Mineral prospecting target area resource decision-making method and system based on minimizing the impact on environmentally sensitive areas

Abstract

The present invention relates to the technical field of geological prospecting, and specifically relates to a resource decision-making method and system for prospecting target areas based on minimizing the impact on environmentally sensitive areas. The method includes: obtaining geological data and environmental data of each prospecting target area in the target exploration area and performing preprocessing; constructing a spatio-temporal coupled dynamic quantification model of environmental sensitivity based on the preprocessed environmental data, and extracting the environmental sensitivity quantification results of each prospecting target area; constructing a geological potential quantification model based on the preprocessed geological data set, and extracting the geological potential quantification results of each prospecting target area; constructing a difference matrix of prospecting target areas based on the environmental sensitivity and geological potential results, and extracting typical prospecting target areas through fuzzy support vector machine classification and density peak clustering; constructing a three-dimensional decision space, and using a multi-objective optimization algorithm to solve the decision-making results of prospecting target areas. The purpose is to improve the efficiency and accuracy of resource decision-making for large-scale prospecting target areas.

Description

Technical Field

[0001] The present invention relates to the technical field of geological prospecting, and specifically relates to a resource decision-making method and system for prospecting target areas based on minimizing the impact on environmentally sensitive areas. Background Art

[0002] In the field of geological and mineral exploration, the accurate delineation of prospecting target areas is the core link of mineral resource exploration. The delineation of prospecting target areas depends on the in-depth integration and comprehensive analysis of multi-disciplinary data such as geology, geophysics, and geochemistry. By carefully sorting out the regional geological background and metallogenic conditions and deeply exploring the distribution law of minerals, prospecting target areas can be delineated and become the core focus areas of mineral exploration work. Although the existing technology can achieve the preliminary delineation of prospecting target areas, there is a lack of an intelligent screening method for large-scale prospecting target area sets.

[0003] Currently, traditional prospecting target area evaluation methods usually rely on expert experience judgment. On the one hand, the static weight allocation dominated by subjective experience leads to the evaluation results being limited by the cognitive biases of experts. Especially in ecologically fragile areas, the cumulative effects of environmental impacts and the natural recovery process cannot be accurately characterized, resulting in biases in the evaluation decisions for prospecting target areas. On the other hand, for large-scale target area sets, the traditional independent target area evaluation mode needs to repeatedly execute multi-disciplinary data interpretation and potential calculation, resulting in the computational complexity increasing exponentially with the number of target areas, thereby causing redundant operations and decision-making delays. These problems limit the efficiency and result accuracy of prospecting target area decision-making, and further restrict the efficiency and scientific nature of mineral exploration in highly sensitive areas. Summary of the Invention

[0004] In order to improve the efficiency and accuracy of resource decision-making for large-scale prospecting target areas, the present invention provides a resource decision-making method and system for prospecting target areas based on minimizing the impact on environmentally sensitive areas. The specific technical solutions adopted are as follows:

[0005] The technical solution of the first aspect of the present invention provides a resource decision-making method for prospecting target areas based on minimizing the impact on environmentally sensitive areas. The method includes:

[0006] Obtain the geological data and environmental data of each prospecting target area in the target exploration area and perform preprocessing;

[0007] Based on the preprocessed environmental data, construct a spatio-temporal coupled dynamic quantification model of environmental sensitivity, and extract the quantification results of environmental sensitivity of each prospecting target area;

[0008] Based on the preprocessed geological data set, construct a geological potential quantification model, and extract the quantification results of geological potential of each prospecting target area;

[0009] Construct a difference matrix of prospecting target areas based on the results of environmental sensitivity and geological potential, and extract typical prospecting target areas through fuzzy support vector machine classification and density peak clustering;

[0010] Construct a three-dimensional decision space and use a multi-objective optimization algorithm to solve the decision results of prospecting target areas.

[0011] Furthermore, obtain the geological data and environmental data of each prospecting target area in the target exploration area and perform preprocessing, including:

[0012] Obtain the vegetation coverage, slope, distance from the ecological protection area, and permafrost stability index of each prospecting target area, construct an environmental data set, and perform standardization preprocessing on the environmental data set;

[0013] Obtain the conjugate fracture density, Bouguer gravity anomaly value, and vertical concentration gradient of primary halos of each prospecting target area, construct a geological data set, and perform normalization preprocessing on the geological data set.

[0014] Furthermore, based on the preprocessed environmental data, construct a spatio-temporal coupling dynamic quantification model of environmental sensitivity, and extract the environmental sensitivity quantification results of each prospecting target area, including:

[0015] Construct a triangular fuzzy judgment matrix, conduct a fuzzy evaluation of the relative importance of environmental factors, and use the fuzzy analytic hierarchy process to extract the weights of environmental factors;

[0016] Couple the weights of environmental factors and the preprocessed environmental data in space and time to extract the historical cumulative effect of environmental impact;

[0017] Based on the historical cumulative effect of environmental impact and the preset time decay coefficient, extract the environmental sensitivity quantification results of each prospecting target area.

[0018] Furthermore, the expression of the environmental sensitivity quantification result is:

[0019]

[0020] In the formula, represents the environmental sensitivity quantification result of the th prospecting target area; represents the starting time; represents the current time; represents the integral time variable; represents the preset time decay coefficient, which is used to reflect the ecological restoration rate; represents the number of environmental factors; represents the th weight of environmental factors; represents the th environmental factor of the Standardized environmental factors.

[0021] Furthermore, based on the preprocessed geological data set, a geological potential quantification model is constructed to extract the geological potential quantification results of each prospecting target area, including:

[0022] Perform dimensionless processing on the preprocessed conjugate fracture density, Bouguer gravity anomaly value, and vertical concentration gradient of primary halo elements;

[0023] Based on the dimensionless conjugate fracture density, Bouguer gravity anomaly value, and vertical concentration gradient of primary halo elements after dimensionless processing, perform multi-source data fusion and construct a geological potential quantification model to extract the geological potential quantification results of each prospecting target area.

