An ore crushing control method, system, electronic device and readable storage medium

Abstract

The application discloses an ore crushing control method and system, electronic equipment and a readable storage medium. The method inputs a multi-scale feature map fused into a feature mapping layer to obtain a multi-scale fusion feature tensor; inputs the multi-scale fusion feature tensor into a candidate region generation layer to determine a target candidate region; inputs the target candidate region into an instance segmentation layer to determine a target instance segmentation mask; calculates a particle equivalent diameter based on the target instance segmentation mask; determines a medium crushing discharge particle size and a fine crushing discharge particle size under a P80 particle size index according to the particle equivalent diameter; calculates a material instantaneous conveying flow based on a continuous time sequence point cloud volume integration model; performs time integration on the material instantaneous conveying flow to obtain a fine crushing discharge throughput; and performs hierarchical control on ore crushing through the medium crushing discharge particle size, the fine crushing discharge particle size, the fine crushing discharge throughput and a fine crushing machine running power. The application can improve the ore crushing efficiency and the qualified rate of the crushing products.

Description

Technical Field

[0001] This application relates to the field of mineral processing technology, and in particular to an ore crushing control method, system, electronic device and readable storage medium. Background Technology

[0002] Ore crushing is a key process in the mineral processing / benefiting industry. Existing related technologies mostly adopt a multi-stage open-circuit or closed-circuit crushing process of coarse crushing, medium crushing, fine crushing and screening. The main equipment for coarse crushing is gyratory crusher and jaw crusher, while medium and fine crushing generally uses cone crusher, impact crusher and hammer crusher. Basic control is achieved by manually setting the crusher discharge port (CSS) and adjusting the feed rate. Some production lines introduce PLC systems or simple sensors to monitor equipment operating parameters.

[0003] However, existing technologies have significant shortcomings: there is a lack of real-time online collaborative control between finished material particle size and output; particle size detection relies heavily on manual sampling and screening, which is inefficient and data is lagging; output adjustment and particle size control are disconnected; parameter adjustment relies on manual experience and cannot dynamically respond to variables such as fluctuations in raw ore properties and equipment wear, which can easily lead to problems such as excessively coarse, excessively fine, or crushed particles or unstable output, resulting in a significant reduction in the production efficiency of the crusher and an increase in energy consumption. Summary of the Invention

[0004] This application aims to propose an ore crushing control method, system, electronic device, and readable storage medium, which can improve ore crushing efficiency, increase the qualification rate of ore crushed products, reduce manual monitoring costs, and achieve precise control of output.

[0005] In a first aspect, embodiments of this application provide a method for controlling ore crushing, the method comprising: The three-dimensional spatial point cloud data of the acquired ore crushing material is preprocessed to obtain preprocessed three-dimensional spatial point cloud data. Feature extraction is performed on the preprocessed 3D spatial point cloud data to obtain multi-scale feature maps; the multi-scale feature maps are then fused and input into the feature mapping layer to obtain a multi-scale fused feature tensor. The multi-scale fusion feature tensor is input into the candidate region generation layer to determine the target candidate region; the target candidate region is input into the instance segmentation layer to determine the target instance segmentation mask. Based on the target instance segmentation mask, calculate the equivalent particle size; based on the equivalent particle size, determine the medium crushing discharge particle size and fine crushing discharge particle size under the P80 particle size index. Based on the continuous time-series point cloud volume integral model, the instantaneous material conveying flow rate is calculated; the instantaneous material conveying flow rate is integrated over time to obtain the fine crushing discharge throughput. The crushing of ore is graded and controlled by the discharge particle size of the medium crusher, the discharge particle size of the fine crusher, the discharge throughput of the fine crusher, and the operating power of the fine crusher.

[0006] In some implementations, inputting the target candidate region into the instance segmentation layer to determine the target instance segmentation mask includes: The target candidate region is subjected to feature pooling to obtain a local feature tensor; Multiply the local feature tensor and the mask convolution weight matrix to obtain the multiplication result; Add the multiplication result and the bias term together to obtain the sum; The summation result is input into the activation function to obtain the initial segmentation mask; The initial segmentation mask and the local feature tensor are input into the boundary attention enhancement module for boundary enhancement to obtain the target instance segmentation mask.

[0007] In some implementations, calculating the instantaneous material transport flow rate based on the continuous time-series point cloud volume integral model includes: Obtain material bulk density, conveyor belt speed, effective detection width of conveyor belt, and detection area length; Construct a continuous temporal point cloud volume integral model that includes the height distribution function of the transport section; The material bulk density, the conveyor belt speed, the effective detection width of the conveyor belt, and the detection area length are input into the continuous time-series point cloud volume integral model to calculate the instantaneous material conveying flow rate.

[0008] In some embodiments, the step of classifying and controlling ore crushing by means of the intermediate crushing discharge particle size, the fine crushing discharge particle size, the fine crushing discharge throughput, and the operating power of the fine crusher includes: When the operating power of the fine crusher is less than a preset fine crushing power control threshold, including: If the particle size of the medium crushing output is less than the medium crushing output particle size control threshold multiplied by a percentage coefficient, the feed rate of the fine crushing is increased to restore the operating power of the fine crusher to the first preset reasonable range, and the discharge port of the medium crushing is adjusted according to the relationship model between the output particle size and the discharge port. If the particle size of the medium crushing output is within the second preset reasonable range, then increase the fine crushing feed rate to restore the operating power of the fine crusher to the first preset reasonable range.

[0009] In some embodiments, the step of classifying and controlling ore crushing by means of the intermediate crushing discharge particle size, the fine crushing discharge particle size, the fine crushing discharge throughput, and the operating power of the fine crusher includes: If the operating power of the fine crusher is greater than the preset fine crushing power control threshold, the fine crushing feed rate is reduced to adjust the operating power of the fine crusher to a first preset reasonable range; when the operating power of the fine crusher returns to the first preset reasonable range, a graded judgment of the fine crushing output is performed, including: If the fine crushing output is within the third preset reasonable range and the fine crushing output particle size is within the fourth preset reasonable range, it is judged as a normal working condition; If the fine crushing output is within the third preset reasonable range, and the fine crushing output particle size is greater than the fine crushing output particle size control threshold multiplied by a percentage coefficient, then the fine crushing discharge port is adjusted according to the relationship model between the output particle size and the discharge port. If the fine crushing output is less than the fine crushing output control threshold multiplied by a percentage factor, and the medium crushing output particle size is within the second preset reasonable range, then it is judged as a normal working condition. If the fine crushing output is less than the fine crushing output control threshold multiplied by a percentage factor, and the medium crushing output particle size is greater than the medium crushing output particle size control threshold multiplied by a percentage factor, then the medium crushing discharge port is adjusted according to the relationship model between the output particle size and the discharge port.

