Point-surface combined large-area foundation settlement high-precision measurement method

Abstract

The invention provides a point-surface combined large-area foundation settlement high-precision measurement method, and relates to the technical field of foundation measurement methods. A large-area soft foundation treatment ground surface settlement DEM is obtained through aerial survey of an unmanned aerial vehicle, then based on stacking material surface settlement measurement data of limited points and high-precision sensing measurement data of surface settlement of a foundation buried before stacking, errors are eliminated through fitting by means of the least square principle through a point-surface combination method, and the surface settlement measurement data of the stacked material are obtained. The settlement error generated by self-compression of the stacked material is eliminated, and then the large-area settlement measurement precision error is eliminated, so that the large-area foundation surface layer settlement DEM with relatively high precision is obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of foundation measurement methods, and in particular to a point-surface combined high-precision measurement method for large-area foundation settlement. Background Art

[0002] In soft foundation treatment projects, it is essential to monitor the settlement of the foundation. On-site settlement monitoring can ensure the safety of construction and prevent the damage and instability of the foundation. At the same time, the monitoring data obtained is also an important basis for calculating the current foundation consolidation degree, determining the preloading time, evaluating the foundation treatment effect, and estimating the post-construction settlement, providing a reference for the arrangement of the entire construction period.

[0003] Although traditional single-point measurement and monitoring methods have been developed and mature with high accuracy, they consume a lot of manpower and material resources, and it is difficult to grasp the overall settlement information of the foundation treatment area from a global perspective. UAV LiDAR (Light Detection and Range-ing, LiDAR) can obtain high-density point cloud data by scanning the target area at low altitude and in multiple time periods. After point cloud filtering, interpolation and difference calculation, the elevation change of the target area can be obtained. Using UAV-mounted LiDAR for settlement monitoring is not only efficient, low in manpower consumption, and not affected by construction and terrain, but also can better grasp the surface characteristics of settlement.

[0004] At present, UAV-mounted LiDAR has been widely used in the fields of topographic surveying and mapping, power inspection, land resources planning and remediation, and surface subsidence monitoring in mining areas. However, the research and application of settlement monitoring in soft soil foundation treatment projects are very rare. In particular, UAV-mounted LiDAR scans the terrain surface. When the preloading method is used for foundation treatment, the foundation is filled with thick piles of material. How to eliminate the error of pile settlement is a key issue that needs to be solved. At the same time, there are problems such as low accuracy in constructing digital elevation models (DEMs) using data collected by drones. Therefore, how to improve the accuracy of DEMs obtained by drone aerial surveys is a hot and difficult issue in using drones for settlement surface monitoring. Summary of the invention

[0005] The invention discloses a point-surface combined large-area foundation settlement high-precision measurement method, aiming to solve the problem of low precision in settlement surface monitoring using unmanned aerial vehicles.

[0006] The present invention adopts the following scheme: A high-precision measurement method for large-area foundation settlement combining points and surfaces comprises the following steps: S1. Plan the route through the drone and obtain the three-dimensional point cloud data of the area to be measured by aerial survey; S2, obtaining a ground point cloud by filtering the acquired three-dimensional point cloud, constructing a ground DEM by using the ground point cloud using an irregular triangulated network interpolation algorithm, and subtracting the ground DEMs obtained at different times to obtain a surface settlement DEM; S3, establish the control points in the area to be measured, and obtain the spatial coordinates of each control point by taking the average value of multiple consecutive measurements using the RTK equipment; and measure the surface settlement of the foundation at the control point; S4, performing a preloading treatment, measuring the plane coordinates of the surface layer of the pile material at the reference point and the elevation value of the location, and obtaining the surface settlement value of the surface layer of the pile material at the point in different periods by performing a difference solution based on the change of the elevation value; S5. The surface settlement value obtained by the GNSS equipment at the reference point in the same period is subtracted from the surface settlement value measured by the settlement plate at the same position to obtain the settlement of the pile material itself in the time period; the least squares principle is used to fit the settlement of the pile material itself at each reference point to obtain the settlement DEM of the pile material itself of the entire surface to be measured, and the settlement DEM of the pile material itself is subtracted from the surface settlement DEM, so as to construct the surface settlement DEM in which the settlement of the pile material itself is eliminated; S6. Establish uniform correction points within the area to be measured, and obtain the spatial coordinates of each correction point by taking the average value of multiple consecutive measurements using the RTK device; S7. Obtain the settlement difference of the settlement plate at each correction point within a certain observation time period, that is, the surface settlement amount; subtract the surface settlement difference obtained by the settlement plate from the surface settlement amount at the corresponding correction point coordinates in the surface settlement DEM that eliminates the settlement of the stockpile itself constructed in step S5 to obtain the surface settlement error value; perform plane fitting on the surface settlement error values ​​and their plane coordinates at all correction points using the least squares principle to obtain the surface settlement error correction surface DEM of the entire settlement observation surface; then superimpose the surface settlement DEM that eliminates the settlement of the stockpile itself and the surface settlement error correction surface DEM to perform difference calculation to reduce the surface settlement DEM error caused by UAV aerial survey.

