Intelligent slope modeling analysis method and system driven by monitoring data

Abstract

The invention discloses a side slope intelligent modeling analysis method and system driven by monitoring data, and relates to the field of intelligent monitoring analysis prevention and control of major side slopes of water conservancy and hydropower engineering, and the method comprises the steps: extracting side slope key control points through collected side slope form information, and generating grid node data according to an AI model generated by side slope grid nodes; on the basis of the grid node data and the two-dimensional node data, a Delaunay triangulation algorithm is used for carrying out grid unit division on the node data to generate a two-dimensional grid model, and a three-dimensional grid model is generated through expansion; and carrying out slope mechanical stability, deformation and safety analysis based on grid data output by the two-dimensional network model and the three-dimensional network model. According to the method, multi-source monitoring data and an AI generation technology are fused, full-process automatic and lightweight side slope grid division is realized, the timeliness of side slope safety monitoring analysis is remarkably improved, and an efficient calculation means is provided for side slope safety analysis and treatment of water conservancy and hydropower engineering.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring, analysis and prevention and control of major slopes in water conservancy and hydropower projects. Specifically, it relates to a method and system for intelligent modeling and analysis of slopes driven by monitoring data. Background Art

[0002] Slope safety monitoring and analysis is a key link in the prevention of geological disasters in water conservancy and hydropower projects. With the development and progress of technology, slope safety monitoring and analysis has entered the digital and intelligent era. The rapid development of multi-source monitoring technologies such as laser point clouds, UAV images, and satellite remote sensing has made it possible to obtain data in real time, and at the same time, it has provided a large amount of large-scale and high-precision data for slope monitoring and safety analysis. This has posed new challenges to traditional slope analysis methods and working modes, and issues such as high efficiency and high adaptability have become the new research focus of slope analysis methods.

[0003] Currently, slope analysis methods mainly rely on numerical simulation tools such as the finite element method, finite difference method, and discrete element method. These methods usually need to be carried out with the help of digital models. Digital modeling of slopes usually requires multiple links such as geometric modeling, shape correction, and mesh generation on the basis of pre-monitoring data, and requires a lot of manual intervention and fine-tuning. Especially for major slopes with complex terrain conditions and rock mass structures, repeated adjustments and modeling lead to low efficiency, and it is easy to introduce various errors, resulting in low efficiency and low precision of model analysis, greatly prolonging the overall evaluation cycle, and hindering the efficient response of slope safety monitoring and analysis. And for slope safety prevention and control, it is necessary to provide safety analysis results in time before disasters occur. Therefore, the above-mentioned cumbersome traditional methods are difficult to meet this requirement.

[0004] Therefore, how to automatically and scientifically fuse a large amount of slope monitoring data with on-site real-time monitoring data efficiently and convert it into a high-quality mesh model suitable for engineering analysis is a key technical issue in the field of slope safety monitoring and analysis. Summary of the Invention

[0005] In order to solve the above problems, the purpose of the present invention is to provide a slope intelligent modeling and analysis technology driven by monitoring data, aiming to effectively improve the overall level of slope safety monitoring and early warning.

[0006] To achieve the above technical purpose, the present application provides a method for intelligent modeling and analysis of slopes driven by monitoring data, including the following steps:

[0007] Extract key control points of the slope through the collected slope shape information, and generate mesh node data according to the AI model generated by the slope mesh nodes;

[0008] Based on grid node data, using two-dimensional node data as the basis, the Delaunay triangulation algorithm is used to divide the node data into grid cells to generate a two-dimensional grid model, and then extended to generate a three-dimensional grid model;

[0009] Based on the grid data output by the two-dimensional and three-dimensional network models, slope mechanical stability, deformation and safety analysis are carried out.

[0010] Preferably, in the process of obtaining the AI model, a generative adversarial network structure is adopted, and in the adversarial training process, it learns to generate complete grid node data under the given control point conditions to construct the AI model. Among them, in the training process, constraints on the minimum distance between nodes and the angle of the connecting line between adjacent nodes are added to the loss function.

[0011] Preferably, at the key control points of the slope, the collected slope shape data and height information are subjected to noise filtering, registration and normalization processing to obtain slope shape information. Through edge detection, curve fitting and point cloud segmentation algorithms, combined with geological exploration data, the key control points of the slope external contour and internal geological stratification line are extracted.

[0012] Preferably, when extracting the key control points of the slope, the RANSAC algorithm is used to verify the robustness of the control point data.

