A method and system for determining distribution parameters of graphite mining areas

Abstract

The present application relates to the field of artificial intelligence technology, and in particular provides a method and system for determining the distribution parameters of graphite mining areas. By introducing a multi-dimensional mining area status assessment vector, a variety of geological characteristics and their interrelationships within the mining area are deeply analyzed, thereby achieving an accurate prediction of resource distribution. In addition, an exploration and production linkage analysis component is innovatively adopted to closely combine geological data with actual mining needs, further improving the practicality and accuracy of the prediction. In this way, various problems existing in traditional resource prediction methods can be solved, providing strong technical support for the sustainable development of graphite mining areas.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method and system for determining distribution parameters of a graphite mining area. Background Art

[0002] In the process of exploration and mining of graphite mining areas, accurate prediction of resource distribution has always been a technical challenge in the industry. Traditional resource assessment methods are often based on single geological data or simple data models, which are difficult to fully reflect the complex geological conditions and resource distribution of the mining area. This leads to frequent problems such as resource waste, high mining costs and even safety accidents in the actual mining process. Summary of the invention

[0003] In order to improve the above problems, the present application provides a method and system for determining distribution parameters of a graphite mining area.

[0004] The present application embodiment provides a graphite mining area distribution parameter determination method, which is applied to a graphite mining area distribution parameter determination system, and the method includes:

[0005] Determine a multi-dimensional mining area state assessment vector corresponding to a graphite mining area sampling data set;

[0006] Through the target exploration and production linkage analysis component in the target resource distribution prediction network, according to several geological distribution relationship vector clusters and the multi-dimensional mining area status assessment vector, the associated geological distribution relationship vectors corresponding to each geological distribution relationship vector in the several geological distribution relationship vector clusters are determined; the several geological distribution relationship vector clusters are obtained by clustering the various to-be-processed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component; the to-be-processed geological distribution relationship vectors are determined according to the original geological distribution relationship vectors, or are sampled from various associated geological distribution relationship vectors output by the previous-order exploration and production linkage analysis component of the target exploration and production linkage analysis component according to the mining feasibility judgment requirements, and the target exploration and production linkage analysis component is a first-order exploration and production linkage analysis component in the target resource distribution prediction network;

[0007] Based on each associated geological distribution relationship vector output by the last-order exploration and production linkage analysis component in the target resource distribution prediction network, the resource distribution prediction viewpoint corresponding to the graphite mining area sampling data set is determined.

[0008] Optionally, when the target exploration and production linkage analysis component is a non-first-order exploration and production linkage analysis component in the target resource distribution prediction network, each to-be-processed geological distribution relationship vector corresponding to the target exploration and production linkage analysis component is determined based on the following steps:

[0009] Based on each associated geological distribution relationship vector output by the previous exploration and production linkage analysis component, determine the resource distribution prediction viewpoint corresponding to each associated geological distribution relationship vector; the resource distribution prediction viewpoint includes the confidence weight corresponding to the ore resource indicated by the associated geological distribution relationship vector;

[0010] According to the confidence weights in the resource distribution prediction viewpoints corresponding to each of the associated geological distribution relationship vectors, an associated geological distribution relationship vector that meets the mining feasibility judgment requirements is selected from the associated geological distribution relationship vectors as the geological distribution relationship vector to be processed corresponding to the target exploration and production linkage analysis component.

[0011] Optionally, the input of the previous-order exploration-production linkage analysis component includes a plurality of geological distribution relationship vector clusters obtained by clustering the geological distribution relationship vectors to be processed corresponding to the previous-order exploration-production linkage analysis component; the output of the previous-order exploration-production linkage analysis component includes associated geological distribution relationship vector clusters corresponding to each of the plurality of geological distribution relationship vector clusters;

[0012] The method of selecting, according to the confidence weights in the resource distribution prediction viewpoints corresponding to the associated geological distribution relationship vectors, the associated geological distribution relationship vectors that meet the mining feasibility determination requirements from the associated geological distribution relationship vectors as the geological distribution relationship vector to be processed corresponding to the target exploration and production linkage analysis component includes:

[0013] For each of the associated geological distribution relationship vector clusters, based on the confidence weights in the resource distribution prediction viewpoints corresponding to each of the associated geological distribution relationship vectors in the associated geological distribution relationship vector cluster, an associated geological distribution relationship vector that meets the mining feasibility judgment requirements is selected from the associated geological distribution relationship vector cluster as the geological distribution relationship vector to be processed corresponding to the target exploration and production linkage analysis component.

[0014] Optionally, when the target exploration and production linkage analysis component is a first-order exploration and production linkage analysis component in the target resource distribution prediction network, each to-be-processed geological distribution relationship vector corresponding to the target exploration and production linkage analysis component is determined based on the following steps:

[0015] Using each of the original geological distribution relationship vectors as each of the to-be-processed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component;

[0016] Alternatively, determine the rock formation structural stability characteristics corresponding to each of the original geological distribution relationship vectors; and select the original geological distribution relationship vector that meets the set stability index from each of the original geological distribution relationship vectors according to the rock formation structural stability characteristics corresponding to each of the original geological distribution relationship vectors as the geological distribution relationship vector to be processed corresponding to the target exploration and production linkage analysis component.

[0017] Optionally, the geological distribution relationship vector cluster is obtained according to any of the following methods:

[0018] According to the set number of clusters, each to-be-processed geological distribution relationship vector corresponding to the target exploration and production linkage analysis component is divided into the plurality of geological distribution relationship vector clusters;

[0019] Determine the rock formation structural stability characteristics corresponding to each of the to-be-processed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component;

[0020] According to the set number of clusters, and based on the rock layer structure stability characteristics corresponding to each of the geological distribution relationship vectors to be processed, each of the geological distribution relationship vectors to be processed is transferred to the plurality of geological distribution relationship vector clusters;

[0021] Determine the positioning labels of each unprocessed geological distribution relationship vector corresponding to the target exploration and production linkage analysis component in the three-dimensional vector relationship network; according to the set number of clusters and the positioning labels of each unprocessed geological distribution relationship vector in the three-dimensional vector relationship network, transfer each of the unprocessed geological distribution relationship vectors to the several geological distribution relationship vector clusters.

[0022] Optionally, the target exploration and production linkage analysis component in the target resource distribution prediction network determines, according to the plurality of geological distribution relationship vector clusters and the multi-dimensional mining area status assessment vector, the associated geological distribution relationship vectors corresponding to the respective geological distribution relationship vectors in the plurality of geological distribution relationship vector clusters, including:

[0023] By using the target exploration and production linkage analysis component, a first feature focusing rule is adopted to determine, according to each of the geological distribution relationship vector clusters, a transient geological distribution relationship vector cluster corresponding to each of the geological distribution relationship vector clusters; the transient geological distribution relationship vector cluster includes the transient geological distribution relationship vectors corresponding to each of the geological distribution relationship vectors in the corresponding geological distribution relationship vector cluster;

[0024] Through the target exploration and production linkage analysis component, the second feature focusing rule is adopted to determine the associated geological distribution relationship vector cluster corresponding to each transient geological distribution relationship vector cluster and the multidimensional mining area status assessment vector; the associated geological distribution relationship vector cluster includes the associated geological distribution relationship vectors corresponding to each transient geological distribution relationship vector in the corresponding transient geological distribution relationship vector cluster.

[0025] Optionally, each order of exploration and production linkage analysis components in the target resource distribution prediction network are the target exploration and production linkage analysis components;

[0026] Alternatively, the first u-order exploration and production linkage analysis components in the target resource distribution prediction network are the target exploration and production linkage analysis components; and u is an integer not less than 1 and less than the total order of exploration and production linkage analysis components in the target resource distribution prediction network.

[0027] Optionally, the method further includes:

[0028] Determine a multi-dimensional mining area historical state assessment vector corresponding to a graphite mining area historical sampling data set;

[0029] Determine, through the target exploration and production linkage analysis component in the target resource distribution prediction network, the historical associated geological distribution relationship vectors corresponding to each of the historical geological distribution relationship vectors in the plurality of historical geological distribution relationship vector clusters according to the plurality of historical geological distribution relationship vector clusters and the multi-dimensional mining area historical state assessment vector;

[0030] The several clusters of historical geological distribution relationship vectors are obtained by clustering the various to-be-processed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component; the to-be-processed geological distribution relationship vectors are determined according to the original geological distribution relationship vector samples, or are sampled from various historically associated geological distribution relationship vectors output by the previous-order exploration and production linkage analysis component of the target exploration and production linkage analysis component according to the mining feasibility judgment requirements; the target exploration and production linkage analysis component is a first-order exploration and production linkage analysis component in the target resource distribution prediction network;

[0031] For each of the historical geological distribution relationship vector clusters, determining the training error corresponding to the historical geological distribution relationship vector cluster according to the historical associated geological distribution relationship vectors corresponding to each of the historical geological distribution relationship vectors and the debugging annotations corresponding to the historical sampling data set of the graphite mining area;

[0032] Determining the training error corresponding to the target exploration and production linkage analysis component based on the training errors corresponding to each of the plurality of historical geological distribution relationship vector clusters;

[0033] The target resource distribution prediction network is debugged according to the training errors corresponding to the exploration and production linkage analysis components of each order in the target resource distribution prediction network.

[0034] Optionally, determining the training error corresponding to the cluster of historical geological distribution relationship vectors according to the historical associated geological distribution relationship vectors corresponding to each of the historical geological distribution relationship vectors and the debugging annotations corresponding to the historical sampling data set of the graphite mining area includes:

[0035] Determine the resource distribution training viewpoints corresponding to each of the historical geological distribution relationship vectors based on the historical geological distribution relationship vectors corresponding to each of the historical geological distribution relationship vectors in the historical geological distribution relationship vector cluster;

[0036] Using a deep decision tree model, according to each of the resource distribution training viewpoints and the debugging annotations, a consistency analysis label corresponding to the historical geological distribution relationship vector cluster is determined; the consistency analysis label is used to indicate the consistency analysis method between each of the resource distribution training viewpoints and the debugging annotations;

[0037] The training error corresponding to the historical geological distribution relationship vector cluster is determined based on the consistency analysis label.

[0038] Optionally, the method further comprises: determining a quantitative relationship between the debugging annotation and the resource distribution training viewpoint based on a current network quality evaluation feature of the target resource distribution prediction network;

[0039] The method of using a deep decision tree model to determine the consistency analysis label corresponding to the historical geological distribution relationship vector cluster according to each of the resource distribution training viewpoints and the debugging annotations includes: using the deep decision tree model to determine the consistency analysis label corresponding to the historical geological distribution relationship vector cluster according to the quantitative relationship and each of the resource distribution training viewpoints and the debugging annotations.

