Side slope potential sliding surface extraction method based on energy dissipation rate

Abstract

The invention discloses a side slope potential sliding surface extraction method based on an energy dissipation rate, belongs to the technical field of side slope potential sliding surface extraction, and can solve the problem that an existing method neglects an energy dissipation mechanism in a side slope instability process, so that the identification precision of a potential sliding surface is low. The method comprises the following steps: S1, determining a centroid coordinate of each grid unit on a two-dimensional numerical model of a slope longitudinal section, and predicting an energy dissipation rate of each grid unit; s2, determining a plurality of maximum value feature points of the energy dissipation rate on the vertical section of the slope according to the centroid coordinates and the energy dissipation rate of each grid unit; and S3, determining the potential sliding surface of the slope according to the plurality of maximum value feature points. The method is used for extracting the potential sliding surface of the slope.

Description

Technical Field

[0001] The invention relates to a method for extracting a potential sliding surface of a slope based on energy dissipation rate, and belongs to the technical field of extracting a potential sliding surface of a slope. Background Art

[0002] In the fields of water conservancy, transportation, mining and other engineering, slope stability analysis can provide scientific decision-making basis for slope design, slope state judgment, slope reinforcement and management, and is of great significance to ensure engineering safety and personnel safety. The extraction of potential sliding surface of slope is a key link in slope stability analysis.

[0003] At present, the extraction of potential sliding surfaces of slopes is usually based on limit equilibrium method, numerical simulation method and unsupervised learning algorithm to identify potential sliding surfaces. These methods mostly rely on indicators such as displacement, plastic strain and shear strain of slope material partitions, while ignoring the energy conversion and dissipation mechanism in the process of slope instability. Therefore, it is difficult to accurately describe the generation of potential sliding surfaces from the perspective of energy conversion, resulting in low recognition accuracy of potential sliding surfaces. At the same time, existing methods are usually only applicable to the identification of a single potential sliding surface, and it is difficult to cope with the identification scenario of multiple potential sliding surfaces coexisting, so there are great limitations. Summary of the invention

[0004] The present invention provides a method for extracting potential sliding surfaces of slopes based on energy dissipation rate, which can solve the problem that the existing methods ignore the energy dissipation mechanism in the process of slope instability, resulting in low recognition accuracy of potential sliding surfaces.

[0005] The present invention provides a method for extracting a potential sliding surface of a slope based on energy dissipation rate, the method comprising: S1. Determine the centroid coordinates of each grid cell on the two-dimensional numerical model of the slope longitudinal section and predict the energy dissipation rate of each grid cell; S2. Determine multiple maximum characteristic points of the energy dissipation rate on the longitudinal section of the slope according to the centroid coordinates and energy dissipation rate of each grid cell; S3. Determine the potential sliding surface of the slope based on multiple maximum feature points.

[0006] Optionally, the energy dissipation rate of each grid unit is predicted in S1, specifically including: Obtain the elastic-plastic parameters and creep parameters of different material partitions of the slope; Based on the elastic-viscoplastic creep damage model, the energy dissipation rate of each grid unit is predicted according to the elastic-plastic parameter and the creep parameter.

[0007] Optionally, the S2 specifically includes: S21, constructing a regular grid model according to the two-dimensional numerical model, and determining the energy dissipation rate of each regular grid on the regular grid model according to the centroid coordinates and energy dissipation rate of each grid unit on the two-dimensional numerical model; S22. Determine a plurality of maximum characteristic points of the energy dissipation rate on the longitudinal section of the slope according to the energy dissipation rate of each regular grid.

[0008] Optionally, determining the energy dissipation rate of each regular grid on the regular grid model in S21 specifically includes: An interpolation algorithm is used to determine the energy dissipation rate of each regular grid on the regular grid model.

[0009] Optionally, the S22 specifically includes: S221, according to the energy dissipation rate of each regular grid, determining a plurality of maximum points of the energy dissipation rate along a horizontal path and / or a vertical path on the regular grid model; S222. Determine multiple maximum characteristic points of the energy dissipation rate on the longitudinal section of the slope based on the multiple maximum points.

[0010] Optionally, the S222 specifically includes: According to the multiple maximum value points and the boundary information of the slope longitudinal section, a convex hull detection algorithm is used to determine the multiple maximum value characteristic points of the energy dissipation rate on the slope longitudinal section.