[0010] In some embodiments, the step of classifying and controlling ore crushing by means of the intermediate crushing discharge particle size, the fine crushing discharge particle size, the fine crushing discharge throughput, and the operating power of the fine crusher includes: When the operating power of the fine crusher is within a first preset reasonable range, including: If the particle size of the fine crushed output is within the fourth preset reasonable range, it is judged as a normal working condition; If the fine crushing discharge particle size is greater than the fine crushing discharge particle size control threshold multiplied by a percentage factor, the fine crushing discharge outlet is adjusted according to the relationship model between discharge particle size and discharge outlet. If the fine crushing discharge particle size is less than the fine crushing discharge particle size control threshold multiplied by a percentage factor, and the medium crushing discharge particle size is within the second preset reasonable range, then the fine crushing discharge outlet is adjusted according to the relationship model between the discharge particle size and the discharge outlet. If the fine crushing discharge particle size is less than the fine crushing discharge particle size control threshold multiplied by a percentage factor, and the medium crushing discharge particle size is less than the medium crushing discharge particle size control threshold multiplied by a percentage factor, then the medium crushing discharge outlet is adjusted according to the relationship model between discharge particle size and discharge outlet.

[0011] In some embodiments, the relationship model between the discharge particle size and the discharge port is a discharge port adjustment prediction model that includes a discharge port target adjustment equation. The discharge port adjustment prediction model can be a linear model or a nonlinear model; the discharge port target adjustment equation is: ; in, This indicates the optimal adjustment amount for the ore discharge outlet. This represents the principal correction factor for particle size deviation. This indicates the actual monitored discharge particle size P80 value. This indicates the preset standard value for the output particle size. This indicates the liner wear correction factor. Indicates the wear coefficient of the liner. This represents the operating condition correction constant.

[0012] Secondly, embodiments of this application also provide an ore crushing control system, the system comprising: The data preprocessing module is used to preprocess the acquired three-dimensional spatial point cloud data of ore crushing materials to obtain preprocessed three-dimensional spatial point cloud data. The feature extraction and fusion module is used to extract features from the preprocessed 3D spatial point cloud data to obtain multi-scale feature maps; the multi-scale feature maps are then fused and input into the feature mapping layer to obtain a multi-scale fused feature tensor. The target instance segmentation module is used to input the multi-scale fused feature tensor into the candidate region generation layer to determine the target candidate region; and input the target candidate region into the instance segmentation layer to determine the target instance segmentation mask. The first data calculation module is used to calculate the equivalent particle size based on the target instance segmentation mask; and to determine the medium crushing discharge particle size and fine crushing discharge particle size under the P80 particle size index according to the equivalent particle size. The second data calculation module is used to calculate the instantaneous material conveying flow rate based on the continuous time-series point cloud volume integral model; and to perform time integration on the instantaneous material conveying flow rate to obtain the fine crushed material discharge throughput. The ore crushing control module is used to classify and control ore crushing by measuring the particle size of the medium crushing output, the particle size of the fine crushing output, the throughput of the fine crushing output, and the operating power of the fine crusher.

[0013] Thirdly, embodiments of this application also provide an electronic device, including at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, the instructions being executed by the at least one control processor to enable the at least one control processor to perform an ore crushing control method as described above.

[0014] Fourthly, embodiments of this application also provide a computer-readable storage medium storing computer-executable instructions for causing a computer to execute an ore crushing control method as described above.

[0015] Compared with the prior art, this application has the following beneficial effects: This application preprocesses the acquired 3D spatial point cloud data of crushed ore to obtain preprocessed 3D spatial point cloud data; extracts features from the preprocessed 3D spatial point cloud data to obtain multi-scale feature maps; fuses the multi-scale feature maps and inputs them into a feature mapping layer to obtain a multi-scale fused feature tensor; inputs the multi-scale fused feature tensor into a candidate region generation layer to determine target candidate regions; inputs the target candidate regions into an instance segmentation layer to determine a target instance segmentation mask; calculates the equivalent particle size based on the target instance segmentation mask; determines the medium crushing discharge particle size and fine crushing discharge particle size under the P80 particle size index based on the equivalent particle size; calculates the instantaneous material conveying flow rate based on a continuous time-series point cloud volume integral model; integrates the instantaneous material conveying flow rate over time to obtain the fine crushing discharge throughput; and performs graded control of ore crushing by using the medium crushing discharge particle size, fine crushing discharge particle size, fine crushing discharge throughput, and the operating power of the fine crusher.

[0016] Thus, by preprocessing the three-dimensional point cloud data, computational redundancy can be reduced and the efficiency of real-time model inference can be improved. The trained model can calculate the particle size of medium crushing, fine crushing, and fine crushing throughput in real time without human intervention, thereby improving the efficiency of ore crushing control, reducing energy consumption, and lowering the cost of manual monitoring. By comprehensively considering the particle size of medium crushing, fine crushing, fine crushing throughput, and the operating power of the fine crusher, the ore crushing can be graded and controlled, thereby improving the qualification rate of ore crushing products and achieving precise production control. Attached Figure Description

[0017] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a schematic flowchart of an embodiment of the ore crushing control method provided in this application; Figure 2 This is a schematic diagram of the particle size and yield online monitoring system architecture in the preferred embodiment of the ore crushing control method provided in this application; Figure 3 This is a schematic diagram of the particle size and yield optimization control logic in the best embodiment of the ore crushing control method provided in this application; Figure 4 This is a schematic diagram of the relationship between P80 and the wear of the crusher discharge port in the preferred embodiment of the ore crushing control method provided in this application; Figure 5 This is a schematic diagram of an embodiment of the ore crushing control system provided in this application; Figure 6This is a schematic diagram of the structure of an embodiment of the electronic device provided in this application. Detailed Implementation

[0018] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0019] In the description of this application, the use of terms such as "first," "second," etc., is for the purpose of distinguishing technical features only and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.

[0020] In the description of this application, it should be understood that the orientation descriptions, such as up, down, etc., are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0021] In the description of this application, it should be noted that, unless otherwise explicitly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.

[0022] To address the problems in related technologies, such as excessively coarse or fine particles, crushing, unstable output, significantly reduced crusher efficiency, and increased energy consumption, this application proposes an ore crushing control method, system, electronic equipment, and readable storage medium.

[0023] Reference Figure 1 This application provides a schematic flowchart of an ore crushing control method. This ore crushing control method is applied to an electronic device, which may be a server or a mobile terminal, etc. Figure 1 As shown, the ore crushing control method may include the following steps S101 to S106.

[0024] Step S101: Preprocess the obtained three-dimensional spatial point cloud data of the crushed ore material to obtain preprocessed three-dimensional spatial point cloud data.

[0025] In this step, a 3D vision acquisition module is used to perform high-density, high-precision, all-around 3D scanning and acquisition of the crushed material on the conveyor belt in real time, accurately obtaining high-density, high-fidelity original 3D spatial point cloud data of the material. The acquired original 3D spatial point cloud data is then normalized, filtered for noise reduction, and spatially registered to obtain preprocessed 3D spatial point cloud data.

[0026] Step S102: Extract features from the preprocessed 3D spatial point cloud data to obtain multi-scale feature maps; fuse the multi-scale feature maps and input them into the feature mapping layer to obtain a multi-scale fused feature tensor.