[0024] Furthermore, based on the environmental sensitivity and geological potential results, construct a prospecting target area difference matrix, and extract typical prospecting target areas through fuzzy support vector machine classification and density peak clustering, including:

[0025] Construct a prospecting target area difference matrix based on environmental sensitivity, geological potential results, and spatial distance differences;

[0026] Using the prospecting target area difference matrix as input, use fuzzy support vector machines to preliminarily classify the prospecting target areas and assign fuzzy membership degrees;

[0027] Use the truncated kernel function to calculate the local density of the prospecting target area, and determine the clustering center in combination with the minimum distance;

[0028] Based on the clustering center, use the density peak clustering algorithm to extract typical prospecting target areas.

[0029] Furthermore, construct a three-dimensional decision space and use a multi-objective optimization algorithm to solve the prospecting target area decision result, including:

[0030] Nonlinearly couple the quantification results of the environmental sensitivity, geological potential, and the magnitude of their vertical gradients of typical prospecting target areas to construct a comprehensive decision-making index;

[0031] Based on the comprehensive decision-making index and preset constraint conditions, construct a two-objective integer programming model for maximizing resource development value and minimizing environmental risk;

[0032] Based on the genetic algorithm, solve the Pareto optimal solution set of the two-objective integer programming model;

[0033] Optimize the Pareto optimal solution set using dynamic diversity preservation and elite retention to extract the final prospecting target area decision result.

[0034] Furthermore, the expression of the comprehensive decision-making index is:

[0035]

[0036] In the formula, represents the comprehensive decision-making index of the th typical ore prospecting target area; represents the quantification result of the geological potential of the th typical ore prospecting target area; represents the quantification result of the environmental sensitivity of the th typical ore prospecting target area; represents the geological potential strengthening coefficient; represents the environmental sensitivity suppression coefficient; represents the gradient smoothing factor; represents the smoothing constant; represents the vertical gradient modulus of the geological potential of the th typical ore prospecting target area.

[0037] Furthermore, the expression of the bi-objective integer programming model is:

[0038]

[0039] In the formula, represents the function of the mining economic value changing with depth, , is the first constant, that is, the benchmark economic value constant, is the second constant, that is, the attenuation coefficient of the mining economic value changing with depth; represents the mining depth; represents the function of the ecological impact decaying with depth, , is the third constant, that is, the ecological impact decay rate constant; represents the total budget constraint; represents the geological potential threshold; represents the environmental sensitivity threshold; represents the 0-1 decision variable; represents the cost of the th typical ore prospecting target area; represents the number of typical ore prospecting target areas.

[0040] The technical solution of the second aspect of the present invention provides an ore prospecting target area resource decision-making system based on minimizing the impact on environmentally sensitive areas. Using the ore prospecting target area resource decision-making method described in the technical solution of the first aspect of the present invention, the system includes:

[0041] A data acquisition module configured to acquire geological data and environmental data of each ore prospecting target area in the target exploration area and perform preprocessing;

[0042] An environmental sensitivity dynamic quantification module, configured to construct a spatio-temporal coupled environmental sensitivity dynamic quantification model based on the preprocessed environmental data, and extract the environmental sensitivity quantification results of each prospecting target area;

[0043] A geological potential quantification module, configured to construct a geological potential quantification model based on the preprocessed geological data set, and extract the geological potential quantification results of each prospecting target area;

[0044] A typical target area extraction module, configured to construct a prospecting target area difference matrix based on the environmental sensitivity and geological potential results, and extract typical target areas through fuzzy support vector machine classification and density peak clustering;

[0045] A multi-objective optimization solution module, configured to construct a three-dimensional decision space and use a multi-objective optimization algorithm to solve the prospecting target area decision result.

[0046] The present invention has the following beneficial effects:

[0047] The prospecting target area resource decision-making method based on minimizing the impact of environmentally sensitive areas provided by the present invention realizes the accurate assessment of the environmental risk and resource potential of prospecting target areas by establishing a spatio-temporal coupled environmental sensitivity dynamic quantification model and a geological potential quantification model; furthermore, the original target area set is compressed into a set of typical target areas by using fuzzy support vector machines and density peak clustering, effectively reducing redundant data and retaining key features, and improving the decision-making efficiency; finally, by constructing a three-dimensional decision space and using a multi-objective optimization algorithm, comprehensively considering factors such as resource development value and environmental risk, balancing the relationship between resource development and environmental protection, ensuring the effective development of resources while minimizing the impact on environmentally sensitive areas to the greatest extent, thereby providing a more scientific solution for the resource decision-making of prospecting target areas; especially in the face of large-scale target area sets and highly sensitive areas, this method significantly improves the accuracy and decision-making efficiency of prospecting target area resource decision-making, and helps to improve the efficiency and scientificity of mineral exploration in highly sensitive areas. Description of the Drawings

[0048] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0049] Figure 1 It is the method flow chart of the prospecting target area resource decision-making method based on minimizing the impact of environmentally sensitive areas provided by an embodiment of the present invention;

[0050] Figure 2Schematic diagram of the resource decision-making system for prospecting target areas based on minimizing the impact on environmentally sensitive areas provided by an embodiment of the present invention. Detailed implementation manners

[0051] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details the specific implementation manners, structures, features, and effects of a method and system for resource decision-making of prospecting target areas based on minimizing the impact on environmentally sensitive areas proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0053] The following specifically describes the specific solutions of a method and system for resource decision-making of prospecting target areas based on minimizing the impact on environmentally sensitive areas provided by the present invention in conjunction with the accompanying drawings.