[0007] Furthermore, the specific steps for obtaining the three-dimensional point cloud and ground point cloud of the area to be measured by drone aerial survey are as follows: S11, planning the flight route and flight parameters in the test area; S12, flying the drone and using the laser radar carried by the drone to scan the surface of the area to be measured; S13, using DJI Terra UAV point cloud processing software to export the three-dimensional point cloud of the area to be measured. Furthermore, the filtering algorithm of the irregular triangulated network adopts a progressive triangulated network filtering algorithm, and the specific steps of filtering the three-dimensional point cloud to obtain the ground point cloud are as follows: S21: Determine the threshold of the filtering model; S22: Divide the survey area into blocks, and select the lowest point of each block in the survey area as a seed point to construct an initial triangulation network; S23: If the distance from the pending point to the nearest triangle and the angle between the line connecting the pending point and the nearest triangle vertex and the triangle are both less than the set threshold, the ground point is included; S24: Repeat steps S22 and S23, and the operation ends when no new points are added to the triangulated network.

[0008] Furthermore, the specific steps of constructing a surface layer settlement model that eliminates the settlement of the pile material itself are as follows: S31: establishing a reference point in the test area, measuring the coordinates of the reference point multiple times continuously by using an RTK device, and taking an average value to obtain the real space coordinates of the reference point; S32: before starting to pile up, a settlement plate is buried under the foundation surface layer at the established reference point, and the settlement plate is connected to an automatic settlement sensor to obtain high-precision surface layer settlement; S33: After the preloading foundation treatment is fully loaded, a settlement rod is set on the surface at the reference point, and the GNSS device is installed on the settlement rod; the settlement of the surface drives the settlement rod to move downward, thereby causing the GNSS device to move downward together. At the same time, the GNSS device measures its own position every hour and uploads data to achieve continuous elevation monitoring of the surface. The surface settlement value of the reference point during the period can be obtained by subtracting the elevation values ​​at different time nodes; S34: For the same reference point, the surface settlement value obtained by GNSS within a certain period of time minus the surface settlement value obtained by the settlement plate is the settlement value of the piled material at the reference point; the settlement of the piled material at all reference points is obtained, and the DEM of the settlement of the piled material in the entire test area is obtained by plane fitting based on the principle of least squares method; S35: superimposing the surface settlement DEM and the settlement DEM of the piled material itself and performing a difference operation to obtain a surface settlement DEM with the settlement of the piled material eliminated on the basis of the surface settlement DEM.

[0009] Furthermore, in step S7, the specific process of using the elevation error value surface model to perform accuracy correction on the surface settlement DEM is as follows: S71: establishing 50m×50m correction points evenly distributed in the area to be measured; measuring the coordinates of the correction points multiple times continuously by using an RTK device, and taking the average value to obtain the real space coordinates of the correction points; S72: The surface settlement error value at the correction point is obtained by subtracting the measured surface settlement value of the settlement plate at each correction point from the surface settlement value at the corresponding coordinate obtained from the surface settlement DEM. The surface settlement error values ​​of all correction points are combined with their plane coordinates to obtain the surface settlement error correction surface DEM of the entire measured area through the least squares plane fitting principle. S73: Superimpose and subtract the surface settlement DEM and the surface settlement correction surface DEM to obtain a final surface settlement DEM. Furthermore, the method further comprises step S8: a verification process, wherein the verification process comprises the following steps: S81: Establish checkpoints in the area to be measured, measure the coordinates of the checkpoints for multiple times through RTK equipment, and take the average value to obtain the real space coordinates of the checkpoints; some checkpoints monitor the settlement of the piled material by installing GNSS equipment and burying settlement plates by driving settlement rods; some checkpoints monitor the settlement of the surface layer of the foundation by burying settlement plates under the surface layer of the foundation; S82: deriving the pile self-settlement value at the corresponding checkpoint position in the pile self-settlement DEM, comparing it with the pile self-settlement value actually measured at the checkpoint, and analyzing the accuracy index to obtain the accuracy of the pile self-settlement DEM; S83: Export the surface settlement value at the corresponding checkpoint in the final surface settlement DEM, compare it with the measured surface settlement value at the checkpoint, and analyze the accuracy index to obtain the settlement monitoring accuracy of the surface settlement DEM after the error is corrected.

[0010] Furthermore, in step S2, a ground point cloud is obtained based on a filtering algorithm of an irregular triangulated network.

