Ore screening method and device, electronic equipment and storage medium

Abstract

The invention provides an ore screening method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining ore multi-scale fractal feature data before crushing, and carrying out the calculation to obtain a comprehensive fractal dimension and a plurality of sub fractal dimensions; dynamically determining crushing and grading parameters based on the comprehensive fractal dimension to obtain coarse-fraction and fine-fraction products; dynamically optimizing photoelectric sensing parameters based on the sub-fractal dimensions, carrying out high-precision photoelectric separation on coarse-grained products, obtaining coarse-grained concentrates in advance, and discarding tailings; and determining ore grinding and magnetoelectric separation parameters based on the fractal dimension, and efficiently recovering the fine-fraction materials. The fractal dimension serves as an intelligent decision-making core penetrating through the whole process, self-adaptive collaboration of crushing, grading, sorting and ore grinding links is achieved, the sorting precision and the resource recovery rate are remarkably improved, meanwhile, system energy consumption is reduced, and important industrial application value is achieved.

Description

Technical Field

[0001] This application relates to the field of ore screening technology, and more specifically, to a method, apparatus, electronic device, and storage medium for ore screening. Background Technology

[0002] In the field of mineral processing technology, traditional ore screening processes suffer from problems such as low sorting accuracy, high energy consumption, poor adaptability, and insufficient system synergy. Specifically, traditional crushing and screening use fixed parameters, which cannot be dynamically adjusted according to the physical characteristics of the ore (such as hardness, fissures, and embedding characteristics). This results in an unsatisfactory particle size distribution of the crushed product, easily leading to over-crushing or under-dissociation, which affects the efficiency of subsequent sorting and causes energy waste.

[0003] Traditional photoelectric sorting systems have fixed parameters, which leads to decreased recognition accuracy and poor sorting stability when the ore composition, color, and texture are complex. In addition, existing processes lack global coordination, with each stage (crushing, grading, sorting, and grinding) operating in isolation and parameter settings disconnected from each other. This makes it difficult to formulate a full-process optimization plan based on ore characteristics, thus restricting the improvement of overall resource recovery rate and economic benefits. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method, apparatus, electronic device and storage medium for ore screening to overcome the problems in the prior art.

[0005] In a first aspect, embodiments of this application provide a method for ore screening, the method comprising: Based on the multi-scale fractal characteristic data of the original ore before the crushing operation, multiple sub-fractal dimensions and the comprehensive fractal dimension were calculated. The raw ore is crushed based on the screening parameters determined by the comprehensive fractal dimension to obtain the undersize product that meets the requirements, and the undersize product is then classified by particle size to obtain coarse-grained product and fine-grained product. Based on the photoelectric sensing parameters determined by the subfractal dimension, the coarse-grained product is photoelectrically separated to obtain coarse-grained concentrate and coarse-grained tailings. Based on the grinding parameters determined by the sub-fractal dimension and the comprehensive fractal dimension, the fine-grained product and the coarse-grained tailings are ground and classified, and the ground and classified product is then subjected to dry magnetic-electric composite separation to obtain fine-grained concentrate and final tailings.

[0006] In some technical solutions of this application, the above-mentioned calculation of multiple sub-fractal dimensions and comprehensive fractal dimensions based on the multi-scale fractal characteristic data of the original ore before crushing operations includes: The fractal dimension of the spectral structure is calculated based on the mineral phase distribution based on hyperspectral image recognition, the fractal dimension of the morphological fracture is calculated based on the digital elevation model generated from laser three-dimensional point cloud data, and the fractal dimension of the microstructure is calculated based on the particle boundary extracted from scanning electron microscope images. The comprehensive fractal dimension is calculated based on the spectral fractal dimension and the first weighting coefficient, the morphological crack fractal dimension and the second weighting coefficient, and the microstructure fractal dimension and the third weighting coefficient; wherein the sum of the first weighting coefficient, the second weighting coefficient and the third weighting coefficient is 1.

[0007] In some technical solutions of this application, the screening parameters include the working size of the discharge port and the screening particle size cutting point; the method determines the screening parameters in the following ways: Based on the comprehensive fractal dimension, the discharge port width of the crushing equipment is determined through a first mapping relationship; wherein, the first mapping relationship is a linear relationship in which the discharge port width is negatively correlated with the comprehensive fractal dimension; Based on the comprehensive fractal dimension, the particle size cutting point of the screening equipment is determined through the second mapping relationship.

[0008] In some technical solutions of this application, the screening parameters determined by the comprehensive fractal dimension described above are used to crush the raw ore, including: If the oversize product does not meet the requirements, it is crushed again until the undersize product is obtained.

[0009] In some technical solutions of this application, the aforementioned photoelectric sensing parameters include the energy value of the X-ray source and the frequency value of the three-dimensional contour scan. The method determines the photoelectric sensing parameters in the following manner: The energy value of the X-ray source is determined based on the spectral fractal dimension in the sub-fractal dimension, wherein the spectral fractal dimension is positively correlated with the energy value; The frequency value of the three-dimensional contour scan is determined based on the shape fractal dimension in the sub-fractal dimension, wherein the shape fractal dimension and the frequency value are negatively correlated.

[0010] In some technical solutions of this application, the above-mentioned grinding parameters include the Bonder index correction coefficient and the target grinding fineness, and the method determines the grinding parameters in the following manner: Based on the combined fractal dimension and micro-fractal dimension of the fine-grained product and the coarse-grained tailings, the Bondon index correction coefficient and the target grinding fineness are determined by calling the database strategy network through the control module of the process parameter calibration process.

[0011] In some technical solutions of this application, the above-mentioned dry magnetic-electric composite separation of the products after grinding and classification includes: Based on the fractal dimension of the spectrum, the fractal dimension of the morphological fractures, and the fractal dimension of the microstructure as inputs, and combined with the spectral data of the hyperspectral image, the content of magnetic minerals and the electrical conductivity index are predicted by calling the magnetoelectric sorting parameter configuration table. The separation field strength of the high-intensity magnetic separator is determined based on the content of the magnetic minerals. The sorting voltage of the high-voltage electric separator is determined based on the conductivity index. Dry magnetoelectric composite sorting is performed based on the determined sorting field strength and sorting voltage.

[0012] Secondly, embodiments of this application provide an apparatus for ore screening, the apparatus comprising: The fractal feature modeling module is used to calculate multiple sub-fractal dimensions and the overall fractal dimension based on the multi-scale fractal feature data of the original ore before the crushing operation. The intelligent crushing and screening module is used to crush the raw ore based on the screening parameters determined by the comprehensive fractal dimension to obtain the undersize product that meets the requirements. The particle size classification module is used to classify the undersize product into coarse-grained and fine-grained products. The photoelectric sorting module is used to perform photoelectric sorting on the coarse-grained product based on the photoelectric sensing parameters determined by the sub-fractal dimension, to obtain coarse-grained concentrate and coarse-grained tailings; The grinding and magnetic-electric composite separation module is communicatively coupled with the photoelectric separation module and the particle size classification module. It is used to perform grinding and classification processing on the fine-grained product and the coarse-grained tailings based on the grinding parameters determined by the sub-fractal dimension and the comprehensive fractal dimension, and to perform dry magnetic-electric composite separation on the product after grinding and classification processing to obtain fine-grained concentrate and final tailings.

