Unmanned coal mining method for steeply inclined thin coal seam

Abstract

The invention provides an unmanned coal mining method for a steeply inclined thin coal seam, which belongs to the technical field of coal mining, and comprises the following steps: constructing a three-dimensional geological model through geological radar and drilling data, dividing the coal seam into a horizontal layer with the thickness of 0.8-1.2 m by adopting a layered stoping method, and optimizing layering by utilizing a multi-feature clustering function. The working surface is divided by adopting a honeycomb hexagonal grid, so that the path planning precision is improved; comprehensively considering factors such as coal seam thickness, inclination angle and coal quality to design an optimal coal mining path; constructing a multi-dimensional weighted graph structure to form a coal mining path navigation graph; a one-way partition stoping strategy is developed to improve the coal mining efficiency; real-time path dynamic planning is realized, and the path is updated according to real-time data such as gas content; the coal mining robot adjusts coal mining parameters accordingly, and the technical problems that unmanned coal mining path planning is low in precision and difficult to adapt to coal seam changes under the complex geological condition of the steeply inclined thin coal seam are effectively solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of coal mining, and specifically relates to a method for unmanned coal mining in steeply inclined thin coal seams. Background Technique

[0002] Coal mining in steeply inclined thin coal seams is an important branch in the field of coal mine exploitation. Traditional coal mining technologies mainly rely on manual or semi-mechanized equipment to arrange working faces along the strike of the coal seam for exploitation. Conventional coal mining methods include inclined slicing method, horizontal slicing method, and working face arrangement method along the strike, etc., and the cutting and coal winning operations are completed through manual operation of coal mining machinery and equipment. With the development of coal mine intelligentization, coal mining robots have begun to be applied in some mining areas with better coal seam conditions, and basic coal mining tasks are executed through preset paths.

[0003] However, traditional coal mining technologies for steeply inclined thin coal seams have many defects in practical applications. First of all, the coal mining path planning usually uses simple rectangular grid division of the working face, which is difficult to accurately adapt to the irregular coal seam boundary; secondly, the path algorithms mostly adopt fixed modes and cannot be dynamically optimized according to multi-dimensional characteristics such as coal seam thickness, dip angle, and coal quality; furthermore, the existing technologies lack a response mechanism for real-time geological changes during coal mining, resulting in coal mining operations being difficult to adapt to the sudden changes in the coal seam.

[0004] Especially in the environment of steeply inclined thin coal seams with a dip angle greater than 45 degrees, due to the complex and changeable geological conditions, uneven coal seam thickness, and poor roof stability, the existing coal mining path planning technologies have low accuracy and are difficult to be dynamically adjusted according to the real-time changes of the coal seam, resulting in problems such as low coal mining efficiency, high safety risks, and insufficient resource recovery rate. There is an urgent need for an unmanned coal mining technical solution that can accurately plan coal mining paths and dynamically adapt to coal seam changes. Summary of the Invention

[0005] In view of this, the present invention provides a method for unmanned coal mining in steeply inclined thin coal seams, which can solve the technical problems of low accuracy of unmanned coal mining path planning and difficulty in adapting to coal seam changes under the complex geological conditions of steeply inclined thin coal seams in the prior art.

[0006] The present invention is implemented as follows: The present invention provides a method for unmanned coal mining in steeply inclined thin coal seams, including: constructing a three-dimensional geological model of the thin coal seam to determine the coal seam dip angle distribution matrix, coal seam thickness data matrix, and fault distribution matrix; dividing the horizontal layers of the thin coal seam and establishing a three-dimensional coal mining grid model; using honeycomb hexagonal grids to divide the working face to form a basic network for calculating coal mining paths; designing the main travel path of the coal mining robot based on the Bellman-Ford algorithm, and comprehensively considering the coal seam thickness data matrix, coal seam dip angle distribution matrix, and coal quality distribution matrix to calculate the optimal coal mining path; constructing a coal mining path navigation map to form a multi-dimensional weighted graph structure; developing a one-way sectional coal winning strategy; realizing the dynamic path planning of the coal mining robot and the adjustment of coal mining parameters.

[0007] Among them, the steps of dividing the steeply inclined thin coal seam into horizontal layers and establishing a three-dimensional coal mining grid model are specifically as follows: Design a working face arranged along the strike, adopt the slicing mining method, divide the steeply inclined thin coal seam into multiple horizontal layers according to the thickness, with each layer having a thickness of 0.8 to 1.2 meters, use a multi-feature clustering function to optimize the layering of the coal seam, and establish a three-dimensional coal mining grid model.

[0008] Among them, the steps of using a honeycomb hexagonal grid to divide the working face to form a basic network for calculating the coal mining path are specifically as follows: Divide the working face into multiple regular hexagonal units with a side length of 0.6 meters to form a basic network for calculating the coal mining path, improving the path planning accuracy; The honeycomb hexagonal grid is specifically a spatial representation method that divides the space of the coal mining working face into a set of regular hexagonal units with a side length of 0.6 meters. Each hexagonal unit shares boundaries with the six adjacent units around it. Compared with the traditional rectangular grid, the hexagonal grid has a constant adjacent unit distance in any direction, reducing the cumulative error of direction deviation and improving the path smoothness.

[0009] Among them, the steps of developing a one-way partition mining strategy are specifically as follows: Divide the working face into multiple mining areas, set a one-way traveling path for each mining area, avoid the coal mining robot passing through the same area repeatedly, and improve the coal mining efficiency.

[0010] Among them, the steps of realizing real-time path dynamic planning are specifically as follows: Based on the layered thickness parameter matrix and the set of boundary points for the horizontal layer division of the working face, realize real-time path dynamic planning. The coal mining robot collects the gas content distribution matrix according to the real-time detection data of the coal seam, periodically updates the grid weights in the local area, and recalculates the optimal path to adapt to the changes in the coal seam.

[0011] Among them, the steps of the coal mining robot adjusting the coal mining parameters to achieve precise cutting and safe and efficient coal mining are specifically as follows: The coal mining robot adjusts the coal mining parameters according to the coal seam thickness data matrix, the coal seam dip distribution matrix, the layered thickness parameter matrix and the set of boundary points for the horizontal layer division of the working face, controls the drum height and cutting depth, and realizes precise cutting and safe and efficient coal mining for the steeply inclined thin coal seam.

[0012] Among them, the slicing mining method is specifically to divide the steeply inclined thin coal seam into multiple parallel layers with a thickness of 0.8 to 1.2 meters in the horizontal direction. The coal mining robot independently completes the coal mining operation within each parallel layer. After the coal mining robot completes the coal mining of one parallel layer, it then enters the next parallel layer. It is applicable to steeply inclined coal seams with an inclination angle greater than 45 degrees. By controlling the roof activity range in layers, the risk of roof collapse is reduced, and the coal recovery rate is improved.

[0013] Among them, the Bellman-Ford algorithm is specifically a graph algorithm for finding the single-source shortest path. By performing multiple rounds of relaxation operations on all edges in the graph, the shortest paths from the starting point to all vertices are calculated. In the weighted graph structure designed for coal mining path planning, the edge weights incorporate the coal seam thickness coefficient, coal seam dip angle coefficient, and coal quality coefficient, and the coal mining path that optimally utilizes resources is achieved by minimizing the comprehensive cost.

[0014] Among them, the multi-feature clustering function is used to scientifically divide the coal mining face according to multiple physical properties of the coal seam. The inputs include the coal seam thickness data matrix, coal seam dip angle distribution matrix, coal quality distribution matrix, fault distribution matrix, and gas content distribution matrix, and the outputs are the stratified thickness parameter matrix and the set of boundary points for the horizontal layer division of the working face.

[0015] Among them, the one-way partition mining strategy is specifically a coal mining operation method designed for the characteristics of steeply inclined thin coal seams. A complete working face is divided into multiple mining areas with similar areas according to the changing characteristics of the coal seam thickness data matrix and the coal seam dip angle distribution matrix. A fixed coal mining direction is set within each mining area, and the coal mining robot completes coal mining in one mining area in a single direction and then enters the next mining area, avoiding the coal mining robot moving back and forth within the working face and reducing the non-productive moving time.

[0016] The present invention realizes the precise coal mining path planning for the complex steeply inclined thin coal seam environment by constructing a three-dimensional geological model of the thin coal seam, adopting the stratified mining method and the honeycomb hexagonal grid to divide the working face, and combining the Bellman-Ford algorithm to design the path of the coal mining robot. This method optimizes the stratification of the coal seam through the multi-feature clustering function, comprehensively considers multi-dimensional data such as coal seam thickness, dip angle, and coal quality, and constructs a multi-dimensional weighted graph structure with weight factors including the coal seam thickness coefficient, coal seam dip angle coefficient, and coal quality coefficient, overcoming the deficiencies of traditional technologies in dealing with irregular boundaries and multi-feature fusion. In particular, through real-time path dynamic planning and adaptive adjustment of coal mining parameters, the problem of difficultly coping with coal seam mutations during the coal mining process is solved, enabling the coal mining robot to update the path planning according to real-time detection data and adapt to coal seam changes.