[0027] In this step, the preprocessed 3D point cloud data is processed using the following formula to extract features, resulting in a multi-scale feature map: ; in, Indicates the first Layer output feature map, Represents a non-linear activation function. Represents the convolution kernel weight matrix. Indicates the first Layer output feature map, This represents the bias parameter.

[0028] The multi-scale feature maps are fused and then input into the feature mapping layer using the following formula to obtain the multi-scale fused feature tensor: ; in, Indicates the first Scale-fused feature maps This indicates the convolution fusion operation. Indicates an upsampling operation. Indicates the first Scale-fused feature map.

[0029] Step S103: Input the multi-scale fusion feature tensor into the candidate region generation layer to determine the target candidate region; input the target candidate region into the instance segmentation layer to determine the target instance segmentation mask.

[0030] In this step, the multi-scale fusion feature tensor is input into the candidate region generation layer, and a region proposal network is used to generate candidate detection regions for material particles. The candidate box coordinates, confidence scores, and target probability distribution are output through sliding window convolution. Overlapping and redundant candidate regions are eliminated based on the non-maximum suppression algorithm, thereby improving the uniqueness and accuracy of particle detection and finally obtaining the target candidate region.

[0031] The target candidate region is pooled to obtain a local feature tensor; the local feature tensor is multiplied by the mask convolution weight matrix to obtain the multiplication result; the multiplication result is added to the bias term to obtain the sum result; the sum result is input into the activation function to obtain the initial segmentation mask; the initial segmentation mask and the local feature tensor are input into the boundary attention enhancement module for boundary enhancement to obtain the target instance segmentation mask.

[0032] Step S104: Calculate the equivalent particle size based on the target instance segmentation mask; determine the medium crushing discharge particle size and fine crushing discharge particle size under the P80 particle size index according to the equivalent particle size.

[0033] In this step, the 3D point cloud contours of each particle are ellipsoidally fitted based on the target instance segmentation mask, and the equivalent particle size is calculated as follows: ; in, Indicates the equivalent particle size. , and These represent the lengths of the three-dimensional principal axes of the particle.

[0034] By statistically analyzing the equivalent particle size distribution of all particles, a cumulative particle size distribution curve can be generated in real time, and the particle size of medium crushing and fine crushing under the P80 particle size index can be further calculated.

[0035] Step S105: Calculate the instantaneous material conveying flow rate based on the continuous time-series point cloud volume integral model; perform time integration on the instantaneous material conveying flow rate to obtain the fine crushing discharge throughput.

[0036] In this step, the material bulk density, conveyor belt speed, effective detection width of the conveyor belt, and detection area length are obtained; a continuous time-series point cloud volume integral model incorporating the conveyor cross-section height distribution function is constructed; the material bulk density, conveyor belt speed, effective detection width of the conveyor belt, and detection area length are input into the continuous time-series point cloud volume integral model to calculate the instantaneous material conveying flow rate: ; in, Indicates the instantaneous flow rate of materials. Indicates the bulk density of the material. Indicates the speed of the conveyor belt. Indicates the effective detection width of the conveyor belt. Indicates the length of the detection area. This represents the height distribution function of the conveyor cross section. Represents the integral variable. The corresponding integration interval is 0 to , The corresponding integration interval is 0 to .

[0037] Then, the instantaneous material conveying flow rate is integrated over time to obtain the fine crushing discharge throughput.

[0038] Step S106: The ore crushing is graded and controlled by the discharge particle size of the medium crusher, the discharge particle size of the fine crusher, the discharge throughput of the fine crusher, and the operating power of the fine crusher.

[0039] In this step, when the operating power of the fine crusher is less than the preset fine crushing power control threshold, the following steps are included: If the particle size of the medium crushing output is less than the medium crushing output particle size control threshold multiplied by a percentage coefficient, the fine crushing feed rate is increased to restore the operating power of the fine crusher to the first preset reasonable range, and the medium crushing discharge port is adjusted according to the relationship model between the output particle size and the discharge port; if the medium crushing output particle size is within the second preset reasonable range, the fine crushing feed rate is increased to restore the operating power of the fine crusher to the first preset reasonable range.

[0040] If the operating power of the fine crusher exceeds the preset fine crushing power control threshold, the fine crushing feed rate is reduced to adjust the operating power of the fine crusher to a first preset reasonable range; once the operating power of the fine crusher returns to the first preset reasonable range, a grading judgment of the fine crushing output is performed, including: If the fine crushing output is within the third preset reasonable range and the fine crushing output particle size is within the fourth preset reasonable range, it is judged as a normal operating condition. If the fine crushing output is within the third preset reasonable range and the fine crushing output particle size is greater than the fine crushing output particle size control threshold multiplied by a percentage coefficient, the fine crushing discharge outlet is adjusted according to the relationship model between the output particle size and the discharge outlet. If the fine crushing output is less than the fine crushing output throughput control threshold multiplied by a percentage coefficient and the medium crushing output particle size is within the second preset reasonable range, it is judged as a normal operating condition. If the fine crushing output is less than the fine crushing output throughput control threshold multiplied by a percentage coefficient and the medium crushing output particle size is greater than the medium crushing output particle size control threshold multiplied by a percentage coefficient, the medium crushing discharge outlet is adjusted according to the relationship model between the output particle size and the discharge outlet.

[0041] When the operating power of the fine crusher is within a first preset reasonable range, including: If the fine crushing output particle size is within the fourth preset reasonable range, it is judged as a normal operating condition; if the fine crushing output particle size is greater than the fine crushing output particle size control threshold multiplied by a percentage coefficient, the fine crushing discharge opening is adjusted according to the relationship model between output particle size and discharge opening; if the fine crushing output particle size is less than the fine crushing output particle size control threshold multiplied by a percentage coefficient, and the medium crushing output particle size is within the second preset reasonable range, the fine crushing discharge opening is adjusted according to the relationship model between output particle size and discharge opening; if the fine crushing output particle size is less than the fine crushing output particle size control threshold multiplied by a percentage coefficient, and the medium crushing output particle size is less than the medium crushing output particle size control threshold multiplied by a percentage coefficient, the medium crushing discharge opening is adjusted according to the relationship model between output particle size and discharge opening. The relationship model between output particle size and discharge opening is a discharge opening adjustment prediction model including a discharge opening target adjustment equation. The discharge opening adjustment prediction model can be a linear model or a nonlinear model; the discharge opening target adjustment equation is: ; in, This indicates the optimal adjustment amount for the ore discharge outlet. This represents the principal correction factor for particle size deviation. This indicates the actual monitored discharge particle size P80 value. This indicates the preset standard value for the output particle size. This indicates the liner wear correction factor. Indicates the wear coefficient of the liner. This represents the operating condition correction constant.

[0042] The linear model described above can be a model known to those skilled in the art, such as a multiple linear regression model, and this embodiment does not impose any specific limitations on it.

