A bridge rotation trajectory tracking and monitoring method and system based on stress detection

Abstract

The present invention provides a bridge rotation trajectory tracking and monitoring method and system based on stress detection, which relates to the technical field of bridge rotation construction. The method comprises: obtaining stress data; obtaining total weight data of the bridge rotation; inputting the stress data and the total weight data into a trained rotation angle prediction model to obtain predicted rotation angle data of the bridge rotation at multiple rotation moments; obtaining angle data; determining the horizontal trajectory error score of the bridge rotation according to the predicted rotation angle data and the angle data; obtaining inclination data and distance data; determining the vertical trajectory error score of the bridge rotation according to the inclination data and the distance data; determining whether the bridge rotation trajectory deviates according to the horizontal trajectory error score of the bridge rotation and the vertical trajectory error score of the bridge rotation. According to the present invention, the bridge rotation trajectory can be monitored based on the relationship between the stress of the bridge rotation ball joint and the rotation angle.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge rotation construction, and in particular to a bridge rotation trajectory tracking and monitoring method and system based on stress detection. Background Art

[0002] In the related technology, CN118817136A relates to the field of bridge engineering technology, specifically a three-point swivel support leg pressure monitoring system, which specifically includes equipment installation, data measurement, data analysis and swivel pressure monitoring. At least three groups of normal monitoring components are selected and installed in a circular array between the support leg structure and the circular slideway. The bridge is rotated and the pressure is monitored in real time through the monitoring components. The beneficial effects are: by starting the top support component to extend a specified distance and against the bottom of the bridge, the monitoring components at each position measure and obtain a set of pressure data respectively, and the slideway is rotated and adjusted to repeat the measurement to obtain multiple sets of data. Then, with the help of the central processing unit, a three-dimensional coordinate model is established, which can accurately judge the abnormal data, and then reversely infer the faulty monitoring component, and then select the normal monitoring component for subsequent swivel pressure monitoring, which can not only ensure the safety of construction, but also accurately analyze the pressure condition of the support leg, rotation posture, load distribution and other conditions.

[0003] CN118150012A discloses a method for determining the total measurement accuracy of a bridge rotation balanced pressure monitoring system, which includes determining the maximum vertical stress of the spherical joint surface of the rotating spherical joint in the bridge rotation system; determining the unbalanced moment of the bridge rotation system; calculating the normal stress of the bridge rotation system when the moment is unbalanced based on the unbalanced moment; dividing the normal stress when the moment is unbalanced by the maximum vertical stress of the spherical joint surface to obtain the total measurement accuracy of the bridge rotation balanced pressure monitoring system. The purpose of this scheme is to provide a method for determining the total measurement accuracy of a bridge rotation balanced pressure monitoring system to solve the difficulty of determining the total measurement accuracy of the balanced pressure monitoring system of the bridge rotation system in the prior art, provide technical indicators for subsequent complex circuit design, ensure the rationality and effectiveness of monitoring data, and reduce the difficulty of system circuit design.

[0004] Therefore, although the relevant technology can analyze the rotation posture, it does not consider the impact of the relationship between the stress of the bridge swivel ball joint and the rotation angle on the monitoring of the bridge rotation trajectory, that is, it is impossible to monitor the bridge rotation trajectory based on the relationship between the stress of the bridge swivel ball joint and the rotation angle.

[0005] The information disclosed in the background technology section of this application is only intended to deepen the understanding of the general background technology of this application, and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0006] The present invention provides a bridge rotation trajectory tracking and monitoring method and system based on stress detection, which can solve the technical problem that the related technology cannot monitor the bridge rotation trajectory according to the relationship between the stress of the bridge rotation ball joint and the rotation angle.

[0007] According to a first aspect of the present invention, a method for tracking and monitoring a bridge rotation trajectory based on stress detection is provided, comprising: setting a stress sensor at a bridge rotation ball joint to obtain stress data of the bridge rotation ball joint at multiple rotation moments; obtaining total weight data of the bridge rotation; inputting the stress data and the total weight data into a trained rotation angle prediction model to obtain predicted rotation angle data of the bridge rotation at multiple rotation moments; obtaining angle data of the bridge rotation at multiple rotation moments; determining a horizontal trajectory error score of the bridge rotation based on the predicted rotation angle data and the angle data; installing an inclination sensor and a laser rangefinder at the end of the bridge to obtain inclination data at multiple rotation moments and distance data from the end of the bridge to the ground; determining a vertical trajectory error score of the bridge rotation based on the inclination data and the distance data; determining whether the bridge rotation trajectory deviates based on the horizontal trajectory error score of the bridge rotation and the vertical trajectory error score of the bridge rotation.

