A method for metallogenic prognosis based on quantitative analysis of network topology

Abstract

This invention discloses a mineralization prediction method based on quantitative analysis of network topology, belonging to the field of mineral exploration. The method includes the following steps: S1, acquiring geological profile maps of the target mining area's exploration lines and performing vectorization processing to obtain a vectorized vein network; S2, calculating two-dimensional mineralization intensity and establishing a two-dimensional profile vein digitization system; S3, extracting node-branch structures and defining topological parameters; S4, determining the quantitative correlation between mineralization enrichment and the topological characteristics of the vectorized vein network; S5, discretizing and calculating the probability distribution of each discrete interval; S6, quantitatively evaluating the reliability of the sampling scheme and determining the optimal sampling parameters; S7, screening spatial regions with high connectivity and high mineralization potential; S8, determining the dominant migration channels of ore-forming fluids; and S9, determining the optimal extension direction of the ore body. Using this mineralization prediction method based on quantitative analysis of network topology, efficient, accurate, and quantitative deep mineralization prediction can be achieved.

Description

Technical Field

[0001] This invention relates to the field of mineral exploration technology, and in particular to a mineralization prediction method based on quantitative analysis of network topology. Background Technology

[0002] Quartz vein gold deposits are an important type of gold mining globally, possessing extremely high value for resource development and strategic reserves. With the continuous depletion of shallow, high-grade, large-scale quartz vein gold resources, exploration of deep ore bodies has become a core direction for ensuring the sustainable supply and succession of gold resources.

[0003] The current deep exploration of quartz vein-type gold deposits faces prominent technical challenges: First, the number of deep quartz veins has been significantly reduced, making it difficult to effectively determine the deep mineralization potential; second, quartz vein-type ore bodies generally exhibit pinch-out and recurrence, branching and complex formations, and disordered spatial distribution, making it impossible to accurately recover the migration paths of ore-forming fluids and extremely difficult to delineate deep exploration target areas; third, the grade variation coefficient of this type of ore body is large, directly increasing the risk of exploration engineering deployment and the complexity of resource evaluation.

[0004] Traditional gold exploration methods rely heavily on statistical analysis of large quantities of physical samples. In scenarios where the amount of samples obtained in deep exploration is limited, such empirical methods are difficult to adapt to the analytical needs of ore body grade fluctuations and cannot support deep mineralization prediction and resource evaluation, resulting in slow growth in newly discovered deep gold resources. Summary of the Invention

[0005] The purpose of this invention is to provide a mineralization prediction method based on quantitative analysis of network topology, thereby solving the aforementioned technical problems.

[0006] To achieve the above objectives, this invention provides a mineralization prediction method based on quantitative analysis of network topology, comprising the following steps: S1. Obtain the geological profile map of the exploration line of the target mining area, and perform vectorization processing on the vein network in the profile to obtain the vectorized vein network. Then, use the complementary cumulative distribution function combined with the power law model, log-normal model and exponential distribution model to quantitatively characterize the vein length distribution characteristics. Quantify the vein's dominant orientation and spatial heterogeneity through the length-weighted equal-area rose diagram and the direction anisotropy index. S2. Construct a two-dimensional profile vein digital system based on two-dimensional mineralization intensity. Calculate the two-dimensional mineralization intensity based on the vein length distribution characteristics and profile area characterized in step S1, and establish a two-dimensional profile vein digital system. S3. Based on the two-dimensional profile ore vein digital system of step S2, extract the node-branch structure, classify and statistically analyze the node type and branch type, define three types of topological parameters: average number of connections per branch, number of branches and node density, and quantitatively characterize the connectivity and structural complexity of the ore vein network. S4. Based on the topological parameters obtained in step S3 and the two-dimensional mineralization intensity obtained in step S2, perform correlation analysis to determine the quantitative correlation between the degree of mineralization enrichment and the topological characteristics of the vectorized vein network. S5. Based on the topological parameters of step S3 and the two-dimensional mineralization intensity of step S2, a multi-scale random circular subsampling method is used to fully cover the geological profile of the exploration line. The topological parameters and the two-dimensional mineralization intensity are discretized and the probability distribution of each discrete interval is calculated. S6. Based on the probability distribution results of step S5, the information entropy formula is used to calculate the parameter distribution entropy value under different sampling scales and sampling numbers, quantitatively evaluate the reliability of the sampling scheme and determine the optimal sampling parameters. S7. Based on the quantitative correlation in step S4 and the optimal sampling parameters in step S6, a two-dimensional topology-mineralization coupling model is constructed. In the two-dimensional topology-mineralization coupling model, spatial regions that meet the topology parameter thresholds are selected and identified as spatial regions with high connectivity and high mineralization potential. S8. Based on the topology parameters of step S3, the vectorized vein network of step S1 is transformed into a weighted directed graph. The Dijkstra algorithm is used to calculate the shortest path of the ore-forming fluid from the inflow end to the outflow end within the profile, and the dominant migration channel of the ore-forming fluid is determined. S9. Considering the vein's dominant orientation data from step S1, and based on the highly connected and mineralized potential areas identified in step S7 and the dominant migration channels of ore-forming fluids determined in step S8, perform spatial overlay analysis to delineate high-potential exploration target areas under dual constraints and determine the optimal extension direction of the ore body.

