Bridge structure anti-deformation monitoring method and system

Abstract

The invention provides a bridge structure anti-deformation monitoring method and system, and relates to the technical field of bridge deformation monitoring. The method comprises the following steps: dividing into a plurality of segmented bridge areas, and setting horizontal test points and vertical test points; determining a first distance and a first yaw angle, and a second distance and a second pitch angle; determining a horizontal anti-deformation coefficient; determining a vertical anti-deformation coefficient; obtaining segmented bridge images; determining actually measured similarity data; acquiring environment data; inputting the environment data and the segmented bridge image at the starting moment of the current monitoring period into a trained similarity prediction model to obtain predicted similarity data of a plurality of segmented bridge areas in the current monitoring period; determining a surface anti-deformation coefficient; and determining the anti-deformation capability of the bridge structure. According to the invention, non-contact monitoring can be carried out, the anti-deformation capability of the bridge structure can be comprehensively evaluated in the horizontal, vertical and environmental directions, the comprehensiveness and accuracy of evaluating the anti-deformation capability of the bridge structure are improved, and the monitoring cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge deformation monitoring, and in particular, to a method and system for monitoring the anti-deformation of a bridge structure. Background Art

[0002] In the current related technologies, although deformation monitoring can be carried out, horizontal test points and vertical test points are not set to evaluate the anti-deformation ability in the horizontal and vertical directions, and the influence of the environment on the deformation of the bridge structure is not considered either. That is, it is impossible to comprehensively evaluate the anti-deformation ability of the bridge structure from the three directions of horizontal, vertical and environment.

[0003] The information disclosed in the background art part of this application is only intended to deepen the understanding of the general background art of this application, and should not be regarded as an admission or any form of suggestion that this information constitutes the prior art known to those skilled in the art. Summary of the Invention

[0004] The present invention provides a method and system for monitoring the anti-deformation of a bridge structure, which can solve the technical problem that the related technologies cannot comprehensively evaluate the anti-deformation ability of the bridge structure from the three directions of horizontal, vertical and environment.

[0005] According to a first aspect of the present invention, a method for monitoring the anti-deformation of a bridge structure is provided, including: dividing the entire bridge into multiple segmented bridge areas, setting a plurality of horizontal test points at the side positions of the bridge decks in the segmented bridge areas, and setting a plurality of vertical test points at the side positions of the bridge piers in the segmented bridge areas, wherein there is only one bridge pier in the segmented bridge area, the horizontal test points are on the same horizontal plane, and the vertical test points are on the same vertical line; setting a laser rangefinder at a first preset position in the segmented bridge area, making the laser emitted by the laser rangefinder irradiate on the horizontal test points and the vertical test points, and determining a first distance and a first yaw angle of the laser rangefinder when measuring the plurality of horizontal test points, and a second distance and a second pitch angle of the laser rangefinder when measuring the plurality of vertical test points; at the start and end moments of the current monitoring period, determining a horizontal anti-deformation coefficient according to the first distance and the first yaw angle; at the start and end moments of the current monitoring period, determining a vertical anti-deformation coefficient according to the second distance and the second pitch angle; acquiring segmented bridge images of the plurality of segmented bridge areas at the start and end moments of the current monitoring period; determining measured similarity data according to the segmented bridge images; acquiring environmental data of the current monitoring period, wherein the environmental data includes temperature data, traffic load data and wind speed data; inputting the environmental data and the segmented bridge images at the start moment of the current monitoring period into a trained similarity prediction model to obtain predicted similarity data of the plurality of segmented bridge areas in the current monitoring period; determining a surface anti-deformation coefficient according to the predicted similarity data and the measured similarity data; determining the anti-deformation ability of the bridge structure in the current monitoring period according to the horizontal anti-deformation coefficient, the vertical anti-deformation coefficient and the surface anti-deformation coefficient.

[0006] Further, at the start and end moments of the current monitoring period, determining the horizontal anti-deformation coefficient according to the first distance and the first yaw angle includes: obtaining a start maximum first distance according to the first distance at the start moment of the current monitoring period; obtaining an end maximum first distance according to the first distance at the end moment of the current monitoring period; determining the horizontal anti-deformation coefficient according to the start maximum first distance, the end maximum first distance, the first distance and the first yaw angle.

[0007] Further, determining the horizontal anti-deformation coefficient according to the start maximum first distance, the end maximum first distance, the first distance and the first yaw angle includes: according to the formula Determine the horizontal anti-deformation coefficient of the e-th segmented bridge area , wherein, Is the start maximum first distance of the e-th segmented bridge area, Is the end maximum first distance of the e-th segmented bridge area, is the first distance of the i-th horizontal test point in the e-th segmented bridge area at the start time of the current monitoring period, is the first distance of the i-th horizontal test point in the e-th segmented bridge area at the end time of the current monitoring period, is the first yaw angle of the i-th horizontal test point in the e-th segmented bridge area at the start time of the current monitoring period, is the first yaw angle of the i-th horizontal test point in the e-th segmented bridge area at the end time of the current monitoring period, n is the number of horizontal test points, max is the maximum value function, i ≤ n, and e, i, and n are all positive integers.

[0008] Further, at the start time and the end time of the current monitoring period, according to the second distance and the second pitch angle, determining the vertical anti-deformation coefficient includes: obtaining the start maximum second distance according to the second distance at the start time of the current monitoring period; obtaining the end maximum second distance according to the second distance at the end time of the current monitoring period; obtaining the pier height; determining the vertical anti-deformation coefficient according to the pier height, the start maximum second distance, the end maximum second distance, the second distance, and the second pitch angle.

[0009] Further, determining the vertical anti-deformation coefficient according to the pier height, the start maximum second distance, the end maximum second distance, the second distance, and the second pitch angle includes: according to the formula determine the vertical anti-deformation coefficient of the e-th segmented bridge area , where is the start maximum second distance of the e-th segmented bridge area, is the end maximum second distance of the e-th segmented bridge area, is the second distance of the j-th vertical test point in the e-th segmented bridge area at the start time of the current monitoring period, is the second distance of the j-th vertical test point in the e-th segmented bridge area at the end time of the current monitoring period, is the second pitch angle of the j-th vertical test point in the e-th segmented bridge area at the start time of the current monitoring period, is the second pitch angle of the j-th vertical test point in the e-th segmented bridge area at the end time of the current monitoring period, is the pier height of the e-th segmented bridge area, m is the number of vertical test points, max is the maximum value function, j ≤ m, and e, j, and m are all positive integers.

[0010] Further, based on the segmented bridge image, measured similarity data is determined, including: through the trained image recognition neural network model, performing feature extraction processing on the segmented bridge image at the start moment of the current monitoring period to obtain start segmented bridge feature vectors of multiple segmented bridge regions; through the trained image recognition neural network model, performing feature extraction processing on the segmented bridge image at the end moment of the current monitoring period to obtain end segmented bridge feature vectors of multiple segmented bridge regions; determining the similarity between the start segmented bridge feature vector and the end segmented bridge feature vector of the e-th segmented bridge region as the measured similarity data of the e-th segmented bridge region.

