Coal seam roof stability prediction method based on multiple structural factors

Abstract

The invention relates to the technical field of geological exploration and coal mine safety mining, in particular to a coal seam roof stability prediction method based on multiple structural factors, which comprises the following steps of: 1, determining structural factors influencing the stability of a coal seam roof and quantitative prediction indexes of each factor, the influence factors comprise roof lithology, a rock mass structure, a geological structure, ground stress and hydrogeology; 2, calculating a prediction index quantized value; 3, generating a single-index grid network by using Surfer software; 4, determining subjective and objective weights of each prediction index; 5, coupling the subjective weight and the objective weight to form a corrected weight; and 6, performing weighted superposition on each index by using ArcGIS software, and performing partition prediction on the stability of the coal seam roof. According to the method, the stability of the coal seam roof can be accurately predicted in a zoning mode, the situation that the coal seam roof is not excavated and then protected is achieved, the accident rate of the roof is reduced, casualties and economic losses are reduced, and technical support is provided for optimizing roof management measures.

Description

Technical Field

[0001] This invention relates to a method for predicting the stability of coal seam roof by zoning, and more particularly to a method for predicting the stability of coal seam roof by regional zoning based on multiple structural factors, belonging to the fields of geological exploration and safe coal mining technology. Background Technology

[0002] my country's coal mining is gradually transitioning to a deeper development stage at an average annual rate of 10-25 meters. As the mining depth continues to increase, the condition of the coal seam roof becomes increasingly complex. Affected by multiple effects such as high temperature, high ground stress, complex geological structure, and intense mining disturbance, roadways often experience deformation and damage such as roof falls, sidewall spalling, and floor heave during the mining process. In particular, when the tectonic stress on the coal seam roof exceeds its ultimate load, it is extremely easy to induce disasters such as rock bursts. Roof accidents account for 31.7% of coal mine safety accidents in the past decade or so, ranking first among all coal mine accidents, seriously affecting the safe mining of deep coal resources.

[0003] Common methods for predicting coal seam roof stability include: the mathematical method used by Jing Hongwen in "Prediction Model and Application of Deep Roadway Surrounding Rock Deformation Based on Fuzzy Grey System"; the artificial intelligence prediction used by Yu Bingbing in "Prediction Model of Roadway Roof Unsettlement Based on SSA-SVM"; and the geological evaluation used by Liu Qinfu in "Study on Geological Characteristics and Stability of Coal Seam Roof of Huainan Coalfield 13-1". Mathematical methods are too objective, leading to incomplete predictions due to the complex geological environment of the coal seam roof, neglecting the analysis of complex geological structural elements. Artificial intelligence methods are complex to operate, theoretically difficult, and have low field applicability. Traditional geological evaluation methods rely too heavily on expert experience, are highly subjective, lack quantitative analysis, and fail to accurately reflect the complexity of actual geological conditions. Therefore, there is an urgent need for a regional prediction method for coal seam roof stability based on quantitative analysis of multiple structural factors and integrating subjective and objective evaluation mechanisms to improve the accuracy and reliability of predictions. Summary of the Invention

[0004] To address the problems existing in current prediction methods, this invention proposes a coal seam roof stability prediction method based on multiple structural factors. This method combines subjective and objective approaches, avoiding the shortcomings of objective methods in insufficient analysis of the actual geological environment, and also avoiding the problem of subjective methods relying too heavily on expert experience. By comprehensively considering the influence of four types of factors on roof stability—geological structure, roof lithology and rock mass structure, geostress, and hydrogeology—a completely new roof stability prediction index system is established. Furthermore, Surfer and ArcGIS software are used to simplify the prediction process and make the prediction results more accurate.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] (1) Identify the tectonic physical factors affecting the stability of the coal seam roof and establish a set of predictive index influencing factors:

[0007] a. Geological structural elements group: Define the fault similarity dimension D s and the deformation coefficient K of the folded plane p To construct indicators;

[0008] b. Lithology and Rock Mass Structure Elements: Defining the Sand-Mud Ratio K of the Roof r The lithological combination L of the roof and the modified rock mass quality coefficient Z are lithological and rock mass structure indicators.

