A continuous beam bridge bearing capacity comprehensive evaluation method
By constructing a three-dimensional damage voxel field and a finite element model with physical constraint correction, the data fusion and safety boundary problems in the load-bearing capacity assessment of existing continuous beam bridges were solved, and accurate assessment and safety control of heavy-load operations were achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA RAILWAY 14TH BUREAU GRP NO 3 ENG CO LTD
- Filing Date
- 2026-01-09
- Publication Date
- 2026-05-22
AI Technical Summary
Existing methods for assessing the load-bearing capacity of continuous beam bridges suffer from several problems, including difficulty in integrating multi-source detection data with mechanical models, unclear safety boundaries for heavy-load operations, and the lack of physical constraints in finite element model corrections leading to multiple solutions. These issues result in large errors in the assessment results and an inability to accurately predict heavy-load conditions.
By constructing a three-dimensional damage voxel field and integrating three-dimensional laser scanning point cloud, ultrasonic detection, and rebound strength data, an initial finite element model is generated. Through equivalent static load tests and physical constraint correction models, heavy-load conditions are identified and a spatiotemporal safety envelope for operation is generated. Operation control instructions are monitored and output in real time.
It enables accurate load-bearing capacity assessment of existing continuous beam bridges, improves the accuracy of the model's description of the structural physical condition, and ensures the safety of heavy-load operations and real-time decision support.
Smart Images
Figure CN121479916B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of bridge load-bearing capacity assessment technology, specifically a comprehensive assessment method for the load-bearing capacity of existing continuous beam bridges. Background Technology
[0002] During their long service life, existing continuous beam bridges inevitably suffer damage such as concrete cracking, strength degradation, and geometric deformation due to environmental erosion and repeated loads. With the advancement of infrastructure construction, it is often necessary to temporarily support special heavy equipment that exceeds the original design standards, such as crawler cranes or beam transport vehicles. The load effect under this special working condition far exceeds the conventional traffic load, and the operation process involves complex changes in mechanical state.
[0003] Existing load-bearing capacity assessment techniques involve engineers using equipment such as 3D laser scanning, ultrasonic testing, and rebound hammers to acquire geometric and material state data of bridges. A finite element model is then established based on the design drawings, and structural damage is simulated by reducing the material's elastic modulus. During the model correction phase, deflection or strain measured in static load tests is typically used as a benchmark, and model parameters are adjusted to reduce calculation errors. For safety analysis under heavy loads, the conventional approach is to select several typical, most unfavorable static conditions, such as the maximum bending moment at mid-span, for verification. If the calculated stress is less than the allowable stress, the solution is deemed feasible.
[0004] However, existing load-bearing capacity assessment technologies suffer from a mapping gap between multi-source detection data and mechanical models. Detection data consists of discrete, heterogeneous point clouds or measurement points, while finite element models are continuous mesh elements. Existing methods often employ homogenization, ignoring the non-uniformity of damage distribution in three-dimensional space. This leads to significant deviations between the initial model stiffness distribution and the actual structure. Mathematical optimization aimed solely at fitting experimental data often falls into ill-posed solutions. Algorithms may use parameter values that violate physical principles to align with data, causing the model to lose its predictive ability for unknown heavy-load conditions. Furthermore, heavy-load equipment operation is a continuously changing process, and traditional static verification cannot cover all critical states. Complex nonlinear calculations are difficult to run in real-time on-site, making it impossible for operators to intuitively grasp the safety margin between their current actions and the structural failure boundary. Therefore, this invention provides a comprehensive load-bearing capacity assessment method for existing continuous beam bridges to address the shortcomings of existing technologies. Summary of the Invention
[0005] To address the shortcomings of existing technologies, this invention provides a comprehensive evaluation method for the bearing capacity of existing continuous beam bridges. This method solves the technical problems in the existing evaluation of the bearing capacity of existing continuous beam bridges, such as the difficulty in effectively integrating apparent test data with mechanical models, unclear safety boundaries for heavy-load operations, and the lack of physical constraints in finite element model corrections leading to multiple solutions.
[0006] To achieve the above objectives, the present invention provides the following technical solution:
[0007] The first aspect of this invention provides a comprehensive evaluation method for the bearing capacity of existing continuous beam bridges. This method integrates multi-source non-destructive testing data and mechanical response data, and achieves structural digital twins through model correction under physical constraints. Specifically, it includes the following steps:
[0008] Voxel-based data fusion and initial model construction are performed. Data on geometric defects, concrete strength, and internal flaws of existing bridges are collected. This heterogeneous data is fused and mapped onto a discretized three-dimensional voxel mesh to construct a three-dimensional damage voxel field characterizing the structural physical state. Based on the damage quantification indices of each voxel in the three-dimensional damage voxel field, an initial finite element model containing initial damage distribution information is generated. This step, through voxelization mapping technology, transforms discrete detection point data into a continuous spatial damage field, solving the problem of poor applicability of traditional point-to-point mapping in complex geometric structures.
[0009] Heavy-load simulation and equivalent test design were conducted. The entire process of the target heavy-load operation was simulated in the initial finite element model to identify the stress control sections and theoretical limit states of the structure. Based on the principle of equivalent internal forces, a loading scheme was calculated to simulate the mechanical effects of heavy-load equipment using standard test vehicles, and an equivalent static load test scheme was developed. This step, through combined loading with standard vehicles, generates internal force effects at the control sections that are proportional to the heavy-load conditions, enabling the evaluation of the structure's high-stress behavior without directly applying extreme loads.
[0010] Obtain actual structural response data. Based on the equivalent static load test scheme, conduct on-site graded static load tests on the existing continuous beam bridge, collect data on the deflection, stress, and crack width changes of the structure under elastic working conditions, and perform noise reduction and standardization processing on the original signals to construct the measured response vector.
[0011] Perform physical constraint-based model correction. Calculate the error between the measured response vector and the theoretically calculated response of the initial finite element model under the same loading conditions. Using damage distribution data in the three-dimensional damage voxel field as prior physical information, determine the range of values for the model stiffness parameters. Iteratively adjust the stiffness parameters of the initial finite element model within this range until the error meets the convergence condition, thus completing the model correction.
[0012] Generate a spatiotemporal safety envelope for the operation and implement dynamic control. Based on the modified finite element model, perform nonlinear bearing capacity derivation for various combinations of working conditions for the target heavy-load operation, calculate the set of working conditions that meet the safety factor requirements, and generate the spatiotemporal safety envelope for the operation. During construction, receive real-time status data of on-site operating equipment, calculate the spatial geometric relationship between the current status point and the spatiotemporal safety envelope for the operation in real time, and output corresponding operation control instructions.
