Roadway geological visualization modeling method and system under joint method

Abstract

The application provides a roadway geological visualization modeling method and system under a combined method, comprehensively adopts a Kriging interpolation algorithm and an inverse distance weighting method, constructs a stratum surface model by using a TIN-based surface modeling method; obtains an intersection line of a fault and a stratum when each stratum is cut by the same fault, so as to determine a control point set of a fault surface and construct a three-dimensional geological model of the fault surface; uses a profile line of a local coordinate system established on each control point along a center line as a section, uses a section expansion modeling method to perform geometric modeling of a driving roadway; and performs dynamic prediction and model updating based on static data of a modeling area and dynamic data collected in a driving process, so as to realize model visualization. The application can take into account the form and internal attributes of a geological body, and at the same time, with the help of deep learning, can learn sample data and mine implicit information and feature representation of the data, so as to more accurately learn and progressively predict geological information.

Description

Technical Field

[0001] This invention belongs to the field of geological visualization modeling technology, specifically relating to a tunnel geological visualization modeling method and system based on a combined approach. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Three-dimensional geological modeling technology, powered by computers, is used for geological analysis in a three-dimensional environment. This technology effectively integrates tools for spatial information management, analysis and prediction, geological interpretation and statistics, entity content analysis, and graphical visualization. Over the past few decades, researchers from various application fields have proposed numerous geological modeling methods based on practical needs.

[0004] Three-dimensional dynamic modeling of geological space has been a research hotspot and also a challenging area in recent years. Based on data sources, it can be divided into four categories: modeling methods based on profiles, scattered data, borehole data, and multi-source data. Based on technical levels, it can be divided into five stages: visualization, measurement, analysis, updating, and temporal modeling. The first three are static stages, and the latter two are dynamic stages. Three-dimensional dynamic geological models based on multi-source data can be analyzed by fusing dynamic and static data, using spatial data models as a foundation.

[0005] Spatial data models are a general term for data constraints, data structures, and data operations. They are constructed to facilitate data management, use, storage, and exchange, and form the basis for the spatial relationships and morphology of geological bodies.

[0006] Based on geometric characteristics, spatial data models can be divided into three main categories: surface-based data models, volume-based data models, and hybrid data models. Different spatial data models are suitable for different spatial forms and have their own focuses.

[0007] Surface-based data modeling methods primarily represent geological conditions in three dimensions by using the surface of a model to enclose a spatial region. The geometric features of geological bodies are generally represented and described by tiny surface elements. Due to limitations in its modeling principle, it is difficult to describe the internal properties of three-dimensional bodies, thus hindering the viewing, editing, and analysis of three-dimensional geological information. However, this modeling method has significant advantages in representing the spatial boundaries, visualization, and geometric transformations of geological bodies. Furthermore, because its modeling theory is relatively mature and its procedural methods are easy to implement, it remains the mainstream approach for constructing three-dimensional stratigraphic models.

[0008] Volume-based data modeling methods abstract three-dimensional geological bodies into a collection of adjacent but non-intersecting three-dimensional unit entities, using these tiny volume elements to represent the internal space of the geological body. This modeling method can characterize the internal properties of geological bodies relatively well, but it is difficult to describe the topological relationships between objects, and the computation process consumes a lot of computer resources. Summary of the Invention

[0009] To address the aforementioned problems, this invention proposes a combined approach for roadway geological visualization modeling and a system. This invention employs multiple data models to comprehensively represent different spatial geological bodies, taking into account both the morphology and internal properties of the geological bodies. Furthermore, by leveraging deep learning, it can mine implicit information and feature representations of the data while learning from sample data, enabling more accurate learning and progressive prediction of geological information.

[0010] According to some embodiments, the present invention adopts the following technical solution:

[0011] A combined approach for roadway geological visualization modeling includes the following steps:

[0012] A stratigraphic surface model was constructed by combining the Kriging interpolation algorithm and the inverse distance weighting method, using a TIN-based surface modeling method.

