A BIM-based method and system for analyzing the real-scene model of road bridges

Abstract

The present invention discloses a method and system for analyzing the actual scene model of a road and bridge based on BIM, which relates to the technical field of road and bridges. The method extracts various characteristic parameters including curvature, wind-induced excitation angular frequency, vehicle excitation frequency, and the number of excited vehicles, etc. based on obtaining the actual trajectory of the stay cable and the discrete point set, constructs a stay cable feature set, and obtains a standardized feature set through preprocessing. Furthermore, based on the standardized feature set, the curvature offset rate O of each stay cable at different times is calculated and preliminarily compared and evaluated with a preset offset threshold Oth to identify the tension imbalance state. If there is an imbalance, a coupled analysis is triggered, and the comprehensive dynamic response intensity Rdyn is calculated based on the wind and vehicle excitation functions to reflect the response energy level of the stay cable per unit time. This process realizes the bidirectional analysis coupling of the structural deformation trend and the dynamic response intensity, improving the accuracy of fatigue potential identification and the response pertinence.

Description

Technical Field

[0001] The present invention relates to the technical field of road and bridge, and particularly to a method and system for analyzing the real-scene model of road and bridge based on BIM. Background Art

[0002] The present invention belongs to the field of intelligent detection and operation and maintenance of infrastructure, and particularly relates to the technologies of digital twin of civil engineering and health monitoring of bridge structures. More specifically, it focuses on the method for analyzing the real-scene model of road and bridge structures based on BIM. In modern urban traffic infrastructure, cable-stayed bridges are widely used due to their high structural efficiency, large span, and adaptability to complex terrains. As an important load-bearing component of the bridge, the geometric state and vibration behavior of the stay cables are directly related to the overall load-bearing performance and structural safety of the bridge. Therefore, the problems of state monitoring and fatigue assessment of stay cables have gradually become the key focus in the operation and maintenance management of road and bridges.

[0003] [[ID=ll]]At present, the structural state assessment of stay cables mainly relies on physically arranged vibration sensor arrays, stress gauges, or fiber optic cable control systems. Although these methods have high accuracy, they have the following problems: The installation and maintenance costs are extremely high. Especially on long-span bridges, it is economically infeasible to install sensors on each stay cable; The ability to obtain continuous data is lacking and it is easily affected by faults or climate interference; It is difficult to cover all stay cables or the entire length area of the stay cables, resulting in monitoring blind spots. Therefore, it is difficult to form a comprehensive, continuous, and dynamic health assessment mechanism for the overall stay cable network of the bridge. In this context, a new method based on the fusion of real-scene image reconstruction and BIM data has gradually attracted research attention, but there is still a lack of a technical path for closed-loop assessment, early warning, and response currently;

[0004] The main reason for the above deficiencies is that the traditional monitoring system highly relies on physical hardware deployment and lacks the coupling analysis ability between the structural geometric state and dynamic data of traffic / wind loads, and cannot detect potential fatigue risks induced by vehicle resonance or wind excitation in a timely manner. Especially under the conditions of frequent heavy loads or sudden wind conditions, the stay cables may show phenomena such as microstructural drift, abnormal bending, or uneven tension. Although cracks do not occur immediately during this process, it has entered the fatigue accumulation stage. If early identification and early warning cannot be achieved, it will lead to irreversible low-cycle fatigue cracks or changes in vibration modes in some stay cables, resulting in wire breaks and loose connections of the stay cables, ultimately affecting the long-term service safety and operation reliability of the entire bridge. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a method and system for analyzing the real-scene model of road and bridge based on BIM, which solves the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: including the following steps:

[0007] S1. Based on the real - scene model of the bridge road and the BIM construction line model, perform preliminary fusion to obtain the real - scene BIM model, and load traffic coupling data and wind force coupling data into the real - scene BIM model;

[0008] S2. Through the real - scene BIM model, extract the cable - stayed cable feature set in real - time, and pre - process the cable - stayed cable feature set to obtain a standardized feature set;

[0009] S3. Based on the standardized feature set, calculate the curvature offset rate O of each cable - stayed cable, and set an offset threshold Oth to make a preliminary comparison and evaluation with the curvature offset rate O to judge the tension state of each current cable - stayed cable;

[0010] S4. After the preliminary comparison and evaluation determines that the cable - stayed cable tension is unbalanced, then perform coupling analysis and calculate and output the comprehensive dynamic response intensity Rdyn;

[0011] S5. Based on the obtained comprehensive dynamic response intensity Rdyn and the curvature offset rate O, perform comprehensive calculation and output the cable - stayed cable fatigue risk factor S, and set a first risk threshold F1 and a second risk threshold F2 to make a secondary comparison and evaluation with the cable - stayed cable fatigue risk factor S, and perform risk level division based on the results of the secondary comparison and evaluation.

[0012] Preferably, the S1 includes S11, S12, and S13;

[0013] S11. Through multiple groups of fixed cameras deployed on the bridge road, take timed multi - angle photos of the structural body of the bridge road to obtain a real - time bridge road structure diagram. Through the SIFT feature extraction algorithm, extract key points from the real - time bridge road structure diagram, perform feature matching on the real - time bridge road structure diagram based on the key points, and use RANSAC to eliminate non - matching point pairs;

[0014] Reconstruct the shooting position of each real - time bridge road structure diagram, use the bundle adjustment algorithm to jointly optimize the internal and external three - dimensional point positions of the camera, and output the pose matrix and sparse point cloud of each real - time bridge road structure diagram;

[0015] Based on the pose matrix, apply the semi - global matching SGM and PatchMatch algorithms to improve the texture edge accuracy and perform dense point cloud reconstruction;

[0016] When converting the dense point cloud into mesh patches, use Delaunay triangulation combined with the real - time bridge road structure diagram to generate a textured real - scene model with UV mapping;

[0017] Use 3D modeling software to establish a BIM construction line model according to the bridge road design drawing data, and the design drawing data includes the positions of bridge towers, cable - stayed cable anchor points, cable lengths, modal frequency parameters, and material information.

[0018] Preferably, in S12, by setting the common control points of the real-scene model and the BIM structure line model, the ICP algorithm is used to perform transformation registration on the set of common control points of the dense point cloud of the real-scene model and the BIM structure line model;

[0019] The common control points include the bridge tower top, the anchorage point, and the intersection point of the bridge deck center line;

[0020] The transformation registration includes rotation, scaling, and translation;

[0021] Based on the obtained real-scene BIM model, the non-cable-stayed area is removed using the point density threshold to obtain the actual trajectory of the cable-stayed cable. Then, using the three-dimensional coordinate extraction algorithm for the actual trajectory of the cable-stayed cable, starting from the anchorage point, the discrete points P of the cable-stayed cable are extracted segment by segment according to the point cloud normal direction, where P j = (x, y, z), and P j represents the j-th discrete point, x represents the horizontal axis coordinate of the cable-stayed cable, y represents the vertical axis coordinate of the cable-stayed cable, and z represents the vertical axis coordinate of the cable-stayed cable;

[0022] S13, by setting the API application programming interface, the real-scene BIM model is connected to the meteorological database in real time to extract the wind force coupling data in real time;

[0023] The wind force coupling data includes the wind speed time-varying curve and the natural frequency;

[0024] The traffic coupling data is obtained through the inductive loop and video recognition on the bridge road;

[0025] The traffic coupling data includes the vehicle speed, the vehicle load, and the vehicle passing frequency Fk of the k-th vehicle;

[0026] Then, the wind force coupling data and the traffic coupling data are loaded into the real-scene BIM model for structural response fusion.

