An optimized control method of a coal quality sampling device

Abstract

This invention relates to the field of coal quality sampling technology and discloses an optimized control method for a coal quality sampling device. The method involves detecting a vehicle carriage positioning signal, acquiring a set of structural feature parameters and a set of coal surface images of the target carriage based on a camera, generating a coal quality heterogeneity distribution map of the target carriage based on historical coal quality sampling datasets, generating a candidate sampling point location matrix based on the structural feature parameter set and the coal quality heterogeneity distribution map, performing a spatial representativeness assessment on the candidate sampling point location matrix to obtain a sampling representativeness index, determining that the sampling representativeness index is higher than a preset sampling start threshold, generating a sampling execution path planning vector based on the candidate sampling point location matrix, executing coal quality sampling operations according to the sampling execution path planning vector, collecting a sample set and process visual feedback streams, and combining spatial representativeness assessment with a closed-loop quality verification mechanism to avoid collisions with the coal-carrying vehicle, thereby achieving adaptive optimization of sampling points and dynamic assurance of sampling effectiveness.

Description

Technical Field

[0001] This invention relates to the field of coal sampling technology, and more specifically, to an optimized control method for a coal sampling device. Background Technology

[0002] Coal sampling refers to the technical process of obtaining representative coal samples from the cargo compartments of vehicles transporting coal, and it is a key link in coal quality inspection and trade settlement.

[0003] Current coal sampling technologies mainly rely on manual shoveling or fixed-point mechanical drilling, with sample collection based on operational experience or preset coordinates. However, manual sampling is susceptible to subjective interference, leading to insufficient sample representativeness, and is characterized by low efficiency and high safety risks. While mechanical sampling can improve efficiency, it often employs a fixed surface single-point sampling mode, which is difficult to adapt to the uneven distribution of coal quality within the coal compartment. It also cannot dynamically adjust the sampling strategy according to the real-time coal quality status, resulting in increased sampling deviation. Furthermore, it lacks a real-time verification and feedback optimization mechanism for sampling quality, affecting the accuracy of coal quality assessment and the reliability of sampling operations. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide an optimized control method for a coal quality sampling device. This method guides a stratified multi-point sampling strategy by constructing a coal quality heterogeneity distribution map, and combines spatial representativeness assessment and closed-loop quality verification mechanisms to achieve adaptive optimization of sampling points and dynamic assurance of sampling effectiveness.

[0005] To achieve the above objectives, the present invention provides an optimized control method for a coal quality sampling device, comprising: In response to the detection of the vehicle compartment being in place, the system acquires a set of structural feature parameters of the target compartment and a set of coal surface images based on pre-deployed cameras. Based on the historical coal quality sampling dataset, a coal quality unevenness distribution map of the target car is generated, wherein the historical coal quality sampling dataset consists of historical sampling records of the same loading capacity level as the target car. A candidate location matrix for sampling points is generated based on the set of structural feature parameters and the coal quality heterogeneity distribution map. Spatial representativeness evaluation is performed on the candidate sampling location matrix to obtain a sampling representativeness index. In response to determining that the sampling representativeness index is higher than a preset sampling start threshold, a sampling execution path planning vector is generated based on the candidate sampling location matrix. The coal quality sampling operation is performed according to the sampling execution path planning vector, and the sample set and process visual feedback stream are collected simultaneously.

[0006] Furthermore, based on historical coal quality sampling datasets, a coal quality non-uniformity distribution map of the target carriage is generated, including: The historical coal quality sampling dataset is preprocessed to obtain a standardized historical sampling dataset; Coal spectral features are extracted from the standardized historical sampling dataset to obtain a coal composition feature vector set; The surface image set of coal is subjected to texture depth analysis to obtain the surface texture feature matrix; The coal quality component feature vector set and the surface texture feature matrix are fused and mapped to obtain the coal quality unevenness distribution map.

