Intelligent road and bridge maintenance supervision method based on digital twinning

Abstract

The invention relates to the technical field of digital twinning, in particular to a road bridge intelligent maintenance supervision method based on digital twinning, and the method comprises the steps: constructing a digital twinning model of a to-be-analyzed bridge; collecting monitoring data arranged at each monitoring point of a to-be-analyzed bridge, and synchronizing the monitoring data to the digital twin model; according to the structure mutation mean difference coefficient and the structure foundation differential settlement characterization coefficient, determining whether a division mode is adjusted from grid division to correlation division; determining a supervision state based on the damage index of the key area and the monitoring abnormity degree, and determining to perform maintenance or maintenance analysis on the to-be-analyzed bridge according to the supervision state; and if maintenance is not needed, determining whether to adjust the duration of the supervision period based on the fitting deviation value. The maintenance decision-making precision can be improved.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, and in particular to an intelligent maintenance and monitoring method for roads and bridges based on digital twins. Background Technology

[0002] As the core of transportation infrastructure, roads and bridges face the combined effects of repeated traffic loads, temperature and humidity changes, and acid rain erosion during long-term service. This makes them prone to damage such as crack propagation, steel reinforcement corrosion, and bearing settlement, directly impacting traffic safety and structural durability. With the expansion of road networks and the increase in service life, traditional maintenance and supervision models rely heavily on manual inspections and periodic sampling tests. These methods suffer from drawbacks such as long inspection cycles, data fragmentation, delayed defect identification, and strong subjectivity in maintenance decisions, resulting in poor maintenance efficiency and accuracy. Therefore, improving the intelligence level of road and bridge maintenance supervision and the timeliness of defect early warning is a technical problem that urgently needs to be solved by those skilled in the art.

[0003] Chinese Patent Publication No. CN108873921A discloses a bridge inspection method and system based on digital twin technology. The method includes: real-time acquisition of strain distribution values, vibration spectrum values, and environmental load spectrum values ​​through a sensor network deployed on the physical bridge, generating a structural response dataset and synchronizing it to the digital twin; calculating damage index values ​​and cumulative damage values ​​based on the structural response dataset; inputting the damage index values ​​and cumulative damage values ​​into preset safety criteria to calculate safety margin coefficient values ​​and failure risk level values; calculating the remaining life prediction value based on the safety margin coefficient value and environmental load spectrum values, and simultaneously correcting the degradation rate value of the digital twin; generating priority maintenance instructions based on the failure risk level value, remaining life prediction value, and safety margin coefficient value, and feeding back maintenance effect data to the digital twin to complete the update after execution. However, the above solution has the following problems: it cannot dynamically adjust the analysis method according to the actual state of the bridge, lacks adaptability to the refined and dynamic maintenance supervision of complex bridge structures, and is difficult to accurately capture the early characteristics of local bridge defects, thus leading to poor accuracy in maintenance decisions. Summary of the Invention

[0004] To address this, the present invention provides an intelligent maintenance and supervision method for roads and bridges based on digital twins. This method overcomes the problems of existing technologies, such as the inability to dynamically adjust analysis methods according to the actual condition of the bridge, insufficient adaptability to the refined and dynamic maintenance and supervision of complex bridge structures, difficulty in accurately capturing the early characteristics of local bridge defects, and consequently poor accuracy in maintenance decisions.

[0005] To achieve the above objectives, the present invention provides an intelligent maintenance and monitoring method for roads and bridges based on digital twins, comprising: Construct a digital twin model of the bridge to be analyzed; Collect monitoring data from each monitoring point set up on the bridge to be analyzed, and synchronize the monitoring data to the digital twin model; The determination of whether to change the partitioning method from grid partitioning to associated partitioning is based on the average difference coefficient of structural mutation and the differential settlement characterization coefficient of structural foundation; The regulatory status is determined based on the damage index and monitoring anomaly of key areas, and maintenance or maintenance analysis is performed on the bridge to be analyzed based on the regulatory status; wherein, the key areas are determined based on anomaly characterization values, and the damage index is determined based on the global impact coefficient; In the maintenance analysis, the judgment criteria are adjusted from the gradient change value to a combination of stress response disturbance degree and gradient change value based on the proximity influence coefficient, and the determination of whether to carry out maintenance is based on the determined judgment criteria. If no maintenance is required, the adjustment of the regulatory cycle duration is determined based on the fitting deviation value.