[0027] Preferably, the S2 includes S21 and S22;

[0028] S21, based on the time BIM model after structural response fusion, the cable-stayed cable feature set is extracted in real time;

[0029] The cable-stayed cable feature set includes the curvature k, the wind-induced excitation angular frequency W wind , the vehicle excitation frequency W veh and the number of exciting vehicles M;

[0030] The curvature k is calculated and extracted by the three-point curvature method; assuming that A, B, and C are three consecutive discrete points (P j-1 , P j , P j+1 ), the specific algorithm formula is: , where k j represents the curvature of the j-th discrete point;

[0031] The wind-induced excitation angular frequency W wind is obtained by extracting the wind speed V, the diameter D of the stay cable, and the Strouhal number St after the real-scene BIM model and the structural response are fused, and then performing dimensionless processing and comprehensive calculation and extraction. The specific algorithm formula is: W wind（i） = 2π(St·V i / D i ), where W wind（i） represents the wind-induced excitation angular frequency of the i-th stay cable, π represents the pi, with a value of 3.14, V i represents the wind speed at the location of the i-th stay cable, D i represents the diameter of the i-th stay cable;

[0032] The vehicle excitation frequency W veh is obtained by extracting the number of excited vehicles M, the vehicle speed v, and the vehicle length cd after the real-scene BIM model and the structural response are fused, and then performing dimensionless processing and calculation and extraction. The specific algorithm formula is: W veh（c） = 2π(v c / cd c ), where W veh（c） represents the vehicle excitation frequency of the c-th vehicle, v c represents the passing vehicle speed of the c-th vehicle, cd c represents the vehicle length of the c-th vehicle;

[0033] S22. Preprocess the obtained stay cable feature set, and the preprocessing includes timestamp alignment and normalization processing;

[0034] The timestamp alignment is performed by adding timestamps to all the data in the stay cable feature set and unifying them based on the time interval [t0, t1], where t0 represents the reference state time and t1 represents the comparison state time;

[0035] The normalization processing is performed by using the Min-Max normalization method to normalize all the data in the stay cable feature set, eliminate the influence of the unit dimension of all the data in the stay cable feature set, and obtain the normalized feature set.

[0036] Preferably, S3 includes S31 and S32;

[0037] S31. Based on the real-time obtained curvature k, calculate and analyze the curvature offset rate O of each stay cable from the reference state time t0 to the comparison state time t1, and analyze the tension balance of each stay cable under the time-varying state;

[0038] The curvature offset rate O is calculated and output through the following algorithm formula;

[0039] ;

[0040] In the formula, O i represents the curvature offset rate of the i-th stay cable, n represents the total number of discrete points, and k ij (t1) represents the curvature of the j-th discrete point of the i-th stay cable at the comparison state moment, and k ij (t0) represents the curvature of the j-th discrete point of the i-th stay cable at the reference state moment. represents the initial average curvature of the i-th stay cable.

[0041] Preferably, in S32, a preset offset threshold Oth is set by the user according to the abnormal offset of the stay cable, and the curvature offset rate O i of the i-th stay cable obtained in real time is preliminarily compared and evaluated with the offset threshold Oth to preliminarily judge the tension state of the stay cable, and based on the evaluation result, trigger coupling analysis. The specific evaluation content is as follows;

[0042] When the curvature offset rate O i of the i-th stay cable < offset threshold Oth, it indicates that the stay cable structure is stable;

[0043] When the curvature offset rate O i of the i-th stay cable ≥ offset threshold Oth, it is preliminarily determined that the stay cable tension is unbalanced, and at this time, trigger coupling analysis.

[0044] Preferably, the S4 includes S41;

[0045] S41. After preliminarily comparing and evaluating and judging that the tension of the current stay cable is unbalanced, coupling analysis is carried out. The coupling analysis extracts the wind-induced excitation angular frequency W wind , vehicle excitation frequency W veh and the number of exciting vehicles M, and conducts the energy response intensity of the coupling wind and traffic on the current stay cable per unit time to obtain the comprehensive dynamic response intensity Rdyn, and analyzes the total structural response generated by the superposition of wind force and traffic within a certain period of time;

[0046] The comprehensive dynamic response intensity Rdyn is calculated and output through the following algorithm formula;

[0047] ;

[0048] In the formula, Rdyn i represents the comprehensive dynamic response intensity of the i-th stay cable, a i represents the wind-induced excitation intensity coefficient of the i-th stay cable, and sin represents the sine function. Represents the vehicle excitation intensity of the c-th vehicle, Represents the phase perturbation term of the c-th vehicle, X i (t) represents the modal filtering window function of the i-th stay cable at time t, and dt represents the time calculus variable.

[0049] Preferably, the S5 includes S51 and S52;

[0050] S51. Based on the comprehensive dynamic response intensity Rdyn and the curvature offset rate O of the current stay cable, perform comprehensive calculation to output the stay cable fatigue risk factor S, fuse the tension non-uniformity and dynamic response, and analyze the fatigue risk of the stay cable;

[0051] The stay cable fatigue risk factor S is calculated and output through the following algorithm formula;

[0052] ;

[0053] In the formula, S i Represents the stay cable fatigue analysis factor of the i-th stay cable, and log represents the logarithmic function.

[0054] Preferably, S52. Based on the stay cable fatigue risk factor S of the historical stay cable under normal conditions, set the first risk threshold F1 and the second risk threshold F2, where the first risk threshold F1 represents the lower limit value of the stay cable fatigue, and the second risk threshold F2 represents the upper limit value of the stay cable fatigue. When the stay cable fatigue analysis factor S of the i-th stay cable obtained in real time i Is compared with the first risk threshold F1 and the second risk threshold F2 for secondary comparison and evaluation to judge the fatigue condition of the stay cable, and based on the results of the secondary comparison and evaluation, perform risk level classification. The specific evaluation content is as follows;

[0055] When the stay cable fatigue analysis factor S of the i-th stay cable i <The first risk threshold F1, it means that there is a preliminary risk for the current i-th stay cable. At this time, the current stay cable is classified as a first-level risk, and a reminder for regular re-modeling is issued;

[0056] When the first risk threshold F1 ≤ the stay cable fatigue analysis factor S of the i-th stay cable i <The second risk threshold F2, it means that there is an abnormality in the current stay cable. At this time, the current stay cable is classified as a second-level risk, and the current stay cable is marked and added to the next cycle operation and maintenance list;

[0057] When the stay cable fatigue analysis factor S of the i-th stay cable i ≥The second risk threshold F2, it means that there is a danger in the current stay cable. At this time, the current stay cable is classified as a third-level risk, an immediate warning is issued, the current stay cable number is output, and it is prompted to perform immediate operation and maintenance.

