Adaptive variable-length mineral resource reserve estimation method, device and equipment

Abstract

The application discloses a kind of self-adapting variable length mineral resources reserves estimation method, device and equipment, it is related to the cross technical field of geostatistics and intelligent mineral resources reserves estimation, can solve the technical problem of low mineral resources reserves estimation precision and poor self-adaptive ability caused by equalization preprocessing and single fixed power index of traditional distance power inverse ratio method.The method comprises: determining the effective sample corresponding to each to-be-estimated block of to-be-estimated mining area, for each to-be-estimated block, according to the adaptive coupling coefficient, the distance attenuation coefficient, the relative distance of each corresponding effective sample, the sample length influence coefficient and the relative original sample length of each corresponding effective sample, the weight term of each corresponding effective sample is constructed, the optimal value of weight term parameter is determined based on particle swarm algorithm, the target grade value of to-be-estimated block is calculated according to the optimal value of parameter and weight term, and the total mineral resources reserves of to-be-estimated mining area is calculated according to the target grade value of all to-be-estimated blocks.

Description

Technical Field

[0001] This invention relates to the interdisciplinary field of geostatistics and intelligent mineral resource estimation, and in particular to an adaptive method, apparatus and equipment for estimating mineral resource reserves with variable sample length. Background Technology

[0002] With the deepening of global mineral resource exploration and development, accurate mineral reserve estimation has become one of the core technologies for mineral resource management and development. The traditional inverse distance weighting (IDW) method is widely used due to its simple principle and high computational efficiency, but it has two inherent drawbacks: Firstly, the inverse power law method simplifies three-dimensional solid samples into point data for processing, fundamentally ignoring the important geostatistical principle of spatial variability. Spatial variability indicates that the volume, geometry, and spatial orientation of a sample significantly affect the estimated variance of regionalized variables. The longer the sample, the larger the spatial average range represented by its grade value, and the amount of information it carries differs from that of a short sample. While traditional equal-length preprocessing simplifies calculations, it clearly leads to distortion of the original data, especially when the sample length variation coefficient is large. This can result in inaccurate identification of thin-layer orebody boundaries and systematic biases in overall estimation, i.e., low estimation accuracy.

[0003] Furthermore, the inverse power law distance method relies on a single, fixed power law distance exponent, the selection of which depends on expert experience and lacks adaptability. In complex ore bodies (such as those with anisotropy or multi-layered structures), a single parameter is insufficient to simultaneously capture spatially correlated global trends and local variations, resulting in poor model generalization and adaptability, as well as strong subjectivity in the modeling results, ultimately leading to low estimation accuracy. Summary of the Invention

[0004] In view of this, the present invention provides an adaptive method, apparatus and equipment for estimating mineral resource reserves with variable sample length, which can solve the technical problems of low estimation accuracy and poor adaptability caused by the traditional distance power inverse ratio method due to equal length preprocessing and a single fixed power exponent.

[0005] According to one aspect of the present invention, an adaptive variable-length mineral resource reserve estimation method is provided, the method comprising: Obtain the coordinates of each block to be estimated in the mining area to be estimated, obtain the sample coordinates, grade value and original sample length of each borehole sample in the mining area to be estimated, determine the effective sample corresponding to each block to be estimated in all the borehole samples based on the coordinates of the block to be estimated and the sample coordinates of the effective sample. For each block to be estimated, the relative distance between the block coordinates and the effective sample coordinates is calculated. The relative original sample length is calculated based on the original sample length of each effective sample. A weight term for each effective sample is constructed based on the adaptive coupling coefficient, distance attenuation coefficient, relative distance of the effective samples, sample length influence coefficient, and relative original sample length. Geological rationality constraints are configured for the parameters of the weight term of the effective samples. The weight term of the effective samples is as follows:

[0006] In the formula, This represents the adaptive coupling coefficient. This represents the distance attenuation coefficient. This represents the sample length influence coefficient. This represents the weight term for valid sample i. This represents the relative distance to the valid sample i. Indicates the relative original sample length of valid sample i; For each block to be estimated, under the geological rationality constraint, based on the particle swarm optimization algorithm, the optimal value of the parameter is determined according to the grade value. The target grade value of the block to be estimated is calculated according to the optimal value of the parameter and the weight term of the effective sample. The sub-mineral resource reserves of the block to be estimated are calculated according to the target grade value. The parameter includes the adaptive coupling coefficient, the distance attenuation coefficient and the sample length influence coefficient. The optimal value of the parameter includes the optimal value of the adaptive coupling coefficient, the optimal value of the distance attenuation coefficient and the optimal value of the sample length influence coefficient. The total mineral resource reserves of the mining area to be estimated are calculated based on the reserves of all the sub-mineral resources.

[0007] Preferably, the step of calculating the relative distance of each valid sample based on the coordinates of the block to be estimated and the coordinates of the valid samples, and calculating the relative original sample length of each valid sample based on the original sample length of each valid sample, includes: Calculate the distance between the coordinates of the block to be estimated and the coordinates of each corresponding valid sample. Add all the distances together and divide by the number of all valid samples to obtain the average distance. Divide each distance by the average distance to obtain the relative distance of each corresponding valid sample. The average original sample length is obtained by summing the original sample lengths of all the valid samples and dividing by the total number of valid samples. The relative original sample length of each valid sample is obtained by dividing the original sample length of each valid sample by the average original sample length.

[0008] Preferably, the geological rationality constraint of the parameter configuration for the weighting term of the valid sample includes: Configure the adaptive coupling coefficient with a geological rationality constraint that is greater than or equal to 0.7 and less than or equal to 0.9; Both the distance attenuation coefficient and the sample length influence coefficient are subject to geological rationality constraints, with the distance attenuation coefficient being greater than the sample length influence coefficient.

[0009] Preferably, the step of determining the optimal value of the parameter based on the grade value under the geological rationality constraint and using a particle swarm optimization algorithm includes: Set the particle swarm size and the maximum number of iterations; In the current iteration, for each particle, under the geological rationality constraint, the current position corresponding to the particle is determined as the current parameter to be optimized value of the parameter of the weight term of the effective sample. The verification block coordinates and actual grade values ​​of each verification block in the mining area to be estimated are obtained. The grade estimate value of each verification block is calculated based on the verification block coordinates and the current parameter to be optimized value. Based on the formula for calculating the mean square error, a fitness function is constructed. The current fitness value of the fitness function is calculated based on all the grade estimates and all the actual grade values. The global optimal fitness value is determined based on the current fitness values ​​of all the particles in the particle swarm, and the position corresponding to the global optimal fitness value is determined as the global optimal position. If the position corresponding to the global optimal fitness value is the current position, then the current position is the global optimal position. If the current iteration number is the maximum iteration number, or the change between the global optimal fitness value and the historical global optimal fitness value is less than a preset threshold, then the iteration stops, and the global optimal position is taken as the optimal value of the parameter. If the current position is the global optimal position, then the current parameter to be optimized value is the optimal value of the parameter.

[0010] Preferably, the step of calculating the grade estimate of each verification block based on the verification block coordinates and the current parameter to be optimized includes: For each of the verification blocks, the effective verification sample in the borehole sample is determined according to the coordinates of the verification block and the coordinates of the sample. The sample coordinates, grade value and original sample length of the effective verification sample are respectively determined as the effective verification sample coordinates, effective verification grade value and effective verification block original sample length. For each of the verification blocks, the relative distance and the relative original sample length of the verification block are calculated. Based on the relative distance of the effective verification samples, the relative original sample length of the effective verification samples, and the current parameter to be optimized, the weight term of each effective verification sample corresponding to the verification block is calculated. Based on the weight term of each effective verification sample and the effective verification grade value, the grade estimate of the verification block is calculated.