[0008] Furthermore, the training steps of the rotation angle prediction model include: obtaining historical angle data of multiple historical bridge rotations at multiple historical rotation moments and historical stress data of historical bridge rotation ball joints; obtaining historical total weight data of multiple historical bridge rotations; processing the historical stress data and the historical total weight data through the rotation angle prediction model to obtain historical predicted rotation angle data at multiple historical rotation moments; determining the loss function of the rotation angle prediction model based on the historical stress data, the historical total weight data, the historical angle data and the historical predicted rotation angle data; training the rotation angle prediction model based on the loss function of the rotation angle prediction model to obtain the trained rotation angle prediction model.

[0009] Further, according to the historical stress data, the historical total weight data, the historical angle data and the historical predicted rotation angle data, determining the loss function of the rotation angle prediction model includes: according to the formula Determine the loss function of the rotation angle prediction model ,in, For the i-th historical bridge rotation in Historical stress data of the historical bridge rotation ball joint at the historical rotation moment, is the standard stress data of the i-th historical bridge rotation, is the historical total weight data of the i-th historical bridge rotation, is the standard total weight data of the bridge swivel, For the i-th historical bridge rotation in Historical angle data of historical rotation moments, For the i-th historical bridge rotation in Historical angle data of historical rotation moments, For the i-th historical bridge rotation in The historical predicted rotation angle data of the historical rotation moments, is the number of historical rotation moments of the i-th historical bridge rotation, is the number of historical bridge rotations in the jth batch, N is the number of training batches, ≤ , i≤ , j≤N, and ,i,j, , and N are both positive integers.

[0010] Further, according to the predicted rotation angle data and the angle data, a horizontal trajectory error score of the bridge rotation is determined, including: setting a rotation angle error threshold; and determining the horizontal trajectory error score of the bridge rotation according to the rotation angle error threshold, the predicted rotation angle data and the angle data.

[0011] Further, according to the rotation angle error threshold, the predicted rotation angle data and the angle data, determining the bridge rotation horizontal trajectory error score includes: according to the formula Determine the horizontal trajectory error score of the bridge rotation ,in, is the angle data of the bridge rotation at the k+1th rotation moment, is the angle data of the bridge rotation at the kth rotation moment, To predict the rotation angle data, is the rotation angle error threshold, K is the number of rotation moments of the current bridge rotation, k≤K, and both k and K are positive integers, and if is a conditional function.

[0012] Furthermore, a vertical trajectory error score of the bridge rotation is determined based on the inclination data and the distance data, including: obtaining standard vertical distance data from the end of the bridge to the ground; and determining the vertical trajectory error score of the bridge rotation based on the standard vertical distance data, the inclination data and the distance data.

[0013] Further, according to the standard vertical distance data, the inclination data and the distance data, determining the bridge rotation vertical trajectory error score includes: according to the formula Determine the vertical trajectory error score for bridge rotation ,in, is the distance data of the bridge rotation at the kth rotation moment, is the standard vertical distance data, is the vertical distance error threshold, is the inclination data of the bridge rotation at the kth rotation moment, is the inclination threshold, K is the number of rotation moments of the current bridge rotation, k≤K, and both k and K are positive integers, and if is a conditional function.

[0014] According to a second aspect of the present invention, a bridge rotation trajectory tracking and monitoring system based on stress detection is provided, comprising: a stress data module, used to set a stress sensor on the bridge rotation ball joint to obtain stress data of the bridge rotation ball joint at multiple rotation moments; a total weight data module, used to obtain total weight data of the bridge rotation; a predicted rotation angle data module, used to input the stress data and the total weight data into a trained rotation angle prediction model to obtain predicted rotation angle data of the bridge rotation at multiple rotation moments; an angle data module, used to obtain angle data of the bridge rotation at multiple rotation moments; a bridge rotation water A horizontal trajectory error scoring module is used to determine the horizontal trajectory error score of the bridge rotation according to the predicted rotation angle data and the angle data; an inclination data and distance data module is used to install an inclination sensor and a laser rangefinder at the end of the bridge to obtain the inclination data at multiple rotation moments and the distance data from the end of the bridge to the ground; a bridge rotation vertical trajectory error scoring module is used to determine the vertical trajectory error score of the bridge rotation according to the inclination data and the distance data; a judgment module is used to determine whether the bridge rotation trajectory deviates according to the horizontal trajectory error score of the bridge rotation and the vertical trajectory error score of the bridge rotation.