[0011] Optionally, the S3 specifically includes: S31, determining clusters using a clustering algorithm based on multiple maximum feature points; S32. Determine the potential sliding surface of the slope using a random sample consensus algorithm based on the clusters.

[0012] Optionally, the S31 specifically includes: S311, removing the maximum value feature points whose energy dissipation rate is less than a preset threshold from the multiple maximum value feature points to obtain a feature point set; S312: Determine clusters according to the feature point set using a density-based clustering algorithm.

[0013] Optionally, the S312 specifically includes: Based on the silhouette score indicator, the hyperparameter optimization algorithm is used to optimize multiple hyperparameters of the density-based clustering algorithm to obtain the best hyperparameter combination; A cluster is determined according to the optimal hyperparameter combination and the feature point set and using the density-based clustering algorithm.

[0014] Optionally, the S32 specifically includes: S321, extracting a current inner point set and a current outer point set from the cluster using a random sample consensus algorithm, and fitting the current inner point set to obtain a current potential sliding surface; S322, extracting the next inner point set and the next outer point set from the current outer point set by using the random sample consensus algorithm, and fitting the next inner point set to obtain the next potential sliding surface; S323: Take the next external point set as the new current external point set, and repeat S322 until the number of external points in the current external point set is less than a preset number.

[0015] The beneficial effects that the present invention can produce include: The present invention can comprehensively reflect the irreversible energy conversion process inside the slope by determining the energy dissipation rate of each grid unit in the two-dimensional numerical model of the slope longitudinal section, thereby accurately locating the local damage area inside the slope and extracting the potential sliding surface from the local damage area, thereby effectively improving the identification accuracy of the potential sliding surface of the slope.

[0016] The present invention uses a clustering algorithm to preliminarily identify and separate the local damage area inside the slope, and then uses a random sample consensus algorithm to identify and extract multiple potential sliding surfaces one by one from the local damage area, so that it can cope with the identification scenario where multiple potential sliding surfaces coexist, and has a wider range of applications. At the same time, the combination of the clustering algorithm and the random sample consensus algorithm significantly improves the recognition accuracy and efficiency of the potential sliding surface of the slope. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flow chart of a method for extracting a potential sliding surface of a slope based on energy dissipation rate provided in an embodiment of the present invention; Figure 2 A schematic diagram of a two-dimensional numerical model of a soft interlayer slope provided by an embodiment of the present invention; Figure 3 A schematic diagram of energy dissipation rate distribution of a two-dimensional numerical model provided in an embodiment of the present invention; Figure 4 A schematic diagram of energy dissipation rate distribution of a regular grid model provided in an embodiment of the present invention; Figure 5 A schematic diagram of a maximum point set C provided by an embodiment of the present invention; Figure 6 A schematic diagram of the distribution of multiple maximum feature points after filtering out external noise points using a convex hull detection algorithm provided in an embodiment of the present invention; Figure 7 A schematic diagram of the result of idealized clustering of a feature point set using the DBSCAN algorithm provided in an embodiment of the present invention; Figure 8A schematic diagram of a fitting curve of a potential sliding surface provided by an embodiment of the present invention; Fig. 9 A schematic diagram of cluster 3 provided in an embodiment of the present invention; Fig.10 A schematic diagram of a process of extracting a main sliding surface from cluster 3 provided in an embodiment of the present invention; Fig.11 A schematic diagram of the process of extracting a secondary sliding surface from cluster 3 provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The present invention is described in detail below in conjunction with embodiments, but the present invention is not limited to these embodiments.

[0019] The embodiment of the present invention provides a method for extracting potential sliding surface of a slope based on energy dissipation rate, such as Figure 1 As shown, the method includes: S1. Determine the centroid coordinates of each grid cell on the two-dimensional numerical model of the slope longitudinal section and predict the energy dissipation rate of each grid cell.

[0020] Specifically, a two-dimensional numerical model of the longitudinal section of the slope can be constructed based on the geological information and geometric information of the slope, and the two-dimensional numerical model has a plurality of grid units.

[0021] For example, taking a slope with a weak interlayer as an example, the material partition of the slope includes soil and a weak interlayer. According to the geological information and geometric information of the slope, a two-dimensional numerical model of the longitudinal section of the slope can be constructed on the XZ plane in the XYZ space rectangular coordinate system, such as Figure 2 Then, normal constraints are applied to the left and right boundaries of the two-dimensional numerical model, and the bottom boundary is completely fixed, and the grid density of the two-dimensional numerical model can be controlled by 1m. Based on the XYZ space rectangular coordinate system, this embodiment can determine the centroid coordinates of each grid unit on the two-dimensional numerical model.

