An intelligent tunnel 3D geological modeling method and system under complex geological conditions

Abstract

The present invention belongs to the technical field of 3D geological modeling and provides a method and system for intelligent 3D tunnel geological modeling under complex geological conditions. The technical solution is as follows: combining 3D tunnel profile data, inverted gravity data from the tunnel site, and magnetic susceptibility to perform data reconstruction to obtain a reconstructed tunnel dataset; training a geological attribute classifier based on the reconstructed tunnel dataset to obtain a trained geological attribute classifier; predicting the geological attributes of each point in the space within the modeling area of ​​the tunnel site based on the trained geological attribute classifier, and constructing a 3D geological model that includes surface undulations and structures in combination with surface elevation data; and performing Boolean operations on the 3D geological model that includes surface undulations and structures and a 3D tunnel volume model obtained by sweeping along the 3D tunnel path to obtain a 3D geological model containing the tunnel. The present invention can construct 3D models of tunnels under extremely complex conditions.

Description

Technical Field

[0001] The present invention belongs to the technical field of three-dimensional geological modeling, and in particular relates to an intelligent three-dimensional geological modeling method and system for tunnels under complex geological conditions. Background Art

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

[0003] Tunnel engineering is widely used in highway, railway, and urban rail transit applications. However, tunnel projects often face complex geological conditions, such as faults, fracture zones, and karst caves. These conditions pose significant challenges to tunnel design, construction, excavation, and operational safety. Three-dimensional geological modeling of tunnels under complex geological conditions is a key technology for ensuring project safety. Accurate 3D geological models provide reliable geological information for project design, construction, and operations, helping to identify potential risk factors and providing a basis for developing appropriate construction plans and safety measures.

[0004] When it comes to tunnel construction under complex geological conditions, traditional 3D geological modeling fails to consider the multi-source heterogeneity of geological survey data and the intricate relationship between geological structure and space. It only uses simple stratigraphic data for modeling, and the resulting 3D geological model is not suitable for tunnel modeling under more complex geological conditions. Furthermore, traditional 3D geological models have long modeling cycles, are difficult to update, and contain little information. Therefore, they cannot effectively guide actual tunnel construction. Summary of the Invention

[0005] In order to solve at least one technical problem existing in the above-mentioned background technology, the present invention provides an intelligent three-dimensional geological modeling method and system for tunnels under complex geological conditions. By introducing multi-source heterogeneous data into the stacking ensemble learning algorithm in machine learning, the obtained three-dimensional geological model is suitable for tunnel modeling under more complex geological conditions. At the same time, it can obtain updated information in real time and update the model, which can highly guide the actual construction of tunnels and other problems.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A first aspect of the present invention provides an intelligent three-dimensional geological modeling method for tunnels under complex geological conditions, comprising the following steps:

[0008] Converting the acquired two-dimensional tunnel profile data into three-dimensional tunnel profile data;

[0009] The data is reconstructed by combining the acquired 3D tunnel profile data, the inverted gravity data and magnetic susceptibility of the tunnel site to obtain a reconstructed tunnel dataset.

[0010] Training a geological attribute classifier based on the reconstructed tunnel data set to obtain a trained geological attribute classifier;

[0011] The trained geological attribute classifier is used to predict the geological attributes of each point in the tunnel site modeling area. Combined with the surface elevation data, a three-dimensional geological model including the surface undulations and structures is constructed.

[0012] A three-dimensional geological model including the surface undulations and structures is subjected to Boolean operations on the three-dimensional tunnel model obtained by sweeping along the three-dimensional path of the tunnel to obtain a three-dimensional geological model including the tunnel.

[0013] Furthermore, the data is reconstructed by combining the acquired three-dimensional tunnel profile data, the inverted gravity data and the magnetic susceptibility of the tunnel site area to obtain a reconstructed tunnel data set, including:

[0014] Based on the acquired 3D tunnel profile data, virtual boreholes are added at complex locations where profile data is sparse or contains faults to generate new virtual borehole data.

[0015] The original borehole data, virtual borehole data and three-dimensional tunnel profile data were resampled and combined with the gravity data and magnetic susceptibility inverted in the tunnel site area to obtain a data set with three-dimensional geographic coordinates, gravity data and magnetic susceptibility as feature engineering and stratigraphic categories as labels.

[0016] Furthermore, the geological attribute classifier is trained based on the reconstructed tunnel data set to obtain the trained geological attribute classifier, including:

[0017] The tunnel dataset is evenly and randomly divided into two parts: training set and test set according to the set ratio;

[0018] The training set data is divided into n parts, and n-1 parts of the data are selected each time. The base classifier is used for training. The base classifier to be used is determined based on the performance of different base classifiers with the best hyperparameters in the test set.