[0009] c. Geostress Element Group: Defines the maximum principal stress value S in the area where the coal seam roof is located as a geostress index.

[0010] d. Hydrogeological element group: Define the unit inflow rate q of the roof as a hydrological index;

[0011] (2) Analyze all influencing factors and calculate the quantitative values ​​of the prediction indicators:

[0012] a. Fault similarity dimension D s Fault quantization is performed using the similarity dimension from fractal theory. The specific steps are as follows: A variable-length grid is used for coverage. An initial cover layer is established with a square grid of reference side length r = 1000m. The grid is then progressively refined at levels of 500×500m and 250×250m. The number of cells containing fault traces, N(r), corresponding to each grid level is counted. A regression line is then plotted in a coordinate system with lgN(r) as the ordinate and lgr as the abscissa. The absolute value of the slope is the similarity dimension, calculated using the following formula:

[0013]

[0014] In the formula, N(r) is the number of units and r is the side length.

[0015] b. Coefficient of deformation of the folded plane K p The sliding window method is used for statistical analysis, and the calculation formula is as follows:

[0016]

[0017] In the formula, h is the elevation difference between two adjacent contour lines, m; L0 is the horizontal distance between two adjacent contour lines, m; L1 is the actual length of the contour line closest to the unit point, m; and L2 is the secant length of the contour line closest to the unit point, m.

[0018] c. Top slab sand-mud ratio K r The ratio of the thickness of sandstone strata to mudstone strata in the top plate is calculated using the following formula:

[0019]

[0020] In the formula, H s Thickness of sandstone strata, in meters; H r The thickness of the mudstone strata is in meters (m).

[0021] d. The top lithological assemblage L is divided into four types: the whole is composed of fine-grained sandy mudstone and mudstone—the completely soft type; the whole is composed of thin layers of mudstone and sandstone interbedded—the soft-hard alternating type; the strata grain size becomes finer upwards—the lower hard and upper soft type; and the strata grain size becomes coarser upwards—the lower soft and upper hard type.

[0022] e. The corrected rock mass quality coefficient Z for the roof is determined using AHP hierarchical analysis to determine the weights of the three indicators—rock mass integrity, structural plane complexity, and rock mass strength—on the stability of the roof rock mass structure. The corrected rock mass quality coefficient Z is then introduced as an indicator to quantitatively describe the strength of the coal seam roof rock mass structure. The calculation formula is as follows:

[0023]

[0024] In the formula, a is the rock mass integrity weight; C is the core recovery rate; b is the structural surface complexity weight; f is the structural surface complexity; c is the rock mass strength weight; and K is the rock mass strength.

[0025] f. Maximum principal stress value S: Collect measured geostress data for the studied area, and use the tectonic analytical method to correct the measured values ​​of vertical principal stress, maximum horizontal principal stress, and minimum horizontal principal stress. The specific calculation formula for the tectonic analytical method is as follows:

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032] In the formula, σ v For vertical principal stress, MPa; σ H The maximum horizontal principal stress is σ, MPa; hσ1, σ2, and σ3 are the measured ground stress values, in MPa; l1, l2, and l3 are the products of the dip cosine and azimuth sine corresponding to the measured ground stress values; m1, m2, and m3 are the products of the dip cosine and azimuth sine corresponding to the measured ground stress values; n1, n2, and n3 are the dip cosine values ​​corresponding to the measured ground stress values.

[0033] Linear regression was used to fit the linear relationship between the vertical principal stress, the maximum horizontal principal stress, the minimum horizontal principal stress and the burial depth in the study area, and then the magnitude of the principal stress was calculated back using the burial depth of the top plate.

