Intelligent evaluation method and system for mineral resource exploration and storage medium

Abstract

The application relates to the technical field of mineral exploration, and discloses an intelligent evaluation method and system for mineral resource exploration and a storage medium. The method comprises the following steps: obtaining multi-source geological data of a region to be evaluated, dividing remote sensing physical data into first remote sensing physical data corresponding to sampling points and second remote sensing physical data not corresponding to the sampling points, constructing a preset model based on geological exploration data and the first remote sensing physical data, estimating geological exploration data of virtual sampling points by using the model, combining the estimated data with original data to generate new multi-source geological data, generating a first geological spatial structure and a plurality of second geological spatial structures by two preset geological modeling methods, selecting the closest one to the first geological spatial structure as a reference, analyzing the reference geological spatial structure, calculating the average abundance of minerals and the volume of ores, and then predicting the reserves of mineral resources and performing evaluation. The application improves the accuracy and efficiency of mineral resource evaluation and reduces the exploration cost.

Description

Technical Field

[0001] This application relates to the field of mineral exploration technology, and in particular to an intelligent evaluation method, system and storage medium for mineral resource exploration. Background Technology

[0002] With the rapid development of the global economy, the demand for mineral resources continues to grow. Traditional mineral resource exploration methods mainly rely on the experience of geologists and field exploration. This method is not only time-consuming and labor-intensive, but also often limits the accuracy and efficiency of exploration under complex geological conditions. In recent years, the vigorous development of emerging technologies such as artificial intelligence, big data, and the Internet of Things has brought new opportunities to the field of mineral resource exploration. Utilizing intelligent technologies to assess mineral resources can more efficiently integrate multi-source data, improve the accuracy of exploration, reduce exploration costs, shorten the exploration cycle, and better meet the needs of modern society for the efficient development and utilization of mineral resources.

[0003] A similar prior art is disclosed in Chinese patent application CN117392337A, which presents an AI-based digital mineral exploration method. This method identifies the mineral to be predicted based on element detection data within the area to be predicted; determines the size of each grid cell corresponding to the mineral based on its distribution within the area; converts the area into a rasterized image according to the determined grid sizes; determines the descriptive vector corresponding to each grid cell based on the actual geological map; expands the descriptive vector of each grid cell based on the coupling relationship between grid cells; inputs the descriptive vector of each grid cell into its corresponding grid cell to form a rasterized geological map; inputs the rasterized geological map into a mineral identification model to obtain mineral prediction results; and generates a mineral distribution map based on the mineral prediction results. This method predicts the probability of the presence of the mineral to be predicted in each grid cell using a rasterized geological map and a multi-branch convolutional neural network model, but it does not assess the mineral content within the area to be predicted. Another Chinese patent application, CN119204462A, discloses an intelligent mineral resource exploration and assessment system. This system collects geological and mineral exploration-related data from multiple data sources; preprocesses the data to obtain preprocessed data; analyzes the rock physical properties and geological structural stress state of the exploration area based on the preprocessed data to determine the correlation between these two factors; calculates a dynamic adjustment coefficient based on this correlation; estimates mineral resource reserves based on the dynamic adjustment coefficient; and finally assesses the mineral resource reserves to arrive at a final assessment result. However, this method, which estimates and assesses mineral resource reserves by analyzing rock physical properties and geological structural stress state, lacks sufficient precision and accuracy when dealing with complex geological conditions.

[0004] Therefore, providing an intelligent assessment method, system, and storage medium for mineral resource exploration to improve the accuracy and reliability of mineral resource exploration assessment is an urgent problem to be solved. Summary of the Invention

[0005] This application provides an intelligent assessment method, system, and storage medium for mineral resource exploration, which improves the efficiency and accuracy of mineral resource assessment.

[0006] Firstly, this application provides an intelligent assessment method for mineral resource exploration, the method comprising:

[0007] Step 1: Obtain multi-source geological data for the area to be evaluated. Multi-source geological data includes remote sensing physical data, remote sensing image data, drilling data, geological structure data, and geological exploration data.

[0008] Step 2: Define the data in the remote sensing physical data that corresponds to the sampling points in the geological exploration data as the first remote sensing physical data, and define the data that does not correspond to the sampling points as the second remote sensing physical data. Construct a preset model based on the geological exploration data and the first remote sensing physical data, input the second remote sensing physical data into the preset model, and obtain the estimated geological exploration data at the virtual sampling points in the area to be evaluated.

[0009] Step 3: Add the estimated geological exploration data to the multi-source geological data to generate the first multi-source geological data, and delete the second remote sensing physical data from the multi-source geological data to generate the second multi-source geological data;

[0010] Step 4: Based on the first multi-source geological data, generate the first geological spatial structure of the area to be evaluated using the first preset geological modeling method; based on the second multi-source geological data, generate N1 second geological spatial structures using the second preset geological modeling method; select the spatial structure that is closest to the first geological spatial structure from all the second geological spatial structures as the reference geological spatial structure.

[0011] Step 5: Analyze the reference geological spatial structure to obtain the average mineral abundance and ore volume, and calculate the predicted mineral resource reserves of the area to be evaluated based on the average mineral abundance and ore volume.

[0012] Step 6: Assess the predicted reserves of mineral resources and obtain the final assessment results.

[0013] In conjunction with the first aspect, in the first implementation of the first aspect of this application, step 2 includes:

[0014] Step 21: Serialize the geological exploration data based on the measurement values ​​of specific elements, set the boundary points, and divide the serialized geological exploration data into N2 first data sets based on the boundary points.

[0015] Step 22: Based on the sampling points corresponding to the geological exploration data in each first data set, the first remote sensing physical data is divided into N2 first data sets to generate a second data set. After traversing all the first data sets, N2 second data sets are obtained.

