Research on 3D Geological and Numerical Unified Model of in Mining Slope Based on Multi-Source Data

Abstract

As mining engineering progresses into the deep excavation phase, the intensification of high pressure, high temperature, strong disturbances, and complex geological conditions becomes increasingly prominent. Researchers perform stability analysis on the excavation area to reduce potential safety hazards during the extraction process. Developing a detailed numerical calculation model that accurately reflects the true geological structure is essential for numerical simulation analysis in mining engineering. Based on the excellent 3D geological modeling capabilities of 3D Mine software, this paper introduces a new 3D geological and numerical unified modeling method (3DMine-Rhino-HyperMesh) involving multi-software coupling and details the specific steps and concepts of this modeling approach. Subsequently, using a certain open-pit mine in Panzhihua as a backdrop, a detailed geological and numerical unified model is established, reflecting the true geological structure of the mining area, and the potential failure mechanisms of the mine slope are analyzed. The results indicate that the modeling method aligns well with the actual geological conditions, enhancing the grid quality of the numerical model and offering a new modeling approach for simulating and analyzing large complex geological entities in mining operations.

1. Introduction

In recent years, as the populace of China inflates and its economic sector thrives, the internal craving for subterranean mineral assets has ascended in a consistent manner. With the surface mineral wealth dwindling at an accelerated pace, the exploration of deeper mineral deposits has become an imperative development. Nevertheless, the exacerbation of mineral extraction has augmented the difficulties faced in mining operations. The intricacies of stratigraphic formations and geological circumstances pose considerable safety perils to the extraction and production of mineral resources. Scholars are endeavoring to lessen safety risks during mining operations by assessing the stability of mining locales. Crafting a sophisticated geological and numerical composite model that mirrors the genuine geological composition is the crux of precise numerical simulation analysis in the domain of mine engineering.

Numerical simulation techniques, particularly finite element and finite difference methods, are extensively utilized in engineering fields. However, the current modeling capabilities of numerical simulation software fall short of constructing intricate geological body models. Consequently, substantial terrain simplification becomes essential during complex geological modeling, leading to a marked discrepancy between the resulting numerical model and the actual geological formations. This compromises the accuracy of the computational outcomes. The use of multi-software coupled three-dimensional numerical modeling methods has emerged as an inevitable trend in the research on methodologies for constructing three-dimensional numerical models.

Commonly used numerical software such as FLAC 6.0, ANSYS 16.0, and ABAQUS 2024 often encounter limitations in their preprocessing modeling capabilities, particularly when constructing three-dimensional models of large and complex geological structures. To address this issue, several researchers have proposed various methods. Yu et al. [1] used Rhino software’s Discrete Element Method (DEM) to model features, imported the solid model into FLAC3D, and performed numerical simulation calculations. Jin et al. [2] employed high-precision digital elevation models of slopes, obtained through unmanned aerial vehicle (UAV) measurements and converted them to FLAC3D models using software such as Surfer 2024 and Rhino. Sun et al. [3] established a detailed 3D model of a tailings pond and its downstream terrain using Rhino, Fluent, and FLAC3D and analyzed the flow states of water and sand in the tailings pond breach under different conditions and terrains. Li et al. [4] constructed a three-dimensional numerical model of a tunnel portal using Rhino 2024, AutoCAD 2024, and FLAC3D 6.0, simulating the entire construction process. Xie et al. [5] employed a combined approach using Rhino, Kubrix, and FLAC3D for the numerical simulation and analysis of their study. Liu et al. [6] utilized Rhino-Grasshopper for parametric modeling to create roof models, which were then analyzed through numerical simulation analysis in ANSYS. Gao et al. [7] proposed a multi-software integrated modeling approach combining 3DMine, Surfer, Rhino, Ansys, and FLAC3D, solving issues related to the 3D numerical model construction of complex cavities and a large number of small-scale mining rooms and pillars. Cui et al. [8] suggested using Surfer as an intermediary platform, where data exported from Surfer were converted via a custom script written in the Fish language (an embedded language in FLAC3D) to generate model data files directly readable using FLAC3D, thereby enabling rapid and precise development of complex three-dimensional geological models during preprocessing. Liao et al. [9] developed a FLAC3D-ANSYS interface program in Visual Basic, realizing intuitive, rapid, and automated modeling in FLAC3D. These contributions underscore the trend towards integrating multiple software tools to overcome the limitations of current numerical simulation software, thereby enhancing the accuracy and efficiency of modeling complex geological entities.

