is accumulated according to the duration of the measures.

[0030] Therefore, the bridge damage evolution process is transformed into a time-sensitive economic loss of traffic, so that bridges with different damage evolution rates can show corresponding differences in the whole life economic evaluation results, providing a measurement basis for traffic loss costs that change dynamically with the damage process.

[0031] In step S104, the full life-cycle economic evaluation result of the target bridge is determined based on the target state curve and the traffic loss cost.

[0032] For example, the life-cycle economic evaluation result of the target bridge consists of three parts: economic parameters of the completed phases, economic parameters of the remaining lifespan, and traffic loss costs. The economic parameters of the completed phases are the actual construction and maintenance economic data of the target bridge since its construction; the economic parameters of the remaining lifespan are determined by the values ​​of the target state curve during the remaining lifespan, including annual maintenance costs and repair costs; and the traffic loss costs are the sum of traffic losses accumulated over time in each traffic functional zone. Discounting these parts to the same reference time using a discount rate yields the life-cycle economic evaluation result of the target bridge. For example, the life-cycle economic evaluation result can be expressed as the net present value over the entire lifespan, or it can be expressed as annual equivalent costs, life-cycle costs, etc., which are not limited in this embodiment.

[0033] Therefore, the full life cycle economic evaluation results are determined based on the target state curve, duration, and traffic loss cost, so that the evaluation results simultaneously reflect the bridge damage evolution process and traffic function loss, providing a quantitative basis for bridge maintenance decisions and traffic control such as load limits, traffic restrictions, and bridge closures.

[0034] In some embodiments, a control for triggering a life-cycle economic evaluation of the target bridge can be displayed on the interactive page. Then, in response to the triggering operation of the control, the dynamic life-cycle economic evaluation method for bridges disclosed herein is executed to obtain the life-cycle economic evaluation results of the bridge, which are then displayed on the interactive page. The life-cycle economic evaluation results may include at least one of the following: the composition ratio of the bridge's life-cycle economic cost, construction phase cost, operation and maintenance cost, risk loss cost, and end-of-life residual value; unit life-cycle cost; risk cost ratio; and comprehensive economic benefit indicators.

[0035] It should be understood that the results of the economic evaluation of the entire life cycle of a bridge can be used to compare and analyze different design schemes, maintenance strategies or repair schemes of the bridge, so as to determine the optimal scheme.

[0036] For example, maintenance plans for the target bridge are generated based on the results of the life-cycle economic evaluation. The life-cycle economic evaluation results include the arrival times of each functional threshold and information on traffic loss costs. The generation process includes: determining the constraint window for maintenance timing based on the arrival times of each functional threshold in the life-cycle economic evaluation results, that is, reserving a maintenance window before the arrival time of the load limit threshold as a common time constraint for candidate maintenance plans; determining the maintenance objects based on the defect detection records of the target bridge, and selecting maintenance events that match the state of the target bridge from the maintenance event units of the engineering case, determining the corresponding maintenance measures and maintenance intensity, and forming candidate maintenance plans based on maintenance costs; for each candidate maintenance plan, simulating the state rollback after implementing the maintenance plan, updating the target state curve based on the state rollback, re-determining the arrival times of each functional threshold, traffic loss costs, and life-cycle economic evaluation results, comparing the updated life-cycle economic evaluation results with the life-cycle economic evaluation results before implementing maintenance, and selecting the candidate maintenance plan with the best updated life-cycle economic evaluation results as the target maintenance plan. This allows the evaluation results to be transformed into actionable maintenance schedules, enabling the integration of evaluation results with maintenance operations and improving the timeliness of treatment for bridge structural defects.

[0037] In some embodiments, detour path information of adjacent bridges in the road network where the target bridge is located can also be obtained, and based on the detour path information, the detour traffic volume of restricted vehicles on each adjacent bridge can be determined. The adjacent bridges are the bridges that bear the detour traffic of the target bridge, and the detour path information represents the detour distance from the target bridge to the adjacent bridge and the capacity of the corresponding detour path. The increase in traffic load of each adjacent bridge can be determined based on the detour traffic volume, and the relationship between the bridge structural damage value of each adjacent bridge and the service time can be corrected based on the increase in traffic load. Based on the corrected relationship, the time when the bridge structural damage value of each adjacent bridge reaches the corresponding functional threshold can be determined. Based on the time when each adjacent bridge reaches the corresponding functional threshold and the detour traffic volume, the additional traffic loss cost can be determined, and based on the additional traffic loss cost, the life-cycle economic evaluation result of the target bridge can be updated.

[0038] It should be understood that when traffic restrictions such as load limits, traffic restrictions, or bridge closures are implemented on the target bridge, restricted vehicles will choose detour routes. These detour vehicles will then travel to adjacent bridges, increasing the traffic load on those bridges and accelerating their damage evolution. This will bring them to their functional thresholds earlier and result in additional traffic losses. Furthermore, the detour routes incur additional transportation costs. All these costs arise from the traffic restrictions imposed on the target bridge and should be included in the overall life-cycle economic evaluation of the target bridge. This is to avoid underestimating costs due to neglecting network effects in single-bridge evaluations, which would prevent accurate decision-making for bridge group-level maintenance and traffic control.

[0039] For example, detour distance is used to measure the path cost of relocating vehicles via detours, and traffic capacity is used to measure the upper limit of the capacity of each detour route for detour traffic. For instance, if the road network where the target bridge is located includes adjacent bridge E and adjacent bridge F, the detour route information includes: the detour distance from the target bridge to adjacent bridge E is 12km, corresponding to provincial highway S1, with a daily capacity of 6,000 vehicles and a remaining capacity of 600 vehicles; the detour distance from the target bridge to adjacent bridge F is 18km, corresponding to county road X2, with a daily capacity of 4,000 vehicles and a remaining capacity of 400 vehicles.

[0040] For example, the traffic restriction level is determined based on the traffic function zone where the target bridge is located, and the detour traffic volume for restricted vehicles on each of the adjacent bridges is determined based on the traffic restriction level and detour route information. The traffic restriction level corresponds one-to-one with the traffic function zone. For instance, the traffic restriction level for a normal traffic zone is the normal level, the traffic restriction level for a load-limited function zone is the load-limited level, the traffic restriction level for a restricted traffic function zone is the traffic restriction level, and the traffic restriction level for a bridge closure function zone is the bridge closure level. When the target bridge is in a load-limited function zone, a restricted traffic function zone, or a bridge closure function zone, the calculation of detour traffic volume is triggered.

[0041] For example, if the target bridge is located in a load-restricted functional area, the corresponding traffic restriction level is the load-restricted level, and the restricted vehicles are heavy-duty vehicles whose axle load exceeds the load limit. According to traffic volume survey data of the road section where the target bridge is located, an average of 1200 heavy-duty vehicles pass through daily, of which 800 heavy-duty vehicles need to detour. Based on the detour route information, the detour traffic volume is allocated according to the proportion of the remaining capacity of each detour route: the remaining capacity of the detour route corresponding to adjacent bridge E is 600 vehicles, and the remaining capacity of the detour route corresponding to adjacent bridge F is 400 vehicles. Therefore, the detour traffic volume borne by adjacent bridge E is 800×600 / (400+600)=480 vehicles / day, and the detour traffic volume borne by adjacent bridge F is 800×400 / (400+600)=320 vehicles / day.