[0015] Technical effect: According to the present invention, by real-time monitoring of the stress data and angle data of the bridge swivel, as well as the relationship between the stress and the swivel angle of the bridge swivel ball joint, the bridge swivel trajectory is monitored to improve the safety of the bridge. Through the trained swivel angle prediction model, combined with the stress data and total weight data, the swivel angle of the bridge swivel at different swivel moments can be predicted more accurately. By comparing with the actual measured angle data, the horizontal trajectory deviation of the bridge swivel can be accurately evaluated to improve the accuracy of monitoring. By installing an inclination sensor and a laser rangefinder to monitor the vertical trajectory deviation of the bridge swivel, it is helpful to have a more comprehensive understanding of the trajectory status of the bridge swivel. When determining the loss function of the rotation angle prediction model, the influence of the historical stress data and the historical total weight data on the rotation angle can be used to determine the influence of the above data on the error of the historical predicted rotation angle data, so as to set the weight based on the influence and the relative difference between the rotation angle between adjacent historical rotation moments and the historical predicted rotation angle data, and based on the characteristic that the shorter the time interval with the first batch, the lower the accuracy, so as to weighted sum the errors output by the rotation angle prediction model of multiple historical bridge rotations in each batch at multiple historical rotation moments, and obtain the loss function, so as to improve the design accuracy and objectivity of the loss function, thereby improving the training efficiency in the training process and improving the accuracy of the rotation angle prediction model. When determining the horizontal trajectory error score of the bridge rotation, the horizontal trajectory error score of the bridge rotation can be determined by the rotation angle error threshold, the predicted rotation angle data and the angle data. By real-time monitoring the change of the angle during the bridge rotation process and comparing the difference between the measured rotation angle and the predicted rotation angle data under ideal conditions, the horizontal trajectory error of the bridge rotation can be identified, so as to facilitate the adoption of corrective measures, reduce the risk of accidents, and improve the safety and stability of the rotation process. When determining the vertical trajectory error score of the bridge rotation, the standard vertical distance data, inclination data and distance data can be used to determine the vertical trajectory error score of the bridge rotation. The difference between the distance data and the standard vertical distance data, as well as the inclination data of the bridge end, can more comprehensively reflect the actual state of the bridge rotation in the vertical direction, and promptly discover the deviation of the trajectory of the bridge rotation in the vertical direction, thereby ensuring the safe operation of the bridge rotation.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only and do not limit the present invention. Other features and aspects of the present invention will become more apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other embodiments can be obtained based on these drawings without creative work.

[0018] Figure 1 A schematic flow chart of a bridge rotation trajectory tracking and monitoring method based on stress detection according to an embodiment of the present invention is exemplarily shown;

[0019] Figure 2 A block diagram of a bridge rotation trajectory tracking and monitoring system based on stress detection according to an embodiment of the present invention is exemplarily shown. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0022] Figure 1A flow chart of a bridge rotation trajectory tracking and monitoring method based on stress detection according to an embodiment of the present invention is exemplarily shown, the method comprising: step S101, setting a stress sensor at a bridge rotation ball joint to obtain stress data of the bridge rotation ball joint at multiple rotation moments; step S102, obtaining total weight data of the bridge rotation; step S103, inputting the stress data and the total weight data into a trained rotation angle prediction model to obtain predicted rotation angle data of the bridge rotation at multiple rotation moments; step S104, obtaining the bridge rotation at multiple rotation moments Step S105, determining a horizontal trajectory error score of the bridge rotation according to the predicted rotation angle data and the angle data; Step S106, installing an inclination sensor and a laser rangefinder at the end of the bridge to obtain inclination data at multiple rotation moments and distance data from the end of the bridge to the ground; Step S107, determining a vertical trajectory error score of the bridge rotation according to the inclination data and the distance data; Step S108, determining whether the bridge rotation trajectory deviates according to the horizontal trajectory error score of the bridge rotation and the vertical trajectory error score of the bridge rotation.

[0023] According to the bridge rotation trajectory tracking and monitoring method based on stress detection in an embodiment of the present invention, the bridge rotation trajectory is monitored by real-time monitoring of the stress data and angle data of the bridge rotation, as well as the relationship between the stress and rotation angle of the bridge rotation ball joint, thereby improving the safety of the bridge. Through the trained rotation angle prediction model, combined with the stress data and total weight data, the rotation angle of the bridge rotation at different rotation moments can be predicted more accurately. By comparing with the actual measured angle data, the horizontal trajectory deviation of the bridge rotation can be accurately evaluated, thereby improving the accuracy of monitoring. By installing an inclination sensor and a laser rangefinder to monitor the vertical trajectory deviation of the bridge rotation, it is helpful to have a more comprehensive understanding of the trajectory status of the bridge rotation.

[0024] According to one embodiment of the present invention, in step S101, the rotation moment is the moment when the bridge swivel rotates, and the interval between adjacent rotation moments can be set to 5 minutes, 10 minutes, etc., and the present invention does not limit this. The bridge swivel ball joint itself, as a core component of the rotation, its stress state is crucial to the safety and stability of the swivel. Therefore, the stress sensor is arranged inside the bridge swivel ball joint structure, that is, between the upper ball joint and the lower ball joint, and at multiple rotation moments, the stress data borne by the bridge swivel ball joint is effectively monitored. Multiple stress sensors can be arranged between the upper ball joint and the lower ball joint, and the stress data obtained can be the average stress.