[0013] Thirdly, embodiments of this application provide an electronic device, a processor, a memory, and a bus. The memory stores machine instructions executed by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine instructions are executed by the processor, the steps of the above-described ore screening method are performed.

[0014] Fourthly, embodiments of this application provide a computer storage medium storing a computer program, which, when run by a processor, executes the steps of the above-described ore screening method.

[0015] The technical solutions provided by the embodiments of this application may include the following beneficial effects: The method of this application includes calculating multiple sub-fractal dimensions and a comprehensive fractal dimension based on the multi-scale fractal characteristic data of the original ore before crushing; crushing the original ore based on screening parameters determined by the comprehensive fractal dimension to obtain undersize products that meet the requirements, and classifying the undersize products by particle size to obtain coarse-grained products and fine-grained products; performing photoelectric separation on the coarse-grained products based on photoelectric sensing parameters determined by the sub-fractal dimensions to obtain coarse-grained concentrate and coarse-grained tailings; performing grinding and classification processing on the fine-grained products and the coarse-grained tailings based on grinding parameters determined by the sub-fractal dimensions and the comprehensive fractal dimension, and performing dry magnetic-electric composite separation on the products after grinding and classification processing to obtain fine-grained concentrate and final tailings.

[0016] This application automatically optimizes crushing and grading parameters based on the physical structure characteristics of the ore, ensuring from the source that the product particle size meets the optimal requirements for subsequent sorting, effectively avoiding energy waste and over-grinding. Based on the obtained multiple sub-fractal dimensions, key parameters are dynamically set to match the photoelectric sorting and fine particle sorting stages, thereby significantly improving the sorting accuracy and efficiency of coarse and fine particles. By integrating the core feature of fractal dimension into the entire sorting process, the problem of fragmented processes and fixed parameters in traditional systems is fundamentally solved, significantly improving the overall resource recovery rate and sorting efficiency while achieving a substantial reduction in system energy consumption.

[0017] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating a method for ore screening provided in an embodiment of this application is shown. Figure 2 A schematic diagram of an ore screening apparatus provided in an embodiment of this application is shown; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the accompanying drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0021] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0022] It should be noted that the term "comprising" will be used in the embodiments of this application to indicate the presence of the features declared thereafter, but does not exclude the addition of other features.

[0023] In the field of material sorting, traditional sorting technologies face numerous challenges. On the one hand, traditional material particle size cutting methods lack precision and efficiency. For example, in the ore crushing process, conventional crushers struggle to precisely control the degree of crushing based on the ore's characteristics and target particle size requirements, resulting in uneven particle size distribution. This not only affects subsequent sorting effects but may also lead to energy waste and excessive equipment wear. Simultaneously, traditional screening equipment features fixed screen openings, unable to adaptively adjust to changes in material properties, making it difficult to meet the demands for efficient screening of materials of different particle sizes.

[0024] On the other hand, traditional photoelectric identification and sorting technologies suffer from low sorting accuracy and poor adaptability. In complex material environments, the differences in physical and chemical properties between different materials may not be significant, making it difficult for traditional photoelectric identification methods to accurately distinguish and identify them, resulting in a high sorting error rate. Moreover, when the characteristics of materials change, such as fluctuations in color, shape, or composition, traditional photoelectric identification and sorting systems often cannot adapt and adjust quickly, affecting the stability and reliability of sorting. Furthermore, in traditional material sorting systems, the particle size distribution stage and the photoelectric identification and sorting stage are independent of each other, lacking effective coupling and synergy, failing to fully leverage their respective advantages, and further limiting the improvement of sorting efficiency and accuracy.

[0025] Based on this, the present application provides a method, apparatus, electronic device and storage medium for ore screening, which are described below through embodiments.

[0026] Figure 1 The diagram illustrates a flow chart of an ore screening method provided in an embodiment of this application, wherein the method includes steps S101-S104; specifically: S101. Based on the multi-scale fractal characteristic data of the original ore before the crushing operation, calculate multiple sub-fractal dimensions and the comprehensive fractal dimension; S102. The raw ore is crushed based on the screening parameters determined by the comprehensive fractal dimension to obtain the screened product that meets the requirements, and the screened product is classified by particle size to obtain coarse-grained product and fine-grained product. S103. Based on the photoelectric sensing parameters determined by the subfractal dimension, the coarse-grained product is photoelectrically separated to obtain coarse-grained concentrate and coarse-grained tailings. S104. Based on the grinding parameters determined by the sub-fractal dimension and the comprehensive fractal dimension, the fine-grained product and the coarse-grained tailings are ground and classified, and the ground and classified product is subjected to dry magnetic-electric composite separation to obtain fine-grained concentrate and final tailings.

[0027] This application automatically optimizes crushing and grading parameters based on the physical structure characteristics of the ore, ensuring from the source that the product particle size meets the optimal requirements for subsequent sorting, effectively avoiding energy waste and over-grinding. Based on the obtained multiple sub-fractal dimensions, key parameters are dynamically set to match the photoelectric sorting and fine particle sorting stages, thereby significantly improving the sorting accuracy and efficiency of coarse and fine particles. By integrating the core feature of fractal dimension into the entire sorting process, the problem of fragmented processes and fixed parameters in traditional systems is fundamentally solved, significantly improving the overall resource recovery rate and sorting efficiency while achieving a substantial reduction in system energy consumption.

[0028] The following describes some embodiments of this application in detail. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0029] This application provides a method for ore screening. This method is not a general ore processing technology, but a specialized method specifically designed by deeply coupling the unique multi-scale structural heterogeneity and mechanical response behavior of ores, which can be quantified by fractal theory. Specifically, the method flow is entirely built around the physical essence of the specific object of ore: it begins with the fractal characteristic analysis of its macroscopic morphology, mineral distribution, and microstructure, and based on this fractal dimension, it dynamically determines key process parameters such as discharge port width, photoelectric sensing parameters, and grinding intensity by establishing mapping models directly related to the physical mechanisms of crushing, sorting, and grinding. The construction of these models directly depends on the fractal characterization parameters unique to the ore, such as hardness, embedding relationship, surface roughness, and composition distribution. Finally, the goal is to achieve synergistic optimization of the entire process, improve resource recovery rate, and reduce system energy consumption as the benchmark for effect evaluation and closed-loop control. This complete technical system, which is deeply rooted in the physical nature of ore materials and the separation mechanism from the analysis object and mechanism model to the control target, makes its scheme logic, parameter correlation and control criteria fundamentally different from the ore separation process with general fixed parameters or other material separation fields.