[0007] Preferably, step S1 specifically includes the following steps: S11. Collect geological profile maps of the exploration line in the target mining area and extract the spatial distribution, morphology and location data of the original veins on the profile. S12. Vectorize the vein network in the geological profile of the exploration line, converting the graphical veins into digital vector data to form a vectorized vein network. S13. Quantitative characterization of vein length distribution characteristics; S131. Extract the length data of a single vein from the vectorized vein network, and after removing outlier data, use the complementary cumulative distribution function to perform statistical calculations on the vein length data to characterize the vein length distribution characteristics and obtain a vein length sample dataset. S132. Using the power law model as the core, simultaneously establish complementary cumulative distribution function expressions for the log-normal model and the exponential distribution model as the fitting benchmark function; S133. Perform optimal fitting on the power law model, log-normal model and exponential distribution model respectively to obtain the fitting function and fitting curve of the three types of models; S134. Based on the degree of fit between the fitting curves of the power law model, log-normal model and exponential distribution model and the measured complementary cumulative distribution function, the model with the highest fitting degree is selected as the optimal distribution model for vein length. S135. Based on the optimal distribution model of vein length determined in step S134, complete the quantitative characterization of vein length distribution. S14. Based on the orientation data of the vectorized vein network, draw a length-weighted equal-area rose diagram to realize the spatial visualization and quantitative determination of the dominant orientation of the vein. S15. Calculate the directional anisotropy index, and quantitatively assess the spatial heterogeneity of the vectorized vein network based on the index to complete the prior analysis of the vein's geometric characteristics.

[0008] Preferably, step S2 specifically includes the following steps: S21. Based on the vectorized vein network and vein length distribution data from step S1, count the number of all veins within the target exploration line profile. Length of a single ore vein Calculate the average length of the ore vein and the total length of the vein within the profile ; Simultaneously, the actual sampling area of ​​the two-dimensional exploration line profile was measured and determined. ; S22, Calculate the two-dimensional mineralization intensity : ; S23. Calculate the fracture frequency and crack strength Complete the construction of a two-dimensional profile ore vein digital system: ; / .

[0009] Preferably, the node types mentioned in step S3 include isolated node I and connected node, and the connected node includes branch node Y and intersection node X; wherein, isolated node I is the endpoint of a vein segment; branch node Y is a node where one vein branches off from another, forming a three-way intersection node; intersection node X is a node where two veins intersect each other, forming a four-way intersection node. Branch types include CC type branches, IC type branches, and II type branches. CC type branches are branches between connecting nodes, used to reflect the development degree of the main vein network; IC type branches are branches between isolated nodes and connecting nodes, used to characterize the secondary connection characteristics of the vein network; and II type branches are branches between isolated nodes, used to represent the terminal branch characteristics of the vein network. The number of connections per branch is the ratio of the total number of connections within the node-branch structure to the total number of branches, used to quantitatively characterize the connectivity index of the vein network in a single profile; the number of branches is the total number of all branches within the node-branch structure, used to reflect the development scale of the vein network; the node density is the total number of intersection nodes X and branch nodes Y within a unit profile area, used to characterize the degree of convergence development and the development status of spatial transmission channels in the vein network.

[0010] Preferably, in step S4, the two-dimensional mineralization intensity of S2 is used as the dependent variable, and the average number of connections per branch, the number of branches, and the node density of S3 are used as independent variables; then, the linear correlation analysis method is used to calculate the correlation coefficient between each topological parameter and the two-dimensional mineralization intensity; finally, based on the correlation coefficient values, the quantitative correlation strength between the degree of mineralization enrichment and the topological characteristics of the vectorized vein network is determined.

[0011] Preferably, step S5 specifically includes the following steps: S51. Set the radius gradient and quantity gradient of the multi-scale random circle so that it covers the entire geological profile of the exploration line. S52. Generate random circles according to the set radial gradient and quantitative gradient, and perform random circle subsampling on a single profile; S53. Extract the average number of connections per branch, number of branches, node density, and two-dimensional mineralization intensity data within each random circle; S54. Set discrete intervals for the extracted average number of connections per branch, number of branches, node density and two-dimensional mineralization intensity data respectively, and complete the data discretization process; S55. Count the number of times the data appears in each discrete interval, and calculate the probability distribution of each discrete interval.

[0012] Preferably, step S6 specifically includes the following steps: S61. Based on the probability distribution results of S5, use the information entropy formula to calculate the parameter distribution entropy value of a single random circle and the average entropy value of all random circles under the same sampling scale. S62. Compare the average entropy values ​​of sampling schemes with different radii and numbers to evaluate the sampling reliability; and select the sampling radius and sampling number with the optimal parameter distribution entropy value as the optimal sampling parameters.

[0013] Preferably, in step S8, the edge weights are defined by the inverse ratio of the topological parameters of adjacent nodes in the weighted directed graph to the two-dimensional mineralization intensity. At the same time, the inflow end of the profile is set as the source vertex and the outflow end as the target vertex. The Dijkstra algorithm is used to calculate the minimum weight path from the source vertex to the target vertex and determine it as the dominant transport channel of the ore-forming fluid.