[0054] A prospecting target area is an area with relatively large prospecting potential determined through comprehensive consideration of various information such as geology, geophysics, and geochemistry in geological prospecting work. It is usually delineated after in-depth research on the regional geological background, metallogenic conditions, and mineral distribution laws, and is the key target area for mineral exploration work. Taking a specific mountain range area as an example, the present invention analyzed the strata, structures, rock types, and previous mineral discovery situations, determined several prospecting target areas, and formed a set of prospecting target areas; based on this, the present invention proposes a method and system for resource decision-making of prospecting target areas based on minimizing the impact on environmentally sensitive areas.

[0055] Please refer to Figure 1 , which shows the flowchart of the method for resource decision-making of prospecting target areas based on minimizing the impact on environmentally sensitive areas provided by an embodiment of the present invention. The method includes:

[0056] Step S100: Obtain the geological data and environmental data of each prospecting target area in the target exploration area, and perform preprocessing;

[0057] Step S100 specifically includes:

[0058] Step S110: Obtain the vegetation coverage, slope, distance from the ecological protection area, and frozen soil stability index of each prospecting target area, construct an environmental data set, and perform standardized preprocessing on the environmental data set; specifically, in this embodiment, satellite remote sensing technology is adopted to select satellite images with higher resolution and appropriate time coverage, such as Landsat 8 OLI or Sentinel - 2 images. The NDVI values of each prospecting target area are calculated using the normalized difference vegetation index. A certain number of ground quadrats are set in each prospecting target area, and by field measuring the ratio of the vegetation coverage area to the quadrat area, the NDVI values calculated by remote sensing are verified and calibrated to improve the accuracy of the vegetation coverage data; for the distance from the ecological protection area, in this embodiment, spatial analysis is performed on the central coordinates of the prospecting target area and the boundary data of the ecological protection area, and the Euclidean distance from the center point of the target area to the nearest boundary of the ecological protection area is calculated to obtain the distance from each prospecting target area to the ecological protection area; the frozen soil stability index is a quantitative index specifically used to evaluate the stability of the frozen soil layer in the prospecting target area. This index mainly considers multiple factors affecting the stability of frozen soil, such as ground temperature, soil moisture content, vegetation coverage, etc., to reflect the stability degree of frozen soil under natural conditions or external disturbances. Since the melting or instability of frozen soil can directly affect the stability of the geological structure, thus having a significant impact on the safety and effectiveness of prospecting activities, this index is included in the environmental data of the prospecting target area in this embodiment.

[0059] Step S120: Obtain the conjugate fracture density, Bouguer gravity anomaly value, and vertical concentration gradient of primary halo elements of each prospecting target area, construct a geological data set, and perform normalization preprocessing on the geological data set; among them, the conjugate fracture density refers to the density of conjugate fractures in a specific direction within the prospecting target area, that is, the number of conjugate fracture intersections per unit area, and its density reflects the complexity of the geological structure and the rock fragmentation situation in this prospecting target area; the Bouguer gravity anomaly value is the difference value calculated based on Bouguer gravity measurement compared with the standard gravity field. Through the Bouguer gravity anomaly value, the inhomogeneity of the underground geological structure and material composition can be inferred; the vertical concentration gradient of primary halo elements refers to the change rate of the concentration of primary halo elements closely related to mineralization in geological samples within the prospecting target area in the vertical direction. This parameter is used to determine the location and extension direction of the ore body, and further evaluate the potential of the ore deposit.

[0060] Specifically, in this embodiment, it is necessary to obtain the geological data set and environmental data set of each prospecting target area in the prospecting target area set; among them, the prospecting target area set can be expressed as: , where represents the th prospecting target area; represents the total number of prospecting target areas; represents the th central coordinate of the prospecting target area; Indicates the geological data of the th prospecting target area; Indicates the environmental data of the th prospecting target area; Then, perform normalization preprocessing on the geological data and standardization preprocessing on the environmental data. The geological data set after normalization preprocessing can be expressed as: , represents the quantity of geological data; The environmental data set after standardization can be expressed as: , represents the quantity of environmental data; The data set of the prospecting target area can be expressed as: ; In this embodiment, by performing standardization processing on the environmental data and normalization processing on the geological data, the dimensional difference between different data indicators can be eliminated, making different types of data comparable. It helps to more accurately reflect the influence of various factors on the prospecting target area when constructing the dynamic quantification model of environmental sensitivity and the geological potential quantification model subsequently, and avoid model deviation caused by different data dimensions.

[0061] Step S200: Construct a spatio-temporal coupled dynamic quantification model of environmental sensitivity based on the preprocessed environmental data, and extract the environmental sensitivity quantification results of each prospecting target area;

[0062] Step S200 specifically includes:

[0063] Step S210: Construct a triangular fuzzy judgment matrix, conduct fuzzy evaluation on the relative importance of environmental factors, and extract the environmental factor weights using the fuzzy analytic hierarchy process; Specifically, the weights of environmental factors in this embodiment are determined by the fuzzy analytic hierarchy process to reflect the importance degree of each environmental factor in the overall environmental sensitivity evaluation; In this embodiment, 5 - 10 experts in the field of geological ecology are selected, and the pairwise importance of environmental factors (vegetation coverage, slope, distance from the protected area, permafrost stability) is evaluated using triangular fuzzy numbers to form a triangular fuzzy judgment matrix. For example: Vegetation coverage vs permafrost stability: The field experts believe that the former is "slightly more important", which can be expressed as a triangular fuzzy number as ; Permafrost stability vs slope: The field experts believe that the former is "significantly more important", which can be expressed as a triangular fuzzy number as ; The triangular fuzzy judgment matrix can be expressed as:

[0064] ,

[0065] In the formula, represents the triangular fuzzy judgment matrix, with a dimension of , used to reflect the relative importance relationship between environmental factors; represents the triangular fuzzy number, located in the matrix in the and the column, reflecting the fuzzy judgment of experts on the th environmental factor relative to the th environmental factor; represents the lower limit of the importance judgment of experts on the th environmental factor relative to the th environmental factor; represents the most likely value of the importance judgment of experts on the th environmental factor relative to the th environmental factor; represents the upper limit of the importance judgment of experts on the th environmental factor relative to the th environmental factor;

[0066] Among them, the calculation formula for the weight of environmental factors is:

[0067]

[0068] In the formula, represents the fourth root of the triangular fuzzy number ; represents the summation variable, which is used to traverse each row of the triangular fuzzy judgment matrix. In this embodiment, by fusing the triangular fuzzy judgments of multiple experts and defuzzification calculation, the subjectivity and uncertainty of expert judgments in the traditional analytic hierarchy process are solved; on the other hand, geometric mean aggregation and centroid method defuzzification are adopted to make the weight distribution retain the consensus of the expert group and avoid the influence of the preferences of a single expert.

[0069] Step S220: Perform spatio-temporal coupling on the environmental factor weights and the preprocessed environmental data, and extract the historical cumulative effect of environmental impact; specifically, in this embodiment, based on the geographical location information of the prospecting target area obtained in step S100, the preprocessed environmental data is spatially located. According to the time information of data collection, combined with the starting time sort the environmental data in time series, generate time series raster data, to ensure the spatio-temporal consistency of the environmental data. For each prospecting target area at different time points , the environmental factor weights are weighted with the standardized environmental factors to obtain the contribution value of each environmental factor to the environmental impact at this time point. Then, the contribution values of different environmental factors at the same time point are accumulated to obtain the comprehensive environmental impact value at this time point. As time goes by, these comprehensive environmental impact values are accumulated in chronological order to form a historical cumulative effect data sequence of environmental impacts; finally, the spatio-temporal raster data realizes the visualization of the environmental state of the target area, and the spatial resolution can reach 10m; in this embodiment, by constructing a spatio-temporal raster data model, the refined spatio-temporal alignment of environmental data is realized. The dynamic sliding window accumulation mechanism can capture short-term ecological disturbances (such as seasonal vegetation changes) and long-term trends (such as permafrost degradation), and the prediction error of the cumulative effect is further reduced compared with the static model.

[0070] Step S230: Based on the historical cumulative effect of environmental impact and a preset time decay coefficient, extract the quantification results of the environmental sensitivity of each prospecting target area, which can be expressed as:

[0071]

[0072] In the formula, represents the quantification result of the environmental sensitivity of the th prospecting target area; represents the starting time; represents the current time, that is, calculate the cumulative effect of environmental impact from the starting time to the current time; represents the integration time variable, which is used to calculate the integral of the cumulative effect of environmental impact over time; represents the preset time decay coefficient, which is used to reflect the ecological restoration rate, that is, the natural restoration process of environmental impact over time; represents the number of environmental factors; represents the th environmental factor weight; represents the th th standardized environmental factor of the

[0073] In summary, in this embodiment, the fuzzy analytic hierarchy process is used to determine the weights of environmental factors, making full use of expert experience and comprehensively considering the relative importance of each environmental factor in the environmental sensitivity assessment, so that the weight distribution is more scientific and reasonable. The spatio-temporal coupling calculation method in this embodiment organically combines the environmental factor weights and environmental data in the time and space dimensions, comprehensively reflecting the historical cumulative effect of environmental impacts, and can accurately capture the dynamic impact of environmental changes on the prospecting target area. At the same time, the environmental sensitivity quantification result is calculated based on the preset time decay coefficient, considering the natural process of ecological restoration, which is more in line with the actual situation and provides a quantitative basis for evaluating the long-term impact of prospecting activities on the environment. Finally, these quantitative results provide important environmental risk assessment indicators for subsequent prospecting target area screening and resource decision-making, helping to minimize the impact on environmentally sensitive areas while ensuring resource development, and thus achieving the coordinated development of resource development and environmental protection.

[0074] Step S300: Construct a geological potential quantification model based on the preprocessed geological data set, and extract the geological potential quantification results of each prospecting target area;

[0075] Step S300 specifically includes:

[0076] Step S310: Perform dimensionless processing on the preprocessed conjugate fracture density, Bouguer gravity anomaly value, and vertical concentration gradient of primary halo elements. Specifically, in this example, based on the obtained and preprocessed conjugate fracture density data, the difference in the influence of conjugate fractures on mineralization under different geological backgrounds is further analyzed, that is, by statistically analyzing the conjugate fracture density and mineralization situation of known ore deposits in the study area, the relationship curve between the conjugate fracture density and the mineralization probability is determined, and the dimensionless fracture density can be expressed as:

[0077]

[0078] In the formula, represents the dimensionless conjugate fracture density, reflecting the multiple relationship of the fracture density relative to the mineralization threshold; represents the fracture density mineralization threshold; represents the original conjugate fracture density;

[0079] For the Bouguer gravity anomaly data normalized in step S100, this embodiment uses the wavelet multi-scale decomposition method to separate the regional field and the local field, and extracts the high-frequency component characterizing the buried rock mass; calculate the gradient modulus and perform range standardization, which can be expressed as:

[0080]

[0081] In the formula, represents the normalized gravity gradient modulus, dimensionless; represents the original gravity gradient modulus; represents the minimum value of the gravity gradient modulus within the study area; represents the maximum value of the gravity gradient modulus within the study area;