[0006] Furthermore, for bridges to be analyzed where the average difference coefficient of structural mutation is greater than or equal to the preset average difference coefficient of structural mutation or the differential settlement characterization coefficient of structural foundation is greater than or equal to the preset differential settlement characterization coefficient of structural foundation, the division method is changed from grid division to correlation division.

[0007] Furthermore, in the association partitioning, the association region is determined based on the diffusion association coefficient and the mechanical response coupling degree, and each association region is recorded as the partitioned region.

[0008] Furthermore, in the grid generation process, the grid size is determined based on the monitored anomaly degree; The grid size is negatively correlated with the degree of anomaly detected.

[0009] Furthermore, the key region is a region defined by an anomaly characterization value that is greater than or equal to a preset anomaly characterization value.

[0010] Furthermore, maintenance is carried out on bridges to be analyzed whose regulatory status is that the damage index is greater than or equal to the preset damage index or the monitoring anomaly degree is greater than or equal to the preset monitoring anomaly degree.

[0011] Furthermore, maintenance analysis is conducted on bridges to be analyzed whose regulatory status is that the damage index is less than the preset damage index and the monitoring anomaly degree is less than the preset monitoring anomaly degree.

[0012] Furthermore, for bridges to be analyzed whose proximity influence coefficient is greater than or equal to the preset proximity influence coefficient, the judgment criterion is adjusted from the gradient change value to a combination of stress response disturbance degree and gradient change value.

[0013] Furthermore, the methods for confirming the damage index include: If the global impact coefficient is greater than the preset global impact coefficient, the damage index is determined based on the global impact coefficient. If the global impact coefficient is less than or equal to the preset global impact coefficient, the damage index is determined based on the applied test anomaly degree.

[0014] Furthermore, the decision on whether to reduce the duration of the regulatory cycle is based on the fitting deviation value; Among them, the duration of the supervision period for bridges to be analyzed whose fitting deviation value is greater than the preset fitting deviation value is reduced and adjusted. The decrease in the duration of the regulatory period is positively correlated with the fitting deviation.

[0015] Compared with the prior art, the beneficial effects of the present invention are that, in the technical solution of the present invention, the abnormal mutation degree of the displacement response of the bridge to be analyzed and the uneven settlement state of the lower pier foundation are effectively reflected by the structural mutation mean difference coefficient and the structural foundation differential settlement characterization coefficient. Then, based on the structural mutation mean difference coefficient and the structural foundation differential settlement characterization coefficient, it is determined whether to adjust the division method from grid division to correlation division. This is conducive to the targeted adaptation of the dynamic deformation characteristics and settlement risk level of the bridge structure, thereby improving the characterization accuracy and damage identification sensitivity of the digital twin model for local structural anomalies of the bridge.

[0016] Furthermore, this invention effectively reflects the degree of mechanical performance degradation in key areas, the threat to the overall structural safety of the bridge, and the degree of abnormal deviation of monitoring indicators across the entire area by using damage indices and monitoring anomalies in key areas. This allows for adaptive selection of maintenance or maintenance analysis based on the regulatory status, which is conducive to accurately matching the actual structural health status of the bridge to formulate differentiated management and control strategies. This avoids resource waste caused by over-maintenance and structural safety risks caused by delayed maintenance, thereby improving the scientific, accurate, and economical nature of bridge maintenance supervision.

[0017] Furthermore, in maintenance analysis, this invention effectively reflects the density of association between key areas and surrounding adjacent areas and the level of influence of neighborhood mechanical coupling through the proximity influence coefficient. Then, based on the proximity influence coefficient, it determines whether to adjust the judgment standard from gradient change value to a combination of stress response interference degree and gradient change value. This is beneficial to adapt to the damage judgment needs under different neighborhood association scenarios, thereby improving the accuracy of early neighborhood coupling damage identification and avoiding the risk of missed or misjudged judgments caused by a single judgment standard.

[0018] Furthermore, this invention effectively reflects the degree to which the mechanical response law of key areas deviates from the healthy state and the potential risk of hidden damage by fitting the deviation value. Then, based on the fitting deviation value, it determines whether to adjust the duration of the supervision cycle. This is conducive to dynamically optimizing the supervision frequency to meet the needs of bridge hidden damage prevention and control, thereby realizing early warning and tracking monitoring of hidden damage and avoiding damage omissions or delayed control caused by fixed supervision cycles. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the intelligent road and bridge maintenance and monitoring method based on digital twins according to the present invention; Figure 2 This is a flowchart illustrating how the present invention determines whether to adjust the partitioning method from grid partitioning to associative partitioning based on the average difference coefficient of structural mutation and the differential settlement characterization coefficient of structural foundation; Figure 3 This is a flowchart illustrating how the present invention determines maintenance or maintenance analysis for a bridge to be analyzed based on its regulatory status. Figure 4 This is a flowchart illustrating the process of determining whether to adjust the duration of the regulatory cycle based on the fitting deviation value, as per the present invention. Detailed Implementation