[0058] A BIM-based real-scene model analysis system for road and bridge, including a model fusion module, a multi-dimensional data extraction module, a stay cable curvature analysis module, a windmill coupling module and a fatigue analysis module;

[0059] The model fusion module obtains a real-scene BIM model through preliminary fusion of the real-scene model of the bridge and road and the BIM construction line model, and loads traffic coupling data and wind force coupling data into the real-scene BIM model;

[0060] The multi-dimensional data extraction module extracts the stay cable feature set in real time through the real-scene BIM model, and preprocesses the stay cable feature set to obtain a standardized feature set;

[0061] The stay cable curvature analysis module calculates the curvature offset rate O of each stay cable based on the standardized feature set, and sets an offset threshold Oth to conduct a preliminary comparison and evaluation with the curvature offset rate O to judge the tension state of each current stay cable;

[0062] When the windmill coupling module determines that the stay cable tension is unbalanced through the preliminary comparison and evaluation, it conducts coupling analysis and calculates and outputs the comprehensive dynamic response intensity Rdyn;

[0063] The fatigue analysis module comprehensively calculates and outputs the stay cable fatigue risk factor S based on the obtained comprehensive dynamic response intensity Rdyn and the curvature offset rate O, sets a first risk threshold F1 and a second risk threshold F2 to conduct a secondary comparison and evaluation with the stay cable fatigue risk factor S, and conducts risk level division based on the results of the secondary comparison and evaluation.

[0064] The present invention provides a BIM-based real-scene model analysis method and system for road and bridge. It has the following beneficial effects:

[0065] (1) This method constructs a high-precision real-scene model by deploying multiple groups of fixed cameras to take timed multi-angle photos of the bridge and road structure body, combining image processing algorithms such as SIFT feature extraction, RANSAC registration, bundle adjustment algorithm and dense point cloud reconstruction; and conducts three-dimensional spatial registration of the real-scene model and the BIM construction line model through the ICP algorithm to form a real-scene BIM model with highly integrated structural semantics and spatial structure. On this basis, wind force coupling data and traffic coupling data are obtained in real time through the API interface, and loaded into the fused real-scene BIM model for structural response coupling analysis, realizing digital identification of the stay cable structure state and dynamic load modeling without relying on traditional vibration sensors or strain gauges, greatly reducing the deployment cost of the monitoring system and improving the integrity of the detection area.

[0066] (2) Based on obtaining the actual trajectory of the stay cable and the discrete point set, this method extracts various characteristic parameters including curvature, wind-induced excitation angular frequency, vehicle excitation frequency, and the number of exciting vehicles, constructs a stay cable feature set, and realizes data standardization through timestamp alignment and Min-Max normalization to obtain a standardized feature set. Furthermore, based on the said feature set, the curvature offset rate O of each stay cable at different moments is calculated and preliminarily compared and evaluated with the preset offset threshold Oth to identify the tension imbalance state. If there is an imbalance, coupling analysis is triggered, and the comprehensive dynamic response intensity Rdyn is calculated based on the wind and vehicle excitation functions to reflect the response energy level of the stay cable per unit time. This process realizes the two-way analysis coupling of the static structural deformation trend and the dynamic dynamic response intensity through formula modeling, improving the accuracy of fatigue potential identification and the response pertinence.

[0067] (3) This method constructs a fatigue risk factor S that fuses the tension state and the dynamic response i , to achieve non-linear comprehensive quantification of the structural state of the stay cable; and introduces the first risk threshold F1 and the second risk threshold F2 to conduct a secondary comparison and evaluation of the risk factor, forming a three-level risk classification strategy: preliminary risk, abnormal risk, and dangerous risk. When the stay cable fatigue risk factor is lower than the first risk threshold F1, a regular modeling reminder is automatically sent; when the risk value is between the first risk threshold F1 and the second risk threshold F2, it is marked to enter the next cycle operation and maintenance plan; when it exceeds the second risk threshold F2, an immediate warning is given and the stay cable number and prompt information are output. This mechanism can realize the early identification of structural abnormalities, the automatic grading of operation and maintenance tasks, and the establishment of a high-risk stay cable priority response mechanism, which helps to improve the initiative, intelligence, and controllability of bridge structure maintenance. Description of the Drawings

[0068] Figure 1 Schematic diagram of the steps of a BIM-based road and bridge real-scene model analysis method of the present invention;

[0069] Figure 2 Schematic diagram of the process of a BIM-based road and bridge real-scene model analysis system of the present invention;

[0070] Figure 3 Schematic diagram of the extraction of three-dimensional discrete points of the stay cable;

[0071] Figure 4 Schematic diagram of the registration of the real-scene model and the BIM construction line. Detailed Implementation Modes

[0072] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0073] Embodiment 1: Please refer to Figure 1 、 Figure 3 and Figure 4 , the present invention provides a method for analyzing the real - scene model of road bridges based on BIM. To achieve the above - mentioned purpose, the present invention is realized through the following technical solutions: including the following steps:

[0074] S1. Based on the real - scene model of the bridge road and the BIM construction line model, perform preliminary fusion to obtain a real - scene BIM model, and load traffic coupling data and wind force coupling data into the real - scene BIM model;

[0075] S2. Extract the cable - stayed cable feature set in real - time through the real - scene BIM model, and pre - process the cable - stayed cable feature set to obtain a standardized feature set;

[0076] S3. Calculate the curvature offset rate O of each cable - stayed cable based on the standardized feature set, and set an offset threshold Oth to conduct a preliminary comparison and evaluation with the curvature offset rate O to judge the tension state of each current cable - stayed cable;

[0077] S4. After the preliminary comparison and evaluation determines that the cable - stayed cable tension is unbalanced, perform coupling analysis and calculate and output the comprehensive dynamic response intensity Rdyn;

[0078] S5. Based on the obtained comprehensive dynamic response intensity Rdyn and the curvature offset rate O, perform comprehensive calculation to output the cable - stayed cable fatigue risk factor S, and set a first risk threshold F1 and a second risk threshold F2 to conduct a secondary comparison and evaluation with the cable - stayed cable fatigue risk factor S, and conduct risk level division based on the results of the secondary comparison and evaluation.

