Bridge abnormity monitoring system and method based on three-dimensional model

Abstract

The invention discloses a bridge abnormity monitoring system and method based on a three-dimensional model, and relates to the technical field of bridge monitoring, and the method comprises the steps: synchronously collecting structure data and environment data through underwater detection equipment, calculating a detection precision correction coefficient through an environment adaptive algorithm, and carrying out the calibration of an initial three-dimensional model; fusing the structural design parameters, the historical defect data and the real-time environment data, calculating a dynamic health degree threshold value, identifying abnormal parameters and delimiting a target monitoring area; setting a monitoring priority and a sampling frequency according to the abnormal degree, extracting abnormal features and calculating an abnormal risk index; triggering graded early warning according to the risk index until the risk is removed; according to the method, multi-source data fusion and dynamic threshold adjustment are realized, and the accuracy of underwater structure anomaly recognition and early warning timeliness are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of bridge monitoring technology, specifically to a bridge anomaly monitoring system and method based on a three-dimensional model. Background Technology

[0002] As a crucial component of transportation infrastructure, the long-term safety and durability of the underwater structure of bridges directly impacts the overall operational safety of the bridge. Because underwater structures are constantly exposed to a concealed environment, they are susceptible to various defects and damage due to factors such as water erosion, corrosion, and biological adhesion. Therefore, effective underwater inspection is of paramount importance. Currently, my country's underwater bridge structure inspection technology primarily relies on a variety of methods, including manual divers' exploration, underwater video observation, robotic detection, 3D laser scanning, and sonar technology.

[0003] However, the effectiveness of existing detection technologies is highly dependent on water visibility. In turbid waters, visible light attenuates drastically, resulting in blurry images acquired by underwater video detection methods, rendering them unusable for anomaly identification. While sonar technology does not rely on optical conditions, its resolution is far lower than that of optical images, making it difficult to clearly identify minute defects such as cracks and spalling. Although 3D laser scanning offers millimeter-level accuracy, its effective range is limited to only 1 to 5 meters, similarly restricted by turbid water, significantly reducing its practicality. Therefore, the "visual blind spot" caused by water turbidity severely restricts the accurate identification of anomalies on underwater structural surfaces using existing technologies. Summary of the Invention

[0004] The purpose of this invention is to provide a bridge anomaly monitoring system and method based on a three-dimensional model to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a bridge anomaly monitoring method based on a three-dimensional model, the bridge anomaly monitoring method comprising: Step S100: Simultaneously collect structural data of the underwater bridge structure and environmental data of the surrounding environment through underwater inspection equipment, calculate the inspection accuracy correction coefficient through an environmental adaptive algorithm, substitute the inspection accuracy correction coefficient into the initial three-dimensional model of the underwater bridge structure, and calibrate the initial three-dimensional model. Step S200: Obtain the structural design parameters and historical defect data of the bridge, integrate and analyze the structural design parameters and historical defect data with real-time environmental data, calculate the dynamic health threshold, identify abnormal parameters in combination with the calibrated three-dimensional model, and capture the target monitoring area according to the distribution of abnormal parameters; Step S300: Based on the initial degree of abnormality of abnormal parameters within the target monitoring area, the monitoring priority is divided, the data sampling frequency is set according to the monitoring priority, and the target monitoring area is continuously collected according to the data sampling frequency; the abnormal features in the collected data are extracted through the calibrated three-dimensional model, and the abnormal risk index is calculated based on the abnormal features; Step S400: Trigger a graded early warning based on the abnormal risk index, and send an early warning notification through the communication system until the risk is eliminated.

[0006] Furthermore, step S100 includes: Step S101: Using a detection equipment group consisting of an underwater robot, underwater sonar detection equipment, and a 3D laser scanner, a comprehensive scan of the underwater structure of the bridge is performed, simultaneously collecting structural data and environmental data. The structural data includes the surface flatness, actual dimensional parameters, and preliminary information on surface defects of the underwater components. The environmental data includes water turbidity, water flow velocity, water temperature, and concentration of corrosive media in the water. The collected structural and environmental data are uploaded to the data processing center to form a structured database. In the above steps, surface smoothness refers to the smoothness or unevenness of the surface of the underwater bridge component, which is usually obtained by laser scanning or sonar measurement and is used to determine whether the surface is worn, eroded or deformed; actual size parameters refer to the specific dimensions of the component in reality (such as length, width and thickness); preliminary information on surface defects refers to surface anomalies identified through preliminary scanning, such as cracks, peeling, holes, etc., which are the basis for subsequent detailed analysis; Step S102: Calculate the detection accuracy correction coefficient using an environment adaptive algorithm; the detection accuracy correction coefficient is obtained by weighting and summing the correction coefficients of each dimension with their corresponding influence weights based on preset turbidity correction coefficient, position offset correction coefficient, and surface detection correction coefficient. Step S103: Adjust the spatial coordinates and component size parameters of the initial three-dimensional model according to the calculated detection accuracy correction coefficient to complete the calibration of the three-dimensional model.