[0013] Step S1 further includes:
[0014] The physical space domain surrounding the existing continuous beam bridge is divided into a three-dimensional voxel grid, with each voxel serving as a data container for storing material property state values.
[0015] Three-dimensional laser scanning point cloud data is selected as the geometric defect data, ultrasonic wave velocity data is selected as the internal defect data, and rebound strength data is selected as the concrete strength data. The point cloud data, wave velocity data, and rebound strength data are mapped to the center coordinates of the global voxel using a spatial interpolation algorithm. The comprehensive damage factor of each voxel is calculated to generate the three-dimensional damage voxel field.
[0016] Identify the set of voxels contained within the geometric space of any element in the initial finite element model. Based on the volume average of the comprehensive damage factor of all voxels in the set of voxels, calculate the material stiffness reduction factor. Use the material stiffness reduction factor to reconstruct the stiffness matrix of the ideal elastic element generated based on the original design material parameters to obtain the initial finite element model.
[0017] The overall damage factor for each voxel is calculated based on a weighted summation, consisting of a first part and a second part:
[0018] The first part is the weighting coefficient of the ultrasonic velocity index multiplied by the ultrasonic velocity relative deviation term, wherein the ultrasonic velocity relative deviation term is 1 minus the ratio of the measured ultrasonic velocity at the voxel position to the theoretical standard wave velocity of non-destructive concrete.
[0019] The second part is the weighting coefficient of the rebound strength index multiplied by the relative deviation term of the rebound strength, wherein the relative deviation term of the rebound strength is 1 minus the ratio of the estimated value of the measured rebound strength at the voxel position to the standard value of the concrete strength.
[0020] In step S2, the design of the equivalent static load test scheme based on the principle of equivalent internal forces, which utilizes a standard vehicle to simulate the effects of heavy-duty equipment, specifically includes:
[0021] A heavy load model is applied to the initial finite element model to calculate and identify the control section that generates the maximum internal force or maximum deformation during the heavy load operation, and extract the theoretical maximum response value of the control section under the most unfavorable working condition and the continuous influence line function along the loading path.
[0022] An equivalent loading calculation model is constructed with the number of test vehicles and the coordinates of the vehicle parking positions as optimization variables. The superposition effect value generated by the test vehicle fleet at the control section is calculated using the continuous influence line function. The superposition effect value is between the lower limit and the upper limit of the proportional coefficient set by the theoretical maximum response value as a constraint condition.
[0023] Solving the equivalent loading calculation model yields the number of vehicles and parking positions that satisfy the constraints. The setting range of the lower and upper limits of the proportional coefficient ensures that the test load induces observable deformation in the existing continuous beam bridge and keeps it in an elastic working state.
[0024] In step S3, the step of constructing the measured response vector further includes:
[0025] Establish an on-site physical sensing network to collect micro-strain signals at the bottom of the control section, vertical deflection signals at the mid-span and quarter-point sections, and crack width opening and closing change signals in existing crack concentration areas.
[0026] The acquired raw signals are denoised and the reference value is subtracted to convert the absolute readings into the net response values caused by the test load.
[0027] The net response values of different physical dimensions are weighted, normalized, and standardized and assembled to form the measured response vector, which includes the measured net vertical deflection value, the measured net strain value, and the measured crack width increment value.
[0028] In step S4, the error between the calculated measured response vector and the theoretically calculated response of the initial finite element model specifically includes:
[0029] The stiffness correction coefficients of each element in the initial finite element model are selected as parameters to be corrected, and the parameter vector to be corrected is constructed.
[0030] Calculate the theoretical response vector of the initial finite element model under the experimental load based on the current parameter vector;
[0031] A target functional is constructed to represent the error, wherein the target functional is the weighted sum of squared residuals or the Euclidean distance between the theoretical response vector and the measured response vector.
[0032] In step S4, the iterative adjustment of the stiffness parameters of the initial finite element model specifically includes:
[0033] For each finite element element corresponding to the parameter to be corrected, the average comprehensive damage factor of the voxels covered by the finite element element in the three-dimensional damage voxel field is read.
[0034] The physical constraint boundary is constructed based on the mean value of the comprehensive damage factor, and the corresponding stiffness correction coefficient of the finite element element is constructed. The physical constraint boundary includes a lower limit constraint value and an upper limit constraint value.
[0035] The constrained optimization algorithm is used for iterative solution. In each iteration step, it is checked whether the generated tentative parameter vector satisfies the physical constraint boundary. For parameters that violate the constraints, they are forcibly pulled back to the feasible region defined by the lower constraint value and the upper constraint value.
[0036] In step S5, the step of generating the job spatiotemporal safety envelope further includes:
[0037] Define a multidimensional state vector containing the plane coordinates of the working equipment, the rotation angle of the lifting equipment, and the current lifting weight, and construct the multidimensional parameter space of the working system.
[0038] In the modified finite element model, the multidimensional parameter space is discretized and simulated, and the corresponding structural safety factor is calculated for each discrete working point.
[0039] Construct the spatiotemporal safety envelope of the operation, which is the set of all working points in the multidimensional parameter space where the structural safety factor is greater than or equal to the minimum allowable safety factor.
[0040] In step S5, the real-time comparison of the spatial positional relationship between the current operation status and the operation spatiotemporal safety envelope, and the output of operation control instructions, specifically includes:
[0041] Receive positioning and status sensor data from on-site operating equipment, assemble them into a real-time status vector, and calculate the boundary distance between the real-time status vector and the operating spatiotemporal safety envelope;
[0042] When the real-time status vector is located inside the work spatiotemporal safety envelope and the distance from the boundary is greater than the warning threshold, it is determined to be a safe zone and a work permission signal is output; when the real-time status vector is located inside the work spatiotemporal safety envelope but the distance from the boundary is less than the warning threshold, it is determined to be a warning zone and an audible and visual alarm signal is output; when the real-time status vector is located outside the work spatiotemporal safety envelope, it is determined to be a danger zone and an emergency stop signal is output.
[0043] In a preferred embodiment, the model correction process introduces physical constraint boundaries based on detection data to avoid distortion of the physical meaning of parameters caused by pure mathematical optimization. Specifically, for each finite element in the initial finite element model corresponding to the parameter to be corrected, the average comprehensive damage factor of the voxels it covers in the three-dimensional damage voxel field is read. ,based on Build correction parameters Physical constraint boundaries:
[0044] ;
[0045] ;
[0046] In the formula, and The first The lower and upper limits of the stiffness correction coefficient for each element; This is the average damage factor of the voxel region corresponding to this unit (derived from laser scanning and ultrasound detection); The radius of the confidence interval for the detection data.
