A method for graded mining of marble powder

Abstract

This invention discloses a method for graded mining of marble powder, comprising: a drilling operation stage, collecting drilled rock powder samples and obtaining chemical composition data; using the chemical composition data as sample values ​​of ore grade at corresponding locations to construct a set of ore grade sample points; based on the above set, constructing a variation function model and executing a Kriging space interpolation algorithm to establish a grade prediction mathematical model; outputting ore grade prediction values ​​according to the grade prediction mathematical model and generating an ore grade zoning map; dividing the area to be mined into multiple blasting operation blocks according to the ore grade zoning map, and determining a set of blasting process parameters matching the ore grade for each blasting operation block; and performing drilling, charging, blasting, and loading operations according to each blasting process parameter set. This invention solves the problems of low operational efficiency, frequent blockages, and increased wear in existing technologies, which restrict the production efficiency and economic benefits of mines.

Description

Technical Field

[0001] This invention relates to the field of mining technology, specifically to a method for graded mining of marble powder. Background Technology

[0002] Marble, as an important building decoration material and industrial powder raw material, plays a vital role in economic development. However, current marble mining in China generally suffers from a focus on extraction rather than evaluation. In the production stage, ore quality control largely relies on traditional methods of visual inspection and experience-based judgment for ore grading. This qualitative evaluation method is heavily influenced by human factors and lacks unified, objective quantitative standards, resulting in the mixing of high-grade ore with low-grade waste rock. This not only reduces the quality stability of the final product but also causes a serious waste of valuable resources.

[0003] Furthermore, under complex geological conditions, the ore body structure is intricate, the ore and rock occurrence conditions are varied, and the spatial distribution of grades differs significantly. Currently, mines lack effective means of quality prediction before blasting, and the ore grade is often only known after blasting. This leads to low operating efficiency of loading equipment when dealing with mixed ore and rock, frequent jamming, and accelerated wear, which seriously restricts the mine's production efficiency and economic benefits. Summary of the Invention

[0004] To address the aforementioned shortcomings of existing technologies, this invention provides a method for graded mining of marble powder ore that enables precise pre-mining prediction, differentiated blasting, and graded loading, thereby improving operational efficiency and economic benefits.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] A method for graded mining of marble powder includes the following steps:

[0007] During the drilling operation, rock powder samples are collected along the area to be mined according to a preset grid. Multi-dimensional physicochemical property analysis is performed on the rock powder samples from each borehole to obtain chemical composition data, which includes carbonate content and impurity content.

[0008] The chemical composition data of rock powder samples from each borehole are used as sample values ​​of ore grade at the corresponding borehole locations to construct a set of ore grade sample points covering the area to be mined.

[0009] Based on the set of ore grade sample points, a variation function model of ore grade is constructed using regionalized variable theory, and the kriging space interpolation algorithm is executed using the variation function model to establish a grade prediction mathematical model that characterizes the three-dimensional spatial distribution law of ore quality in the mining area.

[0010] The grade prediction mathematical model outputs the ore grade prediction values ​​at each spatial point in the mining area divided by a preset grid. The ore grade prediction values ​​are then compared with preset grading thresholds to generate an ore grade zoning map.

[0011] Based on the ore grade zoning map, the area to be mined is divided into multiple blasting operation blocks with different ore grade levels, and a set of blasting process parameters matching the ore grade level is determined for each blasting operation block. The set of blasting process parameters includes hole mesh parameters, charge parameters and detonation timing parameters.

[0012] According to the blasting process parameter set corresponding to each blasting operation block, drilling, charging, blasting and loading operations are carried out respectively, and the blasted ore of different grades is loaded and transported separately in separate sections.

[0013] Preferably, the method for constructing a variation function model of ore grade based on the aforementioned set of ore grade sample points using regionalized variable theory is as follows:

[0014] The normality test and transformation processing are performed on the sample values ​​of ore grade of each sample point in the ore grade sample point set to obtain the processed sample values.

[0015] Based on the processed sample values, the experimental variogram values ​​are calculated at different step sizes and in different spatial directions;

[0016] Select a theoretical variogram model and theoretically fit the experimental variogram values ​​to obtain the nugget constant, sill value, and range parameters.

[0017] The variogram model is constructed based on the selected theoretical variogram model and the obtained nugget constant, sill value, and range parameters.

[0018] The different spatial directions include at least the direction along the ore strike, the dip direction, and the thickness direction.

[0019] Preferably, the method for establishing a grade prediction mathematical model that characterizes the three-dimensional spatial distribution law of ore quality within the mining area by executing the Kriging space interpolation algorithm using the variogram model is as follows:

[0020] The Kriging spatial interpolation algorithm is a co-Kriging method. It obtains surface geological mapping data or geophysical exploration data of the area to be mined as covariates, uses the processed sample values ​​as principal variables, calculates the cross-variance function based on the principal variables and the covariates, and uses the variance function model, the cross-variance function, the principal variables and the covariates to execute the Kriging spatial interpolation algorithm to establish the grade prediction mathematical model.

[0021] Preferably, the method for generating an ore grade zoning map by outputting the predicted ore grade values ​​at each spatial point within the mining area divided by a preset grid according to the grade prediction mathematical model, and comparing each predicted ore grade value with a preset grading threshold is as follows:

[0022] The predicted ore grade values ​​at each spatial point within the mining area, divided by a preset grid and output by the grade prediction mathematical model, are compared with multiple preset grading thresholds. Based on the comparison results, the spatial points corresponding to each predicted ore grade value are divided into multiple grade areas.

[0023] The spatial boundary lines of each grade area are extracted using the contour tracing algorithm. The spatial boundary lines of the same grade area are enclosed to form the area polygons of that grade area. Different area polygons are filled with different colors or patterns. Then, the filled area polygons are superimposed on the topographic and geological map of the area to be mined to generate an ore grade zoning map.