[0007] Furthermore, step S102 includes: Historical water turbidity data is acquired and divided into several turbidity intervals. A corresponding equipment accuracy weight is assigned to each turbidity interval. This equipment accuracy weight includes the weight of the 3D laser scanner and the weight of the underwater sonar detection equipment. The current measured turbidity value of the water body is acquired, the turbidity interval to which the measured turbidity value belongs is determined, and the corresponding equipment accuracy weight for each turbidity interval is obtained. The measured accuracy of the 3D laser scanner and the underwater sonar detection equipment under the measured turbidity value of the water body is obtained respectively. The measured accuracy of each device is multiplied by its corresponding equipment accuracy weight to obtain the weighted accuracy of each device. The sum of the weighted accuracies of all devices is the turbidity correction coefficient. In the above steps, the detection accuracy of the 3D laser scanner depends on the light penetration. The higher the turbidity of the water, the more obvious the light attenuation and the lower the detection resolution. On the other hand, underwater sonar detection equipment acquires data through sound wave reflection and has a stronger ability to resist turbidity interference. Therefore, the detection weight of the two devices needs to be allocated according to the turbidity level. Historical water flow velocity data is extracted and divided into several velocity intervals according to the degree of influence of water flow velocity on the underwater robot's attitude. A corresponding sampling frequency is preset for each velocity interval. The current measured water flow velocity value is obtained, and the velocity interval to which the measured water flow velocity value belongs is determined. Position offset data is collected at the preset sampling frequency. Based on the characteristic dimensions of the underwater bridge structure, the ratio of the position offset to the characteristic dimensions is calculated, and the position offset correction coefficient is obtained by subtracting this ratio from 1. In the above steps, the water flow velocity affects the underwater robot's attitude stability. The faster the water flow, the easier it is for the robot to deviate from the preset detection path, resulting in a deviation between the collected structural data and the actual structural position. Therefore, it is necessary to quantify the impact of this deviation on detection accuracy by calculating a position offset correction coefficient. Position offset data refers to the deviation between the collected structural data and the actual structural position caused by the water flow velocity affecting the underwater robot's attitude. The detection error is quantified by measuring the offset of the robot's actual position from the preset path. The characteristic dimensions of the underwater bridge structure refer to representative dimensions of the underwater bridge components (such as pier diameter and abutment width), which are used as a benchmark to calculate the position offset ratio. For example, the smaller the ratio of the offset to the characteristic dimension, the smaller the impact of the deviation on the detection. Historical water temperature data is extracted and averaged. The absolute value of the difference between the reference temperature and the average value is calculated to obtain the temperature deviation degree. The temperature deviation degree is divided into several continuous intervals according to a preset temperature deviation threshold. A corresponding temperature correction ratio is preset for each interval to form a temperature correction interval. The current measured water temperature is obtained, and the temperature deviation degree between the measured water temperature and the reference temperature is calculated. The temperature correction interval is determined, and the corresponding temperature correction ratio is obtained. The temperature correction ratio and the temperature deviation degree are input into the temperature correction coefficient calculation formula to obtain the temperature correction coefficient. In the above steps, the formula for calculating the temperature correction factor is: ; Where T is the temperature correction factor and k is the temperature correction ratio. Temperature deviation in degrees; In the above steps, the reference temperature refers to the water temperature or standard ambient temperature expected during bridge design, used to compare with the actual temperature; when the actual temperature deviates from the reference temperature, the components may undergo dimensional changes due to thermal expansion and contraction, requiring adjustment using a correction factor; the current measured water temperature is obtained by directly measuring the current water temperature value using sensors. Extract historical water corrosive medium concentration data and obtain the average value. Calculate the absolute value of the difference between the average value and the baseline tolerance concentration, and calculate the ratio of the absolute value to the baseline tolerance concentration to obtain the deviation concentration. Divide the deviation concentration into several continuous intervals according to a preset concentration threshold, and preset a corresponding corrosion correction ratio for each interval to form a concentration correction interval. Obtain the current measured concentration of the water corrosive medium, calculate the deviation concentration between the measured concentration of the water corrosive medium and the baseline tolerance concentration, determine the corresponding concentration correction interval, and obtain the corresponding corrosion correction ratio. Input the corrosion correction ratio into the corrosion correction coefficient calculation formula to obtain the corrosion correction coefficient. In the above steps, the formula for calculating the corrosion correction factor is: ; Where F is the corrosion correction coefficient and m is the corrosion correction ratio; In the above steps, the reference tolerance concentration refers to the maximum safe concentration of corrosive media (such as chloride ions and sulfates) in water that the component materials can withstand, as specified in the bridge design. Exceeding this concentration may accelerate corrosion, and the defect measurement value needs to be adjusted by a correction factor. The current measured concentration of corrosive media in the water is obtained through chemical sensors or sampling analysis to obtain the real-time concentration value of corrosive media in the water. Multiply the temperature correction factor by the corrosion correction factor to obtain the surface inspection correction factor; In the above steps, when the water temperature deviates from the bridge's design reference temperature, the underwater components will undergo dimensional changes due to thermal expansion and contraction. This will cause the deviation between the measured dimensions and the design dimensions to be due to structural defects in the components themselves, and will need to be adjusted using a temperature correction factor. When the concentration of corrosive media in the water exceeds the design tolerance concentration, a hidden corrosion layer may form on the surface of the components (such as a loose concrete surface or rust on the steel surface). This will cause the measured values ​​of surface defects (such as corrosion depth and crack width) to be smaller than the actual values, and will need to be adjusted using a corrosion correction factor. In the above steps, since water temperature and corrosive medium concentration both directly affect the surface of the component, and their effects on the test results are independent and superimposed (for example, the component has dimensional deviations due to temperature deviations and surface defect measurement deviations due to excessive corrosive medium concentrations), it is necessary to multiply the two individual correction coefficients to obtain a surface test correction coefficient that comprehensively reflects the influence of the two factors. This coefficient can be directly used to calibrate the surface dimensions and defect parameters of the component in the subsequent three-dimensional model to ensure that the surface condition of the model is consistent with the actual structure.

[0008] Furthermore, step S200 includes: Step S201: Retrieve structural design parameters from the bridge design archive, including the standard value of material strength, design dimensions, design value of bearing capacity, and design grade of corrosion resistance for underwater components; extract historical defect data from the bridge inspection history database, including the location, extent, and development trend of defects recorded in past inspections; In the above steps, the bridge design archive refers to the database or files storing the original design data of the bridge, including construction drawings, calculation sheets, etc.; structural design parameters refer to key parameters extracted from the design archive, such as standard values ​​of material strength (compressive and tensile strength of component materials), design dimensions, design values ​​of bearing capacity (maximum allowable load), and corrosion resistance design level (corrosion resistance performance of materials); these parameters are the basis for setting the foundation health threshold; the bridge inspection history database refers to the database storing bridge inspection records over the years, including inspection time, location, defect type, etc.; defect development trend refers to the changes of defects over time, such as the average annual expansion rate of cracks and the annual growth rate of corrosion area; by analyzing the development trend, the future defect status can be predicted; Step S202: Integrate and analyze the structural design parameters, historical defect data, and collected real-time environmental data to calculate the dynamic health threshold; specifically: Based on the safety standards of components in the structural design parameters, a basic health threshold is set for each detection index. The data is divided into intervals based on the average annual expansion rate of defects in historical defect data. Each interval is assigned a historical defect impact coefficient. The higher the expansion rate, the larger the impact coefficient. This is used to narrow the range of basic health thresholds. The environmental data includes the concentration of corrosive media in the water, water flow velocity, and water temperature. The corrosion correction coefficient, position offset correction coefficient, and temperature correction coefficient calculated in step S201 are obtained respectively. The product of each correction coefficient is taken as the total environmental impact coefficient. The greater the deviation, the greater the total environmental impact coefficient, which is used to further narrow the range of the basic health threshold. The dynamic health threshold is obtained by dividing the basic health threshold by the product of the historical defect impact coefficient and the total environmental impact coefficient; corresponding dynamic health thresholds are formed for different components and different detection indicators. Step S203: Using the calibrated 3D model as a reference, establish a comparison coordinate system and convert the real-time structural data collected in step S101 to the comparison coordinate system; for the detection points or detection units of the components in the model, extract the actual parameters from the real-time structural data of the corresponding detection points or detection units in the 3D model, and compare the actual parameters with the corresponding dynamic health threshold one by one; if the actual parameter exceeds the dynamic health threshold, mark the parameter as an abnormal parameter and record the location, parameter type, and parameter exceedance range of the abnormal parameter; In the above steps, a detection point refers to a specific location point in the 3D model (such as a coordinate point) used for focused detection; a detection unit refers to a region or component in the model (such as a mesh unit) used for batch detection. Step S204: Based on the distribution of abnormal parameters marked in step S203, divide the abnormal region morphology according to the abnormal parameter type. The abnormal region morphology includes point type, line type, and area type. For different morphological abnormal regions, set the corresponding extension range based on the location of the abnormal parameters to form the target monitoring area. Mark the target monitoring area in the calibrated 3D model, assign a unique number, and associate it with the corresponding abnormal parameter information. In the above steps, setting the corresponding expansion range refers to determining the expansion range of the monitoring area based on the type of abnormal parameters (point, line, area). For example, point defects may expand into a circular area, line defects into a strip area, and area defects into a polygonal area. This ensures that potential risk areas around the anomaly are also included in the monitoring.