[0017] The present invention effectively solves the technical problems of low accuracy in the unmanned coal mining path planning under the complex geological conditions of steeply inclined thin coal seams and difficulty in adapting to coal seam changes, improves the coal mining efficiency and resource recovery rate, reduces the risk of roof collapse, and provides technical support for the intelligent unmanned mining of steeply inclined thin coal seams. Brief Description of the Drawings

[0018] Figure 1 It is a flowchart of the method of the present invention.

[0019] Figure 2 It is a schematic diagram of the overall structure of the unmanned coal mining system for steeply inclined thin coal seams in Embodiment 2.

[0020] Figure 3 It is a schematic structural diagram of the coal mining robot in Embodiment 2.

[0021] Figure 4 It is a schematic diagram of honeycomb hexagonal grid path planning in Embodiment 2. Specific implementation manners

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0023] As Figure 1 shown, it is a flowchart of a method for unmanned coal mining in steeply inclined thin coal seams provided by the present invention. The method includes the following steps:

[0024] S01. Construct a three-dimensional geological model of the thin coal seam through ground penetrating radar and drilling data, determine the coal seam dip distribution matrix, coal seam thickness data matrix, and fault distribution matrix, and provide basic data for the path planning of the coal mining robot;

[0025] S02. Design a working face arranged along the strike, and adopt the stratified mining method. Divide the steeply inclined thin coal seam into multiple horizontal layers according to the thickness, with each layer having a thickness of 0.8 to 1.2 meters. Use the multi-feature clustering function to optimize the layering of the coal seam and establish a three-dimensional coal mining grid model;

[0026] S03. Divide the working face using a honeycomb hexagonal grid, divide the working face into multiple regular hexagonal units with a side length of 0.6 meters, form a basic network for calculating the coal mining path, and improve the path planning accuracy;

[0027] S04. Design the main travel path of the coal mining robot based on the Bellman-Ford algorithm, comprehensively consider the coal seam thickness data matrix, coal seam dip distribution matrix, and coal quality distribution matrix, calculate the optimal coal mining path, and generate a sequence of path node coordinates;

[0028] S05. Construct a coal mining path navigation map, establish weighted connections according to the honeycomb hexagonal grid nodes, and the weight factors include the coal seam thickness coefficient, coal seam dip coefficient, and coal quality coefficient to form a multi-dimensional weighted graph structure;

[0029] S06. Develop a one-way sectional mining strategy, divide the working face into multiple mining areas, set a one-way travel path for each mining area, avoid the coal mining robot passing through the same area repeatedly, and improve the coal mining efficiency;

[0030] S07. Based on the layered thickness parameter matrix and the set of boundary points for the division of the working face horizontal layers, real-time path dynamic planning is achieved. The coal mining robot collects the gas content distribution matrix according to the real-time detection data of the coal seam, periodically updates the grid weights in the local area, recalculates the optimal path, and adapts to the changes in the coal seam.

[0031] S08. The coal mining robot adjusts the coal mining parameters according to the coal seam thickness data matrix, the coal seam dip angle distribution matrix, the layered thickness parameter matrix and the set of boundary points for the division of the working face horizontal layers, controls the drum height and cutting depth, and realizes precise cutting and safe and efficient coal mining for steeply inclined thin coal seams.

[0032] Among them, the stratified mining method specifically divides the steeply inclined thin coal seam into multiple parallel layers with a thickness of 0.8 to 1.2 meters in the horizontal direction. The coal mining robot completes the coal mining operation independently within each parallel layer. After the coal mining robot completes the coal mining in one parallel layer, it enters the next parallel layer. It is applicable to steeply inclined coal seams with an inclination angle greater than 45 degrees. By controlling the roof activity range in layers, the risk of roof collapse is reduced, and the coal recovery rate is increased.

[0033] Among them, the honeycomb hexagonal grid is specifically a spatial representation method that divides the space of the coal mining working face into a set of regular hexagonal units with a side length of 0.6 meters. Each hexagonal unit shares boundaries with the six adjacent units around it. Compared with the traditional rectangular grid, the hexagonal grid has a constant adjacent unit distance in any direction, reduces the cumulative error of direction deviation, improves the path smoothness, and the hexagonal grid shows higher adaptability when dealing with irregular boundaries.

[0034] Among them, the Bellman-Ford algorithm is specifically a graph algorithm for finding the shortest path from a single source. By performing multiple rounds of relaxation operations on all edges in the graph, the shortest paths from the starting point to all vertices are calculated. In the weighted graph structure designed for coal mining path planning, the edge weights incorporate the coal seam thickness coefficient, the coal seam dip angle coefficient, and the coal quality coefficient. The coal mining path that optimally utilizes resources is achieved by minimizing the comprehensive cost, and the Bellman-Ford algorithm can handle negative-weight edges and is suitable for considering the changes in coal mining value brought about by the differences in coal quality.

[0035] Among them, the one-way partition mining strategy is specifically a coal mining operation method designed according to the characteristics of steeply inclined thin coal seams. A complete working face is divided into multiple mining areas with similar areas according to the variation characteristics of the coal seam thickness data matrix and the coal seam dip angle distribution matrix. A fixed coal mining direction is set in each mining area. The coal mining robot completes the coal mining in one mining area in a single direction and then enters the next mining area, avoiding the coal mining robot moving back and forth in the working face, reducing the non-productive moving time, and at the same time, by optimizing the connection design of the mining area boundaries, ensuring the continuous advancement of the coal mining face.

[0036] Among them, the multi-feature clustering function in step S02 is used to scientifically divide the coal mining face according to multiple physical properties of the coal seam. The inputs include the coal seam thickness data matrix, the coal seam dip distribution matrix, the coal quality distribution matrix, the fault distribution matrix, and the gas content distribution matrix, and the outputs are the layered thickness parameter matrix and the set of boundary points for the horizontal layer division of the working face; the coal seam thickness data matrix, the coal seam dip distribution matrix, and the fault distribution matrix are derived from the geological radar and drilling data in step S01; the coal quality distribution matrix is derived from the coal quality test data of the coal seam; the gas content distribution matrix is derived from the gas content monitoring data of the coal seam; the layered thickness parameter matrix and the set of boundary points for the horizontal layer division of the working face are used to guide the real-time path dynamic programming in step S07; the multi-feature clustering function specifically includes the following steps:

[0037] Step 1: For the space of the coal mining face, construct a coal seam feature vector matrix. Each element of the coal seam feature vector matrix includes the coal seam thickness value, coal seam dip value, coal quality grade value, fault distance value, and gas content value of the sampling point. Among them, the coal seam thickness value is derived from the coal seam thickness data matrix, the coal seam dip value is derived from the coal seam dip distribution matrix, the coal quality grade value is derived from the coal quality distribution matrix and represents the comprehensive evaluation of coal calorific value and ash content, the fault distance value is derived from the fault distribution matrix and represents the Euclidean distance from the sampling point to the nearest fault, and the gas content value is derived from the gas content distribution matrix;

[0038] Step 2: For the constructed coal seam feature vector matrix, perform dimensionality reduction processing through the principal component analysis method, extract the feature combinations that mainly affect coal mining safety and efficiency, reduce the computational complexity, and obtain the dimensionality-reduced feature data;

[0039] Step 3: Use the K-means clustering algorithm to perform clustering analysis on the dimensionality-reduced feature data. The initial clustering centers are selected according to the maximum distance principle to ensure the maximum feature difference between different categories and the minimum feature difference within the same category, and obtain the clustering result;

[0040] Step 4: According to the clustering result, combined with the coal seam spatial position information, determine the exact thickness and boundary of each horizontal layer, generate the layered thickness parameter matrix and the set of boundary points for the horizontal layer division of the working face, and provide a basis for the subsequent honeycomb hexagonal grid division.

[0041] Among them, the coal seam thickness data matrix is specifically a two-dimensional or three-dimensional data structure representing the coal seam thickness at different positions of the coal mining face, which is obtained through geological radar detection and drilling sampling, and is used to quantitatively describe the thickness change of the coal seam in space, providing a basis for the path planning and cutting depth control of the coal mining robot.

[0042] Among them, the coal seam dip distribution matrix is specifically a two-dimensional or three-dimensional data structure representing the dip angles of the coal seam at different positions of the coal mining face. It is obtained through geological exploration and dip measurement instruments and is used to quantitatively describe the variation of the dip degree of the coal seam in space, providing a basis for the attitude control and stability guarantee of coal mining robots.

[0043] Among them, the coal quality distribution matrix is specifically a two-dimensional or three-dimensional data structure representing the coal quality at different positions of the coal mining face. It is obtained through coal sample analysis and includes coal quality indicators such as calorific value, ash content, and sulfur content, and is used to quantitatively describe the variation of the coal quality in space, providing a decision-making basis for preferentially mining high-quality coal seams.

[0044] Among them, the fault distribution matrix is specifically a two-dimensional or three-dimensional data structure representing the positions and scales of geological faults in the coal mining face. It is obtained through geological exploration and seismic wave detection and is used to quantitatively describe the fault distribution of the coal seam in space, providing safety protection for coal mining robots to avoid dangerous areas.