[0082] For the vertical concentration gradient of primary halo elements, normalization based on the elemental geochemical background value can be expressed as:

[0083]

[0084] In the formula, represents the normalized vertical concentration gradient of elements, dimensionless; represents the original vertical concentration gradient of elements; represents the elemental geochemical background value; represents the conjugate fracture density, used to reflect the ore - controlling ability of structures; represents the Bouguer gravity anomaly, used to indicate concealed rock masses; represents the concentration of the target element; represents the vertical gradient of the target element; represents the first preset weight, represents the second preset weight, represents the third preset weight; represents the maximum neutral gradient;

[0085] Step S320: According to the conjugate fracture density, Bouguer gravity anomaly value, and vertical concentration gradient of primary halo elements after dimensionless processing, conduct multi - source data fusion and construct a geological potential quantification model, and extract the geological potential quantification results of each prospecting target area;

[0086] Specifically, in this embodiment, principal component analysis is performed on the dimensionless conjugate fracture density, Bouguer gravity anomaly value, and vertical concentration gradient of primary halo elements, and the principal components with a cumulative contribution rate greater than 85% are extracted to eliminate multi - source data redundancy;

[0087] The non - linear quantification model of the address potential model can be expressed as:

[0088]

[0089] In the formula, represents the geological potential value of the th prospecting target area; represents the weight of the th principal component; represents the score of the th principal component; Indicates the number of principal components; in this embodiment, in this step, dimensionality reduction is performed through principal component analysis, and the dimension of multi-source geological data is compressed to 2-3 principal components, the model calculation efficiency is increased by 60%, and at the same time, more than 95% of the original information is retained. These quantitative results provide a basis for subsequent screening of ore prospecting target areas and determination of resource development priorities, help improve the accuracy of ore prospecting target area decision-making, reduce unnecessary exploration costs and environmental damage while ensuring resource development.

[0090] Step S400: Construct a difference matrix of ore prospecting target areas based on the results of environmental sensitivity and geological potential, and extract typical ore prospecting target areas through fuzzy support vector machine classification and density peak clustering;

[0091] Step S400 specifically includes:

[0092] Step S410: Construct a difference matrix of ore prospecting target areas based on the results of environmental sensitivity, geological potential, and spatial distance differences; specifically, in this embodiment, in order to measure the comprehensive differences between ore prospecting target areas, considering the quantitative results of environmental sensitivity, geological potential values, and spatial distance differences, the comprehensive difference can be expressed as:

[0093]

[0094] In the formula, represents the th ore prospecting target area and the th ore prospecting target area; represents the weight of environmental sensitivity difference, represents the weight of geological potential difference; represents the weight of spatial distance difference; in this embodiment, by comprehensively considering the environmental sensitivity, geological potential, and spatial distance differences to construct a difference matrix of ore prospecting target areas, it can comprehensively and accurately measure the comprehensive differences between ore prospecting target areas, avoid the limitations of single-factor analysis, fully reflect the characteristic differences of ore prospecting target areas, and help to more accurately identify and screen groups of ore prospecting target areas with similar characteristics; the difference matrix can be expressed as .

[0095] Step S420: Using the prospecting target area difference matrix as input, the prospecting target areas are preliminarily classified by a fuzzy support vector machine and given fuzzy membership degrees. Specifically, in this embodiment, a fuzzy support vector machine is used to preliminarily classify the prospecting target areas. The prospecting target area difference matrix is used as input to train a fuzzy support vector machine (FSVM) model. The trained FSVM model is used to preliminarily classify all prospecting target areas, and each prospecting target area is assigned to the corresponding category. According to the output results of the model, fuzzy membership degrees are assigned to each prospecting target area. The fuzzy membership degree represents the degree to which a prospecting target area belongs to a certain category, and its value range is between [0, 1], which more accurately reflects the fuzzy boundaries and uncertainties between prospecting target areas. Using a fuzzy support vector machine to preliminarily classify the prospecting target areas and assign fuzzy membership degrees can effectively handle the non-linearity and fuzziness of the prospecting target area data. By standardizing the data and optimizing the model parameters, the accuracy and stability of the classification are improved. The introduction of the fuzzy membership degree more precisely describes the relationship between the prospecting target areas and various categories, provides richer information for subsequent clustering analysis, and helps to discover potential prospecting target area categories and characteristics.

[0096] Step S430: Calculate the local density of the prospecting target areas using a truncated kernel function and determine the clustering centers in combination with the minimum distance.

[0097] Among them, the local density is calculated using a truncated kernel function, and its expression is:

[0098]

[0099] In the formula, represents the local density of the th prospecting target area; represents the comprehensive difference between the th prospecting target area and the th prospecting target area; represents the truncation distance, which is determined by the data distribution; represents the indicator function;

[0100] The minimum distance of samples with higher local density can be expressed as:

[0101]

[0102] Then draw the decision diagram of the local density and the minimum distance . In the decision diagram, select the points where both and are relatively large as the initial clustering centers. At the same time, in combination with the actual geological and environmental conditions, the initial clustering centers are manually screened and adjusted to ensure that the clustering centers can represent the characteristics of different types of prospecting target areas. Finally, according to and Determine the clustering centers, and assign all prospecting target areas to the category where the nearest clustering center is located to form a set of typical prospecting target areas. In this embodiment, by calculating the local density and combining the minimum distance to determine the clustering centers, the distribution characteristics of the data can be fully utilized to accurately identify the dense areas and outliers in the data. Reasonably select the clustering centers to make the clustering results more in line with the actual distribution of prospecting target areas, avoiding the subjectivity and randomness in the selection of clustering centers, and providing a reliable basis for subsequent extraction of typical prospecting target areas.