[0011] Furthermore, in step S3, an automatic settlement monitoring sensor is buried before starting loading, and a settlement plate is placed below the foundation surface layer at the reference point. The settlement plate is connected to the automatic settlement monitoring sensor to measure the surface settlement of the foundation at the reference point.

[0012] Furthermore, in step S4, after the graded loading is completed, a settlement rod is set at the control point, and a GNSS device is installed on the settlement rod to determine the plane coordinates of the surface layer of the loading material at the control point and the elevation value of the location by differential GNSS technology.

[0013] Beneficial effects: The ground point cloud was obtained by irregular triangulation algorithm, and the surface settlement DEM was constructed. On this basis, the interference of the settlement of the piled material itself in the preloading foundation treatment condition on the foundation settlement monitoring was eliminated. Furthermore, the surface settlement error surface was constructed, and the error of the surface settlement DEM model was reduced by DEM difference calculation, which improved the accuracy of the settlement surface monitoring by UAV. The point-surface combination method realized the rapid, high-precision and low-cost measurement of large-area foundation settlement. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 , point cloud map of the area to be measured obtained by drone aerial survey; Figure 2 ,Schematic diagram of point cloud before noise removal; Figure 3 ,The effect of point cloud after removing noise; Figure 4 , schematic diagram of a partial enlargement of the point cloud before extracting ground points through point cloud filtering; Figure 5 , Local magnification effect diagram after point cloud filtering extracts ground points; Figure 6 , the surface settlement DEM generated by point cloud filtering and interpolation algorithm; Figure 7 , spatial layout plan of control points; Figure 8 , the spatial layout plan of the correction points; Fig. 9 , spatial layout plan of checkpoints; Fig.10 , schematic diagram of the arrangement of the settling plate, the settling rod and the GNSS equipment in the embodiment; Fig.11 , schematic diagram of arrangement of the settling plate in the embodiment; Fig.12 , pile loading material self-settlement surface model; Fig.13 , DEM of foundation surface settlement after accuracy correction; DETAILED DESCRIPTION Combination Figures 1 to 13 As shown, this embodiment provides a method for high-precision measurement of large-area foundation settlement by combining points and surfaces, including the following steps: S1. Plan the route through the drone and obtain the three-dimensional point cloud data of the area to be measured by aerial survey; S2, filtering the acquired three-dimensional point cloud by a filtering algorithm based on an irregular triangulated network to obtain a ground point cloud, constructing a ground DEM by using the ground point cloud by an irregular triangulated network interpolation algorithm, and subtracting the ground DEMs obtained at different times to obtain a surface settlement DEM; S3, establish the control points in the area to be measured, and use the RTK equipment to continuously measure multiple times and take the average value to obtain the spatial coordinates of each control point; before starting to pile load, bury the automatic settlement monitoring sensor, the settlement plate is below the foundation surface layer at the control point position, and the settlement plate is connected to the automatic settlement monitoring sensor to measure the surface settlement of the foundation at the control point position; S4. After the preloading treatment is carried out and the graded loading is completed, a settlement rod is set at the control point, on which a GNSS device is installed, and the plane coordinates of the surface layer of the pile material at the control point and the elevation value of the location are measured by differential GNSS technology. The surface settlement value of the surface layer of the pile material at the point in different periods can be obtained by differential solution through the change of the elevation value; S5. The surface settlement value obtained by the GNSS equipment at the reference point in the same period is subtracted from the surface settlement value measured by the settlement plate at the same position to obtain the settlement of the pile material itself in the time period; the least squares principle is used to fit the settlement of the pile material itself at each reference point to obtain the settlement DEM of the pile material itself of the entire surface to be measured, and the settlement DEM of the pile material itself is subtracted from the surface settlement DEM, so as to construct the surface settlement DEM in which the settlement of the pile material itself is eliminated; S6. Establish uniform correction points within the area to be measured (e.g., 50 m × 50 m), and obtain the spatial coordinates of each correction point by taking the average value of multiple consecutive measurements using the RTK device; S7. Obtain the settlement difference of the settlement plate at each correction point within a certain observation time period, that is, the surface settlement amount. Subtract the surface settlement difference obtained by the settlement plate from the surface settlement amount at the corresponding correction point coordinates in the surface settlement DEM that eliminates the settlement of the stockpile itself constructed in step S5 to obtain the surface settlement error value. Perform plane fitting on the surface settlement error values ​​and their plane coordinates at all correction points using the least squares principle to obtain the surface settlement error correction surface DEM of the entire settlement observation surface. Then, superimpose the surface settlement DEM that eliminates the settlement of the stockpile itself and the surface settlement error correction surface DEM to perform difference calculation to reduce the surface settlement DEM error caused by UAV aerial survey, thereby improving the settlement surface monitoring accuracy.