[0014] Preferably, step S9 specifically includes the following steps: S91. Load the dominant orientation data of the ore vein obtained in step S1 and use it as the initial structural constraint condition for the extension direction of the ore body. Simultaneously, the spatial region data with high connectivity and high mineralization potential obtained in step S7 is loaded to form the first prediction constraint layer. Load the dominant migration channel data of ore-forming fluids obtained in step S8 to form a second predictive constraint layer; S92. Perform spatial overlay analysis on the first prediction constraint layer and the second prediction constraint layer, and extract the spatial intersection area between the two. S93. Identify the spatial intersection area as a high-potential exploration target area and complete the dual-constraint delineation of the target area; S94. Based on the dominant orientation of the ore vein, trend fitting is performed along the overall direction of the dominant migration channel of the ore-forming fluid to determine the optimal extension direction of the ore body. S95 outputs the spatial range of high connectivity and high mineralization potential, as well as the optimal extension direction of the ore body, to complete the deep mineralization prediction.

[0015] Therefore, the beneficial effects of the mineralization prediction method based on quantitative analysis of network topology adopted in this invention are as follows: 1. The two-dimensional topology-mineralization coupling model constructed based on this invention (this model is constructed by combining three levels: topology parameter statistical analysis, multi-scale random sampling and information entropy analysis, and shortest path solution for profile fluids, to quantitatively characterize the complexity and connectivity of ore-controlling structural branches and their correlation with gold mineralization enrichment) can quickly filter for high average number of connections per branch through topology parameter thresholds. High node density ), high number of branches ( In this area, potential rich mineral target areas are efficiently delineated, significantly improving the efficiency of exploration target area location; 2. It takes into account the dual ore-controlling effects of structural reactivation and network connectivity. Among them, the geometric analysis of vein network and the digitization of the vein profile serve as the geometric and topological data preparation stage of the invention method, providing a solid foundation for the implementation of the method. The prediction accuracy is significantly better than traditional single geometric analysis or geochemical analysis methods, providing a data-driven technical path for the selection of target areas in the field. 3. The shortest path network constructed based on the Dijkstra algorithm can intuitively reveal the spatial distribution pattern of dominant migration channels of ore-forming fluids. High-grade mineralization is significantly concentrated in areas with dense local shortest paths, directly verifying the key control role of topological connectivity on mineralization enrichment. Simultaneously, it can quickly determine the dominant extension direction of the ore body and indicate the main faults, solving the problem of where the ore body extends at depth. Furthermore, it can compare the shortest path from the inflow end to the outflow end of each profile with the shortest path on the profile plan. , , , By superimposing and comparing the spatial distribution, the superposition area of ​​high topological connectivity region and fluid dominant channel is obtained, and the potential target area is constrained in a dual manner. This provides accurate structural criteria for the identification and positioning of core dominant channels for hydrothermal fluid migration and metal precipitation, and clarifies the node-controlled mineralization enrichment law. 4. The method and process of constructing a two-dimensional topology-mineralization coupling model in this invention can be extended from the two-dimensional profile model to the construction of a three-dimensional fracture network. Combined with geochemical isotope data, it can further improve the spatial prediction accuracy and applicability of hydrothermal fluid mineralization processes, and expand the application scenarios and practical value of the method.

[0016] In summary, the method of this invention uses topology, information entropy, and graph theory algorithms to quantitatively analyze the topological connectivity of quartz vein networks and quantify its intrinsic relationship with mineralization enrichment intensity. Ultimately, it establishes a new data-driven paradigm for mineralization prediction that moves from qualitative morphology to quantitative network analysis. This achieves the goal of predicting the optimal extension direction of ore bodies and rapidly delineating potential rich ore target areas, thus overcoming the technical bottlenecks of high exploration engineering risks and difficult target area prediction in deep mineral exploration.

[0017] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0018] Figure 1 This is a flowchart of a mineralization prediction method based on quantitative analysis of network topology as described in this invention. Figure 2 A ternary phase diagram for the branch and node types of the No. 10 vein in the Linglong Gold Mine described in the example; Figure 3 This is a digitally processed topology model of the No. 10 vein of the Linglong Gold Mine described in the embodiment. Figure 4 The above is a comparison of the spatial distribution of four core parameters of the Linglong Gold Mine No. 41 exploration line profile described in the embodiment. Among them, (a) is the spatial distribution of two-dimensional mineralization intensity, (b) is the spatial distribution of the number of connections per branch, (c) is the spatial distribution of node density, and (d) is the spatial distribution of the number of branches. Figure 5 The following is a quantitative analysis diagram of the correlation between mineralization intensity and three types of topological parameters of exploration line No. 41 in Linglong Gold Mine described in the embodiment. Among them, (a) is the correlation coefficient analysis diagram of mineralization intensity and the number of connections per branch, (b) is the residual distribution diagram of mineralization intensity and the number of connections per branch, (c) is the correlation coefficient analysis diagram of mineralization intensity and the number of branches, (d) is the residual distribution diagram of mineralization intensity and the number of branches, (e) is the correlation coefficient analysis diagram of mineralization intensity and node density, and (f) is the residual distribution diagram of mineralization intensity and node density. Figure 6 This is a parameter analysis diagram of the fracture identification of the main vein network described in the embodiment. Figure 7 This is a coupled model diagram of the fracture network and the dominant flow path of the ore-forming fluid in the No. 10 vein of the Linglong Gold Mine, as described in the embodiment. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the embodiments of the present invention and are not intended to limit the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of this application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout.

[0020] It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, such as a process, method, system, product, or server that includes a series of steps or units, not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or device.

[0021] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0022] like Figure 1 As shown, a mineralization prediction method based on quantitative analysis of network topology includes the following steps: S1. Obtain the geological profile map of the exploration line of the target mining area, and perform vectorization processing on the vein network in the profile to obtain the vectorized vein network. Then, use the complementary cumulative distribution function combined with the power law model, log-normal model and exponential distribution model to quantitatively characterize the vein length distribution characteristics. Quantify the vein's dominant orientation and spatial heterogeneity through the length-weighted equal-area rose diagram and the direction anisotropy index.

