Slope group anchor load remote sensing inversion method based on non-uniform stiffness field reconstruction

Abstract

The present application relates to the field of geotechnical engineering safety monitoring, and specifically discloses a slope group anchor load remote sensing inversion method based on non-uniform stiffness field reconstruction, comprising the following steps: selecting a reference anchor cable on the anchored slope to test real-time load values, and obtaining remote sensing data on the anchored slope; extracting the axial micro-deformation of each reference anchor cable and the anchor cable to be measured within the monitoring period; for each reference anchor cable, using the real load value and the remote sensing deformation value synchronously collected, calculating the in-situ apparent stiffness at its location; deducing the stiffness distribution of the unmonitored area, and constructing a continuous non-uniform stiffness field model covering the entire slope; performing load inversion on the anchor cable to be measured without a load cell; generating a group anchor load cloud chart, and identifying stress relaxation areas or stress concentration areas. The present application can fully consider the spatial variability of the anchoring system stiffness, and can realize load inversion across engineering services through feature migration.

Description

Technical Field

[0001] This invention relates to the field of geotechnical engineering safety monitoring, and in particular to a remote sensing inversion method for slope group anchor load based on non-uniform stiffness field reconstruction. Background Technology

[0002] Currently, the safety evaluation of anchoring systems for steep slopes faces common key problems such as local monitoring blind spots, limitations in parametric characterization, and distortion in mechanical inversion, specifically manifested in the following ways:

[0003] (1) Sparse layout of point measuring points, resulting in spatial monitoring blind spots: Traditional anchor cable force gauges have a monitoring coverage rate of less than 5% in actual projects due to high manufacturing costs and difficult maintenance. Such low-density point monitoring networks are difficult to fully reflect the evolution of the stress state of the slope as a whole, and are prone to ignoring the risk of gradual failure of the group anchors in local areas, thus limiting the effectiveness of overall safety management.

[0004] (2) The lack of direct mapping between remote sensing deformation parameters and stress state leads to limitations in characterization: It is technically feasible to obtain millimeter-level micro-deformation fields of the entire slope using high-precision remote sensing technologies such as ground-based synthetic aperture radar (GB-InSAR) and three-dimensional laser scanning (LiDAR). However, the core control index for evaluating the slope support status is the actual internal tensile force of the support structure, rather than simply the external apparent displacement. The direct conversion of deformation parameters to load parameters lacks clear mechanistic mapping support.

[0005] (3) The calculation model is oversimplified and fails to reflect the spatial variability of stiffness: Existing stress inversion methods are usually based on simplified linear elastic models, assuming that the overall stiffness (K) of the slope is a spatially uniform constant. However, in actual engineering, due to the coupled influence of multiple factors such as the distribution of geological structures, the degree of rock weathering, the quality of anchoring grouting, and the construction technology of lattice support structures, the apparent stiffness of the anchoring system at different spatial locations has significant spatial non-uniformity and variability. If a single stiffness coefficient is used to perform full-field inversion, significant systematic errors will be introduced, reducing the accuracy of the safety assessment.

[0006] Furthermore, for a large number of existing aging slopes without any sensor equipment, their safety assessment faces the problem of missing prior benchmark data required for inversion. Such existing projects without initial benchmarks cannot directly use conventional methods to deduce their internal stress state, thus limiting the general application and promotion of advanced remote sensing inversion technology in a wide range of existing slope projects. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a remote sensing inversion method for slope group anchor load based on non-uniform stiffness field reconstruction. This method can fully consider the spatial variability of the stiffness of the anchoring system and can achieve load inversion across engineering services through feature transfer.

[0008] The objective of this invention is achieved through the following technical solution: a remote sensing inversion method for slope group anchor loads based on non-uniform stiffness field reconstruction, comprising the following steps:

[0009] Step S1: Select a reference anchor cable on the anchored slope to test the real-time load value and obtain remote sensing data on the anchored slope;

[0010] Step S2: Based on remote sensing data, target identification and error correction are performed, and the axial micro-deformation of each reference anchor cable and the anchor cable to be tested is extracted during the monitoring period;

[0011] Step S3: For each reference anchor cable, calculate the in-situ apparent stiffness at its location using the synchronously acquired real load value and remotely sensed deformation.