[0042] For example, the detour traffic volume of each adjacent bridge is converted into an increase in traffic load by vehicle type and axle load. First, the detour traffic volume is categorized by vehicle type, the representative axle load of each vehicle type is determined, and the axle load conversion factor for each vehicle type is determined based on the axle load conversion formula or measured axle load spectrum. The axle load conversion factor increases exponentially with the ratio of the representative axle load to the standard axle load; the larger the representative axle load, the larger the axle load conversion factor. Then, the number of detour vehicles of each vehicle type is multiplied by its axle load conversion factor and summed to obtain the increase in traffic load for each adjacent bridge. Finally, the damage evolution rate of adjacent bridges is corrected based on the increase in traffic load.

[0043] For example, if the adjacent bridge E experiences an average daily increase of 480 heavy-load vehicles, representing a 40% increase in heavy-load traffic compared to the original level, then the traffic load increase is 40%. Then, the following calculation formula is used: Take the load evolution rate index =0.25, therefore the correction factor is determined to be approximately 1.099. Wherein, This indicates the percentage increase in traffic load. The damage evolution model for adjacent bridge E is as follows: (t) = 0.28 + 0.031t, where t represents the remaining service time from the current moment in years. Multiplying the rate-related term in the damage evolution model by a correction factor yields... (t) = 0.28 + 0.0341t. Substituting the bridge closure threshold of 0.90 into the equation... (t), the time when the bridge structural damage value of the adjacent bridge E reaches its bridge closure threshold is re-determined, from the previous year 20.0 to the year 18.2.

[0044] It should be understood that the additional traffic loss cost consists of two parts: First, the traffic loss caused by the early implementation of load restrictions, traffic restrictions or bridge closures due to adjacent bridges reaching their corresponding functional thresholds ahead of schedule is equal to the product of the number of years the functional thresholds were reached ahead of schedule and the annual traffic loss cost per unit of that traffic functional area; Second, the additional transportation mileage cost incurred by detour vehicles traveling on the detour route is related to the detour traffic volume, detour distance and unit mileage transportation cost.

[0045] For example, the additional traffic loss cost is discounted to the same baseline time based on the time of occurrence, and then the discounted additional traffic loss cost is used as an additional cost component to update the life-cycle economic evaluation result of the target bridge. For instance, if the adjacent bridge E reaches the bridge closure threshold 1.8 years ahead of schedule, the traffic loss cost incurred by its bridge closure functional area ahead of schedule is 4.8 million yuan per year, totaling 8.64 million yuan; the additional transportation mileage cost incurred by detour vehicles traveling on the detour route is 650,000 yuan per year, totaling approximately 10.5 million yuan after discounting based on the remaining lifespan of 25 years. Therefore, the additional traffic loss cost is determined to be 19.14 million yuan. According to the life-cycle cost caliber, the life-cycle cost after including the additional traffic loss cost increases by 19.14 million yuan compared to when it is not included. According to the life-cycle net present value caliber, the NPV of the target bridge's life-cycle economic evaluation result decreases by 19.14 million yuan.

[0046] Therefore, the external costs generated by traffic restriction measures such as load limits, traffic restrictions, and bridge closures are internalized into the full life-cycle economic evaluation results of the target bridge, so that the evaluation results reflect the impact of detours and provide a more accurate basis for decision-making on bridge group-level maintenance and traffic control.

[0047] In some embodiments, the traffic load spectrum data of the target bridge can also be acquired, and the load limit value can be determined based on the load limit threshold. The traffic load spectrum data is then truncated using the load limit value as the cutoff point to obtain truncated traffic load spectrum data, where the load limit value is the maximum load value allowed for passage within the load limit functional zone. Based on the traffic load spectrum data and the truncated traffic load spectrum data, the equivalent cumulative damage value before and after the load limit is determined, and a damage evolution rate correction coefficient is determined based on the equivalent cumulative damage value. The equivalent cumulative damage value is obtained by multiplying and summing the traffic frequency and damage contribution corresponding to each load level, and the damage evolution rate correction coefficient is the ratio of the equivalent cumulative damage value after the load limit to the equivalent cumulative damage value before the load limit. Based on the damage evolution rate correction coefficient, the relationship between the bridge structural damage value of the target bridge and its service time is corrected, and based on the corrected relationship, the time when the bridge structural damage value of the target bridge reaches each functional threshold is re-determined to update the life-cycle economic evaluation results of the target bridge.

[0048] For example, the traffic load spectrum data characterizes the load distribution borne by the target bridge before the load limit, including different load levels borne by the target bridge and their corresponding traffic frequencies. The traffic load spectrum data is truncated with the load limit value as the cutoff point, and the traffic frequencies above the load limit value are removed to obtain the truncated traffic load spectrum data. This truncated traffic load spectrum data reflects the load distribution borne by the target bridge after the load limit is implemented.

[0049] For example, traffic load spectrum data, such as axle load statistics for 30 consecutive days, is obtained through a high-speed dynamic weighing system installed on the target bridge. The axle load is divided into multiple load level intervals, and the traffic frequency within each load level interval is then counted to form traffic load spectrum data. The target bridge's load limit threshold is 0.55. Based on the bearing capacity assessment results and traffic management requirements, the corresponding load limit value is determined to be 30 tons through the threshold-load mapping relationship, meaning the maximum allowable load for passage within the load-limited functional interval is 30 tons. The traffic load spectrum data is then truncated using 30 tons as the cutoff point, removing traffic frequencies exceeding 30 tons to obtain truncated traffic load spectrum data.

[0050] For example, the damage contribution can be initially determined by the axle load conversion formula or fatigue damage model, and then calibrated using measured axle load spectra and damage monitoring data to form a preset mapping relationship between load level and damage contribution. The damage evolution rate correction coefficient can be taken as the ratio of the equivalent cumulative damage value after load limiting to the equivalent cumulative damage value before load limiting. The smaller the value, the stronger the effect of load limiting on delaying damage evolution.

[0051] For example, if the load limit is 30 tons, the traffic load spectrum before the load limit is as follows: 5000 vehicles / day traveled in the 0-10 ton range, 2000 vehicles / day traveled in the 10-20 ton range, 1000 vehicles / day traveled in the 20-25 ton range, and 400 vehicles / day traveled in the load range above 30 tons. After truncating the traffic load spectrum data with 30 tons as the cutoff point, the truncated traffic load spectrum data is obtained: 5000 vehicles / day traveled in the 0-10 ton range, 2000 vehicles / day traveled in the 10-20 ton range, 1000 vehicles / day traveled in the 20-25 ton range, and 0 vehicles / day traveled in the load range above 30 tons. Furthermore, based on the preset mapping relationship between load level and damage contribution, the damage contribution corresponding to each load level is determined: the damage contribution for the load level range of 0 to 10 tons is 0.2, the damage contribution for the load level range of 10 to 20 tons is 1.0, the damage contribution for the load level range of 20 to 25 tons is 2.5, and the damage contribution for the load level range of more than 30 tons is 11.25. The damage contribution increases rapidly with the increase of load level.