[0020] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0021] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0022] Please see Figures 1 to 4 As shown, this invention provides an intelligent maintenance and monitoring method for roads and bridges based on digital twins, comprising: Construct a digital twin model of the bridge to be analyzed; Collect monitoring data from each monitoring point set up on the bridge to be analyzed, and synchronize the monitoring data to the digital twin model; The determination of whether to change the partitioning method from grid partitioning to associated partitioning is based on the average difference coefficient of structural mutation and the differential settlement characterization coefficient of structural foundation; The regulatory status is determined based on the damage index and monitoring anomaly of key areas, and maintenance or maintenance analysis is performed on the bridge to be analyzed based on the regulatory status; wherein, the key areas are determined based on anomaly characterization values, and the damage index is determined based on the global impact coefficient; In the maintenance analysis, the judgment criteria are adjusted from the gradient change value to a combination of stress response disturbance degree and gradient change value based on the proximity influence coefficient, and the determination of whether to carry out maintenance is based on the determined judgment criteria. If no maintenance is required, the adjustment of the regulatory cycle duration is determined based on the fitting deviation value.

[0023] The bridge to be analyzed is the bridge that needs to be determined whether it needs maintenance. The bridge to be analyzed can be, but is not limited to, beam bridges, arch bridges, and cable-stayed bridges. The monitoring points initially set in this invention are located at the key stress-bearing parts of the bridge. These key stress-bearing parts include, but are not limited to, the mid-span section of the main beam, the section at the support, the section at the cantilever end, the pier top, the middle and lower part of the pier body, the abutment cap, the bridge deck expansion joint, the support node, and the beam splice joint. This is content that is easy for those skilled in the art to understand, and will not be elaborated on in detail. The monitoring data includes environmental load indicators at each monitoring time within a single regulatory cycle and monitoring indicators at each monitoring point at each detection time. The monitoring indicators include, but are not limited to, strain, deflection, and vibration frequency. The environmental load indicators include, but are not limited to, traffic flow, maximum vehicle speed, ambient humidity, and ambient temperature. For a single regulatory cycle, if all the indicators and parameters of the bridge to be analyzed within that regulatory cycle meet the preset judgment conditions, then that regulatory cycle is recorded as a stable regulatory cycle. The judgment thresholds for the indicators and parameters directly reference the indicator limits explicitly stipulated in current national or industry standards. Specifically, relevant clauses regarding bridge structural condition assessment can be found in standards such as the "Technical Standard for Urban Bridge Maintenance" CJJ 99-2017 and the "Specification for Highway Bridge and Culvert Maintenance" JTG H11-2021. The judgment thresholds include, but are not limited to, allowable values ​​of concrete structural stress, bearing settlement limits, and bridge deck smoothness indicators. This invention preferably selects the following preset judgment conditions: allowable concrete structural stress ≤ 20 MPa (corresponding to ordinary C30 concrete), bearing settlement limit ≤ 10 mm / year, and International Roughness Index (IRI) value ≤ 2.0 m / km.

[0024] The present invention sets a monitoring cycle and performs a monitoring status determination at the end of a single monitoring cycle. The duration of a single monitoring cycle is set to 30 days. Within each monitoring cycle, monitoring time is set at 30-minute intervals starting from the start time. When constructing a digital twin model of the bridge to be analyzed, a full-element digital twin model is built based on the bridge's design drawings, BIM 3D model, and laser scanning point cloud data. This model includes a geometric twin layer, a physical twin layer, and a behavioral twin layer. The geometric twin layer accurately maps the dimensions and spatial positions of key components such as the main beam, piers, bearings, and expansion joints. The physical twin layer embeds core physical parameters of the bridge's main structural materials, such as the elastic modulus, tensile strength, and fatigue life curves. The behavioral twin layer integrates algorithm modules such as the bridge's structural dynamics equations and damage evolution prediction models, enabling the simulation of the bridge's mechanical response and damage development under load and environmental conditions. The system utilizes a monitoring network composed of strain sensors, displacement sensors, acceleration sensors, and temperature and humidity sensors to collect monitoring data in real time at each monitoring moment. The collected raw monitoring data undergoes noise reduction filtering, outlier removal, and standardization preprocessing in sequence. The preprocessed effective data is then transmitted to the digital twin model, driving the model to complete the dynamic update of the virtual-real mapping. This enables real-time linkage between the changes in the physical bridge's state and the mechanical response and deformation visualization of the virtual model, providing accurate digital support for subsequent bridge health status assessment, damage early warning, and maintenance decision optimization. This is a commonly used technical method in the field of science and will not be elaborated upon in detail.