[0023] S2. Construct a two-dimensional profile ore vein digital system based on two-dimensional mineralization intensity. Calculate the two-dimensional mineralization intensity based on the vein length distribution characteristics and profile area characterized in step S1, and establish a two-dimensional profile ore vein digital system.

[0024] S3. Based on the two-dimensional profile digitization system of the ore vein in step S2, extract the node-branch structure, classify and statistically analyze the node type and branch type, define three types of topological parameters: average number of connections per branch, number of branches, and node density, and quantitatively characterize the connectivity and structural complexity of the ore vein network.

[0025] S4. Based on the topological parameters obtained in step S3 and the two-dimensional mineralization intensity obtained in step S2, perform correlation analysis to determine the quantitative correlation between the degree of mineralization enrichment and the topological characteristics of the vectorized vein network.

[0026] S5. Based on the topology parameters of step S3 and the two-dimensional mineralization intensity of step S2, a multi-scale random circular subsampling method is used to fully cover the geological profile of the exploration line. The topology parameters and the two-dimensional mineralization intensity are discretized and the probability distribution of each discrete interval is calculated.

[0027] S6. Based on the probability distribution results of step S5, the information entropy formula is used to calculate the parameter distribution entropy values ​​under different sampling scales and sampling quantities, quantitatively evaluate the reliability of the sampling scheme, and determine the optimal sampling parameters.

[0028] S7. Based on the quantitative correlation in step S4 and the optimal sampling parameters in step S6, a two-dimensional topology-mineralization coupling model is constructed. In the two-dimensional topology-mineralization coupling model, spatial regions that meet the topology parameter thresholds are selected and identified as spatial regions with high connectivity and high mineralization potential.

[0029] S8. Based on the topology parameters of step S3, the vectorized vein network of step S1 is transformed into a weighted directed graph. The Dijkstra algorithm is used to calculate the shortest path of the ore-forming fluid from the inflow end to the outflow end within the profile, and the dominant migration channel of the ore-forming fluid is determined.

[0030] S9. Considering the vein's dominant orientation data from step S1, and based on the highly connected and mineralized potential areas identified in step S7 and the dominant migration channels of ore-forming fluids determined in step S8, perform spatial overlay analysis to delineate high-potential exploration target areas under dual constraints and determine the optimal extension direction of the ore body.

[0031] Step S1 specifically includes the following steps: S11. Collect geological profile maps of the exploration lines in the target mining area and extract the spatial distribution, morphology, and location data of the original veins on the profile.

[0032] S12. Vectorize the vein network in the geological profile of the exploration line, converting the graphical veins into digital vector data to form a vectorized vein network.

[0033] S13. Quantitative characterization of vein length distribution characteristics; S131. Extract the length data of a single vein from the vectorized vein network, and after removing outlier data, use the complementary cumulative distribution function to perform statistical calculations on the vein length data to characterize the vein length distribution characteristics and obtain a vein length sample dataset. S132. Using the power law model as the core, simultaneously establish complementary cumulative distribution function expressions for the log-normal model and the exponential distribution model as the fitting benchmark function; S133. Perform optimal fitting on the power law model, log-normal model and exponential distribution model respectively to obtain the fitting function and fitting curve of the three types of models; S134. Based on the degree of fit between the fitting curves of the power law model, log-normal model and exponential distribution model and the measured complementary cumulative distribution function, the model with the highest fitting degree is selected as the optimal distribution model for vein length. S135. Based on the optimal distribution model of vein length determined in step S134, complete the quantitative characterization of vein length distribution.

[0034] S14. Based on the orientation data of the vectorized vein network, draw a length-weighted equal-area rose diagram to realize spatial visualization and quantitative determination of the advantageous orientation of the vein.

[0035] S15. Calculate the directional anisotropy index, and quantitatively assess the spatial heterogeneity of the vectorized vein network based on the index to complete the prior analysis of the vein's geometric characteristics.

[0036] Step S2 specifically includes the following steps: S21. Based on the vectorized vein network and vein length distribution data from step S1, count the number of all veins within the target exploration line profile. Length of a single ore vein Calculate the average length of the ore vein and the total length of the vein within the profile .

[0037] Simultaneously, the actual sampling area of ​​the two-dimensional exploration line profile was measured and determined. .

[0038] S22, Calculate the two-dimensional mineralization intensity : ; S23. Calculate the fracture frequency and crack strength Complete the construction of a two-dimensional profile ore vein digital system: ; / .

[0039] The node types mentioned in step S3 include isolated node I and connected node. The connected node includes branch node Y and intersection node X. Among them, isolated node I is the endpoint of the vein line segment; branch node Y is a node where one vein branches off from another, forming a three-way intersection node; intersection node X is a node where two veins intersect each other, forming a four-way intersection node.

[0040] Branch types include CC-type branches (connecting node to connecting node, i.e., XX, XY, or YY), IC-type branches (connecting to isolated, i.e., IX or IY), and II-type branches (isolated to isolated). Among them, CC-type branches are branches between connecting nodes, used to reflect the development degree of the main vein network; IC-type branches are branches between isolated nodes and connecting nodes, used to characterize the secondary connection characteristics of the vein network; and II-type branches are branches between isolated nodes, used to represent the terminal branch characteristics of the vein network.