[0024] Preferably, the hole mesh parameters include borehole spacing and row spacing; the charge parameters include single-hole charge amount and charge structure type; and the detonation timing parameters include detonation method, detonation time difference, and detonation sequence.

[0025] Preferably, based on the ore grade zoning map, the area to be mined is divided into multiple blasting operation blocks with different ore grade levels, and a set of blasting process parameters matching the ore grade level is determined for each blasting operation block. The method for determining the blasting process parameter set, including hole mesh parameters, charge parameters, and detonation timing parameters, is as follows:

[0026] According to the ore grade zoning map, the area to be mined is divided into a first blasting operation block, a second blasting operation block, and a third blasting operation block. The predicted ore grade of the first blasting operation block is greater than that of the second blasting operation block, and the predicted ore grade of the third blasting operation block is between the predicted ore grade of the first blasting operation block and the predicted ore grade of the second blasting operation block.

[0027] Specifically, for the first blasting operation block, the blasting process parameters adopted are the first borehole spacing, the first row spacing, the first single-hole charge amount, the air-gap charge structure, and the sequential detonation sequence; for the second blasting operation block, the blasting process parameters adopted are the second borehole spacing, the second row spacing, the second single-hole charge amount, the continuous charge structure, and the inter-row micro-delay detonation sequence; wherein, the first borehole spacing is smaller than the second borehole spacing, the first row spacing is smaller than the second row spacing, and the first single-hole charge amount is smaller than the second single-hole charge amount;

[0028] The borehole spacing in the third blasting operation block is greater than the borehole spacing in the first block and less than the borehole spacing in the second block. The row spacing is greater than the row spacing in the first block and less than the row spacing in the second block. The charge amount per borehole is greater than the charge amount per borehole in the first block and less than the charge amount per borehole in the second block. The charge structure adopts partial air gap charging. The borehole delay time in the initiation sequence is greater than the borehole delay time in the sequential borehole initiation sequence and less than the borehole delay time in the row-to-row micro-delay initiation sequence.

[0029] Preferably, the air-gap charge structure used in the first blasting operation block is as follows: the total charge is divided into an upper charge section and a lower charge section in the borehole, and there is an inert material section between the upper charge section and the lower charge section. The charge density of the lower charge section is greater than that of the upper charge section, and a detonator is installed in the lower charge section.

[0030] The continuous charge structure used in the second blasting operation block is as follows: explosives are continuously loaded along the entire length of the borehole, and the diameter of the explosives is equal to the diameter of the borehole to form a coupled charge state. A detonator is installed at the bottom third of the borehole and a reverse detonation method is adopted.

[0031] Preferably, outlier detection is performed on the chemical composition data of each borehole rock powder sample, and borehole rock powder samples whose chemical composition data deviates from the average value by more than a preset multiple of the standard deviation are removed; the chemical composition data of the removed borehole rock powder samples are filled by interpolation calculation based on the chemical composition data of adjacent borehole rock powder samples; and then the chemical composition data of each borehole rock powder sample are normalized.

[0032] Preferably, the method for shoveling and transporting blasted ore of different grades in separate zones is as follows: after the blasting operation is completed, temporary dividing markers are set up on the surface of the blasted ore pile according to the boundary lines of the ore grade zoning map, and the blasted ore of the first blasting operation block, the third blasting operation block and the second blasting operation block are shoveled in sequence according to the temporary dividing markers; and the ore of different grades is unloaded into different transport vehicles or different transport batches.

[0033] Preferably, the marble powder ore grading mining method further includes a dynamic feedback correction step: after blasting, the distribution of the actual ore grade after the blasting is determined and compared with the ore grade prediction value output by the grade prediction mathematical model, and the deviation value between the actual ore grade value and the ore grade prediction value is calculated; when the deviation value exceeds a preset threshold, the actual ore grade value is added as a new sample point to the ore grade sample point set, and the variation function model is updated based on the ore grade sample point set with the added actual ore grade value.

[0034] Compared with the prior art, the present invention has the following advantages:

[0035] 1. This invention systematically collects borehole rock powder samples during the drilling stage and performs multi-dimensional physicochemical property analysis. The chemical composition data of each borehole rock powder sample is directly mapped to the sample value of ore grade at the corresponding borehole location. Based on this, a variogram model is constructed using regionalized variable theory, and a three-dimensional grade prediction mathematical model is established using the Kriging space interpolation algorithm. This grade prediction mathematical model can quantitatively output the predicted ore grade values ​​at each spatial point within the mining area, providing an objective and quantifiable decision-making basis for subsequent graded mining. This completely changes the traditional subjective mode that relies on visual observation and experience-based judgment, enabling ore quality evaluation to shift from qualitative to quantitative methods and eliminating reliance on manual experience.

[0036] 2. This invention generates an ore grade zoning map based on a grade prediction mathematical model, dividing the area to be mined into multiple blasting operation blocks with different ore grade levels, such as high, medium, and low grade blasting operation blocks. Differentiated blasting process parameters are then matched to each grade block. Specifically, for high-grade blocks, smaller borehole spacing and row spacing, lower single-hole charge, air-spaced charge structure, and sequential detonation timing are used to significantly reduce excessive crushing of high-grade ore by the blast stress wave. For low-grade blocks, the opposite parameters are used to enhance crushing. This zone-specific blasting mode achieves differentiated and precise blasting, maximizing the protection of high-value ore while ensuring crushing effect, reducing dilution rate, and effectively protecting high-grade ore.