[0007] Further, a candidate location matrix for sampling points is generated based on the set of structural feature parameters and the coal quality heterogeneity distribution map, including: Based on the set of structural feature parameters and the coal quality heterogeneity distribution map, the sampling strategy decision for the target carriage is determined; In response to determining that the sampling strategy decision is a hierarchical multi-point sampling mode, a candidate location matrix of sampling points is generated based on the coal quality heterogeneity distribution map.

[0008] Further, based on the set of structural feature parameters and the coal quality heterogeneity distribution map, the sampling strategy decision for the target carriage is determined, including: By analyzing the set of structural feature parameters, the parameters of the car body volume, the parameters of the side panel height, and the parameters of loading flatness are obtained. Calculate the degree of coal quality variation based on the coal quality heterogeneity distribution map; Based on the vehicle compartment volume parameters, the side panel height parameters, and the loading flatness parameters, the range of sampleable space in the vehicle compartment is determined; The sampling strategy decision is determined based on the coupling relationship between the coal quality variation value and the sampleable space range of the carriage.

[0009] Further, in response to determining that the sampling strategy decision is a hierarchical multi-point sampling mode, a candidate location matrix of sampling points is generated based on the coal quality heterogeneity distribution map, including: High-gradient region identification is performed on the coal quality heterogeneity distribution map to obtain a set of regions with significant variations; Based on the sampleable space range of the carriage, the set of significantly mutated regions is spatially constrained and filtered to obtain a set of valid candidate regions; The set of effective candidate regions is discretized by gridding to obtain a set of discrete sampling points; Based on the spatial coordinate encoding of the discrete sampling point set, the candidate position matrix of the sampling points is constructed.

[0010] Furthermore, the spatial representativeness of the candidate sampling point location matrix is ​​evaluated to obtain a sampling representativeness index, including: The spatial distribution uniformity of the candidate sampling point location matrix is ​​calculated to obtain the point dispersion index; The coal quality coverage of the candidate sampling point location matrix is ​​evaluated to obtain the component coverage integrity value; The sampling representativeness index is determined by a weighted fusion of the point dispersion index and the component coverage integrity value.

[0011] Further, in response to determining that the sampling representativeness index is higher than a preset sampling initiation threshold, a sampling execution path planning vector is generated based on the candidate sampling point position matrix, including: Accessibility analysis is performed on each point in the candidate location matrix of the sampling points to obtain the point accessibility vector; Based on the reachability vector of the points and the set of structural feature parameters, the sampling drill bit attitude parameters of each point are calculated to obtain the attitude parameter matrix; The sampling execution path planning vector is obtained by performing path optimization calculations based on the point reachability vector and the attitude parameter matrix.

[0012] Further, based on the point reachability vector and the attitude parameter matrix, path optimization calculation is performed to obtain the sampled execution path planning vector, including: The candidate path segment set is determined based on the point reachability vector; Collision risk detection is performed on the candidate path segment set to obtain a path safety score set; The candidate path segment set is filtered and reorganized based on the path safety score set to obtain the sampled execution path planning vector.

[0013] Furthermore, it also includes: The sampling strategy decision is a single-point surface sampling mode.

[0014] Furthermore, in response to the sampling strategy decision being a single-point surface sampling mode, the following includes: Obtain the coordinates of the geometric center of the carriage and the coordinates of the surface center point to determine the single-point sampling coordinates; Generate a vertical descent sampling path planning vector based on the single-point sampling coordinates; A single coal quality sampling operation is performed based on the path planning vector, and the sample set and process visual feedback stream are collected simultaneously.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention discloses an optimized control method for a coal quality sampling device. The method involves detecting a vehicle carriage positioning signal, acquiring a set of structural feature parameters and a set of coal surface images of the target carriage using a camera, generating a coal quality heterogeneity distribution map of the target carriage based on historical coal quality sampling datasets, generating a candidate sampling point location matrix based on the structural feature parameter set and the coal quality heterogeneity distribution map, evaluating the spatial representativeness of the candidate sampling point location matrix to obtain a sampling representativeness index, determining that the index is higher than a preset sampling initiation threshold, generating a sampling execution path planning vector based on the candidate sampling point location matrix, executing coal quality sampling operations according to the sampling execution path planning vector, collecting sample sets and process visual feedback streams, and combining spatial representativeness evaluation with a closed-loop quality verification mechanism to avoid collisions with the coal-carrying vehicle, thereby achieving adaptive optimization of sampling point locations and dynamic assurance of sampling effectiveness. Attached Figure Description