[0025] Specifically, for bridges to be analyzed where the average difference coefficient of structural mutation is greater than or equal to the preset average difference coefficient of structural mutation or the differential settlement characterization coefficient of structural foundation is greater than or equal to the preset differential settlement characterization coefficient of structural foundation, the division method will be changed from grid division to correlation division.

[0026] Specifically, the structural mutation mean deviation coefficient is the standard deviation of the displacement mutation degree corresponding to each monitoring point. For a single monitoring point, the maximum displacement detected by the displacement sensor set at that monitoring point in the current monitoring cycle is denoted as x (in mm), and the maximum displacement detected in the monitoring cycle prior to and adjacent to the current monitoring cycle is denoted as x0. The displacement mutation degree corresponding to this monitoring point is calculated as (x - x0) / (x0 + ε), where ε is 10. -8 ; The structural foundation differential settlement characterization coefficient = S / S0, where S is the maximum settlement difference between adjacent piers measured by a hydrostatic level during the current regulatory period, in mm, and S0 is the allowable settlement difference, the value of which needs to be determined in combination with the bridge structure type, span size and design specification requirements. In this embodiment, the value is taken as 2 mm. Users can determine the values ​​of the preset structural mutation mean difference coefficient and the preset structural foundation differential settlement characterization coefficient according to the actual application scenario. It can be understood that the preset structural mutation mean difference coefficient and the preset structural foundation differential settlement characterization coefficient effectively reflect the degree of abnormal mutation in the displacement response of the bridge to be analyzed and the uneven settlement state of the lower pier foundation. The smaller the preset structural mutation mean difference coefficient and the preset structural foundation differential settlement characterization coefficient, the easier it is to trigger the correlation division, and the more sensitive it is to the identification of small abnormal deformation and settlement risk of the bridge structure. One preset value of the preset structural mutation mean difference coefficient and the preset structural foundation differential settlement characterization coefficient is provided: the preset structural mutation mean difference coefficient is 0.12 and the preset structural foundation differential settlement characterization coefficient is 1.0.

[0027] Specifically, in the association partitioning, the association region is determined based on the diffusion association coefficient and the mechanical response coupling degree, and each association region is recorded as the partitioned region.

[0028] The three-dimensional structural space of the bridge to be analyzed is uniformly divided into several identical cubes. The greater the accuracy of the user's structural representation and damage identification of the digital twin model, the smaller the volume of the cube. It can be understood that the smaller the volume of the cube, the more accurate the representation of the local structural details of the bridge. A value of 0.125m³ for the volume of the cube is provided (i.e., the side length of the cube is 0.5m).

[0029] Diffusion correlation coefficient = mechanical transmission coefficient / preset mechanical transmission coefficient × first weight coefficient + spatial correlation coefficient / preset spatial correlation coefficient × second weight coefficient, where both the first and second weight coefficients are 0.5; For any two cubes, in the digital twin model, a unit load (1kN vertical force) is applied to cube i, and the resulting stress response value F of cube j and the stress response value F0 of cube i itself are extracted; the mechanical transfer coefficient = F / F0; For any two cubes, the spatial correlation coefficient = 1 / (1+α×L), where α is the attenuation coefficient, and in this embodiment, α is taken as 0.1. , used to control the degree of attenuation of spatial distance on correlation, where L is the straight-line distance between the center coordinates of the two cubes, in meters; Extract the stress at the center of the cube in the current regulatory cycle, and record the stress time history data of the two cubes i and j. , The formula for calculating the mechanical response coupling degree C is:

[0030] Where n is the number of monitoring moments in the current regulatory cycle, for and covariance, for standard deviation for Standard deviation; When determining the associated region based on the diffusion correlation coefficient and mechanical response coupling degree, any two cubes in a single associated region satisfy the association condition. The association condition is that the diffusion correlation coefficient between the two cubes is greater than or equal to the preset diffusion correlation coefficient and the mechanical response coupling degree is greater than or equal to the preset association coupling degree. All cubes in a single associated region are a whole, and adjacent cubes do not satisfy the association condition with any cube in the associated region. Adjacent cubes are cubes located outside the associated region and share a face, edge, or point with any cube in the associated region in space.

