Cable force detection method for fitting cylindrical curved surface by using three-dimensional point cloud

Abstract

The invention provides a cable force detection method for fitting a cylindrical curved surface by using a three-dimensional point cloud. The method comprises the following steps: carrying out preprocessing such as segmentation on original three-dimensional point cloud data, and removing points obviously deviating from a cylindrical curved surface; determining the main direction and the cable plane of the point cloud data, and calculating the cable length and the inclination angle of a segment; parameters such as linear density, gravitational acceleration, section circle radius of the inhaul cable, inclination angle and calculated cable length are given, equation fitting is carried out by utilizing point cloud according to a space curved surface equation of the cable under an overall coordinate system, and cable force is obtained; and according to the statistical characteristics of the projection distance from each point to the cylindrical curved surface, the points with relatively large deviation are further eliminated, so that a more accurate cable force identification value can be obtained. According to the method, an error propagation chain can be greatly shortened, and errors caused by the middle process are avoided; the problem of precision difference of slices at different positions is avoided, noise reduction can be carried out by utilizing statistical characteristics of distances between all points and the curved surface, and the precision is remarkably improved. According to the method, the inclination angle (or linear density) and the cable force can be simultaneously used as undetermined parameters to be identified according to needs.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and particularly to a method for detecting cable forces by fitting a cylindrical surface with three-dimensional point clouds. Background Art

[0002] Stay cables are the main load-bearing components of cable-supported bridges, and the cable force is an important indicator reflecting the structural state of the bridge. Abnormal cable forces can lead to adverse changes in structural internal forces and deformations, thereby affecting the safety and service life of the bridge structure. Therefore, during the engineering operation period, the efficient and accurate detection of cable forces has always been a hot issue.

[0003] In recent years, with the development of information technology, the application of three-dimensional point clouds has become more and more extensive, making it possible to estimate cable forces using three-dimensional point clouds. Currently, common practices include directly projecting onto a two-dimensional plane and then obtaining the cable force by fitting the cable plane equation; or by slicing the three-dimensional point cloud data, fitting the center of the circle, and then fitting the cable plane equation after projecting the center of the circle onto the two-dimensional plane and other specific means.

[0004] However, in practical applications, three-dimensional point cloud data often has the following characteristics: (1) The distribution positions of the point clouds obtained by slicing at different positions are different on the cable cross-section; (2) There are large differences in the density of the point clouds along the axial direction of the cable. The former leads to deviations in the vertical plane during direct two-dimensional projection; the latter leads to different accuracies of the centers of the circles obtained by slicing at different positions, and the point clouds obtained by slicing are not actually located at that cross-section. These will all lead to errors in cable force identification. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for detecting cable forces by fitting a cylindrical surface with three-dimensional point clouds to solve the problems existing in the prior art.

[0006] The technical solution adopted to achieve the purpose of the present invention is as follows. A method for detecting cable forces by fitting a cylindrical surface with three-dimensional point clouds includes the following steps:

[0007] A method for detecting cable forces by fitting a cylindrical surface with three-dimensional point clouds, characterized in that it includes the following steps:

[0008] 1) Scanning the cable-stayed bridge to be measured to obtain the original three-dimensional point cloud data of the stay cables;

[0009] 2) Preprocessing the original three-dimensional point cloud data to screen out the effective point cloud data of each stay cable;

[0010] 3) Calculating the center of the data and translating the data to the origin; determining the main direction and the cable plane using the point cloud data; wherein, performing necessary inversion and rotation operations on the three-dimensional coordinates to ensure that the x, y, and z of the point cloud coordinate system correspond one-to-one with the x, y, and z of the local coordinate system of the cable cylindrical surface equation;

[0011] 4) Calculate the cable length l and the inclination angle θ of each cable; Take any cable segment OAB, the chord length between points OB is l, the included angle between the chord direction and the horizontal direction is θ, and the sag of the cable is f;

[0012] 5) According to the spatial surface equation of the cable, use the point cloud data for equation fitting to obtain the cable force T;

[0013] 6) Judge whether the fitting degree meets the requirements; If it does not meet the requirements, then count the projection distances of each point to the curved cylindrical surface; Use statistical knowledge to further eliminate the points with large deviations; Use the reduced point cloud data to repeat steps 3) - 5) to obtain a new cable force.

[0014] Furthermore, in step 1), the original three-dimensional point cloud data is obtained using a three-dimensional laser scanner.

[0015] Furthermore, in step 1), select to obtain the point cloud data under the weather conditions of no wind or gentle breeze.

[0016] Furthermore, in step 2), the preprocessing includes using clustering and / or registration algorithms to eliminate the points that are obviously not near the surface of the cylindrical surface.