[0034] g. Unit inflow rate q: The unit inflow rate is determined on-site using a pumping test, i.e., pumping water from the borehole and observing the inflow rate when the water level in the well drops by one meter. The calculation formula is:

[0035]

[0036] In the formula, Q is the flow rate during pumping, in L / s; S is the drawdown, in m.

[0037] (3) Use Surfer software to generate a single-index partitioned raster network:

[0038] First, use Excel to complete the basic data editing, where X and Y are the spatial locations of the data points (geodetic coordinate system), and Z is the quantitative value of each prediction index; then import the basic data into Surfer for gridding, and use the Kriging interpolation method for calculation; finally, whiten the part outside the boundary of the study area to generate the raster network of each index.

[0039] (4) Determine the subjective and objective weights of each forecast indicator:

[0040] a. Subjective weights: Subjective weights are determined using the Analytic Hierarchy Process (AHP). The specific steps are as follows:

[0041] First, a hierarchical model is established: the highest layer is set as the stability of the coal seam roof; the basic layers are set as various indicators affecting the stability of the coal seam, namely the roof sand-slurry ratio K. r Top lithological combination L, corrected rock mass quality coefficient Z, fault similarity dimension D s , Coefficient of deformation of the folded plane K p A total of 7 items were included, including the ground stress value S and the unit inflow rate q of the roof.

[0042] Secondly, construct the judgment matrix:

[0043]

[0044] In the formula, a ij The meaning is the degree of importance of the i-th indicator compared to the j-th indicator, for example, a 12This indicates the relative importance of the first indicator compared to the second indicator. ij The scoring criteria are as follows: 1 for two equally important indicators, 3 for indicator i being slightly more important than indicator j, 5 for indicator i being significantly more important than indicator j, 7 for indicator i being strongly more important than indicator j, and 9 for indicator i being extremely more important than indicator j. 2, 4, 6, and 8 are the values ​​when the above scale is directly compromised. The score can be any non-negative number.

[0045] Then, the root method is used to solve for the eigenvectors of the judgment matrix, i.e., the subjective weights of the indicators:

[0046] Step 1: Calculate the nth root of the product of the elements in each row of the judgment matrix A. The formula is as follows:

[0047]

[0048] Step 2: Move M i The formula for normalizing the ratio is as follows:

[0049]

[0050] In the formula, a ij Score the importance of the indicators; W i The subjective weight of the indicator.

[0051] Finally, the consistency of the judgment matrix is ​​checked:

[0052] The first step is to calculate the largest eigenvalue of the judgment matrix. The calculation formula is as follows:

[0053]

[0054] The second step is to calculate the consistency index CI, using the following formula:

[0055]

[0056] The third step is to perform a consistency check, calculated using the following formula:

[0057]

[0058] In the formula, n is the order of the judgment matrix; RI is the average random consistency index, the specific value of which can be obtained from a table.

[0059] When CR < 0.1, matrix A is considered to have passed the consistency test and the calculated weight values ​​are valid; otherwise, the judgment matrix needs to be readjusted until the consistency test is satisfied.

[0060] b. Objective Weights: The objective weight values ​​of each prediction indicator are calculated using the entropy weight method. The entropy weight method is an objective weighting method based on information entropy theory. It determines the weight of each indicator based on the dispersion of the data itself. Its core idea is: the greater the dispersion of an indicator's data (the smaller the information entropy), the more information that indicator provides, and the higher its weight; conversely, the more similar the data (the greater the information entropy), the lower the weight. The specific steps are as follows:

[0061] First, determine the sample evaluation matrix, and then perform Min-Max normalization on the matrix for both positive and negative indicators:

[0062] Positive indicators, i.e., indicators that are beneficial to roof stability:

[0063]

[0064] Negative indicators, i.e., indicators that are detrimental to roof stability:

[0065]

[0066] In the formula, r max r represents the maximum value under the same index. min It is the minimum value under the same indicator.

[0067] Secondly, calculate the information entropy value of the indicator:

[0068]

[0069]

[0070] In the formula, b ij The index is the normalized value; H j This refers to the entropy of the indicator information.