[0037] 3. This invention effectively avoids the mixing of high-quality ore and waste rock through precise zoning before blasting and zoned loading and transportation after blasting. The loading equipment can sequentially load high, medium, and low-grade areas according to temporary dividing markers, with different grades of ore unloaded into different transport vehicles or batches. This refined operation mode greatly reduces blockages and ineffective loading during the loading process, improves single-bucket cycle efficiency, and reduces equipment wear and energy consumption in subsequent crushing and processing.

[0038] 4. After each blast, this invention measures the distribution of the actual ore grade and compares it with the predicted ore grade output by the grade prediction mathematical model, calculating the deviation value. When the deviation value exceeds a preset threshold, the actual ore grade value is added as a new sample point to the original ore grade sample point set, and the variation function model is refitted. This closed-loop feedback mechanism enables the grade prediction mathematical model to continuously self-correct and optimize with the accumulation of production data, gradually improving prediction accuracy, adapting to spatial changes in ore body grade, and achieving long-term stable operation.

[0039] 5. This invention integrates geostatistical modeling, three-dimensional visualization zoning, refined blasting control, and shovel loading scheduling into a unified technical framework, forming a replicable and scalable graded mining technology system for marble powder ore. By improving resource utilization, reducing ineffective blasting and energy consumption, and lowering the waste rock contamination rate, it provides a technical paradigm for marble and non-metallic mines under similar complex geological conditions. Attached Figure Description

[0040] Appendix Figure 1 This is a flowchart of the marble powder ore classification mining method of the present invention;

[0041] Appendix Figure 2 This is a flowchart illustrating the construction of a variation function model for ore grade in the marble powder ore classification and mining method of the present invention.

[0042] Appendix Figure 3 This is a flowchart of the process for generating an ore grade zoning map in the marble powder ore classification mining method of the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0044] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0045] Furthermore, in describing representative embodiments, the specification may have presented methods and / or processes as a specific sequence of steps. However, the method or process should not be limited to the specific order of steps described herein, to the extent that it does not depend on such a specific order. As will be understood by those skilled in the art, other sequences of steps are also possible. Therefore, the specific order of steps set forth in the specification should not be construed as a limitation on the embodiments. Moreover, the method and / or process should not be limited to the steps performed in the written order; those skilled in the art will readily understand that these orders can be varied and still remain within the spirit and scope of the embodiments of this application.

[0046] This specific embodiment provides a method for graded mining of marble powder, as shown in the attached figure. Figure 1 As shown, it includes the following steps:

[0047] Step S1) During the drilling operation, borehole locations are arranged along the area to be mined according to a preset grid (e.g., the grid spacing is set to 5m × 5m, which can be adjusted according to the degree of variation in the ore body). A down-the-hole drill is used for drilling, and a rock powder collection device is installed at the borehole opening to continuously collect the rock powder discharged throughout the entire borehole depth as a borehole rock powder sample for that location. To obtain information on vertical grade variations, each borehole can also be divided into several segments according to depth (e.g., every 2m is a segment). Borehole rock powder samples are collected independently in each segment, and the corresponding borehole number and depth range information are labeled for each segment's borehole rock powder sample.

[0048] The collected borehole rock powder samples were then sent to the laboratory for multi-dimensional physicochemical property analysis to obtain chemical composition data, including the content of carbonate components and impurities. Specifically, the mass percentage of calcium carbonate (CaCO3) in the borehole rock powder samples was determined by chemical titration analysis; the mass percentage of magnesium carbonate (MgCO3), silicon dioxide (SiO2), aluminum oxide (Al2O3), and iron oxide (Fe2O3) was determined by X-ray fluorescence spectrometry (XRF); and the uniaxial compressive strength of the corresponding rock strata was determined using a mechanical testing device.

[0049] Therefore, by collecting borehole rock powder samples and conducting multi-dimensional physicochemical property analysis during the drilling stage, the ore quality evaluation point is shifted from the traditional post-blasting stage to the drilling stage, creating a time window for scientific decision-making before mining. The combined use of chemical titration and X-ray fluorescence spectrometry can ensure both detection accuracy and analytical efficiency, obtaining complete chemical composition data covering major carbonate components and impurities.

[0050] In addition, to ensure the data quality for subsequent modeling, the original chemical composition data needs to be preprocessed. Specifically, outlier detection is performed on the chemical composition data of each borehole powder sample, and samples whose chemical composition data deviates from the mean by more than a preset multiple of the standard deviation (e.g., 3 times the standard deviation) are removed. For the removed borehole powder samples, their chemical composition data is filled in using interpolation calculations (such as linear interpolation or inverse distance interpolation) based on the chemical composition data of adjacent borehole powder samples. Finally, the chemical composition data of each borehole powder sample is normalized to ensure that data of different dimensions participate in subsequent calculations at the same scale.

[0051] Outlier detection and removal from borehole rock powder samples can eliminate interference data caused by sampling contamination, experimental errors, etc., and avoid these abnormal borehole rock powder samples from distorting the fitting results of the variogram model. By filling in the removed borehole rock powder samples through interpolation calculation, the grade values ​​of missing or abnormal locations can be reasonably estimated while maintaining data integrity. Normalization transformation eliminates the influence of differences in dimensions and orders of magnitude between different chemical component indicators, so that the weight of each indicator in subsequent modeling is determined by its actual degree of variation rather than by its absolute value, thus ensuring the scientificity and objectivity of the modeling process.

[0052] Step S2) The chemical composition data (especially the contents of CaCO3 and MgCO3) of the rock powder samples from each borehole are used as sample values ​​of ore grade at the corresponding borehole locations to construct a set of ore grade sample points covering the area to be mined. Each sample point in this set of ore grade sample points contains three-dimensional spatial coordinates (x, y, z) and a corresponding ore grade value (such as the mass fraction of CaCO3). For borehole locations with depth segments, each segment corresponds to one spatial point and one sample point.