[0009] Furthermore, step S300 includes: Step S301: Based on the magnitude of the abnormal parameters exceeding the dynamic health threshold in step S203, the monitoring priority is divided. The greater the magnitude of the exceedance, the higher the monitoring priority. For target monitoring areas with different priorities, the corresponding periodic collection frequency is set. The higher the priority, the higher the collection frequency. Before collection, the deviation between the current environmental data and the environmental data collected for the first time is checked. If the deviation exceeds the preset range, the detection equipment parameters are adjusted. After collection, the data is associated with the unique number of the corresponding target monitoring area. Step S302: Using the calibrated 3D model as a reference, establish a feature extraction network within the target monitoring area; map the data periodically collected in step S301 to the corresponding nodes of the feature extraction network, and extract abnormal features according to defect type. The abnormal features include corrosion defects, crack defects, and deformation defects. Corrosion defects include corrosion area, corrosion depth, and corrosion rate, with the corrosion rate calculated by comparing corrosion area or corrosion depth collected at different times. Crack defects include crack length, crack width, and crack propagation rate, with the crack propagation rate calculated by comparing crack length or crack width collected at different times. Deformation defects include deformation amount, deformation direction, and deformation rate, with the deformation rate calculated by comparing deformation amount collected at different times. Associate the extracted abnormal feature parameters with the collection time and the unique number of the target monitoring area to form a time-series feature database. In the above steps, the corrosion rate is equal to the current corrosion amount minus the previous corrosion amount divided by the time interval, used to quantify the rate of corrosion progress; the crack propagation rate is equal to the current crack size minus the previous crack size divided by the time interval, used to assess the stability of the crack; the deformation rate is equal to the current deformation amount minus the previous deformation amount divided by the time interval, where deformation amount includes displacement, angle change, etc., used to determine whether the component is undergoing continuous deformation. Step S303: Set the weights of the three assessment dimensions—defect severity, defect impact range, and defect development rate—in the risk assessment; for each dimension, divide the scoring interval according to the abnormal characteristic parameters, and determine the score corresponding to each scoring interval; wherein, the defect severity is divided into scoring intervals according to the extent to which the abnormal characteristic parameters exceed the dynamic health threshold, the larger the extent, the higher the score; the defect impact range is divided into scoring intervals according to the degree of influence of the defect location on the structural function of the component, the greater the influence, the higher the score; the defect development rate is divided into scoring intervals according to the rate of expansion or change of the defect, the faster the rate, the higher the score; obtain the scores of each dimension of the target monitoring area, multiply each dimension score by its corresponding weight, and sum them to obtain the abnormal risk index of the target monitoring area; record the abnormal risk index calculated each time to form a risk index change curve.

[0010] Furthermore, step S400 includes: Step S401: Based on the impact of the abnormal risk index on bridge safety, classify the early warning levels, set the risk index range corresponding to each early warning level, and set the response time limit and handling requirements for each early warning level. The higher the early warning level, the shorter the response time limit and the more urgent the handling requirements. In the above steps, the impact of the abnormal risk index on the safe operation of the bridge is assessed based on the magnitude of the abnormal risk index. For example, a risk index of 0.8 may indicate high risk, affecting the bridge's load-bearing capacity and requiring immediate action; a risk index of 0.3 may indicate low risk, requiring only periodic monitoring. The degree of impact is classified into warning levels using preset thresholds. Step S402: Compare the abnormal risk index calculated in step S303 with the preset warning level range to match the corresponding warning level; send warning prompts to relevant departments through the communication system, with different recipients and sending frequencies for different warning levels; wherein, the warning prompt content includes the target monitoring area number, abnormal characteristics, risk index, and corresponding handling requirements; Step S403: Continuously monitor the abnormal risk index of the target monitoring area at the frequency set in step S301, and set the conditions for lifting the warning; if the conditions for lifting the warning are met, send a warning lifting notification to the warning recipient through the communication system, mark the warning status as lifted in the system, and archive the relevant detection data and processing records.

[0011] Furthermore, to better implement the above method, a bridge anomaly monitoring system based on a three-dimensional model is also provided. This bridge anomaly monitoring system includes: a data acquisition module, an intelligent analysis module, a risk assessment module, and an early warning management module. The data acquisition module is used to collect structural and environmental data of the underwater bridge structure through underwater inspection equipment, and to calculate the inspection accuracy correction coefficient through an environmental adaptive algorithm to calibrate the initial three-dimensional model. The intelligent analysis module is used to integrate structural design parameters, historical defect data and real-time environmental data to calculate dynamic health thresholds, identify abnormal parameters and delineate target monitoring areas; The risk assessment module is used to periodically collect data on the target monitoring area, extract abnormal features, and calculate the abnormal risk index. The early warning management module is used to trigger tiered early warnings based on the abnormal risk index and track the early warning status until the risk is eliminated. The data acquisition module is electrically connected to the intelligent analysis module, the intelligent analysis module is electrically connected to the risk assessment module, and the risk assessment module is electrically connected to the early warning management module.