[0015] In the above method, the surface DEM of different periods is first obtained by using UAV aerial survey of the area to be measured and filtering algorithm, and then the surface DEM of the area to be measured obtained at different periods is subtracted to obtain the surface settlement DEM. However, this surface settlement DEM does not eliminate the influence of the settlement of the pile material itself on the surface settlement DEM of the measured area. Therefore, it is necessary to eliminate the settlement value caused by the settlement of the pile material itself. To this end, a certain density of control points are selected in the area to be tested, and the plane coordinates and elevation values ​​of the ground at the control points are determined by differential GNSS technology. Since the surface settlement of the control points obtained by the GNSS equipment includes the settlement of the foundation surface layer and the settlement of the pile material itself piled thereon, it is necessary to subtract the surface settlement at the same position from the surface settlement obtained by the GNSS equipment in the same period to obtain the settlement of the pile material itself within the time period. The settlement of the pile material itself at multiple control points is fitted into the DEM of the pile material settlement itself. The surface settlement DEM is subtracted from the DEM of the pile material settlement itself to eliminate the error caused by the settlement of the pile material itself and obtain the surface settlement DEM.

[0016] At the same time, by burying the settlement plate under the foundation surface before loading, the surface settlement of the foundation at the reference point is measured; the surface settlement data obtained by the settlement plate in conjunction with the automatic settlement monitoring sensor is more accurate than the surface settlement accuracy obtained by drone aerial survey. If the surface settlement is only monitored by burying the settlement plate, it is impossible to grasp the differential settlement of the foundation treatment area from the surface shape as a whole. Therefore, drone aerial survey can be used to construct the surface settlement DEM. In order to improve the accuracy of the surface settlement DEM obtained by drone aerial survey, it is necessary to correct its errors. In this embodiment, a settlement plate is buried to obtain a high-precision foundation surface settlement value. The change in the displacement of the settlement plate at different time points based on the initial displacement is the surface settlement of the correction point within the time period; the surface settlement value at the corresponding position of the foundation surface settlement DEM is derived according to the plane coordinates (x, y) of the correction point, and the surface settlement error value at the correction point is obtained by subtracting the surface settlement value of the correction point from the surface settlement value at the corresponding coordinates of the surface settlement DEM; then the surface settlement error values ​​obtained at all correction points are combined with their plane coordinates to fit into a surface settlement error correction surface DEM, and finally the surface settlement DEM is superimposed and subtracted from the error correction surface DEM to obtain a surface settlement DEM with reduced error and improved accuracy. The surface settlement DEM here is the settlement surface obtained after the surface settlement DEM eliminates the influence of the settlement of the pile itself.

[0017] In the present embodiment, in the process of eliminating the error of the pile material itself and correcting the error, the relative height difference is adopted instead of the absolute height difference. Therefore, the scheme can be used in various occasions and is not affected by factors such as altitude.

[0018] In this embodiment, the specific steps of obtaining the three-dimensional point cloud of the area to be measured by drone aerial survey are as follows: S11, planning the flight route and flight parameters in the test area; The UAV route design mainly includes: altitude, heading and overlap rate. The process is as follows: Determine the flight altitude: According to the formula ; (1) The flight height H corresponding to a certain ground resolution is calculated, and the unit of H is meter. In formula (1), GSD is the ground resolution, and the unit is meter; f is the focal length of the camera lens, and the unit is millimeter; a is the pixel size of the camera, and the unit is micrometer; Overlap rate determination: Determined according to the topographic and geomorphic characteristics of the study area, the recommended setting of heading overlap rate is 80% and lateral overlap rate is 70%, which is suitable for most scenarios. For areas with large terrain fluctuations, when the overlap between the lowest and highest points of the terrain is too different, in order to ensure the overlap at the highest point, the overlap rate can be appropriately increased. For areas with small terrain fluctuations, such as plains, the overlap rate can be appropriately reduced to improve the efficiency of aerial surveys, but it is necessary to ensure that the heading overlap rate is not less than 65% and the lateral overlap rate is not less than 60%; Course determination: determined according to the shape of the study area and arranged along the main axis of the study area; Flight parameters mainly include: flight speed, altitude; Point cloud acquisition parameters mainly include point cloud density. Multiple aerial surveys are carried out at different time points. The drone automatically flies according to the arranged route mission, and the laser radar it carries scans the surface to obtain the point cloud in the area to be measured; then the DJI Terra drone point cloud processing software is used to perform preliminary screening of the point cloud, delete obvious noise points and outliers, obtain a three-dimensional point cloud, export the point cloud in las format, and store it in the computer.