[0025] According to an embodiment of the present invention, in step S102, the total weight data of the bridge swivel may be calculated by the weight of the construction materials of the bridge swivel.

[0026] According to one embodiment of the present invention, in step S103, the stress data and the total weight data are input into a trained rotation angle prediction model. The rotation angle prediction model may be a neural network model, which is trained based on a large amount of historical data through machine learning or deep learning technology, and can analyze the predicted rotation angle data of the bridge rotation at multiple rotation moments.

[0027] According to one embodiment of the present invention, the training step of the rotation angle prediction model includes: obtaining historical angle data of multiple historical bridge rotations at multiple historical rotation moments and historical stress data of historical bridge rotation ball joints; obtaining historical total weight data of multiple historical bridge rotations; processing the historical stress data and the historical total weight data through the rotation angle prediction model to obtain historical predicted rotation angle data at multiple historical rotation moments; determining the loss function of the rotation angle prediction model based on the historical stress data, the historical total weight data, the historical angle data and the historical predicted rotation angle data; training the rotation angle prediction model based on the loss function of the rotation angle prediction model to obtain the trained rotation angle prediction model.

[0028] According to one embodiment of the present invention, during the construction of multiple bridges, multiple historical bridge rotations are searched and determined. The historical bridge rotations are bridge rotations that have been rotated in the bridge history records. Each historical bridge rotation has a different rotation time due to total weight, rotation angle, etc. Therefore, the number of historical rotation moments of each historical bridge rotation is different. For example, when the rotation time is 60 minutes and the interval between adjacent historical rotation moments is set to 10 minutes, the number of historical rotation moments is 6. When the rotation time is 120 minutes and the interval between adjacent historical rotation moments is set to 10 minutes, the number of historical rotation moments is 12. Multiple historical bridge rotations are divided into different batches of historical bridge rotations, and the number of historical bridge rotations in each batch is the same, so that rotation angle prediction models of different batches are trained. For example, , M is the number of historical bridge rotations, , , …, are the number of historical bridge rotations in the 1st, 2nd, …, Nth batches respectively. GPS can be used to record the historical angle data of the historical bridge rotation at multiple historical rotation moments, and at the same time, the historical stress data of the historical bridge rotation ball joint at multiple historical rotation moments can be obtained. When the bridge starts to rotate, the stress state of the bridge rotation ball joint structure will change. As the stress in the bridge rotation ball joint structure increases, a greater torque acts on the ball joint, and the rotation angle will also increase. The greater the total weight of the bridge rotation, the greater the driving force required during the rotation process. Therefore, under the same driving force, the greater the total weight of the bridge rotation, the smaller the rotation angle. The rotation angle prediction model can predict the historical predicted rotation angle data at multiple historical rotation moments based on the relationship between the above stress, total weight and rotation angle, based on the historical stress data and the historical total weight data. The loss function is determined according to the relative difference between the historical predicted rotation angle data and the measured historical angle data. The trained rotation angle prediction model is obtained by feedback adjustment of the loss function.

[0029] According to one embodiment of the present invention, the loss function of the rotation angle prediction model is determined based on the historical stress data, the historical total weight data, the historical angle data and the historical predicted rotation angle data, including: determining the loss function of the rotation angle prediction model according to formula (1): , (1), where For the i-th historical bridge rotation Historical stress data of the historical bridge rotation ball joint at the historical rotation moment, is the standard stress data of the i-th historical bridge rotation, is the historical total weight data of the i-th historical bridge rotation, is the standard total weight data of the bridge swivel, For the i-th historical bridge rotation in Historical angle data of historical rotation moments, For the i-th historical bridge rotation in Historical angle data of historical rotation moments, For the i-th historical bridge rotation in The historical predicted rotation angle data of the historical rotation moments, is the number of historical rotation moments of the i-th historical bridge rotation, is the number of historical bridge rotations in the jth batch, N is the number of training batches, ≤ , i≤ , j≤N, and ,i,j, , and N are both positive integers.