[0045] Among them, the gas content distribution matrix is specifically a two-dimensional or three-dimensional data structure representing the gas concentrations at different positions of the coal mining face. It is obtained through gas detectors and gas sampling analysis and is used to quantitatively describe the variation of the gas content in the coal seam in space, providing a basis for the gas safety monitoring and explosion-proof measures of coal mining robots.

[0046] Among them, the stratified thickness parameter matrix is specifically a two-dimensional data structure representing the thickness values of each horizontal layer after multi-feature clustering optimization, including the starting depth and ending depth information of each layer in the working face, and is used to guide the division of the operation range of the coal mining robot in the vertical direction to ensure the scientificity and safety of stratified coal mining.

[0047] Among them, the set of boundary points for the horizontal layer division of the working face is specifically a set of spatial points representing the boundary contours of each horizontal layer after multi-feature clustering optimization, including the three-dimensional coordinate point sequences of the boundaries of each layer, and is used to accurately describe the boundaries of irregular coal seam layers, guiding the coal mining robot to accurately identify the positions of the layer boundaries and avoid overstepping the boundaries during coal mining.

[0048] The specific implementation manners of the above steps will be described in detail below.

[0049] The specific implementation of step S01 is to use a high-resolution geological radar to scan the coal seam area to obtain the electromagnetic wave reflection data of the coal seam. The scanning frequency is set to 500 MHz to 2 GHz, the radar detection depth is controlled within the range of 50 to 100 m, and the scanning line spacing is maintained at 1 to 2 m. At the same time, drilling holes are arranged, the drilling hole spacing is set to 20 to 30 m, the drilling cores are collected and analyzed to extract the coal seam thickness, dip angle and fault location information. The geological radar reflection data and the drilling core data are input into the 3D geological modeling software, and the Kriging interpolation algorithm is applied to spatially interpolate the coal seam parameters between the sampling points to generate a continuous 3D coal seam model. Based on the 3D model, the coal seam dip angle distribution matrix is extracted, the matrix resolution is 0.5 m × 0.5 m, and the coal seam dip angle value of each grid point on the working face is recorded. For the area where the dip angle is greater than 45 degrees, it is marked as the steeply inclined area. The coal seam thickness data matrix is extracted, the matrix resolution is the same as that of the dip angle distribution matrix, and the coal seam thickness value of each grid point on the working face is recorded. For the area where the thickness is less than 2 m, it is marked as the thin coal seam area. The fault location and scale are identified to generate a fault distribution matrix, and the distance value from each grid point to the nearest fault is recorded in the matrix. The area where the distance is less than 5 m is marked as the fault danger area. Finally, the coal seam dip angle distribution matrix, the coal seam thickness data matrix and the fault distribution matrix are integrated into a unified geological database to provide basic data support for the subsequent path planning of coal mining robots. The purpose of this step is to construct a high-precision 3D geological model through accurate geological data collection and processing to provide reliable geological parameters for unmanned coal mining operations.

[0050] The specific implementation of step S02 is based on the three-dimensional geological model obtained in S01. It is determined that the working face is arranged along the coal seam strike direction, and the length of the working face is determined according to the coal seam stability, usually controlled within the range of 50 - 100m. The slicing mining method is adopted, and the vertical thickness direction of the coal seam is divided into multiple horizontal slices. The thickness of each slice is set within the range of 0.8 - 1.2m. The selection of the slice thickness threshold is based on the maximum cutting height of the coal mining robot and the requirements of roof stability. The multi-feature clustering function combining K-means clustering and fuzzy C-means clustering is used to process the coal seam feature data. The input parameters include the coal seam thickness data matrix, coal seam dip distribution matrix, coal quality distribution matrix, fault distribution matrix, and gas content distribution matrix. Through iterative calculation, similar feature regions are clustered into one class. When the change in the clustering center position is less than 0.05m and the within-class variance is less than the preset threshold of 0.1, the iteration is terminated. Based on the clustering results, the horizontal slices are optimized to ensure that the coal seam characteristics within each slice are relatively uniform, and a slice thickness parameter matrix is generated, which contains the starting and ending depths of each slice. A three-dimensional coal mining grid model is established using the slice data, and the grid resolution is set to 0.2m × 0.2m × 0.2m. The model includes the coal seam geometric shape, physical parameters, and slice boundary information. The purpose of this step is to scientifically divide the coal mining slices, provide a spatial division framework for subsequent path planning, optimize the slicing scheme through multi-feature clustering, and improve the coal mining efficiency and safety.

[0051] The specific implementation of step S03 is to divide the working face of each horizontal slice using a honeycomb hexagonal grid on the basis of the established three-dimensional coal mining grid model, converting the working face space into a plane network composed of regular hexagonal cells with a side length of 0.6m. The selection of the hexagonal side length takes into account the working radius and positioning accuracy of the coal mining robot. When the positioning accuracy of the coal mining robot is higher than 0.1m, the hexagonal side length can be set to 0.6m. The Delaunay triangulation method is used to generate the hexagonal grid. First, the set of working face boundary points is determined, and then grid nodes are evenly arranged inside the working face. The node spacing is kept at about 1m. Based on these nodes, Delaunay triangles are constructed and approximated as regular hexagons. For the irregular hexagonal cells at the working face boundary, an adaptive grid adjustment algorithm is used for optimization to ensure that the area change of the grid cells at the boundary does not exceed 20% of the area of the standard hexagon. Each hexagonal grid node is assigned a unique identifier using a three-digit coding method. The first digit represents the slice number, and the last two digits represent the planar position number. An adjacent relationship table of the hexagonal grid is established to record the identifiers of the six adjacent cells of each hexagonal cell, providing topological relationship data for path planning. The purpose of this step is to establish a spatial grid system suitable for the navigation of the coal mining robot. Compared with the traditional rectangular grid, the hexagonal grid provides a more uniform direction distribution and smaller cumulative error, which is beneficial to improving the path planning accuracy.

[0052] The specific implementation of step S04 is based on the cellular hexagonal grid constructed in S03, and the Bellman-Ford algorithm is applied to design the main travel path of the coal mining robot; a weighted directed graph is constructed, where the vertices are the set of hexagonal grid nodes and the edges are the set of connecting edges between the nodes; the edge weight of each edge in the graph is defined as a comprehensive cost function, and the cost function considers multiple factors: including the thickness cost of the edge, the inclination cost of the edge, and the coal quality cost of the edge. The three are weighted and calculated according to the proportionality coefficients, and the proportionality coefficients are set to 0.4, 0.3, and 0.3 respectively, which can be dynamically adjusted according to the actual mining requirements; the thickness cost is obtained by transforming the ratio of the coal seam thickness at the corresponding position of the edge to the maximum coal seam thickness of the working face; the inclination cost is calculated by the ratio of the coal seam inclination at the corresponding position of the edge to the maximum inclination of the working face; the coal quality cost is obtained by transforming the ratio of the coal quality score at the corresponding position of the edge to the highest coal quality score of the working face; the Bellman-Ford algorithm is used to calculate the shortest path from the starting point to the ending point, and each edge is relaxed in each iteration. The total number of iterations is set to the number of grid nodes minus 1; when the shortest path is stable and there is no negative weight loop, the algorithm terminates; according to the calculation results, the coordinate sequence of the main travel path nodes of the coal mining robot is generated, and the coordinate accuracy is controlled at the centimeter level. The purpose of this step is to optimize the coal mining path based on various coal seam parameters, solve the shortest path problem of the weighted graph through the Bellman-Ford algorithm, and provide the optimal travel route for the coal mining robot.

[0053] The specific implementation of step S05 is based on the cellular hexagonal grid in S03 and the edge weight calculation method in S04 to construct a complete coal mining path navigation map; first, determine the vertex set of the navigation map. The center point of each hexagonal grid is used as a navigation vertex, and the vertex attributes include three-dimensional space coordinates, the number of the affiliated layer, and coal seam characteristic parameters; establish the connection relationship between the vertices. For the center points of adjacent hexagonal cells, bidirectional connecting edges are established, and each edge is assigned a weight value; the edge weight factor is calculated using a comprehensive evaluation method, which includes three main coefficients: the coal seam thickness coefficient, with a value range of 0.2 - 0.5. When the coal seam thickness is close to the optimal working height of the coal mining robot, which is 1m, the coefficient value is smaller; the coal seam inclination coefficient, with a value range of 0.2 - 0.5. When the coal seam inclination is close to the optimal working inclination of the coal mining robot, which is 30 degrees, the coefficient value is smaller; the coal quality coefficient, with a value range of 0.1 - 0.4. When the coal quality grade is higher, the coefficient value is smaller; the three coefficients satisfy the normalization condition, and the sum is 1; the edge weight is finally calculated by the weighted sum of the three coefficients and the corresponding characteristic standardized distance values; construct a multi-dimensional weighted graph structure, including a vertex set, an edge set, and a weight set; use an adjacency list to store the graph structure to improve the graph search efficiency; simplify the navigation map by removing the edges with weights exceeding the threshold of 3.0 to reduce the computational complexity. The purpose of this step is to establish a complete navigation map structure considering the influence of multiple factors, provide accurate spatial navigation information for the coal mining robot, and support path planning in complex environments.