[0016] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart illustrating an optimized control method for a coal sampling device according to an embodiment of the present invention is shown. Detailed Implementation

[0017] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0018] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They 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. Therefore, they should not be construed as limitations on this application.

[0019] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0020] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0021] The following is a description of preferred embodiments of the present invention in conjunction with the accompanying drawings.

[0022] like Figure 1 As shown, an embodiment of the present invention discloses an optimized control method for a coal quality sampling device, comprising: S110: In response to the detection of the vehicle compartment being in place signal, acquire the set of structural feature parameters of the target compartment and the set of coal surface images based on the pre-deployed camera; S120: Generate a coal quality unevenness distribution map of the target car based on the historical coal quality sampling dataset, wherein the historical coal quality sampling dataset is a historical sampling record with the same loading capacity level as the target car; S130: Generate a candidate location matrix for sampling points based on the set of structural feature parameters and the coal quality heterogeneity distribution map; S140: Perform spatial representativeness evaluation on the candidate sampling point position matrix to obtain a sampling representativeness index. In response to determining that the sampling representativeness index is higher than a preset sampling start threshold, generate a sampling execution path planning vector based on the candidate sampling point position matrix. S150: Perform coal quality sampling operation according to the sampling execution path planning vector, and simultaneously collect sample set and process visual feedback stream.

[0023] Specifically, the vehicle's positioning signal is triggered by an inductive loop or lidar deployed at the sampling station entrance. The system confirms positioning when the coal truck has fully entered the sampling station and remained stationary for more than 5 seconds. The set of structural characteristic parameters includes the truck's volume, sideboard height, length and width dimensions, and loading flatness level (divided into four levels: flat, slightly raised, moderately raised, and heavily raised), acquired through scanning with a 3D lidar array positioned above the sampling station. The coal surface image set is acquired using an industrial-grade high-resolution camera, including visible light and near-infrared spectral images from the front, middle, and rear of the truck, used to characterize the surface coal's texture, color distribution, and preliminary moisture content.

[0024] In some embodiments of this application, generating a coal quality non-uniformity distribution map of the target carriage based on historical coal quality sampling datasets includes: The historical coal quality sampling dataset is preprocessed to obtain a standardized historical sampling dataset; Coal spectral features are extracted from the standardized historical sampling dataset to obtain a coal composition feature vector set; The surface image set of coal is subjected to texture depth analysis to obtain the surface texture feature matrix; The coal quality component feature vector set and the surface texture feature matrix are fused and mapped to obtain the coal quality unevenness distribution map.

[0025] Specifically, the historical coal quality sampling dataset was selected based on historical sampling records of the same loading capacity as the target wagon, using historical data of the current vehicle's grade and the same type of coal (with a sample size of no less than 50 sets). The historical coal quality sampling dataset underwent preprocessing, including outlier removal (using box plots to remove data exceeding 1.5 times the interquartile range), missing value imputation (using K-nearest neighbor imputation with K set to 5), and dimension unification (converting all mass parameters to kilograms and calorific value parameters to megajoules per kilogram), resulting in a standardized historical sampling dataset. Coal spectral features were extracted from the standardized historical sampling dataset, and principal component analysis (PCA) was used for dimensionality reduction, retaining the top N principal components with a cumulative variance contribution rate exceeding 85%, forming a coal quality component feature vector set (dimension N×1, including key indicators such as ash, sulfur, volatile matter, and calorific value). Texture depth analysis was performed on a set of coal surface images. The Gray-Level Co-occurrence Matrix (GLCM) algorithm was used to extract four texture features: contrast, correlation, energy, and homogeneity, constructing a surface texture feature matrix (4×3 dimensions, where 3 represents three viewpoints). The coal composition feature vector set was fused and mapped with the surface texture feature matrix. A weighted stitching method (0.6 for spectral features and 0.4 for texture features) was used to construct a coal quality heterogeneity distribution map. This map, based on the overhead view of the car body, was divided into 50×50 grid cells, each storing the deviation of the coal quality parameters from the mean at that location. This method can accurately determine the car body's positioning and quickly obtain comprehensive and accurate structural feature parameters and coal surface image information for the target car body. Targeted analysis and processing can be performed based on the actual conditions of different car bodies, greatly improving the accuracy of subsequent operations. This helps to more accurately assess the quality and grade differences of coal.