[0012] Step S4: Deduce the stiffness distribution in the unmonitored area and construct a continuous non-uniform stiffness field model covering the entire slope;

[0013] Step S5: Perform load inversion on the anchor cable to be tested that is not equipped with a force gauge;

[0014] Step S6: Generate a group anchor load cloud map and identify stress relaxation zones or stress concentration zones.

[0015] The beneficial effects of this invention are: by constructing a non-uniform stiffness field, this invention approximates the actual distribution of mechanical properties of rock and soil, effectively solving the huge inversion error caused by assuming uniform stiffness, realizing remote detection of anchor cables in steep slopes and dangerous areas, and proposing a "cross-engineering feature migration" mechanism, which can scientifically and quantitatively evaluate old, data-free, and sensor-free "three-no" slopes using only ground-penetrating radar. Attached Figure Description

[0016] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description.

[0018] like Figure 1 As shown, the remote sensing inversion method for slope group anchor load based on non-uniform stiffness field reconstruction includes the following steps:

[0019] Step S1: Select a reference anchor cable on the anchored slope to test the real-time load value and obtain remote sensing data on the anchored slope;

[0020] Several anchor cables equipped with anchor cable dynamometers and functioning properly were selected on the anchored slope as reference anchor cables, and their synchronous real-time load values ​​F were obtained.ref ;

[0021] Deploy ground-based interferometric radar (GB-InSAR) or ground-based 3D laser scanner, set the scanning cycle and spatial resolution, and perform a full-coverage scan of the entire slope anchorage area to obtain full-field initial point cloud data containing phase information or 3D coordinates as remote sensing data.

[0022] Step S2: Based on remote sensing data, perform target identification and error correction, and extract the axial micro-deformation during the monitoring period;

[0023] S201. Processing remote sensing data: Identify the center of the anchor head or anchor block area of ​​all anchor cables using a point cloud segmentation algorithm; select natural exposed rock masses located outside the slope excavation influence area, without obvious geological tectonic activity, and with high radar reflection coherence as "stable bedrock areas" (i.e., zero-deformation reference points); use the phase fluctuation of this stable bedrock area as background noise, and perform differential correction using an atmospheric phase screen (APS) compensation algorithm or a point cloud registration algorithm to deduct atmospheric delay disturbances and systematic errors caused by instrument instability, thereby obtaining the absolute deformation field;

[0024] S202. Extracting Axial Micro-deformation: Obtain the spatial inclination and azimuth of each anchor cable based on the slope design drawings or 3D model. From the absolute deformation field obtained in step S201, extract the one-dimensional deformation D corresponding to the center of the anchor head or anchor block area of ​​each anchor cable along the radar line of sight. LOS The deformation is transformed into the true micro-deformation D along the axial direction of each reference anchor cable and the anchor cable under test by projection onto the cosine matrix of the spatial geometric direction. axial The projection transformation formula is as follows:

[0025]

[0026] In the formula, D axial D represents the actual micro-deformation of the anchor cable along its axial direction. LOS For the one-dimensional deformation along the radar line of sight; (l r ,m r , n r ) represents the direction cosine of the radar line-of-sight vector in a three-dimensional Cartesian coordinate system; (l a , m a , n a The direction cosine of the anchor cable axial vector in the same coordinate system is derived from the spatial inclination and azimuth angles.

[0027] Step S3: For each reference anchor cable, calculate the in-situ apparent stiffness at its location using the real-time load value collected synchronously and the deformation measured by remote sensing.

[0028] For each reference anchor cable, the real-time load value F collected synchronously is used. ref And the axial micro-deformation D extracted in step S202 ref The in-situ apparent stiffness K at its location is calculated using the following formula. local :

[0029] K local = F ref / D ref

[0030] In the formula, K local For in-situ apparent stiffness; F ref D represents the real-time load value of the reference anchor cable. ref The axial micro-deformation of the reference anchor cable;

[0031] The in-situ apparent stiffness is used as a comprehensive indicator reflecting the elastic modulus of the free section of the anchor cable, the compression modulus of the soil and rock under the anchor, and the stiffness of the lattice beam structure.

[0032] Step S4: Deduce the stiffness distribution in the unmonitored area and construct a continuous non-uniform stiffness field model covering the entire slope;

[0033] Using the aforementioned reference anchor cable as a spatial reference point, the spatial physical coordinates (x, y) of each reference point are extracted. i , y i And the corresponding in-situ apparent stiffness value K local ; Construct a dataset consisting of multiple known spatial observation points; Input the dataset into the Kriging spatial interpolation algorithm or the support vector regression algorithm to perform multi-point integrated spatial fitting and regression calculation, calculate the spatial variability function and solve the stiffness weight of the unmonitored coordinate points, thereby fitting and reconstructing a continuous non-uniform stiffness field model K(x, y) covering the entire slope.