[0040] An embodiment of the present application provides a graphite mining area distribution parameter determination system, comprising at least one processor and a memory; the memory stores computer execution instructions; the at least one processor executes the computer execution instructions stored in the memory, so that the at least one processor executes the above method.

[0041] An embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. The computer program implements the above method when running.

[0042] In the embodiment of the present application, by introducing a multi-dimensional mining area status assessment vector, a variety of geological characteristics and their interrelationships in the mining area are deeply analyzed, thereby achieving an accurate prediction of resource distribution. In addition, an exploration and production linkage analysis component is innovatively adopted to closely combine geological data with actual mining needs, further improving the practicality and accuracy of the prediction. In this way, various problems existing in traditional resource prediction methods can be solved, providing strong technical support for the sustainable development of graphite mining areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A flow chart of a method for determining graphite mining area distribution parameters provided in an embodiment of the present application.

[0044] Figure 2 A structural schematic diagram of a graphite mining area distribution parameter determination system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to better understand the above technical scheme, the technical scheme of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical scheme of the present application, rather than limitations on the technical scheme of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0046] Figure 1 A graphite mining area distribution parameter determination method is shown, which is applied to a graphite mining area distribution parameter determination system. The method includes S101-S103.

[0047] S101, determining a multidimensional mining area state assessment vector corresponding to a graphite mining area sampling data set.

[0048] In the embodiments of the present application, the graphite mining area sampling data set refers to a series of second-order positioning parameter data collected during exploration and sampling work in the graphite mining area. These data usually cover various aspects of geological, geophysical, geochemical and other information of the mining area. These data sets are obtained by field sampling of rocks, soil, water samples and the like at different locations in the mining area and laboratory analysis. The quality and integrity of the sampling data set are crucial for subsequent resource assessment and mining planning.

[0049] Furthermore, the multidimensional mining area status assessment vector is a comprehensive numerical feature vector used to comprehensively describe and assess the overall status of the graphite mining area. This vector consists of multiple dimensions (or features), each of which represents a certain aspect of the characteristics or status of the mining area. For example, these dimensions may include but are not limited to:

[0050] Geological structural characteristics: describe the geological conditions of the mining area, such as strata, structures, faults, etc., and express their complexity and stability through numerical representation;

[0051] Ore body morphological parameters: quantify the shape, size, thickness, etc. of the ore body to evaluate the mineability and economic value of the ore body;

[0052] Ore grade index: indicates the content and quality of graphite in the ore, and is a key factor in measuring the value of mining resources;

[0053] Environmental impact factors: Consider the ecological environment and hydrogeological conditions around the mining area to assess the potential impact of mining activities on the environment;

[0054] Mining technical conditions: reflect the accessibility of the mining area, the difficulty of mining, and the required technology and equipment conditions.

[0055] Each of the above dimensions will be assigned a value according to specific data and evaluation criteria, and these values ​​are combined to form a multidimensional mining area status assessment vector. The multidimensional mining area status assessment vector not only contains a variety of information about the mining area, but also can reveal the internal connections and laws between these information through numerical analysis methods such as clustering, classification, and regression analysis, thereby providing a scientific basis for the exploration and development of the mining area.

[0056] In detail, during the exploration and resource assessment of graphite mining areas, it is first necessary to conduct comprehensive sampling and data collection work in the mining area. This step includes collecting rock, soil and water samples at different locations and depths in the mining area, and obtaining various geological and chemical data through laboratory analysis. These data constitute the graphite mining area sampling data set, which is the basis for subsequent assessment work. Next, based on the sampling data set, a multidimensional mining area status assessment vector needs to be constructed. The construction of this vector is a systematic process, which involves the quantification and integration of multiple key indicators. Specifically, each dimension will be evaluated in detail numerically based on multiple dimensions such as geological structure, ore body morphology, ore grade, environmental impact and mining technical conditions. For example, for the geological structure dimension, it can be assigned according to factors such as the stability of the stratum, the distribution and activity of the fault; for the ore body morphology dimension, it can be quantified according to parameters such as the thickness, length and width of the ore body. Through the comprehensive evaluation of these dimensions, a multidimensional mining area status assessment vector that fully reflects the state of the graphite mining area can be obtained. The multidimensional mining area status assessment vector not only provides an overall overview of the mining area, but also provides important data support and decision-making basis for subsequent exploration and development work. For example, the multidimensional mining area status assessment vector can be used to identify the most promising mining area in the mining area, or to evaluate the feasibility and risks of different mining plans.

[0057] S102. Determine, through the target exploration and production linkage analysis component in the target resource distribution prediction network, the associated geological distribution relationship vectors corresponding to each geological distribution relationship vector in the plurality of geological distribution relationship vector clusters according to the plurality of geological distribution relationship vector clusters and the multi-dimensional mining area status assessment vector.

[0058] Among them, the several geological distribution relationship vector clusters are obtained by clustering the various geological distribution relationship vectors to be processed corresponding to the target exploration and production linkage analysis component; the geological distribution relationship vectors to be processed are determined according to the original geological distribution relationship vectors, or are sampled from various associated geological distribution relationship vectors output by the previous-order exploration and production linkage analysis component of the target exploration and production linkage analysis component according to the mining feasibility judgment requirements, and the target exploration and production linkage analysis component is a first-order exploration and production linkage analysis component in the target resource distribution prediction network.

[0059] In the embodiment of the present application, the target resource distribution prediction network is a neural network model designed to predict the distribution of mineral resources. It learns and simulates the distribution law of resources in the mining area by analyzing the multi-dimensional data of the mining area, such as geological structure, ore body characteristics, ore grade, etc. The network may contain multiple levels and nodes, each of which is responsible for processing and analyzing specific data features, and finally outputs a comprehensive resource distribution prediction.

[0060] The target exploration and production linkage analysis component is a key component in the target resource distribution prediction network. Its main function is to realize the linkage analysis of exploration and production data. This component can receive and process data such as multi-dimensional mining area status assessment vectors and geological distribution relationship vectors, and predict the resource distribution of the mining area by analyzing the inherent connection and influence between these data. At the same time, it also takes into account the actual needs of exploration and production to ensure that the prediction results are closer to the actual situation.

[0061] The associated geological distribution relationship vector refers to the vector representation associated with specific geological features or ore body distribution after being processed by the target exploration and production linkage analysis component. These vectors capture the inherent connection between different geological features and their possible impact on resource distribution. By analyzing these associated vectors, the distribution of resources in the mining area can be predicted more accurately.

[0062] The mining feasibility judgment requirements refer to a series of conditions and standards that need to be met before mining of mineral resources. These requirements usually include the assessment of the scale, grade, mineability, safety and environmental impact of the ore body. Only ore bodies that meet these judgment requirements will be identified as ore bodies with mining value, and then subsequent mining work will be carried out.

[0063] In the process of determining the resource distribution of graphite mining areas, S102 is a key link. This step mainly uses the target exploration and production linkage analysis component in the target resource distribution prediction network to conduct in-depth analysis of the collected multi-dimensional mining area status evaluation vectors and geological distribution relationship vector clusters. First, the target exploration and production linkage analysis component will receive and process several geological distribution relationship vector clusters. These vector clusters are obtained by processing the original geological distribution relationship vectors through clustering algorithms, and they represent different geological characteristics and ore body distribution patterns in the mining area. Then, the component will perform correlation analysis on these geological distribution relationship vector clusters and multi-dimensional mining area status evaluation vectors. The multi-dimensional mining area status evaluation vector contains the overall status information of the mining area, such as geological structure, ore body morphology, ore grade, etc., which is crucial for accurately predicting resource distribution. In the correlation analysis process, the target exploration and production linkage analysis component will use advanced algorithms and technologies to find the intrinsic connection between each geological distribution relationship vector and the multi-dimensional mining area status evaluation vector. This connection may be manifested as the direct impact of a certain geological feature on resource distribution, or the indirect impact of the interaction between multiple geological features on resource distribution. Finally, through in-depth analysis and calculation, the target exploration and production linkage analysis component can determine the associated geological distribution relationship vectors corresponding to each geological distribution relationship vector. These association vectors not only contain information about the original geological characteristics, but also incorporate the influence of the overall state of the mining area, so they can more accurately reflect the actual distribution of resources in the mining area. In addition, it is worth mentioning that the target exploration and production linkage analysis component will also consider the requirements for mining feasibility judgment during the analysis process. This means that it not only pays attention to the geological characteristics and resource distribution, but also analyzes in combination with actual mining conditions and feasibility. This comprehensive consideration makes the prediction results closer to the actual situation and provides strong guidance for subsequent mining work.

[0064] S103, determining the resource distribution prediction viewpoint corresponding to the graphite mining area sampling data set based on each associated geological distribution relationship vector output by the last-order exploration and production linkage analysis component in the target resource distribution prediction network.

[0065] In the embodiments of the present application, the resource distribution prediction viewpoint refers to a comprehensive and accurate predictive description of the resource distribution in the graphite mining area based on multi-dimensional data collection, processing and analysis. It not only involves the quantity, grade and quality of resources, but also takes into account factors such as the spatial distribution of resources, the difficulty of mining, and possible economic benefits. This prediction viewpoint is the conclusion drawn by the graphite mining area distribution parameter determination system after integrating all relevant information, and provides a scientific basis for further exploration, development planning and production decisions in the mining area.

[0066] In the final stage of the resource distribution determination of the graphite mining area, S103 plays a vital role. The core task of this step is to use the associated geological distribution relationship vectors output by the last-order exploration and production linkage analysis component in the target resource distribution prediction network to form a comprehensive prediction view of the resource distribution of the graphite mining area. First of all, it should be clear that these associated geological distribution relationship vectors are obtained after multiple rounds of exploration and production linkage analysis and data processing, and they contain rich geological information and resource distribution characteristics. These vectors not only reflect the geological structure and ore body morphology of the mining area, but also reflect the ore grade, resource reserves and other information. In S103, the graphite mining area distribution parameter determination system uses these associated geological distribution relationship vectors as input and analyzes and calculates them through a series of complex algorithms and models. In this process, the system can combine historical data, expert knowledge and industry experience to ensure the accuracy and reliability of the prediction results. Finally, the system forms a comprehensive resource distribution prediction view based on these analysis results. This view not only includes the quantity, grade and spatial distribution of graphite resources, but also may involve the exploitability, economic benefits and potential risk factors of resources. Such a prediction viewpoint has extremely important guiding significance for the subsequent development planning, production decision-making and risk management of the mining area. Therefore, S103 is a key link in the process of determining the resource distribution of graphite mining areas. It provides a scientific decision-making basis for the sustainable development of mining areas by combining multi-dimensional geological information and data analysis.