[0012] Furthermore, the data acquisition module includes: an acquisition unit, a correction unit, and a calibration unit; The data acquisition unit, through a detection equipment group consisting of an underwater robot, underwater sonar detection equipment and a 3D laser scanner, performs a comprehensive scan of the underwater structure of the bridge and simultaneously collects structural and environmental data. The correction unit is used to calculate the detection accuracy correction coefficient through an environmental adaptive algorithm. The detection accuracy correction coefficient is obtained by weighted summation of the turbidity correction coefficient, the position offset correction coefficient, the surface detection correction coefficient and their respective influence weights. The calibration unit is used to adjust the spatial coordinates and component size parameters of the initial three-dimensional model according to the calculated detection accuracy correction coefficient, thereby completing the calibration of the three-dimensional model.

[0013] Furthermore, the intelligent analysis module includes: a threshold calculation unit, a comparison unit, and a region delineation unit; The threshold calculation unit is used to retrieve structural design parameters and historical defect data from bridge design archives and historical inspection databases, and integrate them with real-time environmental data for analysis. The dynamic health threshold is calculated by dividing the basic health threshold by the product of the historical defect impact coefficient and the total environmental impact coefficient. The comparison unit is used to establish a comparison coordinate system based on the calibrated 3D model, convert the real-time acquired structural data to this coordinate system, compare the actual parameters with the corresponding dynamic health thresholds one by one, mark abnormal parameters and record the location, type and exceedance of abnormal parameters; The region delineation unit is used to divide the abnormal region into different shapes according to the distribution of the marked abnormal parameters and the parameter type, and to set the extension range based on the location of the abnormal parameters to form the target monitoring area. The target monitoring area is then marked and numbered in the calibrated 3D model.

[0014] Furthermore, the risk assessment module includes: a feature extraction unit, a risk calculation unit, and a time series library unit; The feature extraction unit is used to establish a feature extraction network within the target monitoring area, map the periodically collected data to grid nodes, and extract abnormal features according to the defect type. The abnormal features include feature parameters of corrosion defects, crack defects, and deformation defects. The risk calculation unit is used to set the weights of three assessment dimensions: defect severity, defect impact range, and defect development rate. It scores each dimension based on abnormal characteristic parameters, and then multiplies each dimension score by its corresponding weight and sums the results to obtain the abnormal risk index of the target monitoring area. The time series library unit is used to store the extracted abnormal feature parameters, their collection time, and the target monitoring area number to form a time series feature database, and to record the abnormal risk index calculated each time to form a risk index change curve.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. By dynamically correcting the detection accuracy through an environmental adaptive algorithm, the system overcomes environmental interference such as water turbidity, thereby improving the reliability of 3D model calibration and data acquisition.

[0016] 2. By integrating structural design, historical defects, and environmental data, a dynamic health threshold is constructed to achieve accurate identification and hierarchical monitoring of abnormal areas.

[0017] 3. Trigger tiered early warnings based on abnormal risk indices to achieve closed-loop management from data collection to risk management, thereby enhancing the intelligence and emergency response capabilities of underwater bridge structure monitoring. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the method flow of the bridge anomaly monitoring system and method based on a three-dimensional model according to the present invention; Figure 2 This is a schematic diagram of the system structure of the bridge anomaly monitoring system and method based on a three-dimensional model according to the present invention. Detailed Implementation

[0019] 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.

