A mineral resource estimation system based on multi-source heterogeneous big data fusion

Abstract

The application belongs to the technical field of mineral resources management, and discloses a mineral resource estimation system based on multi-source heterogeneous big data fusion; comprising: a data acquisition module for acquiring mineral characteristic data; a data analysis module for analyzing the mineral characteristic data and evaluating the mining method; a cycle estimation module for estimating the mining cycle based on the mining method; a data collection module for collecting mineral influence data; and a mining optimization module for optimizing the mining scheme according to the mineral influence data and the mining cycle; the application not only overcomes the limitations of traditional static estimation methods and enhances the dynamic management and optimization capabilities of mineral resources, but also balances the economic benefits and environmental impacts through the optimization of the mining scheme, improves the economic benefits and environmental protection level of mineral resources, and thus improves the scientificity and sustainability of mineral development.

Description

Technical Field

[0001] This invention relates to the field of mineral resource management technology, and more specifically, to a mineral resource estimation system based on the fusion of multi-source heterogeneous big data. Background Technology

[0002] As an important strategic resource, the scientific development and sustainable management of mineral resources are crucial to national economic and environmental development. However, traditional mineral resource estimation methods mostly focus on static reserve calculations, typically relying on geological exploration, drilling, and sampling, primarily concerned with static data such as the existing reserves and grade of the ore body. While these traditional methods can provide a preliminary assessment of mineral resources, their inherent limitations make it difficult to meet the comprehensive assessment needs of the mineral mining process. They have significant shortcomings in dynamic management, resource optimization, and sustainable development, and cannot effectively support the optimization of mining plans and the prediction of ore body mining cycles.

[0003] With the advancement of remote sensing, geophysical exploration, and geographic information systems, multi-source heterogeneous data is increasingly being applied to mineral resource estimation. These data, originating from different channels and technologies, cover various dimensions of mineral resources and provide more comprehensive and accurate information for precisely reflecting the mining potential of ore bodies, assessing mining benefits, and evaluating their environmental impact. By fusing multi-source heterogeneous data, the shortcomings of traditional static estimation methods can be overcome, providing more possibilities for accurately reflecting mining potential and environmental impact. Therefore, how to assess the mining cycle of ore bodies and optimize mining schemes based on the fusion of multi-source heterogeneous data has become a key challenge in mineral resource estimation.

[0004] In view of this, the present invention proposes a mineral resource estimation system based on the fusion of multi-source heterogeneous big data to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a mineral resource estimation system based on multi-source heterogeneous big data fusion, comprising:

[0006] The data acquisition module is used to acquire mineral characteristic data;

[0007] The data analysis module is used to analyze mineral characteristic data and evaluate mining methods;

[0008] The cycle estimation module estimates the mining cycle based on the mining method.

[0009] The data collection module is used to collect data on the impact of mineral resources.

[0010] The mining optimization module optimizes mining plans based on mineral impact data and mining cycles.

[0011] Furthermore, the mineral characteristic data includes reserve data, grade data, and distribution data;

[0012] The method for evaluating the mining method includes:

[0013] Set different digital tags for different mining methods and mark them as mode tags; input the mineral characteristic data into the trained mode evaluation model, output the corresponding mode tags, and obtain the corresponding mining methods according to the mode tags; the training process of the mode evaluation model includes:

[0014] The mode evaluation model adopts a random forest regression model. Pre-collect a set of historical data. The historical data includes mineral characteristic data and the corresponding mode tags; sample b tree subsets from the a set of historical data. Each tree subset includes c sets of historical data, and train each decision tree according to the b tree subsets; where a and b are both integers greater than 1, and b < a, a = bc;

[0015] Initialize the model parameters, including the number of decision trees b, the maximum depth d of each decision tree, and the minimum number of samples u for internal nodes;

[0016] Recursively construct the first decision tree using one tree subset, and then sequentially train b - 1 decision trees using the remaining b - 1 tree subsets; the generation process of each decision tree is independent and the same; finally, construct a random forest containing b decision trees;

[0017] Calculate the loss function value on the b decision trees; input the mineral characteristic data into each decision tree, record the decision output, the decision output is the mode tag output by each decision tree and is marked as the prediction tag; calculate the mean square error between the average value of all prediction tags and the true value as the loss function value, and the true value is the mode tag corresponding to the mineral characteristic data input into the decision tree; reduce the loss function value through parameter tuning, and select the model hyperparameters with the smallest loss function value as the model hyperparameters of the mode evaluation model.

[0018] Furthermore, the method for recursively constructing the first decision tree using one tree subset includes:

[0019] Divide the c sets of historical data in one tree subset into a training set α and a validation set β, α p is the p-th set of historical data in the training set α, p ∈ [1, α′], α′ < c, and α′ is the number of historical data in the training set α; use the feature t h to represent one data in the mineral characteristic data, and calculate the information gain rate Z(α, t h ) of the h-th feature t h ;

[0020] For each feature t hThe corresponding information gain ratio Z(α, t) h The calculation is performed, and the maximum information gain ratio Z(α, t) is selected. h The feature t corresponding to ) h As an internal node, and based on the maximum information gain ratio Z(α, t) h The feature t corresponding to ) h The training set α is divided into decision subsets, with each decision subset containing the same amount of historical data.