[0019] The filtering algorithm of the irregular triangulated network adopts a progressive triangulated network filtering algorithm. The specific steps of filtering the three-dimensional point cloud to obtain the ground point cloud are as follows: S21: Determine the threshold of the filtering model; usually in hilly and gully areas, the iteration distance is selected as 1 m and the iteration angle is , in the windy beach area, the iteration distance is 0.5 m and the iteration angle is ; S22: Divide the survey area into blocks, and select the lowest point of each block in the survey area as a seed point to construct an initial triangulation network; S23: If the distance from the pending point to the nearest triangle and the angle between the line connecting the pending point and the nearest triangle vertex and the triangle are both less than the set threshold, the ground point is included; S24: repeat steps S22 and S23, and the operation ends when no new points are added to the triangulated network; The specific steps of constructing the surface DEM through the ground point cloud and subtracting the two phases of surface DEM to obtain the surface settlement DEM are as follows: The ground point cloud was selected to generate a DEM model using the irregular triangulated network (TIN) algorithm. The DEM models obtained at two different times were overlapped, and the raster calculator tool was used to perform difference calculations to obtain the surface settlement DEM within that time period.

[0020] The specific steps to construct a surface settlement model that eliminates the settlement of the pile material itself are as follows: S31: establishing a certain density of control points in the area to be measured, measuring the coordinates of the control points multiple times continuously by using an RTK device, and taking an average value to obtain the actual three-dimensional coordinates of the control points; S32: before starting the loading, a settlement plate is buried under the foundation surface layer at the established reference point, and the settlement plate is connected to an automatic settlement monitoring sensor to obtain the settlement of the foundation surface layer; S33: After the preloading foundation treatment is fully loaded, a settlement rod is set on the ground surface at the reference point. A fixing ring is arranged on the outer sleeve of the settlement rod. The fixing ring is stabilized on the ground by a tripod to limit the horizontal tilt of the settlement rod. The GNSS device is installed on the settlement rod; the settlement rod is displaced downward due to the settlement of the ground surface, thereby causing the GNSS device to move downward together. At the same time, the GNSS device measures its own position every hour and uploads data to realize continuous elevation monitoring of the ground surface. The settlement value of the reference point in the period can be obtained by subtracting the elevation values ​​at different time nodes. The GNSS device can continuously measure the three-dimensional spatial coordinates of the reference point according to the set measurement time interval and upload them to the receiver. The three-dimensional spatial coordinates of the reference point in the target time period are averaged to obtain the actual three-dimensional coordinates of the reference point. The change in the Z coordinate along the gravity direction reflects the settlement of the ground surface. The surface settlement value of the reference point in the period can be obtained by subtracting the elevation values ​​of the reference point at different time nodes. The settlement value of the pile itself can be obtained by subtracting the settlement value of the foundation surface layer measured by the settlement plate from the surface settlement value. S34: For the same reference point, the surface settlement value within a certain period of time minus the surface layer settlement value is the pile settlement value at the reference point; thus, the pile settlement value at all reference points is obtained, and the pile settlement value DEM of the entire area to be measured is obtained by plane fitting based on the principle of least squares method; S35: using a raster calculator tool to perform a difference operation between the surface settlement DEM and the pile settlement DEM, so as to obtain a surface settlement DEM in which the pile settlement is eliminated on the basis of the surface settlement DEM.