[0071] Finally, calculate the index entropy weights:

[0072]

[0073] In the formula, W j H is the entropy weight; j This refers to the entropy of the indicator information.

[0074] (5) Calculate the overall weight:

[0075] The multiplier synthesis and normalization method is used to combine subjective and objective weights to form a combined weight value W. k The calculation formula is as follows:

[0076]

[0077] In the formula, W Aj W represents the weight of the j-th index in the analytic hierarchy process. EjLet be the weight of the j-th index in the entropy weight method.

[0078] (6) Conduct comprehensive zoning prediction:

[0079] The processed raster network of the seven indicators was imported into ArcGIS software. Based on the maximum and minimum values ​​and overall distribution patterns of each indicator, and referring to standards such as DZ / T 0215—2002 "Specifications for Coal and Peat Geological Exploration", MT / T 1091—2008 "Evaluation Standards for Hydrogeological, Engineering Geological and Environmental Geological Exploration in Coal Mining Areas", GB / T 12719-2021 "Specifications for Hydrogeological and Engineering Geological Exploration in Mining Areas", "Detailed Rules for Coal Mine Geological Work", "Interpretation of Detailed Rules for Coal Mine Geological Work", and "Detailed Rules for Coal Mine Water Prevention and Control", three threshold values ​​were determined for each indicator: sand-mud ratio with threshold values ​​of 0.5, 1, and 1.5; rock mass quality coefficient with threshold values ​​of 0.1, 0.3, and 0.5; and stress level with threshold values ​​of 10 MPa, 18 MPa, and 30 MPa. Using MPa as the threshold, fault similarity dimension with thresholds of 1.10, 1.35, and 1.55, fold plane deformation coefficient with thresholds of 0.15, 0.35, and 0.55, and unit inflow with thresholds of 0.1 L / s·m, 1 L / s·m, and 5 L / s·m, the raster network of the indicators was reclassified into four levels using ArcGIS. The specific steps are as follows:

[0080] a. Import the single-index partitioned raster network generated using Surfer software into ArcGIS. Use the spatial analysis tools in ArcToolbox, select reclassification, and reclassify the raster network of the 7 prediction indicators into four categories according to the threshold ranges mentioned above.

[0081] b. Using the spatial analysis tools in ArcToolbox, select the weighted sum under the overlay analysis, input the raster data that has been reclassified and attach its corresponding subjective and objective combination weights to generate a weighted overlay raster network;

[0082] c. The weighted stacked raster network is divided into four categories using the equal spacing method;

[0083] By quantitatively analyzing the multi-structural factor system and determining the index weights using a combination of subjective and objective methods, and then using ArcGIS software for weighted summation, coal seam roof zoning prediction can be completed. This can then guide the refined management of coal seam roof, allowing for different management measures to be taken for different types of roofs, significantly reducing the incidence of coal seam roof disasters and ensuring safe and efficient coal mining.

[0084] (7) Innovation and beneficial effects

[0085] The main innovation of this invention patent lies in:

[0086] a. A novel index system for predicting coal seam roof stability was established. Its key feature is that it comprehensively considers the influence of four factors on roof stability: roof lithology and rock mass structure, geological structure, in-situ stress, and mine hydrogeology. In particular, it incorporates in-situ stress values ​​corrected by structural analysis into the index, and establishes a system including the roof sand-mud ratio K. r Top lithological combination L, corrected rock mass quality coefficient Z, fault similarity dimension D s , Coefficient of deformation of the folded plane K p The prediction model for seven quantitative indicators, including ground stress value S and unit inflow q of the roof, clearly defines the stability judgment threshold range of each indicator.

[0087] b. Applying Surfer and ArcGIS software to coal seam roof stability zoning prediction, characterized in that, firstly, the more user-friendly Surfer software is used to generate the raster networks of each of the seven prediction indicators, and then ArcGIS software is used to perform weighted overlay to complete the intelligent zoning prediction of coal seam roof stability.