Many scholars have carried out extensive research in the field of 3D geological and numerical unified models and stability analysis [10,11,12,13,14,15,16,17,18,19,20]. Wang et al. [21] used 3DLine Rhino FLAC3D coupling modeling technology to simulate and predict the excavation process of mining subsidence area in order to solve the problem of 3D geological and numerical unified model. Liu et al. [22] used the 3DLine Rhino Griddle-FLAC3D coupled modeling method to study the field of three-dimensional geology and numerical modeling, constructed a three-dimensional model of the mining area using this method, and conducted stability analysis on its mining process. Zhu et al. [23] and Lin et al. [24] used the 3DLine Ansys FLAC3D coupling modeling method to establish a three-dimensional geological and numerical model of the study area and used this method to visualize and simulate the mining area. Zeng et al. [25] used Midas FLAC3D coupling technology to visualize and numerically calculate the range of goaf collapse in a certain mining area.

There are many studies on the construction of a 3D geological and numerical unified model for high and steep slopes using multiple software couplings, but the constructed numerical model does not match the real geological conditions, and the grid quality is not ideal. Additionally, the process of simultaneously building visualization and computable models often consumes a significant amount of time. In contrast, the 3DMine mining software excels in 3D visualization model construction. By integrating multi-source information such as geological plans and exploration section maps, it can rapidly construct a comprehensive model of the research area. Through a modeling method that uses triangulation constraints on control points, it creates a 3D geological model that mirrors the true geological terrain. If this advantage is utilized to build a numerical calculation model, a 3D geological and numerical unified model reflecting the true geological body can be established. Leveraging the modeling capabilities of the 3DMine software, an innovative 3DMine-Rhino-Hypermesh coupled modeling method was proposed and validated through a certain open-pit mine slope in Panzhihua. This method innovates in the construction of a 3D geological and numerical unified model for high and steep slopes in open-pit mines, significantly reducing the time costs involved in model construction.

2. Construction Method of a Three-Dimensional Geological and Numerical Unified Model

Due to the current limitations of finite element numerical simulation software in modeling capabilities (such as FLAC3D, ABAQUS, etc.), it is difficult to construct a unified 3D visualization and numerical model for mine slopes. However, utilizing 3D modeling software can create a refined 3D geological model of complex geological bodies, complementing the 3D geological modeling method with numerical simulation software, and establishing a unified 3D geological and numerical model is an inevitable trend. This paper utilizes the advantages of 3D geological modeling software 3DMine to establish a research methodology for constructing a unified 3D geological and numerical model of complex geological bodies through the 3DMine-Rhino-HyperMesh coupled modeling method.

As illustrated in Figure 1, this study employs a coupled modeling approach. Initially, the method involves integrating and importing diverse sources of information, including geological plans and exploration profile maps, using 3DMine software. It employs a modeling technique constrained by control points through a triangular mesh to swiftly establish a surface model for the study area. Subsequently, by integrating stratigraphic boundaries and employing Boolean operations, a three-dimensional geological model is created, accurately reflecting the real geological terrain. Following this, the ground surface and stratigraphic triangular mesh are imported into Rhino software. A uniform point cloud is then laid over the surface, and using the Rhino-cock plugin, a uniform triangular mesh is reconstructed from the point cloud. Lastly, interpolation and Boolean operations are conducted on the reconstructed triangular mesh surface to categorize the model’s mineral and rock properties. Mesh plotting is carried out using HyperMesh software, culminating in the construction of a three-dimensional numerical model.

3. Construction of 3D Geological Model

The construction of a three-dimensional geological model enables the visualization of geological structures and a deeper understanding of internal structures and properties, which enables more detailed explanations and analyses of the mechanisms underlying complex geological processes. 3DMine software, a digital terrain model (DTM) generation software based on contour information, is a crucial tool for this task. The main steps for constructing a three-dimensional geological model using 3DMine software are DTM surface, strata boundaries, and Boolean operations. The following sections will detail these three steps.