[0030] This application provides a method for ore screening, which begins with the analysis and quantification of multi-scale fractal characteristics of raw ore before crushing. Fractal characteristic data refers to image or point cloud data acquired through different observation methods that characterize the complexity and irregularity of the ore's structure from macroscopic to microscopic levels. Specifically, this data covers three main scales: at the macroscopic scale, spectral image data of the ore surface is acquired using a hyperspectral imager, with a wavelength range typically covering 400-2500 nanometers; simultaneously, point cloud data of the surface morphology and macroscopic fractures of the ore block are acquired using a laser 3D scanner, with a scanning accuracy of ±1 to 5 millimeters. At the microscopic scale, representative mineral samples are acquired using an automatic sampling device and observed using a scanning electron microscope at magnifications of 500 to 5000 times to obtain image data showing the mineral particle embedding relationship and microstructure.

[0031] It should be noted that in practical applications, it has been found that image data of ores often introduce noise due to factors such as on-site dust, vibration, and changes in lighting, leading to instability in fractal dimension calculation. To solve this problem, this system establishes a single quantitative index that can guide the entire subsequent process during fractal calculation. It creatively integrates three sub-fractal dimensions with different physical meanings and scales through weighted fusion, effectively improving the robustness and accuracy of fractal feature calculation.

[0032] The so-called sub-fractal dimension refers to a numerical parameter calculated based on data at different scales, used to quantify the structural complexity of a specific aspect. It mainly includes three categories: First, the "spectral fractal dimension," which is calculated using box counting to obtain a binary image of the distribution of a specific mineral phase (such as magnetite) after mineral phase identification of hyperspectral image data (e.g., using spectral angle mapping). This dimension reflects the aggregation, interweaving, and complexity of the target mineral in spatial distribution. Second, the "morphological fracture fractal dimension," which is calculated using projection covering or variogram methods based on a digital elevation model reconstructed from laser 3D scanning point cloud data. This dimension characterizes the surface roughness, undulation characteristics, and the development and complexity of the macroscopic fracture network of the ore. Third, the "microstructural fractal dimension," which is calculated using box counting or perimeter-area methods after extracting particle boundaries from scanning electron microscope images. This dimension characterizes the tortuosity of the mineral particle's own morphology, the complexity of the pore structure, or the tortuosity and complexity of its interlocking boundaries.

[0033] Basic definition and formula of box counting:

[0034] In the formula, For the subfractal dimension, To cover the side length of the box, The minimum number of boxes required to cover the foreground of ore particles, when When it approaches 0, and The relationship is linear, and its slope is the subfractal dimension D. .

[0035] a. Fractal dimension of spectral distribution ( Calculation: Based on hyperspectral data, a binary image of the spatial distribution of the target mineral is extracted. Box counting is used to count the number of mineral facies regions covered by boxes of different sizes. Based on the definition and calculation principle of fractal dimension, the dimension representing the heterogeneity and embedding complexity of the mineral's spatial distribution is calculated. .

[0036] b. Fractal dimension of morphological cracks ( Calculation: The laser-generated 3D point cloud data is registered, denoised, and meshed to generate a digital elevation model (DEM). The fractal dimension of the ore surface roughness and fracture network is calculated using either the projection cover method or the variogram method. .

[0037] c. Fractal dimension of microstructure ( Calculation: Based on scanning electron microscopy (SEM) images, extract the contours of mineral grains or grain boundaries. Using the perimeter-area relationship method, for a set of grains, the perimeter P and area A satisfy a fractal relationship: By performing a linear fit on ln P and ln A, twice the slope represents the difficulty of mineral microscopic dissociation and the complexity of the particle structure itself. .

[0038] Fusion modeling of combined fractal dimension (D): To establish a single quantitative indicator that can guide the entire subsequent process, this invention creatively combines the three sub-fractal dimensions with different physical meanings and scales through weighted fusion. The comprehensive fractal dimension is not directly measured; it is an integrated indicator calculated by weighted fusion of the aforementioned multiple sub-fractal dimensions.

[0039] Where D represents the comprehensive fractal dimension; , , These represent the aforementioned spectral fractal dimension, morphological fracture fractal dimension, and microstructural fractal dimension, respectively; α, β, and γ are pre-calibrated weighting coefficients, the sum of which is 1. Their specific values ​​depend on the degree of influence of the structural characteristics represented by each sub-dimension on subsequent crushing operations, and are typically determined through regression analysis of process test data from a large number of ore samples. Thus, the output includes ( , , , The initial fractal model of the ore is used. The comprehensive fractal dimension D, as a global feature index, comprehensively reflects the overall structural complexity and mechanical behavior tendency of the ore at three levels: composition distribution, macroscopic morphology and microstructure, providing a unified and quantitative core basis for subsequent intelligent decision-making throughout the entire process.

[0040] In traditional crushing and screening processes, equipment parameters are fixed and cannot adapt to the structural characteristics of different ores; the screening cutting points lack specificity, leading to indiscriminate recycling of oversize products, significant energy waste when fine and coarse materials are mixed, and low resource recovery efficiency. This application's embodiments use the comprehensive fractal dimension D as the core basis to dynamically determine the key control parameters for crushing and screening.

[0041] In the specific implementation process, the hyperspectral imager, laser 3D scanner, and scanning electron microscope system first acquire raw data simultaneously or sequentially, and all data are transmitted to the central processing unit via an industrial network. Subsequently, integrated image processing and analysis software is used to preprocess the data (including noise reduction, registration, enhancement, and segmentation), and corresponding fractal calculation algorithms are run (the core of which is box counting, which involves selecting a series of decreasing cover box side lengths, calculating the minimum number of boxes required to cover the target shape, and performing linear fitting in a double logarithmic coordinate system, the slope of which is the desired fractal dimension), thus obtaining the desired fractal dimension. , , The data processing system then calls the preset weighting coefficients (α, β, γ) and automatically calculates the comprehensive fractal dimension D of the current batch of ore according to the aforementioned weighting formula. This completes the entire calculation process from raw multi-scale feature data to a set of fractal dimensions (including multiple sub-dimensions and a comprehensive dimension) with clear physical meaning and process guidance value, laying the data foundation for intelligent adaptive control of subsequent processes.

[0042] After obtaining multiple sub-fractal dimensions and the comprehensive fractal dimension, the central control module does not directly execute the final control parameters on the crushing and screening equipment. Instead, it first uses the comprehensive fractal dimension D as the ore structural feature input to generate the initial control parameter range for crushing and screening operations, in order to avoid control instability caused by fluctuations in ore feed, changes in moisture content, and on-site disturbances.