[0011] Preferably, calculating the target grade value of the block to be estimated based on the optimal value of the parameters and the weight term of the effective samples includes: Substituting the optimal value of the parameter into the weight term of the effective sample, the optimal weight solution for each effective sample corresponding to the block to be estimated is obtained; The grade value of the effective sample is determined as the effective grade value. The product of the effective grade value of the same effective sample and the weighted optimal solution is calculated to obtain the first product value. The first product values ​​of all effective samples are added together and divided by the number of all effective samples to obtain the target grade value of the block to be estimated.

[0012] Preferably, the step of calculating the sub-mineral resource reserves of the block to be estimated based on the target grade value includes: Obtain the volume and density of each of the blocks to be estimated; The sub-mineral resource reserves of the same block to be estimated are obtained by multiplying its volume, density and target grade value. The step of calculating the total mineral resource reserves of the mining area to be estimated based on the reserves of all the sub-mineral resources includes: The total mineral resource reserves of the mining area to be estimated are obtained by summing up the reserves of all the sub-mineral resources.

[0013] According to another aspect of the present invention, an adaptive variable-length mineral resource reserve estimation device is provided, the device comprising: The data acquisition and neighborhood search module is used to acquire the coordinates of each block to be estimated in the mining area to be estimated, acquire the sample coordinates, grade value and original sample length of each borehole sample in the mining area to be estimated, determine the effective sample corresponding to each block to be estimated in all the borehole samples based on the coordinates of the block to be estimated and the sample coordinates of the effective sample. The dimensionless modeling and construction module is used to calculate the relative distance between each effective sample and the coordinates of the block to be estimated, based on the block coordinates and the effective sample coordinates; calculate the relative original sample length based on the original sample length of each effective sample; construct a weight term for each effective sample based on the adaptive coupling coefficient, distance attenuation coefficient, relative distance of the effective samples, sample length influence coefficient, and relative original sample length of the effective samples; and configure geological rationality constraints on the parameters of the weight term of the effective samples. The weight term of the effective samples is as follows:

[0014] In the formula, This represents the adaptive coupling coefficient. This represents the distance attenuation coefficient. This represents the sample length influence coefficient. This represents the weight term for valid sample i. This represents the relative distance to the valid sample i. Indicates the relative original sample length of valid sample i; The parameter adaptive optimization module is used to determine the optimal value of the parameters for each block to be estimated under the geological rationality constraint, based on the particle swarm optimization algorithm and the grade value, calculate the target grade value of the block to be estimated based on the optimal value of the parameters and the weight term of the effective sample, and calculate the sub-mineral resource reserves of the block to be estimated based on the target grade value. The parameters include the adaptive coupling coefficient, the distance attenuation coefficient, and the sample length influence coefficient, and the optimal value of the parameters includes the optimal value of the adaptive coupling coefficient, the optimal value of the distance attenuation coefficient, and the optimal value of the sample length influence coefficient. The reserve calculation module is used to calculate the total mineral resource reserves of the mining area to be estimated based on the reserves of all the sub-mineral resources.

[0015] According to another aspect of the present invention, a storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the above-described adaptive variable-length mineral resource reserve estimation method.

[0016] According to another aspect of the present invention, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor, when executing the program, implements the above-described adaptive variable-length mineral resource reserve estimation method.

[0017] By employing the above technical solutions, this invention provides an adaptive variable-length sample length mineral resource reserve estimation method, apparatus, and equipment. Firstly, by preserving the original sample length of the borehole samples, it eliminates the need for traditional equal-length preprocessing, avoiding information distortion caused by forced equal-length processing and improving estimation accuracy. Secondly, it innovatively constructs a dual-weighted coupling model of sample length and distance, i.e., the weight term of the effective sample. Under geological constraints, it ensures the dominance of spatial distance while incorporating the volume effect represented by the original sample length, thereby accurately characterizing the contribution of samples at different scales in spatial interpolation. Finally, it introduces a particle swarm optimization algorithm for iterative optimization, achieving adaptive optimization of model parameters, avoiding information distortion caused by a single power exponent, and overcoming the shortcomings of traditional methods that rely on fixed parameters and lack intelligent tuning capabilities. This invention significantly improves the accuracy and adaptability of mineral resource reserve estimation.

[0018] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings: Figure 1 A flowchart illustrating an adaptive variable sample length mineral resource reserve estimation method provided by an embodiment of the present invention is shown. Figure 2 A flowchart illustrating another adaptive variable sample length mineral resource reserve estimation method provided by an embodiment of the present invention is shown. Figure 3 This diagram illustrates the structure of an adaptive variable sample length mineral resource reserve estimation device provided in an embodiment of the present invention. Figure 4 This invention provides a schematic diagram of another adaptive variable sample length mineral resource reserve estimation device. Figure 5 The graph shows the estimated results of three methods for estimating mineral resource reserves using real data. Figure 6 The diagram shows the accuracy improvement rate of real data using the adaptive variable sample length mineral resource reserve estimation method provided in the embodiments of the present invention. Figure 7 The diagram shows a comparison of the accuracy indicators of the simulated data and a comparison of the accuracy improvement. Detailed Implementation

[0020] The present invention will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0021] This embodiment provides an adaptive method for estimating mineral resource reserves with variable sample length, such as... Figure 1 As shown, the method includes: 101. Obtain the coordinates of each block to be estimated in the mining area to be estimated, obtain the sample coordinates, grade value and original sample length of each borehole sample in the mining area to be estimated, determine the effective sample corresponding to each block to be estimated in all the borehole samples based on the coordinates of the block to be estimated and the sample coordinates of the effective sample.

[0022] Among them, the mining area to be estimated is the mining area where mineral resource reserves are to be estimated. The blocks to be estimated come from the mining area to be estimated and are three-dimensional blocks virtually divided from the mining area to be estimated. All the blocks to be estimated constitute the mining area to be estimated. What is known about the blocks to be estimated is their coordinates (i.e., the position of the blocks to be estimated in the mining area to be estimated, which can be the coordinates of the center point of the blocks to be estimated, which is not limited here) and volume. It is necessary to calculate the target grade value of each block to be estimated through steps 101-103 of the embodiment in order to estimate the mineral resource reserves of the mining area to be estimated.

[0023] The target grade value for each block to be estimated is calculated based on borehole samples. The borehole samples are ore segments of varying lengths taken from boreholes in the mining area to be estimated. What is known about the borehole samples are their coordinates (i.e., the location of the borehole sample in the mining area to be estimated), grade value (including ore type and content), and original sample length (i.e., the actual length of the borehole sample when it was drilled out underground. The longer the original sample length, the wider the underground area it covers, the more geological information it contains, and the greater its weight).

[0024] Existing technologies change the original sample length of borehole samples to a uniform length, ignoring the different weights represented by different lengths, resulting in information distortion. In contrast, this embodiment retains the original sample length, thus taking into account the different weights represented by different lengths in the subsequent construction of weight terms, making the estimation of mineral resource reserves more accurate.

[0025] Based on the principle of spatial variability, the closer a borehole sample is to a given block, the greater its influence on calculating the target grade value of that block. Therefore, for each block to be estimated, a neighborhood search is performed using preset values ​​for the search ellipsoid parameters. Borehole samples within the range corresponding to these preset values ​​are selected as valid samples for that block. The target grade value of the block is calculated only based on these valid samples. The search ellipsoid parameters include the principal axis direction, radius, and dip angle. Specifically, for each block to be estimated, the distance between the block coordinates and the coordinates of each borehole sample is calculated. If this distance is within the range corresponding to the preset values, then the borehole sample is a valid sample for that block. Ultimately, at least one valid sample is obtained for each block to be estimated.