[0026] In some embodiments of this application, generating a candidate location matrix for sampling points based on the set of structural feature parameters and the coal quality heterogeneity distribution map includes: Based on the set of structural feature parameters and the coal quality heterogeneity distribution map, the sampling strategy decision for the target carriage is determined; In response to determining that the sampling strategy decision is a hierarchical multi-point sampling mode, a candidate location matrix of sampling points is generated based on the coal quality heterogeneity distribution map.

[0027] In some embodiments of this application, determining the sampling strategy decision for the target carriage based on the set of structural feature parameters and the coal quality heterogeneity distribution map includes: By analyzing the set of structural feature parameters, the parameters of the car body volume, the parameters of the side panel height, and the parameters of loading flatness are obtained. Calculate the degree of coal quality variation based on the coal quality heterogeneity distribution map; Based on the vehicle compartment volume parameters, the side panel height parameters, and the loading flatness parameters, the range of sampleable space in the vehicle compartment is determined; The sampling strategy decision is determined based on the coupling relationship between the coal quality variation value and the sampleable space range of the carriage.

[0028] Specifically, by analyzing the structural feature parameter set, the carriage volume parameter, the railing height parameter, and the loading flatness parameter (F, grades 1-4) are obtained. Calculate the coal quality variation degree value (Cv) according to the coal quality unevenness distribution map. The specific steps are as follows: perform zonal statistics on the distribution map (divided into 9 equal-area sub-regions), calculate the mean set (μ1 to μ9) of the coal quality parameters (represented by calorific value) in each sub-region, calculate the variance statistic (σ²) of this mean set, and perform normalization according to the formula Cv = σ² / (μ_max - μ_min), where μ_max and μ_min are the maximum and minimum calorific values of the historical data set, to obtain the coal quality variation degree value within the range of 0-1. According to the carriage volume parameter V, the railing height parameter H, and the loading flatness parameter S, combined with the maximum sampling depth D of the sampling drill bit (a preset fixed value determined according to the equipment model), calculate the carriage's available sampling space range R through the formula: R = V×S×(D / H). According to the coupling relationship between the coal quality variation degree value Cv and the carriage's available sampling space range, determine the sampling strategy decision. The sampling strategy decision is the stratified multi-point sampling mode or the single-point surface sampling mode. Specifically: construct the coupling relationship, and the calculation formula for the coupling coefficient K is: K = Cv×(R / R_max), where R_max is the preset maximum available sampling space threshold (determined according to the sampling station design specifications); the value range of the coupling coefficient K is [0,1]. The larger the K value, the higher the matching degree between the coal quality variation degree and the available sampling space, and the more necessary it is to cover different coal quality regions through stratified multi-point sampling; preset two coupling coefficient thresholds K1 (0.5) and K2 (0.8), and clarify the sampling strategy decision according to the value of the coupling coefficient K: when K≥K2 (≥0.8): the coal quality variation degree is extremely high and the available sampling space is sufficient, and the sampling strategy decision is determined to be the stratified multi-point sampling mode; when K1 ≤ K<K2 (0.5≤K<0.8): the coal quality variation degree is medium and the available sampling space is moderate, and the sampling strategy decision is determined to be the stratified multi-point sampling mode; when K<K1 (<0.5): the coal quality variation degree is low and the available sampling space is limited, and the sampling strategy decision is determined to be the single-point surface sampling mode. Only one surface sample needs to be collected in the center area of the carriage to reflect the coal quality of the whole vehicle. Through the stratified multi-point sampling mode or the single-point surface sampling mode, the intelligent sampling of the coal in the carriage can be ensured, the accuracy of sampling extraction can be guaranteed, and limitations can be avoided.