[0043] The screening parameters include the working size of the discharge port of the crushing equipment and the screening particle size cutting point of the screening equipment. The control command is output as an execution signal through the signal conversion module, which drives the hydraulic adjustment mechanism of the crushing equipment to adjust the discharge port opening, and at the same time drives the vibration adjustment unit of the screening equipment to adjust the screening boundary position, so that the ore material entering the equipment completes actual crushing and grading according to the set parameters. Generally speaking, the higher the D value of the ore, the more complex its structure, the more developed the fractures, or the more entangled the interlocking relationships. Under the action of mechanical external force, it is often easier to break along the weak surface. Therefore, a smaller discharge port width can be used and is suitable to obtain a finer initial crushed product while ensuring processing capacity. Conversely, for dense and hard ores with a low D value, a larger discharge port width is used to avoid equipment overload and control energy consumption.

[0044] In practical implementation, the central control module pre-stores a table of discharge port adjustment parameter ranges corresponding to different fractal dimension intervals. When the comprehensive fractal dimension falls within a preset interval, the control module automatically reads the target range of the discharge port for the corresponding interval and generates an adjustment command to drive the crushing equipment actuator to adjust the discharge port size. This mapping relationship can be implemented in practice through a linear function: setting the initial discharge port width. The (mm) mapping function is:

[0045] in, and These are the maximum and minimum discharge openings, respectively. and This represents the expected range boundary of D. This is the proportionality coefficient. When D exceeds the preset range, the boundary value is taken. For example, when the comprehensive fractal dimension is in the range of 2.3 to 2.5, the control module controls the discharge port adjustment mechanism to set the discharge port width within the range of 180 mm to 250 mm.

[0046] After initial parameter loading is completed, the crushing equipment enters the trial operation phase. Raw ore enters the crusher at the set feed rate and is crushed at the initial discharge opening. The crushed product enters the screening unit via a conveying device. Simultaneously, the online particle size monitoring module collects and statistically analyzes the real-time particle size distribution of the crushed material, obtaining key evaluation indicators such as the proportion of oversized particles, the proportion of over-crushed particles, and the average particle size in the current undersize product.

[0047] The central control module compares the above online detection results with the preset target particle size distribution range. When the proportion of particles larger than the upper limit of the target in the undersize product exceeds the first threshold, it is determined that the current crushing intensity is insufficient. The control module sends a fine-tuning command to the hydraulic adjustment mechanism of the crushing equipment to gradually reduce the width of the discharge port according to the preset step size W. When the proportion of over-crushed particles exceeds the second threshold, it is determined that the crushing intensity is too high. The control module triggers a reverse adjustment command to gradually increase the width of the discharge port, thereby achieving a dynamic balance between processing capacity, particle size control and energy consumption.

[0048] For setting the particle size cut-off point, the control module also generates control signals based on the preset grading control parameter table, and adopts a dynamic adjustment strategy guided by the fractal dimension D. When the comprehensive fractal dimension is less than the first threshold, the control module outputs the first screening boundary parameter to the screening equipment; when the comprehensive fractal dimension is greater than or equal to the threshold, it outputs the second screening boundary parameter, thereby driving the screening equipment to adjust the screen vibration amplitude or the equivalent opening of the screen holes, so as to realize the physical diversion processing of materials in different particle size ranges.

[0049] The particle size classification cutoff point S (mm) can be expressed as a piecewise function: like If the value is less than 2.4, then S = S1 (e.g., 15mm). like If ≥ 2.4, then S = S2 (e.g., 12mm).

[0050] Among them, S1 and S2 can be finely adjusted by linear interpolation based on the D value to ensure that the coarse-grained level (cutting point -100mm) meets the optimal particle size requirements for photoelectric sorting.

[0051] The particle size cutoff point S (the critical size, in mm, that distinguishes between "coarse" and "fine" particles) is also dynamically set based on the D value. For example, when the D value is below a certain threshold (e.g., 2.4 mm), the cutoff point S can be set to 12-15 mm; when the D value is above this threshold, the cutoff point S is set to 8-12 mm. The core purpose of this design is to ensure that the particle size range of the final separated "coarse-grained product" perfectly matches the optimal feed requirements of the subsequent photoelectric sorting unit.

[0052] After parameter fine-tuning, when the online monitoring indicators meet the target range requirements for multiple consecutive sampling cycles, the central control module determines that the crushing and screening system has entered a stable operating state and locks the current discharge port size and screening cut point as the steady-state control parameters for this batch of ore. At this time, the system switches to a continuous stable operation mode, while maintaining low-frequency monitoring to cope with changes in operating conditions caused by changes in ore properties.

[0053] The system employs a multi-stage control mechanism involving fractal prediction, initial setting, online detection, and closed-loop correction. This enables the crushing and screening processes to adapt to changes in ore structural characteristics, achieving stable particle size distribution control. It avoids problems such as over-grinding, excessive cyclic load, and energy waste that exist in traditional fixed-parameter processes, providing stable and controllable feed conditions for subsequent photoelectric separation and fine-particle grinding processes.

[0054] In practice, based on the real-time calculated comprehensive fractal dimension D, the discharge port width W and screening cut point S corresponding to the current batch of ore are determined through the aforementioned mapping model. These parameters are then sent as instructions to the corresponding crushers and screening machines. The crushers adjust their hydraulic systems or mechanical devices to precisely adjust the discharge port to the target size W, after which the raw ore is fed in and crushed. The crushed mixed product is then conveyed to screening equipment with a fractal structure screen surface. This equipment, based on the received instructions, adjusts its screen aperture size or screening behavior to the target cut point S for screening. After screening, two products are obtained: one is "undersize product", which is material with a particle size less than or equal to the cutting point S. Its particle size meets the preset requirements and enters the next stage as a qualified product; the other is "oversize product", which is material with a particle size greater than the cutting point S. It is considered to not meet the current particle size requirements and is automatically returned to the crusher feed port by the conveying equipment for further crushing. This forms a closed loop until all materials are crushed to below the target size, thereby ensuring that the particle size composition of the "undersize product" is stable and meets the design expectations.

[0055] Subsequently, the obtained "undersize product" with a concentrated particle size distribution undergoes particle size classification to separate it into two independent processing paths. This is achieved using another classification device (such as another tension screen or air classifier), with the classification boundary being the dynamically set cutting point S mentioned earlier. After classification, two products are produced: one is a "coarse-grained product," with a particle size range between the cutting point S (e.g., 10 mm) and its upper limit (e.g., 100 mm); the other is a "fine-grained product," with a particle size smaller than the cutting point S. This classification step is crucial, as it achieves "separate processing" of the material, conveying only the coarser particles suitable for rapid photoelectric separation (coarse-grained product) to the photoelectric separation unit, while the fine particles unsuitable for photoelectric separation are directly guided to the subsequent grinding-magnetoelectric separation process, thus optimizing the process flow and rationally allocating energy consumption at the source.