[0030] According to one embodiment of the present invention, in formula (1), For the i-th historical bridge rotation in The historical angle data at the moment of the historical rotation is the same as the historical angle data at the moment of the i-th historical bridge rotation. The difference between the historical angle data at the historical rotation moments, that is, and The rotation angle between the historical rotation moments, For the and The rotation angle between the historical rotation moments is equal to the rotation angle between the i-th historical bridge rotation moments The relative difference between the historical predicted rotation angle data at the historical rotation moments. For the i-th historical bridge rotation in The relative difference between the historical stress data of the spherical joint of the historical bridge rotation at the moment of the historical rotation and the standard stress data of the i-th historical bridge rotation. The larger the relative difference, the greater the historical bridge rotation at the moment of the historical rotation. The greater the relative difference between the historical stress data and the standard stress data of the historical bridge rotation ball joint at a historical rotation moment, the greater the rotation angle. It is the ratio of the historical total weight data of the i-th historical bridge rotation to the standard total weight data of the bridge rotation. The smaller the ratio is, the smaller the historical total weight data is and the larger the rotation angle is. It means that the relative difference between the historical stress data and the standard stress data is positively correlated with the rotation angle, and the historical total weight data is negatively correlated with the rotation angle. For example, when the relative difference between the historical stress data and the standard stress data is larger, the torque acting on the ball joint is larger, and the rotation angle is larger. The smaller the historical total weight data is, the smaller the driving force required during the rotation process is. Under the same driving force, the smaller the historical total weight data is, the larger the rotation angle of the bridge is. Therefore, putting the historical stress data related data in the numerator position means that the historical stress data is larger relative to the standard stress data, that is, The larger the value of , the greater the impact on the error of the historical predicted rotation angle data. The historical total weight data related data are placed in the denominator, indicating that the historical total weight data is smaller than the standard total weight data of the bridge rotation, that is, The smaller the value, the greater the impact on the error of the historical predicted rotation angle data. is the weight of the j-th batch, which is used to reasonably weight the relative errors of different training batches in the loss function. For the j+1-th batch of historical bridge rotations, the accuracy of the historical predicted rotation angle data of the j+1-th batch output by the rotation angle prediction model is usually higher than the accuracy of the historical predicted rotation angle data of the j-th batch. That is, the shorter the time interval between a batch and the first batch, the less accurate its prediction result. In order to improve the training efficiency, the higher its weight is set. Conversely, the more accurate the prediction result, the lower its weight is. Therefore, a higher weight can be given to items with lower accuracy, thereby improving the training intensity and training efficiency.

[0031] According to one embodiment of the present invention, using and , the relative errors of the historical predicted rotation angle data of multiple historical bridge rotations in the jth batch at multiple historical rotation moments are weighted averaged to obtain the training loss function. In the process of training the rotation angle prediction model, the loss function is back-propagated and some parameters inside the model are adjusted to reduce the value of the loss function of the rotation angle prediction model, thereby improving the accuracy of the rotation angle prediction model and obtaining the trained rotation angle prediction model.

[0032] In this way, the influence of historical stress data and historical total weight data on the rotation angle can be used to determine the influence of the above data on the error of historical predicted rotation angle data, and then set weights based on the influence and the relative difference between the rotation angle between adjacent historical rotation moments and the historical predicted rotation angle data, and based on the characteristic that the shorter the time interval with the first batch, the lower the accuracy, so as to perform weighted summation on the errors output by the rotation angle prediction model of multiple historical bridge rotations in each batch at multiple historical rotation moments to obtain a 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 improving the accuracy of the rotation angle prediction model.

[0033] According to an embodiment of the present invention, in step S104, GPS may be used to collect the angle of the bridge rotation in real time.

[0034] According to one embodiment of the present invention, in step S105, a bridge rotation horizontal trajectory error score is determined based on the predicted rotation angle data and the angle data.

[0035] According to one embodiment of the present invention, step S105 includes: setting a rotation angle error threshold; and determining a bridge rotation horizontal trajectory error score according to the rotation angle error threshold, the predicted rotation angle data and the angle data.

[0036] According to one embodiment of the present invention, the rotation angle error threshold is used to measure the allowable deviation between the actual rotation angle of the bridge and the predicted rotation angle data. The predicted rotation angle data is obtained through a trained rotation angle prediction model, representing the rotation angle that the bridge should achieve under ideal conditions. The actual angle data is collected in real time through monitoring equipment, reflecting the actual angle change of the bridge during the actual rotation process.

[0037] According to one embodiment of the present invention, the bridge rotation horizontal trajectory error score is determined according to the rotation angle error threshold, the predicted rotation angle data and the angle data, including: determining the bridge rotation horizontal trajectory error score according to formula (2): , (2), where is the angle data of the bridge rotation at the k+1th rotation moment, is the angle data of the bridge rotation at the kth rotation moment, To predict the rotation angle data, is the rotation angle error threshold, K is the number of rotation moments of the current bridge rotation, k≤K, and both k and K are positive integers, and if is a conditional function.