[0013] Preferably, when performing two-dimensional grid cell division, each grid cell is composed of three nodes, and numbers are assigned to each cell and the nodes it contains to form corresponding relationship data between node numbers and coordinate data, and between cell numbers and node numbers.

[0014] Preferably, when extending and generating a three-dimensional grid model, new node data is generated by equidistant sampling along a preset sweeping path, or rotational sampling is performed around a given axis to generate three-dimensional nodes; then the three-dimensional node data is divided into cells according to the sweeping and rotational paths to form a complete three-dimensional grid model.

[0015] Preferably, when performing slope mechanical stability, deformation and safety analysis, based on the grid data output by the three-dimensional network model, slope mechanical stability, deformation and safety analysis are carried out through finite element analysis or finite difference analysis.

[0016] The present invention discloses a slope intelligent modeling and analysis system driven by monitoring data, including:

[0017] A slope grid node generation AI model training module, which uses a generative adversarial network to train a slope grid node generation AI model that meets the requirements;

[0018] A data acquisition module, which is used to acquire multi-source monitoring data such as laser point clouds, UAV images, satellite data, etc. to obtain the real-time shape of the slope;

[0019] The control point extraction module is used to extract the important geological profile control points of the slope based on the real-time shape and height information of the slope, and form the node coordinate data on the external contour and stratification line of the slope; the node automatic generation module uses the pre-trained slope grid node generation AI model to generate the complete grid node coordinate data according to the control point data, and numbers each node;

[0020] The triangulation module uses the Delaunay triangulation algorithm to divide the generated node data into units, and generates the corresponding relationship between the unit numbers and the included node numbers;

[0021] The data output module is used to output the data of node numbers and coordinate data and the data of unit numbers and the included node numbers;

[0022] The three-dimensional expansion module is used to expand the two-dimensional grid data into three-dimensional grid data by sweeping and rotating, and perform corresponding node and unit division;

[0023] The calculation and analysis module inputs the output data into the calculation software through a script for analysis and calculation.

[0024] The present invention discloses the following technical effects:

[0025] 1. By adopting the generative adversarial network and the fully automated modeling method, the present invention realizes the rapid conversion from multi-source real-time monitoring data to a high-quality slope grid model, can simulate the dynamic response of the slope under different working conditions, thereby improving the real-time performance of slope safety assessment and early warning, and providing an efficient calculation means for slope stability analysis.

[0026] 2. The present invention provides a slope automated modeling method based on monitoring data driving. This method uses the key control points on the external contour and stratification line of the slope as conditional inputs, and automatically generates complete grid nodes through the slope grid node generation AI model trained by the conditional generative adversarial network, which can not only fully reflect the actual shape of the slope, but also effectively avoid the subjectivity and error accumulation in traditional manual modeling.

[0027] 3. The modeling method provided by the present invention realizes the full-process automation of two-dimensional grid automatic division and three-dimensional grid expansion, which not only greatly reduces the manual intervention and the time required for the cumbersome operations of traditional commercial software, but also significantly reduces the consumption of computing resources and engineering costs, and provides reliable technical support for slope safety monitoring and emergency response.

[0028] 4. By bypassing the limitations of traditional cumbersome software, the present invention realizes a lightweight and modular system architecture, fully integrating multi-source monitoring information such as laser point clouds, UAV images, and satellite data, enabling automated slope modeling and subsequent engineering analysis to be completed in a short time, thereby effectively improving the overall efficiency of real-time slope state monitoring and early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0030] Figure 1 is a schematic diagram of the method flow described in the present invention;

[0031] Figure 2 is a schematic diagram of the training method of the slope grid node generation AI model described in the present invention;

[0032] Figure 3 is an example diagram of the automatic modeling result in the post-processing calculation software described in the present invention;

[0033] Figure 4 is a schematic diagram of the usage process described in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0034] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all of them. The components of the embodiments of the present application usually described and illustrated in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application to be protected, but only represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0035] Such as Figures 1 - 4As shown in the figure, the present invention provides a method for intelligent modeling and analysis of slopes driven by monitoring data. This method realizes the accurate acquisition of the real-time slope morphology by integrating multi-source monitoring data; uses the key control points on the external contour and stratification line of the slope as condition inputs, and adopts a generative adversarial network (GAN) to automatically generate complete grid node data; then combines the Delaunay triangulation algorithm to achieve two-dimensional grid element division, and can generate a three-dimensional grid model through sweeping and rotation techniques; finally, the generated node and element data are output for the slope mechanical stability and safety analysis by engineering analysis software; specifically including the following content: training of the slope grid node generation AI model, adopting a generative adversarial network structure, and learning to generate complete grid node data under the given control point conditions during the adversarial training process. To ensure the engineering applicability of the generated results, during the training process, by adding constraints such as the minimum distance between nodes and the angle of the connecting line between adjacent nodes to the loss function, it is ensured that the generated nodes have good geometric properties;