[0020] Example 1: As Figure 1 As shown, the present invention provides a technical solution, a bridge anomaly monitoring method based on a three-dimensional model, the bridge anomaly monitoring method comprising: Step S100: Simultaneously collect structural data of the underwater bridge structure and environmental data of the surrounding environment through underwater inspection equipment, calculate the inspection accuracy correction coefficient through an environmental adaptive algorithm, substitute the inspection accuracy correction coefficient into the initial three-dimensional model of the underwater bridge structure, and calibrate the initial three-dimensional model. Step S200: Obtain the structural design parameters and historical defect data of the bridge, integrate and analyze the structural design parameters and historical defect data with real-time environmental data, calculate the dynamic health threshold, identify abnormal parameters in combination with the calibrated three-dimensional model, and capture the target monitoring area according to the distribution of abnormal parameters; Step S300: Based on the initial degree of abnormality of abnormal parameters within the target monitoring area, the monitoring priority is divided, the data sampling frequency is set according to the monitoring priority, and the target monitoring area is continuously collected according to the data sampling frequency; the abnormal features in the collected data are extracted through the calibrated three-dimensional model, and the abnormal risk index is calculated based on the abnormal features; Step S400: Trigger a graded early warning based on the abnormal risk index, and send an early warning notification through the communication system until the risk is eliminated; Step S100 includes: Step S101: Using a detection equipment group consisting of an underwater robot, underwater sonar detection equipment, and a 3D laser scanner, a comprehensive scan of the underwater structure of the bridge is performed, simultaneously collecting structural data and environmental data. The structural data includes the surface flatness, actual dimensional parameters, and preliminary information on surface defects of the underwater components. The environmental data includes water turbidity, water flow velocity, water temperature, and concentration of corrosive media in the water. The collected structural and environmental data are uploaded to the data processing center to form a structured database. Step S102: Calculate the detection accuracy correction coefficient using an environment adaptive algorithm; the detection accuracy correction coefficient is obtained by weighting and summing the correction coefficients of each dimension with their corresponding influence weights based on preset turbidity correction coefficient, position offset correction coefficient, and surface detection correction coefficient. Step S103: Adjust the spatial coordinates and component size parameters of the initial three-dimensional model according to the calculated detection accuracy correction coefficient to complete the calibration of the three-dimensional model; Step S102 includes: Historical water turbidity data is acquired and divided into several turbidity intervals. A corresponding equipment accuracy weight is assigned to each turbidity interval. This equipment accuracy weight includes the weight of the 3D laser scanner and the weight of the underwater sonar detection equipment. The current measured turbidity value of the water body is acquired, the turbidity interval to which the measured turbidity value belongs is determined, and the corresponding equipment accuracy weight for each turbidity interval is obtained. The measured accuracy of the 3D laser scanner and the underwater sonar detection equipment under the measured turbidity value of the water body is obtained respectively. The measured accuracy of each device is multiplied by its corresponding equipment accuracy weight to obtain the weighted accuracy of each device. The sum of the weighted accuracies of all devices is the turbidity correction coefficient. Historical water flow velocity data is extracted and divided into several velocity intervals according to the degree of influence of water flow velocity on the underwater robot's attitude. A corresponding sampling frequency is preset for each velocity interval. The current measured water flow velocity value is obtained, and the velocity interval to which the measured water flow velocity value belongs is determined. Position offset data is collected at the preset sampling frequency. Based on the characteristic dimensions of the underwater bridge structure, the ratio of the position offset to the characteristic dimensions is calculated, and the position offset correction coefficient is obtained by subtracting this ratio from 1. Historical water temperature data is extracted and averaged. The absolute value of the difference between the reference temperature and the average value is calculated to obtain the temperature deviation degree. The temperature deviation degree is divided into several continuous intervals according to a preset temperature deviation threshold. A corresponding temperature correction ratio is preset for each interval to form a temperature correction interval. The current measured water temperature is obtained, and the temperature deviation degree between the measured water temperature and the reference temperature is calculated. The temperature correction interval is determined, and the corresponding temperature correction ratio is obtained. The temperature correction ratio and the temperature deviation degree are input into the temperature correction coefficient calculation formula to obtain the temperature correction coefficient. Extract historical water corrosive medium concentration data and obtain the average value. Calculate the absolute value of the difference between the average value and the baseline tolerance concentration, and calculate the ratio of the absolute value to the baseline tolerance concentration to obtain the deviation concentration. Divide the deviation concentration into several continuous intervals according to a preset concentration threshold, and preset a corresponding corrosion correction ratio for each interval to form a concentration correction interval. Obtain the current measured concentration of the water corrosive medium, calculate the deviation concentration between the measured concentration of the water corrosive medium and the baseline tolerance concentration, determine the corresponding concentration correction interval, and obtain the corresponding corrosion correction ratio. Input the corrosion correction ratio into the corrosion correction coefficient calculation formula to obtain the corrosion correction coefficient. Multiply the temperature correction factor by the corrosion correction factor to obtain the surface inspection correction factor; Step S200 includes: Step S201: Retrieve structural design parameters from the bridge design archive, including the standard value of material strength, design dimensions, design value of bearing capacity, and design grade of corrosion resistance for underwater components; extract historical defect data from the bridge inspection history database, including the location, extent, and development trend of defects recorded in past inspections; Step S202: Integrate and analyze the structural design parameters, historical defect data, and collected real-time environmental data to calculate the dynamic health threshold; specifically: Based on the safety standards of components in the structural design parameters, a basic health threshold is set for each detection index. The system divides the data into intervals based on the average annual expansion rate of defects in historical defect data. Each interval corresponds to a historical defect impact coefficient; the higher the expansion rate, the larger the impact coefficient. This is used to narrow the range of the basic health threshold. The environmental data includes the concentration of corrosive media in the water, water flow velocity, and water temperature. The corrosion correction coefficient, position offset correction coefficient, and temperature correction coefficient calculated in step S201 are obtained respectively. The product of each correction coefficient is taken as the total environmental impact coefficient. The greater the deviation, the greater the total environmental impact coefficient, which is used to further narrow the range of the basic health threshold. The dynamic health threshold is obtained by dividing the basic health threshold by the product of the historical defect impact coefficient and the total environmental impact coefficient; corresponding dynamic health thresholds are formed for different components and different detection indicators. Step S203: Using the calibrated 3D model as a reference, establish a comparison coordinate system and convert the real-time structural data collected in step S101 to the comparison coordinate system; for the detection points or detection units of the components in the model, extract the actual parameters from the real-time structural data of the corresponding detection points or detection units in the 3D model, and compare the actual parameters with the corresponding dynamic health threshold one by one; if the actual parameter exceeds the dynamic health threshold, mark the parameter as an abnormal parameter and record the location, parameter type, and parameter exceedance range of the abnormal parameter; Step S204: Based on the distribution of abnormal parameters marked in step S203, divide the abnormal region morphology according to the abnormal parameter type. The abnormal region morphology includes point type, line type, and area type. For different morphological abnormal regions, set the corresponding extension range based on the location of the abnormal parameters to form the target monitoring area. Mark the target monitoring area in the calibrated 3D model, assign a unique number, and associate it with the corresponding abnormal parameter information. Step S300 includes: Step S301: Based on the magnitude of the abnormal parameters exceeding the dynamic health threshold in step S203, the monitoring priority is divided. The greater the magnitude of the exceedance, the higher the monitoring priority. For target monitoring areas with different priorities, the corresponding periodic collection frequency is set. The higher the priority, the higher the collection frequency. Before collection, the deviation between the current environmental