[0022] The energy dissipation rate of each grid unit is predicted in S1, which may specifically include: Obtain the elastic-plastic parameters and creep parameters of different material partitions of the slope; Based on the elastic-viscoplastic creep damage model, the energy dissipation rate of each grid element is predicted according to the elastic-plastic parameters and creep parameters.

[0023] Specifically, the elastic-viscoplastic creep damage model is an existing model, which can simulate the failure behavior and instability process inside the slope according to the elastic-plastic parameters and creep parameters of the slope, and predict the energy dissipation rate of each grid unit on the two-dimensional numerical model of the slope at a preset time in the future through iterative calculation.

[0024] For example, the elastic-plastic parameters of the above-mentioned weak interlayer slope are shown in Table 1.

[0025] Table 1 Elastic-plastic parameters of soft interlayer slope

[0026] For example, the creep parameters of the above-mentioned weak interlayer slope are shown in Table 2.

[0027] Table 2 Creep parameters of soft interlayer slope

[0028] Specifically, the calculation step of the elastic-viscoplastic creep damage model is set to 1 hour and the number of iterations is set to 8000 times. The energy dissipation rate of each grid unit on the two-dimensional numerical model of the soft interlayer slope after 8000 hours can be predicted. The energy dissipation rate distribution diagram of the two-dimensional numerical model of the soft interlayer slope is shown in Figure 2. Figure 3 shown.

[0029] S2. According to the centroid coordinates and energy dissipation rate of each grid cell, multiple maximum characteristic points of the energy dissipation rate on the longitudinal section of the slope are determined.

[0030] S2 can specifically include: S21. Construct a regular grid model according to the two-dimensional numerical model, and determine the energy dissipation rate of each regular grid on the regular grid model according to the centroid coordinates and energy dissipation rate of each grid unit on the two-dimensional numerical model.

[0031] In S21, the energy dissipation rate of each regular grid on the regular grid model is determined, which can be specifically: The energy dissipation rate of each regular grid on the regular grid model is determined using an interpolation algorithm.

[0032] Specifically, this embodiment constructs a regular grid model according to the boundary of the two-dimensional numerical model. For example, the regular grid model of the above-mentioned weak interlayer slope is as follows: Figure 4 The rectangle shown in the figure has a boundary of the regular grid model that coincides with the maximum boundary of the two-dimensional numerical model in the X direction and the Z direction. The regular grid model is composed of a plurality of regular grids, each of which is a square with a side length of 0.1 m. Then, this embodiment determines the energy dissipation rate of the regular grid at the corresponding position in the regular grid model according to the centroid coordinates and energy dissipation rate of each grid unit on the two-dimensional numerical model, and uses an interpolation algorithm to determine the energy dissipation rate of all the remaining regular grids in the regular grid model. The schematic diagram of the energy dissipation rate distribution of the regular grid model of the weak interlayer slope is shown in FIG. Figure 4 shown.

[0033] By constructing a regular grid model and interpolating the energy dissipation rate from the two-dimensional numerical model to the regular grid model, the interference caused by the grid inhomogeneity of the two-dimensional numerical model on the subsequent potential sliding surface extraction can be reduced.

[0034] S22. According to the energy dissipation rate of each regular grid, a plurality of maximum characteristic points of the energy dissipation rate on the longitudinal section of the slope are determined.

[0035] S22 may specifically include: S221. According to the energy dissipation rate of each regular grid, multiple maximum points of the energy dissipation rate are determined along the horizontal path and / or the vertical path on the regular grid model.

[0036] Specifically, this embodiment uses the energy dissipation rate as a discrimination index. On a regular grid model, multiple maximum points of the energy dissipation rate can be extracted according to a horizontal search path to form a point set A. Multiple maximum points of the energy dissipation rate can also be extracted according to a vertical search path to form a point set B. In practice, since the extraction accuracy of the maximum points is higher when the search path intersects the potential sliding surface at a larger angle, it is advisable to use a vertical search path to extract the maximum points for the gently inclined sliding surface near the foot of the slope, and it is advisable to use a horizontal search path to extract the maximum points for the steeply inclined sliding surface at the rear edge of the landslide. In order to combine the advantages of both the horizontal search path and the vertical search path, this embodiment takes the union of the point set A extracted by the horizontal search path and the point set B extracted by the vertical search path to obtain the maximum point set C, that is, C=A∪B, which can effectively reduce the possibility of missing maximum points. The distribution of each point in the maximum point set C is shown as follows. Figure 5 shown.