[0021] In step S7, the specific process of using the elevation error value surface model to perform accuracy correction on the foundation surface settlement DEM is as follows: S71: establishing correction points of a certain density evenly distributed in the area to be measured; measuring the coordinates of the control points multiple times continuously by using an RTK device, and taking the average value to obtain the actual three-dimensional coordinates of the correction points; S72: The surface settlement error value at the correction point is obtained by subtracting the settlement measurement value of the buried settlement plate at the correction point from the foundation surface settlement value at the corresponding coordinate of the surface settlement DEM. The settlement error values ​​at all correction points are used to obtain the surface settlement error correction surface DEM of the entire measured area through the least squares plane fitting principle. Specifically, the spatial three-dimensional coordinates of each correction point are measured multiple times by RTK equipment, and the three-dimensional coordinates are obtained by taking the average value, so as to obtain the plane coordinates (x, y) in the three-dimensional coordinates to determine the plane position of the correction point. The displacement of the settlement plate at different time points is subtracted based on the initial displacement to obtain the foundation surface settlement of the correction point within the time period. The foundation surface settlement value at the corresponding position of the foundation settlement DEM is derived according to the plane coordinates (x, y) of the correction point, and the surface settlement error value at the correction point is obtained by subtracting the foundation surface settlement value at the correction point from the foundation surface settlement value at the corresponding coordinate of the foundation surface settlement DEM. Thus, the foundation surface settlement error values ​​at all correction points in the measured area are obtained. S73: Using a raster calculator tool, the foundation surface settlement DEM and the surface settlement error correction surface DEM are superimposed and subtracted to obtain a final foundation surface settlement DEM; In one embodiment, step S8 is further included: verification process, which includes the following process: Establish a certain number of checkpoints in the area to be tested, and some of the checkpoints are used for quality inspection of the settlement model of the pile itself. Use GNSS equipment and buried settlement plates to monitor the settlement of the pile itself: at the checkpoint location, a settlement plate is buried in the surface layer of the foundation below the pile, and the surface settlement of the foundation is monitored based on the automatic settlement monitoring sensor; at the checkpoint location, a settlement rod is driven into the surface layer of the pile, and a fixed ring is placed on the settlement rod. The fixed ring is stabilized on the ground by a tripod to limit the horizontal tilt of the settlement rod. The settlement rod moves as the surface sinks; the GNSS equipment is fixed on the settlement rod and moves as the settlement rod sinks to monitor the settlement of the surface; the surface settlement value at the checkpoint is subtracted from the foundation surface settlement value to obtain the settlement value of the pile itself at the checkpoint; the settlement value at the corresponding point of the DEM model of the pile itself settlement is derived according to the checkpoint coordinates, and it is compared with the settlement value of the checkpoint to obtain the accuracy of the pile itself settlement model and judge its reliability; Another part of the checkpoints is used for quality inspection of the surface model of the foundation surface settlement after error correction. The surface settlement of the foundation is monitored by burying settlement plates under the foundation surface: the settlement plates are buried in the foundation surface below the piled material at the checkpoint location, and the surface settlement is monitored based on the automatic settlement monitoring sensor. The settlement of the foundation drives the settlement plate to sink, and the settlement plate sinking is reflected in the displacement change of the automatic settlement monitoring sensor. The change of the automatic settlement monitoring sensor within a time interval is the foundation surface settlement value within the time period; multiple measurements are performed at the checkpoint using the RTK equipment to obtain the spatial three-dimensional coordinates of the checkpoint, and the two-dimensional coordinates (x, y) are averaged to obtain the plane coordinates of the checkpoint; based on the plane coordinates, the foundation surface settlement DEM after error correction is derived, and the foundation surface settlement value at this point is compared with the settlement value of the checkpoint to determine the accuracy.

[0022] The specific layout of the control points, correction points and check points in the area to be tested is explained as follows: The optimal layout scheme that takes into account the monitoring accuracy, construction and raw material costs: The reference points are used for plane fitting of the settlement surface of the piled material itself. The soil properties of the piled material are uniform, and its natural consolidation settlement will not produce large differential settlement. The reference points are evenly arranged around the test area and densely arranged inside. The correction points involve the accuracy correction of the foundation surface DEM, which requires a higher density of distribution. They should be evenly distributed in the area to be measured, with a distribution interval of 50×50 meters. The check points are used to check the accuracy correction effect of the settlement surface of the pile material itself and the settlement surface of the foundation surface. They are evenly arranged and have a lower density than the control points and correction points.

[0023] In this embodiment, the triangulated irregular network (TIN) filtering algorithm uses the structure of irregular triangulated network to connect discrete point clouds. TIN filtering avoids the error caused by resampling, the original laser foot point elevation information will not be destroyed, and the accuracy is guaranteed. The algorithm operation process is mainly divided into three steps: first, the survey area is divided into blocks, and the lowest point of each block in the survey area is selected as the seed point to construct the initial triangulated network. Then, other laser foot points are judged according to certain criteria, and the points that meet the judgment criteria are encrypted into the initial triangulated network. The judgment criteria are: the distance from the to-be-determined point to the nearest triangle and the angle between the line connecting the to-be-determined point and the nearest triangle vertex and the triangle face are both less than the set threshold. The filtering process is performed by iterative encryption of triangles. Finally, this process is repeated and iterated. The operation ends when no new points are added to the triangulated network.

[0024] In this embodiment, the self-settlement surface model of the pile load is obtained by performing quadratic polynomial plane fitting on the self-settlement of the pile load at the reference point; the specific process is: The difference between the surface settlement measured by the GNSS equipment at the reference point and the settlement of the foundation surface layer measured by the settlement plate and their plane coordinates (x, y) form a functional relationship: △Z=f(x,y), where f(x,y) is a quadratic polynomial function.

[0025] The least squares matrix corresponding to the polynomial-based settlement fitting plane is: ; in( , ) ,( , ) ,....,( , ) is the plane coordinate of the reference point, △ ,△ ,...,△ is the settlement value of the pile material itself at the reference point; , , , ..., The model parameters can be obtained by solving the normal equations, so as to fit the settlement values ​​of the pile at other points in the test area.