[0013] Obtain the intersection line between the fault and the strata when each stratum is cut by the same fault, in order to determine the control point set of the fault plane and construct a three-dimensional geological model of the fault plane;

[0014] The outline of the local coordinate system established at each control point along the centerline is used as the cross section, and the cross section expansion modeling method is used to perform geometric modeling of the tunnel.

[0015] Dynamic prediction and model updates are performed based on static data of the modeling area and dynamic data collected during the tunneling process, enabling model visualization.

[0016] As an alternative implementation method, the process of combining the Kriging interpolation algorithm and the inverse distance weighting method includes first using the inverse distance weighting method to perform preliminary interpolation to obtain a rough model, and then applying the Kriging interpolation method to adjust and optimize the model.

[0017] As an alternative implementation method, the process of constructing a stratigraphic surface model using the TIN-based surface modeling method includes starting from an initial triangle or a side, gradually expanding outwards, and adding new points to the triangular mesh in sequence until all points are contained in the TIN; after each insertion of a new point, the new triangle maintains topological consistency with the existing triangular mesh, the region boundary is used as the initial wavefront, and the mesh is generated inwards according to preset rules. When all wavefronts converge and disappear in the region, the mesh generation is complete.

[0018] As an alternative implementation method, the process of constructing a three-dimensional geological model of the fault plane includes:

[0019] (1) Establishing the initial TIN model: Select initial data points, including fault strike, extension range, dip angle, etc., and establish the fault plane TIN model based on these data points.

[0020] (2) Intersect the TIN model in step (1) with the geological fault plane to obtain the fault intersection line;

[0021] (3) Based on the fault displacement information, the difference in elevation between the corresponding point of the original data and the intersection line of the fault is calculated to determine the scope of influence;

[0022] (4) Based on the properties of normal and reverse faults, the strata are divided into two regions, namely the hanging wall and the footwall of the fault. The fault intersection line in step (3) is inserted in the corresponding region to divide the TIN strata grid in the region.

[0023] (5) Using the range obtained in step (3), construct the stratigraphic TIN model after the displacement of the hanging wall and footwall of the fault. Repeat this step to obtain multiple stratigraphic TIN models, all of which contain the corresponding fault, and finally form a fault structure model.

[0024] As an alternative implementation method, in the process of constructing a three-dimensional geological model of the fault plane, the ground plane is set as the top and bottom plates of the coal seam.

[0025] Fault planes are drawn based on fault strike and dip information, and the coal seam intersection line is obtained by intersecting with the strata.

[0026] Process the coal breakage line and obtain the coal breakage line after the operation based on the breakage distance information;

[0027] The coal seam model is generated by dividing the layer into grids and generating a coal seam model based on fault properties. The model includes fault displacement.

[0028] Repeat the above process to obtain the coal seam roof and floor plane model.

[0029] As an alternative implementation method, the geometric modeling of the tunnel is based on the theories of revolution and stretching. The cross-section adopts the contour line of the local coordinate system. The contour line is established at each control point along the center line of the tunnel. The method of tunneling geometry modeling is used to expand the model of the cross-section.

[0030] As an alternative implementation method, during the process of model visualization, a long short-term memory network deep learning model is used for static data to perform progressive predictions based on historical data, thereby achieving dynamic update operations.

[0031] A tunnel geological visualization modeling system based on a joint approach includes:

[0032] The stratigraphic surface model construction module is configured to comprehensively employ the Kriging interpolation algorithm and the inverse distance weighting method to construct the stratigraphic surface model using a TIN-based surface modeling method.

[0033] The fault plane model construction module is configured to obtain the intersection line between the fault and the strata when each stratum is cut by the same fault, so as to determine the control point set of the fault plane and construct a three-dimensional geological model of the fault plane.

[0034] The tunnel geometry modeling module is configured to use the outline of the local coordinate system established at each control point along the centerline as the cross section, and the tunnel geometry modeling is performed using the cross section expansion modeling method.