[0052] Therefore, based on the passing load spectrum data, the equivalent cumulative damage value before load limitation is determined as follows: the passing frequency of each load level interval is multiplied by its damage contribution and then summed, i.e., 5000×0.2+2000×1.0+1000×2.5+400×11.25=10000. Simultaneously, based on the truncated passing load spectrum data, the equivalent cumulative damage value after load limitation is determined as follows: 5000×0.2+2000×1.0+1000×2.5+0×11.25=5500. The ratio of the equivalent cumulative damage values ​​before and after load limitation is 5500 / 10000=0.55, thus determining the damage evolution rate correction coefficient to be 0.55.

[0053] Furthermore, the damage evolution rate correction factor is 0.55. Using the example above, the damage evolution model is D(t) = 0.32 + 0.028t + 0.0006t. 2 After multiplying the rate-related terms by a correction factor of 0.55, the result is D'(t) = 0.32 + 0.0154t + 0.00033t. 2 Based on the revised damage evolution model, the times when the bridge structure damage value reaches the load limit threshold (t1 = year 11.9), the time when it reaches the traffic restriction threshold (t2 = year 19.7), and the time when it reaches the bridge closure threshold (t3 = year 24.6) are redefined. The duration of the load-limited functional zone is 7.8 years, and the duration of the traffic restriction functional zone is 4.9 years. The traffic loss cost of each traffic functional zone is updated based on the redefined times, thereby updating the life-cycle economic evaluation results of the target bridge.

[0054] Therefore, the traffic load spectrum data is truncated using the load limit value as the cutoff point, and the damage evolution rate correction coefficient is determined according to the superlinear damage contribution relationship. This allows the delaying effect of the load limit on damage evolution to be accurately measured, so that the threshold arrival time and the update of the whole life economic evaluation results are consistent with the actual damage evolution law of the target bridge, thereby providing a more accurate decision-making basis for bridge maintenance and traffic control.

[0055] In some embodiments, the life-cycle economic data includes multiple economic data records, each of which includes the bridge's service time and bridge economic parameters. In step S102, the life-cycle economic data of each engineering case is mapped to a preset coordinate system, including: determining the degradation acceleration factor of the bridge corresponding to each engineering case based on the service environment conditions of the bridge corresponding to each engineering case. The degradation acceleration factor is obtained by multiplying the temperature acceleration component, humidity acceleration component, corrosion acceleration component, and freeze-thaw acceleration component. The temperature acceleration component is obtained from the temperature of the service environment of the corresponding bridge, the humidity acceleration component is obtained from the humidity of the service environment of the corresponding bridge, the corrosion acceleration component is obtained from the corrosion level of the service environment of the corresponding bridge, and the freeze-thaw acceleration component is obtained from the freeze-thaw action of the service environment of the corresponding bridge. Based on the degradation acceleration factor corresponding to each engineering case, the bridge service time in the economic data record corresponding to each engineering case is transformed by time scaling to obtain the environmental equivalent age corresponding to the economic data record in each engineering case. Based on the structural damage detection record and the environmental equivalent age of the corresponding bridge, the bridge structural damage value corresponding to the economic data record is determined. The bridge economic parameters in the economic data record and the corresponding bridge structural damage value are correlated and mapped to the preset coordinate system to obtain the state curve of the engineering case.

[0056] It should be understood that different bridges operate under different environmental conditions, resulting in significant differences in the degree of structural degradation even after the same service time. Therefore, this embodiment of the present disclosure comprehensively considers the accelerating effects of four environmental factors—temperature, humidity, corrosion, and freeze-thaw cycles—on structural degradation. The four accelerating components are multiplied to obtain a degradation acceleration factor, thus taking into account the overall amplification effect of environmental conditions on the degradation rate. Furthermore, through time-scale transformation, the service time of the bridges in the engineering case is converted into an environmentally equivalent age comparable to that of the target bridge. Simultaneously, each detection age in the structural damage detection record undergoes time-scale transformation using the same degradation acceleration factor, yielding a discrete correspondence between the equivalent age and the bridge structural damage value. The environmentally equivalent age of the economic data record is interpolated within this discrete correspondence to determine the corresponding bridge structural damage value, thereby making the economic data of the cases obtained under different environmental conditions comparable on the damage axis.

[0057] For example, the degradation acceleration factor characterizes the degree to which environmental conditions accelerate the evolution of bridge structural damage values.

[0058] For example, the temperature acceleration component is mapped according to the annual average temperature, such as 0.85, 1.00, 1.10, and 1.20 for annual average temperatures below 10℃, between 10℃ and 15℃, between 15℃ and 20℃, and above 20℃, respectively. The humidity acceleration component is mapped according to the annual average relative humidity, such as 0.85, 1.00, 1.10, and 1.20 for annual average relative humidity below 50%, between 50% and 70%, between 70% and 85%, above 85%, or when there is alternating dry and wet conditions, respectively. The corrosion acceleration component is mapped according to the corrosion level, such as 1.00, 1.15, 1.30, and 1.50 for general atmospheric environments, marine atmospheric environments or lightly polluted environments, tidal zones or moderately polluted environments, and splash zones or heavily polluted environments, respectively. The freeze-thaw acceleration component is mapped in different tiers based on the number of freeze-thaw cycles per year. For example, when the number of freeze-thaw cycles per year is less than 5, between 5 and 50, between 50 and 100, or more than 100, the values ​​are 0.95, 1.10, 1.20, and 1.30, respectively.

[0059] For example, the bridge in engineering case G is located in a hot and humid region in southern China. Its service environment conditions are: an average annual temperature of 25℃, an average annual relative humidity of 80%, a moderate corrosion level, and a weak freeze-thaw effect. The acceleration components are: temperature acceleration component 1.20, humidity acceleration component 1.10, corrosion acceleration component 1.30, and freeze-thaw acceleration component 0.95. Therefore, the degradation acceleration factor is 1.20 × 1.10 × 1.30 × 0.95 = 1.63. An economic data record generated after 10 years of service of this bridge, after time scaling transformation, yields an environmental equivalent age of 10 × 1.63 = 16.3 years. In the structural damage detection record of this bridge, the structural damage value corresponding to an equivalent age of 12 years is 0.30, and the structural damage value corresponding to an equivalent age of 18 years is 0.48. Linear interpolation between the detection points corresponding to equivalent ages of 12 and 18 years determines that the structural damage value corresponding to an environmental equivalent age of 16.3 years is approximately 0.43. The bridge's economic parameters (e.g., annual maintenance cost of 1.2 million yuan) in the economic data record are correlated and mapped to a preset coordinate system with the bridge structural damage value of 0.42, resulting in a data point on the state curve of this engineering case. This process is repeated for all economic data records, and each data point is fitted to obtain the state curve for engineering case E.

[0060] Therefore, by normalizing the case economic data using degradation acceleration factors and environmental equivalent ages, the interference of different service environments on the comparability of cases is eliminated, and the accuracy of case knowledge transfer is improved.