[0047] A second aspect of the present invention provides a comprehensive evaluation system for the bearing capacity of existing continuous beam bridges, comprising:
[0048] The multi-source damage data mapping and modeling module is used to collect data on geometric defects, concrete strength and internal defects of existing bridges, fuse and map them into a three-dimensional voxel mesh, construct a three-dimensional damage voxel field, and generate an initial finite element model containing initial damage information based on the three-dimensional damage voxel field.
[0049] The heavy-load simulation and equivalent test design module is used to simulate heavy-load operating conditions in the initial finite element model, identify control sections and theoretical limit states, and design equivalent static load test schemes that use standard vehicles to simulate the effects of heavy-load equipment based on the principle of equivalent internal forces.
[0050] The on-site response data acquisition and processing module is used to perform on-site graded static load tests on existing continuous beam bridges according to the equivalent static load test scheme, collect the deflection, stress and crack width changes of existing continuous beam bridges under elastic working state as actual response data, and construct the measured response vector.
[0051] The physical constraint model correction module is used to calculate the error between the measured response vector and the theoretically calculated response of the initial finite element model, and iteratively adjust the stiffness parameters of the initial finite element model within the parameter range determined by the three-dimensional damage voxel field until the error meets the convergence condition, thus completing the correction of the initial finite element model.
[0052] The spatiotemporal envelope generation and dynamic control module is used to perform nonlinear bearing capacity deduction of the target heavy-load operation based on the modified finite element model, generate the spatiotemporal safety envelope of the operation, and compare the spatial position relationship between the current operation status and the spatiotemporal safety envelope of the operation in real time during construction, and output operation control instructions.
[0053] This invention provides a comprehensive evaluation method for the bearing capacity of existing continuous beam bridges. It has the following beneficial effects:
[0054] 1. This invention constructs a three-dimensional damage voxel field and uses voxels as normalized data carriers to map discrete apparent and internal damage data, such as three-dimensional laser scanning point clouds, ultrasonic speeds, and rebound strength, into a continuously distributed spatial field. This can accurately represent the spatial variability of the stiffness of existing bridge materials, avoid the errors caused by simple homogenization of the damaged area in traditional methods, and improve the accuracy of the initial finite element model in describing the physical state of the structure.
[0055] 2. In the model correction stage, this invention introduces physical constraint boundaries based on measured damage data. By using the stiffness reduction range calculated from the damage voxel field as a hard constraint condition for parameter inversion, the optimization process is forced to be carried out within the physical feasible region. This effectively prevents the distortion of the physical meaning of parameters caused by simply pursuing mathematical convergence, and ensures that the corrected model fits the static load test data in terms of response values and conforms to the actual damage state of the structure in terms of stiffness distribution.
[0056] 3. This invention establishes a spatiotemporal safety envelope for heavy-load operations. By performing multi-dimensional parameter space traversal simulation of the planar coordinates, rotation angle, and load weight of the lifting equipment, a visualized safe operating domain is generated. The real-time monitoring data on site is then compared spatially with this envelope, transforming complex real-time nonlinear mechanical calculations into efficient geometric position determination. This reduces the lag in online calculations and provides an intuitive and real-time decision-making basis for the construction safety of existing continuous beam bridges under extreme heavy-load conditions. Attached Figure Description
[0057] Figure 1 This is a system architecture diagram of the present invention;
[0058] Figure 2 This is a flowchart of the method steps of the present invention;
[0059] Figure 3 This is a flowchart of the three-dimensional damage voxel field construction process of the present invention;
[0060] Figure 4 This is a flowchart of the heavy-load simulation and equivalent test design of the present invention;
[0061] Figure 5 This is a flowchart of the on-site response data acquisition and construction process of the present invention;
[0062] Figure 6 This is a flowchart of the physical constraint type model correction process of the present invention;
[0063] Figure 7 This is a schematic diagram of the spatiotemporal envelope generation and dynamic control of the present invention.
[0064] Among them, 10 is the multi-source damage data mapping and modeling module; 20 is the heavy load simulation and equivalent test design module; 30 is the field response data acquisition and processing module; 40 is the physical constraint model correction module; and 50 is the spatiotemporal envelope generation and dynamic control module. Detailed Implementation
[0065] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0066] See attached document Figure 1 , Figure 1 This is a system architecture diagram according to an embodiment of the present invention. The present invention provides a comprehensive evaluation system for the bearing capacity of existing continuous beam bridges, including a multi-source damage data mapping and modeling module 10, a heavy load simulation and equivalent test design module 20, a field response data acquisition and processing module 30, a physical constraint model correction module 40, and a spatiotemporal envelope generation and dynamic control module 50.
[0067] The multi-source damage data mapping and modeling module 10 is used to process multi-source non-destructive testing data of the bridge. This module receives three-dimensional laser scanning point cloud data, ultrasonic wave velocity data, and rebound strength data of the bridge. The multi-source damage data mapping and modeling module 10 discretizes the physical space surrounding the bridge structure into a three-dimensional voxel mesh and performs spatial interpolation to map the aforementioned detection data to the corresponding voxels to generate a three-dimensional damage voxel field. The multi-source damage data mapping and modeling module 10 further calculates the reduction coefficients of the finite element material properties based on the three-dimensional damage voxel field, establishing an initial finite element model containing initial damage distribution information.
[0068] The heavy-load simulation and equivalent test design module 20 is used to perform virtual working condition simulations. It loads the target heavy-load equipment's load model into the initial finite element model and identifies the control sections and theoretical stress states of the structure through calculation. Based on the principle of equivalent internal forces, the module calculates the loading scheme of the standard test vehicle, determines the number of test vehicles and their load distribution coordinates, and ensures that the internal force effect generated by the test load at the control section reaches the preset proportion of the target heavy-load working condition.
[0069] The on-site response data acquisition and processing module 30 is used to acquire data on the actual mechanical behavior of the structure. The module is connected to displacement sensors, strain sensors, and crack observation instruments installed at key locations on the bridge to acquire displacement signals, strain signals, and crack width signals during the graded static load test. The module performs noise reduction and feature extraction processing on the acquired raw signals to construct a measured response vector containing measured deflection, measured strain, and crack width increments.