[0043] The aforementioned nonlinear model can be a model known to those skilled in the art, such as a BP neural network model, and this embodiment does not impose any specific limitations on it.

[0044] To facilitate understanding by those skilled in the art, a set of preferred embodiments is provided below: This embodiment provides an AI-powered intelligent ore crushing control method that can improve ore crushing efficiency and finished product qualification rate. It aims to solve problems such as uneven particle size of finished product, large output fluctuations, and lagging monitoring of key indicators in existing fine ore crushing production lines, thereby improving the intelligent management and control level of the production line and ultimately increasing crushing efficiency and finished product qualification rate. This method precisely installs visual monitoring equipment at key locations such as the feed inlet, crushing stage, and discharge outlet of the production line to achieve real-time and continuous monitoring of material gradation and flow rate, ensuring the accuracy and timeliness of the monitoring data. Combined with a pre-set intelligent crushing decision control system, the collected monitoring data is rapidly analyzed, calculated, and processed to accurately determine the current operating status of the production line and provide timely and targeted adjustment suggestions. This achieves precise control of finished product quality stability and output, improves crusher production efficiency, reduces manual monitoring costs, and enhances both production efficiency and economic benefits.

[0045] The particle size and yield online monitoring system in this embodiment is adapted to the standard process flow of medium and fine crushing of ore. The system is linked with production equipment to achieve real-time acquisition, transmission, and analysis of key parameters. It also has the capability to connect to a distributed control system (DCS) for centralized management and control of production data. The architecture of the particle size and yield online monitoring system is as follows: Figure 2 As shown, it specifically includes: The ore medium and fine crushing production line adapted in this embodiment uses a continuous belt conveyor to complete the entire material feeding process. The ore to be crushed is uniformly and stably conveyed by the belt conveyor to the pre-screening equipment for raw material screening. The undersize material from the pre-screening is directly conveyed to the finished product stockpile, while the oversize material is conveyed to the medium crusher for medium crushing (pre-screening can be performed depending on the properties of the raw ore; if pre-screening is not required, the raw material is directly conveyed to the medium crusher for processing). The output of the medium crusher is conveyed to the inspection screening equipment for secondary screening. The undersize material from the inspection screening is conveyed to the finished product stockpile, while the oversize material is conveyed to the fine crusher for fine crushing. The output of the fine crusher is again conveyed to the inspection screening equipment for screening. The undersize material enters the finished product stockpile, while the oversize material returns to the fine crusher for cyclic crushing, forming a closed-loop crushing process.

[0046] The core monitoring data of the online monitoring system includes: the operating power of the medium crusher. Operating power of fine crusher , medium crushing discharge particle size Fine crushing output particle size and the amount of fine crushed material discharged .

[0047] The particle size and flow rate data for medium and fine crushing are acquired through laser particle size online monitoring equipment. This equipment is specifically designed for the high-dust, high-vibration, and continuous production environments of mining operations, featuring a highly integrated structure and strong on-site adaptability. It consists of a 3D vision acquisition module, a light source module, a mounting bracket module, and an edge computing module. The light source module provides light to the vision acquisition device within the 3D vision acquisition module. The mounting bracket module supports the vision acquisition device. The high-precision 3D vision image acquisition module is the core data sensing unit at the front end of the equipment. Operating based on the laser triangulation principle, it can perform high-density, high-precision, all-around 3D scanning and acquisition of crushed materials on the conveyor belt in real time, accurately obtaining high-density, high-fidelity original 3D spatial point cloud data of the material. This point cloud data can completely reconstruct the three-dimensional outline, particle size distribution, material flow density, and material flow motion status of the entire material flow, ensuring the integrity and accuracy of the original material data from a hardware perception perspective. It avoids the technical shortcomings of two-dimensional visual inspection, such as missing material thickness dimensions, misjudgment of particle overlap, and large deviations in particle size calculation. The high-precision original three-dimensional spatial point cloud data of the material can be synchronously transmitted to an industrial-grade edge computing processing module in real time. The industrial-grade edge computing module pre-deploys a dedicated deep learning intelligent recognition and computation model that has been trained and adapted to the characteristics of the crushed material.

[0048] This dedicated deep learning intelligent recognition computing model is built upon the theory of deep learning algorithms for convolutional neural network feature extraction and instance segmentation. It employs a hierarchical network architecture consisting of a "point cloud preprocessing layer—multi-scale feature extraction layer—candidate region generation layer—instance segmentation layer—granular parameter parsing layer—traffic parsing layer." Specifically: 1. The point cloud preprocessing layer is used to normalize, filter, reduce noise, and spatially register the raw 3D spatial point cloud data acquired in real time by the 3D vision acquisition module. Specifically, a statistical outlier filtering algorithm is used to remove dust floating noise and abnormal discrete points caused by equipment vibration, and a voxel mesh downsampling method is combined to equalize the point cloud density, so as to reduce computational redundancy and improve the real-time inference efficiency of the model.

[0049] The statistical outlier filtering algorithm described above can identify and remove outliers that deviate from the mean by analyzing the distance distribution between each point and its neighbors, thereby cleaning up noisy data.

[0050] The above-mentioned voxel mesh downsampling method can achieve point cloud downsampling by dividing the point cloud into a 3D voxel mesh and replacing all points in it with a representative point of each voxel.

[0051] It should be noted that the spatial registration in this embodiment can be carried out using point cloud registration methods known to those skilled in the art, such as the Iterative Closest Point (ICP) algorithm. This embodiment does not provide a specific description or limitation of this method.

[0052] 2. The multi-scale feature extraction layer adopts an encoder-decoder convolutional neural network structure. Through multiple convolutional kernels, it abstracts the point cloud projection feature map layer by layer to extract the edge contour features, texture gradient features, spatial geometric features, and particle stacking topology features of the material particles. Addressing the characteristics of the working conditions, such as large particle size range, irregular shape, and severe particle adhesion, a multi-scale feature pyramid fusion mechanism is introduced to achieve simultaneous perception of the overall contour of large particles and the detailed boundaries of small particles.

[0053] The convolutional feature extraction process in the multi-scale feature extraction layer satisfies the following expression: ; in, Indicates the first Layer output feature map; Indicates the first The layer outputs a feature map, which is used as input to the first layer. layer; The kernel weight matrix; These are bias parameters; This is a non-linear activation function. This allows for the extraction of multi-scale feature maps. The convolution kernel weight matrix and bias parameters can be obtained during training.

[0054] To address the complex working conditions characterized by large particle size range, irregular shape, and severe particle adhesion and occlusion in ore, a feature pyramid fusion mechanism is introduced in the multi-scale feature extraction layer to perform lateral fusion of feature maps at different scales. The fusion expression is as follows: ; in, For the first Scale-fused feature maps; Indicates the first Scale-fused feature maps; This is an upsampling operation; This is a convolution fusion operation.