[0037] S222. Determine multiple maximum characteristic points of the energy dissipation rate on the longitudinal section of the slope based on the multiple maximum points.

[0038] S222 may specifically include: According to the boundary information of multiple maximum points and the slope longitudinal section, the convex hull detection algorithm is used to determine the multiple maximum feature points of the energy dissipation rate on the slope longitudinal section.

[0039] In this embodiment, since the area of ​​the regular grid model is larger than that of the two-dimensional numerical model, the above interpolation algorithm will generate multiple external noise points outside the geometric boundary of the slope longitudinal section, and the process of searching and extracting the maximum value points will also extract the maximum value points in these external noise points. Therefore, the maximum value point set C will also contain multiple external noise points, such as Figure 5To eliminate the interference of these external noise points, this implementation can use the convex hull detection algorithm to filter out all the external noise points outside the geometric boundary of the slope longitudinal section in the maximum point set C based on the boundary information of the slope longitudinal section, and then use all the remaining maximum points as maximum feature points. The distribution diagram of multiple maximum feature points after filtering out external noise points using the convex hull detection algorithm is shown in Figure 6 shown.

[0040] S3. Determine the potential sliding surface of the slope based on multiple maximum feature points.

[0041] S3 can specifically include: S31. Determine clusters using a clustering algorithm based on multiple maximum feature points.

[0042] S31 may specifically include: S311 , removing the maximum feature points whose energy dissipation rate is less than a preset threshold from the plurality of maximum feature points to obtain a feature point set.

[0043] Specifically, there may be internal noise points within the geometric boundary of the slope longitudinal section. Some of these internal noise points are numerical errors caused by the strong boundary constraints of the two-dimensional numerical model, and the other part is interference data generated by the dense low energy dissipation area inside the slope. In order to eliminate the interference of these internal noise points, this embodiment arranges multiple maximum feature points from large to small according to the energy dissipation rate, and then calculates the upper quartile of them. Set as the preset threshold, and set the energy dissipation rate less than the upper quartile The maximum feature points are removed, and then all the remaining maximum feature points are taken as the feature point set.

[0044] S312: Determine clusters according to the feature point set using a density-based clustering algorithm.

[0045] Specifically, the density-based clustering algorithm may be Density-Based Spatial Clustering of Applications with Noise (DBSCAN algorithm for short). The DBSCAN algorithm can identify one or more high-density areas in the feature point set, and the high-density area is the local damage area inside the slope. The DBSCAN algorithm marks each high-density area as a cluster, and some of the clusters contain potential sliding surfaces. At the same time, the DBSCAN algorithm can identify sparse areas in the feature point set and mark the sparse areas as noise. The advantage of the DBSCAN algorithm is that the identification of clusters can be completed without pre-specifying the number of clusters, and non-convex clusters such as arc clusters can be effectively identified. Therefore, the preliminary extraction of multiple potential sliding surfaces can be achieved under the condition that the number of potential sliding surfaces is unknown.

[0046] Specifically, the hyperparameters of the DBSCAN algorithm include the neighborhood radius Eps and the minimum number of points MinPts. By adjusting the values ​​of the above hyperparameters, the performance of the DBSCAN algorithm can be optimized. Usually, the values ​​of the hyperparameters are adjusted manually. For example, this embodiment finds through manual adjustment that for the above-mentioned weak interlayer slope, when the neighborhood radius Eps=0.5 and the minimum number of points MinPts=3, the DBSCAN algorithm can achieve the most ideal clustering effect. Figure 7 As shown in the figure, after the DBSCAN algorithm idealized the feature point set of the soft interlayer slope based on the above hyperparameter values, a total of 11 high-density areas, i.e., 11 clusters, were identified, of which two clusters each contained a potential sliding surface, cluster 1 contained the main sliding surface, and cluster 2 contained a secondary sliding surface, showing the most ideal performance of the DBSCAN algorithm.