[0053] Since the borehole rock powder samples are derived from rocks at specific depths, their chemical composition data can accurately reflect the grade characteristics of the ore at that location. Therefore, by establishing a direct mapping relationship between the chemical composition data of each borehole rock powder sample and the ore grade sample values, all subsequent modeling and prediction work is based on traceable and verifiable measured data.

[0054] Step S3) Based on the sample point set of ore grade, a variation function model of ore grade is constructed using regionalized variable theory, and the kriging space interpolation algorithm is executed using the variation function model to establish a grade prediction mathematical model that characterizes the three-dimensional spatial distribution law of ore quality in the mining area.

[0055] For details, see attached. Figure 2 As shown, based on the sample point set of ore grades, the variation function model of ore grade is constructed using regionalized variable theory, including the following steps:

[0056] Step S31) Perform normality tests (such as using QQ plots or Shapiro-Wilk tests) and transformations (such as logarithmic transformations or Box-Cox transformations) on the sample values ​​of ore grades for each sample point in the ore grade sample point set. The transformations are performed under the condition that they do not conform to the normal distribution rule.

[0057] Step S32) Based on the processed sample values, calculate the experimental variation function values ​​at different step lengths (i.e., spatial distances, such as step lengths of 10m, 20m, 30m, etc.) and different spatial directions (at least including the ore strike direction, dip direction, and thickness direction).

[0058] Calculating the experimental variogram values ​​at different step sizes characterizes the degree to which ore grade changes with increasing spatial distance. Calculating the experimental variogram values ​​along the three main directions of ore strike, dip, and thickness reveals the differences in the rate and magnitude of ore grade variation in different directions, i.e., spatial anisotropy. This is particularly important for layered or vein-like ore bodies such as marble, because the variation in ore grade along the vertical thickness direction is usually much greater than that along the strike direction. Accurately capturing this anisotropic characteristic is the foundation for subsequent precise interpolation.

[0059] Step S33) Select a theoretical variogram model (e.g., spherical model, exponential model, or Gaussian model) and theoretically fit the experimental variogram values. The fitting method can be the least squares method or manual fitting to obtain the nugget constant, sill value, and range parameters. The range parameters obtained along different directions are different and are used to characterize the spatial anisotropy of ore grade.

[0060] Fitting the theoretical variogram model can transform discrete experimental variogram values ​​into a continuous mathematical model. Here, the nugget constant represents the sum of sampling error and microscale variation, the sill value represents the overall variation amplitude of regionalized variables, and the range parameter defines the effective distance of the spatial correlation of ore grade. These three parameters comprehensively describe the spatial structural characteristics of ore body grade.

[0061] Step S34) Based on the selected theoretical variogram model and the obtained nugget constant, sill value, and range parameter, construct a complete variogram model.

[0062] By combining the theoretical variogram model with the obtained nugget constant, sill value, and range parameters, a complete variogram model is constructed. Using this variogram model, the corresponding variogram value can be calculated at any spatial location within the mining area, based on its spatial distance vector to each known sample point.

[0063] Specifically, the method for establishing a grade prediction mathematical model that characterizes the three-dimensional spatial distribution law of ore quality in the mining area by using the variogram model to execute the Kriging space interpolation algorithm is as follows: In this specific embodiment, the co-Kriging method is preferred to be used as the Kriging space interpolation algorithm.

[0064] The specific method is as follows: Surface geological mapping data (such as stratigraphic occurrence and lithological boundaries) or geophysical exploration data (such as resistivity and magnetic susceptibility) of the area to be mined are obtained as covariates. The sample values ​​of ore grade at each sample point in the ore grade sample point set, after normality testing and transformation, are used as the main variables. Based on the spatial distribution data of the main and covariates, the cross-variance function is calculated, which describes the correlation between the main and covariates at different spatial distances. Then, using the variogram model, cross-variance function, main variables, covariates, and the Kriging spatial interpolation algorithm, a grade prediction mathematical model characterizing the three-dimensional spatial distribution law of ore quality within the area to be mined is established and output.

[0065] Normality testing and transformation are prerequisites for Kriging spatial interpolation. Kriging interpolation is based on statistical assumptions, requiring data to follow a normal distribution; otherwise, the interpolation results may introduce systematic bias. Through logarithmic transformation or Box-Cox transformation, skewed data is converted into an approximately normal distribution, making the subsequently calculated variogram parameters more statistically meaningful. This mathematical transformation eliminates scaling effects and heteroscedasticity in the original data, ensuring statistical comparability of ore grade data from different regions.

[0066] Furthermore, this scheme employs co-kriging interpolation to obtain optimal weighting coefficients by solving a joint system of equations comprising a variogram model, cross-variograms, principal variables, and covariates. The core characteristics of these weighting coefficients are: known sample points closer to the point to be estimated receive greater weights; variables with stronger spatial correlation receive greater weights; and the sum of all weights satisfies the unbiasedness constraint (the sum of principal variable weights is 1, and the sum of covariate weights is 0). Compared to ordinary kriging using only principal variables, co-kriging exhibits lower estimation variance and higher prediction accuracy, especially in areas with low borehole density. The grade prediction mathematical model established through this step is essentially a mathematical function that takes three-dimensional spatial coordinates as input and outputs the predicted ore grade; it encapsulates the spatial distribution information of all known sample points and the spatial structure information of the ore body.