[0055] Through the aforementioned multi-scale fusion mechanism, simultaneous perception of the overall contour of large-sized particles and the local boundary details of small-sized particles is achieved, improving the feature representation capability of multi-level particles. The feature map after multi-scale fusion (i.e., the fused feature map corresponding to the last scale) is input to the feature mapping layer for unified scale mapping and feature standardization processing, generating the input feature tensor of the candidate region generation layer (i.e., the multi-scale fused feature tensor).

[0056] 3. The candidate region generation layer receives a multi-scale fused feature tensor. The candidate region generation layer uses a region proposal network to generate candidate detection regions for material particles. It outputs the candidate box coordinates, confidence scores, and target probability distribution through a sliding window convolution method. It also eliminates overlapping and redundant candidate regions based on a non-maximum suppression algorithm, thereby improving the uniqueness and accuracy of particle detection and finally obtaining the target candidate region.

[0057] The aforementioned region proposal network can be a neural network structure for object detection tasks, which can quickly and effectively extract candidate regions from images.

[0058] 4. The instance segmentation layer performs initial pixel-level mask prediction on granular targets based on target candidate regions. Its core output expression is: ; in, For the first The initial segmentation mask for each particle instance; For the first Local feature tensors of each target candidate region after feature pooling; The mask convolution weight matrix; For bias terms; is the activation function. The mask convolution weight matrix and bias term can be obtained during training.

[0059] The initial segmentation mask enables coarse-grained separation of the main particle target region. To further improve particle contour segmentation accuracy under complex occlusion, particle adhesion, and boundary blurring conditions, the instance segmentation layer, after generating the initial segmentation mask, inputs the local pooling features (i.e., local feature tensors) and the initial segmentation mask together into the boundary attention enhancement module to perform feature weighting enhancement on the particle boundary region. Specifically: The expression for calculating the boundary attention weights is: ; in, The boundary attention feature response matrix; These are the query matrix, key matrix, and value matrix, respectively. For feature dimensions.

[0060] Based on the boundary attention response matrix, the boundary region of the initial segmentation mask is refined and corrected. The optimization expression is as follows: ; in, The final instance segmentation mask after boundary enhancement (i.e., the target instance segmentation mask). Enhance the weighting coefficients for the boundary.

[0061] By using a boundary attention enhancement mechanism to perform boundary compensation and contour refinement on the initial segmentation results, the edge response characteristics of the particle overlapping area can be strengthened, the particle boundary closure integrity and instance separation accuracy can be improved, thereby accurately distinguishing broken particles that are in contact with each other, partially occluded and stacked and adhered as independent particle instances, and realizing single-particle-level accurate identification in complex material flow environments.

[0062] 5. The grain size parameter analysis layer performs ellipsoid fitting and equivalent grain size calculation on the 3D point cloud contour of each particle based on the instance segmentation result (i.e., the target instance segmentation mask). The expression for calculating the equivalent grain size is as follows: ; in, The equivalent particle size; These represent the lengths of the three-dimensional principal axes of the particles.

[0063] By statistically analyzing the equivalent particle size distribution of all particles, a cumulative particle size distribution curve can be generated in real time, and further calculations can be performed to obtain P80, P50, and the mass percentage parameters of each particle size. In this embodiment, the P80 particle size index is selected.

[0064] It should be noted that the calculation of P80, P50 and the mass percentage parameters of each particle size in this embodiment can be performed using techniques known to those skilled in the art, and this embodiment will not describe them in detail.

[0065] 6. The flow analysis layer calculates the instantaneous material conveying flow rate based on the continuous time-series point cloud volume integral model. Its calculation expression is: ; in, For instantaneous material delivery flow rate; Bulk density of the material; The speed of the conveyor belt; The height distribution function of the conveying section; The effective detection width of the conveyor belt; This represents the length of the detection area.

[0066] The cumulative transmission flow rate is obtained through time integration calculation: ; Ultimately, the deep learning instance segmentation model can output the particle size distribution parameters, P80 particle size index, instantaneous conveying flow rate and cumulative conveying flow rate (which can be the calculated fine crushing discharge throughput) of medium and fine crushed materials in real time, and transmit them synchronously to the intelligent control module of the discharge port, providing real-time operating condition input parameters for the dynamic closed-loop adjustment of the discharge port.

[0067] The particle size and conveying flow data collected by the laser particle size online monitoring equipment are synchronously and in real time transmitted to the data acquisition and control system along with the operating power data of the medium crusher and the fine crusher. The system then performs integrated real-time analysis and calculation on multi-dimensional data such as power, particle size, and discharge throughput, providing data support for subsequent regulation.

[0068] The data analysis and control logic in this embodiment takes the operating power status of the fine crusher as the core entry point, and combines multi-dimensional online monitoring data such as fine crushing output particle size, fine crushing output throughput, and medium crushing output particle size to construct a closed-loop control process with graded judgment and progressive steps. The particle size and output optimization control logic is as follows: Figure 3 As shown. The data analysis and control logic includes the following: First, determine the power of the fine crusher, which is divided into the following three operating conditions: 1. The fine crushing power (i.e., the operating power of the fine crusher) is too low. .

[0069] If the particle size of the medium crushing output (The gradation of the intermediate crusher discharge, with 80% of the passing particle size (i.e., reaching P80), is too small, meaning... ( If the particle size control threshold for intermediate crushing is set, it indicates that the crushing intensity of intermediate crushing is too high and the feed load of fine crushing is insufficient. In this case, the feed rate of fine crushing needs to be increased to restore the fine crushing power to the first preset reasonable range. To ensure both production capacity and crushing efficiency, the discharge opening of the intermediate crusher is adjusted based on the relationship model between the output particle size and the discharge opening. For fine power, This is the threshold for fine power control.

[0070] If the particle size of the medium crushing output is within the second preset reasonable range ( Simply increase the fine crushing feed rate to quickly bring the fine crushing power back to the first preset reasonable range. This ensures the overall production capacity and crushing efficiency of the production line.

[0071] It should be noted that the gradation in this embodiment can be the distribution of particles of different sizes in the medium crushing output, which can be determined through sieve analysis.

[0072] 2. The power consumption for fine crushing is too high ( ).

[0073] If the fine crushing power is detected to be too high, in order to avoid damage to the equipment due to overload, first directly reduce the fine crushing feed rate and adjust the operating power of the fine crusher to the first preset reasonable range. After power recovery, the process enters the stage of determining the grade of fine crushed material output, and targeted control measures are taken based on different monitoring results: If the fine crushing output is within the third preset reasonable range ( Furthermore, the particle size of the finely crushed output is within the reasonable range of the fourth preset ( ),in, To ensure that 80% of the passing particle size is within the range of fine crushed output, If the particle size control threshold for fine crushed output is set, it is considered a normal operating condition.

[0074] If the fine crushing output is within the third preset reasonable range ( ),in This indicates that the throughput control threshold for fine crushing output is too high, but the particle size of the fine crushing output is too large. If the result is negative, it is determined to be wear of the fine crushing liner, and the fine crushing discharge port needs to be adjusted according to the relationship model between the discharge particle size and the discharge port.