[0041] The number of connections per branch is the ratio of the total number of connections within the node-branch structure to the total number of branches, used to quantitatively characterize the connectivity index of the vein network in a single profile; the number of branches is the total number of all branches within the node-branch structure, used to reflect the development scale of the vein network; the node density is the total number of intersection nodes X and branch nodes Y within a unit profile area, used to characterize the degree of convergence development and the development status of spatial transmission channels in the vein network.

[0042] In step S4, the two-dimensional mineralization intensity of S2 is used as the dependent variable, and the average number of connections per branch in S3 is used as the dependent variable. Number of branches Node density Using the independent variable as an example, a linear correlation analysis method is then used to calculate the correlation coefficient between each topological parameter and the two-dimensional mineralization intensity. Finally, based on the correlation coefficient values, the quantitative correlation strength between the degree of mineralization enrichment and the topological characteristics of the vectorized vein network is determined.

[0043] Step S5 specifically includes the following steps: S51. Set the radius gradient and quantity gradient of the multi-scale random circle so that it covers the entire geological profile of the exploration line.

[0044] S52. Generate random circles according to the set axial gradient and quantitative gradient, and perform random circle subsampling on a single profile.

[0045] In the initial multi-scale random sampling, 50, 100, 150, 200, 250, 300, 350, 400, 450, 500, 550, and 600 sampling circles with radii of 50m, 100m, 150m, and 200m are randomly generated for each profile. The circles are classified according to their number and radius scale, and the average entropy of each sampling (e.g., 50 circles with a radius of 100) is calculated, which is the mean of the entropy of all circles.

[0046] S53. Extract the average number of connections per branch, number of branches, node density, and two-dimensional mineralization intensity data within each random circle.

[0047] S54. Set discrete intervals for the extracted average number of connections per branch, number of branches, node density and two-dimensional mineralization intensity data respectively, and complete the data discretization process.

[0048] S55. Count the number of times the data appears in each discrete interval, and calculate the probability distribution of each discrete interval.

[0049] Step S6 specifically includes the following steps: S61. Based on the probability distribution results in S5, use the information entropy formula to calculate the parameter distribution entropy value of a single random circle and the average entropy value of all random circles under the same sampling scale.

[0050] S62. Compare the average entropy values ​​of sampling schemes with different radii and numbers to evaluate the sampling reliability; and select the sampling radius and sampling number with the optimal parameter distribution entropy value as the optimal sampling parameters.

[0051] In this embodiment, firstly, the key attributes of the fracture data ( , , , Discretize it. Specifically, discretize it. Discretize the data, dividing it into intervals [0, 0.3, 0.7, 2], where each interval represents low, medium, high, and very high. Then, calculate the probability distribution for each interval using the information entropy formula. Similarly, set... The discrete interval [0, 5, 10, 20, ... Each interval represents low, medium, high, and very high; set the node density. The discrete interval [0, 2, 5, 10, ... Each interval is low, medium, high, and very high; the discrete intervals for mineralization intensity are set to [0, 2, 5, 10, ...]. Each interval is low, medium, high, and very high.

[0052] In step S8, the edge weights are defined by the inverse ratio of the topological parameters of adjacent nodes in the weighted directed graph to the two-dimensional mineralization intensity. At the same time, the inflow end of the profile is set as the source vertex and the outflow end as the target vertex. The Dijkstra algorithm is used to calculate the minimum weight path from the source vertex to the target vertex and determine it as the dominant migration channel of the ore-forming fluid.

[0053] This approach assumes that hydrothermal fluids in a fracture network prefer to migrate along the "shortest" and most connected channels. Each fracture segment on the profile is then treated as an edge, with the edge weight determined by the inverse ratio of the combined topological parameters of the two adjacent nodes to the two-dimensional mineralization intensity. The algorithm calculates the shortest path (i.e., the dominant fluid channel) from the inflow end to the outflow end of each profile. The shortest path network constructed using this method visually reveals the spatial distribution pattern of the dominant channels for ore-forming fluids. Higher edge frequency and betweenness centrality indicate a higher quartz vein order, serving as the connecting backbone and center of the fracture network, indicating the main fractures. These parameters jointly constrain the dominant extension direction of the ore body, solving the problem of determining the deep extension direction of the ore body.

[0054] Step S9 specifically includes the following steps: S91. Load the dominant orientation data of the ore vein obtained in step S1 and use it as the initial structural constraint condition for the extension direction of the ore body.

[0055] Simultaneously, the spatial region data with high connectivity and high mineralization potential obtained in step S7 is loaded to form the first predictive constraint layer.

[0056] Load the dominant migration channel data of ore-forming fluids obtained in step S8 to form a second predictive constraint layer.

[0057] S92. Perform spatial overlay analysis on the first and second prediction constraint layers to extract the spatial intersection area between them.

[0058] S93. Identify the spatial intersection area as a high-potential exploration target area and complete the dual-constraint delineation of the target area.

[0059] S94. Based on the dominant orientation of the ore vein, trend fitting is performed along the overall direction of the dominant migration channel of the ore-forming fluid to determine the optimal extension direction of the ore body.

[0060] S95 outputs the spatial range of high connectivity and high mineralization potential, as well as the optimal extension direction of the ore body, to complete the deep mineralization prediction.