[0021] For each, based on the maximum information gain ratio Z(α, t) h The decision tree is divided into decision subsets, and the algorithm recursively repeats the process until the number of historical data in the decision subset is less than or equal to the minimum number of samples u of the internal nodes, or the number of recursions is greater than or equal to the maximum depth d of each decision tree, at which point the recursion ends; the trained decision tree model is evaluated using the validation set β, and the construction of the first decision tree model is finally completed.

[0022] Furthermore, Z(α, t) h The expression for ) is: In the formula, h∈[1,k], k is the number of data types in the mineral feature data, and αg is the training set α based on feature t. h The decision subset is divided, g∈[1,k′], where k′ is an integer greater than 1, H(αg) is the information entropy of ag, and H(α) is the information entropy of the training set α. The calculation methods of H(α) and H(αg) are as follows:

[0023]

[0024]

[0025] In the formula, αg p This represents the method label corresponding to the p-th group of historical data in αg.

[0026] Furthermore, the method for estimating the mining cycle includes:

[0027] Based on reserve data, distribution data, and mining methods, the range of mining rates is estimated; a pre-set attenuation factor q is used; the expression for the change in mining rate is: V(t) = V min +(V max -V min )×e -qt In the formula, V(t) is the mining rate corresponding to mining time t, V min V is the minimum extraction rate within the extraction rate range. max The maximum extraction rate is defined as e, where e is the natural constant and t is the extraction time. The expression for the reserves data is: In the formula, S represents the reserve data, and T represents the mining cycle; the numerical solution method is used to solve for the value of the mining cycle from the expression of the reserve data.

[0028] Furthermore, the step of estimating the mining rate range includes:

[0029] Step A1: Set the search interval [m,n] and precision ε;

[0030] Step A2: Determine the search sequence F(x);

[0031] Step A3: Determine the partition coefficient F(r) from the search sequence;

[0032] Step A4: Divide the search interval using a partitioning coefficient to obtain two points ψ1 and ψ2;

[0033] Step A5: Calculate the feasibility of ψ1 and ψ2 respectively;

[0034] Step A6: Update the search range;

[0035] Step A7: Calculate the width υ of the search interval;

[0036] Step A8: If the width υ is less than or equal to the precision ε, then the search interval is taken as the mining rate range; if the width υ is greater than the precision ε, then return to step A3.

[0037] Furthermore, in step A2, the search sequence

[0038] In step A3, the method for determining the partitioning coefficient F(r) is as follows: calculate the initial partitioning coefficient δ′. Subtract the initial splitting coefficient from each value in the search sequence to obtain the numerical difference; sort each numerical difference from smallest to largest, obtain the value in the search sequence corresponding to the numerical difference at the top of the list, and use it as the splitting coefficient F(r);

[0039] In step A4 In the formula, F(r-2) represents the two values ​​in the search sequence that rank first and second to the dividing coefficient F(r), and F(r-1) represents the value in the search sequence that ranks first to the dividing coefficient F(r).

[0040] In step A5, the feasibility is obtained by using the following methods: storage data, distribution data, mode labels, and the corresponding values ​​of points as analysis data, inputting the analysis data into the trained rate estimation model, and predicting the corresponding feasibility.

[0041] In step A6, the method for updating the search interval is as follows: mark the feasibility corresponding to ψ1 as the first feasibility. The feasibility corresponding to ψ2 is marked as the second feasibility. like The search range is then updated to Right now like The search range is then updated to Right now

[0042] In step A7, the maximum value of the search interval is subtracted from the minimum value to obtain the width υ.

[0043] Furthermore, the training process for the rate estimation model includes:

[0044] Collect y sets of analysis data in advance, and set a corresponding feasibility for each of the y sets of analysis data, where y is an integer greater than 1. Convert the analysis data and the corresponding feasibility into a set of corresponding feature vectors.

[0045] Each set of feature vectors is used as input to the rate estimation model. The rate estimation model outputs a set of predicted feasibility corresponding to each set of analyzed data and uses the actual feasibility corresponding to each set of analyzed data as the prediction target. The actual feasibility is the pre-set feasibility corresponding to the analyzed data. The training objective is to minimize the sum of prediction errors of all analyzed data. The rate estimation model is trained until the sum of prediction errors converges and then training stops. The rate estimation model is a deep neural network model.

[0046] Furthermore, the mineral impact data includes historical market data and environmental data; the historical market data includes historical supply and demand data and historical price data; the historical supply and demand data represents the production and consumption of mineral resources at historical moments; and the historical price data represents the transaction prices of mineral resources at historical moments.

[0047] Historical supply and demand data are input into a supply and demand forecasting model to predict future supply and demand data, which includes the production and consumption of mineral resources during the mining cycle. Historical price data are input into a price forecasting model to predict future price data, which includes the transaction price of mineral resources during the mining cycle. Both the supply and demand forecasting model and the price forecasting model are LSTM models in the recurrent neural network model. Based on the future supply and demand data and future price data, the mining plan during the mining cycle is optimized, and the mining plan includes the mining rate and mining method.