[0061] In some embodiments, step S102, determining the selection weight of each engineering case based on preset rules, includes: forming a candidate segment sequence from the segments to be affected in each engineering case according to the increasing order of bridge structure damage values, and performing trial splicing on each candidate segment sequence to determine the constraint correction amount at each splicing position after trial splicing. The segment to be affected is the curve segment located to the right of the state point in the state curve. The candidate segment sequence is formed sequentially from adjacent segments to be affected within the bridge structure damage value interval. The constraint correction amount includes the difference in bridge structure damage values ​​on both sides of the splicing position and the difference in economic parameters on both sides of the splicing position. Based on the constraint correction amount at each splicing position, the selection weight of the engineering cases on both sides of the corresponding splicing position is determined. The selection weight is inversely proportional to the constraint correction amount.

[0062] It should be understood that the target state curve after splicing should remain continuous and smooth at the splicing position. The smaller the difference between the damage values ​​of the bridge structures on both sides of the splicing position and the difference between the economic parameters, the smoother the connection at the splicing position, and the closer the corresponding target state curve is to the actual economic evolution process of the target bridge. Therefore, in this embodiment, the constraint correction amount is used as a measure of splicing quality, and the selected weight is inversely proportional to the constraint correction amount. The candidate segment sequence is composed of adjacent segments to be generated from the bridge structure damage value intervals in sequence, that is, the end of the damage value interval covered by the previous segment to be generated and the starting point of the damage value interval covered by the next segment to be generated are adjacent on the damage axis to ensure that the splicing progresses continuously along the damage axis.

[0063] For example, the sections to be affected in engineering cases E, F, and G are as follows: the section to be affected in engineering case E covers the damage value range [0.32, 0.55], the section to be affected in engineering case F covers the damage value range [0.53, 0.75], and the section to be affected in engineering case G covers the damage value range [0.72, 0.95]. Each section to be affected is arranged in order of increasing bridge structural damage value to form a candidate section sequence EFG. Trial splicing is performed on the candidate section sequence: at the splicing position between the sections to be affected in engineering cases E and F, the difference in bridge structural damage values ​​on both sides is 0.02, and the difference in economic parameters on both sides is 80,000 yuan / year; at the splicing position between the sections to be affected in engineering cases F and G, the difference in bridge structural damage values ​​on both sides is 0.05, and the difference in economic parameters on both sides is 210,000 yuan / year.

[0064] Then, the damage difference and economic parameter difference at each splicing position are normalized to the maximum value among all splicing positions, and then weighted and summed according to preset weight coefficients to obtain the comprehensive constraint correction amount for each splicing position. For example, if the preset weight coefficient is 0.5, the comprehensive constraint correction amount at splicing position EF is 0.5×(0.02 / 0.05)+0.5×(8 / 21)=0.39, and the comprehensive constraint correction amount at splicing position FG is 0.5×(0.05 / 0.05)+0.5×(21 / 21)=1.00. Based on the comprehensive constraint correction amount at each splicing position, the selection weights of the engineering cases on both sides of the corresponding splicing position are determined. The selection weights are inversely proportional to the comprehensive constraint correction amount: the ratio of the selection weights at splicing position EF to those at splicing position FG is 1 / 0.39:1 / 1.00, or approximately 0.72:0.28. Project case E only involves the splicing position EF, and its selection weight is 0.72; project case F involves both splicing positions EF and FG, and its selection weight is the average of the two position weights: (0.72 + 0.28) / 2 = 0.50; project case G only involves the splicing position FG, and its selection weight is 0.28. After normalization, the selection weights of project cases E, F, and G are 0.48, 0.33, and 0.19, respectively.

[0065] Therefore, by using the smoothness of the splicing position as the basis for case selection, the target state curve obtained by splicing is continuous and smooth, avoiding the deviation in economic evaluation caused by the abrupt changes in the connection of case segments, thus providing a more accurate basis for decision-making in bridge maintenance.

[0066] In some embodiments, step S102, determining the selection weight of each engineering case based on preset rules, includes: determining the structural similarity, environmental condition similarity, and operational characteristic similarity between the target bridge and the corresponding bridges of each engineering case; and, based on preset weight coefficients, weighting and summing the structural similarity, environmental condition similarity, and operational characteristic similarity to obtain the comprehensive similarity between the target bridge and the engineering case; and determining the selection weight of the engineering case based on the comprehensive similarity, the data completeness of the engineering case, and the data timeliness value; wherein, the data completeness is obtained by dividing the number of valid data items in the engineering case by the total number of data items, and the valid data items include non-empty economic data, defect data, and risk data fields that have not been identified as abnormal data, and the data timeliness value is obtained by the time difference between the data formation time of the engineering case and the evaluation time of the target bridge.

[0067] It should be understood that the higher the overall similarity between the target bridge and the corresponding bridge in the engineering case, the greater the reference value of the life-cycle economic data of the engineering case for the target bridge; the higher the data completeness, the higher the usability of the engineering case data; the smaller the data timeliness value, the closer the interval between the engineering case data and the current evaluation time, and the closer the current economic and operating conditions reflected by the data. Therefore, in this embodiment of the disclosure, the overall similarity, data completeness, and data timeliness value are used to jointly determine the selection weight of the engineering case.

[0068] For example, based on the bridge type, structural system, span range and material type of the target bridge, preliminary screening of engineering cases in the bridge life-cycle economic knowledge base is carried out to construct a set of candidate engineering cases. Comprehensive similarity calculation is performed on the candidate engineering cases to reduce computational complexity and improve case matching efficiency.

[0069] First, the multidimensional similarity between the target bridge and each engineering case is calculated, including structural feature similarity, environmental condition similarity, and operational feature similarity. The comprehensive similarity is expressed as:

[0070] in, Indicates the overall similarity. For structural feature similarity, For environmental condition similarity, For operational feature similarity, , and These are the corresponding weighting coefficients.

[0071] Based on this, the engineering cases were ranked, and the top cases with the highest overall similarity were selected. m Several engineering cases were selected as candidate reference engineering case studies. m The size can be dynamically determined based on the size of the knowledge base, and is generally taken as... When the overall similarity of engineering cases is lower than a preset threshold... At that time, no knowledge transfer will be involved. If the number of engineering cases that meet the conditions is insufficient, reference cases will be added in descending order of comprehensive similarity until the preset number is reached.

[0072] Furthermore, to improve the reliability and robustness of knowledge transfer, a data quality evaluation mechanism is introduced to calculate the data completeness and timeliness of each engineering case.

[0073] Data completeness is used to characterize the completeness of engineering case data, and its expression is:

[0074] in, For data integrity, This refers to the number of valid data items (i.e., the number of non-empty and usable data fields such as economic, disease, and risk data). The total number of data items (based on the complete data template defined in the knowledge base).

[0075] The data timeliness value is used to characterize the time validity of engineering case data, and its expression is:

[0076] in, For the first i Data timeliness values ​​for individual project cases The time difference (in years) between the time when the engineering case data was generated and the time when the target bridge was evaluated. The time-based benchmark period is usually 10 years, but can be adjusted proportionally according to the bridge's design life.

[0077] The knowledge credibility of engineering cases is constructed based on comprehensive similarity, data completeness, and data timeliness, which is used to characterize the reliability of knowledge transfer from engineering cases to the target bridge.

[0078] in, For the first i The credibility of knowledge in individual engineering case studies , and These are the corresponding weighting coefficients.