[0035] The model visualization module is configured to perform dynamic prediction and model updates based on static data of the modeling area and dynamic data collected during the tunneling process, thereby achieving model visualization.

[0036] A computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the steps in the above method.

[0037] An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, wherein the computer instructions, when executed by the processor, perform the steps in the method described above.

[0038] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0039] This invention comprehensively employs Kriging interpolation and inverse distance weighted interpolation as the mathematically defined spatial interpolation methods used in the construction of this 3D geological model. First, inverse distance weighted interpolation is used for preliminary interpolation to obtain a rough model. Then, Kriging interpolation is applied to adjust and optimize the model. The results of the two methods are compared and verified to select the most suitable interpolation scheme. For the geological surface model, a TIN-based surface modeling method is used to construct the stratigraphic surface model. A boundary advance method is used to construct the TIN and automatically generate the TIN model. An edge is selected from a given set of points as the initial edge, and the point closest to the current boundary is continuously selected to form a new triangle. This process is repeated until all points are added to the triangulation network. For the fault model, the intersection lines of the fault and strata when each stratum is cut by the same fault are obtained to determine the control point set of the fault plane. Then, the boundary advance algorithm is used to construct the 3D geological model of the fault plane. Based on the theories of solids of revolution and tensile bodies, the contour lines of the local coordinate system established along the centerline at each control point are used as cross-sections. The cross-section extension modeling method is used for the geometric modeling of the tunnel.

[0040] This invention also integrates deep learning into the construction of dynamic models. When predicting the dynamic updates of the tunnel model, it introduces Long Short-Term Memory (LSTM) and CNN deep learning models to perform progressive predictions based on historical data, reducing the limitations of relying solely on static data updates and improving the model's predictive ability. By introducing recurrent connections in the time dimension and adding new hidden layers between different time points, the entire neural network can model the sequential relationships between sequences.

[0041] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0042] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0043] Figure 1 A flowchart of a tunnel geological visualization modeling method based on the Kriging-IDW-LSTM joint method provided in one embodiment;

[0044] Figure 2 A flowchart illustrating the basic process of the boundary advancement method as provided in one embodiment;

[0045] Figure 3 A flowchart of fault modeling is provided for one embodiment;

[0046] Figure 4 A flowchart of LSTM modeling provided for one embodiment. Detailed Implementation

[0047] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0048] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0049] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0050] Where there is no conflict, the embodiments and features described in this application may be combined with each other.

[0051] Example 1

[0052] This invention provides a method for visual modeling of tunnel geology, such as... Figure 1 As shown, it includes the following steps:

[0053] Step (1): Select a reasonable spatial interpolation method. One of the important theoretical bases for establishing a geological model is spatial interpolation.

[0054] Simulating geological interfaces typically involves two key steps: first, projecting the geological interface to be simulated onto a reference plane and triangulating the projected area; second, using interpolation based on the original sampled data to obtain the three-dimensional coordinates of the grid nodes, generating a geological surface with three-dimensional spatial properties.

[0055] In the actual modeling process, detailed analytical exploration geological data is used to obtain the spatial coordinate values ​​of the sampling points and reflect the three-dimensional geometric shape of the target interface, but the spatial coordinates of other places besides the sampling points cannot be directly obtained.

[0056] Therefore, it is necessary to establish a certain type of geological surface based on the obtained sampling point data, so that the real sampling points can be passed through or as close as possible to the surface, and the spatial coordinates of other locations can also be calculated using the geological surface.

[0057] In summary, the spatial distribution pattern of the geological surface is derived from the analysis of the original sampling data, which is interpolation. This process involves calculating the spatial coordinates of points in unknown areas. The interpolation algorithms used in this invention mainly include Kriging and inverse distance weighting, primarily applied to the creation of initial geological surfaces, local re-subdivision of surfaces, and surface reconstruction. Its basic principles and characteristics are as follows:

[0058] (1) Kriging interpolation algorithm

[0059] This method focuses on defining the weighting coefficients to ensure that the interpolation function is in its best operating mode. Based on the theory of variograms and the analysis of the corresponding structural composition, it takes values ​​for the data changes in a limited data range, such as a defined stratigraphic domain, and adopts an unbiased optimal estimation method for these data.