(1) The process of constructing a digital terrain model (DTM) surface involves collecting contour data from various sources, such as topographic maps or LiDAR scans. The scattered data points are triangulated using a triangulation method, such as Delaunay or Voronoi triangulation, to construct the DTM surface model, as shown in Schematic Figure 2. The generated DTM surface can be visualized and further analyzed in geographic information systems (GIS) software or other three-dimensional (3D) software. This process enables a deeper understanding of the internal structures and properties of the terrain, which in turn enables more detailed explanations and analyses of the mechanisms underlying complex geological processes.

(2) The construction of strata boundaries primarily relies on geological drilling information. By connecting the strata boundary lines from each profile, the strata boundaries that traverse the outer contour model are formed, as shown in Figure 3. The use of triangulation methods effectively reflects the spatial position and trend of the geological body when connecting the strata boundary lines. Most mining software systems, including 3DMine and DIMine, support various triangulation methods, such as the minimum area method, the equidistant interval method, and the equal angle method. Each method has its advantages, which are summarized in

Table 1. Triangulated network connection method.

Triangulation Network Connection Method	Detailed Description
least area method	The surface area of the connected triangulation network is controlled to the minimum as far as possible.
equal central angle	Let the circumradius of each triangle be the same as far as possible.
Distance equal division method	Try to connect the points on the line segments with equal distance as much as possible.

. The triangulation approach used in 3DMine software is based on the use of contour information to generate a digital terrain model (DTM) surface, as mentioned above.

(3) During the formation and existence of rock masses, various geological processes have an impact on the rock mass, resulting in the presence of various discontinuous geological bodies such as joints, cracks, faults, and shear zones. To display and calculate these complex geological bodies, it is necessary to perform boolean operations and cutting on the surface strata boundaries of the geological model. Figure 4 illustrates the process of boolean operations. During the process of performing boolean operations, it is necessary to consider how to accurately identify and select the appropriate target objects for each operation, as well as how to avoid errors and ensure the accuracy of the results.

By executing a comprehensive three-dimensional geological model of the research domain, the spatial configuration of the geological formations and the stratigraphic constituents are graphically represented in an intuitively enhanced manner, thereby augmenting proficiency in predicting stratigraphic inclinations. Through in-depth research on the spatial morphology and structural distribution of geological formations from a three-dimensional perspective, a wealth of fundamental data can be obtained that can be used for further deep excavation. This approach can provide a more comprehensive understanding of the geological environment of the study area and facilitate better decision-making in future exploration and excavation activities.

4. Construction of the Three-Dimensional Numerical Model

4.1. Modeling Approach

Using the 3DMine mining engineering software, based on contour-based research of regional plans, exploration cross-sectional maps, and other multi-source geological information in .DWG format, combined with the construction method of triangulation networks, it is possible to achieve 3D visualization of complex geological terrain, structural plane attitudes, strata distribution, and geological information within the software. However, the method of constructing triangulation networks, despite its ability to realistically represent complex geological terrains, has become one of the main obstacles to the construction of 3D numerical models. Therefore, in practical applications, it is necessary to conduct in-depth research on this issue to achieve more accurate, efficient, and reliable construction of 3D numerical models.

The modeling capabilities of Rhino software are based on the utilization of embedded surface and draped-curtain functions in Rhino software to process triangulated meshes and construct portable 3D numerical models. This method can reduce the difficulty of model construction to some extent. However, the constructed models lack a strong correspondence with actual geological structures, which seriously undermines the reliability of computational results.

Utilizing a coupled modeling approach that combines multiple software, such as 3DMine-FLAC3D, during model construction, the three-dimensional visual model generated by 3DMine software is directly imported into the FLAC3D numerical software for elastoplastic calculations. This approach leverages existing geological data information, streamlines data processing, effectively addresses the shortcomings of finite element software pre-processing, reduces tedious and time-consuming modeling processes, and thereby shortens design cycles. However, it should be noted that there may be errors during the process of importing intersecting surfaces and bodies, which could potentially lead to modeling failure.