[0047] It is understandable that the above-mentioned optimal clustering effect of the DBSCAN algorithm is based on repeated manual adjustment of hyperparameters, so it is highly dependent on manual work. In order to improve the adjustment efficiency of hyperparameters and the automation of the DBSCAN algorithm, this embodiment can also use a hyperparameter optimization algorithm to automatically optimize the hyperparameter values ​​of the DBSCAN algorithm.

[0048] S312 may specifically include: Based on the silhouette score indicator, the hyperparameter optimization algorithm is used to optimize multiple hyperparameters of the density-based clustering algorithm to obtain the best hyperparameter combination; The clusters are determined based on the optimal hyperparameter combination and feature point set using a density-based clustering algorithm.

[0049] Specifically, the hyperparameter optimization algorithm may be a grid search method, which is an exhaustive search method that searches for the best hyperparameter combination by traversing all possible combinations of multiple hyperparameters.

[0050] Specifically, this embodiment uses the silhouette score as an evaluation index to optimize multiple hyperparameters. First, the grid search method is used to traverse all possible combinations of the neighborhood radius Eps and the minimum number of points MinPts; then, the DBSCAN algorithm is trained using each combination, and the Silhouette score of the DBSCAN algorithm trained using each combination is calculated; finally, the combination corresponding to the best Silhouette score is determined as the best hyperparameter combination. The optimization results of this embodiment show that when the neighborhood radius Eps=1.2 and the minimum number of points MinPts=19, the DBSCAN algorithm obtains the best Silhouette score of 0.36. Therefore, the neighborhood radius Eps=1.2 and the minimum number of points MinPts=19 are the best hyperparameter combination of this embodiment.

[0051] S32. Based on the clusters, the potential sliding surface of the slope is determined using a random sample consensus algorithm.

[0052] Specifically, this embodiment uses the Random Sample Consensus algorithm (RANSAC algorithm for short) to further separate discrete sliding surface feature points, namely, interior points, from the clusters, and performs smooth fitting on the discrete interior points, ultimately achieving automatic extraction of multiple potential sliding surfaces.

[0053] Specifically, this embodiment executes the RANSAC algorithm for each cluster in turn, setting the sampling times , residual threshold , the minimum number of internal points MinInlier = 300. For clusters with less than the minimum number of internal points MinInlier, the RANSAC algorithm directly skips. For clusters with more than or equal to the minimum number of internal points MinInlier, the RANSAC algorithm separates the internal and external points in the clusters through iterative calculation, and then fits multiple discrete internal points to obtain the geometric model of the potential sliding surface. .

[0054] For example, this embodiment uses the RANSAC algorithm to separate the inliers and outliers of cluster 1 and cluster 2 obtained by manually adjusting the hyperparameters, and determines the optimal number of inliers of a potential sliding surface. , the optimal number of interior points of another potential sliding surface , and then the inner points of the two potential sliding surfaces are smoothly fitted to obtain Figure 8 The fitting curve shown in the figure also obtains the parameter vectors of the geometric model of the two potential sliding surfaces. , thus obtaining the geometric model of two potential sliding surfaces and .

[0055] In practice, when the DBSCAN algorithm does not achieve ideal performance due to reasons such as hyperparameter selection, multiple potential sliding surfaces may be classified into the same cluster in clustering. For example, Fig. 9 As shown in the figure, when the DBSCAN algorithm performs clustering based on the best hyperparameter combination (neighborhood radius Eps=1.2 and minimum number of points MinPts=19) obtained by the above automatic optimization, the primary sliding surface and the secondary sliding surface are both classified into the same cluster 3. This is because there are local shear damage noise points along the extension line of the slope near the toe of the slope, and there are X-shaped local shear failure noise points on the trailing edge of the secondary sliding surface. The existence of these noise points makes the point sets of the two potential sliding surfaces nearly connected, causing the DBSCAN algorithm to mistakenly classify the two potential sliding surfaces into one category. To achieve the separation of the two potential sliding surfaces, one method is to manually adjust the neighborhood radius Eps and the minimum number of points MinPts based on prior knowledge, and the other method is to manually remove the above local shear damage noise points and local shear failure noise points to eliminate the influence of noise points on the clustering results. However, both of the above methods cannot achieve automatic extraction of potential sliding surfaces, resulting in low extraction efficiency.

[0056] This embodiment uses the RANSAC algorithm in a nested manner, which can effectively solve the problem of multiple potential sliding surfaces being divided into the same cluster.