[0079] In this embodiment, the method constructs a closed-loop analysis process from multi-source data fusion, structural parameter extraction, state analysis to fatigue risk grading by focusing on potential hidden problems such as tension imbalance, structural fatigue, and vibration resonance that may occur during the service of the cable-stayed bridge. In the specific implementation, first, the three-dimensional real-scene model constructed from the real-scene images of the bridge road is registered and fused with the BIM construction line model to construct a real-scene BIM model with real geometric states and structural semantic attributes, and the environmental coupling data such as traffic and wind force are synchronously loaded to form a structural response analysis benchmark in the unified spatio-temporal domain. Subsequently, the characteristic parameter set of the cable-stayed cable is extracted in real time through this model and standardized to unify the data format and time series coordinates. On this basis, based on the standardized feature set, the curvature offset rate of each cable-stayed cable at different times is calculated and compared with the set offset threshold to judge whether the tension is abnormal from the perspective of geometric shape change; if the tension imbalance is identified, a time-varying superposition model of wind force and traffic load is established to calculate the coupled dynamic response intensity of the cable and evaluate the vibration energy accumulation within a given time window. Finally, by combining the curvature offset rate and the dynamic response intensity, the cable-stayed cable fatigue analysis factor S of the i-th cable-stayed cable with two fused factors is constructed. i And through the first risk threshold F1 and the second risk threshold F2 for hierarchical discrimination, multi-level risk classification and intelligent decision support for the operating state of the cable-stayed cable are realized. Through the above implementation, the present invention effectively realizes non-contact, high-precision, and sustainable monitoring of the structural state of the cable-stayed cable, and solves the problems of high sensor deployment cost, limited coverage, difficult coupling of dynamic loads, and difficult quantification of fatigue risks in the prior art. The proposed method not only improves the accuracy and scope of the perception of the operating state of the cable-stayed bridge, but also enhances the foresight and response efficiency of fatigue risk early warning, providing a low-cost, highly intelligent, and strongly visual comprehensive solution for bridge structure maintenance management, and having significant engineering promotion value and industrial application prospects;

[0080] Embodiment 2: Please refer to Figure 1 、 Figure 3 and Figure 4 , specifically: S1 includes S11, S12, and S13;

[0081] S11. Through multiple groups of fixed cameras deployed on the bridge road, the structural body of the bridge road is photographed at multiple angles at regular intervals to obtain the real-time bridge road structure diagram. The key points are extracted from the real-time bridge road structure diagram through the SIFT feature extraction algorithm, and the feature matching of the real-time bridge road structure diagram is performed based on the key points. RANSAC is used to remove the unmatched point pairs;

[0082] Reconstruct the shooting position of each real-time bridge road structure diagram, use the bundle adjustment algorithm, jointly optimize the internal and external three-dimensional point positions of the camera, and output the attitude matrix and sparse point cloud of each real-time bridge road structure diagram.

[0083] Apply the semi - global matching (SGM) and PatchMatch algorithms based on the pose matrix to improve the texture edge accuracy and perform dense point cloud reconstruction;

[0084] Convert the dense point cloud into mesh patches, and use Delaunay triangulation combined with the real - time bridge road structure diagram to generate a textured real - scene model for UV mapping;

[0085] Use 3D modeling software to establish a BIM construction line model according to the bridge road design drawing data, where the design drawing data includes the positions of bridge towers, cable - stayed cable anchor points, cable lengths, modal frequency parameters, and material information.

[0086] S12: By setting the common control points of the real - scene model and the BIM construction line model, use the ICP algorithm to perform transformation registration on the set of common control points of the dense point cloud of the real - scene model and the BIM construction line model;

[0087] The common control points include the top of the bridge tower, the anchor point, and the intersection of the bridge deck center line;

[0088] The transformation registration includes rotation, scaling, and translation;

[0089] Based on the obtained real - scene BIM model, use the point density threshold to eliminate the non - cable - stayed cable area, obtain the actual trajectory of the cable - stayed cable, and then use the 3D coordinate extraction algorithm to extract the discrete points P of the cable - stayed cable from the anchor point along the point cloud normal direction segment by segment. Among them, P j = (x, y, z), P j represents the j - th discrete point, x represents the horizontal axis coordinate of the cable - stayed cable, y represents the vertical axis coordinate of the cable - stayed cable, and z represents the vertical axis coordinate of the cable - stayed cable;

[0090] S13: Set up an API application interface to connect the real - scene BIM model with the meteorological database in real - time and extract wind - force coupling data in real - time;

[0091] The wind - force coupling data includes the time - varying curve of wind speed and natural frequency;

[0092] Obtain traffic - coupling data through inductive loop and video recognition on the bridge road;

[0093] The traffic - coupling data includes vehicle speed, vehicle load, and the vehicle passing frequency Fk of the k - th vehicle;

[0094] Then load the wind - force coupling data and traffic - coupling data into the real - scene BIM model for structural response fusion.

[0095] In this embodiment, the method constructs a real - scene BIM model by fusing real - scene images and BIM construction line models, providing a unified and accurate spatial basis for subsequent cable - stayed structure state recognition and risk assessment. Specifically, multi - group fixed cameras on the bridge are used to obtain multi - angle real - time images. The SIFT feature extraction and RANSAC algorithm are used to achieve image registration. The bundle adjustment algorithm, SGM, and PatchMatch algorithm are combined to complete the dense point cloud reconstruction, and a high - precision real - scene model with texture mapping is generated through Delaunay triangulation. At the same time, combined with the design drawing information, a BIM construction line model with complete structural semantics is constructed, including the core structural information of the bridge road. On this basis, by setting common control points, the ICP algorithm is used to perform coordinate transformation registration on the real - scene model and the BIM model to ensure their spatial alignment. The cable - stayed cable area and its actual trajectory are extracted based on point density and normal direction in the dense point cloud, and the discrete points of the cable - stayed cable are extracted segment by segment through the three - dimensional coordinate extraction algorithm, laying the spatial analysis accuracy for subsequent curvature and dynamic response calculations. Further, external coupling information is introduced. Through the API interface, it is linked with the meteorological database to extract the time - varying curve of wind speed and the natural frequency of the cable - stayed cable. At the same time, traffic flow information is obtained based on the inductive loop and video recognition algorithm, and the above - mentioned multi - source dynamic data are uniformly loaded into the real - scene BIM model to complete the fusion of the structure - environment coupling input. Through the comprehensive implementation of the above technical path, the present invention not only realizes the deep fusion of the real - scene and BIM and the unity of structural semantics, but also establishes a real - time coupling channel between the wind and traffic load data and the structural geometric state, significantly improving the authenticity of structural modeling, the accuracy of analysis and calculation, and the reliability of subsequent risk identification.