[0060] Based on the theory of function variation and its correlation analysis, it can be concluded that the amount of data in a region has certain spatial correlations. Essentially, this is achieved by using the structural characteristics of function transformation and the basic data of the region to make unbiased optimal estimates of the acquired data. The advantage of Kriging interpolation is its ability to significantly reduce the variance of potential errors. When building geological models, Kriging interpolation can achieve a good fit between the initial sample data and the target geological interface, and the obtained interpolation conclusions are relatively close to the basic state and spatial appearance of the underlying geological body. Kriging interpolation performs well in scenarios where geological interfaces are smooth, continuous, and without significant undulations. However, it performs poorly in fitting numerical points at geological interfaces with large undulations, such as abrupt changes on both sides of large faults, near folds, and irregular rock mass interfaces.

[0061] (2) Inverse Distance Weighting

[0062] Inverse Distance Weighted (IDW) is based on the assumption that known points closer to the target point have a greater influence on the predicted value of the target point. Specifically, the influence of each known point is weighted according to the inverse of its distance from the target point; the closer the distance, the greater the weight. Therefore, in geological modeling, especially when encountering local geological features such as faults, the data points at these features are usually not uniformly distributed, and points near faults are particularly important. IDW can better capture the characteristics of such geological features, effectively compensating for the shortcomings of Kriging interpolation in areas with large undulations. The predicted values ​​are usually weighted averages. This method is simple and intuitive, but it is quite sensitive to data distribution and spatial heterogeneity.

[0063] The basic principle of inverse distance weighted (IDW) estimation is to use the values ​​of known points to estimate the values ​​of unknown points, and to weight the estimates based on the distances between the known and unknown points. Points that are closer to each other have a greater influence on the estimated value. The formula is as follows:

[0064]

[0065] in:

[0066] Z(x) is the estimated value at position x. Zi is the value at the known point i. wi is the weight of the known point i with respect to position x. n is the total number of known points used for interpolation. The weight wi is usually calculated using the following formula:

[0067]

[0068] Where: di is the distance from known point i to unknown point x. p is a non-negative power exponent, usually taking the value of 1 or 2, which determines the rate at which the weight decays with distance.

[0069] Leveraging the simplicity and intuitiveness of IDW and the statistical advantages of Kriging, more accurate predictions can be obtained from geological data, thus increasing the reliability of tunnel modeling. Kriging interpolation can help address issues that IDW may encounter when dealing with spatial heterogeneity and uneven data point distribution. Improved stability: The spatial variability model of the Kriging method can enhance overall numerical stability and prediction reliability.

[0070] In the geological modeling process, this invention first uses IDW for preliminary interpolation to obtain a rough geological model, and then applies Kriging interpolation to adjust and optimize the model.

[0071] The main application of this invention is to establish a high-precision model of the coal and rock geological body surrounding tunnel excavation, accurately representing spatially occurring information, including coal and rock distribution, structure, and stratum thickness, considering only the lithological classification of the strata. Because the TIN model method for representing complex surface models can eliminate redundant data and ensure high fitting accuracy by varying the size and number of triangular meshes, a TIN-based surface modeling method is chosen to construct the stratum surface model.

[0072] The TIN generation algorithm used in this invention is the boundary advancement method. The basic idea is to start from an initial triangle (or one edge) and gradually expand outwards, adding new points to the triangulation mesh sequentially until all points are included in the TIN. The key to this process is ensuring that after each insertion of a new point, the new triangle maintains topological consistency with the existing triangular mesh, and that the entire mesh structure is non-overlapping, continuous, and topologically sound. The region boundary is used as the initial wavefront, and advancement is performed inwards according to certain rules. Mesh generation occurs during this advancement process. The mesh generation is complete when all wavefronts converge and disappear within the region. The specific generation method is as follows: Figure 2 As shown.