[0067] In the technical solution described in S101-S103 above, the graphite mine area distribution parameter determination system first performs a key step when performing detailed resource distribution determination on the graphite mine area, namely, determining the multi-dimensional mine area status assessment vector corresponding to the graphite mine area sampling data set. This step is to collect a variety of data in the mine area, such as geological structure, ore body morphology, ore grade, etc., and then use these data to construct a multi-dimensional status assessment vector. This vector can fully reflect the overall status of the mine area and provide important basic information for subsequent resource distribution prediction.

[0068] Next, the graphite mine area distribution parameter determination system conducts in-depth analysis of several geological distribution relationship vector clusters and multi-dimensional mine area status assessment vectors through the target exploration and production linkage analysis component in its built-in target resource distribution prediction network. These geological distribution relationship vector clusters are obtained by clustering the geological distribution relationship vectors to be processed, and they represent different geological characteristics and ore body distribution conditions in the mine area. The graphite mine area distribution parameter determination system uses these vector clusters and the previously determined multi-dimensional mine area status assessment vectors to further determine the associated geological distribution relationship vectors corresponding to each geological distribution relationship vector. The purpose of this step is to find out the intrinsic connection between various geological characteristics and their possible impact on the distribution of graphite resources.

[0069] In the process of determining the associated geological distribution relationship vector, the graphite mine area distribution parameter determination system also takes into account the linkage between exploration and production. This means that the graphite mine area distribution parameter determination system not only analyzes the current geological data, but also combines the actual mining conditions and feasibility, thereby ensuring that the obtained analysis results are closer to the actual situation and can provide effective guidance for subsequent mining work.

[0070] Finally, the graphite mining area distribution parameter determination system determines the resource distribution prediction viewpoint corresponding to the graphite mining area sampling data set based on the various associated geological distribution relationship vectors output by the last-order exploration and production linkage analysis component in the target resource distribution prediction network, and integrates all the collected and analyzed information. This prediction viewpoint is a comprehensive and accurate description of the distribution of graphite resources in the mining area. It not only includes the quantity and grade of resources, but also considers the feasibility and economy of mining, providing strong support for the development planning and production decision-making of the mining area.

[0071] In this way, the graphite mine distribution parameter measurement system can efficiently and accurately predict the resource distribution of graphite mines through multi-dimensional data collection and analysis, combined with the actual needs of exploration and production, providing an important scientific basis for the sustainable development of the mine.

[0072] It is worth mentioning that in the process of determining the resource distribution of graphite mining areas, several key geological distribution relationship vectors are involved, namely the geological distribution relationship vector to be processed, the original geological distribution relationship vector and the associated geological distribution relationship vector. These vectors play different roles in the determination process and together form the basis for resource distribution prediction.

[0073] In detail, the original geological distribution relationship vector is a vector converted from the original data directly collected from the graphite mining area. These data usually include the geological structure information, ore body morphology, ore grade, etc. of the mining area. For example, an original geological distribution relationship vector may be expressed as: [ore body thickness, ore grade, fault zone density, rock layer dip, ...]. A specific numerical feature vector example can be: [5 meters, 98% carbon content, 3 lines / square kilometer, 45 degrees, ...]. These original data are the basis for subsequent analysis and prediction.

[0074] The unprocessed geological distribution relationship vector is a data vector obtained after preliminary processing or screening based on the original geological distribution relationship vector. This step of processing may include data cleaning, standardization, and normalization. In some cases, the unprocessed geological distribution relationship vector may also be sampled directly from the original data to meet specific analysis needs. For example, if researchers are interested in ores within a specific grade range, they may screen out ore data that meets this grade range from the original data and convert it into a unprocessed geological distribution relationship vector.

[0075] The associated geological distribution relationship vectors are obtained through the exploration and production linkage analysis component in the target resource distribution prediction network. These vectors not only take into account the original geological characteristics, but also combine the actual needs and conditions of exploration and production. For example, an associated geological distribution relationship vector may contain geological information related to specific mining technology or economic indicators. The generation of these vectors usually involves complex algorithms and models that can reveal the intrinsic connection between different geological characteristics and the impact of these characteristics on the distribution of graphite resources and the feasibility of mining.

[0076] In terms of specific values, the associated geological distribution relationship vector may be expressed as a series of weighted or transformed values ​​that reflect the comprehensive impact of various geological factors and production conditions on resource distribution. For example, an associated geological distribution relationship vector may be similar to: [0.85 (representing the weight of a certain geological feature), 0.6 (representing the weight of another geological feature), ...].

[0077] Therefore, the processed geological distribution relationship vector, the original geological distribution relationship vector and the associated geological distribution relationship vector play a vital role in the determination of graphite mining area resource distribution. They not only provide a rich data basis for prediction, but also provide a scientific basis for exploration and production decision-making in mining areas by revealing the intrinsic connection and influence between various geological features.

[0078] In summary, the embodiment of the present application significantly improves the accuracy and efficiency of resource assessment and mining planning in the mining area by constructing a comprehensive and sophisticated graphite mining area resource distribution prediction system. First, by determining the multidimensional mining area state assessment vector corresponding to the graphite mining area sampling data set, the system can capture the geological, ore body and ore characteristics of the mining area in an all-round way, laying a solid foundation for subsequent resource prediction. Secondly, with the help of the exploration and production linkage analysis component in the target resource distribution prediction network, not only a single geological feature is considered, but also the intrinsic connection between the geological distribution relationship vectors is further excavated, so as to accurately determine the associated geological distribution relationship vectors corresponding to each geological distribution relationship vector. This innovative analysis method effectively integrates multidimensional data and actual production needs, making resource prediction closer to the actual situation. Finally, by integrating the associated geological distribution relationship vectors output by the last-order exploration and production linkage analysis component, a comprehensive and accurate resource distribution prediction viewpoint can be formed, which provides a strong scientific basis for exploration, mining and production decisions in the mining area. In general, it not only improves the accuracy of graphite resource prediction, but also greatly enhances the feasibility and economic benefits of mining in the mining area.

[0079] In some optional embodiments, when the target exploration and production linkage analysis component is a non-first-order exploration and production linkage analysis component in the target resource distribution prediction network, each to-be-processed geological distribution relationship vector corresponding to the target exploration and production linkage analysis component is determined based on the following steps: based on each associated geological distribution relationship vector output by the previous-order exploration and production linkage analysis component, determining the resource distribution prediction viewpoint corresponding to each of the associated geological distribution relationship vectors; the resource distribution prediction viewpoint includes the confidence weight corresponding to the ore resource indicated by the associated geological distribution relationship vector; according to the confidence weight in the resource distribution prediction viewpoint corresponding to each of the associated geological distribution relationship vectors, selecting the associated geological distribution relationship vector that meets the mining feasibility judgment requirements from each of the associated geological distribution relationship vectors as the to-be-processed geological distribution relationship vector corresponding to the target exploration and production linkage analysis component.

[0080] In this embodiment, the linkage analysis of exploration and production is a complex multi-stage process involving the collaboration of multiple linkage analysis components of exploration and production. These components are arranged in stages in the target resource distribution prediction network, and each order component performs more in-depth processing based on the analysis results of its previous order component.

[0081] On the basis that the target exploration and production linkage analysis component is not the first-order component in the network, it means that before the target exploration and production linkage analysis component, at least one exploration and production linkage analysis component has performed preliminary processing and analysis on the geological data.

[0082] First, the first-order exploration and production linkage analysis component will output a series of associated geological distribution relationship vectors. These vectors are generated based on the original geological data and preliminary analysis results, and they describe the relationship between different geological features and ore resource distribution in the mining area. Each associated geological distribution relationship vector will have a corresponding resource distribution prediction view.

[0083] The resource distribution prediction view is a comprehensive assessment that not only includes the possible distribution of ore resources indicated by the associated geological distribution relationship vector, but also gives the confidence weight of this distribution. The confidence weight is an important parameter that reflects the reliability and accuracy of the prediction view. A high confidence weight means that the prediction view is more credible, while a low confidence weight may indicate that there is greater uncertainty in the prediction.

[0084] Next, the graphite mining area distribution parameter determination system screens all the associated geological distribution relationship vectors based on these confidence weights. The purpose of this step is to select those vectors that conform to the geological distribution law and meet the requirements for mining feasibility determination. Specifically, the system sets a confidence weight threshold. Only when the confidence weight of the vector exceeds this threshold will it be selected as the geological distribution relationship vector to be processed by the next-level exploration and production linkage analysis component.

[0085] In this way, the graphite mine distribution parameter determination system can gradually screen out the geological areas with the greatest mining potential, while excluding those areas with high uncertainty or high mining difficulty. This not only improves the accuracy of resource prediction, but also provides strong decision-making support for the actual mining of the mine.

[0086] With this design, through the collaboration of multi-order exploration and production linkage analysis components, it is possible to deeply explore the hidden information in geological data and accurately predict the distribution of ore resources. In particular, when the target exploration and production linkage analysis component is not a first-order component, the system can use the analysis results of the previous-order component to further screen out high-confidence geological distribution relationship vectors, thereby greatly improving the accuracy of resource prediction and the feasibility of mining. This prediction method of step-by-step screening and gradual refinement not only reduces unnecessary exploration costs, but also provides a scientific decision-making basis for the sustainable development of mining areas.

[0087] In some other possible embodiments, the input of the previous-order exploration and production linkage analysis component includes a number of geological distribution relationship vector clusters obtained by clustering the geological distribution relationship vectors to be processed corresponding to the previous-order exploration and production linkage analysis component; the output of the previous-order exploration and production linkage analysis component includes the associated geological distribution relationship vector clusters corresponding to each of the several geological distribution relationship vector clusters; and the associated geological distribution relationship vectors that meet the mining feasibility judgment requirements are selected from each of the associated geological distribution relationship vectors according to the confidence weights in the resource distribution prediction viewpoints corresponding to each of the associated geological distribution relationship vectors as the geological distribution relationship vector to be processed corresponding to the target exploration and production linkage analysis component, including: for each of the associated geological distribution relationship vector clusters, according to the confidence weights in the resource distribution prediction viewpoints corresponding to each of the associated geological distribution relationship vectors in the associated geological distribution relationship vector clusters, the associated geological distribution relationship vectors that meet the mining feasibility judgment requirements are selected from the associated geological distribution relationship vector clusters as the geological distribution relationship vector to be processed corresponding to the target exploration and production linkage analysis component.