[0026] 102. For each block to be estimated, calculate the relative distance between the block coordinates and the effective sample coordinates, calculate the relative original length of each effective sample based on the original sample length, construct a weight term for each effective sample based on the adaptive coupling coefficient, distance attenuation coefficient, relative distance of the effective sample, sample length influence coefficient, and relative original sample length of the effective sample, and configure the parameters of the weight term of the effective sample with geological rationality constraints.

[0027] For each block to be estimated, the following operation is performed: Suppose there are n valid samples corresponding to a certain block to be estimated, and the coordinates of the block to be estimated are ( ), calculate the coordinates of the block to be estimated ( ) respectively with the coordinates of n valid samples (the valid sample coordinates of valid sample i are ( The distance between (i) where i is greater than or equal to 1 and less than or equal to n.

[0028]

[0029] in, This represents the effective sample coordinates of effective sample i and the coordinates of the block to be estimated. The distance between them.

[0030]

[0031] in, This indicates the average distance.

[0032]

[0033] in, Represents the relative distance of valid sample i, when When =1, it indicates that the effective sample i is at the average distance. A value less than 1 indicates a distance close to the average. A value greater than 1 indicates a distance farther than the average.

[0034] For each block to be estimated, perform the following operations:

[0035] in, This represents the original sample length of valid sample i. This represents the average original sample length.

[0036]

[0037] in, Indicates the relative original sample length of valid sample i, when When =1, it indicates that the effective sample is at the average original sample length. A value less than 1 indicates that the sample length is shorter than the average original sample length. A value >1 indicates that the sample is longer than the average original sample length.

[0038] It should be noted that, in order to eliminate the difference in physical dimensions between distance and sample length and to construct a unified weighting term, the dimensionless relative distance and the dimensionless relative original sample length are calculated.

[0039] For each block to be estimated, perform the following operations: The weighting term for the effective samples is:

[0040] In the formula, This represents the adaptive coupling coefficient. This represents the distance attenuation coefficient. This represents the sample length influence coefficient. This represents the weight term for valid sample i. This represents the relative distance to the valid sample i. This represents the relative original sample length of valid sample i.

[0041] 103. For each block to be estimated, under the geological rationality constraint, based on the particle swarm optimization algorithm, determine the optimal value of the parameter according to the grade value, calculate the target grade value of the block to be estimated according to the optimal value of the parameter and the weight term of the effective sample, and calculate the sub-mineral resource reserves of the block to be estimated according to the target grade value.

[0042] The parameters include the adaptive coupling coefficient, the distance attenuation coefficient, and the sample length influence coefficient, and the optimal parameter values ​​include the optimal values ​​of the adaptive coupling coefficient, the optimal values ​​of the distance attenuation coefficient, and the optimal values ​​of the sample length influence coefficient.

[0043] The particle swarm optimization algorithm obtains the optimal values ​​of parameters through iterative optimization. The optimal values ​​of parameters are then substituted into the weight terms of the effective samples to obtain the optimal weight solution for each effective sample of the block to be estimated. The target grade value of the block to be estimated is then calculated, and the sub-mineral resource reserves of the block to be estimated are then calculated.

[0044] 104. Calculate the total mineral resource reserves of the mining area to be estimated based on the reserves of all the sub-mineral resources.

[0045] It should be noted that the total mineral resource reserves of the mining area to be estimated are the sum of the sub-mineral resource reserves of all the blocks to be estimated in the mining area, and the reserves of each sub-mineral resource are calculated based on the target grade value of the corresponding block to be estimated.

[0046] The traditional inverse power law method for distance has two inherent limitations that restrict its accuracy and reliability: First, there are limitations in data preprocessing: the traditional inverse power distance method forces equalization preprocessing for borehole samples of varying lengths, which leads to information distortion.

[0047] Second, the limitations of spatial modeling: the traditional inverse power law of distance simplifies three-dimensional solid samples into point data, ignoring their physical volume effect.

[0048] This invention starts with borehole samples and constructs a dual-weighted coupling model of sample length and distance by fusing the sample's physical scale (relative original sample length after dimensionless processing) and spatial distribution characteristics (relative distance after dimensionless processing). This dual-weighted coupling model of sample length and distance constitutes the weight term, thereby achieving a key shift in geological variable estimation from traditional point-based approximation to volumetric modeling. By preserving original sample length information and intelligently optimizing parameters, this scheme effectively overcomes the information distortion and modeling bias caused by traditional methods due to equal-length processing and reliance on a single fixed parameter.

[0049] This invention establishes a dual-weighted coupling model of sample length and distance for synergistic optimization of sample physical scale and spatial distribution characteristics. The specific technical effects are as follows: (1) Preserve original data information and improve algorithm estimation accuracy: Eliminate the equal length preprocessing step to avoid the distortion of grade information caused by the forced uniform length. By constructing a dual-weight coupling model of sample length and distance, the contribution difference of samples of different scales in spatial interpolation can be accurately characterized, thereby effectively reducing interpolation error under non-uniform distribution conditions.

[0050] (2) The construction of the dual-weight coupling model of sample length and distance is reasonable: the calculation formula of the weight term is determined based on the spatial variability theory of geostatistics and the dimensional consistency theory of physics, which ensures the rationality of the method. The calculation formula of the weight term makes reasonable use of the original sample length information while maintaining the distance as the main factor, thus improving the estimation accuracy.

[0051] (3) Data processing and computation efficiency optimization: For the first time, the original sample length is quantified into a computable weight factor, abandoning the traditional manual equalization operation, simplifying the process and improving data processing efficiency; the traditional separate and multi-step process of equalization and spatial interpolation is optimized into a one-step integrated intelligent computation, and the parameters are automatically optimized through adaptive algorithms, reducing manual intervention.

[0052] (4) Parameter adaptive optimization, improved accuracy and strong adaptive capability: Design an intelligent interpolation algorithm with autonomous feedback adjustment capability, use particle swarm algorithm to adaptively optimize the parameters of the weight term, introduce geological rationality constraints in the optimization process, so that the weight term has strong adaptive capability and avoids information distortion caused by a single power exponent.

[0053] (5) Simplify the overall workflow: The traditional equal-length processing and spatial interpolation process is optimized into a one-step integrated intelligent calculation. The weight parameters are automatically determined through an adaptive parameter optimization algorithm, reducing manual intervention.

[0054] This invention provides an adaptive variable-length sample length mineral resource reserve estimation method, apparatus, and equipment. Firstly, by preserving the original sample length of the borehole samples, the traditional equal-length preprocessing is eliminated, avoiding information distortion caused by forced equal-length preprocessing and improving estimation accuracy. Secondly, an innovative dual-weighted coupled model of sample length and distance is constructed, representing the weight term of the effective sample. Under geological constraints, this model ensures the dominance of spatial distance while incorporating the volume effect represented by the original sample length, thus accurately characterizing the contribution of samples at different scales in spatial interpolation. Finally, a particle swarm optimization algorithm is introduced for iterative optimization, achieving adaptive optimization of model parameters. This avoids information distortion caused by a single power exponent and overcomes the shortcomings of traditional methods that rely on fixed parameters and lack intelligent tuning capabilities. This invention significantly improves the accuracy and adaptability of mineral resource reserve estimation.

[0055] Furthermore, as a refinement and extension of the specific implementation methods described above, and to fully illustrate the specific implementation process in this embodiment, another adaptive variable-length mineral resource reserve estimation method is provided, such as... Figure 2 As shown, the method includes: 201. Obtain the coordinates of each block to be estimated in the mining area to be estimated, obtain the sample coordinates, grade value and original sample length of each borehole sample in the mining area to be estimated, determine the effective sample corresponding to each block to be estimated in all the borehole samples based on the coordinates of the block to be estimated and the sample coordinates of the effective sample.