[0031] The user can determine the values ​​of the preset mechanical transfer coefficient and the preset spatial correlation coefficient according to the refined monitoring requirements of the bridge structure. The mechanical transfer coefficient and the spatial correlation coefficient effectively reflect the mechanical transfer correlation and spatial position correlation between any two cubes. The greater the user's requirement for the accuracy of the correlation area division, the smaller the values ​​of the preset mechanical transfer coefficient and the preset spatial correlation coefficient. In this embodiment, the preferred values ​​are: preset mechanical transfer coefficient of 0.2 and preset spatial correlation coefficient of 0.3.

[0032] The preset values ​​of diffusion correlation coefficient and preset mechanical response coupling degree can be determined by the user based on the accuracy requirements for damage identification of the bridge structure and the need for refined division of the associated region. It can be understood that the diffusion correlation coefficient and mechanical response coupling degree reflect the comprehensive correlation strength and mechanical response synergy between any two cubic units. The higher the user's requirements for the sensitivity of local bridge damage identification and the accuracy of defining the boundary of the associated region, the larger the values ​​of the preset diffusion correlation coefficient and preset mechanical response coupling degree will be. This will help to filter out unit combinations with more significant correlation characteristics and improve the accuracy of subsequent analysis. In this embodiment, the preferred value of the preset diffusion correlation coefficient is 0.7, and the preferred value of the preset mechanical response coupling degree is 0.6.

[0033] Specifically, in grid generation, the grid size is determined based on the degree of anomaly detected; The grid size is negatively correlated with the degree of anomaly detected.

[0034] In the mesh generation, the three-dimensional structural space of the bridge to be analyzed is uniformly divided into several identical partition regions. Each partition region is a cube. The mesh size is the side length of a single partition region. The mesh size = preset monitoring anomaly degree / monitoring anomaly degree × size threshold. The size threshold is 1m. The monitoring anomaly degree is the maximum value among the anomalies of each monitoring indicator. The anomaly degree of a single monitoring indicator = |the maximum value of the value of the monitoring indicator monitored in the current regulatory period - the preset value of the monitoring indicator| / the preset value of the monitoring indicator. The user can determine the preset index value corresponding to a single monitoring indicator based on the bridge structure safety management and control requirements. The greater the user's requirement for bridge safety management and control accuracy, the smaller the preset index value corresponding to a single monitoring indicator. A method for determining the preset index value corresponding to a monitoring indicator is provided, which records the average value of the monitoring indicator corresponding to each monitoring time in each stable monitoring cycle as the preset index value. In this embodiment, the preset value for monitoring anomaly is 0.08, but the above value is not limited to this. Users can increase the value based on the service life of the bridge to reduce the false alarm rate caused by the structural performance degradation of old bridges and large fluctuations in monitoring data.

[0035] Specifically, the key region is a region defined by an anomaly characterization value that is greater than or equal to a preset anomaly characterization value.

[0036] Specifically, for a single defined region, the maximum stress at the center of the circumscribed sphere of that region at each monitoring moment in the current monitoring cycle is denoted as a1, and the average stress detected at the centers of the circumscribed spheres of adjacent regions at that monitoring moment is denoted as a2. The anomaly characteristic value of that region is calculated as (a1-a2) / (a1+10). -8 ); The stress corresponding to the center of the circumscribed sphere of a single region at a single monitoring moment is calculated by calling the finite element statics solver of the behavioral twin layer of the digital twin model. Based on the data input at this monitoring moment, the stress distribution field of the region is simulated and calculated. Based on the spatial coordinates of the circumscribed sphere's center, the post-processing interpolation function of the solver is used to accurately locate the center of the sphere in the global stress distribution field. The stress corresponding to this location is extracted as the stress corresponding to the center of the circumscribed sphere of the region at that monitoring moment. This is a common technique used by those skilled in the art, and will not be elaborated on in detail.

[0037] The user can determine the preset anomaly characterization value according to the actual application scenario. The anomaly characterization value reflects the stress concentration degree and structural anomaly degree of the divided area relative to the adjacent area. The higher the user's sensitivity requirement for identifying local stress anomalies in the bridge, the smaller the preset anomaly characterization value will be, so as to achieve accurate capture of early and minor stress anomalies. The present invention provides a preset anomaly characterization value, and the preferred value is 0.1.

[0038] Specifically, maintenance is carried out on bridges to be analyzed whose regulatory status is that the damage index is greater than or equal to the preset damage index or the monitoring anomaly is greater than or equal to the preset monitoring anomaly.