[0048] The steps to optimize the mining plan include:

[0049] Step B1: Construct N mining sets, assign different numerical labels to different mining sets and mark them as set labels, and construct a mining matrix based on all set labels;

[0050] Step B2: Preset the constraint table size and maximum number of iterations, and determine the evaluation function;

[0051] Step B3: Randomly select a set label from the mining matrix as the current solution and add it to the constraint table. The current solution is then taken as the optimal solution.

[0052] Step B4: Based on the mining matrix, obtain the set labels adjacent to the current solution and use them as adjacent solutions;

[0053] Step B5: Calculate the evaluation value for each adjacent solution;

[0054] Step B6: Determine if each adjacent solution exists in the constraint table; if it exists, mark the corresponding adjacent solution as a constraint solution; if it does not exist, treat the corresponding adjacent solution as a candidate solution.

[0055] Step B7: Mark the candidate solution with the largest evaluation value as the optimal candidate solution. If the evaluation value of the constraint solution is greater than the evaluation value of the optimal candidate solution, then update the optimal candidate solution to the constraint solution. If the evaluation value of the constraint solution is less than or equal to the evaluation value of the optimal candidate solution, then delete the constraint solution.

[0056] Step B8: Update the current solution to the optimal candidate solution, and determine whether to update the optimal solution; add all candidate solutions to the constraint table;

[0057] Step B9: Determine if the number of iterations is equal to the maximum number of iterations; if yes, terminate the iteration and obtain the mining set corresponding to the set label of the optimal solution; if no, increment the number of iterations and return to step B4.

[0058] Further, in step B1, one mining method is randomly selected from all mining methods, and a value is randomly selected from the mining rate range to construct a mining set. A total of N mining sets are constructed, and all N mining sets are different.

[0059] In step B2, the expression for the evaluation function is: In the formula, f is the evaluation value, jx is the economic benefit, and hy is the degree of environmental impact. All are weighting coefficients;

[0060] The methods for obtaining economic benefits and environmental impact are as follows: Based on future supply and demand data and future price data, predicted supply and demand data and predicted price data are obtained; the predicted supply and demand data is one data point from the future supply and demand data, and the predicted price data is one data point from the future price data, with the predicted supply and demand data and predicted price data corresponding to the same time period; the predicted supply and demand data, predicted price data, and set labels are used as test data, and the test data are input into a trained benefit prediction model to predict the corresponding economic benefits; the environmental data and set labels are used as research data, and the research data are input into a trained impact prediction model to predict the corresponding environmental impact; the training process of both the benefit prediction model and the impact prediction model is consistent with the training process of the rate estimation model, and both are neural network models;

[0061] In step B8, the method for determining whether to update the optimal solution is as follows: if the evaluation value of the optimal candidate solution is greater than the evaluation value of the optimal solution, then the optimal solution is updated to the optimal candidate solution; if the evaluation value of the optimal candidate solution is less than or equal to the evaluation value of the optimal solution, then the optimal solution is not updated.

[0062] The technical effects and advantages of the mineral resource estimation system based on multi-source heterogeneous big data fusion proposed in this invention are as follows:

[0063] By acquiring mineral characteristic data, a preliminary assessment of mining methods is made; and based on these methods, a dynamic and accurate estimate of the mining rate range and mining cycle of mineral resources is achieved. Simultaneously, by integrating multi-source heterogeneous data, including environmental data and historical market data, dynamic optimization of mining methods and rates throughout the entire mining cycle is realized. This not only overcomes the limitations of traditional static estimation methods and enhances the dynamic management and optimization capabilities of mineral resources, but also achieves a balance between economic benefits and environmental impact by optimizing mining schemes, thereby improving the economic benefits and environmental protection level of mineral resources and enhancing the scientific nature and sustainability of mineral development. Attached Figure Description

[0064] Figure 1 This is a schematic diagram of a mineral resource estimation system based on multi-source heterogeneous big data fusion according to Embodiment 1 of the present invention;

[0065] Figure 2 This is a flowchart of the mining scheme optimization method according to Embodiment 1 of the present invention. Detailed Implementation

[0066] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0067] Example 1

[0068] Please see Figure 1 As shown in the figure, the mineral resource estimation system based on multi-source heterogeneous big data fusion described in this embodiment includes a data acquisition module, a data analysis module, a cycle estimation module, a data collection module, and a mining optimization module; each module is connected by wired and / or wireless means to realize data transmission between modules;

[0069] The data acquisition module is used to acquire mineral characteristic data.

[0070] Mineral resource characteristics data include reserves data, grade data, and distribution data;

[0071] Reserves data refers to the amount of exploitable mineral resources in an ore body, usually expressed as the volume or mass of ore. Grade data refers to the content of valuable minerals (such as metallic minerals) in the ore, which is an important indicator determining the economic value of the mineral. Distribution data refers to the spatial distribution of the ore body on the surface or underground, including the size, shape, depth, and direction of extension of the ore body, used to clarify the structure and characteristics of the ore body. It should be noted that the methods for obtaining mineral characteristic data are all existing technologies, which will not be elaborated on here. Mineral characteristic data will affect the choice of mining method. Different mineral characteristic data are suitable for different mining methods. For example, when the grade data is high and the reserves data is large, the appropriate mining method is open-pit mining to reduce the mining cost per unit of ore. When the grade data is low and the reserves data is small, the appropriate mining method is underground mining to reduce the unnecessary mining of low-grade ore.