[0067] Step S4) Based on the grade prediction mathematical model, output the predicted ore grade values ​​for each spatial point within the mining area divided by a preset grid. Compare each predicted ore grade value with a preset grading threshold to generate an ore grade zoning map, as shown in the attached figure. Figure 3 As shown, the specific method is as follows:

[0068] Step S41) The predicted ore grade values ​​at each spatial point within the mining area, divided by a preset grid and output by the grade prediction mathematical model, are numerically compared with multiple preset grading thresholds. The preset grading thresholds can be pre-set according to the ore's intended use and processing requirements. For example, areas with an ore grade ≥90% CaCO3 are classified as high-grade areas, areas with an ore grade between 60% and 90% CaCO3 are classified as medium-grade areas, and areas with an ore grade <60% are classified as low-grade areas. Based on the comparison results, each spatial point is divided into a corresponding grade area (high, medium, or low).

[0069] This step transforms continuously changing ore grade predictions into discrete grade categories. Essentially, it maps the numerical results output by the mathematical model to grading standards used in engineering practice. The grading threshold can be flexibly adjusted to adapt to different product specifications and processing requirements. For example, when market demand for high-purity marble powder is strong, the lower limit of the threshold for the high-grade zone can be increased; when bulk aggregates are the main product, the grading standards can be relaxed. This flexibility allows this solution to adapt to different market environments and customer needs, demonstrating strong adaptability.

[0070] Step S42) Use a contour tracing algorithm (such as the Marching Squares algorithm) to extract the spatial boundary lines of each grade region. Enclose the spatial boundary lines of regions belonging to the same grade to form closed region polygons of that grade. Fill different region polygons with different colors (e.g., red for high-grade regions, yellow for medium-grade regions, and gray for low-grade regions) or patterns (e.g., diagonal lines for high-grade regions and dots for low-grade regions).

[0071] The principle of contour line tracing algorithms is to determine the location of contour points in regular grid data through linear interpolation, and then connect them according to specific topological rules to form contour lines. This encloses the boundary lines of areas belonging to the same grade, forming closed polygons, thus completing the geometric construction from lines to surfaces, making each grade area a spatially recognizable unit. Different visual characteristics (colors or patterns) are assigned to different polygon areas, utilizing the human eye's visual sensitivity to color, allowing on-site operators to instantly identify the distribution range and boundary locations of areas of different grades without specialized knowledge. This visual representation significantly lowers the barrier to disseminating technical information within the construction team.

[0072] Step S43) Overlay the filled polygons of each region onto the topographic and geological map of the area to be mined to generate a visualized ore grade zoning map.

[0073] By overlaying ore grade zoning information onto traditional topographic and geological maps, a fusion of geological and ore grade information is achieved. Topographic and geological maps provide information on the macroscopic morphology, structural features, and topographic variations of ore bodies, while ore grade zoning information provides spatial distribution information on ore quality. This overlay allows decision-makers to simultaneously access both types of key information in a single view, facilitating comprehensive judgment and scientific decision-making. The generated ore grade zoning map can be printed as two-dimensional drawings and distributed to construction teams, or loaded onto a digital mining platform or tablet computer for digital guidance of on-site operations.

[0074] Step S5) Based on the generated ore grade zoning map, the area to be mined is divided into multiple blasting operation blocks with different ore grade levels.

[0075] In this embodiment, based on the predicted ore grade, the area to be mined is divided into a high-grade area in the first blasting operation block (i.e., step 4), a low-grade area in the second blasting operation block (i.e., step 4), and a medium-grade area in the third blasting operation block (i.e., step 4). The predicted ore grade of the first blasting operation block is greater than that of the second blasting operation block, and the predicted ore grade of the third blasting operation block falls between the two.

[0076] Specifically, a set of blasting process parameters matching the ore grade is determined for each blasting operation block. Hole grid parameters include borehole spacing and row spacing; charge parameters include single-hole charge amount and charge structure type; and detonation sequence parameters include detonation method, detonation time difference, and detonation sequence.

[0077] The borehole network parameters (borehole spacing and row spacing) determine the borehole density and energy input density per unit area; the charge parameters (charge per borehole and charge structure type) determine the total explosive energy within the borehole and its distribution along the borehole depth; and the detonation sequence parameters (detonation method, detonation time difference, and detonation sequence) determine the interaction mode of explosive energy between boreholes. These three parameters are set independently but work synergistically to jointly determine the final blasting effect. Using these parameters as a tiered classification provides a clear operational framework for subsequent differentiated parameter matching.

[0078] Specifically, for the first blasting operation block (high-grade zone), the blasting process parameters adopted are the first borehole spacing, the first row spacing, the first single-hole charge amount, the air-gap charge structure, and the sequential detonation sequence.

[0079] For the second blasting operation block (low-grade zone), the blasting process parameters adopted are the second borehole spacing, the second row spacing, the second single-hole charge amount, the continuous charge structure, and the inter-row micro-delay initiation sequence.

[0080] In this configuration, the spacing between the first and second boreholes is less than the spacing between the second and third boreholes; the spacing between the first and second rows is less than the spacing between the second and third rows; and the charge per borehole is less than the charge per borehole. For example, the spacing between the first and second boreholes is set to 2.5m, and the spacing between the second and third boreholes is set to 4.0m; the spacing between the first and second rows is set to 2.0m, and the spacing between the second and third rows is set to 3.5m; the charge per borehole is set to 30kg, and the charge per borehole is set to 60kg.

[0081] For the third blasting operation block (medium grade zone), the borehole spacing is greater than the first borehole spacing and less than the second borehole spacing, the row spacing is greater than the first row spacing and less than the second row spacing, the single-hole charge is greater than the first single-hole charge and less than the second single-hole charge, the charge structure adopts partial air gap charging (i.e., an air column is set in the upper part of the borehole, and the lower part is continuously charged), and the inter-hole delay time of the initiation sequence is greater than the inter-hole delay time of the sequential initiation sequence and less than the inter-hole delay time of the row-to-row micro-differential initiation sequence.