[0011] Further, the training steps of the similarity prediction model include: obtaining historical segmented bridge images of multiple historical segmented bridge regions at the start and end moments of multiple historical monitoring periods; determining historical measured similarity data based on the historical segmented bridge images; obtaining the third distance from the midpoint position of the pier of each historical segmented bridge region to the midpoint position of the entire bridge; obtaining historical environmental data of multiple historical segmented bridge regions in multiple historical monitoring periods, where the historical environmental data includes historical temperature data, historical traffic load data, and historical wind speed data; through the similarity prediction model, processing the historical environmental data and the historical segmented bridge image at the start moment of the historical monitoring period to obtain historical predicted similarity data of multiple historical segmented bridge regions in multiple historical monitoring periods; determining the loss function of the similarity prediction model based on the third distance, the historical environmental data, the historical measured similarity data, and the historical predicted similarity data; training the similarity prediction model according to the loss function of the similarity prediction model to obtain the trained similarity prediction model.

[0012] Further, determining the loss function of the similarity prediction model based on the third distance, the historical environmental data, the historical measured similarity data, and the historical predicted similarity data includes: according to the formula determining the loss function Loss of the similarity prediction model, where is the historical measured similarity data of the y-th historical segmented bridge region in the h-th historical monitoring period, is the historical predicted similarity data of the y-th historical segmented bridge region in the h-th historical monitoring period, is the third distance from the midpoint position of the pier of the y-th historical segmented bridge region to the midpoint position of the entire bridge, is the third distance from the midpoint position of the pier of the first historical segmented bridge region to the midpoint position of the entire bridge, is the historical temperature data of the y-th historical segmented bridge region in the h-th historical monitoring period, is the standard temperature data, is the historical traffic load data of the y-th historical segmented bridge area in the h-th historical monitoring period, is the standard traffic load data, is the historical wind speed data of the y-th historical segmented bridge area in the h-th historical monitoring period, is the standard wind speed data, H is the number of historical monitoring periods, is the number of historical segmented bridge areas in the x-th training batch, N is the number of training batches, h ≤ H, y ≤ , x ≤ N, and h, x, y, H, and N are all positive integers.

[0013] Furthermore, according to the predicted similarity data and the measured similarity data, determine the surface anti-deformation coefficient, including: according to the formula determine the surface anti-deformation coefficient , where is the measured similarity data of the e-th segmented bridge area, is the predicted similarity data of the e-th segmented bridge area, E is the number of segmented bridge areas, e ≤ E, and e and E are all positive integers.

[0014] According to a second aspect of the present invention, there is provided a bridge structure anti-deformation monitoring system, comprising: a segmented bridge area module for dividing the entire bridge into a plurality of segmented bridge areas, arranging a plurality of horizontal test points at the side positions of the bridge deck in the segmented bridge areas, and arranging a plurality of vertical test points at the side positions of the bridge piers in the segmented bridge areas, wherein there is only one bridge pier in the segmented bridge area, the horizontal test points are on the same horizontal plane, and the vertical test points are on the same vertical line; a measurement module for arranging a laser rangefinder at a first preset position in the segmented bridge area, making the laser emitted by the laser rangefinder irradiate on the horizontal test points and the vertical test points, and determining a first distance and a first yaw angle of the laser rangefinder when measuring the plurality of horizontal test points, and a second distance and a second pitch angle of the laser rangefinder when measuring the plurality of vertical test points; a horizontal anti-deformation coefficient module for determining a horizontal anti-deformation coefficient according to the first distance and the first yaw angle at the start and end times of the current monitoring period; a vertical anti-deformation coefficient module for determining a vertical anti-deformation coefficient according to the second distance and the second pitch angle at the start and end times of the current monitoring period; a segmented bridge image module for acquiring segmented bridge images of the plurality of segmented bridge areas at the start and end times of the current monitoring period; an actual similarity data module for determining actual similarity data according to the segmented bridge images; an environmental data module for acquiring environmental data of the current monitoring period, wherein the environmental data includes temperature data, traffic load data and wind speed data; a predicted similarity data module for inputting the environmental data and the segmented bridge image at the start time of the current monitoring period into a trained similarity prediction model to obtain predicted similarity data of the plurality of segmented bridge areas in the current monitoring period; a surface anti-deformation coefficient module for determining a surface anti-deformation coefficient according to the predicted similarity data and the actual similarity data; and an anti-deformation ability evaluation module for determining the anti-deformation ability of the bridge structure in the current monitoring period according to the horizontal anti-deformation coefficient, the vertical anti-deformation coefficient and the surface anti-deformation coefficient.

[0015] Technical effects: According to the present invention, by setting horizontal test points and vertical test points and using a laser rangefinder for measurement, it helps to understand the deformation conditions of the bridge structure in the horizontal and vertical directions. By combining the influence of environmental data on the deformation of the bridge structure with a similarity prediction model, the deformation trend of the bridge can be predicted more accurately. Through non-contact monitoring, the anti-deformation ability of the bridge structure is comprehensively evaluated from three directions: horizontal, vertical, and environmental, improving the comprehensiveness and accuracy of evaluating the anti-deformation ability of the bridge structure and reducing the monitoring cost. When determining the horizontal anti-deformation coefficient, it can be based on the relative difference between the first yaw angle at the end and the start of the current monitoring period, and based on the characteristic that the farther the horizontal test point is from the first preset position, the less obvious the change in the first yaw angle, weights are set, so as to perform weighted averaging on the relative differences between the first yaw angles at the end and the start of the current monitoring period for multiple horizontal test points to obtain the horizontal anti-deformation coefficient, improving the accuracy of identifying the anti-lateral deformation ability of the bridge deck. When determining the vertical anti-deformation coefficient, it can be based on the ratio of the vertical displacement of the vertical test point to the height of the pier, and based on the characteristic that the farther the vertical test point is from the first preset position, the greater the vertical displacement corresponding to the same pitch angle change, weights are set, so as to perform weighted averaging on the ratios of the vertical displacements of multiple vertical test points to the height of the pier to obtain the vertical anti-deformation coefficient, improving the accuracy of identifying the anti-vertical deformation ability of the pier. When determining the loss function of the similarity prediction model, it can be determined through the influence of historical temperature data, historical traffic load data, and historical wind speed data on the deformation of the bridge, so as to determine the influence of the above data on the error of historical prediction similarity data, and then based on this influence and the relative difference between the historical measured similarity data and the historical prediction similarity data, weights are set based on the characteristic that the closer the historical segmented bridge area is to the midpoint position of the entire bridge, the greater the influence on the error of the historical prediction similarity data, and weights are set based on the characteristic that the lower the accuracy is with a shorter time interval from the first batch, so as to perform weighted summation on the errors of the similarity prediction model outputs of multiple historical segmented bridge areas in multiple historical monitoring periods for each training batch to obtain the loss function, so as to improve the design accuracy and objectivity of the loss function, thereby improving the training efficiency during the training process and the accuracy of the similarity prediction model. When determining the surface anti-deformation coefficient, the surface anti-deformation coefficient can be determined through predicted similarity data and measured similarity data. By comparing the actual deformation amount of the bridge with the predicted deformation amount, it can reflect the strength of the surface anti-deformation ability of the bridge under the influence of environmental factors, improving the reliability and scientificity of the surface anti-deformation coefficient.