[0034] In the embodiments of this application, in order to eliminate the influence of stiffness degradation caused by rheology or damage of the soil and rock mass, the stiffness field is recalculated using the current benchmark data each time monitoring is performed, so as to realize the dynamic update of the continuous non-uniform stiffness field model.

[0035] Step S5: Perform load inversion on the anchor cable to be tested that is not equipped with a force gauge;

[0036] For any anchor cable to be tested that is not equipped with a force gauge, obtain its coordinates and the micro-deformation D obtained by remote sensing. target ;

[0037] Extract the stiffness value K of the anchor cable to be measured from the stiffness field model;

[0038] Inverse calculation of the locking load of the anchor cable: F target = K * Dtarget .

[0039] Step S6: Generate a group anchor load cloud map and identify stress relaxation areas or stress concentration areas;

[0040] (1) Assign the corresponding spatial physical coordinates to all the anchor cable load values ​​calculated in step S5, and generate a two-dimensional or three-dimensional group anchor load cloud map by color contour lines or heat map rendering.

[0041] (2) Set the upper safety threshold and lower allowable threshold of the anchor cable design load in the system. When the load contour line of a local area of ​​the cloud map is lower than the lower allowable threshold, it is automatically identified and delineated as a stress relaxation area with sliding risk; when the load contour line of a local area is higher than the upper safety threshold, it is identified and delineated as a stress concentration area with anchor cable breakage risk.

[0042] In the embodiments of this application, cross-project feature migration is performed for the project to be measured where no force gauge is installed at all;

[0043] It should be noted that this process can serve as a parallel / alternative branch for aging slopes without reference anchors. For this type of project, specifically including:

[0044] (1) Database construction: Based on historical monitoring engineering data, a "geological-anchoring stiffness characteristic database" is constructed. The database uses the geological strength index of rock mass (GSI), uniaxial compressive strength of rock (UCS), weathering degree and support parameters as feature vectors, and the measured apparent stiffness K as label data.

[0045] (2) Retrieval and similarity judgment: Extract the geological survey data and design parameters of the project to be tested to form a target feature vector, and use the K-nearest neighbor (KNN) algorithm or Euclidean distance calculation algorithm to match the reference working condition with the smallest feature distance and the most similarity in the database;

[0046] (3) Generate the initial stiffness model: Extract the corresponding geological features from the local geological logging zones of the project under test based on the apparent stiffness parameters of the matched reference working conditions; assign the apparent stiffness parameters under different geological features as initial values ​​to the anchor cable spatial nodes in the corresponding geological zones of the project under test; for the anchor cable nodes at the junction of different geological logging zones, use spatial distance inverse weighting or interpolation algorithms to perform smooth transition processing of stiffness values, thereby generating the prior initial continuous stiffness field model of the project under test; then use it for load inversion of the anchor cables under test in the project under test. During the inversion process, for any anchor cable under test without a load cell, obtain its coordinates and the micro-deformation D obtained by remote sensing. target Then proceed with the calculation as described in step S6.

[0047] In the embodiments of this application, the method of the present invention is not limited to anchor cables, but is also applicable to the stress inversion of similar deep support structures such as anchor bolts, anti-slide piles, and soil nails;

[0048] In the embodiments of this application, micro-deformation includes, but is not limited to, axial deformation, settlement deformation of anchor pier / pile top, or tilting deformation, as long as a stiffness mapping relationship between them and the force can be established.

[0049] The solution of this application will be further described below with reference to specific embodiments:

[0050] Example 1: Self-closed-loop inversion of a single project;

[0051] The high slope of a hydropower station is mainly composed of granite, with localized fault fracture zones. A total of 200 anchor cables of 2000kN capacity are installed, with a design locking load of 1500kN.

[0052] Reference point identification: Select 10 anchor cables as reference points.

[0053] Reference anchor cable A (located in the fractured zone): The load gauge shows a load of 1200 kN, and radar measurement indicates a cumulative compressive deformation of 8.0 mm at its anchor head. Calculate the apparent stiffness K. A = 1200 / 8.0 = 150kN / mm (The rock mass is relatively soft and has low stiffness).

