A GIS-based three-dimensional geological modeling optimization method

Abstract

This invention discloses a GIS-based method for optimizing 3D geological modeling, belonging to the field of 3D geological modeling technology. The method includes: collecting multi-source geological data of the target area, performing standardization processing, and importing the processed data into a GIS platform to construct a GIS spatial database with spatial index; using an improved K-means and DBSCAN hybrid clustering algorithm, combined with the standardized multi-source geological data, to generate a geological information feature set; using an optimized spatial interpolation algorithm that integrates geostatistics and stochastic algorithms to construct a borehole 3D model and perform geological constraint verification and correction; using a multi-coupled cascaded hybrid density network, combining the input borehole 3D model data and the geological information feature set with GIS topology analysis to generate a 3D geological body model; optimizing the 3D geological body model, iteratively correcting the model through a preset quality assessment system, and outputting an optimized 3D geological model. This invention is applicable to various high-precision 3D geological modeling scenarios.

Description

Technical Field

[0001] This invention relates to the field of three-dimensional geological modeling technology, and in particular to a GIS-based method for optimizing three-dimensional geological modeling. Background Technology

[0002] Currently, 3D geological modeling technology is widely used in geological-related fields. Traditional modeling methods mainly rely on a single data source (such as borehole data) and use simple spatial interpolation algorithms such as Kriging interpolation and inverse distance weighted interpolation to construct geological models, assisting in the manual drawing of geological body boundaries to complete the modeling process. With the increasing accuracy requirements of geological exploration and the growing amount of multi-source geological data, traditional methods are gradually becoming unable to meet the needs of practical applications.

[0003] Existing 3D geological modeling methods suffer from the following core defects: First, they have weak multi-source data integration capabilities, failing to efficiently integrate geological data from sources such as boreholes, geophysical exploration, geological maps, and remote sensing images. This results in inconsistent data formats, excessive redundant data, and missing key information, leading to a weak foundation for modeling data. Second, their modeling accuracy is low, relying solely on traditional geostatistical methods without optimization using stochastic algorithms. This leads to biased interpolation results and insufficient consideration of the spatial correlation of geological bodies, resulting in models that do not accurately reflect actual geological conditions and exhibit poor geological rationality. Third, their automation level is low, with the geological body modeling process heavily reliant on manual intervention. Professionals are required to manually correct geological boundaries and adjust model parameters, which is not only time-consuming and labor-intensive but also prone to model distortion due to human error, making it particularly difficult to handle the modeling needs of complex geological structures. Fourth, their performance and versatility are poor. The models have low computational efficiency, making it difficult to support 3D modeling of large-scale areas. Furthermore, they lack a comprehensive model optimization and quality assessment system, resulting in poor visualization effects and an inability to directly adapt to subsequent spatial queries, engineering design, resource assessment, and other application scenarios.

[0004] In summary, the core pain points of existing technologies are insufficient fusion of multi-source geological data, a single spatial interpolation algorithm, difficulty in balancing automation and accuracy in geological body modeling, and the inability of model performance and versatility to meet the needs of practical applications. There is an urgent need for a three-dimensional geological modeling optimization method that can solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to provide a GIS-based three-dimensional geological modeling optimization method, which aims to solve or improve at least one of the above-mentioned technical problems.

[0006] To achieve the above objectives, the present invention provides the following solution: A GIS-based method for optimizing 3D geological modeling, characterized by the following steps: Collect multi-source geological data of the target area, perform standardized processing such as coordinate system unification, format unification and data cleaning, and import the processed data into the GIS platform to build a GIS spatial database with spatial index; An improved hybrid clustering algorithm combining K-means and DBSCAN was adopted, and combined with standardized multi-source geological data, spatial correlation and attribute similarity were calculated based on GIS spatial analysis to generate a geological information feature set. An optimized spatial interpolation algorithm that integrates geostatistics and stochastic algorithms is used to construct a three-dimensional borehole model and perform geological constraint verification and correction. A multi-coupled cascaded hybrid density network is used to combine the input borehole 3D model data with geological information feature sets and GIS topology analysis to generate a 3D model of the geological body. The three-dimensional model of the geological body is optimized, and the model is iteratively corrected through a preset quality assessment system to output the optimized three-dimensional geological model.