[0016] Step 23: Train N2 second datasets to generate a first model for identifying the dataset to which the first remote sensing physical data belongs. At the same time, train each second dataset separately to construct a preset model for predicting geological exploration data based on the first remote sensing physical data for each second dataset.

[0017] Step 24: Use the first model to divide the second remote sensing physical data into different data sets, which are defined as the assigned data sets. Extract any assigned data set, input the data in any assigned data set into the corresponding preset model, and obtain the estimated geological exploration data corresponding to each data in any assigned data set.

[0018] In conjunction with the first aspect, in the second implementation of the first aspect of this application, the method for setting the boundary point in step 21 is as follows:

[0019] Step 211: Execute a predetermined algorithm on the measured values ​​corresponding to specific elements to generate transformed values, identify the quartiles of all transformed values, extract the transformed values ​​within the first preset range before and after the quartiles, and generate a reference data set.

[0020] Step 212: Fit the data in the reference dataset based on a preset mathematical method to generate a fitting line, and define the value corresponding to each sampling point on the fitting line as the fitting value.

[0021] Step 213: Calculate the difference between the transformed value and the fitted value for each sampling point in the reference data set, and calculate the average of the squares of all differences. Define the square root of the average as the dispersion value.

[0022] Step 214: Set the fluctuation range based on the quartile and dispersion values, set the measured value corresponding to the largest change value within the fluctuation range as the first boundary point, and set the measured value corresponding to the smallest change value within the fluctuation range as the second boundary point.

[0023] In conjunction with the first aspect, in the third implementation of the first aspect of this application, step 2 is followed by:

[0024] Step 11: Extract all measured values ​​corresponding to specific elements from geological exploration data, and delete the values ​​that are less than the first preset value from all measured values ​​to generate the first set of values ​​to be analyzed;

[0025] Step 12: Obtain the content range and average content of a specific element in the Earth's crust in the background area, and set a benchmark value based on the content range. The area to be evaluated is included in the background area.

[0026] Step 13: Delete the values ​​greater than the benchmark value from the first set of values ​​to be analyzed, generate the second set of values ​​to be analyzed, use the average content value as the average value of the second set of values ​​to be analyzed, and determine whether the second set of values ​​to be analyzed meets the preset distribution law. If not, proceed to step 14; if yes, proceed to step 15.

[0027] Step 14: Reduce the baseline value by the second preset value to generate a new baseline value, and define the second set of values ​​to be analyzed as the first set of values ​​to be analyzed, then return to step 13;

[0028] Step 15: Define the maximum value of the second set of numerical values ​​to be analyzed as the critical value of a specific element.

[0029] In conjunction with the first aspect, in the fourth implementation of the first aspect of this application, step 3 includes:

[0030] Traverse the first remote sensing physical data. When the measured value of a specific element at any sampling point is less than the critical value, define the geological exploration data and the first remote sensing physical data corresponding to any sampling point as the first redundant data. Delete the first redundant data from the second multi-source geological data and generate new second multi-source geological data.

[0031] When the measured value of a specific element at any virtual sampling point is less than the critical value, the estimated geological exploration data and the second remote sensing physical data corresponding to any virtual sampling point are defined as the second redundant data. The first redundant data and the second redundant data are deleted from the first multi-source geological data to generate new first multi-source geological data.

[0032] In conjunction with the first aspect, in the fifth implementation of the first aspect of this application, step 4 includes:

[0033] Step 41: Extract features from the first feature parameters of each second geological spatial structure, reduce the total number of first feature parameters to a preset number, and obtain the low-dimensional spatial structure corresponding to each second geological spatial structure.

[0034] Step 42: Based on preset rules, assign values ​​to the non-numerical parameters in each low-dimensional spatial structure. Then, standardize the second feature parameters of all low-dimensional spatial structures. Calculate the Euclidean distance between any two low-dimensional spatial structures based on the standardization results. Group all second geological spatial structures based on the Euclidean distance to obtain N3 sets of spatial structures.

[0035] Step 43: Extract the geological features of the ore deposits for all geological spatial structures, and obtain the first ore body attribute parameters and the second ore body attribute parameters corresponding to the first geological spatial structure and each second geological spatial structure respectively;

[0036] Step 44: Extract any set of spatial structures, perform statistical analysis on the attribute parameters of the second ore bodies of all the second geological spatial structures in any set of spatial structures, and obtain the attribute parameters of the third ore bodies in any set of spatial structures.

[0037] Step 45: After traversing all spatial structure sets, the spatial structure set corresponding to the third ore body attribute parameter with the highest similarity to the first ore body attribute data is defined as a specific structure set. Mineral resource assessments are performed on the first geological spatial structure and each third geological spatial structure to obtain the first mineral resource statistical parameters and the second mineral resource statistical parameters. The third geological spatial structure is the second geological spatial structure contained in the specific structure set.

[0038] Step 46: Define the third geological spatial structure corresponding to the second mineral resource statistical parameter that has the highest similarity to the first mineral resource statistical parameter as the reference geological spatial structure.

[0039] In conjunction with the first aspect, in the sixth implementation of the first aspect of this application, step 42 further includes:

[0040] Calculate the ratio of the total number to N3. When the ratio is greater than the second preset range, divide the area to be evaluated into N3 sub-evaluation areas. Divide the first multi-source geological data and the second multi-source geological data into different sub-evaluation areas. Then return to step 4 and perform mineral reserve prediction for each sub-evaluation area.

[0041] When the ratio is less than the second preset range, a new sampling point is generated based on the coordinates of the existing sampling point. Geological data is collected from the new sampling point. After the collection is completed, return to step 1.

