A method and system for detecting the bridge rotation process based on 3D visualization data

Abstract

This invention provides a method and system for detecting bridge rotation processes based on three-dimensional visualization data, belonging to the field of bridge monitoring technology. The method includes: acquiring real-time sensor data and real-time environmental data at multiple moments during the detection cycle; determining a real-time environmental safety factor based on the real-time environmental data; determining a real-time driving safety factor and a real-time comprehensive safety factor based on the real-time sensor data; processing the real-time environmental safety factor, real-time driving safety factor, and real-time comprehensive safety factor using a trained bridge rotation risk prediction model to determine a bridge rotation risk prediction coefficient; and generating a bridge rotation process detection report based on the bridge rotation risk prediction coefficient, real-time environmental safety factor, real-time driving safety factor, and real-time comprehensive safety factor. According to this invention, the accuracy of bridge rotation process detection can be improved.

Description

Technical Field

[0001] This invention relates to the field of bridge monitoring technology, and in particular to a method and system for detecting the bridge rotation process based on three-dimensional visualization data. Background Technology

[0002] In related technologies, the detection of bridge rotation mainly relies on a combination of sensor detection and manual inspection. That is, it mainly depends on human factors. Over-reliance on human factors may make it difficult to ensure the timeliness and accuracy of data processing, resulting in poor real-time performance of the detection results and limited accuracy of the results.

[0003] The information disclosed in the background section of this application is intended only to enhance the understanding of the general background of this application and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0004] This invention provides a method and system for detecting the bridge rotation process based on three-dimensional visualization data, which can solve the technical problem that related technologies cannot guarantee the real-time performance and accuracy of the detection results.

[0005] According to a first aspect of the present invention, a three-dimensional visualization full-field monitoring method for the entire process of bridge rotation is provided, comprising: acquiring real-time sensor data and real-time environmental data at multiple moments during a detection cycle, wherein the real-time sensor data includes: real-time positioning data, real-time attitude data, real-time stress data, and real-time drive system data, and the real-time environmental data includes: real-time wind speed data and real-time temperature data; determining a real-time environmental safety factor based on the real-time environmental data; determining a real-time drive safety factor and a real-time comprehensive safety factor based on the real-time sensor data; processing the real-time environmental safety factor, the real-time drive safety factor, and the real-time comprehensive safety factor using a trained bridge rotation risk prediction model to determine a bridge rotation risk prediction coefficient; and generating a bridge rotation process detection report based on the bridge rotation risk prediction coefficient, the real-time environmental safety factor, the real-time drive safety factor, and the real-time comprehensive safety factor.

[0006] According to the present invention, determining the real-time environmental safety factor based on the real-time environmental data includes: determining the real-time cantilever end temperature difference and the temperature of key structural points based on the real-time temperature data; determining the temperature uniformity safety factor based on the real-time cantilever end temperature difference; determining the temperature range safety factor based on the temperature of the key structural points; determining the wind speed environmental safety factor based on the real-time wind speed data; and determining the real-time environmental safety factor based on the temperature uniformity safety factor, the temperature range safety factor, and the wind speed environmental safety factor.

[0007] According to the present invention, determining the real-time drive safety factor and the real-time comprehensive safety factor based on the real-time sensing data includes: determining the hydraulic pump working pressure, jack stroke, and motor system temperature rise based on the real-time drive system data; determining the real-time drive safety factor based on the hydraulic pump working pressure, the jack stroke, and the motor system temperature rise; and determining the real-time comprehensive safety factor based on the real-time positioning data, the real-time attitude data, and the real-time stress data.

[0008] According to the present invention, determining the real-time drive safety factor based on the hydraulic pump working pressure, the jack stroke, and the motor system temperature rise includes: determining the working pressure fluctuation value based on the hydraulic pump working pressure; determining the hydraulic pump working safety factor based on the working pressure fluctuation value and the hydraulic pump working pressure; determining the stroke difference between adjacent jacks based on the jack stroke; determining the jack synchronization safety factor based on the stroke difference between adjacent jacks; determining the motor system temperature rise safety factor based on the motor system temperature rise; and determining the real-time drive safety factor based on the hydraulic pump working safety factor, the jack synchronization safety factor, and the motor system temperature rise safety factor.

[0009] According to the present invention, determining the real-time comprehensive safety factor based on the real-time positioning data, the real-time attitude data, and the real-time stress data includes: determining the coordinates of key structural points in a preset coordinate system based on the real-time positioning data, wherein the preset coordinate system is a coordinate system established with a preset point on the ground where the rotating bridge is located as the origin; determining the real-time rotation angle and real-time rotation acceleration based on the real-time attitude data; determining the stress value of key points based on the real-time stress data; obtaining standard key point coordinates, standard rotation angle, standard rotation acceleration, and standard key point stress values ​​by designing a BIM model; determining the key point priority of each structural key point; and determining the real-time comprehensive safety factor based on the key point priority, the key point coordinates, the real-time rotation angle, the real-time rotation acceleration, the key point stress value, the standard key point coordinates, the standard rotation angle, the standard rotation acceleration, and the standard key point stress value.

[0010] According to the present invention, determining a real-time comprehensive safety factor based on the key point priority, the key point coordinates, the real-time rotation angle, the real-time rotation acceleration, the key point stress value, the standard key point coordinates, the standard rotation angle, the standard rotation acceleration, and the standard key point stress value includes: determining a key point state vector based on the key point coordinates and the key point stress value; determining a rotation state vector based on the real-time rotation angle and the real-time rotation acceleration; determining a standard key point state vector based on the standard key point coordinates and the standard key point stress value; determining a standard rotation state vector based on the standard rotation angle and the standard rotation acceleration; and determining a standard rotation state vector based on the formula. Determine the real-time comprehensive safety factor at the i-th moment of the detection cycle. Where min is the function for finding the minimum value. Let K be the coordinates of the k-th keypoint at the i-th time in the detection cycle. Let the stress value of the k-th key point be the stress value at the i-th moment of the detection cycle. Let k be the keypoint state vector at the i-th time of the detection period. for The transpose of , Let the coordinates of the k-th keypoint be the standard keypoint coordinates at the i-th time in the detection cycle. Let the stress value of the k-th key point be the standard key point stress value at the i-th time of the detection cycle. Let K be the standard keypoint state vector of the k-th keypoint at the i-th time in the detection period. The key point priority for the k-th key point. Let be the real-time rotation angle at the i-th moment of the detection cycle. To detect the real-time rotational acceleration at the i-th moment of the detection period, Let be the rotational state vector at the i-th moment of the detection period. for The transpose of , Let be the standard rotation angle at the i-th moment of the detection cycle. Let be the standard rotational acceleration at the i-th moment of the detection period. Let K be the standard rotation state vector at the i-th moment of the detection cycle, K be the number of key points, k ≤ K, and both k and K are positive integers.