[0075] If the output of fine crushed material is too small ( However, the particle size of the intermediate crushing output is within the second preset reasonable range. If the condition is normal, then it is considered to be in normal working condition.

[0076] If the output of fine crushed material is too small ( Furthermore, the particle size of the output material from the intermediate crushing process is relatively large. If the result is negative, it is determined that the fine crushing feed particle size is too large, resulting in a decrease in output. The medium crushing discharge port needs to be adjusted according to the relationship model between the medium crushing discharge particle size and the discharge port.

[0077] 3. The power of fine crushing is within the first preset reasonable range ( ).

[0078] If the monitored fine crushed output particle size is within the fourth preset reasonable range ( The production line is determined to be in normal production status, and the existing operating parameters are kept unchanged.

[0079] If the fine crushed output particle size is too large ( If the result is negative, it is determined that the discharge opening has become larger due to liner wear, and the fine crushing discharge opening needs to be adjusted according to the relationship model between the fine crushing discharge particle size and the discharge opening.

[0080] If the fine crushed output particle size is too small ( And the particle size of the intermediate crushing output is within the second preset reasonable range. If the result is negative, it indicates that the fine crushing discharge opening is too small, and the fine crushing discharge opening needs to be adjusted according to the relationship model between the fine crushing output particle size and the discharge opening.

[0081] If the fine crushed output particle size is too small ( Furthermore, the particle size of the intermediate crushing output is also relatively small. If the result is too small, it is determined that the output of medium crushing is too small, which leads to the output of fine crushing being too small. The discharge port of medium crushing needs to be adjusted according to the relationship model between the particle size of medium crushing output and the discharge port.

[0082] The relationship model between discharge particle size and discharge outlet is a quantitative correlation mathematical model constructed through fitting a large amount of experimental data and iterative optimization by machine learning. Its core input parameter is "actual monitored discharge particle size", and its core output parameter is "optimal adjustment amount of discharge outlet". This ensures that the model can adapt to the control requirements under different working conditions and avoids control deviations caused by a single parameter correlation.

[0083] In the actual production process of crushing equipment, the liners wear continuously over time, causing the crusher's discharge opening to enlarge and consequently coarser the output particle size. By continuously collecting discharge particle size P80 and corresponding discharge opening data based on liner wear through online monitoring, a graph is plotted as follows: Figure 4 The curve showing the relationship between P80 and the wear of the crusher discharge port (CSS) reveals the following dynamic changes in liner wear and discharge particle size: Particle size change during wear stage: Between two adjustments of the discharge port, as the wear of the liner intensifies, the gap between the discharge ports continues to increase, and the output P80 (i.e., the particle size of 80% of the ore particles) gradually increases. This trend reflects the process of the product particle size becoming coarser with wear.

[0084] Particle size response to discharge opening adjustment: After each adjustment of the discharge opening gap, P80 will immediately show a significant drop, indicating that by narrowing the discharge opening, the output particle size has been pulled back to a finer level, proving that discharge opening adjustment is a direct and effective means of controlling product particle size.

[0085] The cumulative effect of irreversible wear: Due to the irreversible wear of the liner, the initial value of P80 after each adjustment of the discharge port will be higher than the initial value after the previous adjustment. This is a direct manifestation of the accumulation of liner wear, changes in cavity shape, and the continuous increase of the "reference gap" of the discharge port.

[0086] Traditional manual adjustment methods rely on operators who periodically monitor P80 data and manually adjust the discharge port based on experience. This approach fails to capture the dynamic patterns reflected in the curve, resulting in significant shortcomings: delayed adjustments lead to P80 exceeding the process limit for extended periods; the adjustment amount depends on experience and cannot match the baseline rise in each wear stage, causing P80 instability even after adjustment; and the particle size rise rate varies greatly across different wear stages, making general experience unsuitable for the entire operating cycle. In this embodiment, based on this large amount of data, a quantitative model establishes a dynamic mapping relationship between the discharge port gap and the output P80, and enables adaptive calibration of the model. The model can identify the particle size rise rate in the current wear stage and the baseline rise after each adjustment in real time, predicting the timing of discharge port adjustments and the target gap value in advance. This effectively solves the problems of low precision, slow response, and poor adaptability of manual adjustments, achieving intelligent, precise, and real-time control of discharge port adjustments, providing core technical support for the stable, efficient, and energy-saving operation of the crushing production line.

[0087] In the process of constructing the model of the relationship between discharge particle size and discharge opening, multiple sets of comparative experiments were conducted to cover the working conditions under different crushing intensities, different material types and different liner wear levels for medium and fine crushing. Data on discharge opening and discharge particle size under the corresponding working conditions were collected.

[0088] The collected raw data underwent preprocessing: First, an outlier removal algorithm was used to eliminate abnormal data caused by equipment failure, detection errors, and other factors during the experiment, ensuring data reliability. Then, data normalization was performed to eliminate the influence of different dimensions (discharge opening degree and discharge particle size) on model fitting, improving data comparability. Based on the preprocessed high-quality dataset (including discharge opening degree data and discharge particle size data), multiple linear regression or BP neural network models (i.e., discharge opening adjustment prediction models) were used for model fitting and training. Cross-validation was used to iteratively optimize the models to improve goodness of fit. The root mean square error and mean absolute error were used as evaluation indicators to screen the optimal model parameters. Finally, a dynamic correlation model between discharge particle size and discharge outlet was constructed, exhibiting high fitting accuracy, high generalization ability, and online adaptive updating capability. Its goodness of fit was... Not less than 0.95, prediction error controlled within Within.

[0089] This correlation model is pre-installed in an industrial-grade edge computing module, enabling real-time model access and high-speed computation without relying on cloud computing power, thus meeting the real-time control needs of industrial production sites. The core mathematical expression of the model (i.e., the target adjustment equation for the discharge outlet included in the discharge outlet adjustment prediction model) is optimized as follows: ; in, The optimal adjustment amount for the discharge opening (unit: mm) indicates that the discharge opening needs to be increased, while a negative value indicates that the discharge opening needs to be decreased. It is the main correction coefficient for particle size deviation, which is precisely calibrated according to the type of crushing section (medium crushing and fine crushing). Its core function is to quantify the influence weight of the deviation between the actual output particle size and the preset standard on the adjustment amount of the discharge port. The P80 value (unit: mm) of the actual output particle size is a core indicator reflecting the actual output particle size; it is obtained through the aforementioned laser particle size online monitoring equipment. The standard value for the output particle size (unit: mm) can be set flexibly according to the different grade requirements of the finished aggregate, adapting to diverse production scenarios; This is the liner wear correction coefficient, calibrated according to the degree of liner wear (new liner, moderate wear, and heavy wear), to compensate for the interference of liner wear on the correlation between discharge particle size and discharge port. The wear coefficient of the liner (dimensionless) ranges from 0 to 1, with a larger value for higher wear levels. The working condition correction constant (unit: mm) is calibrated according to the material type and crushing intensity to further improve the model's adaptability and calculation accuracy.