[0026] The specific fitting of the correction point elevation error surface is as follows: The basic idea of ​​​​the present embodiment to improve the measurement accuracy is to obtain the surface settlement value of the foundation surface settlement surface obtained by the UAV and the surface settlement value at the corresponding point through the surface settlement value at a set of elevation correction points within the measured area. The error of the surface settlement at the correction point is minimized by a fitting method based on the least squares principle, and the error value of the discrete point is fitted into the error value of the surface, that is, the entire error surface is obtained; on the basis of the original foundation surface settlement DEM, the error is eliminated by subtracting the obtained error surface.

[0027] The difference between the surface settlement at the correction point and the surface settlement at the corresponding point of the foundation surface settlement DEM is the surface settlement error value. The surface settlement error value at the correction point is fitted by four-parameter surface fitting to obtain the surface settlement error surface model. Finally, the original surface settlement DEM is corrected according to the surface settlement error to obtain the corrected final surface settlement DEM.

[0028] In a preferred embodiment, the accuracy evaluation index of the checkpoint is: assuming that the surface settlement value of the checkpoint is: ; The surface settlement value at the corresponding point on the surface settlement model is: ; (1) The mean error MRE refers to the average value of the error between the surface settlement value at the checkpoint and the foundation surface settlement model. The formula for the mean error is: ; (2) Mean absolute error (MAE) is an indicator that can reflect the actual deviation, which refers to the average of the absolute values ​​of the deviations between the model settlement and the checkpoint settlement.

[0029] ; (3) Mean error It can evaluate the influence of discrete data on the settlement deviation of the settlement model. The above three evaluation indicators are negatively correlated with the accuracy of the surface settlement DEM, that is, the higher the accuracy of the surface settlement DEM, the smaller the value of the evaluation indicator.

[0030] ; The above three error algorithms can be used for inspection and verification.

[0031] It should be understood that the above are only preferred implementations of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention.

[0032] The above description of the drawings used in the implementation manner only illustrates certain embodiments of the present invention and should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

Claims

1. A high-precision measurement method for large-area foundation settlement combining points and surfaces, characterized in that: The steps include: S1. Plan the route through the drone and obtain the three-dimensional point cloud data of the area to be measured by aerial survey; S2, obtaining a ground point cloud by filtering the acquired three-dimensional point cloud, constructing a ground DEM by using the ground point cloud using an irregular triangulated network interpolation algorithm, and subtracting the ground DEMs obtained at different times to obtain a surface settlement DEM; S3, establish the control points in the area to be measured, and obtain the spatial coordinates of each control point by taking the average value of multiple consecutive measurements using the RTK equipment; and measure the surface settlement of the foundation at the control point; S4, performing a preloading treatment, measuring the plane coordinates of the surface layer of the pile material at the reference point and the elevation value of the location, and obtaining the surface settlement value of the surface layer of the pile material at the point in different periods by performing a difference solution based on the change of the elevation value; S5. The surface settlement value obtained by the GNSS equipment at the reference point in the same period is subtracted from the surface settlement value measured by the settlement plate at the same position to obtain the settlement of the pile material itself in the time period; the least squares principle is used to fit the settlement of the pile material itself at each reference point to obtain the settlement DEM of the pile material itself of the entire surface to be measured, and the settlement DEM of the pile material itself is subtracted from the surface settlement DEM, so as to construct the surface settlement DEM in which the settlement of the pile material itself is eliminated; S6. Establish uniform correction points within the area to be measured, and obtain the spatial coordinates of each correction point by taking the average value of multiple consecutive measurements using the RTK device; S7. Obtain the settlement difference of the settlement plate at each correction point within a certain observation time period, that is, the surface settlement amount; subtract the surface settlement difference obtained by the settlement plate from the surface settlement amount at the corresponding correction point coordinates in the surface settlement DEM that eliminates the settlement of the stockpile itself constructed in step S5 to obtain the surface settlement error value; perform plane fitting on the surface settlement error values ​​and their plane coordinates at all correction points using the least squares principle to obtain the surface settlement error correction surface DEM of the entire settlement observation surface; then superimpose the surface settlement DEM that eliminates the settlement of the stockpile itself and the surface settlement error correction surface DEM to perform difference calculation to reduce the surface settlement DEM error caused by UAV aerial survey.

2. The point-surface combined large-area foundation settlement high-precision measurement method according to claim 1 is characterized in that: The specific steps to obtain the 3D point cloud and ground point cloud of the area to be measured through UAV aerial survey are as follows: S11, planning the flight route and flight parameters in the test area; S12, flying the drone and using the laser radar carried by the drone to scan the surface of the area to be measured; S13, using DJI Terra UAV point cloud processing software to export the three-dimensional point cloud of the area to be measured.