[0073] In the fault modeling section, the approach to establishing faults in this invention is to obtain the intersection lines between faults and strata when each stratum is cut by the same fault, in order to determine the control point set of the fault plane. Then, a boundary advancement algorithm is used to construct a three-dimensional geological model of the fault plane, which involves the following steps:

[0074] (1) Establishing the initial TIN model: Select initial data points, including fault strike, extension range, dip angle, etc., and establish the fault plane TIN model based on these data points.

[0075] (2) Intersect the TIN model in (1) with the geological fault plane to obtain the fault line;

[0076] (3) Based on the fault displacement information, the difference in elevation between the corresponding point of the original data and the intersection line of the fault is calculated to determine the scope of influence;

[0077] (4) Based on the properties of normal and reverse faults, the strata are divided into two regions, namely the hanging wall and the footwall. The fault intersection line in (3) is inserted in the corresponding region to divide the TIN strata grid in the region.

[0078] (5) To build a stratigraphic model after the displacement of the upper and lower plates of the fault in the construction of the stratigraphic TIN model, the displacement range obtained in (3) should be used. This step should be repeated to obtain multiple stratigraphic TIN models, all of which contain the corresponding fault, and finally form a fault structure model.

[0079] The above fault modeling process is described in detail below. Figure 3 For example, assuming the stratum is the top and bottom plate of the coal seam; the fault plane is drawn based on information such as fault strike and dip angle, and the fault-coal intersection line is obtained by intersecting with the stratum; the fault-coal intersection line is processed, and the processed fault-coal intersection line is obtained based on the fault displacement information; the stratum grid is divided, and a coal seam model is generated based on the fault attributes (normal fault or reverse fault), the model including fault displacement; the figure shows the coal seam top and bottom plate stratum model obtained by repeating the above process; it is a complete fault structure model.

[0080] In the geometric modeling of tunneling roadways, this invention is based on the theories of solids of revolution and stretched solids. The cross-section adopts the contour line of the local coordinate system. The contour line is established at each control point along the center line of the roadway. The method of tunneling roadway geometric modeling is used to expand the model of the cross-section. The corresponding roadway cross-section is loaded at each control point of the center line. Then, the Z value of the control point on the center line is used to determine whether it is on the roof or the floor. Then, the coordinates of the control points and the shape of the cross-section are combined to form the roadway cross-section polygon. The generated roadway cross-section polygons are then connected to generate a triangular network, thereby generating the roadway model.

[0081] The constructed tunnel geological model visualization can be dynamically updated based on static data of the modeling area and dynamic data collected during the tunneling process, using a Long Short-Time Memory (LSTM) deep learning model for dynamic updates. In modeling tunnels, relying solely on initial static data will result in a model that cannot accurately reflect the dynamic changes occurring during tunnel excavation. As the tunnel progresses, surrounding geological conditions, soil and rock properties, and other factors may change, potentially over a long period. The LSTM model can learn from these long-term changes, identifying patterns and thus more accurately updating the previous static geological model, making the model more precise.

[0082] First, suitable data for training and updating the LSTM is selected, such as geological data (including rock strata properties, soil conditions, and hydrological conditions); construction progress (including tunneling speed, construction period, and equipment used); and environmental conditions (temperature, humidity, and pressure). Then, a preliminary model is trained. The data selected in this invention can all be classified as time-series data arranged chronologically. The tunnel status is recorded at regular intervals according to time steps. This recorded data is used for training, and an LSTM model is built to predict the tunnel status in the future, thus achieving dynamic updates to the geological model. This basic model supports predicting key indicators at the next time point using historical time-series data. Simultaneously, the LSTM is continuously and dynamically updated. As new geological data, environmental conditions, or construction progress information flows into the LSTM, it continues to be trained to gradually adapt to the new data, thereby achieving more accurate results.