[0090] Compared with the prior art, the technical solution of this embodiment has the following advantages: This embodiment achieves multi-dimensional online monitoring and closed-loop control, improving the stability of finished product quality. By synchronously collecting key parameters such as the operating power of medium / fine crushing equipment, the particle size of medium / fine crushing output, and the throughput of fine crushing output, a hierarchical control logic is constructed with fine crushing power as the core and particle size and output as constraints. This logic can perceive fluctuations in finished product particle shape and output in real time, avoiding the lag of manual detection. By predicting the impact of medium crushing output particle size on fine crushing load, the fine crushing feed rate and discharge port gap are dynamically adjusted to avoid overload or no-load operation of the fine crusher, ensuring that the operating power of the fine crushing equipment remains stable within the high-efficiency range. This embodiment is based on a general medium and fine crushing process flow and can be adapted to various crushing equipment such as cone crushers and impact crushers. It can be quickly deployed and applied without large-scale modification of existing production lines, and is suitable for various ore crushing scenarios such as metal ores, non-metal ores, and sand and gravel aggregates, demonstrating good engineering practicality and economy.

[0091] Reference Figure 5 This application also provides an ore crushing control system, which includes: The data preprocessing module 100 is used to preprocess the acquired three-dimensional spatial point cloud data of ore crushing materials to obtain preprocessed three-dimensional spatial point cloud data. The feature extraction and fusion module 200 is used to extract features from the preprocessed 3D spatial point cloud data to obtain multi-scale feature maps; the multi-scale feature maps are then fused and input into the feature mapping layer to obtain a multi-scale fused feature tensor. The target instance segmentation module 300 is used to input the multi-scale fused feature tensor into the candidate region generation layer to determine the target candidate region; and input the target candidate region into the instance segmentation layer to determine the target instance segmentation mask. The first data calculation module 400 is used to calculate the equivalent particle size based on the target instance segmentation mask; and to determine the medium crushing discharge particle size and fine crushing discharge particle size under the P80 particle size index based on the equivalent particle size. The second data calculation module 500 is used to calculate the instantaneous material conveying flow rate based on the continuous time-series point cloud volume integral model; and to perform time integration on the instantaneous material conveying flow rate to obtain the fine crushing discharge throughput. The ore crushing control module 600 is used to classify and control ore crushing by measuring the discharge particle size of medium crushing, the discharge particle size of fine crushing, the discharge throughput of fine crushing, and the operating power of fine crusher.

[0092] It should be noted that since the ore crushing control system in this embodiment is based on the same inventive concept as the ore crushing control method described above, the corresponding content in the method embodiment is also applicable to this system embodiment, and will not be described in detail here.

[0093] Reference Figure 6 This application also provides an electronic device, which includes: At least one memory; At least one processor; At least one program; The program is stored in memory, and the processor executes at least one program to implement the ore crushing control method described above in this disclosure.

[0094] This electronic device can be any smart terminal, including mobile phones, tablets, personal digital assistants (PDAs), and in-vehicle computers.

[0095] The electronic devices according to embodiments of this application will now be described in detail.

[0096] The processor 1600 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this disclosure. The memory 1700 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1700 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1700 and called and executed by the processor 1600 using the ore crushing control method of the embodiments of this disclosure.

[0097] The input / output interface 1800 is used to implement information input and output. The communication interface 1900 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 2000 transmits information between various components of the device (e.g., processor 1600, memory 1700, input / output interface 1800, and communication interface 1900); The processor 1600, memory 1700, input / output interface 1800 and communication interface 1900 are connected to each other within the device via bus 2000.

[0098] This disclosure also provides a storage medium, which is a computer-readable storage medium storing computer-executable instructions for causing a computer to execute the above-described ore crushing control method.

[0099] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0100] The embodiments described in this disclosure are for the purpose of more clearly illustrating the technical solutions of this disclosure and do not constitute a limitation on the technical solutions provided by this disclosure. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by this disclosure are also applicable to similar technical problems.

[0101] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this disclosure, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0102] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0103] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0104] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application 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 the embodiments of this application 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 a 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.

[0105] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0106] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0107] 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; that is, 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 according to actual needs.

[0108] Furthermore, the functional units in the various embodiments of 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. The integrated unit can be implemented in hardware or as a software functional unit.

[0109] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it 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 all or part 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 multiple instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks. The embodiments of this application have been described in detail above with reference to the accompanying drawings, but this application is not limited to the above embodiments. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of this application.

[0110] The embodiments of this application have been described in detail above with reference to the accompanying drawings. However, this application is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of this application.

Claims

1. A method for controlling ore crushing, characterized in that, The method includes: The three-dimensional spatial point cloud data of the acquired ore crushing material is preprocessed to obtain preprocessed three-dimensional spatial point cloud data. Feature extraction is performed on the preprocessed 3D spatial point cloud data to obtain multi-scale feature maps; the multi-scale feature maps are then fused and input into the feature mapping layer to obtain a multi-scale fused feature tensor. The multi-scale fusion feature tensor is input into the candidate region generation layer to determine the target candidate region; the target candidate region is input into the instance segmentation layer to determine the target instance segmentation mask. Based on the target instance segmentation mask, calculate the equivalent particle size; based on the equivalent particle size, determine the medium crushing discharge particle size and fine crushing discharge particle size under the P80 particle size index. Based on the continuous time-series point cloud volume integral model, the instantaneous material conveying flow rate is calculated; the instantaneous material conveying flow rate is integrated over time to obtain the fine crushed material discharge throughput. The ore crushing process is graded and controlled by the medium crushing discharge particle size, the fine crushing discharge particle size, the fine crushing discharge throughput, and the operating power of the fine crusher, including: When the operating power of the fine crusher is less than a preset fine crushing power control threshold, including: If the particle size of the medium crushing output is less than the medium crushing output particle size control threshold multiplied by a percentage coefficient, the feed rate of the fine crushing is increased to restore the operating power of the fine crusher to the first preset reasonable range, and the discharge port of the medium crushing is adjusted according to the relationship model between the output particle size and the discharge port. If the particle size of the medium crushing output is within the second preset reasonable range, then increase the fine crushing feed rate to restore the operating power of the fine crusher to the first preset reasonable range; If the operating power of the fine crusher is greater than the preset fine crushing power control threshold, the fine crushing feed rate is reduced to adjust the operating power of the fine crusher to a first preset reasonable range; when the operating power of the fine crusher returns to the first preset reasonable range, a graded judgment of the fine crushing output is performed, including: If the fine crushing output is within the third preset reasonable range and the fine crushing output particle size is within the fourth preset reasonable range, it is judged as a normal working condition; If the fine crushing output is within the third preset reasonable range, and the fine crushing output particle size is greater than the fine crushing output particle size control threshold multiplied by a percentage coefficient, then the fine crushing discharge port is adjusted according to the relationship model between the output particle size and the discharge port. If the fine crushing output is less than the fine crushing output control threshold multiplied by a percentage factor, and the medium crushing output particle size is within the second preset reasonable range, then it is judged as a normal working condition. If the fine crushing output is less than the fine crushing output control threshold multiplied by a percentage factor, and the medium crushing output particle size is greater than the medium crushing output particle size control threshold multiplied by a percentage factor, then the medium crushing discharge port is adjusted according to the relationship model between the output particle size and the discharge port. When the operating power of the fine crusher is within a first preset reasonable range, including: If the particle size of the fine crushed output is within the fourth preset reasonable range, it is judged as a normal working condition; If the fine crushing discharge particle size is greater than the fine crushing discharge particle size control threshold multiplied by a percentage factor, the fine crushing discharge outlet is adjusted according to the relationship model between discharge particle size and discharge outlet. If the fine crushing discharge particle size is less than the fine crushing discharge particle size control threshold multiplied by a percentage factor, and the medium crushing discharge particle size is within the second preset reasonable range, then the fine crushing discharge outlet is adjusted according to the relationship model between the discharge particle size and the discharge outlet. If the fine crushing discharge particle size is less than the fine crushing discharge particle size control threshold multiplied by a percentage factor, and the medium crushing discharge particle size is less than the medium crushing discharge particle size control threshold multiplied by a percentage factor, then the medium crushing discharge outlet is adjusted according to the relationship model between discharge particle size and discharge outlet.