[0072] The data analysis module is used to analyze mineral characteristic data and evaluate mining methods.

[0073] Methods for evaluating mining methods include:

[0074] Different numerical labels are assigned to different mining methods, which are then marked as method labels. Mining methods include open-pit mining, underground mining, and hydraulic mining. Mineral characteristic data is input into the trained method evaluation model, which outputs the corresponding method labels. The corresponding mining method is then obtained based on these labels. The specific training process of the method evaluation model includes:

[0075] The method evaluation model adopts a random forest regression model. A set of historical data (a) is collected in advance. The historical data includes mineral characteristic data and method labels corresponding to the mineral characteristic data. The method labels corresponding to the mineral characteristic data are collected by those skilled in the art during the historical mineral resource estimation process. A set of historical data (a) is collected, and the corresponding mining methods are obtained sequentially based on practical experience under the conditions of each set of historical data. The corresponding method labels are set for each set of historical data (a). B tree subsets are sampled from the set of historical data (a). Each tree subset includes c sets of historical data. Each decision tree is trained based on the b tree subsets. Here, a and b are both integers greater than 1, and b < a, a = bc.

[0076] The initial model parameters, including the number of decision trees b, the maximum depth of each decision tree d, and the minimum number of samples per internal node u, are preset by those skilled in the art based on the actual situation.

[0077] The first decision tree is constructed recursively using a subset of trees:

[0078] Divide the historical data of c groups in a subset of trees into a training set α and a validation set β. p Let t be the p-th historical data set in the training set α, where p∈[1,α′], α′<c, and α′ is the number of historical data sets in the training set α; use feature t h Represents a data point in the mineral characteristic data, and calculates the h-th feature t. h Information gain ratio Z(α,t) h ), Z(α,t h The expression for ) is: In the formula, h∈[1,k], k is the number of data types in the mineral feature data, and αg is the training set α based on feature t. h The decision subset is divided, g∈[1,k′], where k′ is an integer greater than 1. H(αg) is the information entropy of αg, and H(α) is the information entropy of the training set α. The calculation methods of H(α) and H(αg) are as follows:

[0079]

[0080]

[0081] In the formula, αg p This represents the method label corresponding to the p-th group of historical data in αg;

[0082] For each feature t h The corresponding information gain ratio Z(α,t) h The calculation is performed, and the maximum information gain ratio Z(α,t) is selected. h The feature t corresponding to ) h As an internal node, and based on the maximum information gain ratio Z(α,t)h The feature t corresponding to ) h Divide the training set α into decision subsets such that the number of historical data points in each decision subset is the same;

[0083] For each, based on the maximum information gain Z(α,t) h The decision tree is divided into decision subsets, and the algorithm recursively repeats the process until the number of historical data in the decision subset is less than or equal to the minimum number of samples u of the internal nodes, or the number of recursions is greater than or equal to the maximum depth d of each decision tree, at which point the recursion ends; the trained decision tree model is evaluated using the validation set β, and the construction of the first decision tree model is finally completed.

[0084] b-1 decision trees are trained sequentially using the remaining b-1 tree subsets; the generation process of each decision tree is independent and the same; finally, a random forest containing b decision trees is constructed.

[0085] Calculate the loss function value on b decision trees; input mineral feature data into each decision tree, record the decision output, which is the mode label output by each decision tree, and mark it as the prediction label; calculate the mean square error between the average of all prediction labels and the true value, and use it as the loss function value, where the true value is the mode label corresponding to the mineral feature data input into the decision tree; reduce the loss function value through parameter tuning, and select the model hyperparameter with the smallest loss function value as the model hyperparameter for mode evaluation model.

[0086] The cycle estimation module estimates the mining cycle based on the mining method.

[0087] Methods for estimating mining cycles include:

[0088] Based on reserve data, distribution data, and mining methods, the mining rate range is estimated; a pre-set attenuation factor q is used, which is preset by those skilled in the art according to actual conditions; the expression for the change in mining rate is: V(t) = V min +(V max -V min )×e -qt In the formula, V(t) is the mining rate corresponding to mining time t, V min V is the minimum extraction rate within the extraction rate range. max The maximum extraction rate is defined as e, where e is the natural constant and t is the extraction time. The expression for the reserves data is: In the formula, S represents the reserve data, and T represents the mining cycle. The numerical value of the mining cycle is obtained from the expression of the reserve data using numerical solution methods (such as Newton's method, bisection method, etc.).

[0089] It should be understood that during the mining process, the mining rate is usually higher in the early stages of mining, but as mining progresses, changes in factors such as ore body conditions, mining methods, equipment use, and ore grade will cause the mining rate to gradually decrease.

[0090] The steps for estimating the range of extraction rates include:

[0091] Step A1: Set the search interval [m,n] and precision ε;

[0092] Step A2: Determine the search sequence F(x);

[0093] Step A3: Determine the partition coefficient F(r) from the search sequence;

[0094] Step A4: Divide the search interval using a partitioning coefficient to obtain two points ψ1 and ψ2;

[0095] Step A5: Calculate the feasibility of ψ1 and ψ2 respectively;

[0096] Step A6: Update the search range;

[0097] Step A7: Calculate the width υ of the search interval;

[0098] Step A8: If the width υ is less than or equal to the precision ε, then the search interval is taken as the mining rate range; if the width υ is greater than the precision ε, then return to step A3.