[0070] The physical constraint model correction module 40 is used to calculate the error between the measured response vector and the theoretically calculated response of the finite element model. The physical constraint model correction module 40 reads the three-dimensional damage voxel field data generated by the multi-source damage data mapping and modeling module 10 as the boundary conditions for parameter correction. Within the range defined by the above boundary conditions, the stiffness parameters of the finite element model are iteratively adjusted until the error meets the convergence condition, and the corrected finite element model is output.
[0071] The spatiotemporal envelope generation and dynamic control module 50 is connected to the physical constraint model correction module 40. The spatiotemporal envelope generation and dynamic control module 50 uses the corrected finite element model to perform traversal simulations of various working condition parameter combinations for the target heavy-load operation, calculates and extracts the operation state points that meet the structural safety indicators, and generates the operation spatiotemporal safety envelope. During construction operations, the spatiotemporal envelope generation and dynamic control module 50 receives real-time status data from the on-site operating equipment, calculates the spatial positional relationship between the current state point and the operation spatiotemporal safety envelope, and outputs operation control commands.
[0072] See attached document Figure 2 , Figure 2 This is a flowchart of a method according to an embodiment of the present invention. The present invention provides a method for comprehensive evaluation of the bearing capacity of existing continuous beam bridges, comprising the following steps:
[0073] S1, through the multi-source damage data mapping and modeling module 10, the collected existing bridge geometric defect data, concrete strength data and internal defect data are fused and mapped into a three-dimensional voxel mesh to construct a three-dimensional damage voxel field, and an initial finite element model containing initial damage information is generated based on the three-dimensional damage voxel field.
[0074] S2, through the heavy load simulation and equivalent test design module 20, simulates the target heavy load operation condition in the initial finite element model, identifies the control section and theoretical limit state, and designs an equivalent static load test scheme that uses a standard vehicle to simulate the effect of heavy load equipment based on the principle of equivalent internal force.
[0075] S3, through the on-site response data acquisition and processing module 30, records the graded static load test data of the actual bridge according to the equivalent static load test scheme, obtains the actual response data of the structure within the elastic range, the actual response data includes the changes in deflection, stress and crack width, and constructs the measured response vector;
[0076] S4, through the physical constraint model correction module 40, calculate the error between the measured response vector and the theoretically calculated response of the initial finite element model, and iteratively adjust the stiffness parameters of the initial finite element model within the parameter value range determined by the three-dimensional damage voxel field until the error meets the convergence condition, thus completing the model correction;
[0077] S5, through the spatiotemporal envelope generation and dynamic control module 50, performs nonlinear bearing capacity deduction on the target heavy-load operation based on the modified finite element model, generates the spatiotemporal safety envelope of the operation, and compares the spatial positional relationship between the current operation status and the spatiotemporal safety envelope of the operation in real time during construction, and outputs operation control instructions.
[0078] The above steps will be explained in detail below with reference to specific embodiments.
[0079] See attached document Figure 3 In step S1, the multi-source damage data mapping and modeling module 10 performs an initial model construction process based on a three-dimensional damage voxel field. This initial model construction process specifically includes three steps: data space discretization, heterogeneous damage data normalization mapping, and finite element stiffness matrix reconstruction.
[0080] S1.1 Data Spatial Discretization and Voxel Mesh Definition:
[0081] The multi-source damage data mapping and modeling module 10 establishes a physical spatial domain surrounding the bridge structure based on the design drawings and on-site measured contour data of the continuous beam bridge to be evaluated. Physical space domain The data is divided into a three-dimensional voxel mesh composed of multiple tiny hexahedral units. Any voxel in the three-dimensional voxel mesh is denoted as . Its spatial center coordinates in the global coordinate system are defined as follows: The resolution of each voxel (i.e., voxel side length) is set based on the highest spatial accuracy of the detected data, for example, 50mm to 100mm. It is configured as an independent data container to store the state values of material properties at that spatial location.
[0082] S1.2, Normalization mapping and field construction of heterogeneous damage data:
[0083] The multi-source damage data mapping and modeling module 10 receives discrete point data from field testing equipment. Data sources include: point cloud coordinate data describing surface spalling and cross-sectional defects obtained by a 3D laser scanner; ultrasonic pulse velocity data obtained by an ultrasonic testing instrument at grid points in the test area; and concrete surface hardness rebound value data obtained by a rebound hammer.
[0084] For the aforementioned multi-source heterogeneous data, the module performs spatial interpolation and normalization. For spatially discrete detection data (such as ultrasonic velocity), ordinary Kriging interpolation or other geostatistical interpolation methods are used to extrapolate the discrete measurement point values to the global voxel range. At the central coordinates, a continuous data field is formed.
[0085] For any voxel Define its comprehensive damage factor Comprehensive damage factors It is a scalar, and its value range is... Where 0 represents an intact state and 1 represents a completely failed state. Comprehensive damage factor. The calculation formula is as follows:
[0086] ;
[0087] In the formula, This represents the measured ultrasonic velocity at the voxel location obtained through interpolation. This represents the theoretical standard wave velocity of non-destructive concrete, and its value is determined based on the wave velocity reference value of the concrete of the design grade. This represents the estimated value of the measured rebound strength at the voxel location obtained through interpolation. This indicates the standard value of concrete strength specified in the design drawings; and These are the weighting coefficients for the ultrasonic velocity index and the rebound strength index, respectively, satisfying... The weighting coefficients are set based on the on-site testing environment and the confidence level of each testing method.
[0088] By traversing the computational space domain All voxels within the structure are used to generate a three-dimensional damage voxel field. This three-dimensional damage voxel field, in the form of a digital matrix, quantitatively characterizes the non-uniform distribution of material property degradation within and on the surface of the bridge structure in three-dimensional space.
[0089] S1.3 Initial finite element model generation based on voxel mapping:
[0090] The multi-source damage data mapping and modeling module 10 establishes a geometric finite element model of the bridge structure. This geometric finite element model contains multiple finite element elements. In order to transfer the three-dimensional damage voxel field data generated in step S1.2 to the finite element model, the multi-source damage data mapping and modeling module 10 executes a voxel-element mapping procedure.
[0091] For any 1 in the finite element model Finite element element Identify the set of voxels contained within the geometric space of the unit. :
[0092] ;
[0093] In the formula, Representing finite element The volume area occupied; Denotes any single voxel in a 3D voxel mesh, where The voxel is in Grid index along the axis; Voxel representation The spatial coordinates of the center point in the global coordinate system; Representing finite element The actual geometric volume region occupied in three-dimensional space. For cross-sectional defect regions identified by appearance scanning, if a voxel is located within the defect space, it is removed from the set or its stiffness contribution is set to zero.