data and the environmental data collected for the first time is checked. If the deviation exceeds the preset range, the detection equipment parameters are adjusted. After collection, the data is associated with the unique number of the corresponding target monitoring area. Step S302: Using the calibrated 3D model as a reference, establish a feature extraction network within the target monitoring area; map the data periodically collected in step S301 to the corresponding nodes of the feature extraction network, and extract abnormal features according to defect type. The abnormal features include corrosion defects, crack defects, and deformation defects. Corrosion defects include corrosion area, corrosion depth, and corrosion rate, with the corrosion rate calculated by comparing corrosion area or corrosion depth collected at different times. Crack defects include crack length, crack width, and crack propagation rate, with the crack propagation rate calculated by comparing crack length or crack width collected at different times. Deformation defects include deformation amount, deformation direction, and deformation rate, with the deformation rate calculated by comparing deformation amount collected at different times. Associate the extracted abnormal feature parameters with the collection time and the unique number of the target monitoring area to form a time-series feature database. Step S303: Set the weights of the three assessment dimensions—defect severity, defect impact range, and defect development rate—in the risk assessment; for each dimension, divide the scoring intervals according to the abnormal characteristic parameters, and determine the score corresponding to each scoring interval; wherein, the defect severity is divided into scoring intervals according to the extent to which the abnormal characteristic parameters exceed the dynamic health threshold, the larger the extent, the higher the score; the defect impact range is divided into scoring intervals according to the degree of influence of the defect location on the structural function of the component, the greater the influence, the higher the score; the defect development rate is divided into scoring intervals according to the rate of expansion or change of the defect, the faster the rate, the higher the score; obtain the scores of each dimension of the target monitoring area, multiply each dimension score by its corresponding weight, and sum them to obtain the abnormal risk index of the target monitoring area; record the abnormal risk index calculated each time to form a risk index change curve; Step S400 includes: Step S401: Based on the impact of the abnormal risk index on bridge safety, classify the early warning levels, set the risk index range corresponding to each early warning level, and set the response time limit and handling requirements for each early warning level. The higher the early warning level, the shorter the response time limit and the more urgent the handling requirements. Step S402: Compare the abnormal risk index calculated in step S303 with the preset warning level range to match the corresponding warning level; send warning prompts to relevant departments through the communication system, with different recipients and sending frequencies for different warning levels; wherein, the warning prompt content includes the target monitoring area number, abnormal characteristics, risk index, and corresponding handling requirements; Step S403: Continuously monitor the abnormal risk index of the target monitoring area at the frequency set in step S301, and set the conditions for lifting the warning; if the conditions for lifting the warning are met, send a warning lifting notification to the warning recipient through the communication system, mark the warning status as lifted in the system, and archive the relevant detection data and processing records. Example 2: Figure 2 As shown, in order to better implement the above method, a bridge anomaly monitoring system based on a three-dimensional model is also provided. The bridge anomaly monitoring system includes: a data acquisition module, an intelligent analysis module, a risk assessment module, and an early warning management module. The data acquisition module is used to collect structural and environmental data of the underwater bridge structure through underwater inspection equipment, and to calculate the inspection accuracy correction coefficient through an environmental adaptive algorithm to calibrate the initial three-dimensional model. The intelligent analysis module is used to integrate structural design parameters, historical defect data and real-time environmental data to calculate dynamic health thresholds, identify abnormal parameters and delineate target monitoring areas; The risk assessment module is used to periodically collect data on the target monitoring area, extract abnormal features, and calculate the abnormal risk index. The early warning management module is used to trigger tiered early warnings based on the abnormal risk index and track the early warning status until the risk is eliminated. The data acquisition module is electrically connected to the intelligent analysis module, the intelligent analysis module is electrically connected to the risk assessment module, and the risk assessment module is electrically connected to the early warning management module. The data acquisition module includes: an acquisition unit, a correction unit, and a calibration unit; The data acquisition unit, through a detection equipment group consisting of an underwater robot, underwater sonar detection equipment and a 3D laser scanner, performs a comprehensive scan of the underwater structure of the bridge and simultaneously collects structural and environmental data. The correction unit is used to calculate the detection accuracy correction coefficient through an environmental adaptive algorithm. The detection accuracy correction coefficient is obtained by weighted summation of the turbidity correction coefficient, the position offset correction coefficient, the surface detection correction coefficient and their respective influence weights. The calibration unit is used to adjust the spatial coordinates and component size parameters of the initial three-dimensional model according to the calculated detection accuracy correction coefficient, thereby completing the calibration of the three-dimensional model. The intelligent analysis module includes: a threshold calculation unit, a comparison unit, and a region delineation unit; The threshold calculation unit is used to retrieve structural design parameters and historical defect data from bridge design archives and historical inspection databases, and integrate them with real-time environmental data for analysis. The dynamic health threshold is calculated by dividing the basic health threshold by the product of the historical defect impact coefficient and the total environmental impact coefficient. The comparison unit is used to establish a comparison coordinate system based on the calibrated 3D model, convert the real-time acquired structural data to this coordinate system, compare the actual parameters with the corresponding dynamic health thresholds one by one, mark abnormal parameters and record the location, type and exceedance of abnormal parameters; The area delineation unit is used to divide the abnormal area shape according to the parameter type based on the distribution of the marked abnormal parameters, and to set the extension range based on the location of the abnormal parameters to form the target monitoring area. The target monitoring area is marked and numbered in the calibrated 3D model. The risk assessment module includes: a feature extraction unit, a risk calculation unit, and a time series library unit. The feature extraction unit is used to establish a feature extraction network within the target monitoring area, map the periodically collected data to grid nodes, and extract abnormal features according to the defect type. The abnormal features include feature parameters of corrosion defects, crack defects, and deformation defects. The risk calculation unit is used to set the weights of three assessment dimensions: defect severity, defect impact range, and defect development rate. It scores each dimension based on abnormal characteristic parameters, and then multiplies each dimension score by its corresponding weight and sums the results to obtain the abnormal risk index of the target monitoring area. The time series library unit is used to store the extracted abnormal feature parameters and their collection time and target monitoring area number to form a time series feature database, and to record the abnormal risk index calculated each time to form a risk index change curve; The early warning management module includes: a level matching unit, a communication sending unit, and a status tracking unit; The level matching unit is used to match the abnormal risk index with the preset warning level range to determine the warning level and the corresponding response requirements; The communication transmission unit is used to send early warning information containing the target area number, abnormal characteristics, and handling requirements to relevant departments through the communication system; The status tracking unit is used to continuously monitor the risk index, determine whether the conditions for lifting the warning are met, and send a warning lifting notification to complete the closed-loop management of the warning. In an embodiment of the present invention, in a cross-river bridge underwater structure monitoring project, this system was used to periodically inspect key components such as piers and abutments. An underwater robot equipped with a 3D laser scanner simultaneously collected surface data and water environment data, and an environmental adaptive algorithm was used to calibrate the initial 3D model. The system integrated design parameters and historical defect data to calculate a dynamic health threshold, identifying three abnormal areas and setting them as high-priority monitoring target areas. Through continuous data collection and feature extraction of the monitoring target areas, the system discovered that one crack was expanding at an accelerated rate, with an abnormal risk index reaching 0.78, triggering a level-two warning. Upon receiving the report, the maintenance department immediately carried out reinforcement treatment, and after the system confirmed that the risk had been eliminated, the treatment record was archived, effectively preventing further structural deterioration.