[0039] The regulatory status includes a first regulatory status and a second regulatory status. The first regulatory status is when the damage index is greater than or equal to the preset damage index or the monitoring anomaly is greater than or equal to the preset monitoring anomaly. The second regulatory status is when the damage index is less than the preset damage index and the monitoring anomaly is less than the preset monitoring anomaly.

[0040] The damage index of the critical area can effectively reflect the degree of mechanical performance degradation of the critical area and the degree of threat to the overall structural safety of the bridge. The higher the user's requirements for the safety of the bridge structure, the smaller the value of the preset damage index. In this embodiment, the preset damage index is 1.1.

[0041] Specifically, maintenance analysis is conducted on bridges whose monitoring status shows a damage index lower than the preset damage index and a monitoring anomaly level lower than the preset monitoring anomaly level.

[0042] Specifically, for bridges to be analyzed whose proximity influence coefficient is greater than or equal to the preset proximity influence coefficient, the judgment criterion will be adjusted from the gradient change value to a combination of stress response disturbance degree and gradient change value.

[0043] Specifically, for bridges to be analyzed whose proximity influence coefficient is less than the preset proximity influence coefficient, the criterion is the gradient change value; The proximity influence coefficient is the average of the proximity reference values ​​corresponding to each key area. The proximity reference value corresponding to a single key area = the number of key areas in the division areas adjacent to the key area / the number of division areas adjacent to the key area. The division areas adjacent to a single key area refer to the division areas that share a surface, edge, or point with the key area in space. Gradient change value = Damage characterization value of the current regulatory cycle - Damage characterization value of the regulatory cycle adjacent to and preceding the current regulatory cycle; The damage characterization value corresponding to a single regulatory cycle = damage index / preset damage index × third weighting coefficient + monitoring anomaly degree / preset monitoring anomaly degree × fourth weighting coefficient, where both the third weighting coefficient and the fourth weighting coefficient are 0.5; The preset gradient change value can be determined by the user based on the actual application scenario. The gradient change value can effectively reflect the sensitivity to fluctuations in the slope of the load-stress curve, thereby accurately capturing subtle deterioration trends in the mechanical properties of key areas. The smaller the gradient change value, the more sensitive the perception of slope fluctuations, and the earlier abnormal mechanical response can be detected in the early stages of damage. The smaller the preset gradient change value, the greater the user's need for early damage identification in key areas and precise control of the damage development process. A method for setting the preset gradient change value is provided, which is 0.4 times the damage characterization value corresponding to the current monitoring cycle.

[0044] The method for confirming the stress response disturbance degree is as follows: the monitoring data of each monitoring point at the end of the current regulatory cycle is synchronized to the digital twin model. For a single key area, the key area is recorded as the target key area, and each key area adjacent to the target key area is recorded as the reference key area. A unit load (1kN vertical force) is applied to the center of the smallest circumscribed sphere of the target key area and the reference key area respectively. The stress generated at the center of the target key area at this time is recorded as F1. The stress generated at the center of the target key area by applying a unit load (1kN vertical force) only at the center of the smallest circumscribed sphere of the target key area is recorded as F2. The ratio of F1 to F2 is recorded as the stress response disturbance degree. Understandably, when applying a single-point load to any defined region within the constructed digital twin model, the model's built-in mechanical simulation module precisely binds the preset load parameters (including load magnitude and direction) to the three-dimensional spatial coordinates of the target loading point. This serves as a constraint condition to drive the digital twin model to perform numerical calculations of the stress and other mechanical responses of the corresponding structural region. Simultaneously, the structural statics / dynamics solver integrated in the behavioral twin layer of the digital twin model is invoked. Monitoring data collected at the end of the current monitoring cycle is used as input parameters and substituted into the solver to conduct a global stress field simulation analysis of the defined region. Then, based on the pre-determined spatial coordinates of the center of the circumscribed geometric sphere of the defined region, the stress value corresponding to the center position is extracted from the global stress distribution results obtained from the simulation. This is a conventional technique known in the field and will not be elaborated upon further.

[0045] Determining whether maintenance is required based on established criteria includes: when the criterion is the gradient change value, maintenance is required for bridges under analysis whose gradient change value is greater than or equal to the preset gradient change value, and maintenance is not required for bridges under analysis whose gradient change value is less than the preset gradient change value; when the criterion is a combination of stress response disturbance degree and gradient change value, maintenance is required for bridges under analysis whose stress response disturbance degree is greater than or equal to the preset stress response disturbance degree or whose gradient change value is greater than or equal to the preset gradient change value, and maintenance is not required for bridges under analysis whose stress response disturbance degree is less than the preset stress response disturbance degree and whose gradient change value is less than the preset gradient change value.