[0094] Calculate finite element Material stiffness reduction factor The material stiffness reduction factor is derived from the set The volume average of the comprehensive damage factor for all voxels was derived:
[0095] ;
[0096] In the formula, For set The total number of voxels contained therein; This represents the comprehensive damage factor.
[0097] Based on the calculated reduction factor Reconstruction unit Element stiffness matrix :
[0098] ;
[0099] In the formula, This is the stiffness matrix of the ideal elastic element generated based on the original design material parameters.
[0100] The assembly and solution process of the finite element stiffness matrix is a conventional technique in the field of computational mechanics. Those skilled in the art can implement it based on existing finite element analysis theory, so it will not be elaborated here.
[0101] The multi-source damage data mapping and modeling module 10 outputs a refined finite element model containing initial damage information. A point-to-point mapping from detection data to mechanical model parameters is achieved through a three-dimensional damage voxel field, reflecting the spatial variability of local structural stiffness and providing a benchmark model consistent with the physical condition for subsequent model correction and load-bearing capacity assessment.
[0102] See attached document Figure 4 In step S2, the heavy load simulation and equivalent test design module 20 performs the transformation process from virtual simulation to physical test scheme. This process specifically includes three implementation stages: target working condition limit state deduction, equivalent loading parameter calculation, and measurement point layout optimization.
[0103] S2.1, Target Heavy Load Condition Limit State Deduction:
[0104] The heavy-load simulation and equivalent test design module 20 calls the initial finite element model generated in step S1 and sets boundary conditions to simulate the actual support state of the bridge. The heavy-load simulation and equivalent test design module 20 reads the technical parameters of the special equipment (such as crawler cranes, beam transport vehicles, etc.) to be subjected to heavy-load operation and establishes a heavy-load model. This heavy-load model not only includes the total weight of the equipment, but also specifically includes the ground pressure distribution characteristics and external geometric dimensions of the equipment.
[0105] A heavy-load model is applied to the initial finite element model to simulate the movement and operation of heavy-load equipment on the bridge deck (such as crane lifting and rotation). The internal force response field (bending moment, shear force) and displacement response field of the structure under various working conditions are calculated using a finite element solver. Based on the calculation results, the control section locations of the bridge structure are identified. The control section is defined as the section location where the maximum internal force or maximum deformation occurs during heavy-load operation, denoted as […]. Extract the theoretical maximum response value of the control section under the most unfavorable working condition. (e.g., maximum positive bending moment or maximum mid-span deflection).
[0106] S2.2 Calculation of loading scheme based on the principle of equivalent internal forces:
[0107] The heavy-load simulation and equivalent test design module 20 designs a static load test scheme based on the equivalent internal force principle. This equivalent internal force principle utilizes a standard test vehicle with a relatively small total weight. By adjusting the vehicle arrangement and parking position, the static load test scheme is designed at the control section. The internal force effect generated at the point has a preset proportional relationship with the heavy load condition.
[0108] The heavy-load simulation and equivalent test design module 20 executes the control section influence line extraction program. In the initial finite element model generated in step S1, the loading path is defined along the preset lane centerline. Load path Discretized into a series of closely spaced nodes, a unit vertical load (F) is applied sequentially to each discrete node. unit =1), call the finite element solver to perform static analysis, extract and record the control sections. The corresponding response values (bending moment or deflection) are obtained at each point, thus constructing a discrete influence line data sequence. Based on this influence line data sequence, a continuous influence line function is generated through spline interpolation. The influence line function represents the effect of a unit load on the longitudinal coordinates of the bridge deck. When the cross section is in a certain position, the mechanical response generated is controlled.
[0109] The Heavy-Load Simulation and Equivalent Test Design Module 20 selects a standard three-axle or four-axle heavy-duty truck as the test loading source, defines the axle load distribution and wheelbase parameters of a single test vehicle, simplifies them into a concentrated force sequence, and constructs a system based on the number of test vehicles n and the coordinates of the first vehicle's stopping position. and vehicle spacing To optimize the computational model of the variables. The computational objective is to ensure that the test vehicle fleet follows the influence line function. The cumulative effect value Response under target heavy load conditions Within the safe equivalent range.
[0110] The constraints of the equivalent loading satisfy the following formula:
[0111] ;
[0112] In the formula, The theoretical maximum response value generated by the heavy-duty equipment at the control section; n is the total number of test vehicles to be used; This refers to the number of axles on a single test vehicle. Let m be the axle load of the m-th axle; For the first The coordinates of the front of the test vehicle; Let m be the distance of the m-th axle relative to the front of the vehicle; The influence line function of the control section extracted using the unit load method is mentioned above; and These are the lower and upper limits of the set equivalent proportional coefficient. The value is set to 0.6 to 0.7 to ensure that the test load can induce sufficient observable deformation in the structure; The value of is configured to be between 0.8 and 0.9, and it must be ensured that the calculated test stress is less than the elastic limit of the material, so as to ensure that the test process is within the elastic working range of the structure and does not cause permanent damage.
[0113] Through the above optimization calculations, the optimal number of vehicles n and its corresponding set of precise parking location coordinates that satisfy the inequality constraints are output. .
[0114] S2.3, Measurement point layout and observation strategy formulation:
[0115] The heavy load simulation and equivalent test design module 20 is based on the determined control section position. Based on the test load distribution, an automatic sensor deployment plan is generated.
[0116] In the control section Strain measuring points are placed in the tension zone at the bottom of the beam to monitor the bending performance of the structure; strain measuring points are placed near the support section to monitor the shear performance of the structure; and deflection measuring points are placed at the mid-span and L / 4 section to monitor the overall stiffness of the structure.
[0117] For the existing crack area detected in step S1, if the existing crack area is located within the influence range of the stress of the control section, it is marked as a key observation object and a crack width observation point is set.
[0118] See attached document Figure 5 In step S3, the field response data acquisition and processing module 30 performs the acquisition and structured processing of the actual bridge test data, which specifically includes three implementation steps: sensor network deployment and signal acquisition, graded loading response recording, and construction of multi-dimensional measured response vectors.