[0021] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A bridge anomaly monitoring method based on a three-dimensional model, characterized in that: The bridge anomaly monitoring method includes: Step S100: Simultaneously collect structural data of the underwater bridge structure and environmental data of the surrounding environment through underwater inspection equipment, calculate the inspection accuracy correction coefficient through an environmental adaptive algorithm, substitute the inspection accuracy correction coefficient into the initial three-dimensional model of the underwater bridge structure, and calibrate the initial three-dimensional model. Step S200: Obtain the structural design parameters and historical defect data of the bridge, integrate and analyze the structural design parameters and historical defect data with real-time environmental data, calculate the dynamic health threshold, identify abnormal parameters in combination with the calibrated three-dimensional model, and capture the target monitoring area according to the distribution of abnormal parameters; Step S300: Based on the initial degree of abnormality of abnormal parameters within the target monitoring area, the monitoring priority is divided, the data sampling frequency is set according to the monitoring priority, and the target monitoring area is continuously collected according to the data sampling frequency; the abnormal features in the collected data are extracted through the calibrated three-dimensional model, and the abnormal risk index is calculated based on the abnormal features; Step S400: Trigger a graded early warning based on the abnormal risk index, and send an early warning notification through the communication system until the risk is eliminated.

2. The bridge anomaly monitoring method based on a three-dimensional model according to claim 1, characterized in that: Step S100 includes: Step S101: Using a detection equipment group consisting of an underwater robot, underwater sonar detection equipment, and a 3D laser scanner, a comprehensive scan of the underwater structure of the bridge is performed, simultaneously collecting structural data and environmental data. The structural data includes the surface flatness, actual dimensional parameters, and preliminary information on surface defects of the underwater components. The environmental data includes water turbidity, water flow velocity, water temperature, and concentration of corrosive media in the water. The collected structural and environmental data are uploaded to the data processing center to form a structured database. Step S102: Calculate the detection accuracy correction coefficient using an environment adaptive algorithm; the detection accuracy correction coefficient is obtained by weighting and summing the correction coefficients of each dimension with their corresponding influence weights based on preset turbidity correction coefficient, position offset correction coefficient, and surface detection correction coefficient. Step S103: Adjust the spatial coordinates and component size parameters of the initial three-dimensional model according to the calculated detection accuracy correction coefficient to complete the calibration of the three-dimensional model.

3. The bridge anomaly monitoring method based on a three-dimensional model according to claim 2, characterized in that: Step S102 includes: Historical water turbidity data is acquired and divided into several turbidity intervals. A corresponding equipment accuracy weight is assigned to each turbidity interval. This equipment accuracy weight includes the weight of the 3D laser scanner and the weight of the underwater sonar detection equipment. The current measured turbidity value of the water body is acquired, the turbidity interval to which the measured turbidity value belongs is determined, and the corresponding equipment accuracy weight for each turbidity interval is obtained. The measured accuracy of the 3D laser scanner and the underwater sonar detection equipment under the measured turbidity value of the water body is obtained respectively. The measured accuracy of each device is multiplied by its corresponding equipment accuracy weight to obtain the weighted accuracy of each device. The sum of the weighted accuracies of all devices is the turbidity correction coefficient. Historical water flow velocity data is extracted and divided into several velocity intervals according to the degree of influence of water flow velocity on the underwater robot's attitude. A corresponding sampling frequency is preset for each velocity interval. The current measured water flow velocity value is obtained, and the velocity interval to which the measured water flow velocity value belongs is determined. Position offset data is collected at the preset sampling frequency. Based on the characteristic dimensions of the underwater bridge structure, the ratio of the position offset to the characteristic dimensions is calculated, and the position offset correction coefficient is obtained by subtracting this ratio from 1. Historical water temperature data is extracted and averaged. The absolute value of the difference between the reference temperature and the average value is calculated to obtain the temperature deviation degree. The temperature deviation degree is divided into several continuous intervals according to a preset temperature deviation threshold. A corresponding temperature correction ratio is preset for each interval to form a temperature correction interval. The current measured water temperature is obtained, and the temperature deviation degree between the measured water temperature and the reference temperature is calculated. The temperature correction interval is determined, and the corresponding temperature correction ratio is obtained. The temperature correction ratio and the temperature deviation degree are input into the temperature correction coefficient calculation formula to obtain the temperature correction coefficient. Extract historical water corrosive medium concentration data and obtain the average value. Calculate the absolute value of the difference between the average value and the baseline tolerance concentration, and calculate the ratio of the absolute value to the baseline tolerance concentration to obtain the deviation concentration. Divide the deviation concentration into several continuous intervals according to a preset concentration threshold, and preset a corresponding corrosion correction ratio for each interval to form a concentration correction interval. Obtain the current measured concentration of the water corrosive medium, calculate the deviation concentration between the measured concentration of the water corrosive medium and the baseline tolerance concentration, determine the corresponding concentration correction interval, and obtain the corresponding corrosion correction ratio. Input the corrosion correction ratio into the corrosion correction coefficient calculation formula to obtain the corrosion correction coefficient. Multiply the temperature correction factor by the corrosion correction factor to obtain the surface inspection correction factor.

4. The bridge anomaly monitoring method based on a three-dimensional model according to claim 3, characterized in that: Step S200 includes: Step S201: Retrieve structural design parameters from the bridge design archive, including the standard value of material strength, design dimensions, design value of bearing capacity, and design grade of corrosion resistance for underwater components; extract historical defect data from the bridge inspection history database, including the location, extent, and development trend of defects recorded in past inspections; Step S202: Integrate and analyze the structural design parameters, historical defect data, and collected real-time environmental data to calculate the dynamic health threshold; specifically: Based on the safety standards of components in the structural design parameters, a basic health threshold is set for each detection index. The intervals are divided according to the average annual expansion rate of defects in historical defect data, and a historical defect impact coefficient is set for each interval. The higher the expansion rate, the greater the impact coefficient. The environmental data includes the concentration of corrosive media in the water, the water flow velocity, and the water temperature. The corrosion correction coefficient, the position offset correction coefficient, and the temperature correction coefficient calculated in step S201 are obtained respectively. The product of each correction coefficient is taken as the total environmental impact coefficient. The dynamic health threshold is obtained by dividing the basic health threshold by the product of the historical defect impact coefficient and the total environmental impact coefficient; corresponding dynamic health thresholds are formed for different components and different detection indicators. Step S203: Using the calibrated 3D model as a reference, establish a comparison coordinate system and convert the real-time structural data collected in step S101 to the comparison coordinate system; for the detection points or detection units of the components in the model, extract the actual parameters from the real-time structural data of the corresponding detection points or detection units in the 3D model, and compare the actual parameters with the corresponding dynamic health threshold one by one; if the actual parameter exceeds the dynamic health threshold, mark the parameter as an abnormal parameter and record the location, parameter type, and parameter exceedance range of the abnormal parameter; Step S204: Based on the distribution of abnormal parameters marked in step S203, divide the abnormal region morphology according to the abnormal parameter type. The abnormal region morphology includes point type, line type, and area type. For different morphological abnormal regions, set the corresponding extension range based on the location of the abnormal parameter to form the target monitoring area. Mark the target monitoring area in the calibrated 3D model, assign a unique number, and associate it with the corresponding abnormal parameter information.