[0016] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and do not limit the present invention. According to the following detailed description of the exemplary embodiments with reference to the accompanying drawings, other features and aspects of the present invention will become clearer. Brief Description of the Drawings

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other embodiments can be obtained based on these drawings; Figure 1 Exemplarily shown is a schematic flowchart of a method for monitoring the anti-deformation of a bridge structure according to an embodiment of the present invention; Figure 2 Exemplarily shown is a flowchart for calculating the horizontal anti-deformation coefficient according to an embodiment of the present invention; Figure 3 Exemplarily shown is a flowchart for calculating the vertical anti-deformation coefficient according to an embodiment of the present invention; Figure 4 Exemplarily shown is a flowchart for calculating the measured similarity data according to an embodiment of the present invention; Figure 5 Exemplarily shown is a block diagram of a bridge structure anti-deformation monitoring system according to an embodiment of the present invention. Detailed implementation manners

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, rather than all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0019] The following will specifically describe the technical solutions of the present invention with specific embodiments. These several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0020] Figure 1A schematic flowchart of a method for monitoring the anti-deformation of a bridge structure according to an embodiment of the present invention is exemplarily shown. The method includes: Step S1, dividing the entire bridge into multiple segmented bridge areas, setting a plurality of horizontal test points at the side positions of the bridge decks in the segmented bridge areas, and setting a plurality of vertical test points at the side positions of the bridge piers in the segmented bridge areas. Among them, there is only one bridge pier in the segmented bridge area, the horizontal test points are on the same horizontal plane, and the vertical test points are on the same vertical line; Step S2, setting a laser rangefinder at a first preset position in the segmented bridge area, making the laser emitted by the laser rangefinder irradiate on the horizontal test points and the vertical test points, and determining the first distance and the first yaw angle of the laser rangefinder when measuring the plurality of horizontal test points, and the second distance and the second pitch angle of the laser rangefinder when measuring the plurality of vertical test points; Step S3, at the start time and the end time of the current monitoring period, determining the horizontal anti-deformation coefficient according to the first distance and the first yaw angle; Step S4, at the start time and the end time of the current monitoring period, determining the vertical anti-deformation coefficient according to the second distance and the second pitch angle; Step S5, obtaining segmented bridge images of the plurality of segmented bridge areas at the start time and the end time of the current monitoring period; Step S6, determining the measured similarity data according to the segmented bridge images; Step S7, obtaining the environmental data of the current monitoring period, where the environmental data includes temperature data, traffic load data, and wind speed data; Step S8, inputting the environmental data and the segmented bridge image at the start time of the current monitoring period into the trained similarity prediction model to obtain the predicted similarity data of the plurality of segmented bridge areas in the current monitoring period; Step S9, determining the surface anti-deformation coefficient according to the predicted similarity data and the measured similarity data; Step S10, determining the anti-deformation ability of the bridge structure in the current monitoring period according to the horizontal anti-deformation coefficient, the vertical anti-deformation coefficient, and the surface anti-deformation coefficient.

[0021] For the method for monitoring the anti-deformation of a bridge structure according to the embodiment of the present invention, by setting horizontal test points and vertical test points and using a laser rangefinder for measurement, it helps to understand the deformation conditions of the bridge structure in the horizontal and vertical directions. By combining the influence of environmental data on the deformation of the bridge structure with the similarity prediction model, the deformation trend of the bridge can be predicted more accurately. Through non-contact monitoring, the anti-deformation ability of the bridge structure is comprehensively evaluated from three directions: horizontal, vertical, and environmental, improving the comprehensiveness and accuracy of evaluating the anti-deformation ability of the bridge structure and reducing the monitoring cost.

[0022] According to an embodiment of the present invention, in step S1, the entire bridge is a bridge built on land. Each segmented bridge area has only one pier, that is, multiple segmented bridge areas are divided along the longitudinal direction (the bridge length direction) according to the piers. Each segmented bridge area includes a pier, a bridge deck, and infrastructure (such as bearings, expansion joints). A plurality of horizontal test points are arranged at the side position of the bridge deck in each segmented bridge area (such as the web or flange of the box girder). All the horizontal test points must be located on the same horizontal plane, that is, at the same height, to monitor the displacement of the bridge deck in the horizontal direction (such as lateral deformation or torsional deformation). At the same time, a plurality of vertical test points are arranged at the side position of the pier in each segmented bridge area (such as the windward side or the main direction of force). Among them, the side position of the bridge deck and the side position of the pier are on the same side of the bridge. All the vertical test points must be located on the same vertical line, that is, on the same axis along the height direction of the pier, to monitor the displacement of the pier in the vertical direction (such as the settlement or upheaval of the pier).

[0023] According to an embodiment of the present invention, in step S2, within the defined segmented bridge area, the first preset position is on the ground of the hard base beside the pier in the segmented bridge area (which will not settle), and is used to install a laser rangefinder. Necessary angular adjustments are made so that the emitted laser can irradiate all the horizontal test points and all the vertical test points, thereby measuring the first distance between the first preset position and the horizontal test points, recording the first yaw angle of the laser rangefinder, and the second distance between the first preset position and the vertical test points, recording the second pitch angle of the laser rangefinder. The yaw angle is used to describe the angle at which the laser rangefinder rotates left or right, and the pitch angle is used to describe the angle at which the laser rangefinder rotates up or down.

[0024] According to an embodiment of the present invention, in step S3, the interval between adjacent moments can be set to 12 hours, 24 hours, etc., and each monitoring period can be set to 3 days, 5 days, etc. The present invention does not limit this. At the start moment of the current monitoring period, the first distance and first yaw angle data of each horizontal test point are measured, and at the end moment of the current monitoring period, the first distance and first yaw angle data of each horizontal test point are measured. The horizontal anti-deformation coefficient reflects the displacement of the bridge deck in the horizontal direction. The first preset position in each segmented bridge area during the current monitoring period remains unchanged.

[0025] Figure 2 Exemplarily shows a flowchart for calculating the horizontal anti-deformation coefficient according to an embodiment of the present invention.

[0026] According to an embodiment of the present invention, step S3 includes: step S31, obtaining the starting maximum first distance according to the first distance at the start time of the current monitoring period; step S32, obtaining the ending maximum first distance according to the first distance at the end time of the current monitoring period; step S33, determining the horizontal anti-deformation coefficient according to the starting maximum first distance, the ending maximum first distance, the first distance, and the first yaw angle.

[0027] According to an embodiment of the present invention, at the start time of the current monitoring period, the first distances measured at each horizontal test point are compared to obtain the starting maximum first distance. Similarly, the ending maximum first distance is obtained. According to the distances of each horizontal test point from the first preset position (the starting maximum first distance, the ending maximum first distance, and the first distance) and the change in the first yaw angle, the horizontally calculated anti-deformation coefficient can objectively and accurately reflect the horizontal anti-deformation ability of the bridge deck, providing an important basis for the safety assessment and maintenance of the bridge.

[0028] According to an embodiment of the present invention, determining the horizontal anti-deformation coefficient according to the starting maximum first distance, the ending maximum first distance, the first distance, and the first yaw angle includes: determining the horizontal anti-deformation coefficient of the e-th segmented bridge area according to formula (1) , (1), where is the starting maximum first distance of the e-th segmented bridge area, is the ending maximum first distance of the e-th segmented bridge area, is the first distance of the i-th horizontal test point in the e-th segmented bridge area at the start time of the current monitoring period, is the first distance of the i-th horizontal test point in the e-th segmented bridge area at the end time of the current monitoring period, is the first yaw angle of the i-th horizontal test point in the e-th segmented bridge area at the start time of the current monitoring period, is the first yaw angle of the i-th horizontal test point in the e-th segmented bridge area at the end time of the current monitoring period, n is the number of horizontal test points, max is the maximum value function, i ≤ n, and e, i, and n are all positive integers.