[0042] Secondly, this application provides an intelligent assessment system for mineral resource exploration, the system comprising:

[0043] The data acquisition module is used to acquire multi-source geological data of the area to be evaluated. The multi-source geological data includes remote sensing physical data, remote sensing image data, drilling data, geological structure data, and geological exploration data.

[0044] The data filling module is used to define the data in the remote sensing physical data that corresponds to the sampling points of the geological exploration data as the first remote sensing physical data, and the data that does not correspond to the sampling points as the second remote sensing physical data. Based on the geological exploration data and the first remote sensing physical data, a preset model is constructed, and the second remote sensing physical data is input into the preset model to obtain the estimated geological exploration data at the virtual sampling points in the area to be evaluated.

[0045] The data grouping module is used to add estimated geological exploration data to multi-source geological data to generate the first multi-source geological data, and to delete the second remote sensing physical data from multi-source geological data to generate the second multi-source geological data.

[0046] The spatial structure generation module is used to generate a first geological spatial structure of the area to be evaluated based on the first multi-source geological data and a first preset geological modeling method, and to generate N1 second geological spatial structures based on the second multi-source geological data and a second preset geological modeling method. The module selects the spatial structure that is closest to the first geological spatial structure from all the second geological spatial structures as a reference geological spatial structure.

[0047] The reserve prediction module is used to analyze the reference geological spatial structure, obtain the average mineral abundance and ore volume, and calculate the predicted mineral resource reserves of the area to be evaluated based on the average mineral abundance and ore volume.

[0048] The intelligent assessment module is used to assess the predicted reserves of mineral resources and obtain the final assessment results.

[0049] A third aspect of this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the aforementioned intelligent evaluation method for mineral resource exploration.

[0050] Compared with the prior art, the beneficial effects of the technical solution of this application are at least as follows:

[0051] 1. By integrating multiple data sources such as remote sensing physical data, remote sensing image data, drilling data, geological structure data, and geological exploration data, the geological characteristics of the area to be evaluated can be more comprehensively reflected, which helps to construct a comprehensive and detailed geological spatial structure model.

[0052] 2. Based on existing data, a preset model is constructed using geological exploration data and first remote sensing physical data. This preset model is used to predict geological exploration data at virtual sampling points, which improves the integrity and accuracy of the data. It can reduce sampling costs while increasing data coverage and accuracy.

[0053] 3. Based on different first and second multi-source geological data, a first geological spatial structure and multiple second geological spatial structures are generated respectively. Based on similarity analysis, the second geological spatial structure that is closest to the first geological spatial structure is selected as the reference geological spatial structure that is closest to the actual situation, thereby reducing the bias of a single model and improving the accuracy and reliability of mineral resource reserve prediction. Attached Figure Description

[0054] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is a schematic diagram of an embodiment of the intelligent assessment method for mineral resource exploration in this application.

[0056] Figure 2 This is a schematic diagram of an embodiment of the method for generating estimated geological exploration data in this application.

[0057] Figure 3 A schematic diagram of an embodiment for setting up new sampling points in this application;

[0058] Figure 4 This is a schematic diagram of one embodiment of an intelligent assessment system for mineral resource exploration in this application. Detailed Implementation

[0059] This application provides an intelligent evaluation method, system, and storage medium for mineral resource exploration. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0060] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the intelligent assessment method for mineral resource exploration in this application includes:

[0061] Step 1: Obtain multi-source geological data for the area to be evaluated. Multi-source geological data includes remote sensing physical data, remote sensing image data, drilling data, geological structure data, and geological exploration data.

[0062] Specifically, multi-source geological data of the area to be monitored is obtained using geological surveys, drilling, remote sensing, and geophysical exploration techniques. The aforementioned remote sensing physical data refers to data related to geophysical characteristics acquired from the Earth's atmosphere using remote sensing technology, such as vertical height relative to sea level, topographic slope, and the ratio of reflected light intensity to incident light intensity at a specific wavelength. Remote sensing image data refers to images of the area to be evaluated acquired using satellite or airborne remote sensing equipment. Drilling data refers to sample data of underground rocks and ores obtained through drilling operations, including rock composition, ore grade, and stratum thickness. Geological structure data refers to data describing the internal structure of a geological region, such as the dip angle, strike, thickness, and distribution characteristics of strata. Geological exploration data refers to geological information obtained through geological exploration methods (such as geological mapping and geochemical exploration), including rock type, ore composition ratio, content, ground stability, and stratum age.

[0063] Step 2: Define the data in the remote sensing physical data that corresponds to the sampling points in the geological exploration data as the first remote sensing physical data, and define the data that does not correspond to the sampling points as the second remote sensing physical data. Construct a preset model based on the geological exploration data and the first remote sensing physical data, input the second remote sensing physical data into the preset model, and obtain the estimated geological exploration data at the virtual sampling points in the area to be evaluated.

[0064] Specifically, remote sensing physical data provides a wealth of information about the Earth's surface and atmosphere, which is correlated with geological exploration data. By establishing mathematical models and using machine learning algorithms, these correlations can be analyzed to predict geological exploration data for unknown areas.

[0065] Actual geological exploration is time-consuming and labor-intensive, and the number of sampling points for geological exploration data acquisition is limited. Remote sensing physical data is easier to acquire, typically possessing high spatial resolution and coverage, but it doesn't perfectly correspond to the sampling points in geological exploration data. This approach divides remote sensing physical data into first-level remote sensing physical data corresponding to the sampling points in geological exploration data and second-level remote sensing physical data not corresponding to any sampling points. By using the first-level remote sensing physical data and the geological exploration data to learn the relationship between the two, a model for predicting geological exploration data is constructed. This model is then used to estimate the geological exploration data at virtual sampling points corresponding to the second-level remote sensing physical data that were not actually sampled. This supplements the geological exploration data, improves the data completeness of the entire area to be evaluated, and provides more comprehensive data support for subsequent geological spatial structure generation and reserve assessment.