[0054] Reference anchor cable B (located in intact rock mass): The load gauge shows a load of 1400kN and a cumulative compressive deformation of 3.5mm.

[0055] Calculate apparent stiffness K B = 1400 / 3.5 = 400 kN / mm (The rock mass is hard and has high rigidity).

[0056] Reconstruction: Spatial interpolation is performed using benchmark data to generate a stiffness field.

[0057] Inversion: Radar measured the micro-deformation of an unknown anchor cable C (located in a geological transition zone) to be 5.0 mm. According to the stiffness field model, the interpolated stiffness at point C is approximately 280 kN / mm.

[0058] Result: Back-calculated current load F of anchor cable C C = 280 * 5.0 = 1400 kN.

[0059] Example 2: Regionalized services across projects;

[0060] There are five high slopes with similar geological conditions (all of them are strongly weathered sandstone) along a highway in a mountainous area.

[0061] Data source: Dense anchor cable force gauges were installed only on slope 1, and radar monitoring was conducted. Analysis showed that under these geological conditions, the apparent stiffness K of the anchor cable was linearly related to the rock mass integrity coefficient Kv: K = 300 * Kv + 20.

[0062] Service targets: Slopes No. 2 to No. 5 have no force gauges installed and are existing old slopes.

[0063] Application: A simple geological sketch of slope No. 2 was performed to determine its rock mass integrity coefficient Kv ≈ 0.4.

[0064] Stiffness estimation: Using the formula for slope No. 1, estimate the average stiffness of slope No. 2:

[0065] K≈ 300* 0.4 + 20 = 140 kN / mm.

[0066] Rapid physical examination: The radar is set up to scan the No. 2 slope, the deformation is measured, and the anchor cable load distribution report of the No. 2 slope is quickly generated based on the estimated stiffness.

[0067] The foregoing description illustrates and describes a preferred embodiment of the present invention. However, as previously stated, it should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the inventive concept described herein through the foregoing teachings or techniques or knowledge in related fields. Any modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A remote sensing inversion method for slope group anchor loads based on non-uniform stiffness field reconstruction, characterized in that, Includes the following steps: Step S1: Select a reference anchor cable on the anchored slope to test the real-time load value and obtain remote sensing data on the anchored slope; Step S2: Based on remote sensing data, target identification and error correction are performed, and the axial micro-deformation of each reference anchor cable and the anchor cable to be tested is extracted during the monitoring period; Step S2 includes: S201. Processing remote sensing data: Identify the center of the anchor head or anchor block area of ​​all anchor cables using a point cloud segmentation algorithm; select natural exposed rock masses located outside the slope excavation influence area, without obvious geological tectonic activity, and with high radar reflection coherence as "stable bedrock areas"; use the phase fluctuation of this stable bedrock area as background noise, and perform differential correction using an atmospheric phase screen compensation algorithm or a point cloud registration algorithm to deduct atmospheric delay disturbances and systematic errors caused by unstable instrument setup, and obtain the absolute deformation field; S202. Extracting Axial Micro-deformation: Obtain the spatial inclination and azimuth of each anchor cable based on the slope design drawings or 3D model; extract the one-dimensional deformation D along the radar line of sight corresponding to the center of the anchor head or anchor block area of ​​each anchor cable from the absolute deformation field obtained in step S201. LOS The deformation is transformed into the true axial micro-deformation D along each reference anchor cable and the anchor cable under test by projection onto the cosine matrix of the spatial geometric direction. axial The projection transformation formula is as follows: ； In the formula, D axial D represents the actual axial micro-deformation of the anchor cable. LOS For the one-dimensional deformation along the radar line of sight; (l r , m r ,n r ) represents the direction cosine of the radar line-of-sight vector in a three-dimensional Cartesian coordinate system; (l a , m a , n a The direction cosine of the anchor cable axial vector in the same coordinate system is derived from the spatial dip angle and azimuth angle. Step S3: For each reference anchor cable, calculate the in-situ apparent stiffness at its location using the synchronously acquired real load value and remotely sensed deformation. Step S3 includes: For each reference anchor cable, the real-time load value F collected synchronously is used. ref The in-situ apparent stiffness K at the location of the axial micro-deformation extracted in step S202 is calculated using the following formula. local : K local = F ref / D ref In the formula, K local For in-situ apparent stiffness; F ref D represents the real-time load value of the reference anchor cable. ref The axial micro-deformation of the reference anchor cable; The in-situ apparent stiffness is used as a comprehensive index to reflect the elastic modulus of the free section of the anchor cable, the compression modulus of the rock and soil under the anchor, and the stiffness of the lattice beam structure. Step S4: Deduce the stiffness distribution in the unmonitored area and construct a continuous non-uniform stiffness field model covering the entire slope; Step S5: Perform load inversion on the anchor cable to be tested that is not equipped with a force gauge; Step S6: Generate a group anchor load cloud map and identify stress relaxation zones or stress concentration zones.