[0083] LSTM can overcome the problems of traditional RNN models, such as vanishing and bursting gradients. Its principle lies in a special gated network structure. Compared to traditional RNN models, LSTM not only outputs information to the hidden layer at each time step but also outputs internal information to the next time step. The internal state of LSTM, denoted as Ct, is the part that processes this internal information. Historical information from previous time steps is recorded in the internal state at each time step. The expression for the internal state of LSTM at each time step is as follows:

[0084]

[0085] h t =O t ·tanh(C t )

[0086]

[0087] W c b c These are the weight matrix and bias of the internal memory unit at time t, respectively, f. t Forget gates utilize weights adjusted by importance to selectively forget non-critical and repetitive information, controlling the amount of information forgotten from the input layer and the previous time step's internal state. The calculation formula is as follows:

[0088] f t =σ(W f x t +U f h t-1 +b f )

[0089] The activation function σ is the Sigmoid function, with a range of 0 to 1. It acts as a gate and is defined as follows:

[0090]

[0091] i t The input gate controls the information that needs to be saved at the current moment; the calculation formula is as follows:

[0092] i t =σ(W o x t +U o h t-1 +b o )

[0093] W o b o These are the weight matrix and bias of the output gate at time t, respectively.

[0094] The most prominent feature of the LSTM network structure is its gating mechanism. The parameters to be learned are E, U, and b, and the elements in the three gating vectors all take values ​​between 0 and 1. The transmission unit state can be manipulated using these three gating mechanisms. Because LSTM can remember content that needs to be memorized for a long time, while ignoring some unimportant content, it is called a Long Short-Term Memory (LTSM) network. The advantages of LTSM are particularly prominent in environments requiring long-term memory. However, because the LSTM model has more parameters and more components, training an LSTM model is much more difficult compared to other models.

[0095] In this invention, during the visualization of the tunnel during excavation, an LSTM model is used for static data to perform progressive predictions based on historical data, reducing the limitations of solely relying on static data updates. The steps for building the LSTM model are as follows, and the model structure is as follows: Figure 4 As shown.

[0096] (1) Normalize the input geological data sample data;

[0097] (2) Divide the original data into a training set and a test set;

[0098] (3) Build an LSTM model and select appropriate model parameters;

[0099] (4) Input the training set data into the model to complete the model training;

[0100] (5) Input the test set data into the LSTM model, test the model, obtain the final result, and evaluate the result.

[0101] Example 2

[0102] A tunnel geological visualization modeling system based on a joint approach includes:

[0103] The stratigraphic surface model construction module is configured to comprehensively employ the Kriging interpolation algorithm and the inverse distance weighting method to construct the stratigraphic surface model using a TIN-based surface modeling method.

[0104] The fault plane model construction module is configured to obtain the intersection line between the fault and the strata when each stratum is cut by the same fault, so as to determine the control point set of the fault plane and construct a three-dimensional geological model of the fault plane.

[0105] The tunnel geometry modeling module is configured to use the outline of the local coordinate system established at each control point along the centerline as the cross section, and the tunnel geometry modeling is performed using the cross section expansion modeling method.

[0106] The model visualization module is configured to perform dynamic prediction and model updates based on static data of the modeling area and dynamic data collected during the tunneling process, thereby achieving model visualization.

[0107] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0108] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0109] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0110] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0111] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made by those skilled in the art without creative effort within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A tunnel geological visualization modeling method under a joint approach, characterized in that, Includes the following steps: A stratigraphic surface model was constructed by combining the Kriging interpolation algorithm and the inverse distance weighting method, using a TIN-based surface modeling method. Obtain the intersection line between the fault and the strata when each stratum is cut by the same fault, in order to determine the control point set of the fault plane and construct a three-dimensional geological model of the fault plane; The outline of the local coordinate system established at each control point along the centerline is used as the cross section, and the cross section expansion modeling method is used to perform geometric modeling of the tunnel. Dynamic prediction and model updates are performed based on static data of the modeling area and dynamic data collected during the tunneling process to achieve model visualization. The process of comprehensively using the Kriging interpolation algorithm and the inverse distance weighting method includes first using the inverse distance weighting method for preliminary interpolation to obtain a rough model, and then applying the Kriging interpolation method to adjust and optimize the model. In the process of realizing model visualization, a long short-term memory network deep learning model is used for static data to perform progressive prediction based on historical data and realize dynamic update operation; the static data includes geological data, construction progress, and environmental conditions; the historical data is time series data arranged in chronological order, and the tunnel status data is recorded at regular intervals according to time steps.