[0038] According to one embodiment of the present invention, in formula (2), It is the difference between the angle data of the bridge rotation at the k+1th rotation time and the angle data of the bridge rotation at the kth rotation time, that is, the rotation angle of the bridge rotation between the k+1th and kth rotation times. When the difference between the rotation angle of the bridge between the k+1th and kth rotation moments and the predicted rotation angle data is less than or equal to the rotation angle error threshold, the conditional function value is 1, otherwise, the conditional function value is 0. Therefore, if the difference between the measured rotation angle and the predicted rotation angle data between adjacent rotation moments is greater than the rotation angle error threshold, it means that the difference between the measured rotation angle and the predicted rotation angle data under ideal conditions is large, and the horizontal trajectory of the bridge rotation deviates from normal. It means that the conditional functions corresponding to the rotation angles of the bridge rotation at multiple rotation moments are averaged to obtain the horizontal trajectory error score of the bridge rotation. The larger the horizontal trajectory error score of the bridge rotation, the more serious the deviation of the track of the bridge rotation in the horizontal direction.

[0039] In this way, the horizontal trajectory error score of the bridge rotation can be determined through the rotation angle error threshold, predicted rotation angle data and angle data. By real-time monitoring the changes in angle during the bridge rotation process and comparing the difference between the measured rotation angle and the predicted rotation angle data under ideal conditions, the horizontal trajectory error of the bridge rotation can be identified, making it easier to take corrective measures, reduce the risk of accidents, and improve the safety and stability of the rotation process.

[0040] According to one embodiment of the present invention, in step S106, the inclination sensor can accurately measure and record the inclination data of the bridge end at multiple rotation moments of the bridge rotation, thereby capturing the small inclination angle of the bridge end relative to the horizontal plane. The laser rangefinder can measure the distance from the bridge end to the ground. The laser rangefinder emits a laser beam perpendicular to the plane where the bridge end is located to the ground, and calculates the distance between the bridge end and the ground by receiving the reflected laser signal.

[0041] According to one embodiment of the present invention, in step S107, a vertical trajectory error score of the bridge rotation is determined based on the inclination data and the distance data.

[0042] According to one embodiment of the present invention, step S107 includes: obtaining standard vertical distance data from the bridge end to the ground; and determining a vertical trajectory error score of the bridge rotation based on the standard vertical distance data, the inclination data and the distance data.

[0043] According to one embodiment of the present invention, before the rotation is performed, a precision measuring device, such as a laser rangefinder, can be used to measure the standard vertical distance data from the bridge end to the ground. The actual measured distance data is compared with the standard vertical distance data, and combined with the inclination data to evaluate the degree of deviation from the vertical trajectory during the bridge rotation process.

[0044] According to one embodiment of the present invention, determining the bridge rotation vertical trajectory error score according to the standard vertical distance data, the inclination data and the distance data includes: determining the bridge rotation vertical trajectory error score according to formula (3): ,

[0045] (3), where is the distance data of the bridge rotation at the kth rotation moment, is the standard vertical distance data, is the vertical distance error threshold, is the inclination data of the bridge rotation at the kth rotation moment, is the inclination threshold, K is the number of rotation moments of the current bridge rotation, k≤K, and both k and K are positive integers, and if is a conditional function.

[0046] According to one embodiment of the present invention, in formula (3), Indicates that the difference between the distance data and the standard vertical distance data at the kth rotation moment of the bridge rotation is less than or equal to the vertical distance error threshold, and when the inclination data of the bridge rotation at the kth rotation moment is less than or equal to the inclination threshold, the conditional function value is 1, otherwise, the conditional function value is 0. The laser rangefinder is perpendicular to the plane where the bridge end is located, thereby emitting a laser beam to the ground. Therefore, the distance data from the bridge end to the ground is a straight-line distance. If the bridge end has an inclination angle relative to the horizontal plane (the inclination data is greater than the inclination threshold), the distance data may be the same as the standard vertical distance data. At this time, the vertical trajectory of the bridge rotation is also deviated from the normal state. It means that the conditional functions corresponding to the distance data and inclination data of the bridge rotation at multiple rotation moments are averaged to obtain the vertical trajectory error score of the bridge rotation. The larger the vertical trajectory error score of the bridge rotation, the more serious the deviation of the track of the bridge rotation in the vertical direction.

[0047] In this way, the vertical trajectory error score of the bridge rotation can be determined through the standard vertical distance data, inclination data and distance data. The difference between the distance data and the standard vertical distance data, as well as the inclination data of the bridge end, can more comprehensively reflect the actual state of the bridge rotation in the vertical direction, and promptly discover the deviation of the trajectory of the bridge rotation in the vertical direction, so that the bridge rotation can operate safely.

[0048] According to one embodiment of the present invention, in step S108, the bridge rotation horizontal track error score and the bridge rotation vertical track error score are compared with their respective score thresholds, and if any one of the two scores is greater than the respective score threshold, it is determined that the bridge rotation track is in a deviated state. If both scores are less than their respective score thresholds, it is determined that the bridge rotation track is in a normal state.