[0099] In step A1 above, the search interval [m,n] and precision ε are preset by those skilled in the art based on the actual situation.

[0100] In step A2 above, the search sequence

[0101] In step A3 above, the method for determining the partitioning coefficient F(r) is as follows: calculate the initial partitioning coefficient δ′. Subtract the initial splitting coefficient from each value in the search sequence to obtain the numerical difference; sort each numerical difference from smallest to largest, obtain the value in the search sequence corresponding to the numerical difference at the top of the list, and use it as the splitting coefficient F(r).

[0102] In step A4 above, In the formula, F(r-2) represents the two values ​​in the search sequence that rank first and second to the dividing coefficient F(r), and F(r-1) represents the value in the search sequence that ranks first to the dividing coefficient F(r).

[0103] In step A5 above, the feasibility is obtained as follows: Reserve data, distribution data, mode labels, and the corresponding numerical values ​​of points are used as analysis data. This analysis data is input into the trained rate estimation model to predict the corresponding feasibility. The training process of the rate estimation model includes:

[0104] y sets of analysis data are collected in advance, and a corresponding feasibility degree is set for each of the y sets of analysis data, where y is an integer greater than 1. The analysis data and the corresponding feasibility degree are converted into a set of feature vectors. The feasibility degree corresponding to the analysis data is determined by a person skilled in the art during the historical mineral resource estimation process. The y sets of analysis data are collected, and each set of analysis data is analyzed in turn based on practical experience. The feasibility score of each set of analysis data is evaluated and used as the corresponding feasibility degree. The corresponding feasibility degree is set for each of the y sets of analysis data in turn.

[0105] Each set of feature vectors is used as input to the rate estimation model. The rate estimation model outputs a set of predicted feasibility values ​​corresponding to each set of analyzed data, and uses the actual feasibility value corresponding to each set of analyzed data as the prediction target. The actual feasibility value is the pre-set feasibility value corresponding to the analyzed data. The training objective is to minimize the sum of prediction errors for all analyzed data. The formula for calculating the prediction error is as follows: Where η w The prediction error is represented by w, where w is the group number of the feature vector corresponding to the analyzed data. Let ρ be the prediction feasibility corresponding to the w-th set of analysis data. w Let w represent the actual feasibility corresponding to the w-th set of analytical data; train the rate estimation model until the sum of prediction errors converges, at which point training stops.

[0106] The aforementioned rate estimation model is specifically a deep neural network model, which includes an input layer, hidden layers, and an output layer. Each hidden layer contains multiple neurons, and each neuron is connected to the neurons in the next layer. The connections contain weights that determine the importance and impact of data transmission in the neural network. An activation function is applied to each neuron between the hidden layer and the output layer. The activation function introduces non-linearity, allowing the network to learn more complex patterns and features.

[0107] In step A6 above, the method for updating the search interval is as follows: mark the feasibility corresponding to ψ1 as the first feasibility. The feasibility corresponding to ψ2 is marked as the second feasibility. like The search range is then updated to Right now like The search range is then updated to Right now

[0108] In step A7 above, the maximum value of the search interval is subtracted from the minimum value to obtain the width υ.

[0109] The data collection module is used to collect data on the impact of mineral resources.

[0110] Mineral impact data includes historical market data and environmental data; historical market data includes historical supply and demand data and historical price data.

[0111] Historical supply and demand data represents the production and consumption of mineral resources at historical moments. This data helps to understand the supply and demand situation of mineral resources in the market. When production is less than consumption, it means that the market demand for mineral resources is greater, so it is necessary to increase the mining rate, and vice versa. Historical supply and demand data is obtained from mineral supply and demand reports published by mineral industry research institutions or mineral resource production and demand reports published by mineral industry associations. Historical price data represents the transaction price of mineral resources at historical moments. The higher the transaction price, the greater the economic benefits of mining mineral resources, so it is necessary to increase the mining rate, and vice versa. Historical price data is obtained from mineral trading platforms or mineral price websites.

[0112] Environmental data refers to information about the natural environment surrounding the ore body, such as water sources, air quality, soil conditions, and vegetation cover. It is used to constrain mineral resource extraction activities to ensure that the extraction process does not cause environmental damage. If the extraction activities bring environmental risks (such as water pollution, air quality deterioration, etc.), it is necessary to adjust the extraction method or reduce the extraction rate to comply with environmental protection requirements and reduce ecological damage. Environmental data is obtained through remote sensing technology and satellite imagery, geographic information systems and environmental monitoring systems (such as water quality monitoring stations, meteorological stations, etc.).

[0113] The mining optimization module optimizes mining plans based on mineral impact data and mining cycles.

[0114] Historical supply and demand data are input into the supply and demand forecasting model to predict future supply and demand data, which includes the production and consumption of mineral resources during the mining cycle. Historical price data are input into the price forecasting model to predict future price data, which includes the transaction price of mineral resources during the mining cycle. Both the supply and demand forecasting model and the price forecasting model are LSTM models in the recurrent neural network model. The LSTM model is an existing technology, and the specific training process will not be elaborated on here. Based on the future supply and demand data and the future price data, the mining plan during the mining cycle is optimized, and the mining plan includes the mining rate and mining method.