2. The method for remote sensing inversion of slope group anchor load based on non-uniform stiffness field reconstruction according to claim 1, characterized in that, Step S1 includes: Several anchor cables equipped with anchor cable dynamometers and functioning properly were selected on the anchored slope as reference anchor cables, and their synchronous real-time load values ​​F were obtained. ref ; Deploy ground-based interferometric radar or ground-based 3D laser scanner, set the scanning cycle and spatial resolution, and perform a full-coverage scan of the entire slope anchorage area to obtain full-field initial point cloud data containing phase information or 3D coordinates as remote sensing data.

3. The method for remote sensing inversion of slope group anchor load based on non-uniform stiffness field reconstruction according to claim 1, characterized in that, Step S4 includes: Using the aforementioned reference anchor cable as a spatial reference point, the spatial physical coordinates (x, y) of each reference point are extracted. i , y i And the corresponding in-situ apparent stiffness value K local ; Construct a dataset consisting of multiple known spatial observation points; Input the dataset into the Kriging spatial interpolation algorithm or the support vector regression algorithm to perform multi-point integrated spatial fitting and regression calculation, calculate the spatial variability function and solve the stiffness weight of the unmonitored coordinate points, thereby fitting and reconstructing a continuous non-uniform stiffness field model K(x, y) covering the entire slope.

4. The method for remote sensing inversion of slope group anchor load based on non-uniform stiffness field reconstruction according to claim 1, characterized in that, Step S5 includes: For any anchor cable to be tested that is not equipped with a force gauge, obtain its coordinates and the micro-deformation D obtained by remote sensing. target ; Extract the stiffness value K of the anchor cable to be measured from the stiffness field model; Inverse calculation of the locking load of the anchor cable: F target = K * D target .

5. The method for remote sensing inversion of slope group anchor load based on non-uniform stiffness field reconstruction according to claim 1, characterized in that, Step S6 specifically includes: Assign the corresponding spatial physical coordinates to all the anchor cable load values ​​calculated in step S5, and use color contour lines or heat maps to generate two-dimensional or three-dimensional group anchor load cloud maps. The system sets an upper safety threshold and a lower allowable threshold for the anchor cable design load. When the load contour line of a local area in the cloud map is lower than the lower allowable threshold, it is automatically identified and delineated as a stress relaxation zone with a risk of slippage. When the load contour line of a local area is higher than the upper safety threshold, it is identified and delineated as a stress concentration zone with a risk of anchor cable breakage.

6. The method for remote sensing inversion of slope group anchor load based on non-uniform stiffness field reconstruction according to claim 1, characterized in that, The method further includes: performing cross-project feature transfer for the project to be measured without force gauges installed, specifically including: Based on historical monitoring project data, a "geological-anchoring stiffness characteristic database" was constructed. This database uses rock mass geological strength index, rock uniaxial compressive strength, weathering degree and support parameters as feature vectors, and measured apparent stiffness as label data. Geological survey data and design parameters of the project to be tested are extracted to form a target feature vector. The K-nearest neighbor algorithm or Euclidean distance calculation algorithm is used to match the reference working condition with the smallest feature distance and the most similarity in the database. The apparent stiffness parameters of the matched reference working conditions are used to extract the corresponding geological features based on the local geological logging zones of the project under test. The apparent stiffness parameters under different geological features are used as initial values ​​and assigned to the anchor cable spatial nodes in the corresponding geological zones of the project under test. For the anchor cable nodes at the boundaries of different geological logging zones, the stiffness values ​​are smoothly transitioned using spatial distance inverse weighting or interpolation algorithms, thereby generating the prior initial continuous stiffness field model of the project under test.

Images