[0036] Multi-source monitoring data collection and preprocessing, using multi-source monitoring means such as laser point cloud, UAV images, and satellite data to collect real-time slope morphology information, and performing noise filtering, registration, and normalization processing on the collected data to ensure accurate and unified slope morphology data;

[0037] Extraction of key control points of the slope, based on the slope morphology data and height information in the monitoring data, using image and point cloud processing algorithms to extract important geological section control points of the slope, that is, the nodes on the external contour and stratification line of the slope, whose coordinates are represented in the form of (x, y), and performing necessary normalization processing to be used as condition inputs for subsequent steps;

[0038] Automated generation of grid nodes, processing and fusing the control point coordinates, and inputting them as condition information into the pre-trained slope grid node generation AI model. The grid node data generated by the generator are numbered at the same time to form a complete two-dimensional node set;

[0039] Two-dimensional grid element division, using the automatically generated node data, adopting the Delaunay triangulation algorithm for grid division, and automatically generating two-dimensional grid elements containing triangular elements. Each element is composed of three nodes, and numbers are assigned to each element and the nodes it contains to form the corresponding relationship data between node numbers and coordinates, and element numbers and node numbers;

[0040] Three-dimensional grid expansion, in engineering applications, if simple three-dimensional slope analysis is required, the obtained two-dimensional grid model can be expanded through sweeping and rotation techniques. The specific method is as follows: generating new node data by equally spaced sampling along the preset sweeping path, or performing rotational sampling around a given axis to generate three-dimensional nodes; then dividing the three-dimensional node data according to the sweeping and rotation paths to form a complete three-dimensional grid model;

[0041] Data output and engineering application: Output the generated 2D and 3D grid data, including node numbers and coordinate information, as well as element numbers and the numbers of the nodes they contain. The output data can be directly imported into finite element and finite difference analysis software that can be directly used for slope mechanical stability, deformation, and safety analysis to achieve real-time monitoring and early warning.

[0042] Multi-source monitoring data collection and preprocessing: The sources include but are not limited to laser point clouds, UAV images, satellite data, etc.

[0043] Training of the AI model for generating slope grid nodes: The methods of imposing constraints in the loss function include but are not limited to adding constraints on the minimum distance between nodes and the angles of the connecting lines between adjacent nodes, etc.

[0044] Training of the AI model for generating slope grid nodes: The training networks used include but are not limited to traditional GAN networks, and there are also other GAN network variants used for optimizing training.

[0045] Training of the AI model for generating slope grid nodes: The structure of the generator includes but is not limited to structures based on convolutional neural networks (CNNs) or graph neural networks (GNNs).

[0046] The 3D grid expansion methods include but are not limited to sweeping and rotation.

[0047] The output data can be directly imported into engineering analysis software including but not limited to finite element and finite difference analysis software, etc.

[0048] The present invention provides a training method for an AI model for generating slope grid nodes, including:

[0049] Construction of the GAN training dataset: Construct a dataset x for generating slope grid nodes, which contains the node coordinates after dividing the grids of real slopes and virtual slopes in commercial software.

[0050] Construction of the GAN network generator: Construct a generator G, whose function is to generate grid node coordinates based on the input random slope contour points z.

[0051] Construction of the GAN network discriminator: Compare the result G(z) generated by the generator with the dataset x to train the discriminator D.

[0052] Optimization and iteration of the GAN network generator: Backpropagate the discrimination result of the discriminator D to the generator G to achieve the iteration of the generator.

[0053] Optimization and iteration of the GAN network discriminator: Repeat the construction process of the GAN network discriminator to achieve the iteration of the discriminator D.

[0054] Training of the slope grid node generation AI model. After several adversarial iterations, the slope grid node generation AI model Gn that meets the requirements is obtained.

[0055] Embodiment: The present invention provides a method for intelligent modeling and analysis of slopes driven by monitoring data, including the following steps:

[0056] Step 1, training of the slope grid node generation AI model;

[0057] In the embodiment of the present invention, the technology of generative adversarial network (GAN) is utilized.