[0029] According to an embodiment of the present invention, in formula (1), is the relative difference between the first yaw angle of the i-th horizontal test point in the e-th segmented bridge area at the end time of the current monitoring period and the first yaw angle of the i-th horizontal test point in the e-th segmented bridge area at the start time of the current monitoring period. The larger this relative difference, the greater the lateral displacement of the bridge deck and the weaker the anti-deformation ability of the bridge deck. is the ratio of the first distance of the i-th horizontal test point in the e-th segmented bridge area at the start time of the current monitoring period to the maximum first distance at the start of the e-th segmented bridge area, that is, the normalized start first distance ratio. is the ratio of the first distance of the i-th horizontal test point in the e-th segmented bridge area at the end time of the current monitoring period to the maximum first distance at the end of the e-th segmented bridge area, that is, the normalized end first distance ratio. is to take the maximum value of the normalized start first distance ratio and the normalized end first distance ratio. The result of 1 minus this maximum value can be obtained. The smaller this result is, the larger the first distance of the i-th horizontal test point in the e-th segmented bridge area is, indicating that the horizontal test point is farther from the first preset position. When the horizontal test point undergoes displacement, the change in the first yaw angle is less obvious, and the ability of this measurement result to characterize the transverse deformation of the bridge deck is weaker. That is, the position change of the horizontal test point far from the first preset position cannot significantly reflect the overall deformation trend of the bridge deck. Therefore, a lower weight is assigned. Using performs a weighted average on the relative difference between the first yaw angles of multiple horizontal test points in the e-th segmented bridge area at the end time and the start time of the current monitoring period, and subtracts this average value from 1 to obtain the horizontal anti-deformation coefficient of the e-th segmented bridge area. The larger this horizontal anti-deformation coefficient is, the less the transverse deformation of the bridge deck occurs, and the greater the anti-deformation ability.

[0030] In this way, based on the relative difference between the first yaw angles at the end time and the start time of the current monitoring period, and based on the characteristic that the farther the horizontal test point is from the first preset position, the less obvious the change in the first yaw angle, weights are set, so as to perform a weighted average on the relative difference between the first yaw angles of multiple horizontal test points at the end time and the start time of the current monitoring period, obtain the horizontal anti-deformation coefficient, and improve the accuracy of identifying the anti-transverse deformation ability of the bridge deck.

[0031] According to an embodiment of the present invention, in step S4, the second distance and the second pitch angle data of each vertical test point are measured at the start time of the current monitoring period, and the second distance and the second pitch angle data of each vertical test point are measured at the end time of the current monitoring period. The vertical anti-deformation coefficient reflects the displacement of the pier in the vertical direction.

[0032] Figure 3 Exemplarily shows a flowchart for calculating the vertical anti-deformation coefficient according to an embodiment of the present invention.

[0033] According to an embodiment of the present invention, step S4 includes: step S41, obtaining the starting maximum second distance according to the second distance at the start time of the current monitoring period; step S42, obtaining the ending maximum second distance according to the second distance at the end time of the current monitoring period; step S43, obtaining the pier height; step S44, determining the vertical anti-deformation coefficient according to the pier height, the starting maximum second distance, the ending maximum second distance, the second distance, and the second pitch angle.

[0034] According to an embodiment of the present invention, at the start time of the current monitoring period, the first distances measured at each vertical test point are compared to obtain the starting maximum second distance. Similarly, the ending maximum second distance is obtained. The pier height can be obtained through the bridge design parameters. According to the distances of each vertical test point from the first preset position (the starting maximum second distance, the ending maximum second distance, and the second distance) and the change in vertical displacement, the calculated vertical anti-deformation coefficient can objectively and accurately reflect the anti-deformation ability of the pier in the vertical direction, providing an important basis for the safety assessment and maintenance of the bridge.

[0035] According to an embodiment of the present invention, determining the vertical anti-deformation coefficient according to the pier height, the starting maximum second distance, the ending maximum second distance, the second distance, and the second pitch angle includes: determining the vertical anti-deformation coefficient of the e-th segmented bridge area according to formula (2) , (2), where is the starting maximum second distance of the e-th segmented bridge area, is the ending maximum second distance of the e-th segmented bridge area, is the second distance of the j-th vertical test point in the e-th segmented bridge area at the start time of the current monitoring period, is the second distance of the j-th vertical test point in the e-th segmented bridge area at the end time of the current monitoring period, is the second pitch angle of the j-th vertical test point in the e-th segmented bridge area at the start time of the current monitoring period, is the second pitch angle of the j-th vertical test point in the e-th segmented bridge area at the end time of the current monitoring period, is the pier height of the e-th segmented bridge area, m is the number of vertical test points, max is the maximum value function, j ≤ m, and e, j, and m are all positive integers.

[0036] According to an embodiment of the present invention, in formula (2), is the vertical displacement of the j-th vertical test point in the e-th segmented bridge area between the end time and the start time of the current monitoring period, is the ratio of the vertical displacement of the j-th vertical test point in the e-th segmented bridge area to the pier height of the e-th segmented bridge area. The larger this ratio, the greater the vertical displacement of the pier and the weaker the pier's anti-deformation ability. is the ratio of the second distance of the j-th vertical test point in the e-th segmented bridge area at the start moment of the current monitoring period to the maximum second distance at the start of the e-th segmented bridge area, that is, the normalized start second distance ratio. is the ratio of the second distance of the j-th vertical test point in the e-th segmented bridge area at the end moment of the current monitoring period to the maximum second distance at the end of the e-th segmented bridge area, that is, the normalized end second distance ratio. is to take the maximum value between the normalized start second distance ratio and the normalized end second distance ratio, and the result of 1 minus this maximum value can be obtained. The larger this result, the greater the second distance of the j-th vertical test point in the e-th segmented bridge area, indicating that the vertical test point is farther from the first preset position, and the greater the vertical displacement corresponding to the same pitch angle change. Therefore, a higher weight is assigned. Using perform a weighted average on the ratio of the vertical displacement of multiple vertical test points in the e-th segmented bridge area to the pier height of the e-th segmented bridge area, and subtract this average value from 1 to obtain the vertical anti-deformation coefficient of the e-th segmented bridge area. The larger this vertical anti-deformation coefficient, the less the vertical deformation of the pier and the greater the anti-deformation ability.

[0037] In this way, based on the ratio of the vertical displacement of the vertical test point to the pier height, and based on the characteristic that the farther the vertical test point is from the first preset position, the greater the vertical displacement corresponding to the same pitch angle change, weights can be set, so as to perform a weighted average on the ratio of the vertical displacement of multiple vertical test points to the pier height, obtain the vertical anti-deformation coefficient, and improve the accuracy of identifying the pier's anti-vertical deformation ability.

[0038] According to an embodiment of the present invention, in step S5, segmented bridge images of multiple segmented bridge areas at the start and end moments of the current monitoring period can be obtained through a high-definition camera.

[0039] According to an embodiment of the present invention, in step S6, based on the segmented bridge images, measured similarity data is determined.