[0011] According to the present invention, the training steps of the bridge rotation risk prediction model include: acquiring historical bridge rotation construction records for multiple historical construction periods; determining historical sensor data and historical environmental data based on the historical bridge rotation construction records; determining historical environmental safety factors, historical driving safety factors, and historical comprehensive safety factors based on the historical sensor data and the historical environmental data; processing the historical environmental safety factors, the historical driving safety factors, and the historical comprehensive safety factors through the bridge rotation risk prediction model to determine sample bridge rotation risk prediction coefficients; and determining historical construction alarms based on the historical bridge rotation construction records. Information; based on the historical construction alarm information, determine the time difference of historical alarm information, the historical alarm level, and the historical construction delay time; based on the time difference of historical alarm information, the historical alarm level, and the historical construction delay time, determine the historical bridge rotation risk coefficient; based on the historical bridge rotation risk coefficient, the sample bridge rotation risk prediction coefficient, the historical environmental safety coefficient, the historical driving safety coefficient, and the historical comprehensive safety coefficient, determine the training loss function of the bridge rotation risk prediction model; train the bridge rotation risk prediction model based on the training loss function to obtain the trained bridge rotation risk prediction model.

[0012] According to the present invention, the training loss function of the bridge rotation risk prediction model is determined based on the historical bridge rotation risk coefficient, the sample bridge rotation risk prediction coefficient, the historical environmental safety coefficient, the historical driving safety coefficient, and the historical comprehensive safety coefficient, including: according to the formula Determine the training loss function for the bridge rotation risk prediction model. ,in, Let be the sample bridge rotation risk prediction coefficient at the r-th moment of the e-th historical construction cycle. The historical bridge rotation risk coefficient is a preset time period after the r-th moment of the e-th historical construction cycle. Let r be the historical environmental safety factor at the r-th moment of the e-th historical construction cycle. To preset the environmental safety factor threshold, Let the historical driving safety factor be the value at the r-th moment of the e-th historical construction cycle. To preset the drive safety factor threshold, Let r be the historical comprehensive safety factor at the r-th moment of the e-th historical construction cycle. The threshold for the preset comprehensive safety factor is defined as follows: E is the number of historical construction cycles, e≤E, R is the number of moments in the historical construction cycles, r≤R, and e, E, r, and R are all positive integers.

[0013] According to a second aspect of the present invention, a three-dimensional visualization full-field monitoring system for the entire process of bridge rotation is provided, comprising: a real-time data module for acquiring real-time sensor data and real-time environmental data at multiple moments during a detection cycle, wherein the real-time sensor data includes real-time positioning data, real-time attitude data, real-time stress data, and real-time drive system data, and the real-time environmental data includes real-time wind speed data and real-time temperature data; an environmental coefficient module for determining a real-time environmental safety factor based on the real-time environmental data; a sensor coefficient module for determining a real-time drive safety factor and a real-time comprehensive safety factor based on the real-time sensor data; a risk prediction module for processing the real-time environmental safety factor, the real-time drive safety factor, and the real-time comprehensive safety factor using a trained bridge rotation risk prediction model to determine a bridge rotation risk prediction coefficient; and a detection report module for generating a bridge rotation process detection report based on the bridge rotation risk prediction coefficient, the real-time environmental safety factor, the real-time drive safety factor, and the real-time comprehensive safety factor.

[0014] Technical Effects: This system accurately acquires real-time sensor data and environmental data during bridge rotation. Based on this data, it assesses the environmental conditions, drive system status, and overall bridge condition, determining the real-time environmental safety factor, real-time drive safety factor, and real-time comprehensive safety factor. Furthermore, a trained bridge rotation risk prediction model processes these factors to predict the severity of potential risks, determining the bridge rotation risk prediction coefficient and generating a detection report, thus improving the accuracy of bridge rotation process detection. When determining the real-time comprehensive safety factor, it considers key point priority, key point coordinates, real-time rotation angle, real-time rotation acceleration, key point stress values, standard key point coordinates, standard rotation angle, standard rotation acceleration, and standard key point stress values. During the calculation, it assesses the safety status of the bridge rotation process from both the safety impact of structural key points on the rotation process and the overall bridge posture, improving the comprehensiveness and accuracy of the real-time comprehensive safety factor. When determining the training loss function for a bridge rotation risk prediction model, it can be based on historical bridge rotation risk coefficients, sample bridge rotation risk prediction coefficients, historical environmental safety coefficients, historical driving safety coefficients, and historical comprehensive safety coefficients. During this determination process, the potential impact of historical environmental safety coefficients, historical driving safety coefficients, and historical comprehensive safety coefficients on the occurrence of risk conditions can be considered to determine the error of the sample bridge rotation risk prediction coefficients. Based on this impact and the relative error of the sample bridge rotation risk prediction coefficients, the training loss function can be set to reduce the training loss function during the training process, thereby more effectively improving the accuracy of the bridge rotation risk prediction model.

[0015] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Other features and aspects of the invention will become clearer from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained based on these drawings without creative effort.

[0017] Figure 1 An exemplary flowchart of a bridge rotation process detection method based on three-dimensional visualization data according to an embodiment of the present invention is shown.