[0119] S3.1 Sensor Network Deployment and Signal Acquisition:
[0120] Based on the measurement point layout scheme output in step S2, the on-site response data acquisition and processing module 30 establishes an on-site physical sensing network. This on-site physical sensing network includes, but is not limited to, the following data acquisition channels: strain sensor channels located at the bottom of each control section of the bridge, used to acquire micro-strain signals of the structure under load, with resistance strain gauges or vibrating wire strain gauges as the sensor types; displacement sensor channels located at the mid-span and quarter-span sections of the bridge, used to acquire vertical deflection signals of the beam, with connecting tube deflectometers, photoelectric deflectometers, or high-precision total stations as the sensor types; and crack observation channels located in areas with concentrated existing cracks, used to acquire signals of crack width opening and closing, with cross-crack displacement gauges or digital crack observation instruments as the sensor types.
[0121] The field response data acquisition and processing module 30 synchronously triggers all the above channels through wired or wireless bus protocols to record the original electrical signals at a fixed sampling frequency (e.g., 10Hz to 50Hz). As for the specific installation process of the sensor hardware and the shielding and grounding treatment of the signal transmission lines, those skilled in the art can implement them using conventional engineering test specifications, which will not be elaborated here.
[0122] S3.2, Hierarchical Loading Response Recording and Preprocessing:
[0123] The on-site response data acquisition and processing module 30, in conjunction with the on-site loading vehicle's mobile operation, executes a graded loading procedure. The loading process proceeds in the order of 0%, 50%, 100%, and unloading back to zero to ensure that the structure is always in an elastic working state and to eliminate the influence of nonlinear settlement of the supports.
[0124] After each load level stabilizes, the field response data acquisition and processing module 30 extracts a stable signal over a period of time for averaging to eliminate random environmental noise. For the acquired raw data, a baseline value subtraction operation is performed to convert the absolute reading into a relative increment caused by the test load.
[0125] For any i The measuring point, at the _th ... Net response value under level load The calculation formula is as follows:
[0126] ;
[0127] In the formula, For the first The measuring point at the ... Average signal measurement value during the stable phase of the load level; This is the initial reference value under no-load conditions before the start of the test; This is the temperature drift correction term, calculated based on the rate of change of ambient temperature during loading and the temperature sensitivity coefficient of the sensor.
[0128] S3.3 Construction of Multidimensional Measured Response Vector:
[0129] To adapt to subsequent model correction algorithms, the field response data acquisition and processing module 30 standardizes and assembles the net response values of different physical dimensions to construct a measured response vector. This measured response vector is used to describe the digital fingerprint of the bridge's actual mechanical behavior under specific test loads.
[0130] Define the measured response vector Its structural form is as follows:
[0131] ;
[0132] In the formula, express Net vertical deflection values at key measuring points; express Net measured strain values at key measuring points; express Measured crack width increments at key observation locations; This represents the matrix transpose symbol.
[0133] The on-site response data acquisition and processing module 30 performs weighted normalization on each element of the vector to eliminate the influence of different physical quantities (such as millimeter-level deflection and micro-strain level strain) on the convergence of the subsequent optimization function. The final generated measured response vector... It is stored in the system database as the input target data for the physical constraint model correction module 40 in step S4.
[0134] See attached document Figure 6 In step S4, the physical constraint model correction module 40 is used to perform the inversion optimization process of model parameters.
[0135] The physical constraint-type model correction module 40 defines the parameter vector to be corrected, selects the material properties of components that affect the overall stiffness and local stress of the structure as the correction objects, mainly including the elastic modulus correction coefficients of each element of the main beam, and defines the correction parameter vector. In this embodiment, Defined as the first The stiffness retention coefficients of each element relative to the ideal undamaged state, and the corrected element stiffness matrix. The calculation formula is:
[0136] ;
[0137] In the formula, This is the stiffness matrix of the ideal elastic element generated based on the original design material parameters.
[0138] The physical constraint model correction module 40 constructs an objective functional aimed at minimizing the error between the theoretical and measured responses, calls the finite element solver, and applies the current parameter vector. Calculate the theoretical response vector of the structure under experimental load. Calculate the theoretical response vector and the measured response vector constructed in step S3. The Euclidean distance or weighted sum of squared errors between them.
[0139] Target functional The mathematical expression is as follows:
[0140] ;
[0141] In the formula, Theoretical response vector The first in The nth element represents the nth element calculated by the model. The response value (such as deflection or strain) at each measuring point; For vectors The first in The nth element represents the measured nth element. The response values at each measuring point; For the first The weighting coefficient of each response component is set according to the signal-to-noise ratio of the measured data and the importance of the physical quantity. Let be the dimension of the response vector.
[0142] The physical constraint model correction module 40 executes a physical constraint injection procedure based on a three-dimensional damage voxel field. By reading the three-dimensional damage voxel field containing spatial damage distribution information generated in step S1, it performs physical constraint injection for each unit to be corrected. Read the average comprehensive damage factor of the voxels covered by the unit to be corrected, and denot it as . ,based on Build correction parameters Physical constraint boundaries:
[0143] ;
[0144] ;
[0145] In the formula, and The first The lower and upper limits of the stiffness correction coefficient for each element; This is the average damage factor of the voxel region corresponding to this unit (derived from laser scanning and ultrasound detection); The confidence interval radius of the test data reflects the accuracy fluctuation range of the non-destructive testing method, and is usually taken as 0.05 to 0.1.
[0146] After determining the objective functional After the constraints are determined, the physical constraint model correction module 40 uses a constraint optimization algorithm (such as a constrained particle swarm optimization algorithm, a genetic algorithm, or a sequential quadratic programming method, SQP) to iteratively solve the problem.
[0147] In each iteration, the constrained optimization algorithm generates a tentative parameter vector. The physical constraint model correction module 40 checks whether each element in the parameter vector satisfies the aforementioned physical constraint boundary conditions. For parameters that violate the constraints, the boundary projection method or the penalty function method is used to force them back to the feasible region. Within the system, the objective functional value is calculated, and the parameter vector is updated based on the gradient direction or population fitness.
[0148] When the objective functional value Less than the preset convergence threshold If the number of iterations reaches the maximum limit, the iteration stops, and the physical constraint model correction module 40 outputs the final optimal parameter vector. Based on this, the finite element model is updated to generate the final digital twin model of the existing bridge. This corrected digital twin model of the existing bridge integrates the appearance inspection data and mechanical response data, and has the ability to make high-fidelity predictions for subsequent heavy-load conditions.
[0149] See attached document Figure 7In step S5, the spatiotemporal envelope generation and dynamic control module 50 executes the implementation process from offline prediction to online control based on the corrected digital twin model output in step S4. This process achieves full-process safety coverage by constructing a multi-dimensional state space.