[0096] Embodiment 3: Please refer to Figure 1 , specifically: S2 includes S21 and S22;

[0097] S21. Based on the time - BIM model after structural response fusion, the cable - stayed cable feature set is extracted in real - time;

[0098] The cable - stayed cable feature set includes curvature k, wind - induced excitation angular frequency W wind , vehicle excitation frequency W veh and the number of exciting vehicles M;

[0099] The curvature k is calculated and extracted by the three - point curvature method; assume that A, B, and C are three consecutive discrete points (P j-1 , P j , P j+1 ), and the specific algorithm formula is: , where k j represents the curvature of the j - th discrete point;

[0100] The wind - induced excitation angular frequency W windBy extracting the wind speed V, cable diameter D, and Strouhal number St after the integration of the real - scene BIM model and structural response, and performing dimensionless processing, comprehensive calculation and extraction are carried out. The specific algorithm formula is: W wind（i） = 2π(St·V i / D i ), where W wind（i） represents the angular frequency of the wind - induced excitation of the i - th stay cable, π represents the pi, with a value of 3.14, V i represents the wind speed at the location of the i - th stay cable, D i represents the diameter of the i - th stay cable;

[0101] The vehicle excitation frequency W veh is calculated and extracted by performing dimensionless processing on the number of exciting vehicles M, vehicle speed v, and vehicle length cd after extracting the real - scene BIM model and integrating the structural response. The specific algorithm formula is: W veh（c） = 2π(v c / cd c ), where W veh（c） represents the vehicle excitation frequency of the c - th vehicle, v c represents the passing vehicle speed of the c - th vehicle, cd c represents the vehicle length of the c - th vehicle;

[0102] S22. Pre - process the obtained stay - cable feature set. The pre - processing includes timestamp alignment and normalization processing;

[0103] Timestamp alignment is carried out by adding timestamps to all data in the stay - cable feature set and unifying them based on the time interval [t0, t1], where t0 represents the reference state time and t1 represents the comparison state time;

[0104] Normalization processing is carried out by using the Min - Max normalization method to normalize all data in the stay - cable feature set, eliminating the influence of the unit dimension of all data in the stay - cable feature set and obtaining a normalized feature set.

[0105] In this embodiment, the method extracts the key physical and geometric response characteristics of the stay cables and completes the standardized preprocessing, providing a unified data basis for subsequent curvature offset calculation and dynamic coupling analysis. Among them, by analyzing the discrete point trajectories of each stay cable in the fusion model, a stay cable feature set is constructed. Each parameter is derived from the structural spatial information mapped by the real-scene BIM model and the dynamic external input data, with clear physical meaning and a closed calculation path. The preprocessing of the extracted stay cable feature set specifically includes two key processes: timestamp alignment and numerical normalization. By introducing a unified time interval [t0, t1], consistent time sequence labels are assigned to all parameters to ensure comparability and synchronization in subsequent analysis; at the same time, through the Min-Max normalization method, data with different dimensions, such as frequency, curvature, and vehicle speed, are converted into standardized dimensionless expressions, avoiding bias in subsequent modeling and evaluation results due to numerical scale differences, thereby generating a standardized feature set.

[0106] Embodiment 4: Please refer to Figure 1 , specifically: S3 includes S31 and S32;

[0107] S31. Based on the real-time obtained curvature k, calculate and analyze the curvature offset rate O of each stay cable from the reference state time t0 to the comparison state time t1, and analyze the tension balance of each stay cable under the time-varying state;

[0108] The curvature offset rate O is calculated and output through the following algorithm formula;

[0109] ;

[0110] In the formula, O i represents the curvature offset rate of the i-th stay cable, n represents the total number of discrete points, k ij (t1) represents the curvature of the j-th discrete point of the i-th stay cable at the comparison state time, k ij (t0) represents the curvature of the j-th discrete point of the i-th stay cable at the reference state time, represents the initial average curvature of the i-th stay cable;

[0111] The physical meaning of the formula is that by selecting n discrete points P on the stay cable and recording their spatial positions in two time periods from the reference state time t0 to the comparison state time t1, curve fitting is performed on each discrete point P to obtain the curvatures k at the two times, and based on the curvatures k at the two times, the curvature k deviation is calculated to analyze the relative change degree of the spatial curvature of the stay cable, so as to analyze whether there is a suspicious change in the spatial geometric state of the stay cable, judge tension imbalance, structural looseness or fatigue, and taking the average is to synthesize the changes of all discrete points to obtain the geometric offset degree of the entire stay cable.

[0112] S32. Based on the abnormal offset of the stay cable, the user sets a preset offset threshold Oth, and preliminarily compares and evaluates the curvature offset rate Oi of the ith stay cable obtained in real time with the offset threshold Oth, preliminarily judges the tension state of the stay cable, and based on the evaluation result, triggers coupling analysis. The specific evaluation content is as follows;

[0113] When the curvature offset rate O of the ith stay cable i < the offset threshold Oth, it indicates that the stay cable structure is stable;

[0114] When the curvature offset rate O of the ith stay cable i ≥ the offset threshold Oth, it is preliminarily determined that the stay cable tension is unbalanced, and at this time, coupling analysis is triggered.

[0115] In this embodiment, this method further completes the quantitative identification and preliminary diagnosis of the stay cable tension state based on the previously extracted and standardized stay cable feature set. Among them, by extracting the curvature of equidistant scatter points of each stay cable at the reference state time t0 and the comparison state time t1, and based on the three-point curvature calculation method and the curvature difference ratio formula, the curvature offset rate O of the ith stay cable is output i , which is used to reflect the relative change degree of the cable spatial shape. This calculation not only considers the small spatial offset trend of multiple points along the length direction of the cable, but also establishes a stable proportional discrimination mechanism by normalizing the initial average curvature, effectively avoiding misjudgment caused by the numerical difference of absolute curvature, and realizing the geometric-level tracking analysis of the stay cable tension balance state. Immediately afterwards, the user-adjustable offset threshold Oth is set, and the curvature offset rate O of the ith stay cable of each cable obtained in real time i is preliminarily compared with the offset threshold Oth to form a lightweight and sensor-independent structural stability evaluation mechanism. Through the above implementation method, the present invention can realize the early identification of potential force imbalance, geometric drift or fatigue risk of the stay cable only by reconstructing the spatial state of the bridge stay cable in two time periods and calculating the curvature without the traditional stay cable tensiometer or fiber optic strain sensor. This method significantly improves the sensitivity and coverage of structural deformation identification, while reducing the dependence of the monitoring system on external devices, and enhancing the flexibility and applicability of the monitoring deployment;

[0116] Example 5: Please refer to Figure 1 , specifically: S4 includes S41;

[0117] S41. After preliminarily comparing and evaluating and judging that the tension of the current stay cable is unbalanced, coupling analysis is carried out. The coupling analysis extracts the wind-induced excitation angular frequency W in the standardized feature set wind , vehicle excitation frequency W vehWith the excitation vehicle number M, the energy response intensity of the coupled wind and traffic to the current stay cable per unit time is carried out, the comprehensive dynamic response intensity Rdyn is obtained, and the total structural response generated by the superposition of wind force and traffic within a certain time period is analyzed;

[0118] The comprehensive dynamic response intensity Rdyn is calculated and output through the following algorithm formula;

[0119] ;

[0120] In the formula, Rdyn i represents the comprehensive dynamic response intensity of the i-th stay cable, a i represents the wind-induced excitation intensity coefficient of the i-th stay cable, sin represents the sine function, represents the vehicle excitation intensity of the c-th vehicle, represents the phase perturbation term of the c-th vehicle, X i (t) represents the modal filtering window function of the i-th stay cable at time t, and dt represents the time calculus variable;

[0121] represents the wind-induced excitation term, which is used to represent the periodic response caused by the periodic action of the wind on the stay cable. If this frequency is close to the natural frequency of the stay cable, resonance may be caused. The amplitude of the wind excitation depends on the wind pressure and the windward area of the stay cable;

[0122] represents the traffic excitation term, which is used to represent that each vehicle brings a periodic impact. Due to the different vehicle speeds and wheelbases, their excitation frequencies are different. When multiple vehicles pass by, a composite frequency response spectrum will be formed. The phase perturbation term represents the timing perturbation caused by the non-simultaneous action of each vehicle;

[0123] The modal filtering window function X of the i-th stay cable at time t i (t) is used to extract only the response frequency band close to the natural frequency and modal frequency of the stay cable, which is equivalent to only calculating the partial response that may cause resonance;

[0124] represents integral solution, which is used to represent the vibration energy accumulation during the detection period;

[0125] Integrate the excitation function to obtain the cumulative intensity of the coupled excitation energy per unit time.