[0046] The value of the preset stress response interference degree can be flexibly determined by the user according to the maintenance accuracy requirements. The stress response interference degree can effectively characterize the disturbance intensity of the load transfer effect of the neighborhood mechanically sensitive key area on the stress field distribution of the target key area. The higher the user's demand for accurate identification, evolution process tracking and risk early warning of the bridge structure neighborhood coupled early damage, the smaller the value of the preset neighborhood key area stress coupling interference coefficient should be. In this embodiment, the preset stress response interference degree is 1.4.

[0047] Specifically, the methods for confirming the damage index include: If the global impact coefficient is greater than the preset global impact coefficient, the damage index is determined based on the global impact coefficient. If the global impact coefficient is less than or equal to the preset global impact coefficient, the damage index is determined based on the applied test anomaly degree.

[0048] Overall influence coefficient = Sum of volumes of all key regions / Sum of volumes of all subdivided regions; When determining the damage index based on the global impact coefficient, the damage index = global impact coefficient / preset global impact coefficient; When determining the damage index based on the applied test anomaly, the monitoring data of each monitoring point at the end of the current regulatory cycle are synchronized to the digital twin model. For a single key area, the minimum circumscribed sphere of that key area is constructed, and the center of the circumscribed sphere is recorded as the load application point. Graded vertical loads of 0kN, 0.1kN, 0.2kN, ..., 1.5kN are applied at the load application point. The stress at the center of the key area under each load level is collected. A two-dimensional load-stress curve is plotted with the load value on the x-axis and the stress value on the y-axis. The line connecting adjacent load-stress data points on the two-dimensional coordinate curve is recorded as a reference segment. The ratio of the maximum slope of each reference segment to the preset slope is recorded as the applied test anomaly, and the maximum applied test anomaly of each key area is recorded as the damage index. In this embodiment, the preset global influence coefficient is set to 0.4. It can be understood that the global influence coefficient can effectively reflect the volume weight ratio of the key area in the overall structure of the road and bridge. Its value is directly related to the degree of influence of the key area on the overall safety performance of the bridge. The larger the preset global influence coefficient is, the smaller the proportion of scenarios will trigger the "determine the damage index based on the global influence coefficient" judgment path.

[0049] In this embodiment, the preset slope is 120 MPa / kN. It can be understood that the preset slope corresponds to the ideal response gradient of load and stress under normal mechanical performance conditions in the critical area. Its value is directly related to the sensitivity of judging the degree of damage in the critical area. The smaller the value of the preset slope, the more lenient the threshold for judging damage. Even if the maximum value of the slope of the load-stress curve is small (i.e. the degree of mechanical performance degradation is relatively mild), it is easy for the applied test anomaly to reach or exceed the judgment threshold, and thus it is easier to judge the corresponding area as a damaged area that needs attention. This will make maintenance decisions more inclined to carry out maintenance in advance, which can effectively reduce the structural safety risk caused by further deterioration of minor damage.

[0050] Specifically, the decision on whether to reduce the duration of the regulatory cycle is based on the fitting deviation value; Among them, the duration of the supervision period for bridges to be analyzed whose fitting deviation value is greater than the preset fitting deviation value is reduced and adjusted. The decrease in the duration of the regulatory period is positively correlated with the fitting deviation.

[0051] Specifically, the duration of the monitoring period for bridges to be analyzed whose fitting deviation value is less than or equal to the preset fitting deviation value does not need to be adjusted.

[0052] The fitting deviation is the maximum value among the sub-fitting deviations corresponding to each key region. The sub-fitting deviation for a single key region is equal to the average goodness of fit for that key region in each stable regulatory cycle minus the goodness of fit for the current regulatory cycle. Based on the data points of the load-stress two-dimensional coordinate curve, a linear fitting is performed using the least squares method to obtain the fitted line. The goodness of fit is calculated as 1 - the sum of squares of the differences between the actual stress value and the corresponding fitted stress value at each load level / the sum of squares of the differences between the actual stress value and the average of all actual stress values ​​at each load level. Using the least squares method for linear fitting is a common technique used by those skilled in the art, and will not be elaborated upon further. The reduction in regulatory cycle duration = (fit deviation value - preset fit deviation value) / preset fit deviation value × duration threshold, where the duration threshold is 8 days.