[0056] In this embodiment, the three-dimensional block model file of the ore body is read to obtain the coordinates and volume of each block to be estimated. Data from borehole samples is imported, and the sample coordinates, grade values, and original sample lengths of each borehole sample are extracted. Outlier filtering and integrity verification are performed on the borehole sample data.

[0057] 202. For each block to be estimated, calculate the relative distance between the block coordinates and the effective sample coordinates, calculate the relative original length of each effective sample based on the original sample length, construct a weight term for each effective sample based on the adaptive coupling coefficient, distance attenuation coefficient, relative distance of the effective sample, sample length influence coefficient, and relative original sample length of the effective sample, and configure the parameters of the weight term of the effective sample with geological rationality constraints.

[0058] In this embodiment, the step of calculating the relative distance between the estimated block coordinates and the effective sample coordinates, and calculating the relative original sample length of each effective sample based on the original sample length of each effective sample, includes: calculating the distance between the estimated block coordinates and the corresponding effective sample coordinates; summing all the distances and dividing by the number of all effective samples to obtain the average distance; dividing each distance by the average distance to obtain the relative distance of each effective sample; summing the original sample lengths of all effective samples and dividing by the number of all effective samples to obtain the average original sample length; and dividing the original sample length of each effective sample by the average original sample length to obtain the relative original sample length of each effective sample.

[0059] The dimensionless processing performed above solves the problem of inconsistent dimensions between distance and length, and enhances the adaptability of the sample length and distance dual-weight coupling model to different geostatistical units. The weight term of the constructed effective sample is the sample length and distance dual-weight coupling model.

[0060] The weighting term for the effective samples is:

[0061] In the formula, This represents the adaptive coupling coefficient. This represents the distance attenuation coefficient. This represents the sample length influence coefficient. This represents the weight term for valid sample i. This represents the relative distance to the valid sample i. This represents the relative original sample length of the effective sample i.

[0062] The geological rationality constraint for configuring the weighting term of the effective sample includes: configuring the adaptive coupling coefficient with a geological rationality constraint that the adaptive coupling coefficient is greater than or equal to 0.7 and less than or equal to 0.9; and configuring the distance attenuation coefficient and the sample length influence coefficient with a geological rationality constraint that the distance attenuation coefficient is greater than the sample length influence coefficient.

[0063] The weighting term for the effective sample, namely the dual-weighted coupling model of sample length and distance, is constructed based on the following principle: (1) Distance weighting term: Based on the principle of spatial variability in geostatistics, the weight assigned to a sample decreases as distance increases. Therefore, the distance weighting term is defined as follows: The distance weight term is negative to ensure that the distance weight term follows... Increases and decreases, distance attenuation coefficient Used to control the drastic changes in decay.

[0064] (2) Sample Length Weighting Term: Based on the support effect principle in geostatistics, the weight assigned to the sample length increases with the increase in length. To characterize this nonlinear relationship and ensure the stability of the calculation, the sample length weighting term adopts the natural logarithm function, and is defined as follows: This term is positive in the exponential function, ensuring that the sample length weight term follows the function. Increase and grow, coefficient Used to adjust the degree of influence of sample length.

[0065] (3) Coupling mechanism: In order to balance the contributions of the distance weight term and the sample length weight term and ensure that the distance factor always dominates in all application scenarios, the sample length and distance dual weight coupling model introduces an adaptive coupling coefficient to realize the adaptive coupling of the distance weight term and the sample length weight term.

[0066] (4) Geological rationality constraints: To ensure the geological significance and reasonableness of the estimation of the sample length and distance dual-weighted coupling model, the following constraints are imposed on the parameters: adaptive coupling coefficient The range [0.7, 0.9] defines a clear geological meaning. The lower limit of 0.7 is a strong constraint on the dominance of distance, ensuring that the distance weighting term is the main contributor under all circumstances; the upper limit of 0.9 is a rationality constraint, preventing the model from degenerating into a traditional model that only considers distance due to the adaptive coupling coefficient infinitely approaching 1, thus preserving the corrective significance of sample length. The distance attenuation coefficient is greater than the sample length influence coefficient, further reinforcing the dominance of the distance factor.

[0067] 203. For each of the blocks to be estimated, under the geological rationality constraint, the optimal value of the parameter is determined based on the grade value using the particle swarm optimization algorithm.

[0068] The parameters include the adaptive coupling coefficient, the distance attenuation coefficient, and the sample length influence coefficient, and the optimal parameter values ​​include the optimal values ​​of the adaptive coupling coefficient, the optimal values ​​of the distance attenuation coefficient, and the optimal values ​​of the sample length influence coefficient.

[0069] Under the geological rationality constraint, based on the particle swarm optimization algorithm, the optimal parameter value of the weight term of the effective sample is determined according to the grade value. This includes: setting the particle swarm size and the maximum number of iterations; at the current iteration number, for each particle, under the geological rationality constraint, determining the current position corresponding to the particle as the current parameter to be optimized value of the weight term of the effective sample; obtaining the verification block coordinates and actual grade value of each verification block in the mining area to be estimated; calculating the grade estimate of each verification block based on the verification block coordinates and the current parameter to be optimized value; constructing a fitness function based on the mean square error calculation formula; and calculating the grade estimate based on all the grade estimates and... The current fitness value of the fitness function is calculated based on all actual quality values; the global optimal fitness value is determined based on the current fitness values ​​of all particles in the particle swarm, and the position corresponding to the global optimal fitness value is determined as the global optimal position. If the position corresponding to the global optimal fitness value is the current position, then the current position is the global optimal position; if the current iteration number is the maximum iteration number, or the change between the global optimal fitness value and the historical global optimal fitness value is less than a preset threshold, then the iteration is stopped, and the global optimal position is taken as the optimal value of the parameters. If the current position is the global optimal position, then the current parameter to be optimized value is the optimal value of the parameters.

[0070] The step of calculating the grade estimate of each verification block based on the verification block coordinates and the current parameter to be optimized includes: for each verification block, determining the effective verification sample in the borehole sample based on the verification block coordinates and the sample coordinates, and determining the sample coordinates, grade value, and original sample length of the effective verification sample as the effective verification sample coordinates, effective verification grade value, and effective verification block original sample length; for each verification block, calculating the relative distance of the verification block and the relative original sample length of the verification block, calculating the weight term of each effective verification sample corresponding to the verification block based on the relative distance of the effective verification sample, the relative original sample length of the effective verification sample, and the current parameter to be optimized, and calculating the grade estimate of the verification block based on the weight term of each effective verification sample and the effective verification grade value.

[0071] Specifically: (1) Initialization: Set the particle swarm size to M (particles m = 1, 2...M), and the maximum number of iterations to K (iterations k = 0, 1, 2...K). Under the geological rationality constraints (i.e., the adaptive coupling coefficient is greater than or equal to 0.7 and less than or equal to 0.9, and the distance attenuation coefficient is greater than the sample length influence coefficient), when k = 0, randomly generate the initial velocity of particle m. relative to the initial position , where the initial position = ( ), corresponding to the initial values ​​to be optimized for the parameters, wherein the initial values ​​to be optimized for the parameters include the initial values ​​to be optimized for the adaptive coupling coefficient. The initial value to be optimized for the distance attenuation coefficient. and the initial value to be optimized for the sample length influence coefficient. The initial velocity represents the adjustment range of the adaptive coupling coefficient, distance attenuation coefficient, and sample length influence coefficient in the next step, i.e., k=1. (2) Definition and evaluation of fitness function: Let particle m be at the current position of the current iteration number k. , = ( The corresponding parameter is the current value to be optimized, which includes the current value to be optimized for the adaptive coupling coefficient. The current value to be optimized for the distance attenuation coefficient. And the current value to be optimized for the sample length influence coefficient. .