[0029] In some embodiments of the present application, in response to determining that the sampling strategy decision is the stratified multi-point sampling mode, generate a sampling point candidate position matrix based on the coal quality unevenness distribution map, including: Identify the high-gradient regions of the coal quality unevenness distribution map to obtain a set of regions with significant variations; Perform spatial constraint screening on the set of regions with significant variations according to the carriage's available sampling space range to obtain a set of effective candidate regions; The set of effective candidate regions is discretized by gridding to obtain a set of discrete sampling points; Based on the spatial coordinate encoding of the discrete sampling point set, the candidate position matrix of the sampling points is constructed.

[0030] Specifically, a gradient detection algorithm is used to analyze the coal quality inhomogeneity distribution map, identifying areas where the gradient change of the quantitative value of coal quality inhomogeneity exceeds a preset gradient threshold (0.3), which are designated as the set of areas with significant variation (these areas have large coal quality differences and are key sampling areas). Combining the sampleable space R of the carriage and the coordinate boundaries, the parts of the set of areas with significant variation that exceed the sampleable range and are located in prohibited sampling areas are removed, resulting in a set of effective candidate areas. The effective candidate area set is then uniformly gridded (the grid size is set according to the sampling accuracy requirements, usually 0.5m×0.5m), and the center point of each grid is used as a discrete sampling point, forming a set of discrete sampling points. Each point in the set of discrete sampling points is encoded with two-dimensional coordinates (with the lower left corner of the carriage as the origin, the horizontal axis as the X-axis, and the vertical axis as the Y-axis). The coordinate values ​​of all points are arranged in rows to construct a matrix of candidate sampling point positions (the rows of the matrix correspond to the point numbers, and the columns correspond to the X and Y coordinate values), ensuring that the distribution of points conforms to the coal quality difference pattern and facilitates subsequent path planning.

[0031] In some embodiments of this application, spatial representativeness evaluation is performed on the candidate location matrix of sampling points to obtain a sampling representativeness index, including: The spatial distribution uniformity of the candidate sampling point location matrix is ​​calculated to obtain the point dispersion index; The coal quality coverage of the candidate sampling point location matrix is ​​evaluated to obtain the component coverage integrity value; The sampling representativeness index is determined by a weighted fusion of the point dispersion index and the component coverage integrity value.

[0032] Specifically, the spatial dispersion formula is used to calculate the spatial distribution uniformity of all points in the candidate sampling location matrix, resulting in the point dispersion index P (value range [0,1], the closer the value is to 1, the more uniform the point distribution, avoiding clustering); the calculation formula is: P=1-(standard deviation of the distance between all points and the center point / preset maximum standard deviation of the distance); combined with the coal quality inhomogeneity distribution map, the quantitative value of coal quality inhomogeneity corresponding to all points in the candidate sampling location matrix is ​​statistically analyzed, and the proportion of the number of covered coal quality intervals to the total number of coal quality intervals is used as the component coverage integrity value Q (value range [0,1]). [1] The closer the value is to 1, the more comprehensive the coal types covered); a weighted fusion algorithm is used (the point dispersion index has a weight of 0.5, and the composition coverage integrity value has a weight of 0.5) to calculate the sampling representativeness index R_rep=0.5×P+0.5×Q; where the preset sampling start threshold is 0.7 (which can be adjusted according to the sampling accuracy requirements). When R_rep>0.7, the point representativeness is determined to meet the standard, and the subsequent path planning steps are entered; when R_rep≤0.7, the areas with significant variation are re-screened, the grid size is adjusted, and the candidate location matrix of sampling points is regenerated until the standard is met.