[0150] The spatiotemporal envelope generation and dynamic control module 50 defines a multidimensional state vector representing the characteristics of heavy-load operations. This multidimensional state vector contains all the key independent variables describing the spatiotemporal location and load state of the operating equipment. For typical mobile heavy-duty operations (such as crawler crane travel and lifting), the state vector... Defined as:
[0151] ;
[0152] In the formula, This indicates the planar coordinates of the geometric center of the equipment's ground contact surface in the bridge deck coordinate system; This indicates the rotation angle of the upper slewing structure of the lifting equipment relative to the chassis; This represents the current lifting weight (including the weight of the lifting gear). These variables constitute the four-dimensional parameter space of the operating system. .
[0153] The spatiotemporal envelope generation and dynamic control module 50, in the corrected finite element model, performs parameter space... Discretized traversal simulation is performed, generating a series of discrete working points according to preset step sizes (e.g., walking step size 1m, turning step size 5°, lifting step size 1t). For each working point... The nonlinear finite element solver is called to perform load-bearing capacity verification and calculate the corresponding structural safety factor. The safety factor calculation comprehensively considers the flexural bearing capacity, shear bearing capacity, and local stability of the component.
[0154] Based on the traversal simulation results, the spatiotemporal envelope generation and dynamic control module 50 constructs the spatiotemporal safety envelope for the operation. The envelope is defined as the set of all operating points that satisfy the safety criteria in the four-dimensional parameter space:
[0155] ;
[0156] In the formula, Represents a multidimensional state vector. This indicates the corresponding operating point. The structural safety factor, To standardize the minimum permissible safety factor (e.g., 1.5), mathematically, It manifests as a closed multidimensional hypervolume, the boundary surface of which is the critical safety surface. Any state point inside the volume represents a safe state of the structure, while points outside the volume represent potential risks of failure. This spatiotemporal safety envelope fully describes the distribution law of the ultimate load that the structure can withstand at different positions and angles.
[0157] During the implementation phase of heavy-load operations, the spatiotemporal envelope generation and dynamic control module 50 receives positioning and status sensor data installed on the operating equipment via a wireless communication interface. The sensor data includes real-time coordinates obtained through differential GPS. Real-time angle obtained through rotary encoder and the real-time load obtained through the torque limiter This allows them to assemble a real-time state vector. .
[0158] The spatiotemporal envelope generation and dynamic control module 50 executes a dynamic determination algorithm based on spatial geometric relationships to calculate the real-time state vector. Safety envelope in time and space of operation Boundary distance Based on the inclusion relationship between points and multidimensional geometric objects, hierarchical control instructions are output in real time:
[0159] when And the distance from the boundary is greater than the warning threshold. If the area is determined to be a safe zone, the system outputs a signal to allow operation.
[0160] when But the distance from the boundary is less than the warning threshold. If the area is identified as a warning zone, the system will output an audible and visual alarm signal to prompt the operator to reduce the speed or stop increasing the load.
[0161] when If the area is identified as a danger zone, the system will output an emergency shutdown signal and can also link with the field control loop to cut off the power to the equipment.
[0162] For the specific iterative process of finite element nonlinear solution and the numerical calculation algorithm for the distance between spatial geometric points and surfaces, those skilled in the art can implement it based on conventional algorithms in computational mechanics and computational geometry, and will not elaborate further here.
[0163] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A comprehensive evaluation method for the bearing capacity of existing continuous beam bridges, characterized in that, Includes the following steps: S1. Collect data on geometric defects, concrete strength, and internal defects of existing bridges, fuse and map them into a three-dimensional voxel mesh, construct a three-dimensional damage voxel field, and generate an initial finite element model containing initial damage information based on the three-dimensional damage voxel field. S2. Simulate heavy-load operation conditions in the initial finite element model, identify control sections and theoretical limit states, and design an equivalent static load test scheme based on the principle of equivalent internal forces to simulate the effects of heavy-load equipment using standard vehicles. S3. Perform on-site graded static load tests on the existing continuous beam bridge according to the equivalent static load test scheme, collect the deflection, stress and crack width changes of the existing continuous beam bridge under elastic working state as actual response data, and construct the measured response vector. S4. Calculate the error between the measured response vector and the theoretically calculated response of the initial finite element model, and within the parameter value range determined by the three-dimensional damage voxel field, iteratively adjust the stiffness parameters of the initial finite element model until the error meets the convergence condition, thus completing the initial finite element model correction. S5. Based on the modified finite element model, perform nonlinear bearing capacity deduction for the target heavy-load operation, generate the spatiotemporal safety envelope of the operation, and compare the spatial positional relationship between the current operation status and the spatiotemporal safety envelope of the operation in real time during construction, and output operation control instructions.
2. The method for comprehensive evaluation of the bearing capacity of existing continuous beam bridges according to claim 1, characterized in that, Step S1 further includes: The physical space domain surrounding the existing continuous beam bridge is divided into a three-dimensional voxel grid, with each voxel serving as a data container for storing material property state values. Three-dimensional laser scanning point cloud data is selected as the geometric defect data, ultrasonic wave velocity data is selected as the internal defect data, and rebound strength data is selected as the concrete strength data. The point cloud data, wave velocity data, and rebound strength data are mapped to the center coordinates of the global voxel using a spatial interpolation algorithm. The comprehensive damage factor of each voxel is calculated to generate the three-dimensional damage voxel field. Identify the set of voxels contained within the geometric space of any element in the initial finite element model. Based on the volume average of the comprehensive damage factor of all voxels in the set of voxels, calculate the material stiffness reduction factor. Use the material stiffness reduction factor to reconstruct the stiffness matrix of the ideal elastic element generated based on the original design material parameters to obtain the initial finite element model.
3. The method for comprehensive evaluation of the bearing capacity of existing continuous beam bridges according to claim 2, characterized in that, The calculation of the comprehensive damage factor for each voxel is based on a weighted summation and consists of a first part and a second part: The first part is the weighting coefficient of the ultrasonic velocity index multiplied by the ultrasonic velocity relative deviation term, wherein the ultrasonic velocity relative deviation term is 1 minus the ratio of the measured ultrasonic velocity at the voxel position to the theoretical standard wave velocity of non-destructive concrete. The second part is the weighting coefficient of the rebound strength index multiplied by the relative deviation term of the rebound strength, wherein the relative deviation term of the rebound strength is 1 minus the ratio of the estimated value of the measured rebound strength at the voxel position to the standard value of the concrete strength.