[0040] Figure 4 Exemplarily shows a flowchart of calculating measured similarity data according to an embodiment of the present invention.

[0041] According to an embodiment of the present invention, step S6 includes: step S61, through a trained image recognition neural network model, which can be a convolutional neural network model, such as a resnet model. The present invention does not limit the specific type of the image recognition neural network model. Feature extraction processing is performed on the segmented bridge image at the start moment of the current monitoring period to obtain start segmented bridge feature vectors of multiple segmented bridge regions; step S62, through the trained image recognition neural network model, feature extraction processing is performed on the segmented bridge image at the end moment of the current monitoring period to obtain end segmented bridge feature vectors of multiple segmented bridge regions; step S63, the similarity between the start segmented bridge feature vector and the end segmented bridge feature vector of the e-th segmented bridge region is determined as the measured similarity data of the e-th segmented bridge region.

[0042] According to an embodiment of the present invention, according to the calculation formula of the measured similarity data , where is the measured similarity data of the e-th segmented bridge region, is the start segmented bridge feature vector of the e-th segmented bridge region, is the end segmented bridge feature vector of the e-th segmented bridge region, is the transposed vector of. The larger the measured similarity data, the more similar the segmented bridge images at the start and end moments of the current monitoring period, that is, the less likely the segmented bridge region is to undergo surface deformation (such as tilting, crack expansion).

[0043] According to an embodiment of the present invention, in step S7, temperature helps to understand the environmental thermal conditions and reflects the deformation caused by the thermal expansion and contraction of the bridge. Traffic load can reflect the deformation caused by the traffic pressure borne by the bridge structure, and wind speed can reflect the deformation generated by the bridge structure under the action of wind. For temperature acquisition, a high-precision temperature sensor can be selected. The difference between the maximum and minimum values of the environmental temperature collected at multiple moments in the current monitoring period and the standard temperature (such as 20 °C) that is relatively large is used as the temperature data. For traffic load acquisition, a dynamic weighing system or devices such as strain gauges installed at key parts of the bridge can be used to collect the traffic load on the entire bridge at the current moment, and the maximum value of the traffic load collected at multiple moments in the current monitoring period is used as the traffic load data. For wind speed acquisition, professional instruments such as anemometers can be used, and the maximum value of the wind speed collected at multiple moments in the current monitoring period is used as the wind speed data.

[0044] According to an embodiment of the present invention, in step S8, the environmental data and the segmented bridge image at the start time of the current monitoring period are input into the trained similarity prediction model, which can be a convolutional neural network model, trained based on a large amount of historical data through machine learning or deep learning techniques, and can analyze the predicted similarity data of multiple segmented bridge areas in the current monitoring period.

[0045] According to an embodiment of the present invention, the training steps of the similarity prediction model include: obtaining historical segmented bridge images of multiple historical segmented bridge areas at the start time and end time of multiple historical monitoring periods; determining historical measured similarity data according to the historical segmented bridge images; obtaining the third distance from the midpoint position of the pier of each historical segmented bridge area to the midpoint position of the entire bridge; obtaining historical environmental data of multiple historical segmented bridge areas in multiple historical monitoring periods, where the historical environmental data includes historical temperature data, historical traffic load data, and historical wind speed data; processing the historical environmental data and the historical segmented bridge image at the start time of the historical monitoring period through the similarity prediction model to obtain historical predicted similarity data of multiple historical segmented bridge areas in multiple historical monitoring periods; determining the loss function of the similarity prediction model according to the third distance, the historical environmental data, the historical measured similarity data, and the historical predicted similarity data; training the similarity prediction model according to the loss function of the similarity prediction model to obtain the trained similarity prediction model.

[0046] According to an embodiment of the present invention, historical segmented bridge images can be obtained through a high-definition camera. The acquisition method of historical measured similarity data is similar to that of measured similarity data, which will not be elaborated here. The midpoint position of the piers in each historical segmented bridge area is the geometric center point of the pier (usually the central coordinates at the bottom or top of the pier), and the midpoint position of the entire bridge is the global geometric center of the bridge structure (the symmetric center point of the main girder). The third distance from the midpoint position of the piers in each historical segmented bridge area to the midpoint position of the entire bridge can be obtained through design drawings or GPS. The acquisition method of historical environmental data is similar to that of environmental data, which will not be elaborated here. When the temperature is high, the bridge structure will expand; when the temperature is low, the bridge structure will contract. This thermal expansion and contraction phenomenon will cause the bridge to deform, that is, the higher or lower the temperature, the greater the deformation of the bridge. When a vehicle travels on the bridge, it will exert pressure on the bridge structure. The greater the traffic load, the greater the deformation of the bridge. When the wind speed is relatively high, it will exert wind load on the bridge, triggering wind-induced vibration of the bridge, that is, the greater the wind speed, the greater the deformation of the bridge. The greater the deformation of the bridge indicates that the segmented bridge images at the start and end times are less similar. The similarity prediction model can predict the historical prediction similarity data of multiple historical segmented bridge areas in multiple historical monitoring periods based on the above relationships between temperature, traffic load, wind speed and the deformation of the bridge, based on the third distance and historical environmental data. Determine the loss function according to the relative difference between the historical measured similarity data and the historical prediction similarity data. Divide the multiple historical segmented bridge areas into different training batch historical segmented bridge areas, and the number of historical segmented bridge areas in each training batch is the same, so as to train the similarity prediction model for different training batches. For example, , where E is the number of historical segmented bridge areas, , , …, are the numbers of historical segmented bridge areas in the 1st, 2nd, …, Nth training batches respectively. Then, through feedback adjustment of the loss function, the trained similarity prediction model is obtained.

[0047] According to an embodiment of the present invention, determining the loss function of the similarity prediction model according to the third distance, historical environmental data, the historical measured similarity data and the historical prediction similarity data includes: determining the loss function Loss of the similarity prediction model according to formula (3), (3), where, is the historical measured similarity data of the yth historical segmented bridge area in the hth historical monitoring period, is the historical prediction similarity data of the yth historical segmented bridge area in the hth historical monitoring period, is the third distance from the midpoint position of the pier in the y-th historical segmented bridge area to the midpoint position of the entire bridge, is the third distance from the midpoint position of the pier in the 1st historical segmented bridge area to the midpoint position of the entire bridge, is the historical temperature data of the y-th historical segmented bridge area in the h-th historical monitoring period, is the standard temperature data, is the historical traffic load data of the y-th historical segmented bridge area in the h-th historical monitoring period, is the standard traffic load data, is the historical wind speed data of the y-th historical segmented bridge area in the h-th historical monitoring period, is the standard wind speed data, H is the number of historical monitoring periods, is the number of historical segmented bridge areas in the x-th training batch, N is the number of training batches, h ≤ H, y ≤ , x ≤ N, and h, x, y, H, and N are all positive integers.