[0079] Based on this, the knowledge credibility is normalized to obtain the selection weight of each engineering case, the expression of which is:

[0080] in, For the first i The selection weight of each engineering case n This represents the total number of engineering cases.

[0081] Therefore, the selection weights are determined by comprehensively considering three dimensions: case similarity, data completeness, and data timeliness. This ensures that the case data used in the splicing process not only matches the target bridge but is also complete and relevant to the current evaluation point in time. This further improves the reliability of case knowledge transfer, thereby enhancing the accuracy of the economic evaluation of the bridge's entire life cycle and providing a more accurate basis for bridge maintenance decisions.

[0082] In some embodiments, new structural state data of the target bridge can be acquired, and based on the new structural state data, the measured bridge structural damage value of the target bridge can be determined. The measured bridge structural damage value is compared with the predicted bridge structural damage value obtained through the target state curve to obtain the comparison result. Based on the comparison result, the degradation rate parameter and state regression amount corresponding to each engineering case are corrected, and the selection weight of each engineering case is updated based on the corrected degradation rate parameter and state regression amount. The degradation rate parameter represents the rate of change of the bridge structural damage value corresponding to the engineering case with the bridge service time, and the state regression amount represents the extent to which the maintenance event corresponding to the engineering case reduces the bridge structural damage value. The updated selection weight is used to update the life-cycle economic evaluation result of the target bridge.

[0083] It should be understood that the target state curve is obtained based on historical engineering case data, and there may be discrepancies between the damage evolution pattern it reflects and the actual degradation process of the target bridge. For example, differences between the case bridge and the target bridge in terms of structural details, material properties, actual load levels, and changes in the service environment or operating conditions of the target bridge may all lead to inconsistencies between the predicted bridge structural damage value and the actual detection result. Therefore, in this embodiment of the present disclosure, after obtaining new structural state data, the measured bridge structural damage value is compared with the predicted bridge structural damage value to identify the direction and degree of deviation between the case knowledge transfer results and the actual degradation process of the target bridge, thereby correcting the selection weight of the engineering cases.

[0084] For example, new structural status data is added, including periodic inspection data and health monitoring data from the new inspection cycle. The measured bridge structural damage value is determined based on this new structural status data, following the same procedure as in step S101. For instance, the measured bridge structural damage value is 0.38. Simultaneously, the predicted bridge structural damage value for the current moment is determined by analyzing the relationship between the bridge structural damage value corresponding to the target status curve and service time. For instance, the predicted bridge structural damage value is 0.35. The measured bridge structural damage value is compared with the predicted bridge structural damage value. The measured value is 0.03 higher than the predicted value, and the comparison result is obtained.

[0085] For example, if the comparison results indicate that the measured bridge structural damage value is greater than the predicted bridge structural damage value, it means that the actual degradation process of the target bridge is faster than the statistical law of cases. In this case, the degradation rate parameter is increased and the state regression amount is decreased, which respectively accelerates damage evolution and reduces repair recovery, thus increasing the predicted damage value. If the comparison results indicate that the measured bridge structural damage value is less than the predicted bridge structural damage value, it means that the actual degradation process of the target bridge is slower than the statistical law of cases. In this case, the degradation rate parameter is decreased and the state regression amount is increased, thus decreasing the predicted damage value. The correction magnitude of the degradation rate parameter can be proportional to the deviation between the measured and predicted bridge structural damage values, and an upper limit can be set for the single correction magnitude to prevent overcorrection. The correction magnitude of the state regression amount can be less than the correction magnitude of the degradation rate parameter. For example, it can be a preset proportion of the degradation rate parameter correction magnitude (e.g., 0.5), or determined based on the ratio of the measured magnitude of the reduction in bridge structural damage value caused by actual repair events to the state regression amount in case statistics. An upper limit can be set for the single correction magnitude to reduce the correction uncertainty caused by the sparse sample of repair events.

[0086] For example, after correcting the degradation rate parameter and the state regression amount, the measured degradation rate of the target bridge is determined by the measured bridge structural damage values ​​over two consecutive detection cycles. The degradation matching degree between the corrected degradation rate parameter and the measured degradation rate is calculated for each engineering case; the closer the degradation rate parameter is to the measured degradation rate, the higher the degradation matching degree. Simultaneously, the corrected state regression amount is compared with the actual maintenance regression amount of the target bridge to determine the maintenance matching degree. The degradation matching degree and maintenance matching degree are incorporated into the selection weights. For example, a weighted sum of comprehensive similarity, degradation matching degree, maintenance matching degree, data completeness, and data timeliness is used as the selection weight. Based on the updated selection weights, the target segment to be affected is reselected, and the target state curve is spliced ​​to update the life-cycle economic evaluation results of the target bridge. This allows the evaluation results to dynamically evolve with the actual structural state of the target bridge, thereby maintaining consistency between the evaluation results and the actual degradation process, improving the accuracy of the life-cycle economic evaluation results in long-term use, and ultimately providing a more accurate long-term decision-making basis for bridge maintenance.

[0087] In some embodiments, the entire lifecycle of each engineering case is decomposed into maintenance event units. Each maintenance event unit includes a trigger state and a state rollback amount. The trigger state represents the bridge structural damage value when the corresponding maintenance event occurs, and the state rollback amount represents the extent to which the bridge structural damage value is reduced by the corresponding maintenance event. In step S102, based on the state point, state curve, and selection weight, a target state curve covering the remaining lifecycle of the target bridge is determined, including: dividing each state curve into completed segments and pending segments along the horizontal axis with the state point as the boundary; selecting target maintenance event units whose trigger state matches the state point in the maintenance event units corresponding to each engineering case based on the selection weight; updating the state point based on the state rollback amount of the target maintenance event unit to obtain an updated state point; determining the target pending segment based on the pending segment corresponding to the engineering case to which the target maintenance event unit belongs, located between the updated state point and the trigger state of the target maintenance event unit; and obtaining the target state curve covering the remaining lifecycle of the target bridge based on the target pending segment.

[0088] It should be understood that maintenance events (such as component replacement, reinforcement, and repair) throughout the bridge's lifespan cause a step-like regression in the bridge's structural damage value. The entire lifespan of the engineering cases exhibits an alternating sequence of continuous damage growth, step-like repair regression, and subsequent growth. Therefore, this embodiment decomposes the entire lifespan of each engineering case into maintenance event units based on maintenance events, characterizing each maintenance event with a trigger state and a state regression amount. The trigger state is the bridge structural damage value at the time the maintenance event occurs, i.e., the trigger node for maintenance. The state regression amount is the extent to which the maintenance event reduces the bridge structural damage value. The state point of the target bridge represents its current bridge structural damage value. For maintenance event units where the trigger state and state point match, the engineering case to which it belongs experiences a maintenance scenario similar to the current state of the target bridge. The pending segment corresponding to this maintenance event unit has reference value for the economic evolution of the target bridge over its remaining lifespan.

[0089] For example, a match is determined when the difference between the bridge structural damage value at the trigger state and the state point is less than a preset matching threshold. When multiple matching maintenance event units exist, the target maintenance event unit is selected based on the selection weight, with higher weights taking priority.