[0054] The specific implementation of step S06 is to develop a one-way sectional mining strategy based on the geological data obtained in S01 and the navigation map constructed in S05. First, the working face is divided based on the coal seam dip distribution matrix. When the dip difference between adjacent areas exceeds 15 degrees, a sectional boundary is considered to be set. The sectional division is refined based on the coal seam thickness data matrix. When the thickness difference between adjacent areas exceeds 0.5 m, the sectional boundary is considered to be adjusted. Considering the location of geological faults comprehensively, the areas on both sides of the fault are compulsorily divided into different mining areas. The working face is usually divided into 3 to 5 mining areas, and the area of each mining area is kept similar, with the error controlled within 15%. The optimal coal mining direction is determined for each mining area. The principal component analysis method is used to determine the principal direction of the change in coal seam characteristics, and the coal mining direction is set to the direction perpendicular to the principal direction to ensure the minimum change in coal seam characteristics during the coal mining process. The coal mining directions of adjacent mining areas are kept as consistent as possible or the included angle is less than 45 degrees to reduce the adjustment range during the transition between mining areas. A transition area is designed at the boundary of each mining area, with the width controlled within 2 to 3 m. The coal mining paths of different mining areas are connected by a smooth curve, and the radius of curvature is not less than 1.5 times the minimum turning radius of the coal mining robot. A priority ranking of mining areas is established. Based on the coal quality distribution matrix and the gas content distribution matrix, the mining areas with high coal quality and low gas content are preferentially mined. The optimal transition path between mining areas is designed to ensure that the coal mining robot can enter the next mining area safely and efficiently after completing one mining area. The purpose of this step is to avoid the repeated movement of the coal mining robot, reduce the non-productive time, and improve the coal mining efficiency by reasonably dividing the mining areas and planning the coal mining direction.

[0055] The specific implementation of step S07 is to develop a real-time path dynamic programming strategy based on the hierarchical thickness parameter matrix generated in S02 and the set of boundary points for the division of the working face horizontal layer. The coal mining robot is equipped with a front detection radar with a detection distance set to 3 - 5 m and a scanning frequency not lower than 1 Hz to collect the information of the coal seam ahead in real time. A gas sensor array is installed with a sampling frequency not lower than 0.5 Hz and a measurement accuracy better than 0.01% to construct a local gas content distribution matrix. The sliding window method is used to update the grid data of the local area. The window size is set to 5 m × 5 m, and the window sliding step length is 1 m. Only the grid points where the difference between the detected data and the original data exceeds the threshold are updated. The thickness difference threshold is set to 0.2 m, the dip angle difference threshold is set to 5 degrees, and the gas content difference threshold is set to 0.1%. Based on the updated local grid data, the D* algorithm is used for dynamic path planning. The D* algorithm adds the ability of dynamic update on the basis of the Bellman-Ford algorithm and triggers replanning when the environmental change exceeds the preset threshold. The triggering conditions for replanning include: the difference between the detected coal seam thickness and the original data exceeds 0.3 m, the difference between the detected coal seam dip angle and the original data exceeds 10 degrees, and the detected gas content exceeds the safety threshold of 1%. The replanning process adopts a local search strategy, only recalculating the path for the affected area, with the search range limited within a radius of 10 m, and calculating a new optimal path segment to replace the corresponding part in the original path. Path smoothing is performed using the cubic spline interpolation method to control the path curvature change rate to be less than 0.2 / m² to ensure the smooth movement of the coal mining robot. The purpose of this step is to achieve the dynamic adjustment of the coal mining path, enabling the coal mining robot to respond to the changes in the coal seam in a timely manner according to the real-time detection data and ensuring the safety and adaptability of the coal mining operation.

[0056] The specific implementation of step S08 is that the coal mining robot adjusts the coal mining parameters according to multi-source coal seam data to achieve precise cutting and safe coal extraction; the coal mining robot is equipped with a laser rangefinder and an inclination sensor to measure the relative position and attitude with the coal seam in real time, the ranging accuracy is better than 0.05m, and the inclination measurement accuracy is better than 0.5 degrees; the drum height is adjusted based on the coal seam thickness data matrix, the drum height setting range is 0.6 - 1.5m, and the adjustment accuracy is 0.01m. When the measured coal seam thickness is less than 0.7m, the drum height is set to the coal seam thickness minus 0.1m to preserve the integrity of the roof and floor; the working attitude of the robot is adjusted based on the coal seam inclination distribution matrix, and the axis of the drum is kept parallel to the coal seam strike through the hydraulic support system. The inclination adaptation range is 0 - 60 degrees. When the coal seam inclination is greater than 50 degrees, an additional stable support system is started to prevent the robot from tipping over; the cutting depth is controlled based on the stratified thickness parameter matrix, the cutting depth setting range is 0.3 - 0.8m. When the working face is close to the stratified boundary, the cutting depth gradually decreases and can be adjusted to a minimum of 0.1m to ensure that the stratified boundary is not exceeded; the boundary points of the horizontal layer division of the working face are used to guide the boundary treatment. When the distance from the stratified boundary is less than 1m, the cutting speed is reduced to 50% of the normal speed to improve the boundary treatment accuracy; a real-time adjustment strategy for coal mining parameters is established, and a fuzzy control algorithm is used. The input variables are coal seam thickness, inclination, and distance from the boundary, and the output variables are drum height, cutting depth, and cutting speed. The fuzzy rule base contains 30 - 50 rules covering various coal seam conditions; a safety monitoring system is set up. When the detected gas concentration exceeds 1.5%, the cutting speed is automatically reduced, and when it exceeds 2%, the coal mining operation is suspended; a roof monitoring system is established, and acoustic detection technology is used to monitor the roof state. When the roof integrity index is lower than 0.8, the cutting depth is reduced, and when it is lower than 0.6, the coal mining in the current area is stopped. The purpose of this step is to achieve precise control of coal mining parameters, adapt to the complex conditions of steeply inclined thin coal seams, ensure coal mining quality and safety, and improve coal recovery rate.

[0057] Specifically, the principle of the present invention is: The core technical principle of the present invention's solution lies in the organic combination of geological modeling, spatial grid division, path algorithm optimization, and dynamic adjustment mechanism, constructing a complete unmanned coal mining path planning and execution system for steeply inclined thin coal seams. First, a three-dimensional geological model of the thin coal seam is constructed through geological exploration data to provide basic data support for path planning; secondly, a multi-feature clustering function is introduced to scientifically stratify the complex coal seam, and the stratification parameters are optimized according to the physical properties of the coal seam to make the stratification more in line with the actual geological characteristics.

[0058] In terms of spatial representation, a honeycomb hexagonal grid is innovatively adopted to replace the traditional rectangular grid. This structure has a constant adjacent cell distance in any direction, reducing the cumulative error of direction deviation, improving the path smoothness, and showing higher adaptability when dealing with irregular boundaries. In the design of the path algorithm, the Bellman-Ford algorithm can handle negative weight edges and is suitable for considering the change in coal mining value caused by the difference in coal quality. By integrating multi-dimensional factors such as coal seam thickness, dip angle, and coal quality, a weight function is constructed to calculate the global optimal coal mining path.

[0059] To cope with the changes in the coal seam, the present invention designs a real-time path dynamic programming mechanism. The coal mining robot collects data such as gas content through real-time detection, periodically updates the grid weights in the local area, and recalculates the optimal path. At the same time, the one-way partition mining strategy is introduced to avoid the coal mining robot moving back and forth in the working face, reducing the non-productive moving time.

[0060] The organic combination of these technical principles enables the present invention to achieve precise planning and dynamic adjustment of the coal mining path under the complex conditions of steeply inclined thin coal seams, thus solving the core technical problems of low path planning accuracy and difficulty in adapting to coal seam changes in the prior art, and having a solid foundation both theoretically and practically.

[0061] A specific embodiment 1 of the present invention is provided below. The specific implementation of each step in this embodiment 1 is described in detail as follows.

[0062] The specific implementation of step S01 is to use a high-resolution geological radar to scan the coal seam area to obtain the electromagnetic wave reflection data of the coal seam. The scanning frequency is set to 500 MHz to 2 GHz, the radar detection depth is controlled within the range of 50 to 100 m, and the scanning line spacing is maintained at 1 to 2 m. At the same time, drilling holes are arranged, the drilling hole spacing is set to 20 to 30 m, the drilling cores are collected and analyzed to extract the coal seam thickness, dip angle, and fault position information. The geological radar reflection data and the drilling core data are input into the 3D geological modeling software, and the Kriging interpolation algorithm is used to perform spatial interpolation on the coal seam parameters between the sampling points to generate a continuous 3D coal seam model. The Kriging interpolation algorithm is calculated based on the random field theory and the variogram. The calculation formula of its estimated value Z * (x0) at the position x0 is:

[0063]

[0064] In the formula, Z * (x0) is the estimated value at the position x0; Z(x i ) is the known observed value at the position x i ; λ i is the weight coefficient corresponding to the position x i ; n is the number of observed points participating in the estimation.