3. The point-surface combined large-area foundation settlement high-precision measurement method according to claim 1 is characterized in that: The filtering algorithm of the irregular triangulated network adopts a progressive triangulated network filtering algorithm. The specific steps of filtering the three-dimensional point cloud to obtain the ground point cloud are as follows: S21: Determine the threshold of the filtering model; S22: Divide the survey area into blocks, and select the lowest point of each block in the survey area as a seed point to construct an initial triangulation network; S23: If the distance from the pending point to the nearest triangle and the angle between the line connecting the pending point and the nearest triangle vertex and the triangle are both less than the set threshold, the ground point is included; S24: Repeat steps S22 and S23, and the operation ends when no new points are added to the triangulated network.

4. The point-surface combined large-area foundation settlement high-precision measurement method according to claim 1 is characterized in that: The specific steps to construct a surface settlement model that eliminates the settlement of the pile material itself are as follows: S31: establishing a reference point in the test area, measuring the coordinates of the reference point multiple times continuously by using an RTK device, and taking an average value to obtain the real space coordinates of the reference point; S32: before starting to pile up, a settlement plate is buried under the foundation surface layer at the established reference point, and the settlement plate is connected to an automatic settlement sensor to obtain high-precision surface layer settlement; S33: After the preloading foundation treatment is fully loaded, a settlement rod is set on the surface at the reference point, and the GNSS device is installed on the settlement rod; the settlement of the surface drives the settlement rod to move downward, thereby causing the GNSS device to move downward together. At the same time, the GNSS device measures its own position every hour and uploads data to achieve continuous elevation monitoring of the surface. The surface settlement value of the reference point during the period can be obtained by subtracting the elevation values ​​at different time nodes; S34: For the same reference point, the surface settlement value obtained by GNSS within a certain period of time minus the surface settlement value obtained by the settlement plate is the settlement value of the piled material at the reference point; the settlement of the piled material at all reference points is obtained, and the DEM of the settlement of the piled material in the entire test area is obtained by plane fitting based on the principle of least squares method; S35: superimposing the surface settlement DEM and the settlement DEM of the piled material itself and performing a difference operation to obtain a surface settlement DEM with the settlement of the piled material eliminated on the basis of the surface settlement DEM.

5. The point-surface combined large-area foundation settlement high-precision measurement method according to claim 1 is characterized in that: In step S7, the specific process of using the elevation error value surface model to perform accuracy correction on the surface settlement DEM is as follows: S71: establishing 50m×50m correction points evenly distributed in the area to be measured; measuring the coordinates of the correction points multiple times continuously by using an RTK device, and taking the average value to obtain the real space coordinates of the correction points; S72: The surface settlement error value at the correction point is obtained by subtracting the measured surface settlement value of the settlement plate at each correction point from the surface settlement value at the corresponding coordinate obtained from the surface settlement DEM. The surface settlement error values ​​of all correction points are combined with their plane coordinates to obtain the surface settlement error correction surface DEM of the entire measured area through the least squares plane fitting principle. S73: Superimpose and subtract the surface settlement DEM and the surface settlement correction surface DEM to obtain a final surface settlement DEM.

6. The point-surface combined large-area foundation settlement high-precision measurement method according to claim 1 is characterized in that: The process also includes step S8: verification process, which includes the following steps: S81: Establish checkpoints in the area to be measured, measure the coordinates of the checkpoints for multiple times through RTK equipment, and take the average value to obtain the real space coordinates of the checkpoints; some checkpoints monitor the settlement of the piled material by installing GNSS equipment and burying settlement plates by driving settlement rods; some checkpoints monitor the settlement of the surface layer of the foundation by burying settlement plates under the surface layer of the foundation; S82: deriving the pile self-settlement value at the corresponding checkpoint position in the pile self-settlement DEM, comparing it with the pile self-settlement value actually measured at the checkpoint, and analyzing the accuracy index to obtain the accuracy of the pile self-settlement DEM; S83: Export the surface settlement value at the corresponding checkpoint in the final surface settlement DEM, compare it with the measured surface settlement value at the checkpoint, and analyze the accuracy index to obtain the settlement monitoring accuracy of the surface settlement DEM after the error is corrected.

7. The point-surface combined large-area foundation settlement high-precision measurement method according to claim 1 is characterized in that: In step S2, a ground point cloud is obtained based on a filtering algorithm of an irregular triangulated network.

8. The point-surface combined large-area foundation settlement high-precision measurement method according to claim 1 is characterized in that: In step S3, an automatic settlement monitoring sensor is buried before starting loading, and a settlement plate is located below the foundation surface layer at the reference point. The settlement plate is connected to the automatic settlement monitoring sensor to measure the surface settlement of the foundation at the reference point.

9. The point-surface combined large-area foundation settlement high-precision measurement method according to claim 1 is characterized in that: In step S4, after the graded loading is completed, a settlement rod is set at the control point, and a GNSS device is installed on the settlement rod to determine the plane coordinates of the surface layer of the loading material at the control point and the elevation value of the location by differential GNSS technology.

Images