3DMine, Rhino, and other software have excellent capabilities in constructing complex geological models, and they each have their advantages in the process of building unified 3D geological and numerical models. HyperMesh software can generate meshes for complex models and convert them into FLAC3D models through interfaces, and this software can also be converted into each other through specific format files. The research method proposed in this paper aims to effectively construct a unified 3D geological and numerical model that reflects complex geological bodies. This method is based on a 3D geological model generated using 3DMine software, where a triangulated mesh model with surface, stratum, and fault information is transferred to Rhino software for reconstruction and interpolation of point clouds to generate surface, stratum, and fault surfaces. The process of constructing the unified 3D geological and numerical model mainly includes two steps: reconstructing the triangulated mesh and laying draped surfaces.

In this section, we describe the two most important steps in detail: (1) reconstructing the triangulated mesh and (2) laying draped surfaces. The process of reconstructing the triangulated mesh begins by extracting uniform point clouds from the triangulated mesh using information such as surface and stratum information, which is then reconstructed using Rhino-Cock plugins to generate 3D solid models. These solid models are then partitioned into meshes using HyperMesh software. Finally, the HyperMesh to FLAC3D program is used to construct a three-dimensional numerical model of complex geological bodies based on the generated 3D solid models.

4.2. Reconstructing Triangulation Network

The DTM surface triangulated mesh is constructed from control points on the contour lines, which accurately reflects the real geological body but also poses certain obstacles to the construction of the 3D numerical model. As shown in Figure 5, in this study, a triangulated mesh reconstruction method was used to import the DTM surface into Rhino software. A uniformly distributed point cloud is laid on the surface, and the Rhino-Cock plugin is used to reconstruct a uniform triangulated mesh from the point cloud. The process aims to transform the irregular DTM triangulated surface into a relatively uniform triangulated surface, providing advantages such as accuracy and ease of conversion for the reconstructed triangulated surface model.

4.3. Laying Cloth Curtain

As shown in Figure 6, the draping surface functionality in Rhino is established using a control point grid (control points), which is reminiscent of interpolation calculation methods. The control point grid is used to delineate complex surface shapes based on the distribution of control points on the surface, given specified boundaries and surfaces. The position of the control points on the draping surface can be arbitrarily adjusted, unlike interpolation methods. The control point grid is composed of control points and lines, arcs, etc., connecting these points. When constructing the draping surface, these control points need to be placed on the surface. By adjusting the position of these control points, the entire control point grid will change, and the surface will be adjusted to a new shape. During this draping interpolation process, Rhino helps users draw, edit, and arrange the control points using basic graphics and other tools to convert the complex control point grid into the desired surface.

5. Engineering Analysis of Mine Slope

Using the example of the Panzhihua City open-pit mine, we will demonstrate the method of constructing a unified 3D geological and numerical model using 3DMine software’s visualization model-building technology and Rhino software’s model-processing capabilities. As shown in Figure 7, the open-pit mine has reached the later stage of a deep-pit open-pit mine, forming multiple areas such as Dongshangtou, Shizishan, and southeast slope, with typical characteristics of a mid-to-low mountain structural erosion landform. The mining marks are relatively deep, with large differences in height, and presented as artificial step-like slopes. The slope angle is between 46° and 52°, and the final slope elevation is between 1000 and 1435 m, with a maximum height of over 400 m. The eastern hill has an overall slope of 50° with riser slopes of 65°, posing a serious threat of geological hazards. Furthermore, the research area encompasses twenty exploration cross-sections and sixteen sets of drilling data. The crux and challenges of model construction lie in the accurate reflection of the area’s distinctive attributes, including slope configuration, stratigraphic layers, and fault systems, within the three-dimensional numerical framework.

5.1. Three-Dimensional Geological and Numerical Unified Model

Before constructing the three-dimensional visualization model using 3DMine software, it is necessary to perform a check and optimization of the drawings. Missing contour line elevations need to be completed, as shown in Figure 8a,b. By integrating geological plan maps and exploration cross-sections, as well as information on faults and stratigraphic boundaries, a modeling method with triangulation network constraints on control points is employed to construct the stratigraphic boundaries of the mining area, as shown in Figure 9a. Based on this, a three-dimensional geological model reflecting the real geological terrain is constructed using the Boolean operation function, as shown in Figure 9b. The construction process of the fault model for mining slopes is the same as that of the geological model. Both are constructed through interpolation and Boolean operations. The constructed model is validated using typical cross-sections, as shown in Figure 10a,b, where the information displayed in the cross-sections is consistent with the dynamic cross-sectional display of the constructed three-dimensional geological model.