[0018] Figure 2 An exemplary flowchart illustrating the determination of the real-time environmental safety factor according to an embodiment of the present invention is shown;

[0019] Figure 3 A flowchart illustrating the determination of the real-time drive safety factor and the real-time integrated safety factor according to an embodiment of the present invention is shown;

[0020] Figure 4 A block diagram of a bridge rotation process detection system based on three-dimensional visualization data according to an embodiment of the present invention is shown as an example. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0023] Figure 1 An exemplary flowchart of a three-dimensional visualization full-field monitoring method for the entire bridge rotation process according to an embodiment of the present invention is shown. The method includes: Step S1, acquiring real-time sensor data and real-time environmental data at multiple moments during the detection cycle, wherein the real-time sensor data includes: real-time positioning data, real-time attitude data, real-time stress data, and real-time drive system data, and the real-time environmental data includes: real-time wind speed data and real-time temperature data; Step S2, determining a real-time environmental safety factor based on the real-time environmental data; Step S3, determining a real-time drive safety factor and a real-time comprehensive safety factor based on the real-time sensor data; Step S4, processing the real-time environmental safety factor, the real-time drive safety factor, and the real-time comprehensive safety factor using a trained bridge rotation risk prediction model to determine a bridge rotation risk prediction coefficient; Step S5, generating a bridge rotation process detection report based on the bridge rotation risk prediction coefficient, the real-time environmental safety factor, the real-time drive safety factor, and the real-time comprehensive safety factor.

[0024] The bridge rotation process detection method based on three-dimensional visualization data according to an embodiment of the present invention can accurately acquire real-time sensor data and real-time environmental data during the bridge rotation process. Based on the real-time sensor data and real-time environmental data, the method assesses the environmental conditions, drive system conditions, and overall bridge conditions to determine the real-time environmental safety factor, real-time drive safety factor, and real-time comprehensive safety factor. Furthermore, the real-time environmental safety factor, real-time drive safety factor, and real-time comprehensive safety factor can be processed using a trained bridge rotation risk prediction model to predict the severity of potential risks, determine the bridge rotation risk prediction coefficient, and generate a detection report, thereby improving the accuracy of bridge rotation process detection.

[0025] According to an embodiment of the present invention, in step S1, real-time sensing data and real-time environmental data are acquired at multiple moments during the detection cycle. The real-time sensing data includes real-time positioning data, real-time attitude data, real-time stress data, and real-time drive system data. The real-time environmental data includes real-time wind speed data and real-time temperature data.

[0026] For example, GNSS receivers are deployed at key structural points of the bridge (e.g., the center of the ball joint, the beam end) to acquire real-time positioning data; fiber optic canopy sensors are installed at key structural points of the bridge to acquire real-time stress data; dual-axis tilt sensors and accelerometers are installed at the core moving parts of the bridge's rotating structure (e.g., the ball joint support point of the bridge rotation project, the key section of the beam) to acquire real-time attitude data of the bridge rotation; real-time drive system data is acquired through sensors (e.g., travel encoders and pressure sensors) installed in the control system; and real-time environmental data is acquired through anemometers and temperature sensors installed at preset locations (e.g., the highest operating point of the bridge, the end of the structure, and key points of the bridge).

[0027] According to one embodiment of the present invention, in step S2, a real-time environmental safety factor is determined based on the real-time environmental data.

[0028] Figure 2 A flowchart for determining the real-time environmental safety factor according to an embodiment of the present invention is shown as an example.

[0029] According to an embodiment of the present invention, step S2 includes: step S21, determining the real-time cantilever end temperature difference and the temperature of the key structural point based on the real-time temperature data; step S22, determining the temperature uniformity safety factor based on the real-time cantilever end temperature difference; step S23, determining the temperature range safety factor based on the temperature of the key structural point; step S24, determining the wind speed environmental safety factor based on the real-time wind speed data; and step S25, determining the real-time environmental safety factor based on the temperature uniformity safety factor, the temperature range safety factor, and the wind speed environmental safety factor.

[0030] For example, based on real-time temperature data, the temperatures of the cantilever ends on both sides of the bridge main body and the temperatures of key structural points (such as ball joints and struts) are determined. The real-time cantilever end temperature difference is then determined based on the temperature difference between the two cantilever ends. A temperature uniformity safety factor is determined based on a preset temperature difference threshold (e.g., 2 degrees Celsius) and the ratio of the real-time cantilever end temperature difference to the preset threshold. A larger temperature uniformity safety factor results in a smaller temperature difference between the two cantilever ends, reducing the likelihood of axis displacement due to temperature differences and increasing safety. If the temperature of a key structural point falls within an ideal temperature range (e.g., 5 degrees Celsius to 25 degrees Celsius), the safety result for that temperature range is 1; otherwise, the safety result for the corresponding temperature range is... The total result is 0. The temperature range safety factor is determined by summing and averaging the results based on the number of key points. The larger the temperature range safety factor, the more structural key points are within the ideal temperature range, the smaller the impact of thermal expansion and contraction of building materials on the rotation construction, and the greater the safety. The wind speed environmental safety factor is determined by the ratio of the difference between the preset wind speed warning threshold (e.g., 6 m / s) and the real-time wind speed data to the preset wind speed warning threshold. The larger the wind speed environmental safety factor, the lower the real-time wind speed, the lower the risk of wind causing the bridge to sway or even collapse, and the higher the safety. The real-time environmental safety factor is determined by summing the temperature uniformity safety factor, the temperature range safety factor, and the wind speed environmental safety factor.

[0031] According to an embodiment of the present invention, in step S3, the real-time drive safety factor and the real-time comprehensive safety factor are determined based on the real-time sensing data.

[0032] Figure 3 A flowchart illustrating the determination of the real-time drive safety factor and the real-time integrated safety factor according to an embodiment of the present invention is shown.

[0033] According to an embodiment of the present invention, step S3 includes: step S31, determining the hydraulic pump working pressure, jack stroke, and motor system temperature rise based on the real-time drive system data; step S32, determining the real-time drive safety factor based on the hydraulic pump working pressure, the jack stroke, and the motor system temperature rise; and step S33, determining the real-time comprehensive safety factor based on the real-time positioning data, the real-time attitude data, and the real-time stress data.