[0061] Example In this embodiment, the ore veins in sections 11-95 of the Linglong gold ore vein network are analyzed as an example. After digitizing the ore vein system of this series of sections, the node-branch structure of each section's ore vein network is extracted, the proportion of various types of nodes and branches is statistically analyzed, and three types of topological parameters are defined: average number of connections per branch, number of branches, and node density. Based on the proportion of node type and branch type, a ternary phase diagram (e.g., ...) is constructed. Figure 2 As shown, taking vein No. 10 as an example, its branch classification and node classification ternary phase diagram can intuitively reflect the network structure characteristics. Through the ternary phase diagram and topology parameter statistics, the differences in vein network structure and spatial distribution characteristics of different profiles are systematically analyzed.

[0062] The vein branching types in this series of profiles show significant differentiation. In most profiles, the proportion of CC-type branches is between 40% and 55%, indicating a relatively well-developed main structure. Specifically, the proportions of CC-type branches in profiles 23, 47, 65, and 95 are 51.9%, 56.3%, 48.7%, and 54.8%, respectively. These profiles correspond to... , The values ​​are relatively high, and the network connectivity is strong, suggesting that it has a large potential for mineralization and enrichment. In contrast, the proportion of CC-type branches in sections 29, 41, 53 and 89 is less than 45%, with section 53 having the lowest proportion at only 38.6%. Meanwhile, the proportion of type II branches reaches 21.7%, indicating that this section is dominated by terminal veins, with weak overall connectivity and a loose vein network structure, which is not conducive to the transport and enrichment of ore-forming fluids. Regarding node types, I nodes (isolated nodes) generally accounted for the highest proportion in sections 11-95, all exceeding 50%, indicating the widespread existence of isolated endpoints in each section. This is consistent with the natural geological characteristics of quartz vein-type gold deposits. Y nodes (branch nodes) and X nodes (intersection nodes) constitute the main body of the vein network intersection structure, and show significant differences between different sections: Y nodes accounted for 46.4% in section 47, showing typical three-way intersection characteristics, suggesting frequent tectonic activity in this area, which is conducive to the diversion and convergence of ore-forming fluids; X nodes accounted for 13.8% in section 65, the highest among all sections, indicating that it has multi-directional intersecting topological characteristics and is a main fault intersection area, providing favorable space for the migration of ore-forming fluids and metal precipitation; while Y nodes accounted for only 24.9% in section 53, and X nodes also accounted for a low proportion (6.1%), indicating poor network connectivity, simple structure, and underdeveloped spatial transmission channels, further confirming the judgment that the mineralization potential of this section is weak.

[0063] Meanwhile, based on the digital processing results of profiles 11-95, a vein node-branch model diagram for each profile was constructed (e.g., Figure 3As shown, this is a node-branch model diagram of vein No. 10, which clearly presents the spatial relationship between veins, nodes, and branches. It intuitively presents the topological skeleton of the vein network in each profile, providing a visual model foundation and solid basic data for the subsequent accurate extraction of topological parameters, correlation analysis between mineralization intensity and topological parameters, multi-scale random sampling, and solution of the dominant migration channels of ore-forming fluids. It also provides a quantitative basis for the preliminary judgment of the mineralization potential of different profiles.

[0064] For typical sections in sections 11-95, a random circular sampling scheme consistent with that of exploration line section 41 was adopted, i.e., 600 samples with a radius of 200m. Full-coverage random circular sub-sampling was performed on each typical section, and the mineralization intensity P21 and three types of topological parameters Cb, Nb, and Nd within each random circle were calculated. The sampling results were consistent with those of section 41, and the mineralization intensity and the three types of topological parameters showed a strong spatial correlation (e.g., Figure 4 As shown, taking section 41 as an example, its mineralization intensity and topological parameter spatial distribution are highly consistent. Among them, the high mineralization areas of sections 47, 65, and 95 are mainly concentrated in the densely branched CC type and the intersection of Y and X nodes. Although the mineralization enrichment range does not form the typical "inverted C shape" of section 41, it generally shows the characteristics of "concentrated distribution and high overlap with the topological connectivity area", and the enrichment depth matches the regional tectonic background. On the other hand, section 53 has poor connectivity, and the overall mineralization intensity is relatively low, with no obvious concentrated enrichment area, further verifying the intrinsic relationship between branch type, topological parameters, and mineralization enrichment.

[0065] Correlation analysis was performed on the mineralization intensity and topological parameters of typical profiles 11-95. The results were consistent with those of profile 41, showing a strong linear correlation. Specifically, the correlation coefficients between mineralization intensity and Cb, Nb, and Nd in profile 47 were 0.682, 0.817, and 0.823, respectively; those in profile 65 were 0.721, 0.853, and 0.861, respectively; and those in profile 53 were 0.356, 0.421, and 0.438, respectively. This fully demonstrates that the stronger the connectivity of the vein network, the higher the degree of mineralization enrichment. Mineralization enrichment is closely related to the convergence and intersection of vein branches. Topologically, vein connectivity and vein enrichment show good consistency (e.g., Figure 5 As shown, this is the correlation analysis result of profile 41, which can intuitively reflect the correlation characteristics between mineralization intensity and topological parameters.