[0056] After completing the crushing and screening operations and obtaining a coarse-grained product with a stable particle size distribution, the central control module uses multiple sub-fractal dimensions as input values ​​for the sensing strategy of the photoelectric sorting unit. This is used to generate the initial operating parameters of the photoelectric sensing system and gradually optimizes the sorting conditions through a dynamic calibration mechanism, thereby improving the identification stability and sorting accuracy under complex ore conditions. Traditional photoelectric sorting technology suffers from low identification accuracy and poor adaptability in complex ore environments. The color, shape, or composition of different ores may vary, leading to a high sorting error rate. Simultaneously, the fixed parameters of the photoelectric system cannot be dynamically adjusted according to ore characteristics, resulting in the failure to recover some coarse concentrate in a timely manner and the presence of useful minerals in the tailings. This application's embodiment uses the comprehensive fractal dimension D as the core basis to dynamically determine the key control parameters for crushing and screening.

[0057] The photoelectric sensing parameters mainly refer to the key variables driving the "vision" and "judgment" functions of the sorting system. Specifically, these include the energy value (E) of the X-ray source used for perspective detection and the frequency value (F) of the three-dimensional contour scanning used for surface morphology capture. The intelligent logic behind their dynamic determination lies in matching the "detection capability" of the sensing system with the physical characteristics of the ore. Specifically, the energy value of the X-ray source is mainly adjusted based on the spectral fractal dimension (Dh). The level of the spectral fractal dimension Dh directly reflects the complexity of the spatial interweaving and encapsulation of useful minerals and gangue minerals within the ore. When the Dh value is high, it indicates a complex distribution and tight interweaving of mineral components, requiring higher-energy X-rays (typically in the range of 120-160 keV, adjusted in 5 keV increments) to achieve effective penetration and clear imaging, thereby accurately distinguishing internal components; conversely, the energy can be appropriately reduced to save energy consumption. The frequency value of the three-dimensional contour scanning is mainly adjusted based on the morphological fractal dimension (Dm). The fractal dimension Dm of morphology characterizes the roughness and irregularity of the ore particle surface. A high Dm value indicates an extremely uneven particle surface, well-developed cracks, or highly irregular shape, significantly increasing the amount of 3D point cloud data acquired per frame at the same image resolution. To avoid overloading the data processing system and ensure the recognition accuracy of each frame, the scanning frequency needs to be appropriately reduced (typically adjusted within the range of 80-120 frames / second). Conversely, for particles with smooth surfaces and regular shapes (low Dm value), the scanning frequency can be increased to accelerate processing. The quantitative relationship upon which the parameter adjustment depends was established through prior calibration experiments on standard samples covering ores with different characteristics.

[0058] In practical implementation, the working energy parameters of the X-ray source are set based on a pre-stored parameter relationship table. This table is established based on historical ore sample experimental data and is used to select corresponding energy parameters within different fractal characteristic ranges. The X-ray source energy in the high-speed photoelectric sensing module... (keV) Pre-stored control parameters:

[0059] in Based on the basic energy (e.g., 140 keV). For adjustment coefficients, This refers to the average spectral fractal dimension in a standard sample library established based on historically representative ore samples. A high fractal dimension (indicating complex composition) necessitates a moderate increase in energy to enhance penetration.

[0060] Scan frequency F (frames / second) pre-stored control parameters:

[0061] in For the maximum frequency, The minimum fractal dimension of the topography in the standard sample library. This is an adjustment coefficient. The rougher the surface (... (The larger the resolution, the lower the scanning frequency can be to balance the data processing load at the same resolution).

[0062] The central control module calls the photoelectric sorting parameter configuration table in the memory based on the sub-fractal dimension information, generates the corresponding X-ray source working energy control command and three-dimensional contour scanning frequency control command, and sends them to the photoelectric sorting execution unit through the communication interface, so that the photoelectric sorting device can operate under the set working conditions and perform scanning, identification and spraying separation operations on the ore particles passing through the conveyor belt.

[0063] Coarse-grained products are uniformly spread into a single layer by a material distribution homogenizer and then sequentially passed through a high-speed photoelectric sensing module composed of a high-speed linear array camera, an X-ray detection unit, and a 3D laser contour scanner. In this module, the material undergoes two simultaneous scanning modes: X-ray transmission scanning at optimized energy E to acquire its internal density and composition information; and 3D contour scanning at optimized frequency F to acquire its precise external shape, volume, and surface texture information. This multi-source information is synchronously transmitted to an image processing and intelligent recognition unit for fusion analysis. This unit integrates a mineral recognition algorithm capable of comprehensively judging the properties of each material particle. The image processing and recognition module performs fusion analysis on the collected data, calculating the recognition confidence index and the estimated false positive rate under the current operating conditions. The recognition confidence index characterizes the stability of the system's results in distinguishing between useful minerals and gangue minerals. When the confidence level of multiple consecutive frames of recognition results falls below a preset threshold, the central control module determines that the current sensing parameters do not adequately match the physical properties of the ore, triggering a dynamic adjustment mechanism for the sensing parameters.

[0064] In the X-ray sensing channel, the central control module fine-tunes the X-ray source energy in stages based on the relative deviation of the compositional fractal dimension Dh. When the compositional fractal dimension is higher than the historical average, the control module gradually increases the X-ray energy in preset steps to enhance the penetration and contrast capabilities of complex embedded structures; when the compositional fractal dimension is lower than the baseline value, the energy is gradually reduced to decrease system energy consumption and reduce redundant radiation. In the 3D contour scanning channel, the central control module progressively adjusts the scanning frequency based on the changes in the morphology fractal dimension Dm and the real-time point cloud data density. When the morphology fractal dimension is high and the point cloud data density exceeds the processing threshold, the system automatically reduces the scanning frequency to avoid data congestion and ensure single-frame recognition accuracy; when the particle surface morphology is relatively regular, the scanning frequency is increased to improve processing throughput.

[0065] The above parameter adjustment process employs a multi-round iterative strategy. After each round of adjustment, the system re-enters a short-time sampling window to re-evaluate the recognition stability. When the system's detection results for multiple consecutive rounds consistently meet the recognition confidence threshold and the false positive rate remains consistently below the set upper limit, the central control module determines that the photoelectric sorting system has entered a stable operating state and locks the current X-ray energy and scanning frequency as the steady-state sorting parameters for this batch of ore.

[0066] During steady-state operation, the identification results are sent in real time to the actuator of the array-type separator. This separator typically consists of a series of high-speed, precise electromagnetic spray valves or pneumatic nozzles. When particles identified as "useful minerals" (i.e., coarse concentrate) reach the corresponding spray valve, the control system triggers a very short-duration (usually less than 1 millisecond) compressed air pulse (pressure adjustable within the range of 0.4-0.7 MPa) to blow them off their original falling trajectory and into the concentrate collection tank; while particles identified as "gangue" (i.e., coarse tailings) remain undisturbed and fall into the tailings collection tank along their natural trajectory. Through this process, efficient and precise separation of materials is achieved at the coarse stage. The coarse concentrate obtained from the separation is an aggregate of useful minerals that has been liberated or has a high enrichment level, which can be used directly as a product or enter subsequent refining; while the coarse tailings, mainly waste rock, are sent to the next stage to be combined with fine-grained products for further processing.