2. The ore crushing control method according to claim 1, characterized in that, The step of inputting the target candidate region into the instance segmentation layer to determine the target instance segmentation mask includes: The target candidate region is subjected to feature pooling to obtain a local feature tensor; Multiply the local feature tensor and the mask convolution weight matrix to obtain the multiplication result; Add the multiplication result and the bias term together to obtain the sum; The summation result is input into the activation function to obtain the initial segmentation mask; The initial segmentation mask and the local feature tensor are input into the boundary attention enhancement module for boundary enhancement to obtain the target instance segmentation mask.

3. The ore crushing control method according to claim 1, characterized in that, The calculation of instantaneous material transport flow rate based on the continuous time-series point cloud volume integral model includes: Obtain material bulk density, conveyor belt speed, effective detection width of conveyor belt, and detection area length; Construct a continuous temporal point cloud volume integral model that includes the height distribution function of the transport section; The material bulk density, the conveyor belt speed, the effective detection width of the conveyor belt, and the detection area length are input into the continuous time-series point cloud volume integral model to calculate the instantaneous material conveying flow rate.

4. The ore crushing control method according to claim 1, characterized in that, The relationship model between the discharge particle size and the discharge port is a discharge port adjustment prediction model that includes a discharge port target adjustment equation. The discharge port adjustment prediction model can be a linear or nonlinear model. The discharge port target adjustment equation is: ； in, This indicates the optimal adjustment amount for the ore discharge outlet. This represents the principal correction factor for particle size deviation. This indicates the actual monitored discharge particle size P80 value. This indicates the preset standard value for the output particle size. This indicates the liner wear correction factor. Indicates the wear coefficient of the liner. This represents the operating condition correction constant.

5. An ore crushing control system, characterized in that, The system includes: The data preprocessing module is used to preprocess the acquired three-dimensional spatial point cloud data of ore crushing materials to obtain preprocessed three-dimensional spatial point cloud data. The feature extraction and fusion module is used to extract features from the preprocessed 3D spatial point cloud data to obtain multi-scale feature maps; the multi-scale feature maps are then fused and input into the feature mapping layer to obtain a multi-scale fused feature tensor. The target instance segmentation module is used to input the multi-scale fused feature tensor into the candidate region generation layer to determine the target candidate region; and input the target candidate region into the instance segmentation layer to determine the target instance segmentation mask. The first data calculation module is used to calculate the equivalent particle size based on the target instance segmentation mask; and to determine the medium crushing discharge particle size and fine crushing discharge particle size under the P80 particle size index according to the equivalent particle size. The second data calculation module is used to calculate the instantaneous material conveying flow rate based on the continuous time-series point cloud volume integral model; and to perform time integration on the instantaneous material conveying flow rate to obtain the fine crushed material discharge throughput. The ore crushing control module is used to perform graded control of ore crushing based on the intermediate crushing discharge particle size, the fine crushing discharge particle size, the fine crushing discharge throughput, and the operating power of the fine crusher, including: When the operating power of the fine crusher is less than a preset fine crushing power control threshold, including: If the particle size of the medium crushing output is less than the medium crushing output particle size control threshold multiplied by a percentage coefficient, the feed rate of the fine crushing is increased to restore the operating power of the fine crusher to the first preset reasonable range, and the discharge port of the medium crushing is adjusted according to the relationship model between the output particle size and the discharge port. If the particle size of the medium crushing output is within the second preset reasonable range, then increase the fine crushing feed rate to restore the operating power of the fine crusher to the first preset reasonable range; If the operating power of the fine crusher is greater than the preset fine crushing power control threshold, the fine crushing feed rate is reduced to adjust the operating power of the fine crusher to a first preset reasonable range; when the operating power of the fine crusher returns to the first preset reasonable range, a graded judgment of the fine crushing output is performed, including: If the fine crushing output is within the third preset reasonable range and the fine crushing output particle size is within the fourth preset reasonable range, it is judged as a normal working condition; If the fine crushing output is within the third preset reasonable range, and the fine crushing output particle size is greater than the fine crushing output particle size control threshold multiplied by a percentage coefficient, then the fine crushing discharge port is adjusted according to the relationship model between the output particle size and the discharge port. If the fine crushing output is less than the fine crushing output control threshold multiplied by a percentage factor, and the medium crushing output particle size is within the second preset reasonable range, then it is judged as a normal working condition. If the fine crushing output is less than the fine crushing output control threshold multiplied by a percentage factor, and the medium crushing output particle size is greater than the medium crushing output particle size control threshold multiplied by a percentage factor, then the medium crushing discharge port is adjusted according to the relationship model between the output particle size and the discharge port. When the operating power of the fine crusher is within a first preset reasonable range, including: If the fine crushed output particle size is within the fourth preset reasonable range, it is judged as a normal working condition; If the fine crushing discharge particle size is greater than the fine crushing discharge particle size control threshold multiplied by a percentage factor, the fine crushing discharge outlet is adjusted according to the relationship model between discharge particle size and discharge outlet. If the fine crushing discharge particle size is less than the fine crushing discharge particle size control threshold multiplied by a percentage factor, and the medium crushing discharge particle size is within the second preset reasonable range, then the fine crushing discharge outlet is adjusted according to the relationship model between the discharge particle size and the discharge outlet. If the fine crushing discharge particle size is less than the fine crushing discharge particle size control threshold multiplied by a percentage factor, and the medium crushing discharge particle size is less than the medium crushing discharge particle size control threshold multiplied by a percentage factor, then the medium crushing discharge outlet is adjusted according to the relationship model between discharge particle size and discharge outlet.

6. An electronic device, characterized in that, It includes at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, which, when executed by the at least one control processor, enable the at least one control processor to perform the ore crushing control method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform the ore crushing control method as described in any one of claims 1 to 4.

Images