[0034] For example, the hydraulic pump's working pressure is obtained through a pressure sensor installed on the hydraulic pump, the jack stroke is obtained through a stroke encoder, and the temperature difference between the motor system and the surrounding environment, i.e., the motor system temperature rise, is obtained through a temperature sensor installed on the motor system. Based on the hydraulic pump's working pressure, the jack stroke, and the motor system temperature rise, the working condition of the drive system is evaluated to determine the real-time drive safety factor. Based on real-time positioning data, real-time attitude data, and real-time stress data, the bridge's attitude, position, and stress safety status during the bridge rotation process is evaluated to determine the real-time comprehensive safety factor.

[0035] According to an embodiment of the present invention, step S32 includes: step S321, determining the working pressure fluctuation value based on the working pressure of the hydraulic pump; step S323, determining the working safety factor of the hydraulic pump based on the working pressure fluctuation value and the working pressure of the hydraulic pump; step S324, determining the stroke difference between adjacent jacks based on the jack stroke; step S325, determining the jack synchronization safety factor based on the stroke difference between adjacent jacks; step S326, determining the motor system temperature rise safety factor based on the motor system temperature rise; and step S327, determining the real-time drive safety factor based on the hydraulic pump working safety factor, the jack synchronization safety factor, and the motor system temperature rise safety factor.

[0036] For example, the working pressure fluctuation value is determined based on the difference between the hydraulic pump's working pressure at the current moment of the detection cycle and the hydraulic pump's working pressure at the previous moment. A first ratio is determined based on the ratio of the difference between the set pressure fluctuation warning threshold (which can be set to 10% of the hydraulic pump's rated pressure) and the set pressure fluctuation warning threshold. The larger the first ratio, the smaller the working pressure fluctuation value, the lower the possibility of oil circuit rupture, and the better the working safety condition of the hydraulic pump. A second ratio is determined based on the ratio of the difference between the set abnormal working pressure peak value (which can be set to 130% of the hydraulic pump's rated pressure) and the set abnormal working pressure peak value. The larger the second ratio, the lower the hydraulic pump's working pressure, the lower the possibility of hydraulic pump overload, and the better the working safety condition of the hydraulic pump. The hydraulic pump's working safety factor is determined by summing the first and second ratios. Based on the jack stroke, determine the stroke difference between adjacent jacks; based on the preset stroke difference threshold (which can be set to 2mm) and the ratio of the difference between the maximum stroke differences of multiple adjacent jacks and the preset stroke difference threshold, determine the jack synchronization safety factor. The larger the jack synchronization safety factor, the smaller the maximum stroke difference of multiple adjacent jacks, the lower the possibility of structural torque exceeding the limit, and the higher the stroke synchronization of multiple jacks; based on the preset temperature rise threshold (which can be set to 60 degrees Celsius) and the ratio of the difference in temperature rise of the motor system and the preset temperature rise threshold, determine the motor system temperature rise safety factor. The larger the motor system temperature rise safety factor, the lower the possibility of motor system burnout; based on the sum of the hydraulic pump working safety factor, the jack synchronization safety factor, and the motor system temperature rise safety factor, determine the real-time drive safety factor.

[0037] According to an embodiment of the present invention, step S33 includes: step S331, determining the coordinates of key points of the structural key points in a preset coordinate system based on the real-time positioning data, wherein the preset coordinate system is a coordinate system established with a preset point on the ground where the rotating bridge is located as the origin; step S332, determining the real-time rotation angle and real-time rotation acceleration based on the real-time attitude data; step S333, determining the stress value of key points based on the real-time stress data; step S334, obtaining the standard key point coordinates, standard rotation angle, standard rotation acceleration, and standard key point stress value by designing a BIM model; step S335, determining the key point priority of each structural key point; step S336, determining the real-time comprehensive safety factor based on the key point priority, the key point coordinates, the real-time rotation angle, the real-time rotation acceleration, the key point stress value, the standard key point coordinates, the standard rotation angle, the standard rotation acceleration, and the standard key point stress value.

[0038] For example, a coordinate system is established with a predetermined point on the ground where the rotating bridge is located (e.g., the rotation support point) as the origin, the ground as the xoy plane, and the vertically upward direction as the z-axis. The coordinates of key structural points in the predetermined coordinate system are determined using a GNSS receiver. Real-time rotation angles and accelerations of the bridge are acquired using dual-axis tilt sensors and accelerometers. Stress values ​​at key structural points are acquired using fiber optic sensors. A design BIM model is established based on the bridge's design parameters, and the bridge rotation process is simulated based on the bridge rotation plan and the design BIM model. The standard key point coordinates and stress values ​​of each structural key point under standard conditions during the bridge rotation process are obtained through the design BIM model, along with the overall quasi-rotation angle and standard rotation acceleration of the bridge. The weights of the structural key points' influence on the overall stability of the bridge are then determined. Based on the severity of failure consequences and real-time control requirements, the priority of each structural critical point is determined. For example, the priority of the critical point at the center of the ball hinge is 10, the priority of the critical point at the connection point between the pier and the abutment is 9, the priority of the critical point at the beam end is 8, the priority of the critical point at the critical section of the beam is 7, the priority of the critical point at the top of the main tower is 6, and the priority of the critical point at the position of the rotation track is 5. The higher the priority, the greater the importance of the structural critical point and the greater its impact on the rotation. Based on the critical point priority, critical point coordinates, real-time rotation angle, real-time rotation acceleration, critical point stress value, standard critical point coordinates, standard rotation angle, standard rotation acceleration, and standard critical point stress value, the safety status of the bridge's attitude, position, speed, and angle during the rotation process is evaluated, and a real-time comprehensive safety factor is determined.