Using Rhino software to handle complex models, as demonstrated in Figure 11, we imported the triangulated mesh of the southeastern research area extracted from 3DMine software. As shown in Figure 12, a uniform point cloud is laid out within the region and the triangulated mesh is then reconstructed. This process aims to transform the irregular DTM triangulated surface into a relatively uniform triangulated surface, resulting in a reconstructed surface model of the triangulated mesh that is accurate and easy to convert. However, if cutting, boolean operations, or importing into HyperMesh software are performed in Rhino, errors in the topological relationships of adjacent areas may occur, leading directly to the failure of numerical model construction. Therefore, in this paper, the grid partitioning process is performed in HyperMesh, where topological relationships and similar issues do not need to be considered.

As shown in Figure 13, the local mine area on the southern slope of this mining site has a total of 6,804,504 elements and 1,252,125 nodes in the solid mesh partition. We believe that this approach significantly improves numerical model construction accuracy and efficiency.

5.2. The Value of Rock Mass Parameters Based on Non-Contact Measurement and Laboratory Experiments

The mechanical properties of the rock mass are ascertained via a synthesis of empirical data derived from laboratory-based rock testing and the extent of structural plane development. While rock parameters may be derived from controlled indoor experiments, the meticulous analysis of structural planes in the engineering assessment of high and mine slopes presents a formidable challenge. Therefore, unmanned aerial vehicle (UAV) photogrammetry technology is employed to capture the three-dimensional model of the study area, and the Shape MetriX software is used to statistically analyze the distribution of structural planes. Based on the data obtained from these two sources, the Chinese National Standard (CNSS) is applied to classify the quality of the rock mass, and recommended values for rock strength and mechanical parameters are provided.

(1)

Laboratory test

For the two typical rock types in the study area, namely, iron ore and fine-grained gabbro, uniaxial rock mechanics tests were conducted using the RMT-401 rock mechanics testing system (as shown in Figure 14). This system involves full-process control, the recording of various measurement data, and documentation preparation. The test specimen size used was a cylindrical shape with a diameter of 50 mm and a height of 10 mm.

In the course of this experimental procedure, a total of six test specimens were subjected to analysis, encompassing three iron ore samples and an equal number of fine-grained gabbro specimens. Prior to the conduct of tests, a comprehensive suite of physical parameters was meticulously measured for each specimen, as delineated in

Table 2. Uniaxial compression test of the specimen.

Rock Samples	Specimen Number	$\mathbf{Compressive} \mathbf{Strength} \mathsf{\sigma }$(MPa)	Elastic Modulus E (GPa)	Modulus of Deformation E0 (GPa)	Poisson Ratio μ
iron ore	Fe-1	193.083	84.297	48.711	0.255
	Fe-2	191.037	103.717	59.339	0.272
	Fe-3	154.368	94.127	42.744	0.271
	mean value	179.496	94.047	50.265	0.266
Fine-grained gabbro	V5-1	114.985	80.052	22.085	0.17
	V5-2	84.943	45.839	22.473	0.25
	V5-3	108.211	41.011	15.047	0.2
	mean value	99.964	62.946	22.279	0.21

.

The results of the uniaxial compression tests of iron ore and fine-grained gabbro are shown in

Table 3. Specifications of uniaxial compression test specimens.

Rock Samples	Specimen Number	Specimen Size (mm)		Quality m (g)	$\mathbf{Density} \mathit{\rho }$ (g/cm3)
		Diameter d	Height h
Ron ore	Fe-1	49.95	100.10	834.49	4.26
	Fe-2	49.98	100.01	846.52	4.32
	Fe-3	50.00	100.02	822.65	4.20
	mean value	49.98	100.04	834.55	4.26
Fine-grained gabbro	V5-1	49.98	100.09	808.84	4.12
	V5-2	50.02	100.06	805.49	4.10
	V5-3	50.01	99.95	816.00	4.17
	mean value	50.00	100.03	810.11	4.13

. The results indicate that the average uniaxial compressive strength of iron ore is 179.496 MPa, the average elastic modulus is 94.047 GPa, and the Poisson’s ratio is 0.266; while the average uniaxial compressive strength of fine-grained gabbro is 99.964 MPa, the average elastic modulus is 62.946 GPa, and the Poisson’s ratio is 0.21.