[0019] The n prediction results obtained by training the base classifier are used as training data for retraining, and the meta-classifier is trained to obtain a trained geological attribute classifier.

[0020] Furthermore, during the training process, the geological attribute classifier obtains the data added during the tunnel construction and excavation process and adds it to the dataset, re-divides the training set and test set, and re-predicts the attributes of the spatial coordinate points in the modeling area to update the established three-dimensional geological model.

[0021] Furthermore, during the training process, the geological attribute classifier also includes obtaining advanced geological prediction data during tunnel excavation to invert the attributes of the tunnel surrounding rock, including stratum occurrence, surrounding rock grade, rock burst grade, compressive strength, elastic modulus, wave velocity, and groundwater, and assigning the attributes of the tunnel surrounding rock to the classifier.

[0022] Furthermore, during the training process of the geological attribute classifier, the performance of the geological attribute classifier is evaluated by drawing a confusion matrix diagram and a ROC curve diagram.

[0023] Furthermore, the predicted values ​​of the geological attribute classifier are combined with the surface elevation data, including:

[0024] Import the geological attributes of each point in the space within the tunnel site modeling area into the visualization software and create a SGrid model. Different lithologies represented by different predicted values ​​are assigned different colors. Ultimately, a three-dimensional geological model of the tunnel with clear geological boundaries, including special geological structures and unfavorable geological bodies, is obtained.

[0025] The surface elevation data is used to constrain the three-dimensional geological model of the tunnel containing special geological structures and unfavorable geological bodies, and a three-dimensional geological model with surface undulations and structures is established.

[0026] A second aspect of the present invention provides an intelligent three-dimensional geological modeling system for tunnels under complex geological conditions, comprising:

[0027] A data conversion module, which is used to convert the acquired two-dimensional tunnel profile data into three-dimensional tunnel profile data;

[0028] A data reconstruction module is used to combine the acquired three-dimensional tunnel profile data, the inverted gravity data and magnetic susceptibility of the tunnel site area to perform data reconstruction and obtain a reconstructed tunnel data set;

[0029] A classifier training module, which is used to train a geological attribute classifier based on the reconstructed tunnel data set to obtain a trained geological attribute classifier;

[0030] The three-dimensional geological model construction module is used to predict the geological attributes of each point in the space within the tunnel site modeling area based on the trained geological attribute classifier, and to construct a three-dimensional geological model containing surface undulations and structures in combination with surface elevation data; the three-dimensional geological model containing surface undulations and structures is subjected to Boolean operations on the three-dimensional geological model containing the surface undulations and structures and the three-dimensional tunnel model obtained by sweeping along the three-dimensional path of the tunnel to obtain a three-dimensional geological model containing the tunnel.

[0031] A third aspect of the present invention provides a computer-readable storage medium.

[0032] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned intelligent three-dimensional geological modeling method for tunnels under complex geological conditions.

[0033] A fourth aspect of the present invention provides a computer device.

[0034] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the intelligent three-dimensional geological modeling method for tunnels under complex geological conditions as described above are implemented.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1. The present invention introduces a stacking integrated machine learning method based on three-dimensional coordinate points, gravity, magnetic susceptibility, lithology and other characteristics. It can predict the geological conditions in the modeling area by analyzing existing sampling data and inversion data. When geological exploration data is limited and sparse, it makes full use of existing data and encrypts the geological data to achieve high-precision three-dimensional geological modeling of strip-shaped bodies such as tunnels. It is suitable for the construction of three-dimensional geological models of strip-shaped tunnels under various complex terrain conditions.

[0037] 2. The present invention can obtain the data added during the tunnel construction and excavation process and add it to the data set. At the same time, it can obtain the advanced geological prediction data during tunnel excavation to invert the properties of the tunnel surrounding rock and train the classifier. Compared with other modeling methods, this method has the advantages of simple modeling process, comprehensive consideration of multiple factors, low cost, higher modeling accuracy under the same available data, faster modeling speed, and the ability to quickly update the established model based on the modification or addition of training data. The model contains richer geological information, which can effectively improve the intelligent level of geological modeling.