[0103] Step S440: Based on the clustering centers, use the density peak clustering algorithm to extract typical prospecting target areas. After the preliminary classification using the fuzzy support vector machine, this embodiment uses the density peak clustering algorithm to further extract typical prospecting target areas. Based on the determined clustering centers, use the density peak clustering algorithm to assign all prospecting target areas to the category where the nearest clustering center is located. During the assignment process, consider the fuzzy membership information of the prospecting target areas. For prospecting target areas with similar fuzzy memberships, make a comprehensive judgment by combining factors such as their geological potential, environmental sensitivity, and spatial location to ensure a more reasonable assignment result. Then analyze the clustering results, and select a set of representative typical prospecting target areas according to indicators such as the scale, compactness of the clustering, and similarity to the known ore-forming areas. Conduct on-site verification of the set of typical prospecting target areas or compare with existing geological data to evaluate the accuracy and reliability of the clustering results. If it is found that there are deviations in the clustering results, relevant parameters can be adjusted in the previous steps or the clustering analysis can be performed again.

[0104] This embodiment extracts typical prospecting target areas based on the clustering centers using the density peak clustering algorithm, which can effectively classify the prospecting target areas according to similarity and extract a set of representative typical target areas. By comprehensively considering multiple factors such as fuzzy membership during clustering assignment and typical target area screening and verification, the accuracy and reliability of typical prospecting target areas are improved. These typical prospecting target areas can provide key references for subsequent resource decisions, reduce unnecessary exploration work, improve decision-making efficiency, and at the same time reduce the impact on environmentally sensitive areas.

[0105] Step S500: Construct a three-dimensional decision space and use a multi-objective optimization algorithm to solve the decision result of the prospecting target area;

[0106] Step S500 specifically includes:

[0107] Step S510: Nonlinearly couple the quantification results of the environmental sensitivity, geological potential of the typical prospecting target areas and the modulus of their vertical gradient to construct a comprehensive decision-making index; The expression of the comprehensive decision-making index is:

[0108]

[0109] In the formula, represents the comprehensive decision-making index for the th typical ore prospecting target area; represents the quantification result of the geological potential of the th typical ore prospecting target area; represents the quantification result of the environmental sensitivity of the th typical ore prospecting target area; represents the geological potential enhancement coefficient, with a value range of 0.3 - 0.5, reflecting the weight of deep mineralization potential; represents the environmental sensitivity suppression coefficient, with a value range of 0.8 - 1.2, used to adjust the influence of environmental sensitivity in comprehensive decision-making, with a value range of 0.8 - 1.2; represents the gradient smoothing factor, with a value range of 1.0 - 2.0, used to suppress the interference of isolated anomalies; represents the smoothing constant, used to prevent the denominator from being zero; represents the vertical gradient modulus of the geological potential of the th typical ore prospecting target area; In this embodiment, by non-linearly coupling the environmental sensitivity, the quantification result of geological potential, and its vertical gradient modulus of the typical ore prospecting target area, a comprehensive decision-making index is constructed, comprehensively considering the influence of various key factors on ore prospecting decision-making. Using a scientific parameter determination method, the index can more accurately reflect the actual value and environmental impact of the ore prospecting target area. This comprehensive decision-making index provides a unified quantification basis for subsequent multi-objective optimization, helps to find a more reasonable balance point between resource development and environmental protection, and improves the scientificity and accuracy of ore prospecting decision-making.

[0110] Step S520: Based on the comprehensive decision-making index and preset constraint conditions, construct a two-objective integer programming model for maximizing resource development value and minimizing environmental risk; then construct a three-dimensional decision space, where the three dimensions of this space are geological potential, environmental sensitivity, and mining depth. Each typical ore prospecting target area corresponds to a point in this three-dimensional space. In order to perform multi-objective optimization decision-making in the three-dimensional decision space, we need to define the variation functions of mining economic value and ecological impact with depth. The variation function of mining economic value with depth can be expressed as: , where , are constants. This function indicates that as the mining depth increases, the mining economic value changes linearly; the ecological impact attenuation function with depth can be expressed as: , where is a constant. This function indicates that as the mining depth increases, the ecological impact decays exponentially;

[0111] The expression of the two-objective integer programming model is:

[0112]

[0113] Wherein, represents the function of the mining economic value varying with depth, , is the first constant, i.e., the benchmark economic value constant, is the second constant, i.e., the attenuation coefficient of the mining economic value varying with depth, which can be fitted through historical cost data; represents the mining depth; represents the function of the ecological impact decaying with depth, , is the third constant, i.e., the ecological impact decay rate constant, which can be determined through ecological restoration experiments; represents the total budget constraint; represents the geological potential threshold; represents the environmental sensitivity threshold; represents a 0-1 decision variable; represents the th cost of a typical prospecting target area; represents the number of typical prospecting target areas; In this embodiment, a bi-objective integer programming model for maximizing the resource development value and minimizing the environmental risk is constructed, clarifying two core objectives in the prospecting decision-making, and by reasonably determining the functions of the mining economic value and the ecological impact varying with depth and various constraint conditions, the prospecting problem is transformed into a mathematical optimization problem. This model construction method can systematically consider various factors such as geological potential, environmental sensitivity, mining depth, cost, and budget, providing a comprehensive and quantitative decision-making framework for the selection of prospecting target areas, and realizing the reasonable development of resources and the effective protection of the environment.