[0033] In some embodiments of this application, in response to determining that the sampling representativeness index is higher than a preset sampling initiation threshold, generating a sampling execution path planning vector based on the sampling point candidate position matrix includes: Accessibility analysis is performed on each point in the candidate location matrix of the sampling points to obtain the point accessibility vector; Based on the reachability vector of the points and the set of structural feature parameters, the sampling drill bit attitude parameters of each point are calculated to obtain the attitude parameter matrix; The sampling execution path planning vector is obtained by performing path optimization calculations based on the point reachability vector and the attitude parameter matrix.

[0034] In some embodiments of this application, path optimization calculations are performed based on the point reachability vector and the attitude parameter matrix to obtain the sampled execution path planning vector, including: The candidate path segment set is determined based on the point reachability vector; Collision risk detection is performed on the candidate path segment set to obtain a path safety score set; The candidate path segment set is filtered and reorganized based on the path safety score set to obtain the sampled execution path planning vector.

[0035] Specifically, based on the structural feature parameter set of the carriage (sideboard height, loading flatness) and the motion parameters of the sampling drill bit (maximum extension length, maximum pitch angle), each point in the candidate sampling point position matrix is ​​analyzed one by one to determine whether the sampling drill bit can reach the point without interference, generating a point reachability vector (vector elements are 0 or 1, 1 indicates reachability, 0 indicates inaccessibility), and eliminating points with reachability of 0; based on the valid point coordinates in the point reachability vector, combined with the kinematic model of the sampling drill bit, the attitude parameters such as the drill bit pitch angle, extension length, and rotation angle corresponding to each valid point are calculated, and an attitude parameter matrix is ​​constructed (matrix rows correspond to valid point indices, and columns correspond to the values ​​of each attitude parameter); candidate path segment generation: based on the coordinate order of the valid points, all possible path combinations (i.e., candidate path segment set) are generated, and each path segment is generated... The path segment corresponds to the walking trajectory of two adjacent points; a collision detection algorithm is used to simulate the process of the mechanical sampling arm walking along each candidate path segment, detect whether there is interference with the side panels of the carriage or coal piles, quantify the collision risk, and obtain a path safety score set (score range [0,1], 1 indicates no collision risk, 0 indicates serious collision risk); the path safety score threshold is set to 0.9, and candidate path segments with a safety score <0.9 are eliminated; for the remaining candidate path segments, the path length (preferring the shortest path) and the difficulty of attitude adjustment (preferring the path with small changes in attitude parameters) are combined to screen and reorganize them to obtain the optimal sampling execution path planning vector (the vector elements are the coordinates of each effective point and the corresponding attitude parameters, arranged in the walking order), to ensure path safety and efficiency, avoid mechanical collisions between vehicles, and avoid accidents.

[0036] In some embodiments of this application, it also includes: The sampling strategy decision is a single-point surface sampling mode.

[0037] In some embodiments of this application, in response to the sampling strategy decision being a single-point surface sampling mode, the following includes: Obtain the coordinates of the geometric center of the carriage and the coordinates of the surface center point to determine the single-point sampling coordinates; Generate a vertical descent sampling path planning vector based on the single-point sampling coordinates; A single coal quality sampling operation is performed based on the path planning vector, and the sample set and process visual feedback stream are collected simultaneously.