[0072] The process involves obtaining the coordinates and actual grade values ​​of the verification blocks in the mining area to be estimated. For each verification block, the distance between its coordinates and the coordinates of each borehole sample is calculated. If this distance falls within the preset range of the search ellipsoid parameters, the borehole sample is considered a valid verification sample for that block. The borehole sample has known coordinates, grade value, and original sample length. The valid verification sample has the following coordinates, grade value, and original sample length: the valid verification block's original sample length. Finally, at least one valid verification sample is obtained for each verification block. It should be noted that the difference between the block to be estimated and the verification block is that the grade value of the block to be estimated is unknown, while the actual grade value is known for the verification block.

[0073] For each verification block, calculate the relative distance to each corresponding valid verification sample. The calculation method is the same as in step 202 of the embodiment for calculating the relative distance of valid samples, and will not be repeated here. Calculate the relative original sample length to each corresponding valid verification sample. The calculation method is the same as in step 202 of the embodiment for calculating the relative original sample length of valid samples, and will not be repeated here. At the current position of particle m and the current iteration number k, i.e., the parameter value is... = ( = Under this condition, a weight term is constructed for each valid validation sample.

[0074] The weighting term for the valid verification sample is:

[0075] When the particle is m and the current iteration number is k, the adaptive coupling coefficient is... The value is the current parameter to be optimized, which is the adaptive coupling coefficient. At particle m and current iteration number k, the distance decay coefficient The value is the current parameter to be optimized, which is the distance attenuation coefficient. When particle m and the current iteration number k, the sample length influence coefficient The value to be optimized is the current parameter value of the sample length influence coefficient. , This represents the weight term for valid validation sample j. This represents the relative distance to valid verification sample j. This represents the relative original sample length of the valid verification sample j.

[0076] For each validation block, the grade estimate of that validation block is calculated based on the weight term of the corresponding valid validation sample. Specifically, the weight term of valid validation sample j is... Multiply the effective verification grade value of the effective verification sample j by the effective verification grade value to obtain the second product value corresponding to the effective verification sample j. Add the second product values ​​corresponding to all effective verification samples corresponding to the verification block, and divide by the sum of the weight terms of all effective verification samples to obtain the grade estimate of the verification block. Then, obtain the grade estimate of each verification block.

[0077] Based on the formula for calculating mean squared error, a fitness function is constructed. For particle m, at the current iteration number k, the current fitness value is:

[0078] Where the number of validation blocks t = 1, 2, ..., T is the total number of validation blocks. This represents the current fitness value of particle m at the current iteration number k. This represents the grade estimate of the verification block t corresponding to particle m. This represents the actual grade value of the verification block t.

[0079] (3) The iterative optimization process is described below. In each iteration, the following operations are performed: Evaluation and Update: Since the particle swarm size is M, for the current iteration number k, the particle swarm parameters have M sets of values, i.e. ( , , For each particle (m=1, 2...M), each iteration corresponds to a fitness value. The fitness value of particle m at the current iteration number k is calculated after the iteration is complete. Then, the largest fitness value among all fitness values ​​obtained by particle m before the current iteration number k is taken as the historical best fitness value of particle m, and the current fitness value is compared. The optimal position of particle m is updated by comparing its current fitness value with the historical best fitness value. If the current fitness value is larger, the optimal position of particle m is updated. Current location And the individual optimal fitness value of particle m. Update to the current fitness value. If the individual's optimal fitness value is large, then do not update; that is, update the individual optimal fitness value of particle m. This refers to the historical optimal fitness value of an individual. The largest fitness value among all particles up to the current iteration number k is taken as the historical global optimal fitness value for the entire population. Then, among all particles, find the particle with the largest current fitness value at the current iteration number k. If its current fitness value is greater than the historical global optimal fitness value for the entire population, update the global optimal position gbest to that particle's current position and update the global optimal fitness value. If its current fitness value is less than the historical global best fitness value, then it is not updated; that is, the global best fitness value is used. It is the historical global optimal fitness value.

[0080] Update particle velocity: Update the current velocity of particle m in the d-th dimension (d=1, 2, 3 correspond to the adaptive coupling coefficient, distance attenuation coefficient, and sample length influence coefficient, respectively) according to the following formula:

[0081] Update particle position: Update the current position of particle m in the d-th dimension (d=1, 2, 3 correspond to the adaptive coupling coefficient, distance attenuation coefficient, and sample length influence coefficient, respectively) according to the following formula:

[0082] in: and Let $\frac{m}{k}$ be the current velocity and current position of particle $m$ at the current iteration number $k$. and These are the next velocity and next position of particle m at the next iteration number k+1, respectively; Let d be the d-th dimension component of the optimal position of particle m. d is the d-th dimension component of the global optimal position; w is the inertial weight, used to balance global and local search capabilities; A learning factor for adjusting the step size of the individual's optimal direction. The learning factor is used to adjust the step size for the globally optimal direction; and is a random number between [0, 1] used to introduce randomness into the search.

[0083] Termination and Output: If the current iteration count is the maximum iteration count, or the change between the global optimal fitness value and the historical global optimal fitness value is less than a preset threshold, then the iteration stops. The global optimal position Gbest ( , , The optimal value of the parameter used as the weight term for effective samples is output, i.e., the optimal value of the parameter of the adaptive coupling coefficient is... = The optimal value of the distance attenuation coefficient parameter is = The optimal value of the parameter for the influence coefficient of sample length is If the position corresponding to the globally optimal fitness value is the current position, then the current position is the globally optimal position, and the current parameter value to be optimized is the optimal parameter value. = , = , ).

[0084] 204. Calculate the target grade value of the block to be estimated based on the optimal value of the parameters and the weight term of the effective sample, and calculate the sub-mineral resource reserves of the block to be estimated based on the target grade value.

[0085] 205. Calculate the total mineral resource reserves of the mining area to be estimated based on the reserves of all the sub-mineral resources.

[0086] For steps 204 and 205 of the embodiment, the step of calculating the target grade value of the block to be estimated based on the optimal value of the parameter and the weight term of the effective sample includes: substituting the optimal value of the parameter into the weight term of the effective sample to obtain the optimal weight solution for each effective sample corresponding to the block to be estimated; determining the grade value of the effective sample as the effective grade value; calculating the product of the effective grade value of the same effective sample and the optimal weight solution to obtain a first product value; adding the first product values ​​of all effective samples and dividing by the number of all effective samples to obtain the target grade value of the block to be estimated.

[0087] The optimal weight solution for effective sample i The calculation formula is as follows: =

[0088] Among them, the optimal value of the adaptive coupling coefficient parameter is The optimal value of the distance attenuation coefficient parameter is The optimal value of the parameter for the influence coefficient of sample length is .

[0089] Target grade value of the block to be estimated q The calculation formula is as follows:

[0090] Among them, there are n valid samples corresponding to the block q to be estimated, corresponding to i=1, 2...n. This indicates the target grade value of the block to be valued. This represents the effective grade value of effective sample i.

[0091] The step of calculating the sub-mineral resource reserves of the block to be estimated based on the target grade value includes: obtaining the volume of each block q to be estimated. With density ; Calculate the volume of the same block to be estimated. The density and the target grade value The product of these terms yields the sub-mineral resource reserves of the block to be estimated. The above is based on the reserves of all the sub-mineral resources. Calculating the total mineral resource reserves of the mining area to be estimated includes: calculating the reserves of all the sub-mineral resources. The total mineral resource reserves of the mining area to be estimated are obtained by adding them together. .