[0053] Understandably, the larger the fitting deviation value, the more obvious the deviation of the mechanical response law in the key area from the healthy state, and there may be hidden damage (such as internal cracks, material fatigue, etc.). It is necessary to shorten the supervision cycle and further verify the damage. When the user has greater requirements for the sensitivity and timeliness of bridge structure maintenance supervision, the preset fitting deviation value should be smaller, so as to achieve accurate capture of small deviations in the mechanical response law. In this embodiment, the preset fitting deviation value is 0.07.

[0054] It should be noted that the minimum threshold for the duration of the regulatory period is 10 days. If the calculated reduction results in the regulatory period being less than the minimum threshold, the regulatory period will be fixed at the minimum threshold.

[0055] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for intelligent maintenance and supervision of roads and bridges based on digital twins, characterized in that, include: Construct a digital twin model of the bridge to be analyzed; Collect monitoring data from each monitoring point set up on the bridge to be analyzed, and synchronize the monitoring data to the digital twin model; The determination of whether to change the partitioning method from grid partitioning to associated partitioning is based on the average difference coefficient of structural mutation and the differential settlement characterization coefficient of structural foundation; The regulatory status is determined based on the damage index and monitoring anomaly of key areas, and maintenance or maintenance analysis is performed on the bridge to be analyzed based on the regulatory status; wherein, the key areas are determined based on anomaly characterization values, and the damage index is determined based on the global impact coefficient; In the maintenance analysis, the judgment criteria are adjusted from the gradient change value to a combination of stress response disturbance degree and gradient change value based on the proximity influence coefficient, and the determination of whether to carry out maintenance is based on the determined judgment criteria. If no maintenance is required, the adjustment of the regulatory cycle duration is determined based on the fitting deviation value.

2. The intelligent maintenance and supervision method for roads and bridges based on digital twins according to claim 1, characterized in that, For bridges to be analyzed where the average difference coefficient of structural mutation is greater than or equal to the preset average difference coefficient of structural mutation or the differential settlement characterization coefficient of the structural foundation is greater than or equal to the preset differential settlement characterization coefficient of the structural foundation, the division method is changed from grid division to correlation division.

3. The intelligent maintenance and supervision method for roads and bridges based on digital twins according to claim 2, characterized in that, In the association partitioning, the association region is determined based on the diffusion association coefficient and the mechanical response coupling degree, and each association region is recorded as the partitioned region.

4. The intelligent maintenance and supervision method for roads and bridges based on digital twins according to claim 3, characterized in that, In grid generation, the grid size is determined based on the monitored anomaly degree; The grid size is negatively correlated with the degree of anomaly detected.

5. The intelligent maintenance and supervision method for roads and bridges based on digital twins according to claim 4, characterized in that, The key region is the region defined by an anomaly characterization value that is greater than or equal to a preset anomaly characterization value.

6. The intelligent maintenance and supervision method for roads and bridges based on digital twins according to claim 1, characterized in that, Maintenance will be carried out on bridges to be analyzed whose monitoring status is that the damage index is greater than or equal to the preset damage index or the monitoring anomaly is greater than or equal to the preset monitoring anomaly.

7. The intelligent maintenance and supervision method for roads and bridges based on digital twins according to claim 6, characterized in that, Maintenance analysis is conducted on bridges whose monitoring status shows a damage index lower than the preset damage index and a monitoring anomaly level lower than the preset monitoring anomaly level.

8. The intelligent maintenance and supervision method for roads and bridges based on digital twins according to claim 7, characterized in that, For bridges to be analyzed whose proximity influence coefficient is greater than or equal to the preset proximity influence coefficient, the judgment criterion will be adjusted from the gradient change value to a combination of stress response disturbance degree and gradient change value.

9. The intelligent maintenance and supervision method for roads and bridges based on digital twins according to claim 7, characterized in that, The methods for confirming the damage index include: If the global impact coefficient is greater than the preset global impact coefficient, the damage index is determined based on the global impact coefficient. If the global impact coefficient is less than or equal to the preset global impact coefficient, the damage index is determined based on the applied test anomaly degree.

10. The intelligent maintenance and supervision method for roads and bridges based on digital twins according to claim 8, characterized in that, The decision on whether to reduce the duration of the regulatory cycle is based on the fitting deviation value. Among them, the duration of the supervision period for bridges to be analyzed whose fitting deviation value is greater than the preset fitting deviation value is reduced and adjusted. The decrease in the duration of the regulatory period is positively correlated with the fitting deviation.

Images