[0082] For the first blasting block, a combination of parameters—small hole spacing, small row spacing, and low charge—was used. The principle behind this is that the small hole mesh parameters result in a more uniform distribution of explosive energy, reducing the risk of localized over-crushing; the low charge per hole reduces the explosive energy acting on a unit volume of ore; the air-spaced charge structure creates an air cushion effect in the borehole, reducing peak pressure but extending the duration of impact; and sequential detonation avoids stress superposition between adjacent boreholes. The combined application of these techniques allows the explosive stress wave to act on the high-grade ore in a gentler manner, reducing excessive fracture development and thus protecting the ore's block size integrity. The second blasting block uses the opposite parameters, aiming to enhance the crushing effect, as the ore value in this area is lower and the block size requirements are relatively lenient. The parameters for the third blasting block fall between the two, achieving a balance between protection and crushing.

[0083] For example, the air-gap charge structure used in the first blasting operation block is as follows: the total charge is divided into an upper charge section and a lower charge section in the borehole, and an inert material section (such as rock powder, sand or air column) is set between the upper charge section and the lower charge section. The charge density of the lower charge section is greater than that of the upper charge section (for example, the lower charge section uses high-velocity explosive and the upper charge section uses low-velocity explosive). A detonator is set in the lower charge section.

[0084] This differentiated charge density design between the upper and lower sections utilizes the principle that explosive energy preferentially propagates along the path of least resistance. The rock mass around the bottom of the borehole is constrained by the rock pillar above, making fracturing more difficult and requiring higher energy input; while the rock mass around the borehole opening is less constrained, and excessive charging can easily cause flyrock and over-crushing. Therefore, a higher charge density is used in the lower charge section to ensure effective fracturing at the bottom of the borehole; a lower charge density is used in the upper charge section to reduce impact on the high-grade ore at the borehole opening. An inert material section isolates the upper and lower charge sections, allowing the explosive energy of the lower charge to preferentially propagate downwards and to the surrounding area, further protecting the upper rock mass. This structural design enables a differentiated spatial distribution of explosive energy, precisely matching the fracturing requirements of rock masses at different depths.

[0085] For example, the continuous charge structure used in the second blasting operation block is as follows: explosives are continuously loaded along the entire length of the borehole, the diameter of the explosives is equal to the diameter of the borehole to form a coupled charge state, and a detonator is set at the bottom third position of the borehole, and a reverse detonation method is adopted.

[0086] Continuous and coupled charging configurations ensure that the shock wave and propellant gases generated during the explosion act entirely on the borehole wall, maximizing energy transfer efficiency. This is suitable for secondary blasting operations requiring strong fracturing effects. The detonator is located one-third of the way up from the bottom of the borehole and uses a reverse initiation method, causing the detonation wave to propagate upwards. This prolongs the time the propellant gases act on the borehole wall, enhancing the fracturing effect. The principle is that during reverse initiation, the detonation wave propagates from the bottom of the borehole to the opening. The rock at the bottom is pre-damaged by the shock wave before the propellant gases expand, and then the propellant gases further expand the fractures under higher pressure, resulting in more thorough fracturing.

[0087] Step S6) Perform drilling, charging, blasting and loading operations according to the blasting process parameter set corresponding to each blasting operation block. Among them, the blasted ore of different grades is loaded and transported separately by section.

[0088] Specifically, during drilling rig positioning, perforations with corresponding hole mesh parameters are drilled within different blocks according to the block boundaries marked on the ore grade zoning map. During charging, operators select the corresponding charging structure and charge amount based on the block identification. During detonation, detonation is carried out according to the preset detonation sequence (high-grade zones are detonated hole by hole, low-grade zones are detonated with slight inter-row delay, and medium-grade zones are detonated with intermediate delay).

[0089] After blasting, the ore is loaded and transported in separate zones. The specific method is as follows: Temporary dividing markers (such as colored flags, marker posts, or lime lines) are placed on the surface of the blasted ore pile according to the boundary lines of the ore grade zoning map. Loading equipment operators (such as excavators or loaders) sequentially load the blasted ore from the first blasting block (high-grade zone), the third blasting block (medium-grade zone), and the second blasting block (low-grade zone) according to the temporary dividing markers. During loading, the high-grade area is loaded first, followed by the medium-grade area, and finally the low-grade area to avoid cross-loading. Ore of different grades is unloaded into different transport vehicles or different transport batches; for example, high-grade ore is transported to a high-end powder processing line, while low-grade ore is transported to a building material aggregate production line or stored as waste rock.

[0090] Zonal loading and separate transportation are the final implementation steps of graded mining technology. Its core value lies in transforming the zoning information obtained through predictive models before blasting into operational guidelines for actual mining operations after blasting. The placement of temporary separating markers on the blast pile surface transforms abstract grade zoning boundaries into physically visible on-site markers, eliminating uncertainty in information transmission. Loading is done sequentially from high to medium to low grade. The principle behind this is to prioritize the removal of high-value ore, preventing secondary mixing when loading low-grade ore later. Simultaneously, removing high-grade areas first creates working space, facilitating the loading of lower-grade areas later. Different grades of ore are unloaded into different vehicles or batches, achieving refined classification from the source of mining. Downstream processing plants can adopt different processing parameters based on the grade of each batch of ore, thereby improving the quality and efficiency of the entire process.

[0091] Step S7) Dynamic feedback correction step.

[0092] After blasting, rapid on-site detection methods (such as handheld XRF analyzers or sending samples to the laboratory) are used to determine the distribution of the actual ore grade after the blast. These measured actual ore grade values ​​are then compared point-by-point with the predicted ore grade values ​​for the corresponding spatial locations output by the grade prediction mathematical model, and the deviation (e.g., mean absolute error or root mean square error) between the actual and predicted ore grade values ​​is calculated.