4. The method for comprehensive evaluation of the bearing capacity of existing continuous beam bridges according to claim 1, characterized in that, In step S2, the design of the equivalent static load test scheme based on the principle of equivalent internal forces, which utilizes a standard vehicle to simulate the effects of heavy-duty equipment, specifically includes: A heavy load model is applied to the initial finite element model to calculate and identify the control section that generates the maximum internal force or maximum deformation during the heavy load operation, and extract the theoretical maximum response value of the control section under the most unfavorable working condition and the continuous influence line function along the loading path. An equivalent loading calculation model is constructed with the number of test vehicles and the coordinates of the vehicle parking positions as optimization variables. The superposition effect value generated by the test vehicle fleet at the control section is calculated using the continuous influence line function. The superposition effect value is between the lower limit and the upper limit of the proportional coefficient set by the theoretical maximum response value as a constraint condition. Solving the equivalent loading calculation model yields the number of vehicles and parking positions that satisfy the constraints. The setting range of the lower and upper limits of the proportional coefficient ensures that the test load induces observable deformation in the existing continuous beam bridge and keeps it in an elastic working state.
5. The method for comprehensive evaluation of the bearing capacity of existing continuous beam bridges according to claim 1, characterized in that, In step S3, the step of constructing the measured response vector further includes: Establish an on-site physical sensing network to collect micro-strain signals at the bottom of the control section, vertical deflection signals at the mid-span and quarter-point sections, and crack width opening and closing change signals in existing crack concentration areas. The acquired raw signals are denoised and the reference value is subtracted to convert the absolute readings into the net response values caused by the test load. The net response values of different physical dimensions are weighted, normalized, and standardized and assembled to form the measured response vector, which includes the measured net vertical deflection value, the measured net strain value, and the measured crack width increment value.
6. The method for comprehensive evaluation of the bearing capacity of existing continuous beam bridges according to claim 1, characterized in that, In step S4, the error between the calculated measured response vector and the theoretically calculated response of the initial finite element model specifically includes: The stiffness correction coefficients of each element in the initial finite element model are selected as parameters to be corrected, and the parameter vector to be corrected is constructed. Calculate the theoretical response vector of the initial finite element model under the experimental load based on the current parameter vector; A target functional is constructed to represent the error, wherein the target functional is the weighted sum of squared residuals or the Euclidean distance between the theoretical response vector and the measured response vector.
7. The method for comprehensive evaluation of the bearing capacity of existing continuous beam bridges according to claim 6, characterized in that, In step S4, the iterative adjustment of the stiffness parameters of the initial finite element model specifically includes: For each finite element element corresponding to the parameter to be corrected, the average comprehensive damage factor of the voxels covered by the finite element element in the three-dimensional damage voxel field is read. The physical constraint boundary is constructed based on the mean value of the comprehensive damage factor, and the corresponding stiffness correction coefficient of the finite element element is constructed. The physical constraint boundary includes a lower limit constraint value and an upper limit constraint value. The constrained optimization algorithm is used for iterative solution. In each iteration step, it is checked whether the generated tentative parameter vector satisfies the physical constraint boundary. For parameters that violate the constraints, they are forcibly pulled back to the feasible region defined by the lower constraint value and the upper constraint value.
8. The method for comprehensive evaluation of the bearing capacity of existing continuous beam bridges according to claim 1, characterized in that, In step S5, the step of generating the job spatiotemporal safety envelope further includes: Define a multidimensional state vector containing the plane coordinates of the working equipment, the rotation angle of the lifting equipment, and the current lifting weight, and construct the multidimensional parameter space of the working system. In the modified finite element model, the multidimensional parameter space is discretized and simulated, and the corresponding structural safety factor is calculated for each discrete working point. Construct the spatiotemporal safety envelope of the operation, which is the set of all working points in the multidimensional parameter space where the structural safety factor is greater than or equal to the minimum allowable safety factor.
9. The method for comprehensive evaluation of the bearing capacity of existing continuous beam bridges according to claim 1, characterized in that, In step S5, the real-time comparison of the spatial positional relationship between the current operation status and the operation's spatiotemporal safety envelope, and the output of operation control instructions, specifically includes: Receive positioning and status sensor data from on-site operating equipment, assemble them into a real-time status vector, and calculate the boundary distance between the real-time status vector and the operating spatiotemporal safety envelope; When the real-time status vector is located inside the work spatiotemporal safety envelope and the distance from the boundary is greater than the warning threshold, it is determined to be a safe zone and a work permission signal is output; when the real-time status vector is located inside the work spatiotemporal safety envelope but the distance from the boundary is less than the warning threshold, it is determined to be a warning zone and an audible and visual alarm signal is output; when the real-time status vector is located outside the work spatiotemporal safety envelope, it is determined to be a danger zone and an emergency stop signal is output.
10. A comprehensive evaluation system for the bearing capacity of existing continuous beam bridges, applied to the comprehensive evaluation method for the bearing capacity of existing continuous beam bridges as described in any one of claims 1-9, characterized in that, include: The multi-source damage data mapping and modeling module is used to collect data on geometric defects, concrete strength and internal defects of existing bridges, fuse and map them into a three-dimensional voxel mesh, construct a three-dimensional damage voxel field, and generate an initial finite element model containing initial damage information based on the three-dimensional damage voxel field. The heavy-load simulation and equivalent test design module is used to simulate heavy-load operating conditions in the initial finite element model, identify control sections and theoretical limit states, and design an equivalent static load test scheme that uses a standard vehicle to simulate the effects of heavy-load equipment based on the principle of equivalent internal forces. The on-site response data acquisition and processing module is used to perform on-site graded static load tests on the existing continuous beam bridge according to the equivalent static load test scheme, collect the deflection, stress and crack width changes of the existing continuous beam bridge under elastic working state as actual response data, and construct the measured response vector. The physical constraint model correction module is used to calculate the error between the measured response vector and the theoretically calculated response of the initial finite element model, and iteratively adjust the stiffness parameters of the initial finite element model within the parameter value range determined by the three-dimensional damage voxel field until the error meets the convergence condition, thus completing the initial finite element model correction. The spatiotemporal envelope generation and dynamic control module is used to perform nonlinear bearing capacity deduction of the target heavy-load operation based on the modified finite element model, generate the spatiotemporal safety envelope of the operation, and compare the spatial positional relationship between the current operation status and the spatiotemporal safety envelope of the operation in real time during construction, and output operation control instructions.