[0092] Sub-mineral resource reserves The calculation formula is as follows:

[0093] in, This represents the volume of the block q to be estimated. This represents the density of the block q to be estimated. This represents the target grade value of the block q to be estimated. This indicates the sub-mineral resource reserves of the block to be estimated.

[0094] Total mineral resource reserves of the mining area to be estimated The calculation formula is as follows:

[0095] Among them, the number of blocks to be estimated in the mining area to be estimated is: indivual, This indicates the total mineral resource reserves of the mining area to be estimated.

[0096] Furthermore, the generalization ability of the sample length and distance dual-weighted coupling model in embodiments 201-205 of this invention is evaluated through comprehensive verification. Then, based on the estimated total mineral resource reserves of the mining area to be estimated and the verification data, a technical report conforming to the specifications is automatically generated.

[0097] Specifically, the comprehensive verification is as follows: The verification process primarily employed real-world mine data, supplemented by simulated data. The verification results demonstrate that this invention outperforms traditional methods in terms of estimation accuracy and adaptability to complex geological conditions.

[0098] First, verification with real data: (1) Validation data The data was verified using real data from a volcanogenic massive sulfide deposit. The dataset contained 8,638 borehole samples and had the following characteristics: the sample length distribution was extremely uneven, ranging from 0.02 to 3.46 meters, and the coefficient of variation (CV) of the sample length was greater than 0.3; the ore body morphology was complex, with multiple thin-layer structures; the grade distribution varied drastically, and the boundary identification was difficult.

[0099] (2) Comparison of schemes The method of this invention is based on an improved inverse distance weighting with dual-weight coupling algorithm (EIDW-DC). Figure 5 The EIDW-DC method, traditional IDW (with a power exponent typically of 2), and ordinary Kriging are compared. Traditional IDW, specifically the inverse distance weighting (IDW) method, is primarily evaluated based on estimation accuracy. The estimation results are shown below. Figure 5 As shown.

[0100] (3) Verification results like Figure 6 As shown, within each grade range, the method of this invention outperforms the traditional inverse distance power ratio method, with an overall accuracy improvement of 1.42%. In particular, the accuracy improvements reach 9.09% and 3.54% in the 0-1 g / t and 3-4 g / t grade ranges, respectively.

[0101] Second, simulated data verification (auxiliary verification). (1) Verification purpose: To assist in verifying the effectiveness of the method of the present invention in a controlled environment.

[0102] (2) Data generation: generate simulated data for 1,500 borehole samples. The sample length adopts the Weber distribution (CV>0.3), ranging from 0.2 to 8.0 meters, and the grade value ranges from 0.1 to 8.0.

[0103] (3) Verification results: Simulation results show that the method of the present invention has a lower mean square error (mean square error is...) than the traditional inverse power law method. Figure 7 MSE and mean absolute error (mean absolute error is...) Figure 7 The computational efficiency was reduced by 2.91% and 1.48% respectively compared to the standard kriging model, demonstrating improved computational efficiency and further validating the effectiveness of the dual-weight coupling model and the parameter adaptive optimization mechanism. Figure 7 In this context, RMSE stands for Root Mean Square Error.

[0104] Report generation: Based on the above reserve estimation results and system verification data, a technical report that conforms to the mineral resource reserve estimation specifications can be automatically generated.

[0105] Experimental verification shows that the present invention has achieved significant improvements in estimation accuracy, model adaptability, and adaptability to complex geological conditions.

[0106] This invention provides an adaptive variable-length sample length mineral resource reserve estimation method, apparatus, and equipment. Firstly, by preserving the original sample length of the borehole samples, the traditional equal-length preprocessing is eliminated, avoiding information distortion caused by forced equal-length preprocessing and improving estimation accuracy. Secondly, an innovative dual-weighted coupled model of sample length and distance is constructed, representing the weight term of the effective sample. Under geological constraints, this model ensures the dominance of spatial distance while incorporating the volume effect represented by the original sample length, thus accurately characterizing the contribution of samples at different scales in spatial interpolation. Finally, a particle swarm optimization algorithm is introduced for iterative optimization, achieving adaptive optimization of model parameters. This avoids information distortion caused by a single power exponent and overcomes the shortcomings of traditional methods that rely on fixed parameters and lack intelligent tuning capabilities. This invention significantly improves the accuracy and adaptability of mineral resource reserve estimation.

[0107] Furthermore, as Figure 1 and Figure 2 The specific implementation of the method shown in this invention provides an adaptive variable-length mineral resource reserve estimation device, such as... Figure 3 As shown, the device includes: a data acquisition and neighborhood search module 31, a dimensionless conversion and construction module 32, a parameter adaptive optimization module 33, and a storage calculation module 34; The data acquisition and neighborhood search module 31 is used to acquire the coordinates of each block to be estimated in the mining area to be estimated, acquire the sample coordinates, grade value and original sample length of each borehole sample in the mining area to be estimated, determine the effective sample corresponding to each block to be estimated in all the borehole samples based on the coordinates of the block to be estimated and the sample coordinates, and determine the sample coordinates of the effective sample as the effective sample coordinates. The dimensionless modeling and construction module 32 is used to calculate the relative distance between each effective sample and the coordinates of the block to be estimated, based on the block coordinates and the effective sample coordinates; calculate the relative original length of each effective sample based on its original length; construct a weight term for each effective sample based on the adaptive coupling coefficient, distance attenuation coefficient, relative distance of the effective samples, sample length influence coefficient, and relative original length of the effective samples; and configure geological rationality constraints on the parameters of the weight term of the effective samples. The weight term of the effective samples is as follows:

[0108] In the formula, This represents the adaptive coupling coefficient. This represents the distance attenuation coefficient. This represents the sample length influence coefficient. This represents the weight term for valid sample i. This represents the relative distance to the valid sample i. Indicates the relative original sample length of valid sample i; The parameter adaptive optimization module 33 is used to determine the optimal value of the parameters for each block to be estimated under the geological rationality constraint, based on the particle swarm optimization algorithm and the grade value, calculate the target grade value of the block to be estimated based on the optimal value of the parameters and the weight term of the effective sample, and calculate the sub-mineral resource reserves of the block to be estimated based on the target grade value. The parameters include the adaptive coupling coefficient, the distance attenuation coefficient and the sample length influence coefficient, and the optimal value of the parameters includes the optimal value of the adaptive coupling coefficient, the optimal value of the distance attenuation coefficient and the optimal value of the sample length influence coefficient. The reserve calculation module 34 is used to calculate the total mineral resource reserves of the mining area to be estimated based on the reserves of all the sub-mineral resources.

[0109] Accordingly, in order to calculate the relative distance of each of the effective samples based on the coordinates of the block to be estimated and the coordinates of the effective samples, and to calculate the relative original sample length of each of the effective samples based on the original sample length of each of the effective samples, the dimensionless transformation and construction module 32 includes: a first dimensionless transformation unit 321 and a second dimensionless transformation unit 322. The first dimensionless unit 321 is specifically used to calculate the distance between the coordinates of the block to be estimated and the coordinates of each corresponding valid sample, add all the distances together and divide by the number of all valid samples to obtain the average distance, and divide each distance by the average distance to obtain the relative distance of each corresponding valid sample. The second dimensionless unit 322 is specifically used to add the original sample lengths of all the valid samples together and divide by the number of all the valid samples to obtain the average original sample length, and to divide the original sample length of each valid sample by the average original sample length to obtain the relative original sample length of each valid sample.

[0110] Accordingly, in order to configure geological rationality constraints on the parameters of the weighting terms of the effective samples, the dimensionless conversion and construction module 32 is specifically used to configure geological rationality constraints on the adaptive coupling coefficient, wherein the adaptive coupling coefficient is greater than or equal to 0.7 and less than or equal to 0.9; and to configure geological rationality constraints on both the distance attenuation coefficient and the sample length influence coefficient, wherein the distance attenuation coefficient is greater than the sample length influence coefficient.