[0067] During steady-state operation, the system maintains a low-frequency online monitoring mechanism. When a drift in ore characteristics is detected, leading to a decrease in recognition confidence, the system automatically re-enters the parameter fine-tuning process, thereby achieving an adaptive response of the photoelectric sorting process to changes in ore properties.

[0068] By employing a multi-stage control strategy involving fractal feature guidance, initial parameter setting, online confidence assessment, dynamic calibration, and steady-state sorting, the photoelectric sorting system no longer relies on fixed parameters for operation. This effectively solves the problems of unstable identification accuracy and poor adaptability of traditional photoelectric sorting under complex ore conditions, while significantly improving the recovery rate of coarse concentrate and reducing the load on subsequent grinding operations.

[0069] After photoelectric separation of coarse-grained products and obtaining coarse-grained concentrate and tailings, the remaining fine-grained products and coarse-grained tailings from photoelectric separation need to be combined. The core problems of traditional fine-grained grinding and separation processes are: fixed grinding parameters cannot adapt to differences in ore microstructure, leading to over-grinding or under-grinding, affecting subsequent separation accuracy; high energy consumption in material mixing, with the combined processing of fine-grained products and coarse-grained tailings increasing energy consumption and reducing separation efficiency; and fixed separation parameters, where the field strength and voltage of magnetic and electrostatic separation cannot be dynamically optimized according to the ore's physical properties, resulting in low fine-grained concentrate recovery and the easy retention of valuable minerals in the tailings. This application's embodiments rely on sub-fractal dimension (Du) and comprehensive fractal dimension (D) to achieve intelligent adaptive control of fine-grained grinding and composite separation.

[0070] For this mixture, subsequent grinding and separation operations are also optimized using intelligent parameters based on fractal characteristics. First, the determination of grinding parameters mainly relies on the overall fractal dimension (D) and the microstructure fractal dimension (Du). Among them, the overall fractal dimension D reflects the overall structural complexity and hardness tendency of the material, while the microstructure fractal dimension Du accurately characterizes the irregularity and dissociation difficulty of the boundaries between mineral particles.

[0071] Specifically, the central control module, based on the ore liberation difficulty level corresponding to the comprehensive fractal dimension and the microstructure fractal dimension, retrieves the target ranges for mill speed, media filling rate, feed rate, and classification cutting particle size from the grinding parameter database, and selects the median value of each parameter range as the initial operating parameters of the grinding system. After the initial parameter loading is completed, the mixture enters the grinding-classification closed-loop system. The online particle size detection module and the circulating load monitoring module synchronously collect the particle size distribution information and return sand status data of the grinding product and transmit them to the central control module in real time.

[0072] The merged material is fed into the fractal grinding system. The central control module compares the current particle size distribution of the ground product with the target particle size control range. When the product particle size is detected to be too coarse, the control module gradually increases the mill speed or media filling rate according to the preset adjustment step size, and simultaneously lowers the grading and cutting particle size. When the product is detected to be too fine or the proportion of over-grinding increases, the mill load is adjusted in the opposite direction and the grading and cutting particle size is appropriately relaxed, thereby forming a dynamic balance between grinding efficiency and energy consumption. After each round of parameter adjustment, the system enters a stable sampling cycle to re-evaluate the particle size distribution of the ground product. When the particle size index stably meets the target range requirements for multiple consecutive sampling cycles, the central control module determines that the grinding system has entered a steady-state operation state and locks the current grinding condition parameters as the steady-state grinding parameters for this batch of ore. The system operates in a closed loop until the product fineness stably reaches the predetermined target (e.g., the content of -0.074 mm is controlled between 50% and 75%). During steady-state grinding operation, the central control module continuously monitors the grinding load variation trend using both the comprehensive fractal dimension and the microstructural fractal dimension. When a drift in ore liberation difficulty is detected, the system synchronously corrects the upper limit of mill power input and the feeding rhythm to avoid abnormal load fluctuations or inefficient operating ranges, thereby achieving coordinated control of fractal structure characteristics and on-site operating conditions. After this grinding and classification process, valuable minerals and gangue in the material achieve more complete individual liberation.

[0073] Subsequently, the qualified fine particles after grinding and classification are subjected to dry magnetic-electric composite separation. This separation process dynamically optimizes the separation conditions based on intelligent prediction of material composition, requiring no water throughout the process and avoiding the water consumption and pollution problems of wet separation. The central control module uses the spectral fractal dimension, morphological fractal dimension, and microstructure fractal dimension as the control inputs for magnetic-electric separation. It queries the corresponding target ranges of magnetic field strength and electric separation voltage in the preset parameter configuration library and selects the median value of the range as the initial operating parameters of the magnetic-electric separation system. Specifically, it includes two series-connected optimized separation stages: the first is dry high-intensity magnetic separation, where the separation field strength (B) is not fixed but is based on the magnetic mineral content (Fe) in the material predicted by fractal characteristics (combined with spectral data). predicted The method is dynamically determined. Its basic principle is: when the predicted magnetic mineral content is high, it indicates that the amount of target minerals requiring magnetic recovery in the material is large or the occurrence state is complex, requiring an appropriate increase in the separation field strength (adjusted within the range of 1.2 to 1.8 Tesla) to ensure the recovery rate; conversely, the field strength can be reduced to save energy. Field strength adjustment can be achieved through a linear model:

[0074] in, Based on the fundamental field strength, This is the adjustment coefficient. Next is high-voltage electric separation, where the separation voltage (U) is based on the material conductivity index predicted by fractal characteristics (U). Dynamic determination. The electrical conductivity characteristics of materials are closely related to their mineral composition. When the predicted conductivity index is high, it indicates that the overall conductivity of the material is good or that the electrical properties of the target mineral and gangue are significantly different. Therefore, it is necessary to adjust the intensity of the high-voltage electric field (voltage adjusted within the range of 30 to 40 kV) to achieve the best sorting effect. The adjustment model is as follows:

[0075] in, Based on the voltage, This is the adjustment coefficient.

[0076] After initial parameter loading, fine-grained materials first enter the magnetic separation equipment. The online grade detection module performs real-time sampling and analysis of the magnetic concentrate and tailings streams to obtain magnetic mineral recovery rate and tailings residue indicators. The central control module compares the detection results with the target separation indicators. When the magnetic mineral loss rate is detected to be higher than the preset threshold, the control module gradually increases the magnetic field strength within the magnetic separation field strength range according to the preset step size. When the concentration of impurities increases or the separation selectivity decreases, the magnetic field strength is gradually reduced or the feed rate is adjusted to maintain the balance between recovery rate and concentrate grade.