[0065] Weight coefficient λ i Obtained by solving the following Kriging equations:

[0066]

[0067] where, γ(x i , x j ) is the variogram value between positions x i and x j ; μ is the Lagrange multiplier; γ(x i , x0) is the variogram value between position x i and the target estimation position x0.

[0068] Based on the 3D model, extract the coal seam dip distribution matrix A with a matrix resolution of 0.5m×0.5m, record the coal seam dip values of each grid point on the working face, and mark the areas with dip angles greater than 45 degrees as steeply inclined areas; the coal seam dip distribution matrix A is expressed as:

[0069] A = [a ij m×n ;

[0070] where, a ij represents the coal seam dip value at the working face coordinates (i, j), in degrees; m and n respectively represent the number of grid points of the working face in the x and y directions.

[0071] Extract the coal seam thickness data matrix H with the same matrix resolution as the dip distribution matrix, record the coal seam thickness values of each grid point on the working face, and mark the areas with thickness less than 2m as thin coal seam areas; the coal seam thickness data matrix H is expressed as:

[0072] H = [h ij m×n ;

[0073] where, h ij represents the coal seam thickness value at the working face coordinates (i, j), in meters; m and n have the same values as those in the dip distribution matrix.

[0074] Identify the fault positions and scales, generate the fault distribution matrix F, record the distance values from each grid point to the nearest fault in the matrix, and mark the areas with distances less than 5m as fault hazard areas; the fault distribution matrix f is expressed as:

[0075] F = [f ij m×n ;

[0076] where, f ij ​​​It represents the distance value from the working face coordinate at (i, j) to the nearest fault, with the unit of meter; m and n are the same as the values in the dip distribution matrix.

[0077] Finally, the coal seam dip distribution matrix A, the coal seam thickness data matrix H, and the fault distribution matrix F are integrated into a unified geological database to provide basic data support for the subsequent path planning of coal mining robots. The purpose of this step is to construct a high-precision 3D geological model through accurate geological data acquisition and processing, providing reliable geological parameters for unmanned coal mining operations.

[0078] The specific implementation method of step S02 is based on the 3D geological model obtained in S01. It is determined that the working face is arranged along the coal seam strike direction, and the working face length is determined according to the coal seam stability, usually controlled within the range of 50 - 100 m; the stratified mining method is adopted, and the coal seam is divided into multiple horizontal layers in the vertical thickness direction, with each layer thickness set within the range of 0.8 - 1.2 m. The selection of the layer thickness threshold is based on the maximum cutting height of the coal mining robot and the requirements of roof stability; the multi-feature clustering function combining K-means clustering and fuzzy C-means clustering is used to process the coal seam feature data, and the input parameters include the coal seam thickness data matrix H, the coal seam dip distribution matrix A, the coal quality distribution matrix Q, the fault distribution matrix F, and the gas content distribution matrix G. The calculation process of the multi-feature clustering function includes the following steps: First, construct the coal seam feature vector matrix V, and each feature vector contains 5 feature components:

[0079] V = {v ij} m×n , where v ij = (h ij , a ij , q ij , f ij , g ij );

[0080] In the formula, v ij represents the feature vector at the working face coordinate (i, j); h ij , a ij , q ij , f ij , g ij respectively represent the coal seam thickness value, coal seam dip value, coal quality grade value, fault distance value, and gas content value at this position.

[0081] Secondly, perform dimensionality reduction processing on the feature vectors through the principal component analysis method, and calculate the covariance matrix C:

[0082]

[0083] In the formula, represents the mean of all feature vectors.

[0084] Solve for the eigenvalues and eigenvectors of the covariance matrix C, and select the top k eigenvectors with the largest contribution rate to form the projection matrix P. The value of k is usually taken as 2 or 3, such that the cumulative contribution rate exceeds 85%. Perform a projection transformation on the original eigenvectors to obtain the dimensionality-reduced eigen-data V':

[0085] V' = {v' ij} m×n , where v' ij = P T v ij ;

[0086] Use the K-means clustering algorithm to perform clustering analysis on the dimensionality-reduced eigen-data. The initial cluster centers are selected using the maximum distance principle, and the clustering process iteratively optimizes the objective function J:

[0087]

[0088] In the formula, K represents the number of clusters; C l represents the l-th cluster set; μ l represents the l-th cluster center; ||v' ij - μ l || represents the Euclidean distance from the eigenvector v' ij to the cluster center μ l .

[0089] When the change in the position of the cluster center is less than 0.05 m and the within-class variance is less than the preset threshold of 0.1, terminate the iteration. Based on the clustering results, optimize the horizontal stratification to ensure that the coal seam characteristics within each stratification are relatively uniform, and generate the stratification thickness parameter matrix L, which contains the starting and ending depths of each stratification:

[0090] L = [l ij p×2 ;

[0091] In the formula, l ij represents the depth value of the i-th stratification at position j. j = 1 represents the starting depth, j = 2 represents the ending depth, and the unit is meters; p represents the number of stratifications.

[0092] Use the stratified data to establish a three-dimensional coal mining grid model. The grid resolution is set to 0.2 m × 0.2 m × 0.2 m. The model contains coal seam geometry, physical parameters, and stratification boundary information. The purpose of this step is to scientifically divide the coal mining stratifications, provide a spatial division framework for subsequent path planning, optimize the stratification scheme through multi-feature clustering, and improve coal mining efficiency and safety.

[0093] ​The specific implementation of step S03 is to divide the working face of each horizontal layer by using a honeycomb hexagonal grid on the basis of the established three-dimensional coal mining grid model, and convert the working face space into a plane network composed of regular hexagonal cells with a side length of 0.6 m; the selection of the hexagonal side length takes into account the working radius and positioning accuracy of the coal mining robot. When the positioning accuracy of the coal mining robot is higher than 0.1 m, the hexagonal side length can be set to 0.6 m; the Thiessen polygon method is used to generate the hexagonal grid. First, determine the set B of working face boundary points:

[0094] B = {(x i , y i )|i = 1, 2,..., b};

[0095] In the formula, (x i , y i ) represents the coordinates of the i-th point on the working face boundary; b represents the total number of boundary points.

[0096] Uniformly arrange the set N of grid nodes inside the working face:

[0097] N = {(u j , v j )|j = 1, 2,..., n};

[0098] In the formula, (u j , v j ) represents the coordinates of the j-th grid node inside the working face; n represents the total number of internal grid nodes.

[0099] Construct Thiessen polygons based on these nodes and approximate the polygons as regular hexagons. The construction of Thiessen polygons is based on the principle of the dual graph. For each internal node (u j , v j ), its Thiessen polygon P j is defined as:

[0100]

[0101] In the formula, d((x, y), (u, v)) represents the Euclidean distance between two points (x, y) and (u, v) on the plane.

[0102] For the irregular hexagonal cells at the working face boundary, an adaptive grid adjustment algorithm is used for optimization to ensure that the area change of the grid cells at the boundary does not exceed 20% of the area of the standard hexagon. The hexagonal grid nodes are assigned a unique identifier ID ijk , and a three-digit coding method is adopted:

[0103] ID ijk = i × 10000 + j × 100 + k;

[0104] Wherein, i represents the stratification number; j and k respectively represent the row number and column number of the planar position.

[0105] Establish an adjacent relationship table R of the hexagonal grid to record the identifiers of the six adjacent cells for each hexagonal cell:

[0106] Where r i = {ID i , {ID i1 , ID i2 ,..., ID i6}};

[0107] Wherein, ID i represents the identifier of the i-th hexagonal cell; ID i1 to ID i6 represent the identifiers of the six cells adjacent to the i-th hexagonal cell.

[0108] The purpose of this step is to establish a spatial grid system suitable for the navigation of coal mining robots. Compared with the traditional rectangular grid, the hexagonal grid provides a more uniform direction distribution and a smaller cumulative error, which is beneficial to improving the path planning accuracy.

[0109] The specific implementation of step S04 is based on the honeycomb hexagonal grid constructed in S03. Apply the Bellman-Ford algorithm to design the main travel path of the coal mining robot; construct a weighted directed graph G = (V, E), where V is the set of hexagonal grid nodes and E is the set of connection edges between nodes; for each edge e = (u, v) ∈ E in the graph, define the edge weight w(e) as a comprehensive cost function, and the cost function considers multiple factors:

[0110] w(e) = α × T(e) + β × S(e) + γ × Q(e);

[0111] Wherein, T(e) represents the thickness cost of edge e, S(e) represents the dip angle cost of edge e, and Q(e) represents the coal quality cost of edge e; α, β, γ are weight coefficients, which are respectively set to 0.4, 0.3, 0.3, and can be dynamically adjusted according to actual mining requirements, and satisfy α + β + γ = 1.

[0112] The calculation formula for the thickness cost T(e) is:

[0113]

[0114] Wherein, G(e) is the coal seam thickness at the position corresponding to edge e, with the unit of meter; H max is the maximum coal seam thickness of the working face, with the unit of meter.

[0115] The calculation formula for the dip angle cost S(e) is:

[0116]

[0117] In the formula, A(e) is the seam dip angle at the position corresponding to edge e, with the unit of degree; A max is the maximum dip angle of the working face, with the unit of degree.