[0126] In this embodiment, after judging that there are signs of tension imbalance in the stay cable, the method further performs the coupled analysis of wind force and traffic excitation to obtain a more realistic and physically driven clear dynamic response evaluation result. Specifically, by extracting the wind-induced excitation angular frequency W in the standardized feature set wind, vehicle excitation frequency W veh and the number of excited vehicles M, a time-domain coupling expression of wind and vehicle loads is constructed. In this expression, the wind-induced excitation term simulates the periodic excitation caused by the change of the wind field in the form of a sine wave, and its amplitude is jointly determined by the wind pressure, the windward area of the cable and the local flow velocity; the traffic excitation term is composed of the superposition of different frequency components generated by the passing of multiple vehicles, and a phase perturbation term is introduced to consider the non-uniformity of the overall excitation caused by the vehicle time sequence distribution. This coupling function is multiplied by the modal filtering window function X i (t) of the i-th stay cable at time t, which is used to screen the excitation signal components close to the modal frequency of the current i-th stay cable, so as to retain only the structural response contribution in the potential resonance interval. Finally, by integrating this function within the monitoring time interval [t0, t1], the comprehensive dynamic response intensity Rdyn i of the i-th stay cable is output, which is used to measure the structural energy input generated by the combined wind and vehicle loads per unit time. Through the implementation of this step, the present invention realizes the targeted fusion of the external excitation environment for secondary response quantification analysis after identifying the initial abnormality of the structure. Compared with the traditional structural monitoring that only relies on a single judgment criterion of tension value or frequency change, this method effectively eliminates the interference of non-resonant perturbations through the coupling simulation of the wind and vehicle double excitation sources and modal filtering processing, and significantly improves the pertinence, physical accuracy and prediction ability of the response analysis. In addition, the obtained Rdyn i provides strong support for the calculation of the fatigue risk factor, enabling the entire risk discrimination system to have a complete progressive logic chain from state perception, abnormality identification to dynamic response quantification, and having high intelligence and engineering practical value.

[0127] Example 6: Please refer to Figure 1 , specifically: S5 includes S51 and S52;

[0128] S51. Based on the comprehensive dynamic response intensity Rdyn and the curvature offset rate O of the current stay cable, a comprehensive calculation is performed to output the stay cable fatigue risk factor S, and the tension non-uniformity and dynamic response are fused to analyze the fatigue risk of the stay cable;

[0129] The stay cable fatigue risk factor S is calculated and output through the following algorithm formula;

[0130] ;

[0131] In the formula, S i represents the stay cable fatigue analysis factor of the i-th stay cable, and log represents the logarithmic function;

[0132] log(1 + Rdyn-) represents the vibration response intensity after non-linear transformation, measuring the amount of external dynamic excitation energy received by the structure. The logarithmic function log is used to non-linearly compress the dynamic response quantity, preventing a small number of extremely large excitation values from dominating the scoring, ensuring that even for moderate excitations, high risk values can be generated when the structure is abnormally obvious, that is, when the curvature deviation rate O- of the i-th stay cable is high. That is, if the structure is already abnormal, even general excitation may lead to problems, providing a comprehensive evaluation mechanism that reflects the non-linear relationship between the structural state and the excitation energy, and is the core index for the final risk grading judgment.

[0133] S52. Based on the stay cable fatigue risk factor S of the historical stay cable under normal conditions, set the first risk threshold F1 and the second risk threshold F2. Among them, the first risk threshold F1 represents the lower limit value of stay cable fatigue, and the second risk threshold F2 represents the upper limit value of stay cable fatigue. When the stay cable fatigue analysis factor S of the i-th stay cable obtained in real time i is compared and evaluated twice with the first risk threshold F1 and the second risk threshold F2 to judge the fatigue condition of the stay cable, and the risk level is divided based on the results of the secondary comparison and evaluation. The specific evaluation content is as follows;

[0134] When the stay cable fatigue analysis factor S of the i-th stay cable i < F1, it indicates that there is a preliminary risk for the current i-th stay cable. At this time, the current stay cable is classified as a first-level risk, and a reminder for regular re-modeling is issued;

[0135] When F1 ≤ the stay cable fatigue analysis factor S of the i-th stay cable i < F2, it indicates that there is an abnormality in the current stay cable. At this time, the current stay cable is classified as a second-level risk, and the current stay cable is marked and added to the next cycle's operation and maintenance list;

[0136] When the stay cable fatigue analysis factor S of the i-th stay cable i ≥ F2, it indicates that there is a danger in the current stay cable. At this time, the current stay cable is classified as a third-level risk, an immediate warning is issued, the current stay cable number is output, and it is prompted to perform immediate operation and maintenance.

[0137] In this embodiment, this method is used to further construct the stay cable fatigue analysis factor S of the i-th stay cable that integrates the geometric state and the excitation response on the basis of having identified the abnormal tension of the stay cable and completed the calculation of the dynamic response intensity i and complete the classification evaluation of the structural risk and the output of response suggestions based on a multi-level threshold system. Specifically, through the stay cable fatigue analysis factor S of the i-th stay cable iCalculate the fatigue risk factor for each stay cable. Set the first risk threshold F1 and the second risk threshold F2 of the fatigue risk factor according to the statistical data of the historical bridge structure under normal operation conditions, and compare the currently calculated stay cable fatigue analysis factor S i value of the i-th stay cable with the first risk threshold F1 and the second risk threshold F2 for a secondary comparison to achieve a hierarchical identification of the fatigue state. Through the above implementation method, the present invention realizes a complete fatigue assessment closed-loop path from spatial geometric deformation, environmental excitation response, quantitative construction of risk factors, and hierarchical judgment output. Compared with the traditional structural safety assessment method that relies on manual experience judgment or a single stress threshold judgment, this method not only integrates the structural state and excitation energy in the assessment dimension, but also constructs a non-linear scoring mechanism through a logarithmic function, improving the response sensitivity of the model in the medium and high risk intervals, significantly enhancing the automatic judgment ability, early warning lead time, and decision-making reliability of the bridge stay cable fatigue monitoring system, and having extremely high practical engineering application and intelligent operation and maintenance promotion value.