[0007] Furthermore, the multi-source geological data includes borehole data, geological profile data, geophysical data, remote sensing image data, and geological map data.

[0008] Furthermore, the hybrid clustering algorithm first uses K-means to perform preliminary clustering of global data to classify geological categories, and then uses DBSCAN to perform fine clustering of local data to extract small-scale geological information; the geological information feature set includes stratigraphic distribution features, structural distribution features, and lithological distribution features.

[0009] Furthermore, the optimized spatial interpolation algorithm specifically involves: using the variogram function in geostatistics to calculate the range and sill value spatial variation characteristics of borehole data, and determining the range of spatial correlation of the data; using the variogram function calculation results and the geological information feature set as input features for interpolation training of the random forest algorithm; and incorporating geological constraints such as stratigraphic contact relationships and structural strike during the interpolation process.

[0010] Furthermore, the multi-coupled cascaded hybrid density network sequentially comprises an input layer, a feature extraction layer, a coupling layer, and an output layer, wherein: The input layer is used to input borehole 3D model data and geological information feature sets; The feature extraction layer is used to extract the spatial morphological and attribute features of geological bodies; The coupling layer is used to integrate geostatistical spatial correlation calculations with random algorithm uncertainty analysis; The output layer is used to output three-dimensional models of geological bodies, including strata, faults, and rock masses.

[0011] Furthermore, the quality assessment system includes three core indicators: accuracy indicators, geological rationality indicators, and performance indicators.

[0012] Furthermore, optimizing the three-dimensional model of the geological body includes: refining the mesh in key areas and simplifying the mesh in non-key areas.

[0013] Furthermore, the optimized 3D geological model is output in obj, stl, and shp formats, which can be directly integrated into GIS platforms, engineering design software, and resource assessment software.

[0014] The present invention discloses the following technical effects: This invention utilizes standardized integration of GIS multi-source data, hybrid clustering information extraction, geostatistical and random forest optimization interpolation, and automated modeling of multi-coupled cascaded hybrid density networks. It significantly improves the accuracy, efficiency, and automation level of 3D geological modeling, effectively solving problems such as poor data fusion, low interpolation accuracy, high reliance on manual intervention, and insufficient model performance in traditional methods. It enables accurate representation of complex geological structures and rapid modeling of large-scale areas. The model is compatible with multiple platform outputs and can be widely applied to scenarios such as mineral exploration, engineering geology, disaster assessment, and underground space planning. It has strong practicality and promotional value. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

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

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] like Figure 1 As shown, this invention provides a GIS-based three-dimensional geological modeling optimization method, comprising the following steps: Step 1: Integration and Preprocessing of Multi-Source Geological Data Multi-source geological data was collected for the target area, including borehole data (borehole coordinates, depth, stratigraphic thickness, lithological description, sampling data, etc.), geological profile data (profile location, stratigraphic interfaces, fault parameters, etc.), geophysical data (seismic exploration data, electrical resistivity tomography data, etc.), remote sensing imagery (high-resolution satellite imagery, aerial remote sensing imagery), and geological map data (1:5000-1:10000 scale geological maps, including stratigraphic zoning, structural distribution, etc.). The collected data underwent standardization: all data were unified to the CGCS2000 National Geodetic Coordinate System, and the 1985 National Height Datum was adopted; data in different formats such as shp, dwg, txt, and tif were converted to GIS-compatible shp or GeoJSON formats; outliers in borehole and geophysical data were removed using the Z-score algorithm; redundant data was removed through duplicate data screening; and missing key data were supplemented using neighboring data interpolation. Standardized multi-source data is imported into GIS platforms such as ArcGIS or QGIS to build a GIS spatial database and create an R tree spatial index, enabling rapid data retrieval, querying, and management.