[0048] According to an embodiment of the present invention, in formula (3), is the relative difference between the historical measured similarity data and the historical predicted similarity data of the y-th historical segmented bridge area in the h-th historical monitoring period. is the ratio between the historical wind speed data and the standard wind speed data of the y-th historical segmented bridge area in the h-th historical monitoring period. The larger this ratio, the larger the historical wind speed data and the greater the deformation of the bridge. Thus, the change in the historical wind speed data has a greater impact on the error of the historical predicted similarity data. Among them, the standard wind speed data can be 10m / s, is the ratio between the historical traffic load data and the standard traffic load data of the y-th historical segmented bridge area in the h-th historical monitoring period. The larger this ratio, the larger the historical traffic load data and the greater the deformation of the bridge. Thus, the change in the historical traffic load data has a greater impact on the error of the historical predicted similarity data. Among them, the standard traffic load data can be the maximum load-bearing capacity of the bridge, is the relative difference between the historical temperature data and the standard temperature data of the y-th historical segmented bridge area in the h-th historical monitoring period. The larger this relative difference, the greater the difference between the historical temperature data and the standard temperature data (for example, high temperature weather or low temperature weather), and the greater the deformation of the bridge. Thus, the change in the difference between the historical temperature data and the standard temperature data has a greater impact on the error of the historical predicted similarity data. Among them, the standard temperature data can be 20℃, It is shown that the historical wind speed data, historical traffic load data, and historical temperature data are positively correlated with the deformation of the bridge. For example, when strong winds blow, the bridge will vibrate continuously and cause deformation. Overloaded vehicles cause more serious damage to the bridge, which may lead to problems such as cracks and deformation of the bridge. In high-temperature or low-temperature weather, the bridge structure will expand and contract thermally. Therefore, the relevant data of historical wind speed data, historical traffic load data, and historical temperature data are placed in the numerator position. The larger the historical wind speed data, historical traffic load data, and historical temperature data are relative to their respective standard data, that is, , and the larger the values are, the greater the impact on the error of the historical prediction similarity data. is the ratio between the third distance from the midpoint position of the pier in the y-th historical segmented bridge area to the midpoint position of the entire bridge and the third distance from the midpoint position of the pier in the first historical segmented bridge area to the midpoint position of the entire bridge. Among them, the pier in the first historical segmented bridge area is the pier closest to the bridgehead or the bridge tail. The result of 1 minus this ratio can be obtained. The larger this result is, the smaller the third distance is, indicating that the pier is closer to the midpoint position of the entire bridge. For example, in the middle part of the bridge, the historical segmented bridge area at this position is more likely to generate deformation, and the greater the impact on the error of the historical prediction similarity data. Therefore, its weight value is higher. is the weight value of the x-th training batch, which is used to reasonably weight the relative errors of different training batches in the loss function. For the historical segmented bridge area of the (x + 1)-th training batch, the accuracy of the historical prediction similarity data of the (x + 1)-th training batch output by the similarity prediction model is usually higher than that of the historical prediction similarity data of the x-th training batch. That is, the shorter the time interval between a certain training batch and the first training batch, the less accurate its prediction result. To improve the training efficiency, the higher its weight value is set. On the contrary, the more accurate the prediction result is, the lower its weight value is. Thus, higher weights can be assigned to the items with lower accuracy, thereby enhancing the training intensity and training efficiency.

[0049] According to an embodiment of the present invention, by using , and , a weighted average is performed on the relative differences of the historical prediction similarity data of multiple historical segmented bridge areas in multiple historical monitoring cycles of the x-th training batch to obtain a training loss function. During the process of training the similarity prediction model, by performing backpropagation on the loss function, some parameters inside the model are adjusted to reduce the value of the loss function of the similarity prediction model, thereby improving the accuracy of the similarity prediction model and obtaining a trained similarity prediction model.

[0050] In this way, based on the influence of historical temperature data, historical traffic load data, and historical wind speed data on the deformation of the bridge, the influence of the above data on the error of the historical prediction similarity data can be determined. Then, based on this influence and the relative difference between the historical measured similarity data and the historical prediction similarity data, weights are set according to the characteristics that the closer the historical segmented bridge area is to the midpoint position of the entire bridge, the easier it is to generate deformation and the greater the influence on the error of the historical prediction similarity data. Also, weights are set according to the characteristic that the shorter the time interval from the first batch, the lower the accuracy. Thus, the errors of the similarity prediction model outputs for multiple historical segmented bridge areas in multiple historical monitoring cycles in each training batch are weighted and summed to obtain a loss function, so as to improve the design accuracy and objectivity of the loss function, thereby enhancing the training efficiency during the training process and improving the accuracy of the similarity prediction model.

[0051] According to an embodiment of the present invention, in step S9, based on the predicted similarity data and the measured similarity data, the surface anti-deformation coefficient is determined.

[0052] According to an embodiment of the present invention, determining the surface anti-deformation coefficient based on the predicted similarity data and the measured similarity data includes: determining the surface anti-deformation coefficient according to formula (4) , (4), where is the measured similarity data of the e-th segmented bridge area, is the predicted similarity data of the e-th segmented bridge area, E is the number of segmented bridge areas, e ≤ E, and both e and E are positive integers.

[0053] According to an embodiment of the present invention, in formula (4), is the relative difference between the measured similarity data and the predicted similarity data of the e-th segmented bridge area. The larger this relative difference is, the larger the measured similarity data is, indicating that the actual deformation amount of the bridge is less than the predicted deformation amount (the deformation amount predicted under the influence of environmental factors), that is, the stronger the surface anti-deformation ability of the bridge even under the influence of environmental factors. The smaller this relative difference is, the smaller the measured similarity data is, indicating that the actual deformation amount of the bridge is more than the predicted deformation amount, that is, the weaker the surface anti-deformation ability of the bridge under the influence of environmental factors. Taking the average of the relative differences between the measured similarity data and the predicted similarity data of multiple segmented bridge areas, the surface anti-deformation coefficient can be obtained. The larger this surface anti-deformation coefficient is, the stronger the surface anti-deformation ability of the bridge.

[0054] In this way, the surface anti-deformation coefficient can be determined by predicting the similarity data and the measured similarity data, and by comparing the actual deformation amount of the bridge with the predicted deformation amount, the strength of the surface anti-deformation ability of the bridge under the influence of environmental factors can be reflected, and the reliability and scientificity of the surface anti-deformation coefficient can be improved.

[0055] According to an embodiment of the present invention, in step S10, the horizontal anti-deformation coefficients of multiple segmented bridge areas are averaged to obtain the average horizontal anti-deformation coefficient, the vertical anti-deformation coefficients of multiple segmented bridge areas are averaged to obtain the average vertical anti-deformation coefficient, and the average horizontal anti-deformation coefficient, the average vertical anti-deformation coefficient, and the surface anti-deformation coefficient are averaged to obtain the anti-deformation coefficient. The greater the anti-deformation coefficient, the stronger the anti-deformation ability of the bridge structure in the current monitoring period.