[0138] Example 7: Please refer to Figure 1 and Figure 2 , a BIM-based road and bridge real-scene model analysis system, including a model fusion module, a multi-dimensional data extraction module, a stay cable curvature analysis module, a windmill coupling module, and a fatigue analysis module;

[0139] The model fusion module performs preliminary fusion on the real-scene model of the bridge road and the BIM construction line model to obtain a real-scene BIM model, and loads traffic coupling data and wind force coupling data into the real-scene BIM model;

[0140] The multi-dimensional data extraction module extracts the stay cable feature set in real time through the real-scene BIM model, and preprocesses the stay cable feature set to obtain a standardized feature set;

[0141] The stay cable curvature analysis module calculates the curvature offset rate O of each stay cable based on the standardized feature set, and sets an offset threshold Oth to make a preliminary comparison and evaluation with the curvature offset rate O to judge the tension state of each current stay cable;

[0142] When the windmill coupling module determines that the stay cable tension is unbalanced through the preliminary comparison and evaluation, it performs coupling analysis and calculates and outputs the comprehensive dynamic response intensity Rdyn;

[0143] The fatigue analysis module comprehensively calculates and outputs the stay cable fatigue risk factor S based on the obtained comprehensive dynamic response intensity Rdyn and the curvature offset rate O, sets the first risk threshold F1 and the second risk threshold F2 to make a secondary comparison and evaluation with the stay cable fatigue risk factor S, and divides the risk level based on the results of the secondary comparison and evaluation.

[0144] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention.

Claims

1. A method for analyzing the real - scene model of road and bridge based on BIM, characterized in that: It includes the following steps: S1. Based on the real - scene model and BIM construction line model of the bridge road, conduct preliminary fusion to obtain a real - scene BIM model, and load traffic coupling data and wind force coupling data into the real - scene BIM model; S2. Extract the cable - stayed cable feature set in real - time through the real - scene BIM model, and pre - process the cable - stayed cable feature set to obtain a standardized feature set; S3. Calculate the curvature offset rate O of each cable - stayed cable based on the standardized feature set, set an offset threshold Oth, and conduct a preliminary comparison and evaluation with the curvature offset rate O to judge the tension state of each current cable - stayed cable; S4. After the preliminary comparison and evaluation determines that the cable - stayed cable tension is unbalanced, conduct a coupling analysis and calculate and output the comprehensive dynamic response intensity Rdyn; S5. Based on the obtained comprehensive dynamic response intensity Rdyn and curvature offset rate O, conduct a comprehensive calculation to output the cable - stayed cable fatigue risk factor S, set a first risk threshold F1 and a second risk threshold F2, conduct a secondary comparison and evaluation with the cable - stayed cable fatigue risk factor S, and conduct risk level division based on the results of the secondary comparison and evaluation.

2. The method for analyzing the real - scene model of road and bridge based on BIM according to claim 1, wherein: The S1 includes S11, S12, and S13; S11. Through multiple groups of fixed cameras deployed on the bridge road, take timed multi - angle photos of the structural body of the bridge road to obtain a real - time bridge road structure diagram. Use the SIFT feature extraction algorithm to extract key points from the real - time bridge road structure diagram, perform feature matching on the real - time bridge road structure diagram based on the key points, and use RANSAC to eliminate non - matching point pairs; Reconstruct the shooting position of each real - time bridge road structure diagram, use the bundle adjustment algorithm to jointly optimize the internal and external three - dimensional point positions of the camera, and output the pose matrix and sparse point cloud of each real - time bridge road structure diagram; Based on the pose matrix, apply the semi - global matching SGM and PatchMatch algorithms to improve the texture edge accuracy and conduct dense point cloud reconstruction; Convert the dense point cloud into mesh patches, and use Delaunay triangulation combined with the real - time bridge road structure diagram to generate a textured real - scene model through UV mapping; Use 3D modeling software to establish a BIM construction line model according to the bridge road design drawing data, and the design drawing data includes tower, cable - stayed cable anchor point position, cable length, modal frequency parameters, and material information.

3. The method for analyzing the real - scene model of road and bridge based on BIM according to claim 2, wherein: S12. By setting the common control points of the real - scene model and the BIM construction line model, use the ICP algorithm to perform transformation registration on the common control point set of the dense point cloud of the real - scene model and the BIM construction line model; The common control points include the top of the bridge tower, the anchorage point, and the intersection of the bridge deck center line; The transformation registration includes rotation, scaling, and translation; Based on the obtained real-life BIM model, the point density threshold is used to eliminate the non-cable area and obtain the actual trajectory of the cable. The three-dimensional coordinate extraction algorithm is used to extract the discrete points P of the cable segment by segment starting from the anchor point according to the normal direction of the point cloud. j = (x, y, z), P j represents the jth discrete point, x represents the horizontal axis coordinate of the inclined cable, y represents the vertical axis coordinate of the inclined cable, and z represents the vertical axis coordinate of the inclined cable; S13. Set up an API application program interface to connect the real - scene BIM model with the meteorological database in real - time and extract wind force coupling data in real - time; The wind force coupling data includes the wind speed time - varying curve and the natural frequency; Obtain traffic coupling data by installing inductive loop detectors and video recognition on the bridge road; The traffic coupling data includes vehicle speed, vehicle load, and the vehicle passing frequency Fk of the k - th vehicle; Load the wind power coupling data and traffic coupling data into the real - scene BIM model for structural response fusion.

4. The method for analyzing the real - scene model of road and bridge based on BIM according to claim 3, wherein: The S2 includes S21 and S22; S21, Extract the cable - stayed cable feature set in real - time based on the time - based BIM model after structural response fusion; The cable-stayed cable feature set includes curvature k, wind-induced excitation angular frequency W wind , vehicle excitation frequency W veh and the number of exciting vehicles M; The curvature k is calculated and extracted by the three-point curvature method; assume that A, B, and C are three consecutive discrete points (P j-1 , P j , P j+1 ), and the specific algorithm formula is: , where k j represents the curvature of the j-th discrete point; The wind-induced excitation angular frequency W wind After extracting the wind speed V, the stay cable diameter D, and the Strouhal number St after the structural response fusion of the real-scene BIM model, and performing dimensionless processing, it is comprehensively calculated and extracted. The specific algorithm formula is: W wind（i） = 2π(St·V i / D i ), where W wind（i） represents the wind-induced excitation angular frequency of the i-th stay cable, π represents the circumference ratio, with a value of 3.14, V i represents the wind speed at the position of the i-th stay cable, and D i represents the diameter of the i-th stay cable; The vehicle excitation frequency W veh After dimensionless processing of the number of exciting vehicles M, vehicle speed v, and vehicle length cd after extracting the real - scene BIM model and fusing the structural response, the calculation and extraction are carried out. The specific algorithm formula is: W veh（c） = 2π(v c / cd c ), where W veh（c） represents the vehicle excitation frequency of the c - th vehicle, v c represents the passing vehicle speed of the c - th vehicle, and cd c represents the vehicle length of the c - th vehicle; S22, Pre - process the obtained cable - stayed cable feature set, and the pre - processing includes timestamp alignment and normalization; The timestamp alignment is carried out by adding timestamps to all data in the cable - stayed cable feature set and unifying them based on the time interval [t0, t1], where t0 represents the reference state time and t1 represents the comparison state time; The normalization is carried out by using the Min - Max normalization method to normalize all data in the cable - stayed cable feature set, eliminate the influence of the unit dimension of all data in the cable - stayed cable feature set, and obtain the normalized feature set.