[0093] The calculation of deviation values ​​forms the basis for quantitatively evaluating the model's prediction accuracy. By comparing the actual measured values ​​with the model's predicted values ​​point by point, it is possible to identify which spatial regions have larger prediction deviations, thus allowing for targeted corrections. The mean absolute error reflects the overall level of prediction error, while the root mean square error is more sensitive to large deviations. Using both together provides a comprehensive evaluation of model performance. This evaluation result provides a quantitative basis for whether to trigger a subsequent model update.

[0094] When the deviation exceeds a preset threshold (e.g., mean absolute error exceeds 5%), the model update procedure is initiated: the actual ore grade value measured this time is added as a new sample point to the ore grade sample point set. Then, based on the ore grade sample point set with these newly added sample points, the variogram model construction steps are re-executed (i.e., normality test, experimental variogram calculation, and theoretical fitting are re-performed), thereby updating the variogram model. Simultaneously, the preset grading threshold can be appropriately adjusted according to the deviation (e.g., adjusting the lower boundary of the high-grade zone from 90% to 88%).

[0095] The dynamic feedback correction mechanism enables closed-loop self-optimization of the mathematical model. As mining progresses, newly exposed orebody interfaces provide new measured data points for model updates. Incorporating these new sample points into the original sample set increases the spatial distribution density of the sample points, especially in areas where boreholes were previously sparse, providing new data support. Refitting the variogram function allows spatial structure parameters such as the nugget constant and range parameter to continuously optimize with increasing information. This continuous optimization mechanism enables the mathematical model to adapt to gradual changes in ore body grade, avoiding long-term prediction biases caused by limited initial modeling samples, and ensuring the long-term stable operation of the graded mining technology under actual production conditions.

[0096] In summary, the marble powder ore grading mining method provided by this invention forms a complete and refined mining technology process with perforation analysis, geological modeling, grade zoning, differential blasting, grading loading, and dynamic feedback as the main lines.

[0097] The advantages of this method are as follows: First, by acquiring the chemical composition data of borehole rock powder samples during the drilling stage and mapping it to sample values ​​of ore grade, the quality evaluation node is significantly moved forward from after blasting to the drilling stage, creating conditions for scientific decision-making before mining. Second, by utilizing variogram analysis and Kriging interpolation techniques in geostatistics, a three-dimensional prediction model capable of characterizing the spatial variability of grade in complex marble ore bodies is constructed, solving the problem of the lack of quantitative prediction tools in the production stage of non-metallic mines. Third, differentiated blasting parameters are matched for different grade areas based on the ore grade zoning map, especially for high-grade areas, where a combination of small-hole mesh, low charge, air gap, and sequential detonation is adopted, significantly reducing blasting damage to high-value ores. Finally, closed-loop optimization is achieved through zoned loading and transportation and dynamic feedback correction.

[0098] This method improves the recovery rate of high-grade ore and significantly reduces the waste rock mixing rate by accurately identifying high-grade ore bodies and implementing protective blasting, thus directly enhancing the economic benefits of the mine. On the other hand, it increases loading efficiency by avoiding blockage caused by mixed ore and rock, reducing equipment wear and energy consumption. Furthermore, this method provides a replicable and scalable solution for the refined mining of marble and similar non-metallic mines, promoting the industry's transformation and upgrading towards digitalization and green development.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit the technical solutions. Those skilled in the art should understand that any modifications or equivalent substitutions to the technical solutions of the present invention without departing from the spirit and scope of the present invention should be covered within the scope of the claims of the present invention.

Claims

1. A method for graded mining of marble powder, characterized in that, Includes the following steps: During the drilling operation, borehole rock powder samples are collected along the area to be mined according to a preset grid. Multi-dimensional physicochemical property analysis is performed on each borehole rock powder sample to obtain chemical composition data, which includes carbonate content and impurity content. The chemical composition data of rock powder samples from each borehole are used as sample values ​​of ore grade at the corresponding borehole locations to construct a set of ore grade sample points covering the area to be mined. Based on the set of ore grade sample points, a variation function model of ore grade is constructed using regionalized variable theory, and the kriging space interpolation algorithm is executed using the variation function model to establish a grade prediction mathematical model that characterizes the three-dimensional spatial distribution law of ore quality in the mining area. The grade prediction mathematical model outputs the ore grade prediction values ​​at each spatial point in the mining area divided by a preset grid. The ore grade prediction values ​​are then compared with preset grading thresholds to generate an ore grade zoning map. Based on the ore grade zoning map, the area to be mined is divided into multiple blasting operation blocks with different ore grade levels, and a set of blasting process parameters matching the ore grade level is determined for each blasting operation block. The set of blasting process parameters includes hole mesh parameters, charge parameters and detonation timing parameters. According to the blasting process parameter set corresponding to each blasting operation block, drilling, charging, blasting and loading operations are carried out respectively, and the blasted ore of different grades is loaded and transported separately in separate sections.

2. The method for graded mining of marble powder ore according to claim 1, characterized in that, Based on the aforementioned set of ore grade sample points, the method for constructing a variation function model of ore grade using regionalized variable theory is as follows: The normality test and transformation processing are performed on the sample values ​​of ore grade of each sample point in the ore grade sample point set to obtain the processed sample values. Based on the processed sample values, the experimental variogram values ​​are calculated at different step sizes and in different spatial directions; Select a theoretical variogram model and theoretically fit the experimental variogram values ​​to obtain the nugget constant, sill value, and range parameters. The variogram model is constructed based on the selected theoretical variogram model and the obtained nugget constant, sill value, and range parameters. The different spatial directions include at least the direction along the ore strike, the dip direction, and the thickness direction.