[0058] a) Construct a slope grid node generation dataset x, which contains the node coordinates after dividing the grids of real slopes and virtual slopes in commercial software. The commercial software used in the embodiment of the present invention is ANSYS. In it, batch grid division operations are performed on real slopes and virtual slopes, and the node coordinates after dividing the grids are output in batches.

[0059] b) Construct a generator G, whose function is to generate grid node coordinates based on the input of random slope contour points z. In the embodiment of the present invention, Python programming is used to implement the generation and input of random slope contour points and the writing of the generator. The generator adopts a structure based on convolutional neural network (CNN) or graph neural network (GNN). Its multi-layer convolution and upsampling modules can fully capture the spatial features of the input and gradually generate a fine node distribution.

[0060] c) Compare the result G(z) generated by the generator with the dataset x to train the discriminator D. In the embodiment of the present invention, Python programming is used to implement the training of the discriminator D.

[0061] d) Backpropagate the discrimination result of the discriminator D to the generator G to realize the iteration of the generator. In the embodiment of the present invention, Python programming is used to implement the backpropagation of the discrimination result of the discriminator D to the generator G to realize the iteration of the generator.

[0062] e) In the embodiment of the present invention, Python programming is used to repeat the process of c to realize the iteration of the discriminator D.

[0063] f) In the embodiment of the present invention, Python programming is used to repeat the process of d to realize the adversarial iteration of the generator G and the discriminator D. During the training process, in addition to the adversarial loss, a geometric constraint term is added to the loss function, including the minimum distance constraint between nodes and the adjacent node connection angle constraint, to ensure that the generated nodes meet the quality requirements of the grid units in the project.

[0064] g) After several adversarial iterations, the slope grid node generation AI model Gn that meets the requirements is obtained. Figure 2 The training flow chart is shown.

[0065] Step 2, Multi-source monitoring data acquisition and preprocessing;

[0066] In actual engineering, the morphological information of the slope is collected in real time by a variety of sensors. In this embodiment, the following monitoring means are adopted. Laser point cloud acquisition: Use a terrestrial laser scanner to obtain high-density point cloud data of the slope surface and obtain a fine three-dimensional contour of the slope; UAV image acquisition: Through UAV aerial photography, obtain orthophotos and high-resolution video data of the slope to provide auxiliary reference for the point cloud data; Satellite remote sensing data: Use high-resolution satellite images to obtain information on the surrounding environment of the slope and large-scale terrain changes.

[0067] The data collected above passes through a preprocessing module. The specific processing contents include:

[0068] Noise filtering: Adopt statistical filtering or clustering filtering methods to remove noise points generated by equipment errors or environmental interference;

[0069] Data registration: Unify the coordinate systems of data from different devices and use relevant algorithms to achieve multi-source data registration;

[0070] Data normalization: Perform normalization processing on the slope elevation and horizontal distance to ensure the scale consistency between different data sources;

[0071] Step 3, Extraction of key control points of the slope;

[0072] The preprocessed data contains rich geometric information of the slope. Through image processing algorithms such as edge detection, curve fitting, and point cloud segmentation, combined with geological exploration data, key control points of the external contour and internal geological stratification lines of the slope are extracted. The specific steps are as follows: a) Contour detection: Detect the slope contour and extract the contour boundary; b) Stratification line extraction: Based on the geological exploration results and the elevation change of the slope, use the piecewise fitting method to extract the stratification lines reflecting different formation interfaces; c) Control point screening: Sample the detected contours and stratification lines, extract representative control points, and store their (x, y) coordinates. To eliminate local anomalies, the RANSAC algorithm is used to verify the robustness of the control point data.

[0073] Step 4, Automatic generation of grid nodes;

[0074] In this embodiment, the trained slope grid node generation AI model Gn is used to automatically generate complete grid node data under the condition of control points. The specific content is as follows:

[0075] The coordinates of the slope external contour and the layered line control points obtained in step 3 are processed and fused with a random noise vector of a certain dimension, and then input into the slope grid node generation AI model. This model can automatically complete and generate the coordinates of all internal nodes of the slope under the condition of given slope key control points, and number each generated node.