[0020] Step 2: Information Clustering Extraction A hybrid clustering algorithm combining improved K-means and DBSCAN is employed. First, the K-means algorithm performs preliminary clustering of the global data, quickly classifying it into major stratigraphic, structural, and lithological categories. Then, the DBSCAN algorithm performs fine-grained clustering of local data, extracting small-scale geological information such as fault zones and lithological zoning. Based on GIS spatial analysis capabilities, the spatial correlation (e.g., stratigraphic similarity between adjacent boreholes) and attribute similarity (e.g., lithological composition similarity) of each cluster unit are calculated to generate a geological information feature set, which is used as input parameters for the clustering algorithm to improve accuracy. The clustering results are then filtered and merged, removing invalid cluster units and retaining those consistent with actual geological conditions. Finally, key geological information such as stratigraphic distribution characteristics, structural distribution characteristics, and lithological distribution characteristics are generated, serving as constraints for subsequent modeling.

[0021] Step 3: Drilling Plane Visualization and Model Building Based on a GIS platform, borehole data is projected onto a plane to generate borehole distribution maps (labeled with borehole coordinates, depth, and stratum name), stratum thickness contour maps, and lithological zoning maps. This visually demonstrates the spatial distribution of boreholes and their correlation with geological information, providing a basis for adjusting interpolation algorithm parameters. An optimized spatial interpolation model integrating geostatistical variograms and random forest algorithms is constructed: A variogram is used to calculate spatial variation characteristics such as range and sill values ​​in the borehole data, determining the spatial correlation range of the data; the variogram calculation results and the geological information feature set extracted in step 2 are used together as input features for the random forest algorithm to train the interpolation model to achieve three-dimensional interpolation of the borehole data; geological constraints such as stratigraphic contact relationships and structural strikes are incorporated during the interpolation process to avoid abnormal areas that do not conform to geological laws. The geological constraints of the interpolated three-dimensional borehole model are verified by comparing the deviation between the actual borehole data and the interpolation results. If the deviation exceeds a preset threshold of 5%, parameters such as the number of random forest decision trees and the type of variogram are adjusted, and interpolation is repeated until the accuracy requirements are met.

[0022] Step 4: Geological Model Construction A multi-coupled cascaded hybrid density network is constructed, integrating stochastic algorithms and geostatistical calculations. This network comprises an input layer, a feature extraction layer, a coupling layer, and an output layer. The input layer takes into account the borehole 3D model data constructed in step 3 and the geological information feature set extracted in step 2. The feature extraction layer extracts the spatial morphological and attribute features of geological bodies. The coupling layer couples the spatial correlation calculated by geostatistical calculations with the uncertainty analysis of stochastic algorithms to optimize the fitting accuracy of geological body boundaries. The output layer outputs the 3D model contour data of geological bodies such as strata, faults, and rock masses. Based on GIS topology analysis functions, the output 3D model contour data of geological bodies is processed to achieve spatial stitching and trimming of geological bodies, verify topological relationships such as strata contact relationships and fault-strata cutting relationships, and remove topological conflicts such as overlapping and gaps in geological bodies to ensure the geometric rationality of the model. For complex geological structures such as reverse faults, folds, and magmatic intrusions, the fine modeling function of the hybrid density network, combined with profile analysis and buffer analysis in GIS spatial analysis, automatically fits the boundaries of complex geological bodies, reducing manual intervention.

[0023] Step 5: Overall optimization of 3D geological modeling and accuracy optimization: Based on the information clustering results from step 2, grid densification is applied to key areas such as ore body distribution areas, engineering sites, and geological hazard zones, while grid simplification is applied to non-key areas such as ordinary strata far from the engineering area. Cross-validation is used to compare the deviations between model data and actual geological exploration data, iteratively adjusting modeling parameters until the average model deviation is ≤3%. Model performance optimization: A multi-scale model is constructed using LOD (Level of Detail) technology. A simplified model is used for long-distance viewing to improve rendering efficiency, while a high-precision model is used for close-up viewing to ensure visualization effects. Model data is compressed and stored in a lightweight format to improve loading speed and computational efficiency. Model quality assessment and iterative correction: A quality assessment system is constructed, including accuracy indicators (deviation between model and actual geological data, interpolation accuracy), geological rationality indicators (correctness of geological body topological relationships, accuracy of geological structure expression), and performance indicators (model loading speed, rendering efficiency, data size). A model optimization report is automatically generated based on the assessment results, and iterative corrections are made to address any issues until the model meets all assessment indicators.