[0111] Accordingly, in order to determine the optimal parameter value based on the grade value using the particle swarm optimization algorithm under the geological rationality constraint, the parameter adaptive optimization module 33 is specifically used to set the particle swarm size and the maximum number of iterations; at the current iteration number, for each particle, under the geological rationality constraint, the current position corresponding to the particle is determined as the current parameter to be optimized value of the parameter of the weight term of the effective sample; the coordinates of the verification block and the actual grade value of each verification block in the mining area to be estimated are obtained; the grade estimate value of each verification block is calculated based on the verification block coordinates and the current parameter to be optimized value; a fitness function is constructed based on the formula for calculating the mean square error; and the grade is estimated based on all the grades. The current fitness value of the fitness function is calculated based on the estimated value and all the actual quality values. The global optimal fitness value is determined based on the current fitness values ​​of all the particles in the particle swarm. The position corresponding to the global optimal fitness value is determined as the global optimal position. If the position corresponding to the global optimal fitness value is the current position, then the current position is the global optimal position. If the current iteration number is the maximum iteration number, or if the change between the global optimal fitness value and the historical global optimal fitness value is less than a preset threshold, then the iteration stops, and the global optimal position is used as the optimal parameter value. If the current position is the global optimal position, then the current parameter to be optimized value is the optimal parameter value.

[0112] Accordingly, in order to calculate the grade estimate of each verification block based on the verification block coordinates and the current parameter to be optimized value, the parameter adaptive optimization module 33 is specifically used to determine the effective verification sample in the borehole sample based on the verification block coordinates and the sample coordinates for each verification block, and to determine the sample coordinates, grade value and original sample length of the effective verification sample as the effective verification sample coordinates, effective verification grade value and effective verification block original sample length respectively; for each verification block, to calculate the relative distance of the verification block and the relative original sample length of the verification block, to calculate the weight term of each effective verification sample corresponding to the verification block based on the relative distance of the effective verification sample, the relative original sample length of the effective verification sample and the current parameter to be optimized value, and to calculate the grade estimate of the verification block based on the weight term of each effective verification sample and the effective verification grade value.

[0113] Accordingly, in order to calculate the target grade value of the block to be estimated based on the optimal value of the parameters and the weight term of the effective samples, the parameter adaptive optimization module 33 is specifically used to substitute the optimal value of the parameters into the weight term of the effective samples to obtain the optimal weight solution for each effective sample corresponding to the block to be estimated; determine the grade value of the effective sample as the effective grade value; calculate the product of the effective grade value of the same effective sample and the optimal weight solution to obtain a first product value; add the first product values ​​of all effective samples and divide by the number of all effective samples to obtain the target grade value of the block to be estimated.

[0114] Accordingly, in order to calculate the sub-mineral resource reserves of the block to be estimated based on the target grade value, the parameter adaptive optimization module 33 is specifically used to obtain the volume and density of each block to be estimated; calculate the product of the volume, density and the target grade value of the same block to be estimated to obtain the sub-mineral resource reserves of the block to be estimated; in order to calculate the total mineral resource reserves of the mining area to be estimated based on all the sub-mineral resource reserves, the reserve calculation module 34 is specifically used to add up all the sub-mineral resource reserves to obtain the total mineral resource reserves of the mining area to be estimated.

[0115] It should be noted that other corresponding descriptions of the functional units involved in the adaptive variable sample length mineral resource reserve estimation device provided in this embodiment can be found in [reference needed]. Figures 1 to 2 The corresponding description will not be repeated here.

[0116] Based on the above, Figures 1 to 2 Accordingly, this embodiment also provides a storage medium, which may be volatile or non-volatile, storing a computer program that, when executed by a processor, implements the above-described method. Figures 1 to 2 The method shown is an adaptive variable-length mineral resource reserve estimation method.

[0117] Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, portable hard drive, etc.) and includes several methods for enabling a computer device (such as a personal computer, server, or network device, etc.) to execute various implementation scenarios of the present invention.

[0118] Based on the above, Figures 1 to 2 The method shown and Figure 3 , Figure 4To achieve the above objectives, the present invention also provides a computer device, specifically a personal computer, server, network device, etc., as shown in the illustrated embodiment. This computer device includes a storage medium and a processor; the storage medium stores a computer program; the processor executes the computer program to achieve the above-described objectives. Figure 1 and Figure 2 The method shown is an adaptive variable-length mineral resource reserve estimation method.

[0119] Optionally, the computer device may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.

[0120] Those skilled in the art will understand that the computer device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0121] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned computer device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the non-volatile storage medium, as well as communication with other hardware and software in the information processing entity device.

[0122] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary general-purpose hardware platform, or it can be implemented by hardware.

[0123] This invention provides an adaptive variable-length sample length mineral resource reserve estimation method, apparatus, and equipment. Firstly, by preserving the original sample length of the borehole samples, the traditional equal-length preprocessing is eliminated, avoiding information distortion caused by forced equal-length preprocessing and improving estimation accuracy. Secondly, an innovative dual-weighted coupled model of sample length and distance is constructed, representing the weight term of the effective sample. Under geological constraints, this model ensures the dominance of spatial distance while incorporating the volume effect represented by the original sample length, thus accurately characterizing the contribution of samples at different scales in spatial interpolation. Finally, a particle swarm optimization algorithm is introduced for iterative optimization, achieving adaptive optimization of model parameters. This avoids information distortion caused by a single power exponent and overcomes the shortcomings of traditional methods that rely on fixed parameters and lack intelligent tuning capabilities. This invention significantly improves the accuracy and adaptability of mineral resource reserve estimation.

[0124] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention. Those skilled in the art will understand that the modules in the apparatus of the embodiment can be distributed within the apparatus of the embodiment as described, or can be located in one or more apparatuses different from this embodiment, with corresponding changes. The modules of the above-described embodiment can be combined into one module, or further divided into multiple sub-modules. The above-described serial numbers are for descriptive purposes only and do not represent the superiority or inferiority of the embodiment. The above disclosures are only a few specific embodiments of the present invention; however, the present invention is not limited thereto, and any variations conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. An adaptive method for estimating mineral resource reserves with variable sample length, characterized in that, The method includes: The coordinates of each block to be estimated in the mining area to be estimated are obtained. The sample coordinates, grade value, and original sample length of each borehole sample in the mining area to be estimated are obtained. Based on the coordinates of the block to be estimated and the sample coordinates, the effective sample corresponding to each block to be estimated is determined from all the borehole samples. The sample coordinates of the effective sample are determined as the effective sample coordinates. For each block to be estimated, the distance between the coordinates of the block to be estimated and the sample coordinates of each borehole sample is calculated. The borehole samples corresponding to the distances within the preset value range of the search ellipsoid parameters are selected as the effective samples corresponding to the block to be estimated. For each block to be estimated, the relative distance between the block coordinates and the effective sample coordinates is calculated. The relative original sample length is calculated based on the original sample length of each effective sample. A weight term for each effective sample is constructed based on the adaptive coupling coefficient, distance attenuation coefficient, relative distance of the effective samples, sample length influence coefficient, and relative original sample length. Geological rationality constraints are configured for the parameters of the weight term of the effective samples. The weight term of the effective samples is as follows: In the formula, This represents the adaptive coupling coefficient. This represents the distance attenuation coefficient. This represents the sample length influence coefficient. This represents the weight term for valid sample i. This represents the relative distance to the valid sample i. Indicates the relative original sample length of valid sample i; The geological rationality constraints for the parameter configuration of the weighting term of the valid samples include: Configure the adaptive coupling coefficient with a geological rationality constraint that is greater than or equal to 0.7 and less than or equal to 0.9; Both the distance attenuation coefficient and the sample length influence coefficient are subject to a geologically reasonable constraint that the distance attenuation coefficient is greater than the sample length influence coefficient. For each block to be estimated, under the geological rationality constraint, based on the particle swarm optimization algorithm, the optimal value of the parameter is determined according to the grade value. The target grade value of the block to be estimated is calculated according to the optimal value of the parameter and the weight term of the effective sample. The sub-mineral resource reserves of the block to be estimated are calculated according to the target grade value. The parameter includes the adaptive coupling coefficient, the distance attenuation coefficient and the sample length influence coefficient. The optimal value of the parameter includes the optimal value of the adaptive coupling coefficient, the optimal value of the distance attenuation coefficient and the optimal value of the sample length influence coefficient. The total mineral resource reserves of the mining area to be estimated are calculated based on the reserves of all the sub-mineral resources.