[0066] Step 3: Add the estimated geological exploration data to the multi-source geological data to generate the first multi-source geological data, and delete the second remote sensing physical data from the multi-source geological data to generate the second multi-source geological data.

[0067] Specifically, by adding estimated geological exploration data to multi-source geological data, data gaps between actual sampling points can be filled, making the first multi-source geological data more complete. The first geological spatial structure generated based on the first multi-source geological data can more accurately reflect the actual geological conditions. Deleting the second remote sensing physical data from the original multi-source geological data can reduce redundant information in the data, improve data quality and usability, and reduce the impact of redundant data on the geological spatial structure by generating multiple second geological spatial structures based on the second multi-source geological data.

[0068] Step 4: Based on the first multi-source geological data, generate the first geological spatial structure of the area to be evaluated using the first preset geological modeling method. Based on the second multi-source geological data, generate N1 second geological spatial structures using the second preset geological modeling method. Select the spatial structure that is closest to the first geological spatial structure from all the second geological spatial structures as the reference geological spatial structure.

[0069] Specifically, the geological spatial structure is a three-dimensional spatial model of the area to be evaluated. The first pre-set geological modeling method is to generate the geological spatial structure using geological modeling software (such as GOCAD, Surpac, Leapfrog, etc.). The second pre-set geological modeling method is to construct the geological spatial structure using a multi-point geostatistical method. The multi-point geostatistical method is a statistical-based geological modeling method used to generate three-dimensional geological models. It captures complex geological patterns and structures using training images and generates multiple possible geological models by calculating conditional probabilities. Constructing a geological spatial structure reflecting the spatial distribution of geological bodies using the multi-point geostatistical method can capture more complex geological structures, such as faults, folds, and veins, and can generate multiple geological spatial structures. Each geological spatial structure is generated based on conditions of known data points, providing different possible realizations of orebody characteristics, thereby enabling the assessment of the uncertainty of orebody characteristics.

[0070] The first geological spatial structure is a three-dimensional spatial model generated based on more complete data, which can more comprehensively reflect the geological characteristics of the area to be evaluated. The second geological spatial structure is a three-dimensional spatial model that takes into account the uncertainty of ore body characteristics. By comparing the similarity between multiple second geological spatial structures and the first geological spatial structure, the closest structure is selected as the reference geological spatial structure. This can reduce the errors that may be caused by a single modeling method. By selecting the geological spatial structure that is most similar to the actual geological data, and using the reference geological spatial structure that is closest to the actual geological conditions as the basis for mineral resource reserve prediction, the accuracy of resource estimation can be improved.

[0071] Step 5: Analyze the reference geological spatial structure to obtain the average mineral abundance and ore volume, and calculate the predicted mineral resource reserves of the area to be evaluated based on the average mineral abundance and ore volume.

[0072] Specifically, average mineral abundance refers to the average content of the target mineral within a reference geological spatial structure; ore volume refers to the total volume of the ore within the reference geological spatial structure. The product of average mineral abundance and ore volume is used as the predicted mineral resource reserves.

[0073] Step 6: Assess the predicted reserves of mineral resources and obtain the final assessment results.

[0074] Specifically, the assessment results include mineral resource reserve assessment results (estimate the total amount of all mineral resources in the region, including different types of mineral resources (such as metallic minerals, non-metallic minerals, energy minerals, etc.)), mineral resource economic value assessment results (estimate the economic value of mineral resources based on ore grade, market price, etc.), and mineral resource development feasibility assessment (assess the mining technology conditions of the deposit, including ore body occurrence conditions, rock stability, groundwater conditions, etc.).

[0075] In one specific embodiment, the process of performing step 2 may specifically include the following steps:

[0076] Step 21: Serialize the geological exploration data based on the measurement values ​​of specific elements, set a boundary point, and divide the serialized geological exploration data into N2 first data sets based on the boundary point.

[0077] Step 22: Based on the sampling points corresponding to the geological exploration data in each first data set, the first remote sensing physical data is divided into N2 first data sets to generate a second data set. After traversing all the first data sets, N2 second data sets are obtained.

[0078] Step 23: Train N2 second datasets to generate a first model for identifying the dataset to which the first remote sensing physical data belongs. At the same time, train each second dataset separately to construct a preset model for predicting geological exploration data based on the first remote sensing physical data for each second dataset.

[0079] Step 24: Use the first model to divide the second remote sensing physical data into different data sets, which are defined as the assigned data sets. Extract any assigned data set, input the data in any assigned data set into the corresponding preset model, and obtain the estimated geological exploration data corresponding to each data in any assigned data set.

[0080] The flowchart of the method for generating estimated geological exploration data is as follows: Figure 2 As shown. Specifically, a specific element is a key element related to mineral resources, such as gold, copper, and iron. The measured value refers to the degree or content of a certain element or compound in the Earth's crust, usually expressed as a mass percentage or ppm (parts per million). Based on the boundary points, the serialized geological exploration data is divided into different sets, which allows data with the same or similar content of a specific element to be grouped into one set. The data in each set have similar characteristics (low content of the specific element, moderate content of the specific element, high content of the specific element). Based on the spatial correspondence between sampling points and remote sensing physical data, the remote sensing physical data is correspondingly assigned to different data sets and stored in correspondence with its corresponding remote sensing physical data, generating a data pair set (i.e., the second data set).