[0076] Step 5, two-dimensional grid element division:

[0077] Based on the two-dimensional node data output by the slope grid node generation AI model, use the Delaunay triangulation algorithm to divide the node data into grid elements. The specific implementation steps are as follows: a) Node data input: Input the two-dimensional node data generated in step 3 into the Delaunay triangulation module; b) Element division: Use the Delaunay algorithm to automatically construct triangular elements to ensure that each generated triangular element does not contain too small interior angles, thus ensuring the geometric rationality of the grid elements; c) Number management: Assign unique numbers to each divided triangular element and the nodes that make it up, forming a corresponding relationship dataset between node numbers and coordinate data, and element numbers and the included node numbers.

[0078] Step 6, three-dimensional grid expansion;

[0079] In some engineering applications, a two-dimensional grid model needs to be expanded to generate a three-dimensional grid model for more complex slope mechanics analysis. The specific methods include: a) Sweeping method: Extend the two-dimensional cross-section along a predetermined direction (such as the positive direction or the vertical direction of the slope), and sample at equal intervals in the extension direction to generate new node data; b) Rotation method: For slopes with rotational symmetry characteristics, rotate the two-dimensional cross-section around a specific axis to sample and generate new three-dimensional nodes; c) Three-dimensional element division: Trace the expanded three-dimensional node data along the sweeping and rotation paths, and then implement element division to generate a three-dimensional grid model that meets engineering requirements.

[0080] Step 7, data output and engineering application;

[0081] Output the element node data generated in step 5 or step 6. Specifically, it includes: a) Data format: The output file contains node numbers and coordinate information, as well as the corresponding relationship between each element number and the node numbers that make it up; b) Engineering interface: The output data can be directly imported into finite element, finite difference analysis software or other slope analysis software for subsequent engineering calculations such as analysis and stability assessment. Figure 3 This is the generated case; c) Real-time update: In the slope monitoring system, by regularly collecting and processing monitoring data, the dynamic update of the grid model can be realized, which can reflect the slope changes in real time and provide accurate data support for the early warning system.

[0082] SeeFigure 4 , which is the module operation diagram of this method: the slope grid node generation AI model training module uses a generative adversarial network to train an AI model for generating slope grid nodes that meet the requirements; the data acquisition module is used to collect multi-source monitoring data such as laser point clouds, UAV images, and satellite data to obtain the real-time shape of the slope; the control point extraction module is used to extract important geological profile control points of the slope based on the real-time shape and height information of the slope to form node coordinate data on the external contour and stratification line of the slope; the node automatic generation module uses the pre-trained AI model for generating slope grid nodes to generate complete grid node coordinate data according to the control point data and numbers each node; the triangulation module uses the Delaunay triangulation algorithm to divide the generated node data into units and generate the corresponding relationship between the unit numbers and the included node numbers; the data output module is used to output the data of node numbers and coordinate data and the data of unit numbers and the included node numbers; the 3D expansion module is used to expand the 2D grid data into 3D grid data by sweeping and rotating and perform corresponding node and unit division; the calculation and analysis module inputs the output data into calculation software through a script for analysis and calculation.

[0083] In the embodiment of the present invention, the multi-source monitoring data collection and preprocessing are sourced from, including but not limited to, laser point clouds, UAV images, satellite data, etc.

[0084] In the training of the AI model for generating slope grid nodes provided in the embodiment of the present invention, the method of imposing constraints in the loss function includes but is not limited to adding constraints such as the minimum distance between nodes and the angle of the connecting line between adjacent nodes.

[0085] In the training of the AI model for generating slope grid nodes provided in the embodiment of the present invention, the training network used includes but is not limited to traditional GAN networks, and there are also other GAN network variants used for optimizing the training.

[0086] In the training of the AI model for generating slope grid nodes provided in the embodiment of the present invention, the structure of the generator includes but is not limited to a structure based on a convolutional neural network (CNN) or a graph neural network (GNN).

[0087] In the embodiment of the present invention, the 3D grid expansion methods include but are not limited to sweeping and rotating.

[0088] In the embodiment of the present invention, the output data can be directly imported into engineering analysis software including but not limited to finite element and finite difference analysis software, etc.