[0039] According to an embodiment of the present invention, step S336 includes: determining a key point state vector based on the key point coordinates and the key point stress value; determining a rotation state vector based on the real-time rotation angle and the real-time rotation acceleration; determining a standard key point state vector based on the standard key point coordinates and the standard key point stress value; determining a standard rotation state vector based on the standard rotation angle and the standard rotation acceleration; and determining the real-time comprehensive safety factor at the i-th moment of the detection cycle according to formula (1). ,

[0040] (1)

[0041] Where min is the function for finding the minimum value. Let K be the coordinates of the k-th keypoint at the i-th time in the detection cycle. Let the stress value of the k-th key point be the stress value at the i-th moment of the detection cycle. Let k be the keypoint state vector at the i-th time of the detection period. for The transpose of , Let the coordinates of the k-th keypoint be the standard keypoint coordinates at the i-th time in the detection cycle. Let the stress value of the k-th key point be the standard key point stress value at the i-th time of the detection cycle. Let K be the standard keypoint state vector of the k-th keypoint at the i-th time in the detection period. The key point priority for the k-th key point. Let be the real-time rotation angle at the i-th moment of the detection cycle. To detect the real-time rotational acceleration at the i-th moment of the detection period, Let be the rotational state vector at the i-th moment of the detection period. for The transpose of , Let be the standard rotation angle at the i-th moment of the detection cycle. Let be the standard rotational acceleration at the i-th moment of the detection period. Let K be the standard rotation state vector at the i-th moment of the detection cycle, K be the number of key points, k ≤ K, and both k and K are positive integers.

[0042] According to one embodiment of the present invention, The cosine similarity between the state vector of the k-th keypoint at time i in the detection cycle and the state vector of the standard keypoint is calculated. The larger the cosine similarity, the closer the keypoint coordinates and stress value of the k-th keypoint at time i in the detection cycle are to the preset standard. To find the minimum cosine similarity between the state vectors of the K keypoints at time i of the detection period and the state vectors of the standard keypoints, the above minimum value taking process can be used to determine the situation with the most severe anomaly among the K keypoints at time i of the detection period. Let be the ratio of the minimum cosine similarity between the state vector of a key point and the state vector of a standard key point at the i-th moment of the detection period, to the corresponding key point priority. This represents the safety factor at the structural key point during the rotation process, determined based on the anomalies and importance of the structural key point. A higher key point priority indicates greater importance, which amplifies anomalies at that key point and has a more severe impact on the rotation process. The lower the value, The cosine similarity between the rotation state vector at time i of the detection period and the standard rotation state vector is given. The greater the cosine similarity, the closer the real-time rotation angle and real-time rotation acceleration of the bridge at time i of the detection period are to the standard rotation angle and standard rotation acceleration.

[0043] In this way, the real-time comprehensive safety factor can be determined based on the priority of key points, key point coordinates, real-time rotation angle, real-time rotation acceleration, key point stress value, standard key point coordinates, standard rotation angle, standard rotation acceleration, and standard key point stress value. During the calculation process, the safety status of the bridge during the rotation process can be evaluated from two aspects: the safety impact of structural key points on the rotation process and the overall attitude of the bridge. This improves the comprehensiveness and accuracy of the real-time comprehensive safety factor.

[0044] According to an embodiment of the present invention, in step S4, the real-time environmental safety factor, the real-time driving safety factor, and the real-time comprehensive safety factor are processed by the trained bridge rotation risk prediction model to determine the bridge rotation risk prediction coefficient.

[0045] According to an embodiment of the present invention, the training steps of the bridge rotation risk prediction model include: acquiring historical bridge rotation construction records for multiple historical construction cycles; determining historical sensor data and historical environmental data based on the historical bridge rotation construction records; determining historical environmental safety coefficient, historical driving safety coefficient, and historical comprehensive safety coefficient based on the historical sensor data and the historical environmental data; processing the historical environmental safety coefficient, the historical driving safety coefficient, and the historical comprehensive safety coefficient through the bridge rotation risk prediction model to determine sample bridge rotation risk prediction coefficients; determining historical construction alarm information based on the historical bridge rotation construction records; determining historical alarm information time difference, historical alarm level, and historical construction delay time based on the historical construction alarm information; determining historical bridge rotation risk coefficients based on the historical alarm information time difference, the historical alarm level, and the historical construction delay time; and determining the training loss function of the bridge rotation risk prediction model based on the historical bridge rotation risk coefficients, the sample bridge rotation risk prediction coefficients, the historical environmental safety coefficients, the historical driving safety coefficients, and the historical comprehensive safety coefficients.

[0046] The bridge rotation risk prediction model is trained according to the training loss function to obtain the trained bridge rotation risk prediction model.

[0047] For example, obtain the construction records of historical bridges similar to the current bridge that have already been constructed, i.e., historical bridge rotation construction records; in the historical bridge rotation construction records, identify historical sensor data and historical environmental data. Historical sensor data includes: historical positioning data, historical attitude data, historical stress data, and historical drive system data; historical environmental data includes: historical wind speed data and historical temperature data; based on the historical sensor data, determine the historical drive safety factor and the historical comprehensive safety factor. The method for determining the historical drive safety factor and the historical comprehensive safety factor is the same as the method for determining the real-time drive safety factor and the real-time comprehensive safety factor, and will not be repeated here. Based on the historical environmental data, determine the historical environmental safety factor. The method for determining the environmental safety factor is the same as that for determining the real-time environmental safety factor, and will not be repeated here. Based on the bridge rotation risk prediction model, the historical environmental safety factor, historical driving safety factor, and historical comprehensive safety factor are processed. The urgency of potential dangerous situations within a preset time period (e.g., 30 minutes) following the corresponding time point of the historical environmental safety factor is predicted to determine the sample bridge rotation risk prediction coefficient. Based on historical bridge rotation construction records, historical construction alarm information for a preset time period following the acquisition time point of historical sensor data and historical environmental data is determined. For example, when acquiring historical sensor data and historical environmental data at the initial moment of the first historical construction cycle, the first historical construction cycle... The alarm information recorded within 30 minutes after the initial moment of the period is called historical construction alarm information. Based on this historical alarm information, the time difference between the time the alarm was issued and the time the historical sensor and environmental data were acquired is determined; this is the historical alarm information time difference. Alarms are typically classified into three levels: Level 1, Level 2, and Level 3, with Level 1 being the most severe. The historical alarm level is determined based on the historical construction alarm information. The construction delay time caused by this alarm is also determined based on the historical construction alarm information; this is the historical construction delay time. A third ratio is determined based on the ratio of a preset time difference threshold (which can be set to 1 minute) to the historical alarm information time difference. The smaller the historical alarm information time difference, the more likely a hazard will occur in a short time. The larger the third ratio, the higher the risk level. A fourth ratio is determined based on the ratio of a preset alarm level threshold (which can be set to level 3) to historical alarm levels. The smaller the historical alarm level, the higher the risk level, and the larger the fourth ratio. A fifth ratio is determined based on the ratio of historical construction delay time to a preset delay time (which can be set to 1 hour). The longer the historical construction delay time, the more difficult the hazard management, and the larger the fifth ratio. The historical bridge rotation risk coefficient is determined by summing the third, fourth, and fifth ratios. The training loss function for the bridge rotation risk prediction model is determined based on the historical bridge rotation risk coefficient, the sample bridge rotation risk prediction coefficient, the historical environmental safety coefficient, the historical driving safety coefficient, and the historical comprehensive safety coefficient.The bridge rotation risk prediction model is trained based on the training loss function to improve its accuracy in predicting bridge rotation risks, thus obtaining a trained bridge rotation risk prediction model.