[0088] Based on this embodiment, exploration and production linkage analysis is a refined, multi-level process, and each stage involves in-depth processing and analysis of geological data. This embodiment will elaborate on how to process geological distribution relationship vectors through clustering in the exploration and production linkage analysis process, and screen out key vectors that meet the requirements of mining feasibility.

[0089] First, the first-order exploration and production linkage analysis component receives a series of geological distribution relationship vectors to be processed as input. These vectors are generated based on the original geological survey data and preliminary analysis results, describing the potential relationship between different geological characteristics and graphite resource distribution in the mining area. In order to process these data more effectively, the system first clusters these vectors to form several geological distribution relationship vector clusters. Each cluster represents a specific geological distribution pattern or resource enrichment feature.

[0090] Next, the first-order exploration and production linkage analysis component will conduct in-depth analysis of these vector clusters and output the associated geological distribution relationship vector clusters corresponding to each cluster. These associated vector clusters not only retain the core information of the original vectors, but also incorporate more insights into resource distribution and mining feasibility.

[0091] Then, the graphite mining area distribution parameter determination system conducts further screening based on the resource distribution prediction viewpoints of each vector in each associated geological distribution relationship vector cluster, especially the confidence weight. The confidence weight reflects the accuracy and reliability of the prediction viewpoint and is an important indicator for evaluating the feasibility of mining.

[0092] For each cluster of associated geological distribution relationship vectors, the system selects those associated geological distribution relationship vectors that conform to the geological distribution law and meet the requirements for mining feasibility judgment according to the confidence weight of each vector. These selected vectors will be regarded as the geological distribution relationship vectors to be processed by the target exploration and production linkage analysis component.

[0093] In this way, the graphite mining area distribution parameter determination system can accurately identify the geological areas with the greatest mining potential and economic benefits, providing strong support for subsequent resource development and utilization. In detail, through cluster analysis technology, the complex geological information is effectively integrated, and key geological distribution relationship vector clusters are extracted from it. Furthermore, using the scientific indicator of confidence weight, the system can accurately screen out geological areas that meet the feasibility requirements of mining, greatly improving the efficiency and accuracy of resource exploration. This intelligent data processing and decision support capability not only provides a solid technical guarantee for the sustainable development of graphite mining areas, but also injects new impetus into the transformation, upgrading and green development of the mining industry.

[0094] Under some replaceable technical ideas, when the target exploration and production linkage analysis component is the first-order exploration and production linkage analysis component in the target resource distribution prediction network, the various unprocessed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component are determined based on the following steps: each of the original geological distribution relationship vectors is used as the various unprocessed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component; or, the rock formation structure stability characteristics corresponding to each of the original geological distribution relationship vectors are determined; according to the rock formation structure stability characteristics corresponding to each of the original geological distribution relationship vectors, the original geological distribution relationship vectors that meet the set stability index are selected from each of the original geological distribution relationship vectors as the unprocessed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component.

[0095] In the embodiment of the present application, exploration and production linkage analysis is a core link, especially when it comes to the first-order exploration and production linkage analysis component, the data processing method is particularly important. This embodiment will elaborate on how to determine the geological distribution relationship vector to be processed corresponding to the first-order exploration and production linkage analysis component.

[0096] First, the first-order exploration and production linkage analysis component plays the role of preliminary analysis and screening of geological data in the system. The input it receives is the original geological distribution relationship vector, which is usually data directly obtained through geological surveys, reflecting the original geological characteristics of the mining area and the preliminary distribution of graphite resources.

[0097] In detail, the system can directly use these original geological distribution relationship vectors as the vectors to be processed by the first-order exploration and production linkage analysis component. This method is suitable for situations with high data quality and less noise, and can maximize the preservation of the integrity and authenticity of the original data.

[0098] However, in actual operation, the original geological data often contains a lot of redundant information and noise, and direct use may affect the accuracy of the analysis. Therefore, the system also provides another more sophisticated processing method.

[0099] In this way, the system will first determine the rock structure stability characteristics corresponding to each original geological distribution relationship vector. Rock structure stability is an important geological indicator that reflects the firmness and compressive resistance of the rock layer, which is crucial to the safety of graphite mining.

[0100] Next, the system will select vectors that meet the set stability index from the original geological distribution relationship vectors based on these stability characteristics. This set stability index is usually a threshold. Only when the stability of the rock formation reaches a certain level, the geological area is considered to have the potential for mining.

[0101] The original geological distribution relationship vectors screened in this way will be used as the vectors to be processed in the first-order exploration and production linkage analysis component. This processing method not only removes redundant and noisy data, but also ensures the accuracy and effectiveness of subsequent analysis. In this way, the original geological distribution relationship vectors can be used directly, or they can be screened according to the stability characteristics of the rock structure, ensuring the accuracy and effectiveness of the data. This processing method not only improves the accuracy of the analysis, but also provides reliable data support for subsequent exploration and production activities, thereby helping to improve the mining efficiency and safety of graphite mining areas.

[0102] Under some other preferred technical ideas, the geological distribution relationship vector cluster is obtained according to any of the following methods: according to the set number of clustering clusters, each of the to-be-processed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component is divided into the several geological distribution relationship vector clusters; the rock formation structural stability characteristics corresponding to each of the to-be-processed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component are determined; according to the set number of clustering clusters, each of the to-be-processed geological distribution relationship vectors is transferred to the several geological distribution relationship vector clusters according to the rock formation structural stability characteristics corresponding to each of the to-be-processed geological distribution relationship vectors; the positioning label of each to-be-processed geological distribution relationship vector corresponding to the target exploration and production linkage analysis component in the three-dimensional vector relationship network is determined; according to the set number of clustering clusters, each of the to-be-processed geological distribution relationship vectors is transferred to the several geological distribution relationship vector clusters according to the positioning label of each of the to-be-processed geological distribution relationship vectors in the three-dimensional vector relationship network.

[0103] In this embodiment, the acquisition of geological distribution relationship vector clusters is a key link, which helps the system to understand the geological structure and resource distribution of the mining area more accurately. This embodiment will introduce in detail several preferred technical ideas for acquiring geological distribution relationship vector clusters.

[0104] First, the system can use a direct clustering method. In this method, the system will divide the various geological distribution relationship vectors to be processed corresponding to the target exploration and production linkage analysis component according to the pre-set number of clusters. This division is based on the similarity or distance between vectors, ensuring that the vectors within each cluster have high similarity in geological characteristics, while the vectors between different clusters are quite different. In this way, the system can obtain several clusters of geological distribution relationship vectors with clear geological characteristics.

[0105] Secondly, the system can also consider the characteristics of rock structure stability for clustering. With this idea, the system will first determine the rock structure stability characteristics corresponding to each geological distribution relationship vector to be processed. These characteristics may include the hardness, compressive strength, joint development, etc. of the rock formation. Then, the system will divide the geological distribution relationship vectors to be processed into different clusters based on these characteristics and the set number of clustering clusters. This method not only considers the similarity of geological vectors, but also combines the actual stability of the rock structure, making the clustering results more in line with the actual geological situation.

[0106] Finally, the system can also use the three-dimensional vector relationship network for clustering. In this method, the system will determine a positioning label in the three-dimensional vector relationship network for each geological distribution relationship vector to be processed. This label can be a three-dimensional coordinate, representing the position of the vector in the geological space. Then, based on these positioning labels and the set number of clusters, the system will divide the geological distribution relationship vectors to be processed into different clusters. This method can more intuitively reflect the spatial distribution of geological vectors, which helps the system to understand the geological structure of the mining area more accurately.

[0107] In this way, a variety of flexible and efficient methods are used to obtain geological distribution relationship vector clusters. These methods not only take into account the similarity of geological vectors, but also combine the stability and spatial distribution of rock layer structures, making the clustering results more accurate and comprehensive. This precise geological distribution relationship vector cluster division helps the system to have a deeper understanding of the geological characteristics of the mining area and provide strong data support for subsequent resource exploration and development. At the same time, the flexibility and diversity of these methods also enable the system to adapt to the exploration needs under different geological conditions, improving the versatility and practicality of the system.

[0108] It is worth mentioning that the rock formation structural stability characteristics corresponding to each of the to-be-processed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component are determined, including:

[0109] First, conduct preliminary exploration of the rock structure in the area to be explored by using geological radar or acoustic wave detection equipment to obtain basic physical properties of the rock formation, including but not limited to parameters such as thickness, density, and acoustic wave propagation speed of the rock formation;

[0110] Next, the data processing module in the target exploration and production linkage analysis component is used to pre-process the data obtained from the preliminary exploration, including data cleaning, outlier removal and standardization, to ensure the accuracy and consistency of the data;

[0111] Then, based on the preprocessed data, a machine learning algorithm, such as support vector machine (SVM) or random forest, is used to train a rock structure stability classification model;

[0112] During the model training process, the model is tuned based on the knowledge and experience of geological experts to improve the model's ability to identify the stability characteristics of rock structures;

[0113] After the training is completed, the geological distribution relationship vector to be processed is input into the trained rock formation structure stability classification model to obtain the rock formation structure stability feature labels corresponding to each geological distribution relationship vector to be processed. These labels can represent the stable, relatively stable, unstable and other states of the rock formation structure;

[0114] Finally, according to these labels, the geological distribution relationship vectors to be processed with similar rock structure stability characteristics are classified into one category, thereby providing an accurate basis for the subsequent division of geological distribution relationship vector clusters.

[0115] Through the above technical solution, the rock structure stability characteristics corresponding to each of the geological distribution relationship vectors to be processed corresponding to the target exploration and production linkage analysis component can be accurately determined, thereby improving the accuracy and efficiency of geological exploration.

[0116] In a preferred technical idea, the target exploration and production linkage analysis component in the target resource distribution prediction network determines the associated geological distribution relationship vectors corresponding to each geological distribution relationship vector in the multiple geological distribution relationship vector clusters according to the multiple geological distribution relationship vector clusters and the multidimensional mining area status assessment vector, including: through the target exploration and production linkage analysis component, using the first feature focusing rule, according to each geological distribution relationship vector cluster, determining the transient geological distribution relationship vector cluster corresponding to each geological distribution relationship vector cluster; the transient geological distribution relationship vector cluster includes the transient geological distribution relationship vectors corresponding to each geological distribution relationship vector in the corresponding geological distribution relationship vector cluster; through the target exploration and production linkage analysis component, using the second feature focusing rule, according to each transient geological distribution relationship vector cluster and the multidimensional mining area status assessment vector, determining the associated geological distribution relationship vector cluster corresponding to each transient geological distribution relationship vector cluster; the associated geological distribution relationship vector cluster includes the associated geological distribution relationship vectors corresponding to each transient geological distribution relationship vector in the corresponding transient geological distribution relationship vector cluster.