[0056] The method for monitoring the anti-deformation of a bridge structure according to an embodiment of the present invention, by setting horizontal test points and vertical test points and using a laser rangefinder for measurement, helps to understand the deformation conditions of the bridge structure in the horizontal and vertical directions. By considering the influence of environmental data on the deformation of the bridge structure and combining with a similarity prediction model, the deformation trend of the bridge can be predicted more accurately. Through non-contact monitoring, the anti-deformation ability of the bridge structure is comprehensively evaluated from three directions: horizontal, vertical, and environmental, improving the comprehensiveness and accuracy of the evaluation of the anti-deformation ability of the bridge structure and reducing the monitoring cost. When determining the horizontal anti-deformation coefficient, it can be based on the relative difference between the first yaw angle at the end and the start of the current monitoring period, and based on the characteristic that the farther the horizontal test point is from the first preset position, the less obvious the change in the first yaw angle, weights are set, so as to perform weighted averaging on the relative differences between the first yaw angles at the end and the start of the current monitoring period for multiple horizontal test points to obtain the horizontal anti-deformation coefficient, improving the accuracy of identifying the anti-lateral deformation ability of the bridge deck. When determining the vertical anti-deformation coefficient, it can be based on the ratio of the vertical displacement of the vertical test point to the height of the bridge pier, and based on the characteristic that the farther the vertical test point is from the first preset position, the greater the vertical displacement corresponding to the same pitch angle change, weights are set, so as to perform weighted averaging on the ratios of the vertical displacements of multiple vertical test points to the height of the bridge pier to obtain the vertical anti-deformation coefficient, improving the accuracy of identifying the anti-vertical deformation ability of the bridge pier. When determining the loss function of the similarity prediction model, it can be determined by the influence of historical temperature data, historical traffic load data, and historical wind speed data on the deformation of the bridge, so as to determine the influence of the above data on the error of the historical prediction similarity data. Then, based on this influence and the relative difference between the historical measured similarity data and the historical prediction similarity data, weights are set based on the characteristic that the closer the historical segmented bridge area is to the midpoint position of the entire bridge, the more likely it is to deform and the greater the influence on the error of the historical prediction similarity data, and based on the characteristic that the lower the accuracy is with a shorter time interval from the first batch, weights are set, so as to perform weighted summation on the errors of the similarity prediction model outputs of multiple historical segmented bridge areas in multiple historical monitoring periods for each training batch to obtain the loss function, so as to improve the design accuracy and objectivity of the loss function, thereby improving the training efficiency during the training process and the accuracy of the similarity prediction model. When determining the surface anti-deformation coefficient, it can be determined by the predicted similarity data and the measured similarity data. By comparing the actual deformation amount of the bridge with the predicted deformation amount, the strength of the surface anti-deformation ability of the bridge under the influence of environmental factors can be reflected, improving the reliability and scientificity of the surface anti-deformation coefficient.

[0057] Figure 5A block diagram of a bridge structure anti-deformation monitoring system according to an embodiment of the present invention is exemplarily shown. The system includes: a segmented bridge area module for dividing the entire bridge into multiple segmented bridge areas, setting a plurality of horizontal test points at the side positions of the bridge deck in the segmented bridge areas, and setting a plurality of vertical test points at the side positions of the bridge piers in the segmented bridge areas. Among them, there is only one bridge pier in the segmented bridge area, the horizontal test points are on the same horizontal plane, and the vertical test points are on the same vertical line; a measurement module for setting a laser rangefinder at a first preset position in the segmented bridge area, making the laser emitted by the laser rangefinder irradiate on the horizontal test points and the vertical test points, and determining the first distance and the first yaw angle of the laser rangefinder when measuring a plurality of horizontal test points, and the second distance and the second pitch angle of the laser rangefinder when measuring a plurality of vertical test points; a horizontal anti-deformation coefficient module for determining the horizontal anti-deformation coefficient according to the first distance and the first yaw angle at the start and end times of the current monitoring period; a vertical anti-deformation coefficient module for determining the vertical anti-deformation coefficient according to the second distance and the second pitch angle at the start and end times of the current monitoring period; a segmented bridge image module for acquiring segmented bridge images of a plurality of segmented bridge areas at the start and end times of the current monitoring period; an actual similarity data module for determining the actual similarity data according to the segmented bridge images; an environmental data module for acquiring the environmental data of the current monitoring period, where the environmental data includes temperature data, traffic load data, and wind speed data; a predicted similarity data module for inputting the environmental data and the segmented bridge images at the start time of the current monitoring period into a trained similarity prediction model to obtain the predicted similarity data of a plurality of segmented bridge areas in the current monitoring period; a surface anti-deformation coefficient module for determining the surface anti-deformation coefficient according to the predicted similarity data and the actual similarity data; an anti-deformation ability evaluation module for determining the anti-deformation ability of the bridge structure in the current monitoring period according to the horizontal anti-deformation coefficient, the vertical anti-deformation coefficient, and the surface anti-deformation coefficient.

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

[0059] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the drawings are only 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 the embodiments of the present invention can have any deformation or modification without departing from the principle.

Claims

1. A method for monitoring the anti-deformation of a bridge structure, characterized in that Including: Dividing the entire bridge into multiple segmented bridge areas, arranging a plurality of horizontal test points at the side positions of the bridge decks in the segmented bridge areas, and arranging a plurality of vertical test points at the side positions of the bridge piers in the segmented bridge areas, wherein there is only one bridge pier in the segmented bridge area, the horizontal test points are on the same horizontal plane, and the vertical test points are on the same vertical line; arranging a laser rangefinder at a first preset position in the segmented bridge area, making the laser emitted by the laser rangefinder irradiate on the horizontal test points and the vertical test points, and determining the first distance and the first yaw angle of the laser rangefinder when measuring a plurality of horizontal test points, and the second distance and the second pitch angle of the laser rangefinder when measuring a plurality of vertical test points; at the start and end moments of the current monitoring period, determining the horizontal anti-deformation coefficient according to the first distance and the first yaw angle; at the start and end moments of the current monitoring period, determining the vertical anti-deformation coefficient according to the second distance and the second pitch angle; acquiring segmented bridge images of a plurality of segmented bridge areas at the start and end moments of the current monitoring period; determining the measured similarity data according to the segmented bridge images; acquiring the environmental data of the current monitoring period, wherein the environmental data includes temperature data, traffic load data and wind speed data; inputting the environmental data and the segmented bridge image at the start moment of the current monitoring period into a trained similarity prediction model to obtain the predicted similarity data of a plurality of segmented bridge areas in the current monitoring period; determining the surface anti-deformation coefficient according to the predicted similarity data and the measured similarity data; determining the anti-deformation ability of the bridge structure in the current monitoring period according to the horizontal anti-deformation coefficient, the vertical anti-deformation coefficient and the surface anti-deformation coefficient.

2. The method for monitoring the anti-deformation of a bridge structure according to claim 1, characterized in that At the start and end moments of the current monitoring period, determining the horizontal anti-deformation coefficient according to the first distance and the first yaw angle includes: acquiring the start maximum first distance according to the first distance at the start moment of the current monitoring period; acquiring the end maximum first distance according to the first distance at the end moment of the current monitoring period; determining the horizontal anti-deformation coefficient according to the start maximum first distance, the end maximum first distance, the first distance and the first yaw angle.

3. The method for monitoring the anti-deformation of a bridge structure according to claim 2, wherein, Determine the horizontal anti-deformation coefficient according to the starting maximum first distance, the ending maximum first distance, the first distance, and the first yaw angle, including: According to the formula Determine the horizontal anti-deformation coefficient of the e-th segmented bridge area , where is the starting maximum first distance of the e-th segmented bridge area, is the ending maximum first distance of the e-th segmented bridge area, is the first distance of the i-th horizontal test point in the e-th segmented bridge area at the start moment of the current monitoring period, is the first distance of the i-th horizontal test point in the e-th segmented bridge area at the end moment of the current monitoring period, is the first yaw angle of the i-th horizontal test point in the e-th segmented bridge area at the start moment of the current monitoring period, is the first yaw angle of the i-th horizontal test point in the e-th segmented bridge area at the end moment of the current monitoring period, n is the number of horizontal test points, max is the maximum value function, i ≤ n, and e, i, and n are all positive integers.