[0048] According to an embodiment of the present invention, the training loss function of the bridge rotation risk prediction model is determined based on the historical bridge rotation risk coefficient, the sample bridge rotation risk prediction coefficient, the historical environmental safety coefficient, the historical driving safety coefficient, and the historical comprehensive safety coefficient, including: determining the training loss function of the bridge rotation risk prediction model according to formula (2). ,

[0049] (2)

[0050] in, Let be the sample bridge rotation risk prediction coefficient at the r-th moment of the e-th historical construction cycle. The historical bridge rotation risk coefficient is a preset time period after the r-th moment of the e-th historical construction cycle. Let r be the historical environmental safety factor at the r-th moment of the e-th historical construction cycle. To preset the environmental safety factor threshold, Let the historical driving safety factor be the value at the r-th moment of the e-th historical construction cycle. To preset the drive safety factor threshold, Let r be the historical comprehensive safety factor at the r-th moment of the e-th historical construction cycle. The threshold for the preset comprehensive safety factor is defined as follows: E is the number of historical construction cycles, e≤E, R is the number of moments in the historical construction cycles, r≤R, and e, E, r, and R are all positive integers.

[0051] According to one embodiment of the present invention, This is the ratio of the historical environmental safety factor at the r-th moment of the e-th historical construction cycle to the preset environmental safety factor threshold. The larger this ratio, the larger the historical environmental safety factor. The preset environmental safety factor threshold can be set to 1. This is the ratio of the historical drive safety factor at the r-th moment of the e-th historical construction cycle to the preset drive safety factor threshold. The larger this ratio, the larger the historical drive safety factor. The preset drive safety factor threshold can be set to 1. This is the ratio of the historical comprehensive safety factor at the r-th moment of the e-th historical construction cycle to the preset comprehensive safety factor threshold. The larger this ratio, the larger the historical comprehensive safety factor. The preset comprehensive safety factor threshold can be set to 1. This indicates a negative correlation between the historical environmental safety factor, historical driving safety factor, historical comprehensive safety factor, and the sample bridge rotation risk prediction coefficient. For example, a larger historical environmental safety factor indicates a smaller impact of temperature and wind on construction safety, resulting in a smaller sample bridge rotation risk prediction coefficient. Conversely, a larger sample bridge rotation risk prediction coefficient indicates a more normal operating condition of the driving system, further reducing the risk prediction coefficient. Similarly, a larger historical comprehensive safety factor indicates a more normal state of key structural points and overall bridge rotation acceleration during the rotation process, also leading to a lower risk prediction coefficient. Therefore, placing the related terms of the historical environmental safety factor, historical driving safety factor, and historical comprehensive safety factor in the denominator represents... , and The larger the value, the smaller the predicted coefficient of bridge rotation risk.

[0052] According to one embodiment of the present invention, The relative error between the predicted bridge rotation risk coefficient at time r of the e-th historical construction cycle and the historical bridge rotation risk coefficient over a preset period after time r of the e-th historical construction cycle is used to... The training loss function is obtained by weighted averaging the relative errors between the sample bridge rotation risk prediction coefficient at time r of the e-th historical construction cycle and the historical bridge rotation risk coefficient over a preset period after time r of the e-th historical construction cycle. During training, this training loss function is reduced, thereby reducing the error between the sample bridge rotation risk prediction coefficient and the historical bridge rotation risk coefficient. This improves the prediction accuracy of the bridge rotation risk prediction model and ultimately enhances its overall accuracy.

[0053] In this way, the training loss function of the bridge rotation risk prediction model can be determined based on historical bridge rotation risk coefficients, sample bridge rotation risk prediction coefficients, historical environmental safety coefficients, historical driving safety coefficients, and historical comprehensive safety coefficients. During the determination process, the possible impact of historical environmental safety coefficients, historical driving safety coefficients, and historical comprehensive safety coefficients on the occurrence of risk conditions can be used to determine the influence of the above data on the error of the sample bridge rotation risk prediction coefficients. Based on this influence and the relative error of the sample bridge rotation risk prediction coefficients, the training loss function can be set, thereby reducing the training loss function during the training process of the bridge rotation risk prediction model and improving the accuracy of the bridge rotation risk prediction model in a more targeted manner.

[0054] According to an embodiment of the present invention, in step S5, a bridge rotation process detection report is generated based on the bridge rotation risk prediction coefficient, the real-time environmental safety coefficient, the real-time driving safety coefficient, and the real-time comprehensive safety coefficient.

[0055] For example, based on the real-time environmental safety factor, the environmental conditions during the bridge rotation process are determined. The larger the real-time environmental safety factor, the more suitable the environmental conditions are for construction. Based on the real-time drive safety factor, the working condition of the drive system during the bridge rotation process is determined. The larger the real-time drive safety factor, the better the working condition of the drive system. Based on the real-time comprehensive safety factor, the stress, position, and overall attitude of key points during the bridge rotation process are determined. The larger the real-time comprehensive safety factor, the better the overall condition of the bridge. Based on the bridge rotation risk prediction coefficient, the severity of possible risks is determined. The larger the bridge rotation risk prediction coefficient, the more severe the possible risks.