(2)

Statistical results of joints

To accurately estimate the distribution of fractures, three photo-control points were placed in the study area to determine the spatial coordinates of the model. Then, the DJI Phantom 4 RTK drone was used to perform aerial photography in the study area to obtain 91 high-resolution images, which were imported into Shape MetriX software for 3D reconstruction, by incorporating the coordinates of three known photo-control points, a 3D model of the slope was obtained, as shown in Figure 15a, with a length of 19.2 m, width of 6.15 m, and height of 4.35 m, composed of 78,756 point clouds.

On this basis, structural analysis was conducted on a total of 94 complete structural surfaces. The orientations of the 94 structural surfaces were subjected to K-means clustering analysis, as shown in Figure 15b. The results indicated that the dominant fracture sets in the area can be classified into three groups: J1, consisting of 30 structural surfaces with a strike of 212.9° and a dip of 29.73°; J2, consisting of 32 structural surfaces with a strike of 234.71° and a dip of 57.89°; and J3, consisting of 32 structural surfaces with a strike of 184.44° and a dip of 62.05°. The trace statistics of the three fracture sets are shown in Figure 16, with average spacing values of 0.87 m, 1.12 m, and 2.5 m, respectively.

(3)

Rock mass quality classification

Drawing upon the findings from the aforementioned indoor rock mechanics experiments and statistical analyses of fractures, the Chinese National Standard classification system, a paragon of methodological precision, was applied to evaluate the quality of the rock mass within the study area. The saturated uniaxial compressive strength of the rock served as a proxy for its mineral rigidity, whereas the count of rock volume joints and the mean spacing between structural planes were employed to gauge the integrity of the rock mass. The collated outcomes are encapsulated in

Table 4. Qualitative estimation methods.

Volumetric Joint Count of Rock Mass (Strip/m3)	Development Degree of Structural Plane		Bonding Degree	Rock Mass Integrity	Kv
	Number of Classes	Average Distance (m)
<3	1~2	>1.0	integration	integrity	>0.75
3~10	1~2	>1.0	Poor combination	relatively complete	0.75~0.55
	2~3	1.0~0.4	integration
10~20	2~3	1.0~0.4	Poor combination	less cracked	0.55~0.35
	≥3	0.4~0.2	integration
20~35	≥3	0.4~0.2	Poor combination	crushing	0.35~0.15
>35	\	\	Poor bonding	extremely fractured	<0.15

.

Due to the presence of multiple dominant fracture sets in the rock mass, a method for calculating the average spacing of multiple dominant fracture sets is proposed:

$${\displaystyle \frac{1}{d_{ev}}}={\displaystyle \frac{1}{d_1}}+{\displaystyle \frac{1}{d_2}}+\dots +{\displaystyle \frac{1}{d_n}}$$

In the formula, d_ev represents the average spacing of fractures in the rock mass, d_i represents the average spacing of the i-th group of fractures, and n represents the number of dominant fracture sets.

Based on the formula above, the average spacing of fractures in various locations can be calculated. The average spacing of fractures in the selected area is 0.41 m. Based on Table 4, the value of the selected area can be estimated to be 0.35. According to the saturated uniaxial compressive strength and integrity of the rock mass, the quality classification of the rock mass can be conducted, as shown in

Table 5. Rock mass quality grading in the selected area.

Lithologic Characters	${\mathit{R}}_{\mathit{c}}$ (MPa)	BQ	Classification Results
Fine-grained gabbro	99.964	372	$III$
Fe	179.496	372	$III$

.

In summary, the rock mass primarily belongs to the category of fair to poor rock quality. Based on the suggested values from the ‘Engineering Rock Mass Classification Standard’ and considering the results of laboratory tests, the recommended values for the physical and mechanical parameters of the rock mass are presented in

Table 6. Recommended values for physical and mechanical parameters of rock mass.