[0089] In summary, the present invention precisely geometrically models based on multi-source real-time monitoring data of a slope, extracts key control points such as the external contour and stratification lines of the slope; through an artificial intelligence model trained by a generative adversarial network, it realizes the automatic generation of grid node data for the entire slope area; uses the Delaunay triangulation algorithm to divide the generated node data into grid cells, and at the same time expands the two-dimensional grid into a three-dimensional grid model through sweeping and rotation techniques as needed; then further imports the generated two-dimensional and three-dimensional grid data into engineering analysis software to realize the mechanical responses (such as deformation, stress distribution, and stability, etc.) of the slope under different working conditions, as well as real-time calculation and evaluation. Through the above process, the dynamic changes of the slope under natural and construction conditions can be effectively simulated, and the key engineering parameters required for the safety evaluation of the slope can be obtained in real time. This method provides an efficient and intelligent solution for the automated modeling and analysis of major slopes in water conservancy and hydropower projects, greatly improving the real-time performance and accuracy of slope safety monitoring and early warning, and at the same time significantly reducing the time and economic costs of manual modeling.

[0090] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 or multiple blocks.

[0091] In the description of the present invention, it should be understood that the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "a plurality" means two or more, unless otherwise specifically defined.

[0092] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A slope intelligent modeling and analysis method driven by monitoring data, characterized in that It includes the following steps: Extract the key control points of the slope from the collected slope morphology information, and generate grid node data according to the AI model generated by the slope grid nodes; Based on the grid node data, using the two-dimensional node data as the basis, use the Delaunay triangulation algorithm to divide the node data into grid cells to generate a two-dimensional grid model, and expand it to generate a three-dimensional grid model; Based on the grid data output by the two-dimensional and three-dimensional network models, conduct slope mechanical stability, deformation and safety analysis.

2. The method for intelligent modeling and analysis of a slope driven by monitoring data according to claim 1, wherein: In the process of obtaining the AI model, a generative adversarial network structure is adopted, and in the adversarial training process, learn to generate complete grid node data under the given control points, and construct the AI model. Among them, in the training process, add constraints on the minimum distance between nodes and the angle of the connecting line between adjacent nodes to the loss function.

3. The method for intelligent modeling and analysis of a slope driven by monitoring data according to claim 2, wherein: When obtaining the key control points of the slope, perform noise filtering, registration and normalization processing on the collected slope morphology data and height information to obtain the slope morphology information, and extract the key control points of the slope external contour and internal geological stratification line through edge detection, curve fitting and point cloud segmentation algorithms, combined with geological exploration data.

4. The method for intelligent modeling and analysis of a slope driven by monitoring data according to claim 3, wherein: When extracting the key control points of the slope, use the RANSAC algorithm to verify the robustness of the control point data.

5. The method for intelligent modeling and analysis of a slope driven by monitoring data according to claim 4, wherein: When performing grid cell division, each grid cell is composed of three nodes, and number each cell and the nodes it contains to form data on the correspondence between node numbers and coordinates, and cell numbers and node numbers.

6. The method for intelligent modeling and analysis of a slope driven by monitoring data according to claim 5, wherein: When expanding and generating a three-dimensional grid model, generate new node data by equidistant sampling along a preset sweeping path, or perform rotational sampling around a given axis to generate three-dimensional nodes; then divide the three-dimensional node data according to the sweeping and rotational paths to form a complete three-dimensional grid model.

7. The method for intelligent modeling and analysis of a slope driven by monitoring data according to claim 6, wherein: When conducting slope mechanical stability, deformation and safety analysis, based on the grid data output by the two-dimensional and three-dimensional network models, conduct slope mechanical stability, deformation and safety analysis through finite element analysis or finite difference analysis.

8. An intelligent modeling analysis system for slopes driven by monitoring data, characterized in that, This system is used to implement a method for intelligent modeling and analysis of a slope driven by monitoring data as described in any one of claims 1-7. This system includes: An AI model training module for generating slope grid nodes, which uses a generative adversarial network to train an AI model for generating slope grid nodes that meets the requirements; The data acquisition module is used to collect multi-source monitoring data such as laser point clouds, UAV images, satellite data, etc. to obtain the real-time morphology of the slope; The control point extraction module is used to extract important geological section control points of the slope based on the real-time morphology and height information of the slope, and form the node coordinate data on the external contour and stratification line of the slope; The node automatic generation module uses a pre-trained slope grid node generation AI model to generate complete grid node coordinate data according to the control point data and number each node; The triangulation module uses the Delaunay triangulation algorithm to divide the generated node data into units, and generates the corresponding relationship between the unit number and the included node number; The data output module is used to output the data of the node number and coordinate data and the data of the unit number and the included node number; The three-dimensional expansion module is used to expand the two-dimensional grid data into three-dimensional grid data through sweeping and rotation, and perform corresponding node and unit division; The calculation and analysis module inputs the output data into the calculation software through a script for analysis and calculation.

Images