[0081] The first model and the preset model are neural network models. A corresponding preset model is constructed for each second dataset to improve the model's predictive ability. During data estimation, the second remote sensing physical data is first classified based on the first model, dividing it into different datasets. This achieves classification and prediction of remote sensing data in unsampled areas, associating them with similar geological conditions. Then, based on the dataset, the preset model corresponding to that dataset is extracted. Based on this preset model, the estimated geological exploration data corresponding to each remote sensing physical data point in that dataset is obtained, enabling more accurate prediction of geological exploration data for virtual sampling points and improving the accuracy of geological exploration data estimation.

[0082] In one specific embodiment, the method for setting the boundary point in step 21 is as follows:

[0083] Step 211: Execute a predetermined algorithm on the measured values ​​corresponding to specific elements to generate transformed values, identify the quartiles of all transformed values, extract the transformed values ​​within the first preset range before and after the quartiles, and generate a reference data set.

[0084] Step 212: Fit the data in the reference dataset based on a preset mathematical method to generate a fitting line, and define the value corresponding to each sampling point on the fitting line as the fitting value.

[0085] Step 213: Calculate the difference between the transformed value and the fitted value for each sampling point in the reference data set, and calculate the average of the squares of all differences. Define the square root of the average as the dispersion value.

[0086] Step 214: Set the fluctuation range based on the quartile and dispersion values, set the measured value corresponding to the largest change value within the fluctuation range as the first boundary point, and set the measured value corresponding to the smallest change value within the fluctuation range as the second boundary point.

[0087] Specifically, the pre-defined algorithms include data standardization, data normalization, logarithmic transformation, logarithmic-linear transformation, square root transformation, etc. Executing the pre-defined algorithms is to handle the nonlinear relationships and heteroscedasticity of the data, so as to eliminate the influence of dimensions, compress the dynamic range, and effectively adjust the data distribution, which helps to improve the performance and effectiveness of the model.

[0088] The transformed values ​​within a first preset range before and after the quartile are extracted to generate a reference data set. For example, the first preset range is set to generate the reference data set based on 10% of the transformed values ​​before and after the quartile (e.g., if the quartile is 10, the reference data set is generated based on numbers between [9, 11]). The specific first preset range can be set according to the experience of those skilled in the art or according to the actual application scenario; this embodiment does not limit this. Subsequently, the data in the reference data set is fitted using a preset mathematical method (e.g., linear regression, curve fitting, etc.) to generate a fitted line.

[0089] Setting the fluctuation range based on quartiles and dispersion values ​​allows for a more reasonable division of data intervals. For example, the fluctuation range could be [M-2×SD, M+2×SD], where M is the quartile and SD is the dispersion value. Determining the boundary point by the maximum and minimum values ​​within the fluctuation range enables more accurate segmentation of the data set and improves the accuracy of data classification.

[0090] In one specific embodiment, step 2 is followed by:

[0091] Step 11: Extract all measurement values ​​corresponding to specific elements from the geological exploration data, and delete the values ​​that are less than the first preset value from all measurement values ​​to generate the first set of values ​​to be analyzed.

[0092] Step 12: Obtain the content range and average content of a specific element in the Earth's crust within the background area, and set a baseline value based on the content range. The area to be evaluated is included within the background area.

[0093] Step 13: Delete the values ​​greater than the benchmark value from the first set of values ​​to be analyzed, and generate the second set of values ​​to be analyzed. Use the average content value as the average value of the second set of values ​​to be analyzed. Determine whether the second set of values ​​to be analyzed meets the preset distribution law. If not, proceed to step 14; if yes, proceed to step 15.

[0094] Step 14: Reduce the baseline value by the second preset value to generate a new baseline value, and define the second set of values ​​to be analyzed as the first set of values ​​to be analyzed, then return to step 13.

[0095] Step 15: Define the maximum value of the second set of numerical values ​​to be analyzed as the critical value of a specific element.

[0096] Specifically, the specific elements are key elements related to mineral resources, such as gold, copper, and iron. The first preset value is a background value or detection limit. By deleting values ​​lower than the first preset value, data that may not have mineral value is removed, noise is reduced, and data processing efficiency is improved.

[0097] The background area is a larger region encompassing the area to be assessed, typically including the area under assessment and its surrounding geological environment. The range of a specific element's abundance in the Earth's crust within the background area refers to the normal abundance range of that element in the crust within an area unaffected by mineralization; this range is usually obtained through geological surveys and statistical analysis.

[0098] First, the maximum value or a percentage (e.g., 90%) of the content range of a specific element in the background area is set as the baseline value. For example, if the copper (Cu) content in the background area ranges from 0.1% to 2.0%, then 2.0% or 90% of it (i.e., 1.8%) can be used as the baseline value. Next, the measured values ​​of specific elements extracted from geological exploration data are analyzed to form a second set of values ​​to be analyzed. If this set of values ​​does not meet the preset distribution pattern (such as a symmetrical unimodal asymptotic distribution, Gaussian distribution, or bell-shaped distribution), this usually means that there are measured values ​​corresponding to mineralization anomalies in the data. Mineralization can cause the content of certain elements to be significantly higher than the background value, forming geochemical anomalies. These anomalies usually appear as high values ​​at the tail of the data distribution. Therefore, it is necessary to reduce the baseline value according to the second preset value (e.g., 0.1%) to generate a new baseline value and then judge it again. If the data in the second set of values ​​to be analyzed follow the preset distribution pattern, it usually means that these data mainly reflect the distribution of the background value (i.e., areas without obvious mineralization). At this point, the maximum value of the data in the second set of values ​​to be analyzed when it follows a preset distribution pattern can be used as the critical value for determining whether mineralization exists. This critical value is used to distinguish between background values ​​and mineralization anomalies, thereby more accurately identifying mineralized areas.