[0038] Specifically, the top planar contour of the carriage sideboard is analyzed using 3D LiDAR scanning data to calculate the geometric midpoints of the carriage's length and width directions, obtaining the geometric center coordinates (Xc, Yc, Zc), where Zc is the elevation of the carriage floor. Stereoscopic vision reconstruction is performed on the coal surface image set to identify the coordinates of the highest point of the coal pile surface (Xs, Ys, Zs), and the surface center coordinates are calculated as (Xc, Yc, Zs). The single-point sampling coordinates are determined to be (Xc, Yc, Zs - 0.1 meters), meaning a 0.1-meter layer thickness of surface cover is reserved below the surface center point to ensure the sampling head contacts the actual coal rather than dust or scattered debris. By integrating the dual positioning of the carriage's geometric center and the coal surface center, the single-point sampling location is ensured to be both at the physical center of the carriage (ensuring spatial representativeness) and aligned with the actual surface of the coal pile (avoiding scattered areas at the edges), solving the technical problems of high arbitrariness and sample deviation caused by point offset in traditional single-point sampling. By reserving a 0.1-meter surface cover layer, surface weathering debris generated during transportation is effectively filtered, improving sample authenticity and testing accuracy. The specific implementation of generating a vertical descent sampling path planning vector based on the single-point sampling coordinates includes: setting the initial position of the sampling drill bit base, calculating the horizontal distance D1 and azimuth angle θ between the base and the single-point sampling coordinates. When D1 ≤ 4 meters (drill bit working radius), the boom pitch angle α = 90 degrees (vertically downward), the forearm pitch angle β = 0 degrees, and the sampling head rotation angle γ = 0 degrees, forming a vertical attitude path planning vector (Xc, Yc, Zs-0.1, 90, 0, 0). The sampling descent stroke is set to a standard surface sampling depth of 0.3 meters, generating a straight interpolation path from a safe height (Zs+0.5 meters) to the target depth (Zs-0.1-0.3 = Zs-0.4 meters), with the path speed set to 0.1 meters per second to ensure a smooth and impact-free sampling process. By planning the vertical descent path, the robotic arm's motion trajectory is simplified to the maximum extent, reducing the number of joint posture changes, significantly shortening the time of a single sampling operation, and improving sampling efficiency. The vertical posture ensures orthogonal contact between the sampling head and the coal surface, avoiding sample contamination and lateral stress on the robotic arm caused by oblique drilling, extending equipment lifespan and ensuring operational safety. A complete sampling process traceability archive is established (video recording of drilling depth and cross-sectional status), providing a visual basis for subsequent quality consistency verification.

[0039] In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0040] Although the invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, as long as there is no structural conflict, the features in the embodiments disclosed in this invention can be combined with each other in any way. The fact that not all of these combinations are described in this specification is merely for the sake of brevity and resource conservation.

[0041] It will be understood by those skilled in the art that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An optimized control method for a coal quality sampling device, characterized in that, include: In response to the detection of the vehicle compartment being in place, the system acquires a set of structural feature parameters of the target compartment and a set of coal surface images based on pre-deployed cameras. Based on the historical coal quality sampling dataset, a coal quality unevenness distribution map of the target car is generated, wherein the historical coal quality sampling dataset consists of historical sampling records of the same loading capacity level as the target car. A candidate location matrix for sampling points is generated based on the set of structural feature parameters and the coal quality heterogeneity distribution map. Spatial representativeness evaluation is performed on the candidate sampling location matrix to obtain a sampling representativeness index. In response to determining that the sampling representativeness index is higher than a preset sampling start threshold, a sampling execution path planning vector is generated based on the candidate sampling location matrix. The coal quality sampling operation is performed according to the sampling execution path planning vector, and the sample set and process visual feedback stream are collected simultaneously.

2. The optimized control method for the coal quality sampling device according to claim 1, characterized in that, Based on historical coal quality sampling datasets, a coal quality non-uniformity distribution map of the target carriage is generated, including: The historical coal quality sampling dataset is preprocessed to obtain a standardized historical sampling dataset; Coal spectral features are extracted from the standardized historical sampling dataset to obtain a coal composition feature vector set; The surface image set of coal is subjected to texture depth analysis to obtain the surface texture feature matrix; The coal quality component feature vector set and the surface texture feature matrix are fused and mapped to obtain the coal quality unevenness distribution map.