[0115] like Figure 2 As shown, the steps to optimize the mining plan include:

[0116] Step B1: Construct N mining sets, assign different numerical labels to different mining sets and mark them as set labels, and construct a mining matrix based on all set labels;

[0117] Step B2: Preset the constraint table size and maximum number of iterations, and determine the evaluation function;

[0118] Step B3: Randomly select a set label from the mining matrix as the current solution and add it to the constraint table. The current solution is then taken as the optimal solution.

[0119] Step B4: Based on the mining matrix, obtain the set labels adjacent to the current solution and use them as adjacent solutions;

[0120] Step B5: Calculate the evaluation value for each adjacent solution;

[0121] Step B6: Determine if each adjacent solution exists in the constraint table; if it exists, mark the corresponding adjacent solution as a constraint solution; if it does not exist, treat the corresponding adjacent solution as a candidate solution.

[0122] Step B7: Mark the candidate solution with the largest evaluation value as the optimal candidate solution. If the evaluation value of the constraint solution is greater than the evaluation value of the optimal candidate solution, then update the optimal candidate solution to the constraint solution. If the evaluation value of the constraint solution is less than or equal to the evaluation value of the optimal candidate solution, then delete the constraint solution.

[0123] Step B8: Update the current solution to the optimal candidate solution, and determine whether to update the optimal solution; add all candidate solutions to the constraint table;

[0124] Step B9: Determine if the number of iterations is equal to the maximum number of iterations; if yes, terminate the iteration and obtain the mining set corresponding to the set label of the optimal solution; if no, increment the number of iterations and return to step B4.

[0125] In step B1 above, a mining method is randomly selected from all mining methods, and a value is randomly selected from the mining rate range to construct a mining set. A total of N mining sets are constructed, and all N mining sets are different.

[0126] In step B2 above, the constraint table size and the maximum number of iterations are preset by those skilled in the art based on the actual situation; the expression of the evaluation function is: In the formula, f is the evaluation value, jx is the economic benefit, and hy is the degree of environmental impact. All are weighting coefficients; the specific values ​​of the weighting coefficients in the formula can be set according to the actual situation. The weighting coefficients reflect the degree of influence of economic benefits and environmental impact on the evaluation value. Those skilled in the art can preset the corresponding weighting coefficients according to the actual degree of influence of economic benefits and environmental impact on the evaluation value, so as to accurately evaluate the comprehensive benefits of mining methods and mining rates, that is, to achieve a balance between economic benefits and environmental impact, so that the mining scheme can meet economic goals while minimizing the negative impact on the environment; the evaluation value is calculated using dimensionless calculation.

[0127] The methods for obtaining economic benefits and environmental impact are as follows: Based on future supply and demand data and future price data, predicted supply and demand data and predicted price data are obtained; the predicted supply and demand data is one data point from the future supply and demand data, and the predicted price data is one data point from the future price data, with the predicted supply and demand data and predicted price data corresponding to the same time period; the predicted supply and demand data, predicted price data, and set labels are used as test data, and the test data are input into a trained benefit prediction model to predict the corresponding economic benefits; the environmental data and set labels are used as research data, and the research data are input into a trained impact prediction model to predict the corresponding environmental impact; the training process of both the benefit prediction model and the impact prediction model is consistent with the training process of the rate estimation model, and both are neural network models.

[0128] In step B8 above, the method for determining whether to update the optimal solution is as follows: if the evaluation value of the optimal candidate solution is greater than the evaluation value of the optimal solution, then the optimal solution is updated to the optimal candidate solution; if the evaluation value of the optimal candidate solution is less than or equal to the evaluation value of the optimal solution, then the optimal solution is not updated.

[0129] This embodiment acquires mineral characteristic data to preliminarily assess mining methods; and based on these methods, it achieves dynamic and accurate estimation of the mining rate range and mining cycle of mineral resources. Simultaneously, it integrates multi-source heterogeneous data, including environmental data and historical market data, to achieve dynamic optimization of mining methods and rates throughout the entire mining cycle. This not only overcomes the limitations of traditional static estimation methods and enhances the dynamic management and optimization capabilities of mineral resources, but also achieves a balance between economic benefits and environmental impact by optimizing mining schemes, thereby improving the economic benefits and environmental protection level of mineral resources and enhancing the scientific nature and sustainability of mineral development.

[0130] Example 2

[0131] This application also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable code, which, when executed by the one or more processors, can perform a mineral resource estimation system based on the fusion of multi-source heterogeneous big data as described above.

[0132] The methods or systems according to the embodiments of this application can also be implemented using the architecture of the electronic device shown in this application. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to a network, input / output, a hard disk, etc. The storage device in the electronic device, such as a ROM or hard disk, may store a mineral resource estimation system based on multi-source heterogeneous big data fusion provided in this application. Furthermore, the electronic device may also include a user interface. Of course, the architecture shown in this application is merely exemplary; when implementing different devices, one or more components in the electronic device shown in this application may be omitted according to actual needs.