[0099] Specifically, the background area is a broader area than the area to be evaluated, and its data reflects the general range of the content of a specific element in the crust in this area. Using the above-mentioned average content as the average value of the second set of values ​​to be analyzed, it is possible to determine whether the second set of values ​​to be analyzed meets the preset distribution law, which can reduce the possibility of misjudgment and improve the reliability of judging the preset distribution law.

[0100] In one specific embodiment, the process of performing step 3 may specifically include the following steps:

[0101] (1) Traverse the first remote sensing physical data. When the measured value of a specific element at any sampling point is less than the critical value, define the geological exploration data and the first remote sensing physical data corresponding to any sampling point as the first redundant data. Delete the first redundant data from the second multi-source geological data and generate new second multi-source geological data.

[0102] (2) Traverse the estimated geological exploration data. When the measured value of a specific element at any virtual sampling point is less than the critical value, define the estimated geological exploration data and the second remote sensing physical data corresponding to any virtual sampling point as the second redundant data. Delete the first redundant data and the second redundant data from the first multi-source geological data to generate new first multi-source geological data.

[0103] Specifically, removing redundant data that is irrelevant to mineral resources or is below a critical value can reduce noise in the data, improve data quality and usability, and enhance the efficiency and accuracy of subsequent geological modeling and resource assessment.

[0104] In one specific embodiment, the process of performing step 4 may specifically include the following steps:

[0105] Step 41: Extract features from the first feature parameters of each second geological spatial structure, reduce the total number of first feature parameters to a preset number, and obtain the low-dimensional spatial structure corresponding to each second geological spatial structure.

[0106] Step 42: Based on preset rules, assign values ​​to the non-numerical parameters in each low-dimensional spatial structure. Then, standardize the second feature parameters of all low-dimensional spatial structures. Calculate the Euclidean distance between any two low-dimensional spatial structures based on the standardization results. Group all second geological spatial structures based on the Euclidean distance to obtain N3 sets of spatial structures.

[0107] Step 43: Extract the geological features of the ore deposits from all geological spatial structures, and obtain the first ore body attribute parameters and the second ore body attribute parameters corresponding to the first geological spatial structure and each second geological spatial structure.

[0108] Step 44: Extract any set of spatial structures, perform statistical analysis on the attribute parameters of the second ore bodies of all the second geological spatial structures in any set of spatial structures, and obtain the attribute parameters of the third ore bodies in any set of spatial structures.

[0109] Step 45: After traversing all spatial structure sets, define the spatial structure set corresponding to the third ore body attribute parameter with the highest similarity to the first ore body attribute data as a specific structure set. Perform mineral resource assessment on the first geological spatial structure and each third geological spatial structure to obtain the first mineral resource statistical parameters and the second mineral resource statistical parameters. The third geological spatial structure is the second geological spatial structure contained in the specific structure set.

[0110] Step 46: Define the third geological spatial structure corresponding to the second mineral resource statistical parameter that has the highest similarity to the first mineral resource statistical parameter as the reference geological spatial structure.

[0111] Specifically, the number of second geological spatial structures generated by the second pre-defined geological modeling method can be very large and diverse. Directly selecting a reference geological spatial structure from all second geological spatial structures is not only computationally expensive but may also overlook some important geological features. By grouping the large number of geological spatial structures into multiple smaller sets, each set containing geological spatial structures with similar geological features, the number of geological spatial structures that need to be evaluated can be significantly reduced, improving computational efficiency.

[0112] The second geological spatial structure is a high-dimensional three-dimensional spatial model, which usually contains a large number of feature parameters. Before grouping, dimensionality reduction is performed using techniques such as multidimensional scaling (MDS) and principal component analysis (PCA) to reduce the number of feature parameters, thereby reducing computational complexity and improving computational efficiency.

[0113] In geological modeling, geological spatial structures typically contain various geological features and parameters. These parameters can be continuous numerical values ​​(such as the size and grade of ore bodies) or discrete categories (such as the type of ore body, the type of rock strata, etc.). To calculate the distances between these geological spatial structures, it is necessary to convert discrete categories into numerical values ​​(i.e., assign values ​​to non-numerical parameters). For example, for ore body types: Category 1 represents ore body type A, Category 2 represents ore body type B, and during the assignment process, ore body type A is assigned a value of 1, and ore body type B is assigned a value of 2; for rock strata types: Category 1 represents sandstone, Category 2 represents shale, Category 3 represents limestone, etc., and during the assignment process, sandstone is assigned a value of 1, shale is assigned a value of 2, and limestone is assigned a value of 3.

[0114] Specifically, ore body attribute parameters include the morphology, distribution, location, volume, and extent of the ore body. Statistical analysis is performed on the content of useful minerals in the ore of each third geological spatial structure to obtain the average content, standard deviation, coefficient of variation, and distribution as statistical parameters for mineral resources.

[0115] In one specific embodiment, the process of performing step 42 further includes the following steps:

[0116] (1) Calculate the ratio of the total quantity to N3. When the ratio is greater than the second preset range, divide the area to be evaluated into N3 sub-evaluation areas, divide the first multi-source geological data and the second multi-source geological data into different sub-evaluation areas, and then return to step 4 to predict mineral reserves for each sub-evaluation area.

[0117] (2) When the ratio is less than the second preset range, a new sampling point is generated based on the existing sampling point coordinates. Geological data is collected from the new sampling point. After the collection is completed, return to step 1.