[0118] The calculation formula for the coal quality cost Q(e) is:

[0119]

[0120] In the formula, C(e) is the coal quality score at the position corresponding to edge e, dimensionless; C max is the highest coal quality score of the working face, dimensionless.

[0121] The Bellman-Ford algorithm calculates the shortest paths from the starting point s to all other nodes through relaxation operations. Initialize the distance array D:

[0122] D[s] = 0, D[v] = ∞ for all v ∈ V, v ≠ s;

[0123] Execute |V|-1 iterations, and perform relaxation operations on all edges in each iteration:

[0124] D[v] = min(D[v], D[u] + w(u, v)) for all edges (u, v) ∈ E;

[0125] Check whether there is a negative weight cycle:

[0126] For all edges (u, v) ∈ E, if D[v] > D[u] + w(u, v), then there is a negative weight cycle.

[0127] When the shortest path is stable and there is no negative weight cycle, the algorithm terminates; according to the calculation results, generate the coordinate sequence P of the main travel path of the coal mining robot:

[0128] P = {(x1, y1), (x2, y2),..., (x n , y n )};

[0129] In the formula, (x i , y i ) represents the coordinates of the i-th node on the path, and the coordinate accuracy is controlled at the centimeter level.

[0130] The purpose of this step is to optimize the coal mining path based on various coal seam parameters, solve the shortest path problem of the weighted graph through the Bellman-Ford algorithm, and provide the optimal travel route for the coal mining robot.

[0131] The specific implementation of step S05 is to construct a complete coal mining path navigation map based on the honeycomb hexagonal grid of S03 and the edge weight calculation method of S04; first, determine the vertex set V of the navigation map, with the center point of each hexagonal grid as a navigation vertex, and the vertex attributes include three-dimensional space coordinates, the number of the layer to which it belongs, and coal seam characteristic parameters; establish the connection relationship between vertices. For the center points of adjacent hexagonal cells, establish two-way connection edges, and assign a weight value to each edge; the edge weight factor is calculated using a comprehensive evaluation method, which includes three main coefficients: the coal seam thickness coefficient W h , with a value range of 0.2 to 0.5. When the coal seam thickness is close to the optimal working height of 1 m of the coal mining robot, the coefficient value is smaller; the coal seam dip coefficient W a , with a value range of 0.2 to 0.5. When the coal seam dip is close to the optimal working dip of 30 degrees of the coal mining robot, the coefficient value is smaller; the coal quality coefficient W q , with a value range of 0.1 to 0.4. When the coal quality grade is higher, the coefficient value is smaller; the three coefficients satisfy the normalization condition:

[0132] W h +W a +W q =1;

[0133] The final calculation formula for the edge weight is:

[0134] W=W h ×D h +W a ×D a +W q ×D q ;

[0135] In the formula, D h , D a , D q are the normalized distance values of the coal seam thickness, dip, and coal quality respectively, and the calculation formulas are as follows:

[0136]

[0137] In the formula, h is the coal seam thickness at the current position, in meters; h opt is the optimal working height of the coal mining robot, usually 1 m; h max and h min are the maximum and minimum coal seam thicknesses of the working face, in meters respectively.

[0138]

[0139] In the formula, a is the coal seam dip at the current position, in degrees; a opt is the optimal working dip of the coal mining robot, usually 30 degrees; a max and a minare the maximum and minimum seam dip angles of the working face, in degrees.

[0140]

[0141] In the formula, q is the coal quality grade value at the current position, dimensionless; q max and q min are the highest and lowest coal quality grade values of the working face, respectively, dimensionless.

[0142] Construct a multi-dimensional weighted graph structure G=(V, E, W), where V is the vertex set, E is the edge set, and W is the weight set; use an adjacency list to store the graph structure to improve the graph search efficiency; simplify the navigation graph by removing edges with weights exceeding the threshold of 3.0 to reduce the computational complexity. The purpose of this step is to establish a complete navigation graph structure considering multiple factors, provide accurate spatial navigation information for the coal mining robot, and support path planning in complex environments.

[0143] The specific implementation of step S06 is based on the geological data obtained in S01 and the navigation graph constructed in S05 to develop a one-way partition mining strategy; first, divide the working face based on the seam dip angle distribution matrix A, and consider setting partition boundaries when the dip angle difference between adjacent areas exceeds 15 degrees; refine the partition based on the coal seam thickness data matrix H, and consider adjusting the partition boundary when the thickness difference between adjacent areas exceeds 0.5 m; comprehensively consider the location of geological faults, and force the areas on both sides of the fault to be divided into different mining areas; the working face is usually divided into 3-5 mining areas, and the area of each mining area is kept similar, with the error controlled within 15%; determine the optimal coal mining direction for each mining area, use the principal component analysis method to determine the main direction of the change in coal seam characteristics, and set the coal mining direction to the direction perpendicular to the main direction to ensure the minimum change in coal seam characteristics during the coal mining process; the coal mining directions of adjacent mining areas should be kept as consistent as possible or the included angle should be less than 45 degrees to reduce the adjustment range during the transition between mining areas; design a transition area at the boundary of each mining area, with the width controlled within 2-3 m, and use a smooth curve to connect the coal mining paths of different mining areas, and the radius of curvature is not less than 1.5 times the minimum turning radius of the coal mining robot; establish a priority ranking for mining areas, based on the coal quality distribution matrix Q and the gas content distribution matrix G, and give priority to mining areas with high coal quality and low gas content; design the optimal transition path between mining areas to ensure that the coal mining robot can safely and efficiently enter the next mining area after completing one mining area. The purpose of this step is to avoid repeated movement of the coal mining robot, reduce non-productive time, and improve coal mining efficiency by reasonably dividing mining areas and planning coal mining directions.

[0144] The specific implementation of step S07 is to develop a real-time path dynamic programming strategy based on the layered thickness parameter matrix L generated in S02 and the set B of the division boundary points of the working face horizontal layer. The coal mining robot is equipped with a front detection radar, the detection distance is set to 3 - 5 m, the scanning frequency is not less than 1 Hz, and the coal seam information in front is collected in real time. A gas sensor array is installed, the sampling frequency is not less than 0.5 Hz, and the measurement accuracy is better than 0.01%, to construct a local gas content distribution matrix G′:

[0145] G′ = [g′ ij m′×n′ ;

[0146] In the formula, g′ ij represents the gas content value at the local area coordinates (i, j), and the unit is percentage; m′ and n′ respectively represent the number of grid cells of the local area in the x and y directions.

[0147] The sliding window method is used to update the local area grid data. The window size is set to 5 m × 5 m, and the window sliding step length is 1 m. Only the grid points where the difference between the detected data and the original data exceeds the threshold are updated. The thickness difference threshold is set to 0.2 m, the dip angle difference threshold is set to 5 degrees, and the gas content difference threshold is set to 0.1%. Based on the updated local grid data, the D* algorithm is used for dynamic path planning. The D* algorithm adds dynamic update ability on the basis of the Bellman-Ford algorithm, and triggers replanning when the environment changes exceed the preset threshold. The replanning trigger conditions include: the difference between the detected coal seam thickness and the original data exceeds 0.3 m, the difference between the detected coal seam dip angle and the original data exceeds 10 degrees, and the detected gas content exceeds the safety threshold of 1%. The replanning process adopts a local search strategy, and only the affected area is recalculated for the path. The search range is limited within a radius of 10 m, and the corresponding part in the original path is replaced by calculating a new optimal path segment. Path smoothing processing is carried out using the cubic spline interpolation method to control the path curvature change rate to be less than 0.2 / m 2 , ensuring the smooth movement of the coal mining robot. The purpose of this step is to realize the dynamic adjustment of the coal mining path, so that the coal mining robot can respond to the coal seam changes in time according to the real-time detection data, and ensure the safety and adaptability of the coal mining operation.

[0148] The specific implementation of step S08 is that the coal mining robot adjusts the coal mining parameters according to the multi-source coal seam data to achieve precise cutting and safe coal extraction. The coal mining robot is equipped with a laser rangefinder and an inclination sensor to measure the relative position and attitude with the coal seam in real time. The ranging accuracy is better than 0.05 m, and the inclination measurement accuracy is better than 0.5 degrees. Based on the coal seam thickness data matrix H, the drum height h r is adjusted. The drum height setting range is 0.6 - 1.5 m, and the adjustment accuracy is 0.01 m. When the measured coal seam thickness h mea ​When it is less than 0.7 m, the drum height is set to the coal seam thickness minus 0.1 m:

[0149]

[0150] In the formula, h r is the set value of the drum height, in meters; h mea is the measured coal seam thickness, in meters.

[0151] Based on the coal seam dip distribution matrix A, adjust the working attitude angle θ of the robot. Keep the drum axis parallel to the coal seam strike through the hydraulic support system. The dip adaptation range is 0 to 60 degrees. When the coal seam dip a mea is greater than 50 degrees, start the additional stable support system to prevent the robot from tipping over:

[0152]

[0153] In the formula, θ is the set value of the working attitude angle of the robot, in degrees; a mea is the measured coal seam dip, in degrees; δ is the additional stable angle compensation value, usually set to 5 to 10 degrees.