2. The method according to claim 1, characterized in that, The step of calculating the relative distance of each valid sample based on the coordinates of the block to be estimated and the coordinates of the valid samples, and calculating the relative original sample length of each valid sample based on the original sample length of each valid sample, includes: Calculate the distance between the coordinates of the block to be estimated and the coordinates of each corresponding valid sample. Add all the distances together and divide by the number of all valid samples to obtain the average distance. Divide each distance by the average distance to obtain the relative distance of each corresponding valid sample. The average original sample length is obtained by summing the original sample lengths of all the valid samples and dividing by the total number of valid samples. The relative original sample length of each valid sample is obtained by dividing the original sample length of each valid sample by the average original sample length.

3. The method according to claim 1, characterized in that, Under the geological rationality constraints, determining the optimal value of the parameter based on the grade value using the particle swarm optimization algorithm includes: Set the particle swarm size and the maximum number of iterations; In the current iteration, for each particle, under the geological rationality constraint, the current position corresponding to the particle is determined as the current parameter to be optimized value of the parameter of the weight term of the effective sample. The verification block coordinates and actual grade values ​​of each verification block in the mining area to be estimated are obtained. The grade estimate value of each verification block is calculated based on the verification block coordinates and the current parameter to be optimized value. Based on the formula for calculating the mean square error, a fitness function is constructed. The current fitness value of the fitness function is calculated based on all the grade estimates and all the actual grade values. The global optimal fitness value is determined based on the current fitness values ​​of all the particles in the particle swarm, and the position corresponding to the global optimal fitness value is determined as the global optimal position. If the position corresponding to the global optimal fitness value is the current position, then the current position is the global optimal position. If the current iteration number is the maximum iteration number, or the change between the global optimal fitness value and the historical global optimal fitness value is less than a preset threshold, then the iteration stops, and the global optimal position is taken as the optimal value of the parameter. If the current position is the global optimal position, then the current parameter to be optimized value is the optimal value of the parameter.

4. The method according to claim 3, characterized in that, The step of calculating the grade estimate for each verification block based on the verification block coordinates and the current parameter to be optimized includes: For each of the verification blocks, the effective verification sample in the borehole sample is determined according to the coordinates of the verification block and the coordinates of the sample. The sample coordinates, grade value and original sample length of the effective verification sample are respectively determined as the effective verification sample coordinates, effective verification grade value and effective verification block original sample length. For each of the verification blocks, the relative distance and the relative original sample length of the verification block are calculated. Based on the relative distance of the effective verification samples, the relative original sample length of the effective verification samples, and the current parameter to be optimized, the weight term of each effective verification sample corresponding to the verification block is calculated. Based on the weight term of each effective verification sample and the effective verification grade value, the grade estimate of the verification block is calculated.

5. The method according to claim 1, characterized in that, The step of calculating the target grade value of the block to be estimated based on the optimal value of the parameters and the weight term of the effective samples includes: Substituting the optimal value of the parameter into the weight term of the effective sample, the optimal weight solution for each effective sample corresponding to the block to be estimated is obtained; The grade value of the effective sample is determined as the effective grade value. The product of the effective grade value of the same effective sample and the weighted optimal solution is calculated to obtain the first product value. The first product values ​​of all effective samples are added together and divided by the number of all effective samples to obtain the target grade value of the block to be estimated.

6. The method according to claim 1, characterized in that, The calculation of the sub-mineral resource reserves of the block to be estimated based on the target grade value includes: Obtain the volume and density of each of the blocks to be estimated; The sub-mineral resource reserves of the same block to be estimated are obtained by multiplying its volume, density and target grade value. The step of calculating the total mineral resource reserves of the mining area to be estimated based on the reserves of all the sub-mineral resources includes: The total mineral resource reserves of the mining area to be estimated are obtained by summing up the reserves of all the sub-mineral resources.

7. An adaptive variable-length mineral resource reserve estimation device, characterized in that, The device includes: The data acquisition and neighborhood search module is used to acquire the coordinates of each block to be estimated in the mining area to be estimated, acquire the sample coordinates, grade value, and original sample length of each borehole sample in the mining area to be estimated, determine the effective sample corresponding to each block to be estimated from all the borehole samples based on the coordinates of the block to be estimated and the sample coordinates of the effective sample, wherein, for each block to be estimated, the distance between the coordinates of the block to be estimated and the sample coordinates of each borehole sample is calculated, and the borehole samples corresponding to the distance within the preset value range of the search ellipsoid parameters are selected as the effective samples corresponding to the block to be estimated; The dimensionless modeling and construction module is used to calculate the relative distance between each effective sample and the coordinates of the block to be estimated, based on the block coordinates and the effective sample coordinates; calculate the relative original sample length based on the original sample length of each effective sample; construct a weight term for each effective sample based on the adaptive coupling coefficient, distance attenuation coefficient, relative distance of the effective samples, sample length influence coefficient, and relative original sample length of the effective samples; and configure geological rationality constraints on the parameters of the weight term of the effective samples. The weight term of the effective samples is as follows: In the formula, This represents the adaptive coupling coefficient. This represents the distance attenuation coefficient. This represents the sample length influence coefficient. This represents the weight term for valid sample i. This represents the relative distance to the valid sample i. Indicates the relative original sample length of valid sample i; The dimensionless conversion and construction module is used to configure the adaptive coupling coefficient with a geological rationality constraint that the adaptive coupling coefficient is greater than or equal to 0.7 and less than or equal to 0.9, and to configure the distance attenuation coefficient and the sample length influence coefficient with a geological rationality constraint that the distance attenuation coefficient is greater than the sample length influence coefficient. The parameter adaptive optimization module is used to determine the optimal value of the parameters for each block to be estimated under the geological rationality constraint, based on the particle swarm optimization algorithm and the grade value, calculate the target grade value of the block to be estimated based on the optimal value of the parameters and the weight term of the effective sample, and calculate the sub-mineral resource reserves of the block to be estimated based on the target grade value. The parameters include the adaptive coupling coefficient, the distance attenuation coefficient, and the sample length influence coefficient, and the optimal value of the parameters includes the optimal value of the adaptive coupling coefficient, the optimal value of the distance attenuation coefficient, and the optimal value of the sample length influence coefficient. The reserve calculation module is used to calculate the total mineral resource reserves of the mining area to be estimated based on the reserves of all the sub-mineral resources.

8. A storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the adaptive variable-length mineral resource reserve estimation method according to any one of claims 1 to 6.

9. A computer device comprising a memory, a processor, and a computer program stored on a storage medium and executable on the processor, characterized in that, When the processor executes the program, it implements the adaptive variable-length mineral resource reserve estimation method according to any one of claims 1 to 6.

Images