[0057] Specifically, S32 may include: S321, extracting the current inner point set and the current outer point set from the cluster using a random sample consensus algorithm, and fitting the current inner point set to obtain the current potential sliding surface; S322, extracting the next inner point set and the next outer point set from the current outer point set using a random sample consensus algorithm, and fitting the next inner point set to obtain the next potential sliding surface; S323: Take the next external point set as the new current external point set, and repeat S322 until the number of external points in the current external point set is less than a preset number.

[0058] Specifically, the preset number is the minimum number of inliers MinInlier.

[0059] For example, for Fig. 9 Cluster 3 contains two potential sliding surfaces. First, the RANSAC algorithm is used to perform multiple rounds of iterations to extract internal points in cluster 3 until the number of internal points of the main sliding surface is maximized, thereby obtaining the internal point set of the main sliding surface, as shown in Fig.10 shown. Fig.10Iter-1 to Iter-13 in the figure represent the best fit of the inliers obtained after each iteration in the process of extracting the inliers of the main sliding surface through 13 iterations. The colors of Iter-1 to Iter-13 from light to dark represent the change of the best fit of the inliers during the iteration process. It can be seen that the best fit of the inliers is initially distributed near the shallow sliding surface (Iter-1 to Iter-5), and then the iteration converges to the deep sliding surface with more inliers (Iter-6 to Iter-13).

[0060] The internal point set of the main sliding surface is removed from cluster 3 and fitted to obtain the geometric model of the main sliding surface. Then, the RANSAC algorithm is used to repeatedly extract the next internal point set from the remaining external point set of cluster 3 to obtain the internal point set of the secondary sliding surface, as shown in Fig.11 shown. Fig.11 Iter-1 to Iter-9 in the figure represent the best fit of the inliers obtained after each iteration in the process of extracting the inliers of the secondary sliding surface for 9 iterations. It can be seen that due to the reduction in the number of feature points in the external point set, the extraction of the secondary sliding surface tends to converge from Iter-2. Subsequently, the internal point set of the secondary sliding surface is removed and fitted to obtain the geometric model of the secondary sliding surface. Finally, after judgment, since the number of remaining external points is less than the minimum number of internal points MinInlier, the RANSAC algorithm stops extracting inliers.

[0061] This embodiment uses the RANSAC algorithm in a nested manner, so that potential sliding surfaces that intersect or overlap in space can be separated and extracted one by one, thereby improving the extraction accuracy of the potential sliding surfaces.

[0062] Specifically, the implementation process of the nested RANSAC algorithm is as follows: Step 1. Initialize clustering The label of the potential sliding surface in and the optimal number of interior points . Define the number of sampling times , residual threshold , the minimum number of internal points MinInlier. Suppose the potential sliding surface The geometric model is ,in Representing geometric models Parameter vector, for example, this embodiment uses a quadratic curve ,but ,in, and Potential sliding surface The geometric model exist The horizontal and vertical coordinates in the plane rectangular coordinate system, , and They are control parameters.

[0063] Step 2: In clustering Conducted Repeat sampling: 1) Randomly select 3 points to estimate the potential sliding surface The parameter vector Since the potential sliding surface of the slope in practice is convex, if , then skip this sampling directly.

[0064] 2) For Points , ,in, and Points exist Calculate the horizontal and vertical coordinates in the rectangular coordinate system When the parameter vector is And the horizontal axis is The vertical coordinate of , then calculate With point The vertical coordinate The residual between .

[0065] 3) If , then it is called point is an internal point. Calculate the residuals of all points and count the number of internal points , which satisfies The total number of all points is the number of internal points.

[0066] 4) If and ,but ,and The optimal parameter vector .

[0067] Step 3. Data output: 1) Get the potential sliding surface The geometric model The parameter vector and its interior point set .

[0068] 2) Potential sliding surface The interior point set From the cluster Remove, that is .

[0069] 3) If the clustering If the number of remaining feature points in is still greater than MinInlier, repeat Step 2 (let , ); otherwise, stop extracting the potential sliding surface.