[0038] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0040] Figure 1 This is the overall technical roadmap of the intelligent three-dimensional geological modeling method for tunnels under complex geological conditions provided by the embodiment of the present invention;

[0041] Figure 2 is a schematic diagram of comprehensive data processing provided by an embodiment of the present invention;

[0042] Figure 3 Schematic diagram of the stacking algorithm training process provided by an embodiment of the present invention;

[0043] Figure 4 It is a three-dimensional geological model constructed by the intelligent tunnel three-dimensional geological modeling method under complex geological conditions provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0045] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0046] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0047] In view of the characteristics of traditional 3D geological modeling mentioned in the background technology, such as long cycle, much manual interaction, and strong professionalism, the present invention

[0048] Example 1

[0049] like Figure 1 As shown, this embodiment provides an intelligent three-dimensional geological modeling method for tunnels under complex geological conditions, including the following steps:

[0050] Step 1: Obtain two-dimensional tunnel profile data and use simulation methods to obtain inverted gravity data and magnetic susceptibility of the tunnel site;

[0051] Step 2: Convert the acquired 2D tunnel profile data into 3D tunnel profile data;

[0052] The distance D1 is calculated based on the actual coordinates of two drill holes on the cross-section diagram. The formula is:

[0053] D1=sqart((x2-x1) 2 +(y2-y1) 2 ),

[0054] Where (x1, y1) is the coordinate of hole 1, and (x2, y2) is the coordinate of hole 2.

[0055] In CAD, the distance between the two holes is measured as D2, and the scaling ratio a is obtained by comparing the distance on the drawing with the actual distance. The distance on the drawing is converted into the size of the actual distance, the actual coordinate position is determined, and the coordinate orientation of the drawing is completed.

[0056] The actual three-dimensional coordinates on the section are obtained through mathematical calculations. The x-coordinate in the two-dimensional coordinate system is multiplied by the Cos(θ) and Sin(θ) values ​​of the angle between the section and the x-axis of the actual longitude and latitude coordinates. The depth Z value is the y-coordinate of any point on the map:

[0057]

[0058] Actual 3D coordinates of the section:

[0059]

[0060] Step 3: Reconstruct the data by combining the acquired 3D tunnel profile data, the inverted gravity data of the tunnel site, and the magnetic susceptibility to obtain a reconstructed data set.

[0061] like Figure 2 As shown, in order to effectively apply the data to machine learning algorithm training, the previously obtained data needs to be reconstructed.

[0062] Based on the 3D tunnel profile data, new data is generated by adding virtual boreholes at complex and special locations such as sparse profile data and faults. All the acquired data are then resampled and, combined with the inverted gravity and magnetic data models, discretized into a series of spatial points with attributes. The attribute features of the data can be selected from 3D geographic coordinates, density and magnetic susceptibility, and stratigraphic properties. The data are then resampled at regular intervals to convert them into stratigraphic sequences with spatial coordinates and stratigraphic numbers. The dataset includes datasets labeled with 3D geographic coordinates (x, y, z), gravity data, magnetic susceptibility for feature engineering, and stratigraphic categories (A, B, C, D, etc.).

[0063] In this embodiment, the drilling data are resampled at a certain distance. The sampling distance is determined according to the size of the model and is generally selected to be 0.5-5m.

[0064] In particular, geological structures such as faults can be regarded as a special stratigraphic lithology (such as fault rocks, caves, etc.).

[0065] Step 4: Train the geological attribute classifier based on the data set reconstructed in step 3, obtain the trained geological attribute classifier and verify it, such as Figure 3 As shown, the following steps are included:

[0066] Step 401: Divide the data set into two parts, a training set and a test set, uniformly and randomly according to a set ratio.

[0067] Step 402: Divide the training set data into n parts, select (n-1) parts of the data each time, and use the base classifier for training. According to the performance of different base classifiers with optimal hyperparameters in the test set, the adaptability of each type of base classifier is evaluated. If a base classifier performs too poorly, the base classifier can be replaced until satisfaction is achieved.

[0068] The base classifiers can be the same or different. Due to the high-dimensional data, the base classifier can choose a complex classifier, such as KNN, DT, RF, SVM, BP and other classifiers.

[0069] Step 403: The n prediction results obtained by training the base classifier are used as training data for retraining. Since the retrained meta-classifier is oriented towards low-dimensional data, a simple classifier can be selected for training, such as LR, LDA, XGBoost, etc.

[0070] Step 404: During model training, the relevant data generated due to construction excavation is added to the dataset for further training;

[0071] If relevant data is added due to construction excavation, the new data can be incorporated into the dataset, and the training set and test set can be re-divided to train the model. The attributes of the spatial coordinate points in the modeling area can be re-predicted to achieve the purpose of updating the established three-dimensional geological model. Through continuous updating, the established three-dimensional geological model of the tunnel is constantly approaching the actual three-dimensional geological conditions.