[0066] Based on this, shortest path networks were constructed for the node-branch models of typical profiles 11-95 using Dijkstra's algorithm (e.g. Figure 7As shown, taking vein No. 10 as an example, its shortest path network clearly presents the dominant migration channels of ore-forming fluids. Significant differences exist in the number of nodes, edges, and paths across different sections: section No. 47 has 58 nodes, 51 edges, and 27 paths; section No. 65 has 62 nodes, 55 edges, and 29 paths; and section No. 53 has only 42 nodes, 38 edges, and 18 paths. By calculating the edge frequency and betweenness centrality (e.g., ...) of each section... Figure 6 As shown in the figure, the analysis diagram of the identification parameters of the main trunk of vein No. 10 can be used to determine the quartz vein order. It was found that the regions with higher connection frequency and betweenness centrality correspond to a higher proportion of CC-type branches and better topological parameters, and all of them are mineralized enrichment areas. Among them, sections No. 47 and No. 65 identified 2-3 dominant migration paths, and their optimal fluid migration paths correspond to the main trunk fracture, which highly overlaps with the mineralized enrichment areas. Section No. 53 has only 1 dominant migration path due to poor network connectivity, and the mineralization intensity on the path is relatively low, which further confirms the controlling role of topological connectivity on the migration of ore-forming fluids.

[0067] In summary, through topological analysis, multi-scale random sampling, correlation analysis, and shortest path determination of ore-forming fluids in the vein network of sections 11-95, the correlation between the differences in vein network structure and mineralization enrichment in different sections has been clarified. Based on branch type, topological parameters, and dominant fluid transport channels, the mineralization potential of each section can be accurately determined, providing scientific and quantitative technical support for the delineation of high-potential exploration target areas and the determination of the optimal extension direction of ore bodies in the deep section 11-95 of the Linglong Gold Mine.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A metallogenic prognosis method based on quantitative analysis of network topology, characterized in that: Includes the following steps: S1. Obtain the geological profile map of the exploration line of the target mining area, and perform vectorization processing on the vein network in the profile to obtain the vectorized vein network. Then, use the complementary cumulative distribution function combined with the power law model, log-normal model and exponential distribution model to quantitatively characterize the vein length distribution characteristics. Quantify the vein's dominant orientation and spatial heterogeneity through the length-weighted equal-area rose diagram and the direction anisotropy index. S2. Construct a two-dimensional profile vein digital system based on two-dimensional mineralization intensity. Calculate the two-dimensional mineralization intensity based on the vein length distribution characteristics and profile area characterized in step S1, and establish a two-dimensional profile vein digital system. S3. Based on the two-dimensional profile vein digitization system of step S2, extract the node-branch structure, classify and statistically analyze the node type and branch type, and define three types of topological parameters: average number of connections per branch, number of branches, and node density. S4. Based on the topological parameters obtained in step S3 and the two-dimensional mineralization intensity obtained in step S2, perform correlation analysis to determine the quantitative correlation between the degree of mineralization enrichment and the topological characteristics of the vectorized vein network. S5. Based on the topological parameters of step S3 and the two-dimensional mineralization intensity of step S2, a multi-scale random circular subsampling method is used to fully cover the geological profile of the exploration line. The topological parameters and the two-dimensional mineralization intensity are discretized and the probability distribution of each discrete interval is calculated. S6. Based on the probability distribution results of step S5, the information entropy formula is used to calculate the parameter distribution entropy value under different sampling scales and sampling numbers, quantitatively evaluate the reliability of the sampling scheme and determine the optimal sampling parameters. S7. Based on the quantitative correlation in step S4 and the optimal sampling parameters in step S6, a two-dimensional topology-mineralization coupling model is constructed. In the two-dimensional topology-mineralization coupling model, spatial regions that meet the topology parameter thresholds are selected and identified as spatial regions with high connectivity and high mineralization potential. S8. Based on the topology parameters of step S3, the vectorized vein network of step S1 is transformed into a weighted directed graph. The Dijkstra algorithm is used to calculate the shortest path of the ore-forming fluid from the inflow end to the outflow end within the profile, and the dominant migration channel of the ore-forming fluid is determined. S9. Considering the vein's dominant orientation data from step S1, and based on the highly connected and mineralized potential areas identified in step S7 and the dominant migration channels of ore-forming fluids determined in step S8, perform spatial overlay analysis to delineate high-potential exploration target areas under dual constraints and determine the optimal extension direction of the ore body.

2. The metallogenic prognosis method based on quantitative analysis of network topology according to claim 1, characterized in that: Step S1 specifically includes the following steps: S11. Collect geological profile maps of the exploration line in the target mining area and extract the spatial distribution, morphology and location data of the original veins on the profile. S12. Vectorize the vein network in the geological profile of the exploration line, converting the graphical veins into digital vector data to form a vectorized vein network. S13. Quantitative characterization of vein length distribution characteristics; S131. Extract the length data of a single vein from the vectorized vein network, and after removing outlier data, use the complementary cumulative distribution function to perform statistical calculations on the vein length data to characterize the vein length distribution characteristics and obtain a vein length sample dataset. S132. Using the power law model as the core, simultaneously establish complementary cumulative distribution function expressions for the log-normal model and the exponential distribution model as the fitting benchmark function; S133. Perform optimal fitting on the power law model, log-normal model and exponential distribution model respectively to obtain the fitting function and fitting curve of the three types of models; S134. Based on the degree of fit between the fitting curves of the power law model, log-normal model and exponential distribution model and the measured complementary cumulative distribution function, the model with the highest fitting degree is selected as the optimal distribution model for vein length. S135. Based on the optimal distribution model of vein length determined in step S134, complete the quantitative characterization of vein length distribution. S14. Based on the orientation data of the vectorized vein network, draw a length-weighted equal-area rose diagram to realize the spatial visualization and quantitative determination of the dominant orientation of the vein. S15. Calculate the directional anisotropy index, and quantitatively assess the spatial heterogeneity of the vectorized vein network based on the index to complete the prior analysis of the vein's geometric characteristics.