[0070] The present invention takes into account that the DBSCAN algorithm is sensitive to the two hyperparameters of neighborhood radius Eps and minimum number of points MinPts. Therefore, only the DBSCAN algorithm is used to perform preliminary clustering of the feature point set to reduce the probability of failure in extracting the potential sliding surface and ensure the robustness of the extraction program. Then, the RANSAC algorithm is used to further extract and fit the inner points from the clusters obtained by the DBSCAN algorithm to achieve accurate extraction of the potential sliding surface. At the same time, through the nested use of the RANSAC algorithm, even if the DBSCAN algorithm fails to achieve accurate clustering when multiple potential sliding surfaces coexist, the RANSAC algorithm can still accurately lock multiple potential sliding surfaces under the background of dense noise points, demonstrating the advantages of the present invention in the identification of complex failure patterns of slopes.

[0071] The above are only a few embodiments of the present application and do not constitute any form of limitation to the present application. Although the present application is disclosed as above with preferred embodiments, it is not intended to limit the present application. Any technician familiar with the profession, without departing from the scope of the technical solution of the present application, using the technical content disclosed above to make slight changes or modifications are equivalent to equivalent implementation cases and fall within the scope of the technical solution.

Claims

1. A method for extracting potential sliding surface of slope based on energy dissipation rate, characterized in that: The method comprises: S1. Determine the centroid coordinates of each grid cell on the two-dimensional numerical model of the slope longitudinal section and predict the energy dissipation rate of each grid cell; S2. Determine multiple maximum characteristic points of the energy dissipation rate on the longitudinal section of the slope according to the centroid coordinates and energy dissipation rate of each grid cell; S3. Determine the potential sliding surface of the slope based on multiple maximum feature points.

2. The method according to claim 1, characterized in that The energy dissipation rate of each grid unit is predicted in S1, specifically including: Obtain the elastic-plastic parameters and creep parameters of different material partitions of the slope; Based on the elastic-viscoplastic creep damage model, the energy dissipation rate of each grid unit is predicted according to the elastic-plastic parameter and the creep parameter.

3. The method according to claim 1, characterized in that The S2 specifically includes: S21, constructing a regular grid model according to the two-dimensional numerical model, and determining the energy dissipation rate of each regular grid on the regular grid model according to the centroid coordinates and energy dissipation rate of each grid unit on the two-dimensional numerical model; S22. Determine a plurality of maximum characteristic points of the energy dissipation rate on the longitudinal section of the slope according to the energy dissipation rate of each regular grid.

4. The method according to claim 3, characterized in that Determining the energy dissipation rate of each regular grid on the regular grid model in S21 specifically includes: An interpolation algorithm is used to determine the energy dissipation rate of each regular grid on the regular grid model.

5. The method according to claim 3, characterized in that: The S22 specifically includes: S221, according to the energy dissipation rate of each regular grid, determining a plurality of maximum points of the energy dissipation rate along a horizontal path and / or a vertical path on the regular grid model; S222. Determine multiple maximum characteristic points of the energy dissipation rate on the longitudinal section of the slope based on the multiple maximum points.

6. The method according to claim 5, characterized in that The S222 specifically includes: According to the multiple maximum value points and the boundary information of the slope longitudinal section, a convex hull detection algorithm is used to determine the multiple maximum value characteristic points of the energy dissipation rate on the slope longitudinal section.

7. The method according to claim 1, characterized in that The S3 specifically includes: S31, determining clusters using a clustering algorithm based on multiple maximum feature points; S32. Determine the potential sliding surface of the slope using a random sample consensus algorithm based on the clusters.

8. The method according to claim 7, characterized in that The S31 specifically includes: S311, removing the maximum value feature points whose energy dissipation rate is less than a preset threshold from the multiple maximum value feature points to obtain a feature point set; S312: Determine clusters according to the feature point set using a density-based clustering algorithm.

9. The method according to claim 8, characterized in that The S312 specifically includes: Based on the silhouette score indicator, the hyperparameter optimization algorithm is used to optimize multiple hyperparameters of the density-based clustering algorithm to obtain the best hyperparameter combination; A cluster is determined according to the optimal hyperparameter combination and the feature point set and using the density-based clustering algorithm.

10. The method according to claim 7, characterized in that The S32 specifically includes: S321, extracting a current inner point set and a current outer point set from the cluster using a random sample consensus algorithm, and fitting the current inner point set to obtain a current potential sliding surface; S322, extracting the next inner point set and the next outer point set from the current outer point set by using the random sample consensus algorithm, and fitting the next inner point set to obtain the next potential sliding surface; S323: Take the next external point set as the new current external point set, and repeat S322 until the number of external points in the current external point set is less than a preset number.

Images