[0024] Step 6: Model Output and Application The optimized 3D geological model can be exported to multiple compatible formats such as obj, stl, and shp, and can be directly imported into GIS platforms, engineering design software such as AutoCAD, and resource assessment software for use in scenarios such as mineral resource reserve calculation, engineering site selection, geological hazard assessment, and urban underground space planning.

[0025] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0026] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A GIS-based method for optimizing three-dimensional geological modeling, characterized in that, Includes the following steps: Collect multi-source geological data of the target area, perform standardized processing such as coordinate system unification, format unification and data cleaning, and import the processed data into the GIS platform to build a GIS spatial database with spatial index; An improved K-means and DBSCAN hybrid clustering algorithm was adopted, combined with standardized multi-source geological data, and spatial correlation and attribute similarity were calculated based on GIS spatial analysis to generate a geological information feature set; An optimized spatial interpolation algorithm that integrates geostatistics and stochastic algorithms is used to construct a three-dimensional borehole model and perform geological constraint verification and correction. A multi-coupled cascaded hybrid density network is used to combine borehole 3D model data with geological information feature sets and GIS topology analysis to generate a 3D model of the geological body. The three-dimensional model of the geological body is optimized, and the model is iteratively corrected through a preset quality assessment system to output the optimized three-dimensional geological model.

2. The GIS-based three-dimensional geological modeling optimization method according to claim 1, characterized in that, The multi-source geological data includes borehole data, geological profile data, geophysical data, remote sensing image data, and geological map data.

3. The GIS-based three-dimensional geological modeling optimization method according to claim 1, characterized in that, The hybrid clustering algorithm first uses K-means to perform preliminary clustering of global data to classify geological categories, and then uses DBSCAN to perform fine clustering of local data to extract small-scale geological information; the geological information feature set includes stratigraphic distribution features, structural distribution features, and lithological distribution features.

4. The GIS-based three-dimensional geological modeling optimization method according to claim 1, characterized in that, The optimized spatial interpolation algorithm specifically involves: using the variogram function from geostatistics to calculate the range and sill value spatial variation characteristics of borehole data, and determining the range of spatial correlation of the data; using the variogram function calculation results and the geological information feature set as input features for interpolation training of the random forest algorithm; and incorporating geological constraints such as stratigraphic contact relationships and structural strike during the interpolation process.

5. The GIS-based three-dimensional geological modeling optimization method according to claim 1, characterized in that, The multi-coupled cascaded hybrid density network sequentially comprises an input layer, a feature extraction layer, a coupling layer, and an output layer, wherein: The input layer is used to input borehole 3D model data and geological information feature sets; The feature extraction layer is used to extract the spatial morphological and attribute features of geological bodies; The coupling layer is used to integrate geostatistical spatial correlation calculations with random algorithm uncertainty analysis; The output layer is used to output three-dimensional models of geological bodies, including strata, faults, and rock masses.

6. The GIS-based three-dimensional geological modeling optimization method according to claim 1, characterized in that, The quality assessment system includes three core indicators: accuracy indicators, geological rationality indicators, and performance indicators.

7. The GIS-based three-dimensional geological modeling optimization method according to claim 1, characterized in that, Optimizing the three-dimensional model of the geological body includes: refining the mesh in key areas and simplifying the mesh in non-key areas.

8. The GIS-based three-dimensional geological modeling optimization method according to claim 1, characterized in that, The optimized 3D geological model is output in obj, stl, and shp formats, and can be directly integrated into GIS platforms, engineering design software, and resource assessment software.

Images