[0077] After the magnetic separation is stabilized, the tailings enter the electrostatic separation unit. The central control module fine-tunes the electrostatic separation voltage in stages based on the online conductivity separation effect detection results. When insufficient separation of effective minerals is detected, the electrostatic separation voltage is gradually increased to enhance the effect of electrical differences; when an increased proportion of non-target minerals is detected as misseparated, the electrostatic separation voltage is appropriately reduced to suppress interference separation phenomena.

[0078] The above-mentioned magnetic separation and electro-electro-separation parameter adjustment process adopts a multi-round sampling, evaluation, and correction strategy. When the fine-grained concentrate grade, recovery rate, and tailings loss rate consistently meet the set process indicators for multiple consecutive detection cycles, the central control module determines that the magnetic-electric composite separation system has entered a steady-state operation state and locks the current magnetic field strength and electro-electro-separation voltage as the steady-state separation parameters for this batch of ore.

[0079] During steady-state operation, the system maintains a low-frequency online monitoring mechanism. When changes in ore properties or deviations in separation performance indicators are detected, the system automatically re-enters the parameter fine-tuning process, thereby achieving an adaptive response of the grinding and magnetic-electric composite separation process to changes in ore structural characteristics.

[0080] In an optional implementation, the method in this embodiment can run on an intelligent integrated control platform. This platform, as the system core, interacts with each unit in real time via industrial Ethernet. It serves as the central hub for the intelligent algorithms of the aforementioned units, integrating local optimizations such as fractal feature perception, personalized crushing, photoelectric sorting, and fine particle grinding into a global optimization decision, forming a complete "perception-decision-execution" intelligent closed loop. The platform's deep control module calls database algorithms to maximize comprehensive sorting efficiency, achieving intelligent monitoring and optimization throughout the entire process. The system processes data from 200-500 monitoring points in real time, including operating parameters of each device (current, pressure, valve opening, etc.), material flow rate, grade online analysis data, and fractal feature data calculated by each unit, constructing a global state vector. The deep control module calls the database to optimize decisions: a. State space: It includes all relevant sensor data, fractal model data, and production targets (such as concentrate grade, recovery rate, and energy consumption).

[0081] b. Motion space: It is a set of joint adjustment instructions for adjustable parameters of each unit (such as discharge port, cutting point, X-ray energy, magnetic field strength, mill feed rate, etc.).

[0082] c. Reward function:

[0083] in to These are the weighting coefficients.

[0084] d. Algorithm Implementation: Employ either the Deep Deterministic Policy Gradient (DDPG) algorithm or the Proximal Policy Optimization (PPO) algorithm based on the Actor-Critic framework. The agent, based on the current state... Output actions through the Actor network It acts on the environment (actual production system); the Critic network evaluates the state-action value Q( , Through continuous interactive training, the policy network converges to a policy that maximizes long-term cumulative rewards. .

[0085] With a response time of less than 100ms and a control accuracy of over 95%, it can achieve 72 hours of continuous unattended operation.

[0086] Through the aforementioned mechanism, the intelligent integrated control platform achieves global coordination and dynamic optimization of the aforementioned unit solutions: The platform first receives and integrates local data from each unit, including fractal characteristics, equipment parameters, and online grades, to build a comprehensive global situational awareness; it breaks through the local optimization of each unit based on fixed rules, and through the control module calling the database, it dynamically weighs and makes collaborative decisions among multiple mutually restrictive objectives such as crushing particle size, sorting accuracy, grinding fineness, and energy consumption to find the global optimal solution for the system; finally, through continuous interaction with the production environment and automatic adjustment of strategies based on reward feedback, it forms a closed-loop autonomous and evolutionary capability, enabling the system to adapt to fluctuations in ore properties and continuously optimize.

[0087] Figure 2 This invention provides a schematic diagram of the structure of an ore screening apparatus according to an embodiment of the present application. The apparatus includes: The fractal feature modeling module is used to calculate multiple sub-fractal dimensions and the overall fractal dimension based on the multi-scale fractal feature data of the original ore before the crushing operation. The intelligent crushing and screening module is used to crush the raw ore based on the screening parameters determined by the comprehensive fractal dimension to obtain the undersize product that meets the requirements; and to classify the undersize product by particle size to obtain coarse-grained product and fine-grained product. The photoelectric sorting module is used to perform photoelectric sorting on the coarse-grained product based on the photoelectric sensing parameters determined by the sub-fractal dimension, to obtain coarse-grained concentrate and coarse-grained tailings; The grinding and magnetic-electric composite separation module is communicatively coupled with the photoelectric separation module and the particle size classification module. It is used to perform grinding and classification processing on the fine-grained product and the coarse-grained tailings based on the grinding parameters determined by the sub-fractal dimension and the comprehensive fractal dimension, and to perform dry magnetic-electric composite separation on the product after grinding and classification processing to obtain fine-grained concentrate and final tailings.

[0088] The calculation of multiple sub-fractal dimensions and a comprehensive fractal dimension based on the multi-scale fractal characteristic data of the original ore before crushing operations includes: The fractal dimension of the spectral structure is calculated based on the mineral phase distribution based on hyperspectral image recognition, the fractal dimension of the morphological fracture is calculated based on the digital elevation model generated from laser three-dimensional point cloud data, and the fractal dimension of the microstructure is calculated based on the particle boundary extracted from scanning electron microscope images. The comprehensive fractal dimension is calculated based on the spectral fractal dimension and the first weighting coefficient, the morphological crack fractal dimension and the second weighting coefficient, and the microstructure fractal dimension and the third weighting coefficient; wherein the sum of the first weighting coefficient, the second weighting coefficient and the third weighting coefficient is 1.

[0089] The screening parameters include the working size of the discharge port and the screening particle size cut point; the device determines the screening parameters in the following ways: Based on the comprehensive fractal dimension, the discharge port width of the crushing equipment is determined through a first mapping relationship; wherein, the first mapping relationship is a linear relationship in which the discharge port width is negatively correlated with the comprehensive fractal dimension; Based on the comprehensive fractal dimension, the particle size cutting point of the screening equipment is determined through the second mapping relationship.

[0090] If the oversize product does not meet the requirements, it is crushed again until the undersize product is obtained.

[0091] The photoelectric sensing parameters include the energy value of the X-ray source and the frequency value of the three-dimensional contour scan. The device determines the photoelectric sensing parameters in the following manner: The energy value of the X-ray source is determined based on the spectral fractal dimension in the sub-fractal dimension, wherein the spectral fractal dimension is positively correlated with the energy value; The frequency value of the three-dimensional contour scan is determined based on the shape fractal dimension in the sub-fractal dimension, wherein the shape fractal dimension and the frequency value are negatively correlated.

[0092] The grinding parameters include the Bonder's index correction factor and the target grinding fineness, and the apparatus determines the grinding parameters in the following manner: Based on the combined fractal dimension and micro-fractal dimension of the fine-grained product and the coarse-grained tailings, the Bondon index correction coefficient and the target grinding fineness are determined by calling the database strategy network through the control module of process parameter calibration.