[0090] For example, the entire lifecycle of engineering case H is decomposed into three maintenance event units: maintenance event unit U1 has a trigger state of 0.30 and a state rollback amount of 0.08, corresponding to one prestressed reinforcement maintenance event; maintenance event units U2 and U3 correspond to two other maintenance events, respectively. The target bridge's state point is P(0.32, 85). The difference between its bridge structural damage value of 0.32 and the trigger state of maintenance event unit U1 (0.30) is 0.02, which is less than the preset matching threshold of 0.05, thus a match is determined; however, it does not match the trigger states of maintenance event units U2 and U3. The target maintenance event unit is selected based on a selection weight. For example, the selection weight for engineering case H is 0.45, the highest among all matching cases, so maintenance event unit U1 is selected as the target maintenance event unit. The state point is updated based on the state rollback amount of target maintenance event unit U1 (0.08), and the updated bridge structural damage value is 0.32 - 0.08 = 0.24. Based on the segment to be generated corresponding to engineering case H, located between the update state point (i.e., bridge structure damage value 0.24) and the trigger state of the target maintenance event unit U1 (i.e., bridge structure damage value 0.30), the target segment to be generated is determined. This target segment to be generated represents the evolution law of economic parameters of engineering case H during the period from 0.24 to 0.30 before the prestressed reinforcement maintenance occurs. It is used as the target segment to be generated to participate in the splicing of the target state curve.

[0091] Therefore, by characterizing the step correction of the damage path by maintenance through maintenance event units, the selected target segment to be affected can reflect the impact of maintenance events on economic evolution, further improving the accuracy of the target state curve in characterizing the economic evolution of the remaining lifespan, thus obtaining more accurate economic evaluation results of the bridge's whole lifespan, and providing more accurate decision-making basis for bridge maintenance investment and traffic restriction.

[0092] In some embodiments, in step S104, multiple types of differences between the target bridge and the corresponding bridges in each engineering case are determined, and feature weights corresponding to each type of difference are determined based on the entropy weight method. The multiple types of differences are weighted based on the feature weights to obtain bridge difference features. The multiple types of differences include differences in structural parameters, material parameters, environmental conditions, traffic loads, and operating conditions. The bridge difference features are input into a domain adaptive model to obtain the migration correction amount of the bridge economic parameters. The domain adaptive model is trained using the feature data of the engineering cases as source domain samples and the feature data of the target bridge as target domain samples. Based on the migration correction amount, the bridge economic parameters determined along the target state curve are corrected, and the full life-cycle economic evaluation result of the target bridge is determined based on the corrected bridge economic parameters and the traffic loss cost.

[0093] It should be understood that there are differences between the target bridge and the corresponding bridge in the engineering case in terms of structure, materials, environment, traffic, and operation. These differences can lead to deviations when the economic parameters of the case are transferred to the target bridge. Therefore, this embodiment of the disclosure uses the entropy weight method to determine the feature weights based on the information content of each difference dimension. The smaller the information entropy of the difference dimension, the richer its distinguishing information, and the larger the corresponding feature weight. Then, the weighted fusion of multiple differences into bridge difference features is input into the domain adaptive model. The domain adaptive model learns the mapping relationship of economic parameters between the engineering case domain and the target bridge domain by minimizing the difference between the feature distribution of the source domain and the feature distribution of the target domain, and outputs the transfer correction amount of the bridge economic parameters, thereby performing inter-domain correction on the economic parameters obtained from the case transfer.

[0094] For example, the various differences between the target bridge and the corresponding bridges in each of the described engineering cases can be obtained based on the basic information of the target bridge. This basic information includes, but is not limited to: bridge structural parameters, used to characterize the structural composition and stress characteristics of the target bridge, including bridge type, span arrangement, structural system, pier and abutment forms, and superstructure and substructure construction parameters, derived from bridge design drawings, BIM models, or as-built data; material parameters, used to characterize the performance of the main structural materials of the bridge, including concrete strength grade, steel type and strength grade, durability indicators, and material aging characteristics, derived from design documents or test reports; environmental condition parameters, used to characterize the impact of the service environment on the bridge, including environmental factors such as temperature, humidity, corrosion level, freeze-thaw environment, wind load environment, and hydrological conditions, derived from environmental monitoring data; traffic load parameters, used to characterize the level of traffic load the bridge bears, including design traffic level, traffic flow, vehicle axle load distribution, and overload probability, derived from traffic survey data or traffic monitoring systems; and operational condition parameters, used to characterize the bridge's operational status, including operational years, maintenance level, inspection frequency, maintenance strategies, and traffic control conditions, derived from the bridge operation management system.

[0095] The aforementioned basic information is obtained through design documents, testing data, operation and management data, and historical maintenance records, and is then standardized so that data from different sources can be used for subsequent knowledge transfer calculations and life-cycle economic evaluation under a unified dimension and unified expression system.

[0096] For example, the domain adaptation model includes a feature extraction network, a domain classifier, and a regression prediction network. The feature extraction network extracts domain-invariant feature representations from the bridge's differential features. The domain classifier, connected to the feature extraction network via a gradient inversion layer, distinguishes between source and target domain samples. The regression prediction network outputs transfer-correction amounts for the bridge's economic parameters based on these domain-invariant feature representations. During training, the first training objective is to minimize the maximum mean difference between the source and target domain feature distributions, and the second training objective is to minimize the prediction error of the source domain sample's economic parameters.

[0097] For example, the domain-adaptive transfer learning model achieves feature alignment between the source and target domains by jointly optimizing the economic parameter prediction loss and the domain distribution difference loss. Its optimization objective function is:

[0098] in, To predict losses based on economic parameters, The loss is the difference in domain distribution between the source and target domains. Adapt weight coefficients to the domain.

[0099] 099. For example, multiple types of differences include structural parameter differences, such as span and structural form differences; material parameter differences, such as concrete strength grade differences; environmental condition differences, such as temperature and humidity differences; traffic load differences, such as daily traffic volume and heavy vehicle ratio differences; and operational condition differences, such as maintenance management level differences. After normalizing the values ​​of these multiple types of differences, a difference vector is obtained. The feature weights for each difference dimension are determined using the entropy weight method. For example, the feature weight for structural parameter differences is 0.30, the feature weight for material parameter differences is 0.25, the feature weight for environmental condition differences is 0.20, the feature weight for traffic load differences is 0.15, and the feature weight for operational condition differences is 0.10. Based on these feature weights, the multiple types of differences are weighted to obtain bridge difference features. These bridge difference features are input into a domain adaptive model to obtain the migration correction amount for the bridge's economic parameters. For example, the migration correction amount is -120,000 yuan / year, meaning that the economic parameters obtained from the case migration need to be reduced by 120,000 yuan / year due to inter-domain differences. The bridge economic parameters determined along the target state curve are corrected based on the migration correction amount. Based on the corrected bridge economic parameters and the traffic loss cost of each traffic functional zone, the full life-cycle economic evaluation result of the target bridge is determined.

[0100] Therefore, by integrating multiple types of differences through the entropy weight method and performing inter-domain correction through a domain adaptive model, the distribution offset between the case domain and the target domain is eliminated, the accuracy of the migration economic parameters is improved, and the reliability of the whole life economic evaluation results is improved, thus providing a more accurate decision-making basis for bridge maintenance and traffic restriction control.