[0117] In the graphite mining area distribution parameter determination system, determining the associated geological distribution relationship vector is a key step, which helps to more accurately predict the resource distribution of the mining area. Under this preferred technical idea, this task will be completed through the target exploration and production linkage analysis component in the target resource distribution prediction network, combined with several geological distribution relationship vector clusters and multi-dimensional mining area status assessment vectors.

[0118] First, the geological distribution relationship vector cluster is a group of geological distribution relationship vectors with similar geological characteristics obtained through cluster analysis. The multidimensional mining area status assessment vector is a vector that comprehensively assesses various status information of the mining area, including but not limited to geological structure, rock formation characteristics, mineral resource distribution, etc.

[0119] Next, two feature focusing rules will be used to gradually determine the associated geological distribution relationship vectors. In the first step, the first feature focusing rule is used to determine the corresponding transient geological distribution relationship vector cluster based on each geological distribution relationship vector cluster. This rule mainly focuses on extracting and strengthening the key geological features within each vector cluster. In this way, a set of more prominent and concentrated transient geological distribution relationship vectors can be obtained, which better represent the geological characteristics of the original vector cluster. In the second step, the second feature focusing rule is used to combine the multidimensional mining area status assessment vector and the transient geological distribution relationship vector cluster obtained in the previous step to determine the associated geological distribution relationship vector cluster. This rule further considers the overall status of the mining area, so that the associated geological distribution relationship vector not only reflects the local geological characteristics, but also incorporates the global information of the mining area. In this way, the associated geological distribution relationship vector obtained is more accurate and comprehensive, and can better guide subsequent exploration and production activities.

[0120] In the whole process, the target exploration and production linkage analysis component plays a core role. It is responsible for processing and analyzing a large amount of geological data and extracting useful information according to specific rules. This analysis method not only improves the efficiency of data processing, but also enhances the accuracy and reliability of prediction results.

[0121] In this way, the graphite mine distribution parameter determination system can more accurately determine the associated geological distribution relationship vector, thereby improving the prediction ability of the resource distribution in the mine. This method not only takes into account the local characteristics of the geological data, but also incorporates the overall status information of the mine, making the prediction results more comprehensive and accurate. This improvement not only helps guide explorers to locate mineral resources more effectively, but also provides strong data support for the sustainable development of the mine, thereby improving the mining efficiency and economic benefits of the entire mining industry.

[0122] It is worth mentioning that in the above technical solution, the first feature focusing rule and the second feature focusing rule are the core invention points, which are crucial to improving the prediction accuracy of resource distribution in graphite mining areas.

[0123] The first feature focusing rule focuses on extracting and enhancing key geological features from each geological distribution relationship vector cluster. The implementation of this rule includes the following steps:

[0124] Feature extraction: First, the system analyzes each geological distribution relationship vector cluster to identify and extract the most representative geological features. These features can include key information such as rock type, mineral composition, structural structure, etc., which is the basis for subsequent analysis;

[0125] Feature enhancement: After extracting key features, the first feature focusing rule will enhance these features. This means that the system will highlight or weight these features so that they occupy a more important position in subsequent analysis. By enhancing key features, the system can more accurately capture the essential attributes of each geological distribution relationship vector cluster;

[0126] Generate transient geological distribution relationship vector clusters: Based on the extracted and enhanced features, the system generates transient geological distribution relationship vector clusters. These transient vector clusters are refined representations of the original geological distribution relationship vector clusters. They retain the most critical geological information and provide a high-quality data foundation for the subsequent second feature focusing rules.

[0127] The second feature focusing rule is based on the first feature focusing rule, and further combines the multi-dimensional mining area status assessment vector to determine the associated geological distribution relationship vector cluster. The implementation of this rule includes the following key steps:

[0128] Global information fusion: The system combines the multi-dimensional mining area status assessment vector with the transient geological distribution relationship vector cluster. This means that in addition to considering local geological characteristics, the overall status of the mining area (such as climate, topography, hydrology, etc.) is also included in the analysis scope. This fusion of global information helps to improve the accuracy and comprehensiveness of the prediction;

[0129] Feature association analysis: After integrating the global information, the second feature focus rule will conduct in-depth feature association analysis. The system will look for potential connections and rules between transient geological distribution relationship vectors and multi-dimensional mining area status assessment vectors. This association analysis helps to reveal the mutual influence and dependence between different geological features and mining area status;

[0130] Generate associated geological distribution relationship vector clusters: Based on the results of the above association analysis, the system will generate associated geological distribution relationship vector clusters. These vector clusters not only reflect the local geological characteristics, but also incorporate the global status information of the mining area. Therefore, they can more accurately predict the resource distribution of the mining area and provide strong data support for subsequent exploration and production activities.

[0131] It can be seen that the first feature focusing rule and the second feature focusing rule together constitute the core technology of the graphite mining area distribution parameter determination system. Through the application of these two rules, the system can more accurately determine the associated geological distribution relationship vector cluster, thereby improving the prediction accuracy and reliability of the resource distribution of the entire mining area.

[0132] In some examples, each order of exploration and production linkage analysis components in the target resource distribution prediction network is the target exploration and production linkage analysis component; or, the first u order exploration and production linkage analysis components in the target resource distribution prediction network are the target exploration and production linkage analysis components; u is an integer not less than 1 and less than the total number of orders of exploration and production linkage analysis components in the target resource distribution prediction network.

[0133] In this embodiment, the target resource distribution prediction network plays a vital role. The network is composed of multiple exploration and production linkage analysis components, which are processed and analyzed in stages in the network.

[0134] First of all, it is necessary to clarify that "target exploration and production linkage analysis components" refer to key components responsible for analysis and processing tasks at a specific stage in the network. In some examples, all exploration and production linkage analysis components in the target resource distribution prediction network are set as target exploration and production linkage analysis components. This means that at each stage in the network, there are dedicated components responsible for processing and analyzing data, and each component undertakes key tasks and works together to complete the resource distribution prediction of the entire mining area.

[0135] In addition, there is another configuration in which only the first u-order E&P linkage analysis components in the target resource distribution prediction network are set as target E&P linkage analysis components. Here, "u" is a specific integer that is not less than 1 and less than the total order of E&P linkage analysis components in the network. This configuration means that in the first u stages of the network, there are dedicated components for in-depth data processing and analysis, while components after the u-th order may undertake other types of tasks or perform different levels of analysis.

[0136] Such a design is highly flexible and targeted. By adjusting the value of u, the number of components in the network responsible for key analysis tasks can be flexibly changed to adapt to mining environments and prediction needs of different complexities. For example, for mining areas with relatively simple geological conditions, a smaller u value can be set to reduce unnecessary analysis stages and improve efficiency; while for mining areas with complex geological conditions and uneven resource distribution, a larger u value can be set to ensure that detailed analysis can be carried out in multiple stages in the early stage, thereby improving the accuracy and reliability of the prediction.

[0137] In this way, the graphite mining area distribution parameter determination system realizes efficient and accurate prediction of mining area resource distribution by flexibly configuring the target exploration and production linkage analysis components in the target resource distribution prediction network. This configuration method not only improves the accuracy of prediction, but also enhances the adaptability and scalability of the system. Whether facing simple or complex geological environments, the system can adjust the number and stages of key analysis components to achieve the best prediction effect, thereby providing strong data support and decision-making basis for exploration and production activities in the mining area.

[0138] In some further exemplary embodiments, the method further includes: determining a multidimensional mining area historical status assessment vector corresponding to a historical sampling data set of a graphite mining area; determining, through a target exploration and production linkage analysis component in a target resource distribution prediction network, a historical associated geological distribution relationship vector corresponding to each of the historical geological distribution relationship vectors in the plurality of historical geological distribution relationship vector clusters according to a plurality of historical geological distribution relationship vector clusters and the multidimensional mining area historical status assessment vector; the plurality of historical geological distribution relationship vector clusters are obtained by clustering each of the to-be-processed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component; the to-be-processed geological distribution relationship vector is determined according to an original geological distribution relationship vector sample, or is obtained from a previous-stage exploration vector of the target exploration and production linkage analysis component according to mining feasibility determination requirements; The target exploration and production linkage analysis component is sampled from each historical associated geological distribution relationship vector output by the exploration and production linkage analysis component; the target exploration and production linkage analysis component is a first-order exploration and production linkage analysis component in the target resource distribution prediction network; for each cluster of the historical geological distribution relationship vectors, the training error corresponding to the cluster of the historical geological distribution relationship vectors is determined according to the historical associated geological distribution relationship vectors corresponding to each of the historical geological distribution relationship vectors and the debugging annotations corresponding to the historical sampling data set of the graphite mining area; based on the training errors corresponding to each of the several clusters of historical geological distribution relationship vectors, the training error corresponding to the target exploration and production linkage analysis component is determined; according to the training errors corresponding to each of the exploration and production linkage analysis components of each order in the target resource distribution prediction network, the target resource distribution prediction network is debugged.

[0139] In this exemplary embodiment, the system uses a method based on historical data and machine learning to optimize and improve the prediction accuracy of resource distribution.

[0140] First, the system determines the multidimensional mining area historical status evaluation vector corresponding to the historical sampling data set of the graphite mining area. This vector is a comprehensive evaluation indicator, which is generated based on multiple parameters in the historical sampling data set (such as geological structure, rock composition, mining area environment, etc.) to comprehensively reflect the historical status of the mining area.

[0141] Next, the system will determine the historical associated geological distribution relationship vectors corresponding to each of these historical geological distribution relationship vectors through the target exploration and production linkage analysis component in the target resource distribution prediction network, combined with several historical geological distribution relationship vector clusters and the above-mentioned multi-dimensional mining area historical status assessment vector. The "historical geological distribution relationship vector cluster" here is obtained by clustering the various unprocessed geological distribution relationship vectors processed by the target exploration and production linkage analysis component. These unprocessed geological distribution relationship vectors can be determined based on the original geological distribution relationship vector samples, or sampled from the historical associated geological distribution relationship vectors output by the previous exploration and production linkage analysis component according to the mining feasibility judgment requirements.

[0142] After determining the historical associated geological distribution relationship vector, the system will calculate the training error corresponding to each cluster of historical geological distribution relationship vectors according to the historical associated geological distribution relationship vectors corresponding to each historical geological distribution relationship vector within it and the debugging annotations corresponding to the historical sampling data set of the graphite mining area. This training error reflects the difference between the system prediction results and the actual historical data.