5. The method for analyzing the real - scene model of road and bridge based on BIM according to claim 4, wherein: The S3 includes S31 and S32; S31, Based on the real - time obtained curvature k, calculate and analyze the curvature offset rate O of each cable - stayed cable from the reference state time t0 to the comparison state time t1, and analyze the tension balance of each cable - stayed cable under the time - varying state; The curvature offset rate O is calculated and output through the following algorithm formula; ； Where, O i represents the curvature deviation rate of the i-th stay cable, n represents the total number of discrete points, k ij (t1) represents the curvature of the j-th discrete point of the i-th stay cable at the comparison state time, k ij (t_{0}) represents the curvature of the j-th discrete point of the i-th stay cable at the reference state time, represents the initial average curvature of the i-th stay cable.

6. The method for analyzing the real - scene model of road and bridge based on BIM according to claim 5, wherein: S32. According to the abnormal offset of the stay cable, the user sets a preset offset threshold Oth, and compares the curvature offset rate O of the ith stay cable obtained in real time with the offset threshold Oth for preliminary comparative evaluation, preliminarily judge the tension state of the stay cable, and based on the evaluation result, trigger coupling analysis. The specific evaluation content is as follows; i And based on the evaluation results, trigger coupling analysis. The specific evaluation content is as follows; When the curvature deviation rate O of the i-th stay cable i < the deviation threshold Oth, it indicates that the stay cable structure is stable; When the curvature deviation rate O of the i-th stay cable i ≥ the deviation threshold Oth, it is preliminarily determined that the stay cable tension is unbalanced, and at this time, the coupling analysis is triggered.

7. The method for analyzing the real - scene model of road and bridge based on BIM according to claim 6, wherein: The S4 includes S41; S41. After initially comparing and evaluating to determine the tension imbalance of the current stay cable, coupling analysis is performed. The coupling analysis extracts the wind-induced excitation angular frequency W in the standardized feature set wind , the vehicle excitation frequency W veh and the number of exciting vehicles M, and performs the energy response intensity of the coupling wind and traffic on the current stay cable per unit time to obtain the comprehensive dynamic response intensity Rdyn, and analyzes the total structural response generated by the superposition of wind force and traffic within a certain period of time; The comprehensive dynamic response intensity Rdyn is calculated and output through the following algorithm formula; ； where, Rdyn i represents the comprehensive dynamic response intensity of the i-th stay cable, a i represents the wind-induced excitation intensity coefficient of the i-th stay cable, sin represents the sine function, represents the vehicle excitation intensity of the c-th vehicle, represents the phase perturbation term of the c-th vehicle, X i (t) represents the modal filtering window function of the i-th stay cable at time t, dt represents the time calculus variable.

8. A method for analyzing a real - scene model of a road and bridge based on BIM according to claim 6, characterized in that: The S5 includes S51 and S52; S51, Based on the comprehensive dynamic response intensity Rdyn and the curvature offset rate O of the current cable - stayed cable, comprehensively calculate and output the cable - stayed cable fatigue risk factor S, fuse the tension non - uniformity and dynamic response, and analyze the fatigue risk of the cable - stayed cable; The cable - stayed cable fatigue risk factor S is calculated and output through the following algorithm formula; ； Where S i represents the cable fatigue analysis factor of the i-th stay cable, and log represents the logarithmic function.

9. The method for analyzing the real - scene model of road and bridge based on BIM according to claim 8, wherein: S52. Set the first risk threshold F1 and the second risk threshold F2 based on the cable fatigue risk factor S of the historical stay cable under normal conditions, where the first risk threshold F1 represents the lower limit value of the stay cable fatigue, and the second risk threshold F2 represents the upper limit value of the stay cable fatigue. When the cable fatigue analysis factor S of the i-th stay cable obtained in real time i is compared and evaluated twice with the first risk threshold F1 and the second risk threshold F2 to judge the fatigue condition of the stay cable, and the risk level is divided based on the results of the secondary comparison and evaluation. The specific evaluation content is as follows; When the cable-stayed cable fatigue analysis factor S of the i-th cable-stayed cable i < the first risk threshold F1, it indicates that there is a preliminary risk for the current i-th cable-stayed cable. At this time, the current cable-stayed cable is classified as a first-level risk, and a reminder for regular re-modeling is issued; When the first risk threshold F1 ≤ the cable fatigue analysis factor S of the i-th stay cable i <When the second risk threshold F2, it indicates that the current stay cable is abnormal. At this time, the current stay cable is classified as a secondary risk, marked, and added to the operation and maintenance list for the next cycle; When the stay cable fatigue analysis factor S of the i-th stay cable i ≥ the second risk threshold F2, it indicates that the current stay cable is in danger. At this time, the current stay cable is classified as a third-level risk, an immediate warning is issued, the current stay cable number is output, and it is prompted to perform maintenance immediately.

10. A BIM-based road and bridge real-scene model analysis system, which is applied to a BIM-based road and bridge real-scene model analysis method according to any one of claims 1-9, and is characterized in that: It includes a model fusion module, a multi - dimensional data extraction module, a cable - stayed cable curvature analysis module, a windmill coupling module, and a fatigue analysis module; The model fusion module initially fuses the real - scene model of the bridge road and the BIM construction line model to obtain the real - scene BIM model, and loads the traffic coupling data and wind power coupling data into the real - scene BIM model; The multi - dimensional data extraction module extracts the cable - stayed cable feature set in real - time through the real - scene BIM model, and pre - processes the cable - stayed cable feature set to obtain the normalized feature set; The cable - stayed cable curvature analysis module calculates the curvature offset rate O of each cable - stayed cable based on the normalized feature set, sets the offset threshold Oth and makes a preliminary comparison and evaluation with the curvature offset rate O to judge the tension state of each current cable - stayed cable; The windmill coupling module performs coupling analysis and calculates and outputs the comprehensive dynamic response intensity Rdyn after judging the cable - stayed cable tension imbalance through the preliminary comparison and evaluation; The fatigue analysis module comprehensively calculates and outputs the cable - stayed cable fatigue risk factor S based on the obtained comprehensive dynamic response intensity Rdyn and the curvature offset rate O, sets the first risk threshold F1 and the second risk threshold F2 to make a secondary comparison and evaluation with the cable - stayed cable fatigue risk factor S, and divides the risk level based on the result of the secondary comparison and evaluation.

Images