[0133] Example 3

[0134] One embodiment of this application discloses a computer-readable storage medium. The computer-readable storage medium stores computer-readable instructions. When executed by a processor, the computer-readable instructions can perform a mineral resource estimation system based on multi-source heterogeneous big data fusion according to an embodiment of this application, as described with reference to the above figures. The storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0135] Furthermore, according to embodiments of this application, the processes described in the above-referenced flowcharts can be implemented as computer software programs. For example, this application provides a non-transitory machine-readable storage medium storing machine-readable instructions that can be executed by a processor to perform instructions corresponding to the method steps provided in this application, such as a mineral resource estimation system based on multi-source heterogeneous big data fusion. When this computer program is executed by a central processing unit (CPU), it performs the functions defined in the method of this application.

[0136] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0137] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A mineral resource estimation system based on multi-source heterogeneous big data fusion, characterized in that, include: The data acquisition module is used to acquire mineral characteristic data, including reserve data, grade data, and distribution data. The data analysis module is used to analyze mineral characteristic data and evaluate mining methods; the methods for evaluating mining methods include: Different numerical labels are assigned to different mining methods and marked as method labels; mineral characteristic data is input into the trained method evaluation model, which outputs the corresponding method labels, and the corresponding mining method is obtained based on the method labels; the training process of the method evaluation model includes: The method evaluation model employs a random forest regression model. A set of historical data (a) is pre-collected, including mineral characteristic data and corresponding method labels. From this set of historical data, b tree subsets are sampled, each containing c sets of historical data. Each decision tree is trained based on these b tree subsets. Here, a and b are both integers greater than 1. , ; Initialize the model parameters, including the number of decision trees b, the maximum depth of each decision tree d, and the minimum number of samples u for each internal node; The first decision tree is constructed recursively using a subset of trees, and then the remaining trees are used sequentially. Training a subset of trees b decision trees; each decision tree is generated independently and in the same way; finally, a random forest containing b decision trees is constructed. Calculate the loss function value on b decision trees; input mineral feature data into each decision tree, record the decision output, which is the mode label output by each decision tree, and mark it as the prediction label; calculate the mean square error between the average of all prediction labels and the true value, and use it as the loss function value, where the true value is the mode label corresponding to the mineral feature data input into the decision tree; reduce the loss function value through parameter tuning, and select the model hyperparameter with the smallest loss function value as the model hyperparameter for mode evaluation; The method of recursively constructing the first decision tree using a subset of trees includes: Divide the c groups of historical data in a subset into a training set. and verification set , For training set The p-th group of historical data, , , For training set The amount of historical data in the data; using features This represents a data point in the mineral characteristic data; calculate the h-th characteristic. Information gain rate ; For each feature The corresponding information gain rate Perform calculations and select the maximum information gain ratio. Corresponding features As an internal node, and based on the maximum information gain rate Corresponding features training set Divide the data into decision subsets, with each subset containing the same amount of historical data. For each based on the maximum information gain rate The algorithm recursively processes the divided decision subsets until the number of historical data in the subset is less than or equal to the minimum number of samples u of the internal nodes, or the number of recursions is greater than or equal to the maximum depth d of each decision tree, at which point the recursion ends. The trained decision tree model is then validated using a validation set. The evaluation was conducted, and the first decision tree model was finally completed. The expression is: In the formula, k is the number of data types in the mineral characteristic data. For training set Based on characteristics The decision subset of the partition, , It is an integer greater than 1. for Information entropy For training set Information entropy, where and The calculation methods include: ； ； In the formula, express The method label corresponding to the p-th group of historical data; The cycle estimation module estimates the mining cycle based on the mining method. The data collection module is used to collect data on the impact of mineral resources. The mining optimization module optimizes mining plans based on mineral impact data and mining cycles.

2. The mineral resource estimation system based on multi-source heterogeneous big data fusion according to claim 1, characterized in that, The method for estimating the mining cycle includes: Based on reserve data, distribution data, and mining methods, estimate the range of mining rates; preset attenuation factors. The expression for the change in mining rate is: In the formula, This represents the mining rate at mining time t. The minimum mining rate within the mining rate range. This represents the maximum extraction rate within the extraction rate range. It is a natural constant. For mining time; the expression for reserve data is: In the formula, For reserve data, The mining cycle is determined by numerical methods, which are used to solve for the value of the mining cycle from the expression of the reserve data.

3. A mineral resource estimation system based on multi-source heterogeneous big data fusion according to claim 2, characterized in that, The step of estimating the mining rate range includes: Step A1: Set the search range and accuracy ; Step A2: Determine the search sequence ; Step A3: Determine the partition coefficients from the search sequence ; Step A4: Divide the search interval using a partitioning coefficient to obtain two points. and ; Step A5: Calculate separately and Corresponding feasibility; Step A6: Update the search range; Step A7: Calculate the width of the search interval ; Step A8: If the width Less than or equal to precision If the width is... Greater than precision Then return to step A3.