[0143] Then, based on the training errors corresponding to each of these historical geological distribution relationship vector clusters, the system will determine the total training error corresponding to the target exploration and production linkage analysis component. This total training error is a comprehensive evaluation of the prediction performance of the component.

[0144] Finally, the system will debug the entire network based on the training errors of each level of exploration and production linkage analysis component in the target resource distribution prediction network. This debugging process may include adjusting network parameters, optimizing algorithms, etc. to improve the prediction accuracy and stability of the network.

[0145] Through these steps, the graphite mine distribution parameter determination system can make full use of historical data and machine learning technology to continuously optimize its prediction performance. This training method based on historical data can not only improve the system's prediction accuracy of the resource distribution in the mine, but also enhance the system's adaptability and robustness. Through continuous iteration and optimization, the system can more accurately predict the resource distribution in the mine, thereby providing more reliable data support and decision-making basis for the exploration, mining and production of the mine.

[0146] In the next step, the training error corresponding to the historical geological distribution relationship vector cluster is determined based on the historical associated geological distribution relationship vector corresponding to each historical geological distribution relationship vector and the debugging annotation corresponding to the historical sampling data set of the graphite mining area, including: based on the historical associated geological distribution relationship vector corresponding to each historical geological distribution relationship vector in the historical geological distribution relationship vector cluster, determining the resource distribution training viewpoint corresponding to each historical associated geological distribution relationship vector; using a deep decision tree model, according to each resource distribution training viewpoint and the debugging annotation, determining the consistency analysis label corresponding to the historical geological distribution relationship vector cluster; the consistency analysis label is used to indicate the consistency analysis method between each resource distribution training viewpoint and the debugging annotation; based on the consistency analysis label, determining the training error corresponding to the historical geological distribution relationship vector cluster.

[0147] In the operation of the graphite mining area distribution parameter determination system, determining the training error corresponding to the historical geological distribution relationship vector cluster is a key step, which is helpful for the subsequent optimization and debugging of the system.

[0148] First, the system will determine the resource distribution training viewpoints corresponding to each historical geological distribution relationship vector in the historical geological distribution relationship vector cluster based on the historical geological distribution relationship vectors corresponding to each historical geological distribution relationship vector. These training viewpoints are essentially a prediction or judgment of the resource distribution based on historical data.

[0149] Next, the system will use a deep decision tree model, which is a machine learning model that can perform decision analysis based on input data. Here, the system will take the previously obtained resource distribution training viewpoints and the debugging annotations corresponding to the historical sampling data set of the graphite mining area as input, and use the deep decision tree model to determine the consistency analysis label corresponding to the historical geological distribution relationship vector cluster. This consistency analysis label can be regarded as an evaluation indicator to measure the degree of consistency between the resource distribution training viewpoints and the debugging annotations.

[0150] Specifically, the consistency analysis labels indicate how the consistency analysis between the training viewpoints and the debug annotations for each resource distribution is performed. For example, the labels may indicate which training viewpoints are highly consistent with the debug annotations and which are significantly different.

[0151] Finally, based on this consistency analysis label, the system can determine the training error corresponding to the historical geological distribution relationship vector cluster. This training error actually reflects the degree of difference between the system prediction results and the actual historical data, and is an important basis for subsequent optimization and debugging of the system.

[0152] In this way, the graphite mining area distribution parameter determination system can accurately determine the training error corresponding to the historical geological distribution relationship vector cluster. This not only helps the system understand the accuracy of its own prediction results, but also provides a clear direction for the subsequent optimization of the system. Through the consistency analysis of the deep decision tree model, the system can more accurately identify the difference between the prediction results and the actual data, so as to make targeted adjustments and improvements. This data-driven optimization method will help improve the overall performance and prediction accuracy of the graphite mining area distribution parameter determination system, and provide more reliable data support for the exploration and development of the mining area.

[0153] In some other possible embodiments, the method further includes: determining the quantitative relationship between the debugging annotation and the resource distribution training viewpoint based on the current network quality evaluation characteristics of the target resource distribution prediction network. Based on this, the use of the deep decision tree model to determine the consistency analysis label corresponding to the historical geological distribution relationship vector cluster according to each of the resource distribution training viewpoints and the debugging annotations includes: using the deep decision tree model, according to the quantitative relationship, according to each of the resource distribution training viewpoints and the debugging annotations, determining the consistency analysis label corresponding to the historical geological distribution relationship vector cluster.

[0154] In this embodiment, the system further introduces the current network quality evaluation feature to determine the quantitative relationship between the debugging annotation and the resource distribution training viewpoint. This step is crucial to improving the accuracy of resource distribution prediction and the debugging effect of the network.

[0155] First, the system determines the quantitative relationship between debugging annotations and resource distribution training viewpoints based on the current network quality evaluation features of the target resource distribution prediction network. The "network quality evaluation features" here can include multiple aspects such as network stability, prediction accuracy, and response speed, which together reflect the current performance and status of the network. By deeply analyzing these features, the system can more accurately understand the connection between debugging annotations and training viewpoints, thereby establishing a quantitative relationship model.

[0156] Next, when the system uses a deep decision tree model to determine the consistency analysis label corresponding to the historical geological distribution relationship vector cluster, it will be based on the quantitative relationship established earlier. Specifically, the system will use the resource distribution training viewpoints and debugging annotations as input according to the quantitative relationship and analyze them through the deep decision tree model. In this way, when judging consistency, the model will not only consider the content of the training viewpoints and debugging annotations themselves, but also combine the quantitative relationship between them, so as to obtain more accurate analysis results.

[0157] In this process, the deep decision tree model plays an important role. It can automatically learn and identify complex associations between data based on input data and quantitative relationships, and then generate consistency analysis labels. These labels not only reflect the degree of consistency between training views and debugging annotations, but also provide a strong basis for subsequent network debugging.

[0158] By introducing network quality evaluation features and quantitative relationships, the graphite mine distribution parameter determination system can more accurately evaluate the consistency between resource distribution training views and debugging annotations. This evaluation method not only improves the prediction accuracy, but also enhances the adaptability and robustness of the system. Using the deep decision tree model for consistency analysis can more objectively evaluate the degree of agreement between the prediction results and the actual data, thereby providing more reliable decision support for the exploration and development of the mining area. In general, these improvements have jointly improved the overall performance and practical value of the graphite mine distribution parameter determination system.

[0159] It is worth mentioning that the deep decision tree model is used to determine the consistency analysis labels corresponding to the historical geological distribution relationship vector clusters according to the quantitative relationship, the resource distribution training viewpoints and the debugging annotations, including: first, based on the quantitative relationship, a comprehensive evaluation index is constructed, which can integrate the correlation, importance and quantitative characteristics between the resource distribution training viewpoints and the debugging annotations; then, each resource distribution training viewpoint and the debugging annotation are vectorized so that they can be used as inputs of the deep decision tree model; then, the deep decision tree model is used to process these vectorized data, wherein each layer of the model nodes will divide the data according to the comprehensive evaluation index and gradually go deeper into the finer-grained features of the data; during the training process of the deep decision tree model, the structure and parameters of the tree are continuously adjusted through the optimization algorithm to ensure that the model can accurately capture the consistency relationship between the resource distribution training viewpoints and the debugging annotations; finally, according to the output results of the deep decision tree model, the consistency analysis labels corresponding to the historical geological distribution relationship vector clusters are determined, and these labels can accurately reflect the degree of consistency between each resource distribution training viewpoint and the debugging annotations.

[0160] Through the above technical solution, we can make full use of the powerful representation learning ability of the deep decision tree model to accurately capture the complex relationship between resource distribution training views and debugging annotations, thereby obtaining accurate consistency analysis labels and providing strong decision support for subsequent geological exploration and resource development.

[0161] In some independent embodiments, the resource distribution prediction viewpoint corresponding to the graphite mining area sampling data set is determined based on each associated geological distribution relationship vector output by the last-order exploration and production linkage analysis component in the target resource distribution prediction network, including:

[0162] First, aggregate the associated geological distribution relationship vectors output by the last-order exploration and production linkage analysis component to form a comprehensive geological distribution relationship representation. This step can use weighted average, maximum pooling or other aggregation functions to ensure that the information of all associated geological distribution relationship vectors can be effectively integrated;

[0163] Next, a decoder network or regression model is used to convert the comprehensive geological distribution relationship representation into a specific resource distribution prediction. This decoder network or regression model can be a neural network based on deep learning, such as a multi-layer perceptron (MLP) or a convolutional neural network (CNN), or a traditional statistical regression model, such as linear regression or support vector regression (SVR);

[0164] During the conversion process, the decoder network or regression model can be trained and optimized by combining historical sampling data of graphite mining areas, geological background knowledge and expert experience to improve the accuracy and reliability of resource distribution prediction;

[0165] Finally, according to the output of the decoder network or regression model, the resource distribution prediction viewpoints corresponding to the graphite mining area sampling data set are determined. These viewpoints can be presented in the form of graphics, tables or texts to provide decision makers with intuitive and easy-to-understand resource distribution prediction information.

[0166] Through the above technical solution, it is possible to fully utilize the associated geological distribution relationship vector output by the last-order exploration and production linkage analysis component in the target resource distribution prediction network, accurately predict the resource distribution of the graphite mining area, and provide strong support for the exploration and development of the mining area.

[0167] In some independent embodiments, the method further includes: generating exploration and production annotation plan graphics of the target graphite mining area pointed to by the graphite mining area sampling data set based on the resource distribution prediction viewpoint.

[0168] In detail, after performing the resource distribution prediction task, the graphite mine area distribution parameter determination system will further use its rich functions and advanced algorithms to generate detailed exploration and production annotation plan graphics for the target graphite mine area based on the obtained resource distribution prediction viewpoints. Specifically, the system will first parse the resource distribution prediction viewpoints, which are obtained based on the in-depth analysis and decoding of the associated geological distribution relationship vectors, reflecting the possible resource distribution of the graphite mine area. The high-performance computing module inside the system will further process the data and render the images of these prediction viewpoints to form an intuitive and clear graphical representation. In the process of generating exploration and production annotation plan graphics, the graphite mine area distribution parameter determination system will combine geological principles, mine area exploration specifications and actual production needs to ensure that the generated graphics are both scientific and practical. These graphics may include thermal maps, contour maps or three-dimensional model maps of resource distribution, as well as the corresponding exploration drilling location, depth and expected production annotations. In addition, the system will also give priority division and mining suggestions in the plan graphics based on factors such as the density of resource distribution, quality grade and mining difficulty. These suggestions can help decision makers to rationally plan exploration paths and production plans, thereby improving exploration efficiency and resource utilization. The final generated exploration and production annotation plan graphics will be displayed to users through the system's user interface, or exported in the form of electronic documents for users to refer to during field exploration and production.