5. The bridge anomaly monitoring method based on a three-dimensional model according to claim 4, characterized in that: Step S300 includes: Step S301: Based on the magnitude of the abnormal parameters exceeding the dynamic health threshold in step S203, the monitoring priority is divided. The greater the magnitude of the exceedance, the higher the monitoring priority. For target monitoring areas with different priorities, the corresponding periodic collection frequency is set. The higher the priority, the higher the collection frequency. Before collection, the deviation between the current environmental data and the environmental data collected for the first time is checked. If the deviation exceeds the preset range, the detection equipment parameters are adjusted. After collection, the data is associated with the unique number of the corresponding target monitoring area. Step S302: Using the calibrated 3D model as a reference, establish a feature extraction network within the target monitoring area; map the data periodically collected in step S301 to the corresponding nodes of the feature extraction network, and extract abnormal features according to defect type. The abnormal features include corrosion defects, crack defects, and deformation defects. Corrosion defects include corrosion area, corrosion depth, and corrosion rate, with the corrosion rate calculated by comparing corrosion area or corrosion depth collected at different times. Crack defects include crack length, crack width, and crack propagation rate, with the crack propagation rate calculated by comparing crack length or crack width collected at different times. Deformation defects include deformation amount, deformation direction, and deformation rate, with the deformation rate calculated by comparing deformation amount collected at different times. Associate the extracted abnormal feature parameters with the collection time and the unique number of the target monitoring area to form a time-series feature database. Step S303: Set the weights of the three assessment dimensions—defect severity, defect impact range, and defect development rate—in the risk assessment; for each dimension, divide the scoring interval according to the abnormal characteristic parameters, and determine the score corresponding to each scoring interval; wherein, the defect severity is divided into scoring intervals according to the extent to which the abnormal characteristic parameters exceed the dynamic health threshold, the larger the extent, the higher the score; the defect impact range is divided into scoring intervals according to the degree of influence of the defect location on the structural function of the component, the greater the influence, the higher the score; the defect development rate is divided into scoring intervals according to the rate of expansion or change of the defect, the faster the rate, the higher the score; obtain the scores of each dimension of the target monitoring area, multiply each dimension score by its corresponding weight, and sum them to obtain the abnormal risk index of the target monitoring area; record the abnormal risk index calculated each time to form a risk index change curve.

6. The bridge anomaly monitoring method based on a three-dimensional model according to claim 5, characterized in that: Step S400 includes: Step S401: Based on the impact of the abnormal risk index on bridge safety, classify the early warning levels, set the risk index range corresponding to each early warning level, and set the response time limit and handling requirements for each early warning level. The higher the early warning level, the shorter the response time limit and the more urgent the handling requirements. Step S402: Compare the abnormal risk index calculated in step S303 with the preset warning level range to match the corresponding warning level; send warning prompts to relevant departments through the communication system, with different recipients and sending frequencies for different warning levels; wherein, the warning prompt content includes the target monitoring area number, abnormal characteristics, risk index, and corresponding handling requirements; Step S403: Continuously monitor the abnormal risk index of the target monitoring area at the frequency set in step S301, and set the conditions for lifting the warning; if the conditions for lifting the warning are met, send a warning lifting notification to the warning recipient through the communication system, mark the warning status as lifted in the system, and archive the relevant detection data and processing records.

7. A bridge anomaly monitoring system based on a three-dimensional model, used to execute the bridge anomaly monitoring method based on a three-dimensional model as described in any one of claims 1-6, characterized in that: The bridge anomaly monitoring system includes: a data acquisition module, an intelligent analysis module, a risk assessment module, and an early warning management module; The data acquisition module is used to collect structural data and environmental data of the underwater structure of the bridge through underwater detection equipment, and to calculate the detection accuracy correction coefficient through an environmental adaptive algorithm to calibrate the initial three-dimensional model. The intelligent analysis module is used to integrate structural design parameters, historical defect data and real-time environmental data, calculate dynamic health threshold, identify abnormal parameters and delineate target monitoring areas; The risk assessment module is used to periodically collect data on the target monitoring area, extract abnormal features, and calculate the abnormal risk index. The early warning management module is used to trigger tiered early warnings based on the abnormal risk index and track the early warning status until the risk is eliminated. The data acquisition module is electrically connected to the intelligent analysis module, the intelligent analysis module is electrically connected to the risk assessment module, and the risk assessment module is electrically connected to the early warning management module.

8. The bridge anomaly monitoring system based on a three-dimensional model according to claim 7, characterized in that: The data acquisition module includes: an acquisition unit, a correction unit, and a calibration unit; The acquisition unit uses a detection equipment group consisting of an underwater robot, underwater sonar detection equipment and a three-dimensional laser scanner to perform a comprehensive scan of the underwater structure of the bridge and simultaneously acquire structural data and environmental data. The correction unit is used to calculate the detection accuracy correction coefficient through an environmental adaptive algorithm. The detection accuracy correction coefficient is obtained by weighted summation of the turbidity correction coefficient, the position offset correction coefficient, the surface detection correction coefficient and their respective influence weights. The calibration unit is used to adjust the spatial coordinates and component size parameters of the initial three-dimensional model according to the calculated detection accuracy correction coefficient, thereby completing the calibration of the three-dimensional model.

9. The bridge anomaly monitoring system based on a three-dimensional model according to claim 7, characterized in that: The intelligent analysis module includes: a threshold calculation unit, a comparison unit, and a region delineation unit; The threshold calculation unit is used to retrieve structural design parameters and historical defect data from bridge design archives and historical inspection databases, and integrate and analyze them with real-time environmental data. The dynamic health threshold is calculated by dividing the basic health threshold by the product of the historical defect impact coefficient and the total environmental impact coefficient. The comparison unit is used to establish a comparison coordinate system based on the calibrated three-dimensional model, convert the real-time collected structural data to the coordinate system, compare the actual parameters with the corresponding dynamic health thresholds one by one, mark abnormal parameters and record the location, type and exceedance of abnormal parameters. The region delineation unit is used to divide the abnormal region shape according to the parameter type based on the distribution of the marked abnormal parameters, and to set the extension range based on the location of the abnormal parameters to form the target monitoring area. The target monitoring area is then marked and numbered in the calibrated three-dimensional model.

10. The bridge anomaly monitoring system based on a three-dimensional model according to claim 7, characterized in that: The risk assessment module includes: a feature extraction unit, a risk calculation unit, and a time series library unit; The feature extraction unit is used to establish a feature extraction network within the target monitoring area, map periodically collected data to grid nodes, and extract abnormal features according to defect type. The abnormal features include feature parameters of corrosion defects, crack defects, and deformation defects. The risk calculation unit is used to set the weights of three assessment dimensions: defect severity, defect impact range, and defect development rate. It scores each dimension based on abnormal characteristic parameters, and then multiplies each dimension score by its corresponding weight and sums the results to obtain the abnormal risk index of the target monitoring area. The time series library unit is used to store the extracted abnormal feature parameters, their collection time, and the target monitoring area number to form a time series feature database, and to record the abnormal risk index calculated each time to form a risk index change curve.

Images