4. A mineral resource estimation system based on multi-source heterogeneous big data fusion as described in claim 3, characterized in that, In step A2, the search sequence ; In step A3, the partitioning coefficient is determined. The method is as follows: calculate the initial partition coefficients. , Subtract the initial splitting coefficient from each value in the search sequence to obtain the numerical difference; sort each numerical difference in ascending order, and obtain the value in the search sequence corresponding to the first numerical difference, which is then used as the splitting coefficient. ; In step A4 , In the formula, For the search sequence ranked by the partition coefficient The first two values, For the search sequence ranked by the partition coefficient The value of the previous digit; In step A5, the feasibility is obtained by using the following methods: storage data, distribution data, mode labels, and the corresponding values ​​of points as analysis data, inputting the analysis data into the trained rate estimation model, and predicting the corresponding feasibility. In step A6, the method for updating the search interval is as follows: The corresponding feasibility level is marked as the first feasibility level. ,Will The corresponding feasibility level is marked as the second feasibility level. ;like Then the search range is updated to ,Right now ;like Then the search range is updated to ,Right now ; In step A7, the maximum value of the search interval is subtracted from the minimum value to obtain the width. .

5. A mineral resource estimation system based on multi-source heterogeneous big data fusion according to claim 4, characterized in that, The training process for the rate estimation model includes: Collect y sets of analysis data in advance, and set a corresponding feasibility for each of the y sets of analysis data, where y is an integer greater than 1. Convert the analysis data and the corresponding feasibility into a set of corresponding feature vectors. Each set of feature vectors is used as input to the rate estimation model. The rate estimation model outputs a set of predicted feasibility corresponding to each set of analyzed data and uses the actual feasibility corresponding to each set of analyzed data as the prediction target. The actual feasibility is the pre-set feasibility corresponding to the analyzed data. The training objective is to minimize the sum of prediction errors of all analyzed data. The rate estimation model is trained until the sum of prediction errors converges and then training stops. The rate estimation model is a deep neural network model.

6. A mineral resource estimation system based on multi-source heterogeneous big data fusion according to claim 5, characterized in that, The mineral impact data includes historical market data and environmental data; the historical market data includes historical supply and demand data and historical price data; the historical supply and demand data represents the production and consumption of mineral resources at a historical moment; the historical price data represents the transaction price of mineral resources at a historical moment. Historical supply and demand data are input into a supply and demand forecasting model to predict future supply and demand data, which includes the production and consumption of mineral resources during the mining cycle. Historical price data are input into a price forecasting model to predict future price data, which includes the transaction price of mineral resources during the mining cycle. Both the supply and demand forecasting model and the price forecasting model are LSTM models in the recurrent neural network model. Based on the future supply and demand data and future price data, the mining plan during the mining cycle is optimized, and the mining plan includes the mining rate and mining method. The steps for optimizing the mining scheme include: Step B1: Construct N mining sets, assign different numerical labels to different mining sets and mark them as set labels, and construct a mining matrix based on all set labels; Step B2: Preset the constraint table size and maximum number of iterations, and determine the evaluation function; Step B3: Randomly select a set label from the mining matrix as the current solution and add it to the constraint table. The current solution is then taken as the optimal solution. Step B4: Based on the mining matrix, obtain the set labels adjacent to the current solution and use them as adjacent solutions; Step B5: Calculate the evaluation value for each adjacent solution; Step B6: Determine if each adjacent solution exists in the constraint table; if it exists, mark the corresponding adjacent solution as a constraint solution; if it does not exist, treat the corresponding adjacent solution as a candidate solution. Step B7: Mark the candidate solution with the largest evaluation value as the optimal candidate solution. If the evaluation value of the constraint solution is greater than the evaluation value of the optimal candidate solution, then update the optimal candidate solution to the constraint solution. If the evaluation value of the constraint solution is less than or equal to the evaluation value of the optimal candidate solution, then delete the constraint solution. Step B8: Update the current solution to the optimal candidate solution, and determine whether to update the optimal solution; add all candidate solutions to the constraint table; Step B9: Determine if the number of iterations is equal to the maximum number of iterations; if yes, terminate the iteration and obtain the mining set corresponding to the set label of the optimal solution; if no, increment the number of iterations and return to step B4.

7. A mineral resource estimation system based on multi-source heterogeneous big data fusion according to claim 6, characterized in that, In step B1, one mining method is randomly selected from all mining methods, and a value is randomly selected from the mining rate range to construct a mining set. A total of N mining sets are constructed, and all N mining sets are different. In step B2, the expression for the evaluation function is: In the formula, As an evaluation value, For economic benefits, For environmental impact, , All are weighting coefficients; The methods for obtaining economic benefits and environmental impact are as follows: Based on future supply and demand data and future price data, predicted supply and demand data and predicted price data are obtained; the predicted supply and demand data is one data point from the future supply and demand data, and the predicted price data is one data point from the future price data, with the predicted supply and demand data and predicted price data corresponding to the same time period; the predicted supply and demand data, predicted price data, and set labels are used as test data, and the test data are input into a trained benefit prediction model to predict the corresponding economic benefits; the environmental data and set labels are used as research data, and the research data are input into a trained impact prediction model to predict the corresponding environmental impact; the training process of both the benefit prediction model and the impact prediction model is consistent with the training process of the rate estimation model, and both are neural network models; In step B8, the method for determining whether to update the optimal solution is as follows: if the evaluation value of the optimal candidate solution is greater than the evaluation value of the optimal solution, then the optimal solution is updated to the optimal candidate solution; if the evaluation value of the optimal candidate solution is less than or equal to the evaluation value of the optimal solution, then the optimal solution is not updated.

Images