[0169] Through this series of operations of the graphite mining area distribution parameter determination system, the accuracy of graphite mining area exploration and production efficiency can be significantly improved. The system not only provides scientific resource distribution predictions, but also gives specific exploration and production suggestions, greatly reducing exploration risks and blindness, and also provides strong technical support for the sustainable development of graphite mining areas. This intelligent exploration and production method will undoubtedly bring revolutionary changes to the development and utilization of graphite mineral resources.

[0170] In some other independent embodiments, based on the resource distribution prediction viewpoint, an exploration and production annotation plan image and text of the target graphite mining area pointed to by the graphite mining area sampling data set is generated, including: using the resource point location information of the resource distribution prediction viewpoint to obtain the mining area spatial mapping characteristics of the resource distribution prediction viewpoint; using a number of image and text annotation feature subsets obtained by feature embedding processing of the image and text annotation feature set, respectively performing semantic enhancement on the mining area spatial mapping features to obtain a number of semantic enhancement mapping features, wherein the image and text annotation feature subsets include at least one annotation mask set, and the annotation mask set includes a plurality of feature relationship network groups that do not overlap and have the same image and text feature members at corresponding mask labels; based on the several semantic enhancement mapping features, the exploration and production behavior annotation features of the mining area spatial mapping features are obtained; and the exploration and production annotation plan image and text are generated according to the exploration and production behavior annotation features.

[0171] In this embodiment, the graphite mining area distribution parameter determination system will further generate exploration and production annotation plan graphics for the target graphite mining area based on the resource distribution prediction viewpoint.

[0172] First, the graphite mining area distribution parameter determination system will extract the resource point location information from the resource distribution prediction point of view. This information accurately points out the possible location of graphite resources in the mining area, which is the key basis for subsequent exploration and production.

[0173] Then, the system uses this positioning information to construct a spatial mapping feature of the mining area from the perspective of resource distribution prediction. This mapping feature is actually a digital mining area model that accurately reflects the spatial distribution of graphite resources in the mining area.

[0174] Then, the system will introduce a feature set of graphic annotations, which is extracted from a large amount of historical exploration and production graphic data through deep learning technology. The system will embed this feature set to obtain several subsets of graphic annotation features. These subsets contain rich semantic information related to geology, exploration and production, which is of great significance for guiding actual exploration and production activities.

[0175] Among these feature subsets, an important component is the annotation mask set. The annotation mask set consists of multiple non-overlapping feature relationship network groups, each network group corresponds to a specific mask label, and the image and text feature members in the network group are the same. These annotation mask sets play a key role in the subsequent semantic enhancement process.

[0176] Next, the system uses these feature subsets to semantically enhance the spatial mapping features of the mining area. This process is achieved by fusing the semantic information in the feature subsets with the spatial mapping features of the mining area. The result of the fusion is a series of semantically enhanced mapping features that not only retain the original spatial distribution information, but also incorporate rich geological and exploration production semantics.

[0177] Based on these semantically enhanced mapping features, the system can further extract exploration and production behavior annotation features of the mining area spatial mapping features. These annotation features directly reflect the exploration and production behaviors that should be carried out at a specific spatial location and are the core basis for generating exploration and production annotation plan graphics.

[0178] Finally, based on these exploration and production behavior annotation features, the graphite mine area distribution parameter determination system will generate exploration and production annotation plan graphics. These graphics not only mark the distribution of graphite resources and exploration and production suggestions in detail, but also show the geological structure and exploration and production plan of the entire mining area in an intuitive and easy-to-understand way.

[0179] In this way, the graphite mine distribution parameter determination system can provide users with scientific and accurate exploration and production guidance, thereby greatly improving the exploration efficiency and production efficiency of graphite mines. At the same time, because the system incorporates rich geological and exploration and production semantic information, the generated plan graphics are more in line with actual production needs, further improving the intelligent level of graphite mine development.

[0180] In some other independent embodiments, before using several graphic annotation feature subsets obtained by feature embedding processing of the graphic annotation feature set to semantically enhance the spatial mapping features of the mining area to obtain several semantically enhanced mapping features, the method also includes: in the graphic annotation feature set, screening out a first feature relationship network that meets the first annotation rule; wherein the first positioning parameter and the second positioning parameter of the graphic feature members in the first feature relationship network are the same variables; using the graphic feature members in the first feature relationship network, obtaining a transient annotation relationship network; wherein the transient annotation relationship network includes at least one pair of feature relationship network groups, and the two feature relationship network groups included in each pair of the feature relationship network groups are respectively in the query area of ​​the index feature set of the transient annotation relationship network; based on the feature relationship network groups included in the transient annotation relationship network, performing at least one feature embedding process on the transient annotation relationship network to obtain the graphic annotation feature subset.

[0181] In this embodiment, the graphite mining area distribution parameter determination system performs a series of complex feature processing and screening tasks before executing semantic enhancement to ensure that the graphic and text annotation feature subset used for semantic enhancement has a high degree of accuracy and relevance.

[0182] First, the graphite mining area distribution parameter determination system will screen out the first feature relationship network that meets the first annotation rule from the image and text annotation feature set. The "first annotation rule" here may refer to the consistency of the positioning parameters of the image and text feature members in the feature relationship network. Specifically, the system will check the image and text feature members in each feature relationship network to ensure that their first positioning parameters and second positioning parameters are the same variables. These two positioning parameters may refer to the position coordinates, size, or other information related to spatial positioning of the feature in the image. Through this screening step, the system can ensure that the selected feature relationship network has a high degree of consistency in spatial positioning, which is crucial for subsequent semantic enhancement processing.

[0183] Next, the system uses the image and text feature members in the first feature relationship network that have been screened out to construct a transient annotation relationship network. This transient annotation relationship network contains at least one pair of feature relationship network groups, and the two feature relationship network groups contained in each pair of feature relationship network groups are respectively in the query area of ​​the index feature set of the transient annotation relationship network. This means that each pair of feature relationship network groups has a clear spatial position and mutual relationship in the transient annotation relationship network, which facilitates the system to perform subsequent feature embedding processing.

[0184] Then, based on the feature network group contained in this transient annotation network, the system will perform at least one feature embedding process on the transient annotation network. Feature embedding is a technology that maps high-dimensional data to low-dimensional space, which can retain the main features of the data while reducing the complexity of calculation. Through this step, the system can extract the core features in the transient annotation network and form a feature subset of the text and image annotation.

[0185] These feature subsets of graphic annotations are then used to semantically enhance the spatial mapping features of the mining area, thereby obtaining several semantically enhanced mapping features. These enhanced mapping features not only retain the original spatial information, but also incorporate rich semantic information, providing more accurate and comprehensive data support for subsequent exploration and production behavior annotation and plan graphic generation.

[0186] In this way, the graphite mine distribution parameter determination system can generate more accurate and useful exploration and production annotation plan graphics. These graphics can not only help users better understand the geological structure and resource distribution of the mining area, but also provide scientific guidance and suggestions for actual exploration and production activities. Therefore, the implementation of this technical solution plays an important role in improving the exploration efficiency and production efficiency of graphite mining areas.

[0187] In some other independent embodiments, when multiple first feature relationship networks are screened and the multiple first feature relationship networks do not intersect with each other, there is only one transient annotation relationship network, and the transient annotation relationship network is obtained by using the graphic feature members in the first feature relationship network with the largest dimension among the multiple first feature relationship networks; or, when multiple first feature relationship networks are screened and the multiple first feature relationship networks do not intersect with each other, there are multiple transient annotation relationship networks, and each of the transient annotation relationship networks is obtained by using the graphic feature members in a corresponding first feature relationship network.

[0188] Based on this embodiment, when the graphite mining area distribution parameter determination system screens the first feature relationship network, it may encounter the situation that multiple first feature relationship networks are screened and these relationship networks do not intersect each other. At this time, the system will construct a transient annotated relationship network according to the specific situation. The following are two possible processing methods: First, if the multiple first feature relationship networks screened do not intersect each other, the system can select a first feature relationship network with the largest dimension to construct a transient annotated relationship network. The "dimension" here may refer to the number of features, complexity or information contained in the feature relationship network. The system uses the graphic feature members in the first feature relationship network with the largest dimension to construct a transient annotated relationship network. This transient annotated relationship network can retain the information in the original feature relationship network to the greatest extent, and provide a rich data foundation for subsequent feature embedding processing and semantic enhancement. Secondly, another processing method is that the system constructs a transient annotated relationship network for each screened first feature relationship network. This means that if N non-intersecting first feature relationship networks are screened, the system will construct N transient annotated relationship networks. Each transient annotated relationship network is obtained using the graphic feature members in a corresponding first feature relationship network. This approach can process the information in each feature relationship network more carefully and avoid information loss or confusion, but it may increase the complexity of subsequent processing.

[0189] Regardless of which method is chosen to construct transient annotation relationship networks, the system will ensure that these relationship networks can accurately reflect the feature relationships in the original data and provide strong support for subsequent feature embedding processing and semantic enhancement.

[0190] Through these two processing methods, the graphite mining area distribution parameter determination system can flexibly respond to different situations and ensure the accuracy and practicality of the generated exploration and production annotation plan graphics. This not only improves the exploration efficiency of the graphite mining area, but also provides scientific data support for the sustainable development of the mining area. At the same time, the flexibility and accuracy of the system also enhance the user's trust in the measurement results, laying a solid foundation for the rational planning and development of the graphite mining area.

[0191] Further, Figure 2 The structure diagram of a graphite mining area distribution parameter determination system 200 provided in an embodiment of the present application. Figure 2 The graphite mining area distribution parameter determination system 200 shown includes a processor 210, and the processor 210 can call and run a computer program from a memory to implement the method in the embodiment of the present application.

[0192] Alternatively, if Figure 2 As shown, the graphite mining area distribution parameter determination system 200 may further include a memory 230. The processor 210 may call and run a computer program from the memory 230 to implement the method in the embodiment of the present application.

[0193] The memory 230 may be a separate device independent of the processor 210 , or may be integrated into the processor 210 .

[0194] Alternatively, if Figure 2 As shown, the graphite mine distribution parameter determination system 200 may also include a transceiver 220, and the processor 210 may control the transceiver 220 to interact with other devices, specifically, to send information or data to other devices, or to receive information or data sent by other devices.