[0056] The bridge rotation process detection method based on three-dimensional visualization data according to embodiments of the present invention can accurately acquire real-time sensor data and real-time environmental data during the bridge rotation process. Based on the real-time sensor data and real-time environmental data, the environmental conditions, drive system conditions, and overall bridge conditions are evaluated to determine the real-time environmental safety factor, real-time drive safety factor, and real-time comprehensive safety factor. Furthermore, the real-time environmental safety factor, real-time drive safety factor, and real-time comprehensive safety factor can be processed using a trained bridge rotation risk prediction model to predict the severity of potential risks, determine the bridge rotation risk prediction coefficient, and generate a detection report, thus improving the accuracy of bridge rotation process detection. When determining the real-time comprehensive safety factor, it can be determined based on key point priority, key point coordinates, real-time rotation angle, real-time rotation acceleration, key point stress value, standard key point coordinates, standard rotation angle, standard rotation acceleration, and standard key point stress value. During the calculation process, the safety status of the bridge rotation process can be evaluated from two aspects: the safety impact of structural key points on the rotation process and the overall bridge attitude, thus improving the comprehensiveness and accuracy of the real-time comprehensive safety factor. When determining the training loss function for a bridge rotation risk prediction model, it can be based on historical bridge rotation risk coefficients, sample bridge rotation risk prediction coefficients, historical environmental safety coefficients, historical driving safety coefficients, and historical comprehensive safety coefficients. During this determination process, the potential impact of historical environmental safety coefficients, historical driving safety coefficients, and historical comprehensive safety coefficients on the occurrence of risk conditions can be considered to determine the error of the sample bridge rotation risk prediction coefficients. Based on this impact and the relative error of the sample bridge rotation risk prediction coefficients, the training loss function can be set to reduce the training loss function during the training process, thereby more effectively improving the accuracy of the bridge rotation risk prediction model.

[0057] Figure 4An exemplary block diagram of a bridge rotation process detection system based on three-dimensional visualization data according to an embodiment of the present invention is shown. The system includes: a real-time data module, used to acquire real-time sensor data and real-time environmental data at multiple moments during the detection cycle, wherein the real-time sensor data includes: real-time positioning data, real-time attitude data, real-time stress data, and real-time drive system data, and the real-time environmental data includes: real-time wind speed data and real-time temperature data; an environmental coefficient module, used to determine a real-time environmental safety factor based on the real-time environmental data; a sensor coefficient module, used to determine a real-time drive safety factor and a real-time comprehensive safety factor based on the real-time sensor data; a risk prediction module, used to process the real-time environmental safety factor, the real-time drive safety factor, and the real-time comprehensive safety factor through a trained bridge rotation risk prediction model to determine a bridge rotation risk prediction coefficient; and a detection report module, used to generate a bridge rotation process detection report based on the bridge rotation risk prediction coefficient, the real-time environmental safety factor, the real-time drive safety factor, and the real-time comprehensive safety factor.

[0058] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0059] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are merely examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functions and structural principles of the present invention have been demonstrated and explained in the embodiments, and any variations or modifications may be made to the implementation of the present invention without departing from the stated principles.

Claims

1. A method for detecting the bridge rotation process based on three-dimensional visualization data, characterized in that, include: At multiple points during the detection cycle, real-time sensor data and real-time environmental data are acquired. The real-time sensor data includes real-time positioning data, real-time attitude data, real-time stress data, and real-time drive system data. The real-time environmental data includes real-time wind speed data and real-time temperature data. Based on the real-time environmental data, a real-time environmental safety factor is determined. Based on the real-time sensor data, a real-time drive safety factor and a real-time comprehensive safety factor are determined. The real-time environmental safety factor, the real-time drive safety factor, and the real-time comprehensive safety factor are processed using a trained bridge rotation risk prediction model to determine a bridge rotation risk prediction coefficient. Based on the bridge rotation risk prediction coefficient, the real-time environmental safety factor, the real-time drive safety factor, and the real-time comprehensive safety factor, a bridge rotation process detection report is generated. Based on the real-time sensing data, the real-time drive safety factor and the real-time comprehensive safety factor are determined, including: determining the hydraulic pump working pressure, jack stroke, and motor system temperature rise based on the real-time drive system data; determining the real-time drive safety factor based on the hydraulic pump working pressure, the jack stroke, and the motor system temperature rise; and determining the real-time comprehensive safety factor based on the real-time positioning data, the real-time attitude data, and the real-time stress data. Based on the real-time positioning data, the real-time attitude data, and the real-time stress data, a real-time comprehensive safety factor is determined, including: determining the coordinates of key structural points in a preset coordinate system based on the real-time positioning data, wherein the preset coordinate system is a coordinate system established with a preset point on the ground where the rotating bridge is located as the origin; determining the real-time rotation angle and real-time rotation acceleration based on the real-time attitude data; determining the stress value of key points based on the real-time stress data; obtaining standard key point coordinates, standard rotation angle, standard rotation acceleration, and standard key point stress values ​​by designing a BIM model; determining the key point priority of each structural key point; and determining the real-time comprehensive safety factor based on the key point priority, the key point coordinates, the real-time rotation angle, the real-time rotation acceleration, the key point stress value, the standard key point coordinates, the standard rotation angle, the standard rotation acceleration, and the standard key point stress value.

2. The bridge rotation process detection method based on three-dimensional visualization data according to claim 1, characterized in that, Based on the real-time environmental data, the real-time environmental safety factor is determined, including: based on the real-time temperature data, determining the real-time temperature difference at the cantilever end and the temperature at key structural points; based on the real-time temperature difference at the cantilever end, determining the temperature uniformity safety factor; based on the temperature at the key structural points, determining the temperature range safety factor; based on the real-time wind speed data, determining the wind speed environmental safety factor; and based on the temperature uniformity safety factor, the temperature range safety factor, and the wind speed environmental safety factor, determining the real-time environmental safety factor.