[0072] Based on the advanced geological prediction data during tunnel excavation, the surrounding rock properties of the tunnel, such as the stratigraphic occurrence, surrounding rock grade, strength, elastic modulus, wave velocity, and groundwater, are inverted. The three-dimensional geographic coordinates and the above attributes are exported as (x, y, z, attribute 1, attribute 2...attribute n). The x, y, z coordinates should be consistent with the subsequent SGrid grid coordinates. The attributes are stored in the grid model to further enrich the model information and provide data support for tunnel excavation and subsequent support design.

[0073] Step 405: Draw a confusion matrix diagram and a ROC curve diagram to evaluate the performance of the geological attribute classifier.

[0074] In the above stacking algorithm, each base learner is trained independently and can use different algorithms or parameters. The different characteristics and advantages of the base classifiers can be fully utilized, thereby improving the performance of the overall model.

[0075] Step 5: Based on the predicted values ​​of the geological attribute classifier and the surface elevation data, a three-dimensional geological model with surface undulations and structures is constructed. The specific steps include the following:

[0076] Step 501: Use the trained classifier to predict the geological attributes of each point in the modeled area of ​​the tunnel site. The attributes of a point in space can be represented as the attributes of a cube centered at that point. To improve classification accuracy and model precision, the cube size is preferably set within a preset range.

[0077] Step 502: Import the data predicted by the classifier into visualization software such as GOCAD, create a SGrid model, and assign different colors to different lithologies represented by different predicted values, and finally obtain a three-dimensional geological model of the tunnel with clear geological boundaries, including special geological structures and unfavorable geological bodies.

[0078] Step 503: Using the surface elevation data to constrain the three-dimensional geological model including the geological structure, a three-dimensional geological model with surface undulations and including the structure is established.

[0079] In this embodiment, when using elevation data to constrain model cutting, it should be ensured that the surface elevation data model range is larger than the modeling range.

[0080] Step 6: Sweep the 2D tunnel contour along the 3D tunnel path to form a 3D tunnel model. Perform Boolean operations on the 3D tunnel model and the 3D geological model obtained in step 5 to finally obtain a 3D geological model containing the tunnel, thus realizing intelligent 3D tunnel geological modeling under complex geological conditions. Figure 4 Shown is the final constructed three-dimensional geological model including the tunnel.

[0081] Example 2

[0082] This embodiment provides an intelligent 3D geological modeling system for tunnels under complex geological conditions, including:

[0083] A data conversion module, which is used to convert the acquired two-dimensional tunnel profile data into three-dimensional tunnel profile data;

[0084] A data reconstruction module is used to combine the acquired three-dimensional tunnel profile data, the inverted gravity data and magnetic susceptibility of the tunnel site area to perform data reconstruction and obtain a reconstructed tunnel data set;

[0085] A classifier training module, which is used to train a geological attribute classifier based on the reconstructed tunnel data set to obtain a trained geological attribute classifier;

[0086] The three-dimensional geological model construction module is used to predict the geological attributes of each point in the space within the tunnel site modeling area based on the trained geological attribute classifier, and to construct a three-dimensional geological model containing surface undulations and structures in combination with surface elevation data; the three-dimensional geological model containing surface undulations and structures is subjected to Boolean operations on the three-dimensional geological model containing the surface undulations and structures and the three-dimensional tunnel model obtained by sweeping along the three-dimensional path of the tunnel to obtain a three-dimensional geological model containing the tunnel.

[0087] Example 3

[0088] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps of the intelligent three-dimensional geological modeling method for tunnels under complex geological conditions as described above are implemented.

[0089] Example 4

[0090] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the intelligent three-dimensional geological modeling method for tunnels under complex geological conditions as described above are implemented.

[0091] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage, etc.) containing computer-usable program code.

[0092] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts 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, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0093] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0094] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0095] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0096] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. An intelligent three-dimensional geological modeling method for tunnels under complex geological conditions, characterized in that: The steps include: Converting the acquired two-dimensional tunnel profile data into three-dimensional tunnel profile data; The data is reconstructed by combining the acquired 3D tunnel profile data, the inverted gravity data and magnetic susceptibility of the tunnel site to obtain a reconstructed tunnel dataset. Training a geological attribute classifier based on the reconstructed tunnel data set to obtain a trained geological attribute classifier; The trained geological attribute classifier is used to predict the geological attributes of each point in the tunnel site modeling area. Combined with the surface elevation data, a three-dimensional geological model including the surface undulations and structures is constructed. A three-dimensional geological model including the surface undulations and structures is subjected to Boolean operations on the three-dimensional tunnel model obtained by sweeping along the three-dimensional path of the tunnel to obtain a three-dimensional geological model including the tunnel.