[0049] According to the bridge rotation trajectory tracking and monitoring method based on stress detection in an embodiment of the present invention, the bridge rotation trajectory is monitored by real-time monitoring of the stress data and angle data of the bridge rotation, as well as the relationship between the stress and rotation angle of the bridge rotation ball joint, thereby improving the safety of the bridge. Through the trained rotation angle prediction model, combined with the stress data and total weight data, the rotation angle of the bridge rotation at different rotation moments can be predicted more accurately. By comparing with the actual measured angle data, the horizontal trajectory deviation of the bridge rotation can be accurately evaluated, thereby improving the accuracy of monitoring. By installing an inclination sensor and a laser rangefinder to monitor the vertical trajectory deviation of the bridge rotation, it is helpful to have a more comprehensive understanding of the trajectory status of the bridge rotation. When determining the loss function of the rotation angle prediction model, the influence of the historical stress data and the historical total weight data on the rotation angle can be used to determine the influence of the above data on the error of the historical predicted rotation angle data, so as to set the weight based on the influence and the relative difference between the rotation angle between adjacent historical rotation moments and the historical predicted rotation angle data, and based on the characteristic that the shorter the time interval with the first batch, the lower the accuracy, so as to weighted sum the errors output by the rotation angle prediction model of multiple historical bridge rotations in each batch at multiple historical rotation moments, and obtain the loss function, so as to improve the design accuracy and objectivity of the loss function, thereby improving the training efficiency in the training process and improving the accuracy of the rotation angle prediction model. When determining the horizontal trajectory error score of the bridge rotation, the horizontal trajectory error score of the bridge rotation can be determined by the rotation angle error threshold, the predicted rotation angle data and the angle data. By real-time monitoring the change of the angle during the bridge rotation process and comparing the difference between the measured rotation angle and the predicted rotation angle data under ideal conditions, the horizontal trajectory error of the bridge rotation can be identified, so as to facilitate the adoption of corrective measures, reduce the risk of accidents, and improve the safety and stability of the rotation process. When determining the vertical trajectory error score of the bridge rotation, the standard vertical distance data, inclination data and distance data can be used to determine the vertical trajectory error score of the bridge rotation. The difference between the distance data and the standard vertical distance data, as well as the inclination data of the bridge end, can more comprehensively reflect the actual state of the bridge rotation in the vertical direction, and promptly discover the deviation of the trajectory of the bridge rotation in the vertical direction, thereby ensuring the safe operation of the bridge rotation.

[0050] Figure 2A block diagram of a bridge rotation trajectory tracking and monitoring system based on stress detection according to an embodiment of the present invention is exemplarily shown, wherein the system comprises: a stress data module, which is used to set a stress sensor on a bridge rotation ball joint to obtain stress data of the bridge rotation ball joint at multiple rotation moments; a total weight data module, which is used to obtain total weight data of the bridge rotation; a predicted rotation angle data module, which is used to input the stress data and the total weight data into a trained rotation angle prediction model to obtain predicted rotation angle data of the bridge rotation at multiple rotation moments; an angle data module, which is used to obtain angle data of the bridge rotation at multiple rotation moments; A beam rotation horizontal trajectory error scoring module is used to determine the horizontal trajectory error score of the bridge rotation based on the predicted rotation angle data and the angle data; an inclination data and distance data module is used to install an inclination sensor and a laser rangefinder at the end of the bridge to obtain the inclination data at multiple rotation moments and the distance data from the end of the bridge to the ground; a bridge rotation vertical trajectory error scoring module is used to determine the vertical trajectory error score of the bridge rotation based on the inclination data and the distance data; a judgment module is used to determine whether the bridge rotation trajectory deviates based on the bridge rotation horizontal trajectory error score and the bridge rotation vertical trajectory error score.

[0051] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0052] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.