[0154] Control the cutting depth d based on the layered thickness parameter matrix L c , the cutting depth setting range is 0.3 to 0.8 m. When the working face is close to the layered boundary d b is close to the layered boundary, the cutting depth gradually decreases, and the minimum can be adjusted to 0.1 m to ensure that it does not exceed the layered boundary:

[0155]

[0156] In the formula, d c is the set value of the cutting depth, in meters; d max is the maximum cutting depth, usually set to 0.8 m; d b is the distance from the layered boundary, in meters; λ is the adjustment coefficient, usually set to 2.0.

[0157] Adopt the working face horizontal layer division boundary point set B to guide the boundary treatment. When the distance from the layered boundary is less than 1 m, the cutting speed v c is reduced to 50% of the normal speed to improve the boundary treatment accuracy:

[0158]

[0159] In the formula, v c is the set value of the cutting speed, in m / min; v max is the maximum cutting speed, usually set to 3 m / min; d b is the distance from the layered boundary, in meters.

[0160] Establish a real-time adjustment strategy for coal mining parameters. Adopt the fuzzy control algorithm. The input variables are coal seam thickness, dip angle, and distance from the boundary. The output variables are drum height, cutting depth, and cutting speed. The fuzzy rule base contains 30 - 50 rules, covering various coal seam conditions. The fuzzy control algorithm is based on the following fuzzy inference process:

[0161] First, perform fuzzification. Convert the input variables x (coal seam thickness, coal seam dip angle, distance from the boundary) into fuzzy sets:

[0162]

[0163] In the formula, μ A (x) is the membership function value of the input variable x corresponding to the fuzzy set A; a, b, c are membership function parameters, determined according to the actual coal seam conditions.

[0164] Then, based on the fuzzy rule base, perform fuzzy inference. The fuzzy rule adopts the Mamdani inference model. The general form of rule R i is:

[0165] If x1 is A1 and x2 is A2 and x3 is A3, then y is B;

[0166] Among them, x1, x2, x3 respectively represent coal seam thickness, coal seam dip angle, and distance from the boundary; A1, A2, A3 respectively represent the corresponding fuzzy sets; y represents the output variable (drum height, cutting depth, or cutting speed); B represents the fuzzy set corresponding to the output variable.

[0167] Finally, perform defuzzification. Use the centroid method to calculate the precise output value y * :

[0168]

[0169] In the formula, y * is the precise output value after defuzzification; y i is the i-th discrete point of the output variable; μ B (y i ) is the membership function value corresponding to this point; n is the number of discrete points.

[0170] Set up a safety monitoring system. When the detected gas concentration g exceeds 1.5%, automatically reduce the cutting speed. When it exceeds 2%, suspend the coal mining operation:

[0171]

[0172] In the formula, v c is the cutting speed setting value, with the unit of m / min; v maxis the maximum cutting speed, usually set to 3 m / min; v normal is the normal cutting speed, calculated by the aforementioned fuzzy control algorithm; g is the gas concentration, in percentage.

[0173] Establish a roof monitoring system, use acoustic wave detection technology to monitor the roof state. When the roof integrity index I r is lower than 0.8, reduce the cutting depth. When it is lower than 0.6, stop coal mining in the current area:

[0174]

[0175] In the formula, d c is the set value of the cutting depth, in meters; d normal is the normal cutting depth, calculated by the aforementioned fuzzy control algorithm; I r is the roof integrity index, dimensionless, with a value range of 0 to 1, calculated from the reflection wave characteristics obtained by acoustic wave detection:

[0176]

[0177] In the formula, f i is the score of the i-th acoustic wave characteristic parameter, with a value range of 0 to 1; w i is the weight coefficient of the i-th characteristic parameter; k is the number of characteristic parameters, usually taking 3 to 5.

[0178] The purpose of this step is to achieve precise regulation of coal mining parameters, adapt to the complex conditions of steeply inclined thin coal seams, ensure coal mining quality and safety, and improve coal recovery rate.

[0179] To better understand and implement the present invention, Example 2 of a specific application scenario of the present invention is provided below: Workers implemented the unmanned coal mining technology for steeply inclined thin coal seams in a certain mining area. The average dip angle of the coal seams in this mining area is 52 degrees, and the coal seam thickness ranges from 0.9 to 1.8 m, belonging to typical steeply inclined thin coal seams. Due to the complex geological conditions in this mining area, the traditional manual coal mining method has problems such as high safety risks and low recovery rates, and there is an urgent need to introduce unmanned coal mining technology to improve safety and coal mining efficiency.

[0180] First, the workers scanned the target working face with a high-resolution geological radar with a frequency of 1.2 GHz. The scanning line spacing was 1.5 m, and the detection depth was 75 m. At the same time, 25 drilling holes were arranged, with a drilling hole spacing of 25 m, and the cores were analyzed in detail. A three-dimensional geological model of the coal seam was constructed through the Kriging interpolation algorithm. This model accurately describes the geometric shape and physical properties of the coal seam, with a resolution of 0.5 m × 0.5 m. According to the model data, the relevant feature matrix was extracted, as shown in Table 1:

[0181] Table 1 Basic Parameters of Coal Seam Feature Matrix

[0182]

[0183]

[0184] Based on the acquired geological data, the staff used the slicing mining method to divide the coal seam into multiple horizontal slices. First, the multi-feature clustering function was used to analyze the coal seam characteristics. The dimension of the feature vector was 5, including coal seam thickness, dip angle, coal quality grade, fault distance, and gas content. Through principal component analysis for dimensionality reduction, 2 principal components were extracted, and the cumulative contribution rate was 87.3%. The K-means clustering algorithm was used to cluster the feature data. After 16 iterations, it converged, and the within-class variance dropped to 0.08. According to the clustering results, the coal seam was divided into 6 horizontal slices in the vertical thickness direction, and the average slice thickness was 1.05m. As Figure 2 shown, the green dashed line in the figure is the planned robot travel path. The structure of the coal mining robot used is as Figure 3 shown.

[0185] Table 2 Horizontal Slice Parameter Table

[0186]

[0187] Subsequently, the staff used a honeycomb hexagonal grid to divide each horizontal slice, and the grid side length was set to 0.6m. The Thiessen polygon method was used to generate the hexagonal grid, and a total of 2784 hexagonal units were divided on the working face. As Figure 4 shown. Each hexagonal unit was assigned a unique identifier, using a three-digit coding method. The first digit represents the slice number (1 - 6), and the last two digits represent the planar position number (01 - 99). An adjacency relation table was constructed based on the hexagonal grid to record the connection relationship between each unit and the adjacent six units, providing topological data support for path planning.

[0188] In terms of path planning, the staff applied the Bellman-Ford algorithm to design the main travel path of the coal mining robot. First, a weighted directed graph was constructed, with the number of vertices being 2784 and the number of edges being 16704. For each edge, the weight value was calculated according to the comprehensive cost function. The weight coefficients of the thickness cost, dip angle cost, and coal quality cost were set to 0.4, 0.3, and 0.3 respectively. The shortest path from the starting point to the ending point was calculated through the Bellman-Ford algorithm. After 2783 iterations, the algorithm converged, and a coordinate sequence of the main travel path of the coal mining robot was generated, including 387 path nodes.

[0189] Table 3 Path Planning Parameter Table

[0190] Parameter Name Parameter Value Unit Description Weight Coefficient α of Comprehensive Cost Function 0.4 Dimensionless Weight of Thickness Cost Weight Coefficient β of Comprehensive Cost Function 0.3 Dimensionless Weight of Dip Angle Cost Weight Coefficient γ of Comprehensive Cost Function 0.3 Dimensionless Weight of Coal Quality Cost Maximum Coal Seam Thickness of Working Face 1.8 m <![CDATA[H max value]]> Maximum Dip Angle of Working Face 58 Degree <![CDATA[A max value]]> Highest Coal Quality Score of Working Face 5 Dimensionless <![CDATA[C max value]]> Number of Path Nodes 387 pcs Total Number of Optimal Path Nodes Path Length 232.2 m Total Length of Optimal Path

[0191] Based on the hexagonal grid and edge weight calculation method, the staff constructed a complete coal mining path navigation map. In the navigation map, the edge weight calculation adopts the coal seam thickness coefficient, dip angle coefficient, and coal quality coefficient, and their value ranges are 0.2 - 0.5, 0.2 - 0.5, and 0.1 - 0.4 respectively. The optimal working height of the coal mining robot is set at 1.0 m, and the optimal working dip angle is set at 30 degrees. The navigation map is simplified by removing the edges with weights exceeding 2.8 to reduce the computational complexity.

[0192] Subsequently, based on the obtained geological data and the constructed navigation map, the staff developed a one-way partition mining strategy. The working face is divided into 4 mining areas, and the area of each mining area is about 125 m 2 , and the area error is controlled within 10%. The optimal coal mining direction is determined for each mining area, the width of the transition area between mining areas is set at 2.5 m, and a smooth curve with a curvature radius of 3.6 m is used to connect the coal mining paths of different mining areas. According to the coal quality distribution matrix and the gas content distribution matrix, the mining areas are mined in the priority order shown in Table 2.