Rock Mass	γ (kN/m3)	C (kPa)	φ (°)	E (GPa)	ν
Fine-grained gabbro	31.00	800	39	10	0.2
mining area	34.00	700	39	7.2	0.27
fault	25.00	50	26	0.09	0.3

.

5.3. Three-Dimensional Stability Analysis of Mine Slope

Based on laboratory test results and field investigations, stability calculations, and analyses were conducted on the excavation process of the southeastern slope region. Figure 17a–c presents contour maps of the slope geo stress field, displacement field, and plastic zone distribution after excavation. The specific analysis results are as follows:

The spatial distribution of overall slope geo stress as shown in Figure 17a displays variability, with a gradual increase from top to bottom, and the stress near the slope surface exhibits a parallel characteristic to the slope face. Figure 17b indicates that the displacement of the excavated slope is predominantly vertical and upward, and the displacement disturbance caused by the excavation process occurs in localized areas. Certain areas near the slope base exhibit a vertical rebound, and certain displacement can also be observed at a certain depth below the slope bottom, indicating that the excavation-induced disturbance has a certain influence on the stress distribution in the lower part. The overall displacement of the slope surface tends to gradually increase from the slope top to toe while decreasing gradually from the slope surface towards the interior. Through observation of Figure 17c, it is found that the number of plastic elements within the slope body increases. Faults F03 and F111’ cut into the slope top at 1438–144 m, forming wedge-shaped bodies located on the slope platform at 1232–1238 m. This suggests that the internal connectivity of the slope body after excavation has further enhanced, indicating potential risks of sliding and failure. Similarly, faults F111 and F122 cut into the slope top at 144 m, forming a wedge-shaped body located on the slope platform at 1285–1287 m. This indicates potential damage. Figure 17a–c represents contour maps of the slope geo stress field, displacement field, and plastic zone distribution after excavation. The presented figures provide a detailed analysis of the post-excavation deformation characteristics of slopes, indicating potential risks and damage.

The plastic zone of the slope rock mass Is mainly distributed near faults and at the foot of the slope. Rock mass failure is mainly characterized by shear failure and tension-shear mixed failure. Although the plastic zone cannot directly reflect the instability and slope failure of the entire rock mass slope, it can reflect the development trend and dangerous positions of local instability and failure. Based on the distribution of the plastic zone in some sections, the probability of instability and slope failure in these areas is relatively high. Therefore, during the mining process of the ore body, attention should be paid to the condition of these slopes, and timely monitoring should be conducted to detect the germination and evolution process of potential geological disasters early.

6. Conclusions

In response to the challenges in developing multi-source data fusion models for high and steep slopes, leveraging the advanced modeling capabilities of the 3DMine mining engineering software, an innovative 3DMine-RhinoHyperMesh multi-software coupling modeling approach is introduced. The study details the modeling process and performs a stability analysis, leading to the following findings:

(1)

The three-dimensional (3D) geological model visualization is constructed using 3DMine software, integrating various data sources such as geological plan maps and exploration cross-sections. Employing a modeling technique that utilizes triangular mesh control points, a highly precise 3D geological model representing the authentic geological terrain is generated. This model offers advantages including exceptional precision, comprehensive information, and lifelike visual realism.

(2)

Following an analysis of the strengths and limitations of different modeling approaches, this paper adopts a multi-software coupled modeling approach termed “3DMine-Rhino-HyperMesh” to establish a unified 3D geological and numerical model reflecting the genuine geological structure. The innovation lies in leveraging Rhino software’s triangulation network reconstruction method to seamlessly convert the 3D geological model into a 3D numerical model. This process transforms irregular DTM triangulated surfaces into more uniform ones and utilizes embedded surfaces to construct the 3D numerical model. The resulting model exhibits excellent conformity with the actual geological formations, enhancing the quality of the numerical grid.

(3)

Using a specific open pit in Panzhihua City as a case study, this research employs a multi-software coupling approach to construct a unified 3D geological and numerical model and assesses the potential landslide risks during mining operations. The analysis reveals a heterogeneous spatial distribution of overall slope geo stress, progressively increasing from top to bottom. Stress near the slope surface aligns parallel to the slope, with fault formations manifesting as unstable wedge-shaped bodies. Post-excavation, internal connectivity within the slope improves, with fault fracture zones posing risks of deep-seated landslides.