[0118] Specifically, when the ratio of the total quantity to N3 is greater than the second preset range, it can be determined that the ore body structure (quantity, distribution, stratigraphy, etc.) of the area to be evaluated is relatively complex. To improve the accuracy of ore body reserve prediction, the area to be evaluated is divided into multiple sub-evaluation areas, and each sub-evaluation area is evaluated separately. When the ratio of the total quantity to N3 is greater than the second preset range, it indicates insufficient data. To improve the accuracy of ore body reserve prediction, new sampling points need to be added to collect more geological data. Preferably, new sampling points are set according to the distance between existing sampling points, such as... Figure 3 As shown, P1, P2, P3, and P4 are existing sampling points. Since the distances between P1 and P4, P2 and P4, and P3 and P4 are relatively large, new sampling points P5, P6, and P7 are set.

[0119] The above describes the intelligent evaluation method for mineral resource exploration in the embodiments of this application. The following describes the intelligent evaluation system for mineral resource exploration in the embodiments of this application. Please refer to [link / reference]. Figure 4 One embodiment of the intelligent assessment system for mineral resource exploration in this application includes:

[0120] The data acquisition module 10 is used to acquire multi-source geological data of the area to be evaluated. The multi-source geological data includes remote sensing physical data, remote sensing image data, drilling data, geological structure data, and geological exploration data.

[0121] The data filling module 20 is used to define the data in the remote sensing physical data that corresponds to the sampling points of the geological exploration data as the first remote sensing physical data, and the data that does not correspond to the sampling points as the second remote sensing physical data. Based on the geological exploration data and the first remote sensing physical data, a preset model is constructed, and the second remote sensing physical data is input into the preset model to obtain the estimated geological exploration data at the virtual sampling points in the area to be evaluated.

[0122] The data grouping module 30 is used to add estimated geological exploration data to multi-source geological data to generate first multi-source geological data, and to delete second remote sensing physical data from multi-source geological data to generate second multi-source geological data.

[0123] The spatial structure generation module 40 is used to generate a first geological spatial structure of the area to be evaluated based on the first multi-source geological data and a first preset geological modeling method, generate N1 second geological spatial structures based on the second multi-source geological data and a second preset geological modeling method, and select the spatial structure that is closest to the first geological spatial structure from all the second geological spatial structures as the reference geological spatial structure.

[0124] The reserve prediction module 50 is used to analyze the reference geological spatial structure, obtain the average mineral abundance and ore volume, and calculate the predicted mineral resource reserves of the area to be evaluated based on the average mineral abundance and ore volume.

[0125] The intelligent assessment module 60 is used to assess the predicted reserves of mineral resources and obtain the final assessment results.

[0126] This application also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the intelligent evaluation method for mineral resource exploration.

[0127] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0128] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0129] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An intelligent assessment method for mineral resource exploration, characterized in that, The method includes: Step 1: Obtain multi-source geological data for the area to be evaluated. The multi-source geological data includes remote sensing physical data, remote sensing image data, drilling data, geological structure data, and geological exploration data. Step 2: Define the data in the remote sensing physical data that corresponds to the sampling point of the geological exploration data as the first remote sensing physical data, and define the data that does not correspond to the sampling point as the second remote sensing physical data. Construct a preset model based on the geological exploration data and the first remote sensing physical data, input the second remote sensing physical data into the preset model, and obtain the estimated geological exploration data at the virtual sampling point in the area to be evaluated. Step 3: Add the estimated geological exploration data to the multi-source geological data to generate the first multi-source geological data, and delete the second remote sensing physical data from the multi-source geological data to generate the second multi-source geological data; Step 4: Based on the first multi-source geological data, generate the first geological spatial structure of the area to be evaluated using the first preset geological modeling method; based on the second multi-source geological data, generate N1 second geological spatial structures using the second preset geological modeling method; and select the spatial structure that is closest to the first geological spatial structure from all the second geological spatial structures as the reference geological spatial structure. Step 5: Analyze the reference geological spatial structure to obtain the average mineral abundance and ore volume, and calculate the predicted mineral resource reserves of the area to be evaluated based on the average mineral abundance and ore volume. Step 6: Assess the predicted reserves of the mineral resources and obtain the final assessment results; Step 2 includes: Step 21: Serialize the geological exploration data based on the measurement values ​​of specific elements, set a boundary point, and divide the serialized geological exploration data into N2 first data sets based on the boundary point. Step 22: Based on the sampling points corresponding to the geological exploration data in each first data set, the first remote sensing physical data is divided into N2 first data sets to generate a second data set. After traversing all the first data sets, N2 second data sets are obtained. Step 23: Train N2 second data sets to generate a first model for identifying the data set to which the first remote sensing physical data belongs. At the same time, train each second data set separately to construct the preset model corresponding to each second data set for predicting the geological exploration data based on the first remote sensing physical data. Step 24: Use the first model to divide the second remote sensing physical data into different data sets, which are defined as belonging data sets. Extract any belonging data set, input the data in any belonging data set into the corresponding preset model, and obtain the estimated geological exploration data corresponding to each data in any belonging data set. Step 2 is followed by: Step 11: Extract all measured values ​​corresponding to specific elements from the geological exploration data, and delete the values ​​that are less than the first preset value from all the measured values ​​to generate a first set of values ​​to be analyzed; Step 12: Obtain the content range and average content of the specific element in the Earth's crust in the background area, and set a benchmark value based on the content range, wherein the area to be evaluated is included within the background area; Step 13: Delete the values ​​greater than the benchmark value from the first set of values ​​to be analyzed, generate a second set of values ​​to be analyzed, use the average content value as the average value of the second set of values ​​to be analyzed, and determine whether the second set of values ​​to be analyzed meets the preset distribution law. If not, proceed to step 14; if yes, proceed to step 15. Step 14: Reduce the benchmark value by a second preset value to generate a new benchmark value, and define the second set of values ​​to be analyzed as the first set of values ​​to be analyzed, then return to step 13; Step 15: Define the maximum value of the second set of values ​​to be analyzed as the critical value of the specific element; Step 4 includes: Step 41: Extract features from the first feature parameters of each second geological spatial structure, reduce the total number of the first feature parameters to a preset number, and obtain the low-dimensional spatial structure corresponding to each second geological spatial structure. Step 42: Based on preset rules, assign values ​​to the non-numerical parameters in each low-dimensional spatial structure, then standardize the second feature parameters of all low-dimensional spatial structures, calculate the Euclidean distance between any two low-dimensional spatial structures based on the standardization results, and group all second geological spatial structures based on the Euclidean distance to obtain N3 spatial structure sets. Step 43: Extract the geological features of the ore deposits from all geological spatial structures, and obtain the first ore body attribute parameters and the second ore body attribute parameters corresponding to the first geological spatial structure and each second geological spatial structure respectively; Step 44: Extract any set of spatial structures, perform statistical analysis on the second ore body attribute parameters of all second geological spatial structures in any set of spatial structures, and obtain the third ore body attribute parameters of any set of spatial structures. Step 45: After traversing all spatial structure sets, the spatial structure set corresponding to the third ore body attribute parameter with the highest similarity to the first ore body attribute parameter is defined as a specific structure set. Mineral resource assessment is performed on the first geological spatial structure and each third geological spatial structure to obtain the first mineral resource statistical parameter and the second mineral resource statistical parameter. The third geological spatial structure is the second geological spatial structure included in the specific structure set. Step 46: Define the third geological spatial structure corresponding to the second mineral resource statistical parameter that has the highest similarity to the first mineral resource statistical parameter as the reference geological spatial structure.