[0114] Step S530: Solve the Pareto optimal solution set of the bi-objective integer programming model based on the genetic algorithm; Specifically, in this embodiment, the Pareto optimal solution set is solved based on the improved genetic algorithm, including the following steps:

[0115] Step S531: Perform binary encoding on the 0-1 decision variable , where each decision variable corresponds to a binary bit. It should be noted that, in order to improve the search efficiency of the algorithm, the encoding length is designed according to the number of typical prospecting target areas and the complexity of the problem, and the problems of low computational efficiency or limited search space caused by too long or too short encoding should be avoided;

[0116] Step S532: Randomly generate a certain number of initial populations;

[0117] Step S533: Configure two fitness functions corresponding to the two objective functions in the bi-objective integer programming model, i.e., maximizing the resource development value and minimizing the environmental risk;

[0118] Step S534: The roulette wheel selection method is combined with the elite retention strategy to ensure that excellent individuals can be retained in the next-generation population. At the same time, a certain degree of randomness is introduced through the roulette wheel selection method to avoid the algorithm falling into a local optimum. For multi-point crossover, the selected individuals are crossed according to the crossover probability to generate new individuals. Then, an adaptive mutation strategy is adopted, and the mutation probability is dynamically adjusted according to the evolutionary situation of the population.

[0119] Step S535: Keep performing genetic operations until the preset maximum number of iterations is met, and finally obtain the Pareto optimal solution set of the bi-objective integer programming model. This optimal solution set contains a series of solutions that achieve a balance between resource development value and environmental risk.

[0120] In summary, this embodiment solves the Pareto optimal solution set of the bi-objective integer programming model based on the improved genetic algorithm, utilizes the global search ability of the genetic algorithm, and searches for the optimal solution that meets the dual objectives of resource development and environmental protection in a complex solution space, avoiding falling into a local optimal solution.

[0121] Step S540: Optimize the Pareto optimal solution set by adopting dynamic diversity preservation and elite retention, and extract the final decision result for the prospecting target area. Specifically, dynamic diversity preservation includes regularly calculating the population entropy value. When the entropy value is lower than the preset threshold, randomly inject new individuals, and at the same time, the mutation probability decreases with the number of iterations and is adjusted using an exponential decay function. Elite retention specifically means retaining the optimal individual in the current population in each generation. Then, screen the Pareto optimal solution set, remove the inferior solutions, and retain the superior solutions. Finally, select one or more solutions from the optimized Pareto optimal solution set as the final decision result.

[0122] In summary, the present invention can accurately evaluate the environmental risk and resource potential of the prospecting target area respectively by constructing a spatio-temporal coupled dynamic quantification model of environmental sensitivity and a geological potential quantification model, enabling decision-makers to have a clearer understanding of each target area. Furthermore, typical prospecting target areas are extracted using fuzzy support vector machines and density peak clustering, effectively reducing the data processing volume while retaining key features and improving the decision-making efficiency. Finally, the multi-objective optimization in the three-dimensional decision space effectively balances resource development and environmental protection, reducing exploration costs and environmental risks, and providing efficient and scientific decision-making support for mineral exploration in highly sensitive areas. During the optimization process, the dynamic diversity preservation and elite retention strategies ensure the quality of the Pareto optimal solution set, further improving the decision-making accuracy. Ultimately, this method can provide a more scientific solution for resource decision-making in prospecting target areas. Especially when facing a large-scale target area set and highly sensitive regions, it significantly improves the accuracy and efficiency of resource decision-making in prospecting target areas, reduces unnecessary exploration costs and environmental damage, promotes the efficient development of geological prospecting work in complex environments, and realizes the coordinated development of mineral resource exploration and ecological environment protection.

[0123] Please refer to Figure 2 , which shows a schematic structural diagram of a prospecting target area resource decision-making system based on minimizing the impact on environmentally sensitive areas provided by an embodiment of the present invention. The system includes:

[0124] A data acquisition module configured to acquire geological data and environmental data of each prospecting target area in the target exploration area and perform preprocessing;

[0125] An environmental sensitivity dynamic quantification module configured to construct a spatio-temporal coupled environmental sensitivity dynamic quantification model based on the preprocessed environmental data and extract the environmental sensitivity quantification results of each prospecting target area;

[0126] A geological potential quantification module configured to construct a geological potential quantification model based on the preprocessed geological data set and extract the geological potential quantification results of each prospecting target area;

[0127] A typical target area extraction module configured to construct a prospecting target area difference matrix based on the environmental sensitivity and geological potential results and extract typical target areas through fuzzy support vector machine classification and density peak clustering;

[0128] A multi-objective optimization solution module configured to construct a three-dimensional decision space and use a multi-objective optimization algorithm to solve the prospecting target area decision result.

[0129] It should be noted that: the above sequence of embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0130] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