[0101] In some embodiments, step S104, based on the target state curve and the traffic loss cost, determines the life-cycle economic evaluation result of the target bridge, including: determining the first bridge economic parameter based on the target state curve; determining the probability of risk events and the economic loss of risk events based on the target bridge's defect detection records, operation and maintenance records, and risk event statistics of each engineering case; correcting the first bridge economic parameter based on the probability of risk events and the economic loss of risk events to obtain the second bridge economic parameter; correcting the second bridge economic parameter based on the time decay characteristics and environmental differences of the engineering cases corresponding to the target state curve to obtain the third bridge economic parameter; and obtaining the life-cycle economic evaluation result of the target bridge based on the third bridge economic parameter, the economic parameters of the target bridge in the current stage, and the traffic loss cost of each traffic functional area.

[0102] For example, based on the relationship between the bridge structural damage value of the target bridge and the service time, the bridge structural damage value corresponding to each moment in the remaining service life is located on the target state curve, and the bridge economic parameter at the corresponding damage value is taken to obtain the economic parameter for each year in the remaining service life, which is used as the first bridge economic parameter.

[0103] For example, based on the target state curve, the first economic parameter for the bridge is determined to be 1.2 million yuan per year for each year of its remaining lifespan. Based on the target bridge's defect detection records (e.g., excessive bearing deformation, continuous crack development), operation and maintenance records (e.g., bearings not yet replaced), and the statistical results of risk events in various engineering cases (e.g., the probability of bearing failure risk events is 0.02 per year, and the economic loss per risk event is 3 million yuan), the probability of risk events occurring for the target bridge is determined to be 0.02 per year, the economic loss per risk event is 3 million yuan, and the expected risk cost is 0.02 × 300 = 60,000 yuan per year. Adding this expected risk cost to the first bridge's economic parameter yields the second bridge's economic parameter of 1.26 million yuan per year. Then, based on the time decay characteristics and environmental differences of the engineering cases corresponding to the target state curve, the second bridge's economic parameter is adjusted. Regarding the time decay characteristics, the time difference between the data formation time of the engineering cases corresponding to the target state curve and the evaluation time of the target bridge is 3 years, the annual decay rate is taken as 0.02, and the time decay correction coefficient is (1-0.02). 3≈0.94. Regarding environmental differences, the case bridge is in a general atmospheric environment with a degradation acceleration factor of 1.10, while the target bridge is in a marine atmospheric environment with a degradation acceleration factor of 1.30. Taking a cost-environmental sensitivity coefficient of 0.3, the environmental difference correction coefficient is 1 + 0.3 × (1.30 / 1.10 - 1) ≈ 1.05. After time decay correction and environmental difference correction, the economic parameters of the second bridge are used to obtain the economic parameters of the third bridge: 126 × 0.94 × 1.05 ≈ 1.24 million yuan / year. Finally, based on the economic parameters of the third bridge, the economic parameters of the target bridge in its current stage, and the traffic loss costs of each traffic function zone, the life-cycle economic evaluation results of the target bridge are obtained. Among them, the economic parameters of the target bridge in its current stage are, for example, the present value of the cumulative economic costs incurred from construction to the current evaluation time, which is 32 million yuan. The traffic loss costs of each traffic function zone are determined according to step S103.

[0104] Therefore, by correcting for risk events, data timeliness, and environmental differences, the life-cycle economic evaluation results can simultaneously reflect the individual risk characteristics of the target bridge and the timeliness characteristics of case data. Combined with the economic parameters of the already occurred stages, a complete evaluation covering the entire life cycle is formed, further improving the accuracy of the evaluation results and thus providing a more accurate basis for bridge maintenance investment decisions.

[0105] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0106] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0107] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

Claims

1. A dynamic evaluation method for the economic efficiency of bridges throughout their entire life cycle based on knowledge transfer from engineering cases, characterized in that, include: Based on the structural state data of the target bridge, the state point corresponding to the target bridge is determined in a preset coordinate system. The preset coordinate system uses the bridge structural damage value as the horizontal axis and the bridge economic parameter as the vertical axis. The horizontal axis is divided into different traffic function intervals by the functional thresholds of the target bridge. The functional thresholds include load limit threshold, traffic restriction threshold and bridge closure threshold. The full life-cycle economic data of each engineering case are mapped to the preset coordinate system to obtain multiple state curves. The selection weight of each engineering case is determined based on preset rules. Based on the state points, the state curves and the selection weights, a target state curve covering the remaining life of the target bridge is determined. Based on the target state curve and the relationship between the bridge structure damage value of the target bridge and the service time, the time when the bridge structure damage value of the target bridge reaches each of the functional thresholds is determined, and based on the time, the passage loss cost of the corresponding passage functional section is determined. Based on the target state curve and the traffic loss cost, the life-cycle economic evaluation result of the target bridge is determined.

2. The method for dynamic evaluation of bridge life-cycle economic efficiency based on engineering case knowledge transfer as described in claim 1, characterized in that, The method further includes: Obtain the detour route information of adjacent bridges in the road network where the target bridge is located, and determine the detour traffic volume of the restricted vehicles on each of the adjacent bridges based on the detour route information. The adjacent bridges are the bridges that bear the detour traffic of the target bridge, and the detour route information represents the detour distance from the target bridge to the adjacent bridge and the traffic capacity of the corresponding detour route. Based on the detour traffic volume, the increase in traffic load of each of the adjacent bridges is determined, and the relationship between the bridge structural damage value of each of the adjacent bridges and the service time is corrected based on the increase in traffic load. Based on the corrected relationship, the time when the bridge structural damage value of each of the adjacent bridges reaches the corresponding functional threshold is determined. Based on the time when each of the adjacent bridges reaches its corresponding functional threshold and the detour traffic volume, the additional traffic loss cost is determined, and based on the additional traffic loss cost, the life-cycle economic evaluation result of the target bridge is updated.

3. The method for dynamic evaluation of bridge life-cycle economic efficiency based on engineering case knowledge transfer as described in claim 1, characterized in that, The method further includes: The traffic load spectrum data of the target bridge is obtained, and the load limit value is determined based on the load limit threshold. The traffic load spectrum data is truncated with the load limit value as the cutoff point to obtain the truncated traffic load spectrum data. The load limit value is the maximum load value allowed to pass through the load limit functional area corresponding to the load limit threshold. Based on the passing load spectrum data and the truncated passing load spectrum data, the equivalent cumulative damage value before and after the load limit is determined, and the damage evolution rate correction coefficient is determined based on the equivalent cumulative damage value. The equivalent cumulative damage value is obtained by multiplying and summing the passing frequency and damage contribution corresponding to each load level, and the damage evolution rate correction coefficient is the ratio of the equivalent cumulative damage value after the load limit to the equivalent cumulative damage value before the load limit. Based on the damage evolution rate correction coefficient, the relationship between the bridge structural damage value of the target bridge and its service time is corrected, and based on the corrected relationship, the time when the bridge structural damage value of the target bridge reaches each of the functional thresholds is re-determined, so as to update the life-cycle economic evaluation results of the target bridge.