[0017] Furthermore, in step 2), for the point cloud data scanned by multiple stations, use the clustering algorithm to separate different data clusters.

[0018] Furthermore, in step 3), use the PCA method to determine the main direction, and the first principal component represents the main direction of the cable.

[0019] Furthermore, in step 5), use the minimization of the sum of the squares of the distances from the data points to the cylindrical surface to optimize the parameters.

[0020] Furthermore, in step 6), use the coefficient of determination, mean square error, and residual as the goodness-of-fit parameters to judge whether to iterate again.

[0021] Furthermore, in step 6), it also has the content of identifying the inclination angle or linear density.

[0022] The technical effects of the present invention are beyond doubt: greatly shorten the chain of error propagation, avoid the errors brought by the intermediate process; avoid the problem of different slice accuracies at different positions, and can use the statistical characteristics of the distances from all points to the surface for noise reduction, significantly improving the accuracy. The present invention can also identify the inclination angle (or linear density) and the cable force simultaneously as undetermined parameters according to needs. Brief Description of the Drawings

[0023] Figure 1 It is a schematic flow chart of a cable force detection method for fitting a cylindrical surface using three-dimensional point clouds; Figure 2It is the force analysis diagram of any cable segment OAB. Specific implementation mode

[0024] The present invention will be further described below in conjunction with embodiments, but it should not be understood that the above-mentioned subject scope of the present invention is limited to the following embodiments. Without departing from the above technical idea of the present invention, various substitutions and changes should be included in the protection scope of the present invention according to the common general knowledge and conventional means in the art.

[0025] Embodiment 1:

[0026] Under the action of cable force, the cable will undergo axial elastic elongation and spatial configuration adjustment, and its geometric shape in the equilibrium state strictly satisfies the mathematical characteristics of the cylindrical surface. There is a clear physical relationship between the cable force and the parameters of the cylindrical surface: the change of the cable force changes the axial strain through the material constitutive relationship, directly affecting the length of the cylindrical axis; at the same time, it changes the spatial configuration through the geometric relationship. Although the measured three-dimensional point cloud has measurement noise and uneven distribution (including differences in axial sampling intervals and circumferential angle offsets of cross-sections), robust parameter estimation can be achieved by establishing a least squares cylindrical surface fitting model with constraints. This model uses the differential geometric characteristics of the cylindrical surface (generatrix straight line constraint, cross-section circle orthogonality constraint) to construct a dual filtering mechanism: in the axial dimension, the influence of sparse point cloud distribution is eliminated through parameter regression, and in the radial dimension, the outlier interference in the cross-section direction is suppressed by minimizing the roundness error.

[0027] This embodiment provides a cable force detection method for fitting a cylindrical surface using three-dimensional point clouds, including the following steps:

[0028] 1) Scan the cable-stayed bridge to be measured to obtain the original three-dimensional point cloud data of the stay cables;

[0029] 2) Preprocess the original three-dimensional point cloud data to screen out the effective point cloud data of each stay cable;

[0030] 3) Calculate the center of the data and translate the data to the origin; determine the main direction and the cable plane using the point cloud data; among them, necessary inversion and rotation operations are performed on the three-dimensional coordinates to ensure that the x, y, and z of the point cloud coordinate system correspond one-to-one with the x, y, and z of the local coordinate system of the cable cylindrical surface equation;

[0031] 4) Calculate the cable length l and the inclination angle θ of each cable;

[0032] 5) According to the spatial surface equation of the cable, use the point cloud data for univariate fitting to obtain the cable force T; the dimensionless equation of the cylindrical surface of the cable in the local coordinate system is as follows;

[0033]

[0034] Among them, x is the chordwise coordinate of the cable, y is the coordinate perpendicular to x in the vertical plane, and z is the transverse coordinate. Optionally, the cylindrical surface equation can also consider the flexural rigidity EI;

[0035] 6) Determine whether the fitting degree meets the requirements; if not, count the projection distances of each point to the curved cylindrical surface; using statistical knowledge, further eliminate the points with large deviations; using the reduced point cloud data, repeat steps 4) - 5) to obtain a new cable force. Use the coefficient of determination, mean square error, and residual as goodness-of-fit parameters to determine whether to iterate again.

[0036] 7) Optionally, if in addition to the cable force T as the parameter to be identified, the inclination angle or linear density also needs to be identified, then in step 5), both the cable force T and the linear density m (or inclination angle) can be regarded as unknowns for multivariate function fitting.