2. The intelligent three-dimensional geological modeling method for tunnels under complex geological conditions according to claim 1, characterized in that: The data is reconstructed by combining the acquired three-dimensional tunnel profile data, the inverted gravity data and the magnetic susceptibility of the tunnel site area to obtain a reconstructed tunnel data set, including: Based on the acquired 3D tunnel profile data, virtual boreholes are added at complex locations where profile data is sparse or contains faults to generate new virtual borehole data. The original borehole data, virtual borehole data and three-dimensional tunnel profile data were resampled and combined with the gravity data and magnetic susceptibility inverted in the tunnel site area to obtain a data set with three-dimensional geographic coordinates, gravity data and magnetic susceptibility as feature engineering and stratigraphic categories as labels.

3. The intelligent three-dimensional geological modeling method for tunnels under complex geological conditions according to claim 1, characterized in that: The step of training a geological attribute classifier based on the reconstructed tunnel data set to obtain a trained geological attribute classifier includes: The tunnel dataset is evenly and randomly divided into two parts: training set and test set according to the set ratio; The training set data is divided into n parts, and n-1 parts of the data are selected each time. The base classifier is used for training. The base classifier to be used is determined based on the performance of different base classifiers with the best hyperparameters in the test set. The n prediction results obtained by training the base classifier are used as training data for retraining, and the meta-classifier is trained to obtain a trained geological attribute classifier.

4. The intelligent three-dimensional geological modeling method for tunnels under complex geological conditions according to claim 1, characterized in that: During the training process, the geological attribute classifier obtains the data added during the tunnel construction and excavation process and adds it to the dataset. The training set and test set are re-divided, and the attributes of the spatial coordinate points in the modeling area are re-predicted to update the established three-dimensional geological model.

5. The intelligent three-dimensional geological modeling method for tunnels under complex geological conditions according to claim 1, characterized in that: During the training process, the geological attribute classifier also includes obtaining advanced geological prediction data during tunnel excavation to invert the properties of the tunnel surrounding rock, including stratum occurrence, surrounding rock grade, rock burst level, compressive strength, elastic modulus, wave velocity, and groundwater, and assigning the properties of the tunnel surrounding rock to the classifier.

6. The intelligent three-dimensional geological modeling method for tunnels under complex geological conditions according to claim 1, characterized in that: During the training process of the geological attribute classifier, the performance of the geological attribute classifier is evaluated by drawing a confusion matrix diagram and a ROC curve diagram.

7. The intelligent three-dimensional geological modeling method for tunnels under complex geological conditions according to claim 1, characterized in that: Combine the predictions from the geological attribute classifier with surface elevation data, including: Import the geological attributes of each point in the space within the tunnel site modeling area into the visualization software and create a SGrid model. Different lithologies represented by different predicted values ​​are assigned different colors. Ultimately, a three-dimensional geological model of the tunnel with clear geological boundaries, including special geological structures and unfavorable geological bodies, is obtained. The surface elevation data is used to constrain the three-dimensional geological model of the tunnel containing special geological structures and unfavorable geological bodies, and a three-dimensional geological model with surface undulations and structures is established.

8. An intelligent three-dimensional geological modeling system for tunnels under complex geological conditions, characterized by: include: A data conversion module, which is used to convert the acquired two-dimensional tunnel profile data into three-dimensional tunnel profile data; A data reconstruction module is used to combine the acquired three-dimensional tunnel profile data, the inverted gravity data and magnetic susceptibility of the tunnel site area to perform data reconstruction and obtain a reconstructed tunnel data set; A classifier training module, which is used to train a geological attribute classifier based on the reconstructed tunnel data set to obtain a trained geological attribute classifier; The three-dimensional geological model construction module is used to predict the geological attributes of each point in the space within the tunnel site modeling area based on the trained geological attribute classifier, and to construct a three-dimensional geological model containing surface undulations and structures in combination with surface elevation data; the three-dimensional geological model containing surface undulations and structures is subjected to Boolean operations on the three-dimensional geological model containing the surface undulations and structures and the three-dimensional tunnel model obtained by sweeping along the three-dimensional path of the tunnel to obtain a three-dimensional geological model containing the tunnel.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the intelligent three-dimensional geological modeling method for tunnels under complex geological conditions as described in any one of claims 1 to 7 are implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the intelligent three-dimensional geological modeling method for tunnels under complex geological conditions are implemented as described in any one of claims 1-7.

Images