3. The mineralization prediction method based on quantitative analysis of network topology according to claim 1, characterized in that: Step S2 specifically includes the following steps: S21. Based on the vectorized vein network and vein length distribution data from step S1, count the number of all veins within the target exploration line profile. Length of a single ore vein Calculate the average length of the ore vein and the total length of the vein within the profile ; Simultaneously, the actual sampling area of ​​the two-dimensional exploration line profile was measured and determined. ; S22, Calculate the two-dimensional mineralization intensity : ； S23. Calculate the fracture frequency and crack strength Complete the construction of a two-dimensional profile ore vein digital system: ； / 。 4. The mineralization prediction method based on quantitative analysis of network topology according to claim 1, characterized in that: The node types mentioned in step S3 include isolated nodes I and connected nodes. Connected nodes include branch nodes Y and intersection nodes X. Among them, isolated node I is the endpoint of a vein segment; branch node Y is a node where one vein branches off from another, forming a three-way intersection; intersection node X is a node where two veins intersect each other, forming a four-way intersection. Branch types include CC type branches, IC type branches, and II type branches. CC type branches are branches between connecting nodes, used to reflect the development degree of the main vein network; IC type branches are branches between isolated nodes and connecting nodes, used to characterize the secondary connection characteristics of the vein network; and II type branches are branches between isolated nodes, used to represent the terminal branch characteristics of the vein network. The number of connections per branch is the ratio of the total number of connections within the node-branch structure to the total number of branches, used to quantitatively characterize the connectivity index of the vein network in a single profile; the number of branches is the total number of all branches within the node-branch structure, used to reflect the development scale of the vein network; the node density is the total number of intersection nodes X and branch nodes Y within a unit profile area, used to characterize the degree of convergence development and the development status of spatial transmission channels in the vein network.

5. The mineralization prediction method based on quantitative analysis of network topology according to claim 1, characterized in that: In step S4, the two-dimensional mineralization intensity of S2 is used as the dependent variable, and the average number of connections per branch, the number of branches, and the node density of S3 are used as independent variables. Then, the linear correlation analysis method is used to calculate the correlation coefficient between each topological parameter and the two-dimensional mineralization intensity. Finally, based on the correlation coefficient values, the quantitative correlation strength between the mineralization enrichment degree and the topological characteristics of the vectorized vein network is determined.

6. The mineralization prediction method based on quantitative analysis of network topology according to claim 1, characterized in that: Step S5 specifically includes the following steps: S51. Set the radius gradient and quantity gradient of the multi-scale random circle so that it covers the entire geological profile of the exploration line. S52. Generate random circles according to the set radial gradient and quantitative gradient, and perform random circle subsampling on a single profile; S53. Extract the average number of connections per branch, number of branches, node density, and two-dimensional mineralization intensity data within each random circle; S54. Set discrete intervals for the extracted average number of connections per branch, number of branches, node density and two-dimensional mineralization intensity data respectively, and complete the data discretization process; S55. Count the number of times the data appears in each discrete interval, and calculate the probability distribution of each discrete interval.

7. The mineralization prediction method based on quantitative analysis of network topology according to claim 1, characterized in that: Step S6 specifically includes the following steps: S61. Based on the probability distribution results of S5, use the information entropy formula to calculate the parameter distribution entropy value of a single random circle and the average entropy value of all random circles under the same sampling scale. S62. Compare the average entropy values ​​of sampling schemes with different radii and numbers to evaluate the sampling reliability; and select the sampling radius and sampling number with the optimal parameter distribution entropy value as the optimal sampling parameters.

8. The mineralization prediction method based on quantitative analysis of network topology according to claim 1, characterized in that: In step S8, the edge weights are defined by the inverse ratio of the topological parameters of adjacent nodes in the weighted directed graph to the two-dimensional mineralization intensity. At the same time, the inflow end of the profile is set as the source vertex and the outflow end as the target vertex. The Dijkstra algorithm is used to calculate the minimum weight path from the source vertex to the target vertex and determine it as the dominant migration channel of the ore-forming fluid.

9. The mineralization prediction method based on quantitative analysis of network topology according to claim 1, characterized in that: Step S9 specifically includes the following steps: S91. Load the dominant orientation data of the ore vein obtained in step S1 and use it as the initial structural constraint condition for the extension direction of the ore body. Simultaneously, the spatial region data with high connectivity and high mineralization potential obtained in step S7 is loaded to form the first prediction constraint layer. Load the dominant migration channel data of ore-forming fluids obtained in step S8 to form a second predictive constraint layer; S92. Perform spatial overlay analysis on the first prediction constraint layer and the second prediction constraint layer, and extract the spatial intersection area between the two. S93. Identify the spatial intersection area as a high-potential exploration target area and complete the dual-constraint delineation of the target area; S94. Based on the dominant orientation of the ore vein, trend fitting is performed along the overall direction of the dominant migration channel of the ore-forming fluid to determine the optimal extension direction of the ore body. S95 outputs the spatial range of high connectivity and high mineralization potential, as well as the optimal extension direction of the ore body, to complete the deep mineralization prediction.

Images