Claims

1. A bridge rotation trajectory tracking and monitoring method based on stress detection, characterized in that: include: A stress sensor is arranged on the bridge swivel spherical joint to obtain stress data of the bridge swivel spherical joint at multiple swivel moments; Obtain the total weight data of the bridge rotation; Input the stress data and the total weight data into a trained rotation angle prediction model to obtain predicted rotation angle data of the bridge rotation at multiple rotation moments; obtain angle data of the bridge rotation at multiple rotation moments; determine the horizontal trajectory error score of the bridge rotation based on the predicted rotation angle data and the angle data; install an inclination sensor and a laser rangefinder at the end of the bridge to obtain inclination data and distance data from the end of the bridge to the ground at multiple rotation moments; determine the vertical trajectory error score of the bridge rotation based on the inclination data and the distance data; determine whether the bridge rotation trajectory deviates based on the horizontal trajectory error score of the bridge rotation and the vertical trajectory error score of the bridge rotation; The training steps of the rotation angle prediction model include: obtaining historical angle data of multiple historical bridge rotations at multiple historical rotation moments and historical stress data of historical bridge rotation ball joints; obtaining historical total weight data of multiple historical bridge rotations; processing the historical stress data and the historical total weight data through the rotation angle prediction model to obtain historical predicted rotation angle data at multiple historical rotation moments; determining the loss function of the rotation angle prediction model based on the historical stress data, the historical total weight data, the historical angle data and the historical predicted rotation angle data; training the rotation angle prediction model based on the loss function of the rotation angle prediction model to obtain the trained rotation angle prediction model; Determining the loss function of the rotation angle prediction model according to the historical stress data, the historical total weight data, the historical angle data and the historical predicted rotation angle data includes: according to the formula Determine the loss function of the rotation angle prediction model ,in, For the i-th historical bridge rotation Historical stress data of the historical bridge rotation ball joint at the historical rotation moment, is the standard stress data of the i-th historical bridge rotation, is the historical total weight data of the i-th historical bridge rotation, is the standard total weight data of the bridge swivel, For the i-th historical bridge rotation Historical angle data of historical rotation moments, For the i-th historical bridge rotation Historical angle data of historical rotation moments, For the i-th historical bridge rotation The historical predicted rotation angle data of the historical rotation moments, is the number of historical rotation moments of the i-th historical bridge rotation, is the number of historical bridge rotations in the jth batch, N is the number of training batches, ≤ , i≤ , j≤N, and ,i,j, , and N are both positive integers.

2. The bridge rotation trajectory tracking and monitoring method based on stress detection according to claim 1 is characterized in that: Determining a bridge rotation horizontal trajectory error score according to the predicted rotation angle data and the angle data includes: setting a rotation angle error threshold; determining a bridge rotation horizontal trajectory error score according to the rotation angle error threshold, the predicted rotation angle data and the angle data.

3. The bridge rotation trajectory tracking and monitoring method based on stress detection according to claim 2 is characterized in that: Determining the bridge rotation horizontal trajectory error score according to the rotation angle error threshold, the predicted rotation angle data and the angle data includes: according to the formula Determine the horizontal trajectory error score of the bridge rotation ,in, is the angle data of the bridge rotation at the k+1th rotation moment, is the angle data of the bridge rotation at the kth rotation moment, To predict the rotation angle data, is the rotation angle error threshold, K is the number of rotation moments of the current bridge rotation, k≤K, and both k and K are positive integers, and if is a conditional function.

4. The bridge rotation trajectory tracking and monitoring method based on stress detection according to claim 3 is characterized in that: Determine the vertical trajectory error score of the bridge rotation according to the inclination data and the distance data, including: obtaining standard vertical distance data from the end of the bridge to the ground; determine the vertical trajectory error score of the bridge rotation according to the standard vertical distance data, the inclination data and the distance data.

5. The bridge rotation trajectory tracking and monitoring method based on stress detection according to claim 4 is characterized in that: Determining the bridge rotation vertical trajectory error score according to the standard vertical distance data, the inclination data and the distance data includes: according to the formula Determine the vertical trajectory error score for bridge rotation ,in, is the distance data of the bridge rotation at the kth rotation moment, is the standard vertical distance data, is the vertical distance error threshold, is the inclination data of the bridge rotation at the kth rotation moment, is the inclination threshold, K is the number of rotation moments of the current bridge rotation, k≤K, and both k and K are positive integers, and if is a conditional function.

6. A bridge rotation trajectory tracking and monitoring system based on stress detection for executing the bridge rotation trajectory tracking and monitoring method based on stress detection as described in any one of claims 1 to 5, characterized in that: include: A stress data module is used to set a stress sensor on the bridge swivel ball joint to obtain stress data of the bridge swivel ball joint at multiple rotation moments; A total weight data module is used to obtain the total weight data of the bridge rotation; a predicted rotation angle data module is used to input the stress data and the total weight data into a trained rotation angle prediction model to obtain the predicted rotation angle data of the bridge rotation at multiple rotation moments; an angle data module is used to obtain the angle data of the bridge rotation at multiple rotation moments; a bridge rotation horizontal trajectory error scoring module is used to determine the bridge rotation horizontal trajectory error score based on the predicted rotation angle data and the angle data; an inclination data and distance data module is used to install an inclination sensor and a laser rangefinder at the end of the bridge to obtain the inclination data at multiple rotation moments and the distance data from the end of the bridge to the ground; a bridge rotation vertical trajectory error scoring module is used to determine the bridge rotation vertical trajectory error score based on the inclination data and the distance data; a judgment module is used to determine whether the bridge rotation trajectory deviates based on the bridge rotation horizontal trajectory error score and the bridge rotation vertical trajectory error score.

Images