[0037] This embodiment has the following characteristics: (1) fewer steps, greatly shortening the error propagation chain and avoiding errors caused by intermediate processes; (2) using all point clouds to fit the cylindrical surface, avoiding the problem of different position accuracies, and can use the statistical characteristics of the distances between all points and the cylindrical surface for noise reduction, significantly improving the accuracy; (3) multiple parameters including the cable force can be identified simultaneously according to needs, such as the inclination angle or linear density, etc.

[0038] Embodiment 2:

[0039] The main content of this embodiment is the same as that of Embodiment 1. Among them, in step 1), the point cloud data is acquired under the weather conditions of no wind or gentle breeze.

[0040] Embodiment 3:

[0041] The main content of this embodiment is the same as that of Embodiment 1 or 2. Among them, in step 2), for the point cloud data scanned by multiple stations, clustering algorithms are used to separate different data clusters. The preprocessing includes using clustering and / or registration algorithms to eliminate the points that are obviously not near the surface of the cylindrical surface.

[0042] Embodiment 4:

[0043] The main content of this embodiment is the same as any one of Embodiments 1 - 3. Among them, calculate the center of the data and translate the data to the origin; use the PCA method to determine the main direction, and the first principal component represents the main direction of the cable. In step 4), the cable length is determined by truncating according to the point cloud data density or intercepting according to the segmentation of the component itself; the inclination angle is determined by referring to the value in the design document, truncating according to the point cloud data density, or intercepting according to the segmentation of the component itself;

[0044] Embodiment 5:

[0045] The main content of this embodiment is the same as any one of Embodiments 1 to 4. Among them, in step 5), the sum of the squares of the distances from the data points to the cylindrical surface is used as the objective to optimize the parameters.

Claims

1. A cable force detection method using a three-dimensional point cloud to fit a cylindrical surface, characterized in that It includes the following steps: 1) Scan the cable-stayed bridge to be measured to obtain the original three-dimensional point cloud data of the stay cables; 2) Preprocess the original three-dimensional point cloud data to screen out the effective point cloud data of each stay cable; 3) Calculate the center of the data and translate the data to the origin; Use the point cloud data to determine the main direction and the cable plane; Among them, perform necessary inversion and rotation operations on the three-dimensional coordinates to ensure that the x, y, and z of the point cloud coordinate system correspond one-to-one with the x, y, and z of the local coordinate system of the cable cylindrical surface equation; Take any cable segment OAB, the chord length between points OB is l, the chord direction and the horizontal direction angle is θ, and the sag of the stay cable is f; 4) Calculate the cable length l and the inclination angle θ of each cable; 5) According to the spatial surface equation of the stay cable, use the point cloud data for equation fitting to obtain the cable force T; 6) Judge whether the fitting degree meets the requirements; If it does not meet the requirements, then count the projection distances of each point to the curved cylindrical surface; Use statistical knowledge to further eliminate points with larger deviations; Use the reduced point cloud data to repeat steps 3) to 5) to obtain a new cable force.

2. The cable force detection method for fitting a cylindrical surface using three-dimensional point clouds according to claim 1, wherein: In step 1), a three-dimensional laser scanner is used to obtain the original three-dimensional point cloud data.

3. The cable force detection method for fitting a cylindrical surface using three-dimensional point clouds according to claim 1, characterized in that: In step 1), it is selected to obtain the point cloud data under the weather conditions of no wind or gentle breeze.

4. The cable force detection method for fitting a cylindrical surface using three-dimensional point clouds according to claim 1, characterized in that: In step 2), the preprocessing includes using clustering and / or registration algorithms to eliminate points that are obviously not near the surface of the cylindrical surface.

5. A cable force detection method for fitting a cylindrical surface using three-dimensional point clouds according to claim 1, characterized in that: In step 2), for the point cloud data that is not scanned in one station, use the clustering algorithm to separate different data piles.

6. The cable force detection method for fitting a cylindrical surface using three-dimensional point clouds according to claim 1, characterized in that: In step 3), the PCA method is used to determine the main direction, and the first principal component represents the main direction of the stay cable.

7. A cable force detection method for fitting a cylindrical surface using three-dimensional point clouds according to claim 1, characterized in that: In step 5), the sum of the squares of the distances from the data points to the cylindrical surface is minimized to optimize the parameters.

8. A cable force detection method for fitting a cylindrical surface using three-dimensional point clouds according to claim 1, characterized in that: In step 6), the coefficient of determination, mean square error, and residual are used as goodness-of-fit parameters to judge whether to iterate again.

9. A cable force detection method for fitting a cylindrical surface using three-dimensional point clouds according to claim 1, characterized in that: In step 6), it also has the content of identifying the inclination angle or linear density.

Images