4. The method for dynamic evaluation of bridge life-cycle economic efficiency based on engineering case knowledge transfer according to any one of claims 1-3, characterized in that, The full life-cycle economic data includes multiple economic data records, each of which includes the bridge's service time and economic parameters. Mapping the full life-cycle economic data of each engineering case to the preset coordinate system includes: Based on the service environment conditions of the bridges corresponding to each engineering case, the degradation acceleration factor of the bridges corresponding to each engineering case is determined. The degradation acceleration factor is obtained by multiplying the temperature acceleration component, humidity acceleration component, corrosion acceleration component and freeze-thaw acceleration component. The temperature acceleration component is obtained from the temperature of the service environment of the corresponding bridge. The humidity acceleration component is obtained from the humidity of the service environment of the corresponding bridge. The corrosion acceleration component is obtained from the corrosion level of the service environment of the corresponding bridge. The freeze-thaw acceleration component is obtained from the freeze-thaw effect of the service environment of the corresponding bridge. Based on the degradation acceleration factor corresponding to each of the engineering cases, the bridge service time in the economic data records corresponding to each of the engineering cases is transformed by time scaling to obtain the environmental equivalent age corresponding to the economic data records in each of the engineering cases. Based on the structural damage detection records of the corresponding bridge and the environmental equivalent age, the bridge structural damage value corresponding to the economic data record is determined. The bridge economic parameters in the economic data records are associated and mapped with the corresponding bridge structural damage value to the preset coordinate system to obtain the state curve of the engineering case.

5. The method for dynamic evaluation of bridge life-cycle economic efficiency based on engineering case knowledge transfer according to any one of claims 1-3, characterized in that, The determination of the selection weight of each engineering case based on preset rules includes: The segments to be affected in each of the aforementioned engineering cases are arranged into a candidate segment sequence in order of increasing bridge structural damage value. Trial splicing is performed on each candidate segment sequence to determine the constraint correction amount at each splicing position after trial splicing. The segment to be affected is the curve segment located to the right of the state point in the state curve. The candidate segment sequence is formed by adjacent segments to be affected in the bridge structural damage value interval in sequence. The constraint correction amount includes the difference in bridge structural damage value on both sides of the splicing position and the difference in economic parameters on both sides of the splicing position. Based on the constraint correction amount at each splicing position, the selection weight of the engineering cases on both sides of the corresponding splicing position is determined, wherein the selection weight is inversely proportional to the constraint correction amount.

6. The method for dynamic evaluation of bridge life-cycle economic efficiency based on engineering case knowledge transfer according to any one of claims 1-3, characterized in that, The determination of the selection weight of each engineering case based on preset rules includes: The structural similarity, environmental condition similarity, and operational characteristic similarity between the target bridge and the corresponding bridges of each of the engineering cases are determined. Based on preset weight coefficients, the structural similarity, environmental condition similarity, and operational characteristic similarity are weighted and summed to obtain the comprehensive similarity between the target bridge and the engineering cases. Based on the comprehensive similarity, the data completeness and data timeliness of the engineering cases, the selection weight of the engineering cases is determined. The data completeness is obtained by dividing the number of valid data items in the engineering case by the total number of data items. The valid data items include non-empty economic data, defect data, and risk data fields that are not identified as abnormal data. The data timeliness value is obtained by the time difference between the data formation time of the engineering case and the evaluation time of the target bridge.

7. The method for dynamic evaluation of bridge life-cycle economic efficiency based on engineering case knowledge transfer according to any one of claims 1-3, characterized in that, The method further includes: Acquire new structural state data of the target bridge, and based on the new structural state data, determine the measured bridge structural damage value of the target bridge. Compare the measured bridge structural damage value with the predicted bridge structural damage value obtained through the target state curve to obtain the comparison result. Based on the comparison results, the degradation rate parameter and state regression amount corresponding to each engineering case are corrected, and the selection weight of each engineering case is updated based on the corrected degradation rate parameter and state regression amount. The degradation rate parameter represents the rate of change of the bridge structural damage value corresponding to the engineering case with the bridge service time, and the state regression amount represents the extent to which the maintenance event corresponding to the engineering case reduces the bridge structural damage value. The updated selection weight is used to update the life-cycle economic evaluation results of the target bridge.

8. The method for dynamic evaluation of bridge life-cycle economic efficiency based on engineering case knowledge transfer according to any one of claims 1-3, characterized in that, The entire lifecycle of each engineering case is decomposed into maintenance event units. Each maintenance event unit includes a trigger state and a state rollback amount. The trigger state represents the bridge structural damage value when the corresponding maintenance event occurs, and the state rollback amount represents the extent to which the bridge structural damage value is reduced by the corresponding maintenance event. The determination of a target state curve covering the remaining lifecycle of the target bridge based on the state points, the state curve, and the selection weights includes: Using the state point as the boundary, each state curve is divided into a completed segment and a segment to be generated along the horizontal axis. Based on the selection weight, a target maintenance event unit whose trigger state matches the state point is selected from the maintenance event units corresponding to each engineering case. The state point is updated based on the state rollback amount of the target maintenance event unit to obtain the updated state point. Based on the pending segment corresponding to the engineering case to which the target maintenance event unit belongs, located between the update state point and the trigger state of the target maintenance event unit, the target pending segment is determined, and based on the target pending segment, the target state curve covering the remaining lifespan of the target bridge is obtained.

9. The method for dynamic evaluation of bridge life-cycle economic efficiency based on engineering case knowledge transfer according to any one of claims 1-3, characterized in that, The determination of the life-cycle economic evaluation result of the target bridge based on the target state curve and the traffic loss cost includes: The target bridge and the corresponding bridges in each of the engineering cases are identified as having multiple types of differences. The feature weights corresponding to the multiple types of differences are determined based on the entropy weight method. The multiple types of differences are then weighted based on the feature weights to obtain the bridge difference features. The multiple types of differences include differences in structural parameters, material parameters, environmental conditions, traffic loads, and operating conditions. The bridge difference features are input into the domain adaptive model to obtain the migration correction amount of the bridge economic parameters. The domain adaptive model is obtained by training with the feature data of the engineering case as the source domain sample and the feature data of the target bridge as the target domain sample. Based on the migration correction amount, the bridge economic parameters determined along the target state curve are corrected, and based on the corrected bridge economic parameters and the traffic loss cost, the life-cycle economic evaluation result of the target bridge is determined.

10. The method for dynamic evaluation of bridge life-cycle economic efficiency based on engineering case knowledge transfer according to any one of claims 1-3, characterized in that, The determination of the life-cycle economic evaluation result of the target bridge based on the target state curve and the traffic loss cost includes: Based on the target state curve, determine the first bridge economic parameters; Based on the defect detection records, operation and maintenance records of the target bridge, and the statistical results of risk events in each of the engineering cases, the probability of occurrence of risk events and the economic loss of risk events for the target bridge are determined. Based on the probability of occurrence of risk events and the economic loss of risk events, the first bridge economic parameters are corrected to obtain the second bridge economic parameters. Based on the time decay characteristics and environmental differences of the engineering cases corresponding to the target state curve, the second bridge economic parameters are corrected to obtain the third bridge economic parameters. Based on the third bridge economic parameters, the economic parameters of the target bridge in the current stage, and the traffic loss cost, the full life-cycle economic evaluation result of the target bridge is obtained.

Images