3. The optimized control method for the coal quality sampling device according to claim 1, characterized in that, A candidate location matrix for sampling points is generated based on the set of structural feature parameters and the coal quality heterogeneity distribution map, including: Based on the set of structural feature parameters and the coal quality heterogeneity distribution map, the sampling strategy decision for the target carriage is determined; In response to determining that the sampling strategy decision is a hierarchical multi-point sampling mode, a candidate location matrix of sampling points is generated based on the coal quality heterogeneity distribution map.

4. The optimized control method for the coal quality sampling device according to claim 3, characterized in that, Based on the set of structural feature parameters and the coal quality heterogeneity distribution map, the sampling strategy decision for the target carriage is determined, including: By analyzing the set of structural feature parameters, the parameters of the car body volume, the parameters of the side panel height, and the parameters of loading flatness are obtained. Calculate the degree of coal quality variation based on the coal quality heterogeneity distribution map; Based on the vehicle compartment volume parameters, the side panel height parameters, and the loading flatness parameters, the range of sampleable space in the vehicle compartment is determined; The sampling strategy decision is determined based on the coupling relationship between the coal quality variation value and the sampleable space range of the carriage.

5. The optimized control method for the coal quality sampling device according to claim 3, characterized in that, In response to determining that the sampling strategy decision is a hierarchical multi-point sampling mode, a candidate location matrix of sampling points is generated based on the coal quality heterogeneity distribution map, including: High-gradient region identification is performed on the coal quality heterogeneity distribution map to obtain a set of regions with significant variations; Based on the sampleable space range of the carriage, the set of significantly mutated regions is spatially constrained and filtered to obtain a set of valid candidate regions; The set of effective candidate regions is discretized by gridding to obtain a set of discrete sampling points; Based on the spatial coordinate encoding of the discrete sampling point set, the candidate position matrix of the sampling points is constructed.

6. The optimized control method for the coal quality sampling device according to claim 1, characterized in that, The spatial representativeness of the candidate location matrix of the sampling points is evaluated to obtain the sampling representativeness index, including: The spatial distribution uniformity of the candidate sampling point location matrix is ​​calculated to obtain the point dispersion index; The coal quality coverage of the candidate sampling point location matrix is ​​evaluated to obtain the component coverage integrity value; The sampling representativeness index is determined by a weighted fusion of the point dispersion index and the component coverage integrity value.

7. The optimized control method for the coal quality sampling device according to claim 1, characterized in that, In response to determining that the sampling representativeness index is higher than a preset sampling initiation threshold, a sampling execution path planning vector is generated based on the candidate sampling point position matrix, including: Accessibility analysis is performed on each point in the candidate location matrix of the sampling points to obtain the point accessibility vector; Based on the reachability vector of the points and the set of structural feature parameters, the sampling drill bit attitude parameters of each point are calculated to obtain the attitude parameter matrix; The sampling execution path planning vector is obtained by performing path optimization calculations based on the point reachability vector and the attitude parameter matrix.

8. The optimized control method for the coal quality sampling device according to claim 7, characterized in that, Based on the point reachability vector and the attitude parameter matrix, path optimization calculation is performed to obtain the sampled execution path planning vector, including: The candidate path segment set is determined based on the point reachability vector; Collision risk detection is performed on the candidate path segment set to obtain a path safety score set; The candidate path segment set is filtered and reorganized based on the path safety score set to obtain the sampled execution path planning vector.

9. The optimized control method for the coal quality sampling device according to claim 3, characterized in that, Also includes: The sampling strategy decision is a single-point surface sampling mode.

10. The optimized control method for the coal quality sampling device according to claim 9, characterized in that, In response to the sampling strategy decision to adopt a single-point surface sampling mode, the following is included: Obtain the coordinates of the geometric center of the carriage and the coordinates of the surface center point to determine the single-point sampling coordinates; Generate a vertical descent sampling path planning vector based on the single-point sampling coordinates; A single coal quality sampling operation is performed based on the path planning vector, and the sample set and process visual feedback stream are collected simultaneously.

Images