[0093] The dry magnetic-electric composite separation of the product after grinding and classification includes: Based on the fractal dimension of the spectrum, the fractal dimension of the morphological fracture, and the fractal dimension of the microstructure as inputs, and combined with the spectral data of the hyperspectral image, the magnetic mineral content and electrical conductivity index are predicted through a pre-set regression analysis model. The separation field strength of the high-intensity magnetic separator is determined based on the content of the magnetic minerals. The sorting voltage of the high-voltage electric separator is determined based on the conductivity index. Dry magnetoelectric composite sorting is performed based on the determined sorting field strength and sorting voltage.

[0094] like Figure 3 As shown, this application provides an electronic device for performing the ore screening method described in this application. The device includes a memory, a processor, a bus, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the ore screening method described above.

[0095] Specifically, the aforementioned memory and processor can be general-purpose memory and processor, without any specific limitations. When the processor runs the computer program stored in the memory, it can execute the aforementioned ore screening method.

[0096] Corresponding to the ore screening method in this application, this application embodiment also provides a computer storage medium storing a computer program, which is executed by a processor to perform the steps of the above-described ore screening method.

[0097] Specifically, the storage medium can be a general-purpose storage medium, such as a removable disk or hard disk. When the computer program on the storage medium is run, it can perform the above-mentioned ore screening method.

[0098] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. The system embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interface; the indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0099] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0100] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0101] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0102] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0103] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A method for screening ores, characterized in that, The method includes: Based on the multi-scale fractal characteristic data of the original ore before the crushing operation, multiple sub-fractal dimensions and the comprehensive fractal dimension were calculated. The raw ore is crushed based on the screening parameters determined by the comprehensive fractal dimension to obtain the undersize product that meets the requirements, and the undersize product is then classified by particle size to obtain coarse-grained product and fine-grained product. Based on the photoelectric sensing parameters determined by the subfractal dimension, the coarse-grained product is photoelectrically separated to obtain coarse-grained concentrate and coarse-grained tailings. Based on the grinding parameters determined by the sub-fractal dimension and the comprehensive fractal dimension, the fine-grained product and the coarse-grained tailings are ground and classified, and the ground and classified product is then subjected to dry magnetic-electric composite separation to obtain fine-grained concentrate and final tailings.

2. The method according to claim 1, characterized in that, The calculation of multiple sub-fractal dimensions and a comprehensive fractal dimension based on the multi-scale fractal characteristic data of the original ore before crushing operations includes: The fractal dimension of the spectral structure is calculated based on the mineral phase distribution based on hyperspectral image recognition, the fractal dimension of the morphological fracture is calculated based on the digital elevation model generated from laser three-dimensional point cloud data, and the fractal dimension of the microstructure is calculated based on the particle boundary extracted from scanning electron microscope images. The comprehensive fractal dimension is calculated based on the spectral fractal dimension and the first weighting coefficient, the morphological crack fractal dimension and the second weighting coefficient, and the microstructure fractal dimension and the third weighting coefficient; wherein the sum of the first weighting coefficient, the second weighting coefficient and the third weighting coefficient is 1.

3. The method according to claim 1, characterized in that, The screening parameters include the working size of the discharge port and the screening particle size cut point; the method determines the screening parameters in the following ways: Based on the comprehensive fractal dimension, the discharge port width of the crushing equipment is determined through a first mapping relationship; wherein, the first mapping relationship is a linear relationship in which the discharge port width is negatively correlated with the comprehensive fractal dimension; Based on the comprehensive fractal dimension, the particle size cutting point of the screening equipment is determined through the second mapping relationship.

4. The method according to claim 1, characterized in that, The screening parameters determined by the comprehensive fractal dimension are used to crush the raw ore, including: If the oversize product does not meet the requirements, it is crushed again until the undersize product is obtained.

5. The method according to claim 1, characterized in that, The photoelectric sensing parameters include the energy value of the X-ray source and the frequency value of the three-dimensional contour scan. The method determines the photoelectric sensing parameters in the following manner: The energy value of the X-ray source is determined based on the spectral fractal dimension in the sub-fractal dimension, wherein the spectral fractal dimension is positively correlated with the energy value; The frequency value of the three-dimensional contour scan is determined based on the shape fractal dimension in the sub-fractal dimension, wherein the shape fractal dimension and the frequency value are negatively correlated.

6. The method according to claim 1, characterized in that, The grinding parameters include the Bonder's index correction factor and the target grinding fineness, and the method determines the grinding parameters in the following manner: Based on the combined fractal dimension and micro-fractal dimension of the fine-grained product and the coarse-grained tailings, the Bondon index correction coefficient and the target grinding fineness are determined by calling the database strategy network through the control module of the process parameter calibration process.

7. The method according to claim 1, characterized in that, The dry magnetic-electric composite separation of the product after grinding and classification includes: Based on the fractal dimension of the spectrum, the fractal dimension of the morphological fractures, and the fractal dimension of the microstructure as inputs, and combined with the spectral data of the hyperspectral image, the content of magnetic minerals and the electrical conductivity index are predicted by calling the magnetoelectric sorting parameter configuration table. The separation field strength of the high-intensity magnetic separator is determined based on the content of the magnetic minerals. The sorting voltage of the high-voltage electric separator is determined based on the conductivity index. Dry magnetoelectric composite sorting is carried out based on the determined sorting field strength and sorting voltage.

8. An apparatus for screening ore, characterized in that, The device includes: The fractal feature modeling module is used to calculate multiple sub-fractal dimensions and the overall fractal dimension based on the multi-scale fractal feature data of the original ore before the crushing operation. The intelligent crushing and screening module is used to crush the raw ore based on the screening parameters determined by the comprehensive fractal dimension to obtain the undersize product that meets the requirements; and to classify the undersize product by particle size to obtain coarse-grained product and fine-grained product. The photoelectric sorting module is used to perform photoelectric sorting on the coarse-grained product based on the photoelectric sensing parameters determined by the sub-fractal dimension, to obtain coarse-grained concentrate and coarse-grained tailings; The grinding and magnetic-electric composite separation module is communicatively coupled with the photoelectric separation module and the particle size classification module. It is used to perform grinding and classification processing on the fine-grained product and the coarse-grained tailings based on the grinding parameters determined by the sub-fractal dimension and the comprehensive fractal dimension, and to perform dry magnetic-electric composite separation on the product after grinding and classification processing to obtain fine-grained concentrate and final tailings.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine instructions that the processor executes. When the electronic device is running, the processor communicates with the memory via the bus. When the machine instructions are executed by the processor, they perform the steps of the ore screening method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The computer storage medium stores a computer program that, when executed by a processor, performs the steps of the ore screening method as described in any one of claims 1 to 7.

Images