4. The method for monitoring the anti-deformation of a bridge structure according to claim 1, characterized in that, At the start and end moments of the current monitoring period, determining the vertical anti-deformation coefficient according to the second distance and the second pitch angle includes: acquiring the start maximum second distance according to the second distance at the start moment of the current monitoring period; acquiring the end maximum second distance according to the second distance at the end moment of the current monitoring period; acquiring the height of the bridge pier; determining the vertical anti-deformation coefficient according to the height of the bridge pier, the start maximum second distance, the end maximum second distance, the second distance and the second pitch angle.

5. The method for monitoring the anti-deformation of a bridge structure according to claim 4, characterized in that, Determine the vertical anti-deformation coefficient according to the pier height, the starting maximum second distance, the ending maximum second distance, the second distance, and the second pitch angle, including: According to the formula Determine the vertical anti-deformation coefficient of the e-th segmented bridge area , where is the starting maximum second distance of the e-th segmented bridge area, is the ending maximum second distance of the e-th segmented bridge area, is the second distance of the j-th vertical test point in the e-th segmented bridge area at the start moment of the current monitoring period, is the second distance of the j-th vertical test point in the e-th segmented bridge area at the end moment of the current monitoring period, is the second pitch angle of the j-th vertical test point in the e-th segmented bridge area at the start moment of the current monitoring period, is the second pitch angle of the j-th vertical test point in the e-th segmented bridge area at the end moment of the current monitoring period, is the pier height of the e-th segmented bridge area, m is the number of vertical test points, max is the maximum value function, j ≤ m, and e, j, and m are all positive integers.

6. The method for monitoring the anti-deformation of a bridge structure according to claim 1, wherein Based on the segmented bridge image, determine the measured similarity data, including: through the trained image recognition neural network model, perform feature extraction processing on the segmented bridge image at the start moment of the current monitoring period to obtain the start segmented bridge feature vectors of multiple segmented bridge regions; through the trained image recognition neural network model, perform feature extraction processing on the segmented bridge image at the end moment of the current monitoring period to obtain the end segmented bridge feature vectors of multiple segmented bridge regions; determine the similarity between the start segmented bridge feature vector and the end segmented bridge feature vector of the e-th segmented bridge region as the measured similarity data of the e-th segmented bridge region.

7. The method for monitoring the anti-deformation of a bridge structure according to claim 1, characterized in that, The training steps of the similarity prediction model include: obtaining historical segmented bridge images of multiple historical segmented bridge regions at the start and end moments of multiple historical monitoring periods; determining historical measured similarity data based on the historical segmented bridge images; obtaining the third distance between the midpoint position of the pier of each historical segmented bridge region and the midpoint position of the entire bridge; obtaining historical environmental data of multiple historical segmented bridge regions in multiple historical monitoring periods, where the historical environmental data includes historical temperature data, historical traffic load data, and historical wind speed data; through the similarity prediction model, process the historical environmental data and the historical segmented bridge images at the start moment of the historical monitoring period to obtain historical predicted similarity data of multiple historical segmented bridge regions in multiple historical monitoring periods; determine the loss function of the similarity prediction model based on the third distance, the historical environmental data, the historical measured similarity data, and the historical predicted similarity data; train the similarity prediction model according to the loss function of the similarity prediction model to obtain the trained similarity prediction model.

8. The method for monitoring the anti-deformation of a bridge structure according to claim 7, characterized in that, Determine the loss function of the similarity prediction model according to the third distance, historical environmental data, the historical measured similarity data, and the historical predicted similarity data, including: According to the formula Determine the loss function Loss of the similarity prediction model, where is the historical measured similarity data of the y-th historical segmented bridge area in the h-th historical monitoring period, is the historical predicted similarity data of the y-th historical segmented bridge area in the h-th historical monitoring period, is the third distance from the midpoint position of the piers in the y-th historical segmented bridge area to the midpoint position of the entire bridge, is the third distance from the midpoint position of the piers in the 1st historical segmented bridge area to the midpoint position of the entire bridge, is the historical temperature data of the y-th historical segmented bridge area in the h-th historical monitoring period, is the standard temperature data, is the historical traffic load data of the y-th historical segmented bridge area in the h-th historical monitoring period, is the standard traffic load data, is the historical wind speed data of the y-th historical segmented bridge area in the h-th historical monitoring period, is the standard wind speed data, H is the number of historical monitoring periods, is the number of historical segmented bridge areas in the x-th training batch, N is the number of training batches, h ≤ H, y ≤ and x ≤ N, and h, x, y, H, and N are all positive integers.

9. The method for monitoring the anti-deformation of a bridge structure according to claim 1, wherein, Determine the surface anti-deformation coefficient according to the predicted similarity data and the measured similarity data, including: According to the formula Determine the surface anti-deformation coefficient , where is the measured similarity data of the e-th segmented bridge area, is the predicted similarity data of the e-th segmented bridge area, E is the number of segmented bridge areas, e ≤ E, and both e and E are positive integers.

10. A bridge structure anti-deformation monitoring system for implementing the bridge structure anti-deformation monitoring method according to any one of claims 1-9, characterized in that, Including: Segmented bridge area module, which is used to divide the entire bridge into multiple segmented bridge areas, set multiple horizontal test points at the side positions of the bridge deck in the segmented bridge areas, and set multiple vertical test points at the side positions of the bridge piers in the segmented bridge areas. Among them, there is only one bridge pier in the segmented bridge area, the horizontal test points are on the same horizontal plane, and the vertical test points are on the same vertical line; Measurement module, which is used to set a laser rangefinder at the first preset position in the segmented bridge area, make the laser emitted by the laser rangefinder irradiate on the horizontal test points and the vertical test points, and determine the first distance and the first yaw angle of the laser rangefinder when measuring multiple horizontal test points, as well as the second distance and the second pitch angle of the laser rangefinder when measuring multiple vertical test points; Horizontal anti-deformation coefficient module, which is used to determine the horizontal anti-deformation coefficient according to the first distance and the first yaw angle at the start and end times of the current monitoring period; Vertical anti-deformation coefficient module, which is used to determine the vertical anti-deformation coefficient according to the second distance and the second pitch angle at the start and end times of the current monitoring period; Segmented bridge image module, which is used to obtain segmented bridge images of multiple segmented bridge areas at the start and end times of the current monitoring period; Measured similarity data module, which is used to determine the measured similarity data according to the segmented bridge images; Environmental data module, which is used to obtain the environmental data of the current monitoring period, where the environmental data includes temperature data, traffic load data and wind speed data; Predicted similarity data module, which is used to input the environmental data and the segmented bridge images at the start time of the current monitoring period into the trained similarity prediction model to obtain the predicted similarity data of multiple segmented bridge areas in the current monitoring period; Surface anti-deformation coefficient module, which is used to determine the surface anti-deformation coefficient according to the predicted similarity data and the measured similarity data; Anti-deformation ability evaluation module, which is used to determine the anti-deformation ability of the bridge structure in the current monitoring period according to the horizontal anti-deformation coefficient, the vertical anti-deformation coefficient and the surface anti-deformation coefficient.

Images