[0195] Optionally, the graphite mining area distribution parameter determination system 200 can implement the corresponding processes corresponding to the storage engine or components in the storage engine (such as a processing module) or a device deployed with a storage engine in each method of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0196] It should be understood that the processor of the embodiment of the present application may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiment can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to be executed, or a combination of hardware and software modules in the decoding processor to be executed. The software module may be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0197] It can be understood that the memory in the embodiments of the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0198] It should be understood that the above-mentioned memory is exemplary but not restrictive. For example, the memory in the embodiments of the present application may also be static random access memory (static RAM, SRAM), dynamic random access memory (dynamic RAM, DRAM), synchronous dynamic random access memory (synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (synch link DRAM, SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DR RAM), etc. That is to say, the memory in the embodiments of the present application is intended to include but not limited to these and any other suitable types of memory.

[0199] Based on the above, a computer-readable storage medium is provided, on which a computer program is stored, and the computer program implements the above method when running.

[0200] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations.

Claims

1. A method for determining distribution parameters of graphite mining areas, characterized in that: Applied to a graphite mining area distribution parameter determination system, the method comprises: Determine a multi-dimensional mining area state assessment vector corresponding to a graphite mining area sampling data set; Through the target exploration and production linkage analysis component in the target resource distribution prediction network, according to several geological distribution relationship vector clusters and the multi-dimensional mining area status assessment vector, the associated geological distribution relationship vectors corresponding to each geological distribution relationship vector in the several geological distribution relationship vector clusters are determined; the several geological distribution relationship vector clusters are obtained by clustering the various to-be-processed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component; the to-be-processed geological distribution relationship vectors are determined according to the original geological distribution relationship vectors, or are sampled from various associated geological distribution relationship vectors output by the previous-order exploration and production linkage analysis component of the target exploration and production linkage analysis component according to the mining feasibility judgment requirements, and the target exploration and production linkage analysis component is a first-order exploration and production linkage analysis component in the target resource distribution prediction network; Determine the resource distribution prediction viewpoint corresponding to the graphite mining area sampling data set based on each associated geological distribution relationship vector output by the last-order exploration and production linkage analysis component in the target resource distribution prediction network; When the target exploration and production linkage analysis component is a non-first-order exploration and production linkage analysis component in the target resource distribution prediction network, each to-be-processed geological distribution relationship vector corresponding to the target exploration and production linkage analysis component is determined based on the following steps: Based on each associated geological distribution relationship vector output by the previous exploration and production linkage analysis component, determine the resource distribution prediction viewpoint corresponding to each associated geological distribution relationship vector; the resource distribution prediction viewpoint includes the confidence weight corresponding to the ore resource indicated by the associated geological distribution relationship vector; According to the confidence weights in the resource distribution prediction viewpoints corresponding to the respective associated geological distribution relationship vectors, the associated geological distribution relationship vectors that meet the mining feasibility judgment requirements are selected from the associated geological distribution relationship vectors as the geological distribution relationship vector to be processed corresponding to the target exploration and production linkage analysis component; The input of the previous-order exploration and production linkage analysis component includes a plurality of geological distribution relationship vector clusters obtained by clustering the geological distribution relationship vectors to be processed corresponding to the previous-order exploration and production linkage analysis component; the output of the previous-order exploration and production linkage analysis component includes the associated geological distribution relationship vector clusters corresponding to each of the plurality of geological distribution relationship vector clusters; The method of selecting, according to the confidence weights in the resource distribution prediction viewpoints corresponding to the associated geological distribution relationship vectors, the associated geological distribution relationship vectors that meet the mining feasibility determination requirements from the associated geological distribution relationship vectors as the geological distribution relationship vector to be processed corresponding to the target exploration and production linkage analysis component includes: For each of the associated geological distribution relationship vector clusters, according to the confidence weights in the resource distribution prediction viewpoints corresponding to the respective associated geological distribution relationship vectors in the associated geological distribution relationship vector clusters, an associated geological distribution relationship vector that meets the mining feasibility judgment requirements is selected from the associated geological distribution relationship vector clusters as the geological distribution relationship vector to be processed corresponding to the target exploration and production linkage analysis component; The target exploration and production linkage analysis component in the target resource distribution prediction network determines, according to a plurality of geological distribution relationship vector clusters and the multi-dimensional mining area status assessment vector, the associated geological distribution relationship vectors corresponding to each of the geological distribution relationship vectors in the plurality of geological distribution relationship vector clusters, including: By using the target exploration and production linkage analysis component, a first feature focusing rule is adopted to determine, according to each of the geological distribution relationship vector clusters, a transient geological distribution relationship vector cluster corresponding to each of the geological distribution relationship vector clusters; the transient geological distribution relationship vector cluster includes the transient geological distribution relationship vectors corresponding to each of the geological distribution relationship vectors in the corresponding geological distribution relationship vector cluster; Through the target exploration and production linkage analysis component, the second feature focusing rule is adopted to determine the associated geological distribution relationship vector cluster corresponding to each transient geological distribution relationship vector cluster and the multidimensional mining area status assessment vector; the associated geological distribution relationship vector cluster includes the associated geological distribution relationship vectors corresponding to each transient geological distribution relationship vector in the corresponding transient geological distribution relationship vector cluster.

2. The method according to claim 1, characterized in that When the target exploration and production linkage analysis component is a first-order exploration and production linkage analysis component in the target resource distribution prediction network, each to-be-processed geological distribution relationship vector corresponding to the target exploration and production linkage analysis component is determined based on the following steps: Using each of the original geological distribution relationship vectors as each of the to-be-processed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component; Alternatively, determine the rock formation structural stability characteristics corresponding to each of the original geological distribution relationship vectors; and select the original geological distribution relationship vector that meets the set stability index from each of the original geological distribution relationship vectors according to the rock formation structural stability characteristics corresponding to each of the original geological distribution relationship vectors as the geological distribution relationship vector to be processed corresponding to the target exploration and production linkage analysis component.

3. The method according to any one of claims 1 to 2, characterized in that: The geological distribution relationship vector cluster is obtained by any of the following methods: According to the set number of clusters, each to-be-processed geological distribution relationship vector corresponding to the target exploration and production linkage analysis component is divided into the plurality of geological distribution relationship vector clusters; Determine the rock formation structural stability characteristics corresponding to each of the to-be-processed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component; According to the set number of clusters, and based on the rock layer structure stability characteristics corresponding to each of the geological distribution relationship vectors to be processed, each of the geological distribution relationship vectors to be processed is transferred to the plurality of geological distribution relationship vector clusters; Determine the positioning labels of each unprocessed geological distribution relationship vector corresponding to the target exploration and production linkage analysis component in the three-dimensional vector relationship network; according to the set number of clusters and the positioning labels of each unprocessed geological distribution relationship vector in the three-dimensional vector relationship network, transfer each of the unprocessed geological distribution relationship vectors to the several geological distribution relationship vector clusters.

4. The method according to any one of claims 1 to 2, characterized in that: All the exploration and production linkage analysis components of each order in the target resource distribution prediction network are the target exploration and production linkage analysis components; Alternatively, the first u-order exploration and production linkage analysis components in the target resource distribution prediction network are the target exploration and production linkage analysis components; and u is an integer not less than 1 and less than the total order of exploration and production linkage analysis components in the target resource distribution prediction network.

5. The method according to claim 1, characterized in that The method further comprises: Determine a multi-dimensional mining area historical state assessment vector corresponding to a graphite mining area historical sampling data set; Determine, through the target exploration and production linkage analysis component in the target resource distribution prediction network, the historical associated geological distribution relationship vectors corresponding to each of the historical geological distribution relationship vectors in the plurality of historical geological distribution relationship vector clusters according to the plurality of historical geological distribution relationship vector clusters and the multi-dimensional mining area historical state assessment vector; The several clusters of historical geological distribution relationship vectors are obtained by clustering the various to-be-processed geological distribution relationship vectors corresponding to the target exploration and production linkage analysis component; the to-be-processed geological distribution relationship vectors are determined according to the original geological distribution relationship vector samples, or are sampled from various historically associated geological distribution relationship vectors output by the previous-order exploration and production linkage analysis component of the target exploration and production linkage analysis component according to the mining feasibility judgment requirements; the target exploration and production linkage analysis component is a first-order exploration and production linkage analysis component in the target resource distribution prediction network; For each of the historical geological distribution relationship vector clusters, determining the training error corresponding to the historical geological distribution relationship vector cluster according to the historical associated geological distribution relationship vectors corresponding to each of the historical geological distribution relationship vectors and the debugging annotations corresponding to the historical sampling data set of the graphite mining area; Determining the training error corresponding to the target exploration and production linkage analysis component based on the training errors corresponding to each of the plurality of historical geological distribution relationship vector clusters; The target resource distribution prediction network is debugged according to the training errors corresponding to the exploration and production linkage analysis components of each order in the target resource distribution prediction network.

6. The method according to claim 5, characterized in that The determining of the training error corresponding to the cluster of historical geological distribution relationship vectors according to the historical associated geological distribution relationship vectors corresponding to each of the historical geological distribution relationship vectors and the debugging annotations corresponding to the historical sampling data set of the graphite mining area comprises: Determine the resource distribution training viewpoints corresponding to each of the historical geological distribution relationship vectors based on the historical geological distribution relationship vectors corresponding to each of the historical geological distribution relationship vectors in the historical geological distribution relationship vector cluster; Using a deep decision tree model, according to each of the resource distribution training viewpoints and the debugging annotations, a consistency analysis label corresponding to the historical geological distribution relationship vector cluster is determined; the consistency analysis label is used to indicate the consistency analysis method between each of the resource distribution training viewpoints and the debugging annotations; Determining the training error corresponding to the historical geological distribution relationship vector cluster based on the consistency analysis label; The method also includes: determining the quantitative relationship between the debugging annotation and the resource distribution training viewpoint based on the current network quality evaluation characteristics of the target resource distribution prediction network; using the deep decision tree model to determine the consistency analysis label corresponding to the historical geological distribution relationship vector cluster according to each of the resource distribution training viewpoints and the debugging annotation, including: using the deep decision tree model, according to the quantitative relationship, according to each of the resource distribution training viewpoints and the debugging annotation, determining the consistency analysis label corresponding to the historical geological distribution relationship vector cluster.

7. A graphite mining area distribution parameter determination system, characterized in that: The method comprises at least one processor and a memory; the memory stores computer-executable instructions; the at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the method according to any one of claims 1 to 6.

Images