3. The bridge rotation process detection method based on three-dimensional visualization data according to claim 2, characterized in that, The real-time drive safety factor is determined based on the hydraulic pump's working pressure, the jack's stroke, and the motor system's temperature rise. This includes: determining the working pressure fluctuation value based on the hydraulic pump's working pressure; determining the hydraulic pump's working safety factor based on the working pressure fluctuation value and the hydraulic pump's working pressure; determining the stroke difference between adjacent jacks based on the jack's stroke; determining the jack synchronization safety factor based on the stroke difference between adjacent jacks; determining the motor system's temperature rise safety factor based on the motor system's temperature rise; and determining the real-time drive safety factor based on the hydraulic pump's working safety factor, the jack synchronization safety factor, and the motor system's temperature rise safety factor.

4. The bridge rotation process detection method based on three-dimensional visualization data according to claim 3, characterized in that, The real-time comprehensive safety factor is determined based on the key point priority, key point coordinates, real-time rotation angle, real-time rotation acceleration, key point stress value, standard key point coordinates, standard rotation angle, standard rotation acceleration, and standard key point stress value. This includes: determining a key point state vector based on the key point coordinates and key point stress value; determining a rotation state vector based on the real-time rotation angle and real-time rotation acceleration; determining a standard key point state vector based on the standard key point coordinates and standard key point stress value; determining a standard rotation state vector based on the standard rotation angle and standard rotation acceleration; and determining a standard rotation state vector based on the formula... Determine the real-time comprehensive safety factor at the i-th moment of the detection cycle. Where min is the function for finding the minimum value. Let K be the coordinates of the k-th keypoint at the i-th time in the detection cycle. Let the stress value of the k-th key point be the stress value at the i-th moment of the detection cycle. Let k be the keypoint state vector at the i-th time of the detection period. for The transpose of , Let the coordinates of the k-th keypoint be the standard keypoint coordinates at the i-th time in the detection cycle. Let the stress value of the k-th key point be the standard key point stress value at the i-th moment of the detection cycle. Let K be the standard keypoint state vector of the k-th keypoint at the i-th time in the detection cycle. The key point priority for the k-th key point. This refers to the real-time rotation angle at the i-th moment of the detection cycle. To detect the real-time rotational acceleration at the i-th moment of the detection period, Let be the rotational state vector at the i-th moment of the detection period. for The transpose of , Let be the standard rotation angle at the i-th moment of the detection cycle. Let be the standard rotational acceleration at the i-th moment of the detection period. Let K be the standard rotation state vector at the i-th moment of the detection cycle, K be the number of key points, k ≤ K, and both k and K are positive integers.

5. The bridge rotation process detection method based on three-dimensional visualization data according to claim 1, characterized in that, The training steps of the bridge rotation risk prediction model include: acquiring historical bridge rotation construction records for multiple historical construction periods; determining historical sensor data and historical environmental data based on the historical bridge rotation construction records; determining historical environmental safety factors, historical driving safety factors, and historical comprehensive safety factors based on the historical sensor data and historical environmental data; processing the historical environmental safety factors, historical driving safety factors, and historical comprehensive safety factors through the bridge rotation risk prediction model to determine sample bridge rotation risk prediction coefficients; and determining historical construction alarm information based on the historical bridge rotation construction records. Based on the historical construction alarm information, determine the time difference of the historical alarm information, the historical alarm level, and the historical construction delay time; based on the time difference of the historical alarm information, the historical alarm level, and the historical construction delay time, determine the historical bridge rotation risk coefficient; based on the historical bridge rotation risk coefficient, the sample bridge rotation risk prediction coefficient, the historical environmental safety coefficient, the historical driving safety coefficient, and the historical comprehensive safety coefficient, determine the training loss function of the bridge rotation risk prediction model; train the bridge rotation risk prediction model based on the training loss function to obtain the trained bridge rotation risk prediction model.

6. The bridge rotation process detection method based on three-dimensional visualization data according to claim 5, characterized in that, Based on the historical bridge rotation risk coefficient, the sample bridge rotation risk prediction coefficient, the historical environmental safety coefficient, the historical driving safety coefficient, and the historical comprehensive safety coefficient, the training loss function of the bridge rotation risk prediction model is determined, including: according to the formula Determine the training loss function for the bridge rotation risk prediction model. ,in, Let be the sample bridge rotation risk prediction coefficient at the r-th moment of the e-th historical construction cycle. The historical bridge rotation risk coefficient is a preset time period after the r-th moment of the e-th historical construction cycle. Let r be the historical environmental safety factor at the r-th moment of the e-th historical construction cycle. To preset the environmental safety factor threshold, Let the historical driving safety factor be the value at the r-th moment of the e-th historical construction cycle. To preset the drive safety factor threshold, Let r be the historical comprehensive safety factor at the r-th moment of the e-th historical construction cycle. The threshold for the preset comprehensive safety factor is defined as follows: E is the number of historical construction cycles, e≤E, R is the number of moments in the historical construction cycles, r≤R, and e, E, r, and R are all positive integers.

7. A bridge rotation process detection system based on three-dimensional visualization data, used to implement the bridge rotation process detection method based on three-dimensional visualization data as described in any one of claims 1-6, characterized in that, include: The system includes a real-time data module for acquiring real-time sensor data and real-time environmental data at multiple points during the detection cycle. The real-time sensor data includes real-time positioning data, real-time attitude data, real-time stress data, and real-time drive system data. The real-time environmental data includes real-time wind speed data and real-time temperature data. An environmental coefficient module is used to determine a real-time environmental safety factor based on the real-time environmental data. A sensor coefficient module is used to determine a real-time drive safety factor and a real-time comprehensive safety factor based on the real-time sensor data. A risk prediction module is used to process the real-time environmental safety factor, the real-time drive safety factor, and the real-time comprehensive safety factor using a trained bridge rotation risk prediction model to determine a bridge rotation risk prediction coefficient. A detection report module is used to generate a bridge rotation process detection report based on the bridge rotation risk prediction coefficient, the real-time environmental safety factor, the real-time drive safety factor, and the real-time comprehensive safety factor.

Images