2. The tunnel geological visualization modeling method under the combined approach as described in claim 1, characterized in that, The process of constructing a stratigraphic surface model using a TIN-based surface modeling method involves starting from an initial triangle or edge and gradually expanding outwards, adding new points to the triangular mesh sequentially until all points are contained in the TIN. After each insertion of a new point, the new triangle maintains topological consistency with the existing triangular mesh. The region boundary is used as the initial wavefront, and the mesh is generated as it advances into the region according to preset rules. When all wavefronts converge and disappear within the region, the mesh generation is complete.

3. The tunnel geological visualization modeling method under the combined approach as described in claim 1, characterized in that, The process of constructing a three-dimensional geological model of a fault plane includes: (1) Establishment of the initial TIN model: Select initial data points, including fault strike, extension range, and dip angle, and establish the fault plane TIN model based on these data points; (2) Intersect the TIN model in step (1) with the geological fault plane to obtain the fault intersection line; (3) Based on the fault displacement information, the difference in elevation between the corresponding point of the original data and the intersection line of the fault is calculated to determine the scope of influence; (4) Based on the properties of normal and reverse faults, the strata are divided into two regions, namely the hanging wall and the footwall. The fault intersection line in step (3) is inserted in the corresponding region to divide the TIN strata grid in the region. (5) Using the range obtained in step (3), construct the stratigraphic TIN model after the displacement of the hanging wall and footwall of the fault. Repeat this step to obtain multiple stratigraphic TIN models, all of which contain the corresponding fault, and finally form a fault structure model.

4. The tunnel geological visualization modeling method under the combined approach as described in claim 1, characterized in that, In the process of constructing a three-dimensional geological model of the fault plane, the ground plane is assumed to be the top and bottom plates of the coal seam. Fault planes are drawn based on fault strike and dip information, and the coal seam intersection line is obtained by intersecting with the strata. Process the coal breakage line and obtain the coal breakage line after the operation based on the breakage distance information; The coal seam model is generated by dividing the layer into grids and generating a coal seam model based on fault properties. The model includes fault displacement. Repeat the above process to obtain the coal seam roof and floor plane model.

5. The tunnel geological visualization modeling method under the combined approach as described in claim 1, characterized in that, The geometric modeling of the tunnel is based on the theories of solids of revolution and extruded solids. The cross-section adopts the contour line of the local coordinate system. The contour line is established at each control point along the center line of the tunnel. The method of tunneling geometric modeling is used to expand the model of the cross-section.

6. A tunnel geological visualization modeling system based on a combined method, implementing the steps of the method according to any one of claims 1-5, characterized in that, include: The stratigraphic surface model construction module is configured to comprehensively employ the Kriging interpolation algorithm and the inverse distance weighting method to construct the stratigraphic surface model using a TIN-based surface modeling method. The fault plane model construction module is configured to obtain the intersection line between the fault and the strata when each stratum is cut by the same fault, so as to determine the control point set of the fault plane and construct a three-dimensional geological model of the fault plane. The tunnel geometry modeling module is configured to use the outline of the local coordinate system established at each control point along the centerline as the cross section, and the tunnel geometry modeling is performed using the cross section expansion modeling method. The model visualization module is configured to perform dynamic prediction and model updates based on static data of the modeling area and dynamic data collected during the tunneling process, thereby achieving model visualization.

7. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, implement the steps of the method according to any one of claims 1-5.

8. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, implement the steps of the method according to any one of claims 1-5.

Images