[0193] During actual coal mining operations, the coal mining robot is equipped with a forward detection radar, with the detection distance set at 4 m and the scanning frequency at 1.5 Hz. The sampling frequency of the gas sensor array is 0.8 Hz, and the measurement accuracy is 0.005%. The sliding window method is used to update the local area grid data, with the window size of 5 m × 5 m and the sliding step of 1 m. The thickness difference threshold is set at 0.2 m, the dip angle difference threshold is set at 5 degrees, and the gas content difference threshold is set at 0.1%. When the detected environmental changes exceed the preset threshold, the D-star algorithm is triggered for dynamic path replanning, and the search range is limited within a radius of 10 m.

[0194] Table 4 Parameters of the Coal Mining Robot

[0195] Parameter Type Parameter Value Unit Description Setting Range of Drum Height 0.6～1.5 m Can be Adjusted According to Coal Seam Thickness Setting Range of Cutting Depth 0.3～0.8 m Can be Adjusted According to Laminated Boundary Dip Angle Adaptation Range 0～60 Degree Support High Dip Angle Working Laser Ranging Accuracy 0.03 m Higher than the Required 0.05 m Dip Angle Measurement Accuracy 0.3 Degree Higher than the Required 0.5 Degree Forward Detection Distance 4 m Real-time Coal Seam Detection Gas Detection Accuracy 0.005 ％ High-precision Safety Monitoring

[0196] The coal mining robot adjusts the coal mining parameters according to multi-source coal seam data to achieve precise cutting and safe mining. The drum height is automatically adjusted according to the coal seam thickness. When the coal seam thickness is less than 0.7 m, the drum height is set at the coal seam thickness minus 0.1 m. The working posture of the robot is adjusted according to the coal seam dip angle. When the dip angle is greater than 50 degrees, an additional stable support system is activated to prevent tipping. The cutting depth is dynamically adjusted according to the distance from the stratification boundary. When the distance from the boundary is less than 1 m, the cutting speed is reduced to 50% of the normal speed. The safety monitoring system monitors the gas concentration in real time. When the gas concentration exceeds 1.5%, the cutting speed is automatically reduced, and when it exceeds 2%, the coal mining operation is suspended.

[0197] Through the unmanned coal mining method of this embodiment, the staff has successfully achieved the safe and efficient mining of steeply inclined thin coal seams. Compared with the traditional coal mining method, the present invention has significant advantages. The traditional mining of steeply inclined thin coal seams mainly relies on manual top coal caving or simple mechanized mining, which has problems such as high safety risks, high labor intensity, and low coal recovery rate. Workers need to operate in a high dip angle environment, and safety accidents occur frequently; roof management is difficult, and the risk of collapse is high; the coal mining efficiency is low, and resource waste is serious.

[0198] The unmanned coal mining method of the present invention completely solves these problems. By adopting the stratified mining method and the honeycomb hexagonal grid planning, the coal recovery rate has been increased from 65% of the traditional method to more than 85%; by adopting intelligent path planning and real-time dynamic adjustment, the coal mining efficiency has been increased by 2.3 times; coal mining robots replace manual operations, eliminating safety accidents; precise parameter control has significantly improved the coal mining quality, the size of coal blocks after mining is uniform, and the generation of fine coal powder is reduced. After implementing the present invention, the coal mining cost of this mining area has been reduced by 37%, the incidence of safety accidents has dropped to zero, and the comprehensive economic benefits have been significantly improved.

[0199] It should be noted that the detailed explanations of the variables involved in the present invention are shown in Tables 5 and 6 below.

[0200] Table 5 Variable Explanation Table (First Part)

[0201]

[0202]

[0203] Table 6 Variable Explanation Table (Second Part)

[0204]

[0205]

[0206] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A method for unmanned coal mining in steeply inclined thin coal seams, characterized in that: include: Construct a three-dimensional geological model of thin coal seams to determine the coal seam dip distribution matrix, coal seam thickness data matrix and fault distribution matrix; Divide thin coal seams into horizontal layers and establish a three-dimensional coal mining grid model; use honeycomb hexagonal grids to divide the working face to form a basic network for coal mining path calculation; design the main travel path of the coal mining robot based on the Bellman Ford algorithm, and calculate the optimal coal mining path by comprehensively considering the coal seam thickness data matrix, coal seam inclination distribution matrix and coal quality distribution matrix; construct a coal mining path navigation map to form a multi-dimensional weighted graph structure; develop a one-way partitioned mining strategy; Realize the dynamic path planning and mining parameter adjustment of the coal mining robot.

2. The method for unmanned coal mining in steeply inclined thin coal seams according to claim 1, characterized in that: The steps of dividing the thin coal seam into horizontal layers and establishing a three-dimensional coal mining grid model are specifically to design a working face arrangement along the strike, adopt a layered mining method, divide the steeply inclined thin coal seam into multiple horizontal layers according to thickness, and each layer is 0.8 to 1.2 meters thick. The coal seam is layered and optimized using a multi-feature clustering function to establish a three-dimensional coal mining grid model.

3. The method for unmanned coal mining in steeply inclined thin coal seams according to claim 2, characterized in that: The step of using a honeycomb hexagonal grid to divide the working face to form a basic network for coal mining path calculation is specifically to divide the working face into multiple regular hexagonal units with a side length of 0.6 meters to form a basic network for coal mining path calculation, thereby improving the path planning accuracy; the honeycomb hexagonal grid is specifically a spatial representation method that divides the coal mining working face space into a set of regular regular hexagonal units with a side length of 0.6 meters, each hexagonal unit shares a boundary with the six adjacent units around it, and compared with the traditional rectangular grid, the hexagonal grid has a constant adjacent unit distance in any direction, thereby reducing the cumulative error of directional deviation and improving the path smoothness.

4. The method for unmanned coal mining in steeply inclined thin coal seams according to claim 3 is characterized in that: The steps of developing the one-way partitioned mining strategy are specifically to divide the working face into multiple mining areas, set a one-way travel path for each mining area, avoid the coal mining robot repeatedly passing through the same area, and improve the coal mining efficiency.

5. The method for unmanned coal mining in steeply inclined thin coal seams according to claim 4, characterized in that: The steps for realizing real-time path dynamic planning are specifically based on the layer thickness parameter matrix and the boundary point set of the horizontal layer division of the working face to realize real-time path dynamic planning. The coal mining robot collects the gas content distribution matrix according to the real-time detection data of the coal seam, periodically updates the local area grid weight, and recalculates the optimal path to adapt to the changes in the coal seam.

6. The method for unmanned coal mining in steeply inclined thin coal seams according to claim 5, characterized in that: The steps of the coal mining robot adjusting coal mining parameters to achieve precise cutting and safe and efficient mining are as follows: the coal mining robot adjusts the coal mining parameters according to the coal seam thickness data matrix, the coal seam inclination distribution matrix, the layered thickness parameter matrix and the set of boundary points of the horizontal layer division of the working face, controls the drum height and the cutting depth, and achieves precise cutting and safe and efficient mining of steeply inclined thin coal seams.

7. The method for unmanned coal mining in steeply inclined thin coal seams according to claim 6, characterized in that: The layered mining method specifically divides the steeply inclined thin coal seam into multiple parallel layers with a thickness of 0.8 to 1.2 meters in the horizontal direction. The coal mining robot completes the coal mining operation independently in each parallel layer. After completing the coal mining of one parallel layer, the coal mining robot enters the next parallel layer. It is suitable for steeply inclined coal seams with an inclination angle greater than 45 degrees. By controlling the range of roof movement by layering, the risk of roof collapse is reduced and the coal recovery rate is improved.

8. The method for unmanned coal mining in steeply inclined thin coal seams according to claim 7, characterized in that: The Bellman-Ford algorithm is specifically a graph algorithm for finding the single-source shortest path. The shortest path from the starting point to all vertices is calculated by performing multiple rounds of relaxation operations on all edges in the graph. In the weighted graph structure designed for coal mining path planning, the edge weight integrates the coal seam thickness coefficient, coal seam inclination coefficient and coal quality coefficient, and realizes the coal mining path with optimal resource utilization by minimizing the comprehensive cost.

9. The method for unmanned coal mining in steeply inclined thin coal seams according to claim 8, characterized in that: The multi-feature clustering function is used to scientifically divide the coal mining working face according to multiple physical properties of the coal seam. The input includes the coal seam thickness data matrix, the coal seam inclination distribution matrix, the coal quality distribution matrix, the fault distribution matrix and the gas content distribution matrix. The output is the layered thickness parameter matrix and the working face horizontal layer division boundary point set.

10. The method for unmanned coal mining in steeply inclined thin coal seams according to claim 9, characterized in that: The one-way zoning mining strategy is specifically a coal mining operation method designed for the characteristics of steeply inclined thin coal seams. A complete working face is divided into multiple mining areas with similar areas according to the change characteristics of the coal seam thickness data matrix and the coal seam inclination distribution matrix. A fixed coal mining direction is set in each mining area. The coal mining robot completes coal mining in a single direction in a mining area and then enters the next mining area, avoiding the coal mining robot from moving back and forth in the working face and reducing non-productive movement time.

Images