2. The intelligent assessment method for mineral resource exploration according to claim 1, characterized in that, In step 21, the method for setting the boundary point is as follows: Step 211: Execute a predetermined algorithm on the measured values ​​corresponding to the specific element to generate transformed values, identify the quartiles of all transformed values, extract the transformed values ​​within a first preset range before and after the quartiles, and generate a reference data set. Step 212: Fit the data in the reference data set based on a preset mathematical method to generate a fitting line, and define the value corresponding to each sampling point on the fitting line as the fitting value. Step 213: Calculate the difference between the transformed value and the fitted value for each sampling point in the reference data set, and calculate the average of the squares of all differences. Define the square root of the average value as the dispersion value. Step 214: Set the fluctuation range based on the quantile and the dispersion value, set the measured value corresponding to the largest change value in the fluctuation range as the first boundary point, and set the measured value corresponding to the smallest change value in the fluctuation range as the second boundary point.

3. The intelligent assessment method for mineral resource exploration according to claim 1, characterized in that, Step 3 includes: When the measured value of the specific element at any sampling point is less than the threshold value, the geological exploration data and the first remote sensing physical data corresponding to any sampling point are defined as first redundant data. The first redundant data is deleted from the second multi-source geological data to generate new second multi-source geological data. When the measured value of a specific element at any virtual sampling point is less than the critical value, the estimated geological exploration data and the second remote sensing physical data corresponding to any virtual sampling point are defined as second redundant data. The first redundant data and the second redundant data are deleted from the first multi-source geological data to generate new first multi-source geological data.

4. The intelligent assessment method for mineral resource exploration according to claim 1, characterized in that, Step 42 further includes: Calculate the ratio of the total quantity to N3. When the ratio is greater than the second preset range, divide the area to be evaluated into N3 sub-evaluation areas. Divide the first multi-source geological data and the second multi-source geological data into different sub-evaluation areas. Then return to step 4 and perform mineral reserve prediction for each sub-evaluation area. When the ratio is less than the second preset range, a new sampling point is generated based on the existing sampling point coordinates, and geological data is collected from the new sampling point. After the collection is completed, the process returns to step 1.

5. An intelligent assessment system for mineral resource exploration, used to implement the method as described in any one of claims 1 to 4, characterized in that, The system includes: The data acquisition module is used to acquire multi-source geological data of the area to be evaluated, including remote sensing physical data, remote sensing image data, drilling data, geological structure data, and geological exploration data. The data filling module is used to define the data in the remote sensing physical data that corresponds to the sampling point of the geological exploration data as the first remote sensing physical data, and the data that does not correspond to the sampling point as the second remote sensing physical data. Based on the geological exploration data and the first remote sensing physical data, a preset model is constructed, and the second remote sensing physical data is input into the preset model to obtain the estimated geological exploration data at the virtual sampling point in the area to be evaluated. The data grouping module is used to add the estimated geological exploration data to the multi-source geological data to generate the first multi-source geological data, and to delete the second remote sensing physical data from the multi-source geological data to generate the second multi-source geological data. The spatial structure generation module is used to generate a first geological spatial structure of the area to be evaluated based on the first multi-source geological data using a first preset geological modeling method, generate N1 second geological spatial structures based on the second multi-source geological data using a second preset geological modeling method, and select the spatial structure that is closest to the first geological spatial structure from all the second geological spatial structures as a reference geological spatial structure. The reserve prediction module is used to analyze the reference geological spatial structure, obtain the average mineral abundance and ore volume, and calculate the predicted mineral resource reserves of the area to be evaluated based on the average mineral abundance and ore volume. The intelligent assessment module is used to assess the predicted reserves of the mineral resources and obtain the final assessment results.

6. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the intelligent evaluation method for mineral resource exploration as described in any one of claims 1-4.

Images