3. The method for graded mining of marble powder ore according to claim 2, characterized in that, The method for establishing a grade prediction mathematical model that characterizes the three-dimensional spatial distribution law of ore quality in the mining area by using the aforementioned variogram model and executing the Kriging space interpolation algorithm is as follows: The Kriging spatial interpolation algorithm is a co-Kriging method. It obtains surface geological mapping data or geophysical exploration data of the area to be mined as covariates, uses the processed sample values ​​as principal variables, calculates the cross-variance function based on the principal variables and the covariates, and uses the variance function model, the cross-variance function, the principal variables and the covariates to execute the Kriging spatial interpolation algorithm to establish the grade prediction mathematical model.

4. The method for graded mining of marble powder ore according to claim 1, characterized in that, The method for generating an ore grade zoning map by outputting the predicted ore grade values ​​at each spatial point within the mining area according to a preset grid based on the grade prediction mathematical model, and comparing each predicted ore grade value with a preset grading threshold, is as follows: The predicted ore grade values ​​at each spatial point within the mining area, divided by a preset grid and output by the grade prediction mathematical model, are compared with multiple preset grading thresholds. Based on the comparison results, the spatial points corresponding to each predicted ore grade value are divided into multiple grade areas. The contour tracing algorithm is used to extract the spatial boundary lines of each level of region. The spatial boundary lines of the same level of region are enclosed to form the region polygon of that level of region. Different region polygons are filled with different colors or patterns. The filled polygons of each region are superimposed on the topographic and geological map of the area to be mined to generate an ore grade zoning map.

5. The method for graded mining of marble powder ore according to claim 1, characterized in that, The hole mesh parameters include borehole spacing and row spacing; the charge parameters include single-hole charge amount and charge structure type; the detonation timing parameters include detonation method, detonation time difference, and detonation sequence.

6. The method for graded mining of marble powder ore according to claim 5, characterized in that, Based on the ore grade zoning map, the area to be mined is divided into multiple blasting operation blocks with different ore grade levels. A set of blasting process parameters matching the ore grade level is determined for each blasting operation block. The method for determining the blasting process parameter set, including hole mesh parameters, charge parameters, and detonation timing parameters, is as follows: According to the ore grade zoning map, the area to be mined is divided into a first blasting operation block, a second blasting operation block, and a third blasting operation block. The predicted ore grade of the first blasting operation block is greater than that of the second blasting operation block, and the predicted ore grade of the third blasting operation block is between the predicted ore grade of the first blasting operation block and the predicted ore grade of the second blasting operation block. Specifically, for the first blasting operation block, the blasting process parameters adopted are the first borehole spacing, the first row spacing, the first single-hole charge amount, the air-gap charge structure, and the sequential detonation sequence; for the second blasting operation block, the blasting process parameters adopted are the second borehole spacing, the second row spacing, the second single-hole charge amount, the continuous charge structure, and the inter-row micro-delay detonation sequence; wherein, the first borehole spacing is smaller than the second borehole spacing, the first row spacing is smaller than the second row spacing, and the first single-hole charge amount is smaller than the second single-hole charge amount; The borehole spacing in the third blasting operation block is greater than the borehole spacing in the first block and less than the borehole spacing in the second block. The row spacing is greater than the row spacing in the first block and less than the row spacing in the second block. The charge amount per borehole is greater than the charge amount per borehole in the first block and less than the charge amount per borehole in the second block. The charge structure adopts partial air gap charging. The borehole delay time in the initiation sequence is greater than the borehole delay time in the sequential borehole initiation sequence and less than the borehole delay time in the row-to-row micro-delay initiation sequence.

7. The method for graded mining of marble powder ore according to claim 6, characterized in that, The air-gap charge structure used in the first blasting operation block is as follows: the total charge is divided into an upper charge section and a lower charge section in the borehole, and there is an inert material section between the upper charge section and the lower charge section. The charge density of the lower charge section is greater than that of the upper charge section, and a detonator is installed in the lower charge section. The continuous charge structure used in the second blasting operation block is as follows: explosives are continuously loaded along the entire length of the borehole, and the diameter of the explosives is equal to the diameter of the borehole to form a coupled charge state. A detonator is installed at the bottom third of the borehole and a reverse detonation method is adopted.

8. The method for graded mining of marble powder ore according to claim 1, characterized in that, Outlier detection was performed on the chemical composition data of each borehole rock powder sample, and borehole rock powder samples whose chemical composition data deviated from the mean by more than a preset multiple of the standard deviation were removed. The chemical composition data of the removed borehole rock powder samples were filled by interpolation calculation based on the chemical composition data of adjacent borehole rock powder samples. Then, the chemical composition data of each borehole rock powder sample were normalized.

9. The method for graded mining of marble powder ore according to claim 1, characterized in that, The method for shoveling and transporting blasted ore of different grades in separate zones is as follows: After the blasting operation is completed, temporary dividing markers are set up on the surface of the blasted ore pile according to the boundary lines of the ore grade zoning map. The blasted ore of the first blasting operation block, the third blasting operation block and the second blasting operation block are shoveled in sequence according to the temporary dividing markers. Ore of different grades is unloaded into different transport vehicles or different transport batches.

10. The method for graded mining of marble powder ore according to claim 1, characterized in that, The marble powder ore grading mining method further includes a dynamic feedback correction step: after blasting, the distribution of the actual ore grade after the blasting is determined and compared with the ore grade prediction value output by the grade prediction mathematical model, and the deviation value between the actual ore grade value and the ore grade prediction value is calculated; when the deviation value exceeds a preset threshold, the actual ore grade value is added as a new sample point to the ore grade sample point set, and the variation function model is updated based on the ore grade sample point set with the added actual ore grade value.

Images