Claims

1. A resource decision-making method for prospecting target areas based on minimizing the impact of environmentally sensitive areas, characterized in that: The method comprises: Obtain geological and environmental data of each prospecting target area in the target exploration area and perform preprocessing, including: Obtain the vegetation coverage, slope, distance to the ecological protection area and permafrost stability index of each prospecting target area, construct an environmental data set and perform standardized preprocessing on the environmental data set; Obtain the conjugate fracture density, Bouguer gravity anomaly, and vertical concentration gradient of primary halo elements in each prospecting target area, construct a geological data set, and perform normalization preprocessing on the geological data set; Based on the pre-processed environmental data, a dynamic quantitative model of environmental sensitivity with temporal and spatial coupling is constructed to extract the quantitative results of environmental sensitivity of each prospecting target area, including: A triangular fuzzy judgment matrix was constructed to conduct fuzzy evaluation on the relative importance of environmental factors, and the fuzzy analytic hierarchy process was used to extract the weights of environmental factors. The environmental factor weights and pre-processed environmental data are coupled in time and space to extract the historical cumulative effects of environmental impacts; Based on the historical cumulative effect of environmental impact and the preset time attenuation coefficient, the quantitative results of environmental sensitivity of each prospecting target area are extracted; A geological potential quantitative model is constructed based on the preprocessed geological data set to extract the quantitative results of the geological potential of each prospecting target area, including: The pre-processed conjugate fracture density, Bouguer gravity anomaly, and vertical concentration gradient of primary halo elements are dimensionlessly processed. According to the dimensionless processed conjugate fracture density, Bouguer gravity anomaly, and vertical concentration gradient of primary halo elements, multi-source data fusion is performed to build a geological potential quantitative model, and the geological potential quantitative results of each prospecting target area are extracted, including: principal component analysis of the dimensionless processed conjugate fracture density, Bouguer gravity anomaly, and vertical concentration gradient of primary halo elements, and extraction of principal components with cumulative contribution rates greater than 85%. The nonlinear quantitative model of geological potential quantification can be expressed as: In the formula, Indicates The geological potential value of each prospecting target area; Indicates The weights of the principal components; Indicates principal component scores; represents the coefficient of the quadratic term, which is used to enhance the discrimination of high potential areas; represents the number of principal components; Based on the environmental sensitivity and geological potential results, a prospecting target area difference matrix is ​​constructed, and typical prospecting target areas are extracted through fuzzy support vector machine classification and density peak clustering, including: Based on environmental sensitivity, geological potential results and spatial distance differences, a prospecting target area difference matrix is ​​constructed. The prospecting target area difference can be expressed as: In the formula, Indicates The prospecting target area and Comprehensive differences between prospecting targets; represents the difference weight of environmental sensitivity, represents the weight of geological potential difference; represents the spatial distance difference weight; the difference matrix can be expressed as ; Indicates Quantification results of environmental sensitivity of each prospecting target area; Indicates Quantification results of environmental sensitivity of each prospecting target area; Indicates The geological potential value of each prospecting target area; Indicates The center coordinates of the prospecting target area; Indicates The center coordinates of the prospecting target area; taking the prospecting target area difference matrix as input, the fuzzy support vector machine is used to preliminarily classify the prospecting target area and assign fuzzy membership; The truncated kernel function is used to calculate the local density of the prospecting target area, and the cluster center is determined in combination with the minimum distance; Based on the cluster centers, the density peak clustering algorithm is used to extract typical prospecting target areas; Construct a three-dimensional decision space and use a multi-objective optimization algorithm to solve the decision results of the prospecting target area.

2. The method for decision-making on mineral prospecting target area resources according to claim 1, characterized in that: The expression of the environmental sensitivity quantification result is: In the formula, Indicates Quantification results of environmental sensitivity of each prospecting target area; Indicates the start time; Indicates the current time; represents the integral time variable; Indicates the preset time attenuation coefficient, which is used to reflect the ecological recovery rate; Indicates the number of environmental factors; Indicates The weight of environmental factors; Indicates The first A standardized environmental factor.

3. The method for decision-making on mineral prospecting target area resources according to any one of claims 1 to 2, characterized in that: Construct a three-dimensional decision space and use a multi-objective optimization algorithm to solve the decision results of the prospecting target area, including: The environmental sensitivity of typical prospecting targets, the quantitative results of geological potential and their vertical gradient modulus are nonlinearly coupled to construct a comprehensive decision-making index. Based on comprehensive decision-making indicators and preset constraints, a dual-objective integer programming model is constructed to maximize resource development value and minimize environmental risks; Solving the Pareto optimal solution set of the bi-objective integer programming model based on genetic algorithm; Dynamic diversity maintenance and elite retention are used to optimize the Pareto optimal solution set and extract the final decision-making results of mineral exploration target areas.

4. The method for decision-making on mineral prospecting target area resources according to claim 3, characterized in that: The expression of comprehensive decision-making index is: In the formula, Indicates Comprehensive decision-making indicators for typical prospecting targets; Indicates Quantification results of geological potential of typical prospecting target areas; Indicates Quantification results of environmental sensitivity of typical prospecting target areas; represents the geological potential enhancement coefficient; represents the environmental sensitivity inhibition coefficient; represents the gradient smoothing factor; represents the smoothing constant; Indicates The vertical gradient modulus of the geological potential of a typical prospecting target area.

5. The method for decision-making on mineral prospecting target area resources according to claim 4, characterized in that: The expression of the bi-objective integer programming model is: In the formula, represents the function of the economic value of mining changing with depth, , is the first constant, i.e. the benchmark economic value constant, is the second constant, i.e., the attenuation coefficient of the economic value of mining with depth; Indicates the mining depth; represents the attenuation function of ecological impact with depth, , is the third constant, i.e., the ecological impact attenuation rate constant; represents the total budget constraint; represents the geological potential threshold; represents the environmental sensitivity threshold; represents a 0-1 decision variable; Indicates The cost of a typical prospecting target area; Indicates the number of typical prospecting target areas.

6. A resource decision-making system for prospecting target areas based on minimizing the impact of environmentally sensitive areas, characterized in that: The method for resource decision-making in prospecting target areas based on minimizing the impact of environmentally sensitive areas as described in any one of claims 1 to 5 is adopted, and the system comprises: A data acquisition module is configured to acquire geological data and environmental data of each prospecting target area in the target exploration area and perform preprocessing; The environmental sensitivity dynamic quantification module is configured to construct a spatiotemporally coupled environmental sensitivity dynamic quantification model based on the preprocessed environmental data, and extract the environmental sensitivity quantification results of each prospecting target area; A geological potential quantification module is configured to construct a geological potential quantification model based on the preprocessed geological data set and extract the geological potential quantification results of each prospecting target area; Typical target area extraction module, configured to construct a prospecting target area difference matrix based on environmental sensitivity and geological potential results, and extract typical target areas through fuzzy support vector machine classification and density peak clustering; The multi-objective optimization solution module is configured to construct a three-dimensional decision space and use a multi-objective optimization algorithm to solve the decision results of the prospecting target area.

Images