[0088] c. In determining the weights of indicators for predicting the stability of coal seam roof, a combined subjective and objective weighting method is adopted. Its key feature is that, firstly, a judgment matrix is ​​constructed based on expert experience; subjective weights of each indicator are obtained through the Analytic Hierarchy Process (AHP); then, the entropy weight method is used to calculate the information entropy of the original data to obtain objective weights reflecting the degree of data variation; finally, the coupling of subjective and objective weights is achieved through a multiplier synthesis and normalization method, and a comprehensive weight value is obtained after normalization. This effectively solves the contradiction between the subjective arbitrariness and objective mechanicalness of traditional single weighting methods, significantly improving the scientificity and reliability of the roof stability prediction indicator system.

[0089] The main beneficial effects of this invention patent are as follows:

[0090] a. The coal seam roof stability prediction method based on multiple structural factors is simple to operate, easy to put into direct use, and has been verified by actual engineering projects, showing high accuracy in prediction results;

[0091] b. It integrates multiple structural physical factors that affect roof stability, significantly reducing the risk of misjudging roof stability due to structural complexity when predicting coal seam roof stability;

[0092] c. Guide differentiated roof management based on roof stability prediction results: reduce unnecessary anchor bolt usage in stable roofs to save support costs; strengthen support in unstable roofs to extend roof maintenance cycles, save maintenance costs, and improve project efficiency. Attached Figure Description

[0093] Figure 1 This is a flowchart of the process of the present invention.

[0094] Figure 2 This is a flowchart for consistency verification.

[0095] Figure 3 This is a partitioning diagram of the roof unit of the No. 2 coal seam in a certain mine.

[0096] Figure 4 This is a lithological zoning map of the roof of the No. 2 coal seam in a certain mine.

[0097] Figure 5 This is a lithological zoning map of the roof of the No. 2 coal seam in a certain mine.

[0098] Figure 6 This is a zoning map showing the complexity of faults in the roof of the No. 2 coal seam in a certain mine.

[0099] Figure 7 This is a zoning map showing the complexity of the roof folds in the No. 2 coal seam of a certain mine.

[0100] Figure 8 This is a zoning diagram of the roof structure type of the No. 2 coal seam in a certain mine.

[0101] Figure 9 This is a zoning map of the geostress field at the location of the roof of the No. 2 coal seam in a certain mine.

[0102] Figure 10 This is a zoning diagram of the roof stability of the No. 2 coal seam in a certain mine.

[0103] Figure 11 The measured support resistance of working face 1 in the stable zone of the roof of coal seam No. 2 in a certain mine.

[0104] Figure 12 The measured support resistance of working face 2 in the general stable zone of the No. 2 coal seam roof of a certain mine. Detailed Implementation

[0105] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present invention. These all fall within the scope of protection of the present invention.

[0106] Example

[0107] Taking the No. 2 coal seam roof of a coal mine in Central China as the research object, the average burial depth of the coal seam exceeds 700 meters. Its main roof lithology is sandy mudstone and sandstone. The vertical principal stress is large, and there is a risk of roof disaster, which poses a serious threat to mine production and the life safety of workers. Therefore, it is crucial to predict the stability of the coal seam roof.

[0108] As attached Figure 1The stability prediction of the coal seam roof can be completed by following the workflow of this invention. The specific steps are as follows:

[0109] (1) Determine the tectonic physical factors affecting the stability of the coal seam roof:

[0110] a. Geological structural elements group: Define the fault similarity dimension D s With the deformation coefficient K of the folded plane p To construct indicators;

[0111] b. Lithology and Rock Mass Structure Elements: Defining the Sand-Mud Ratio K of the Roof r The lithological combination L of the roof and the modified rock mass quality coefficient Z are lithological and rock mass structure indicators.

[0112] c. Geostress Element Group: Defines the maximum principal stress value S in the area where the coal seam roof is located as a geostress index.

[0113] d. Hydrogeological elements group: Define the unit inflow q of the roof as a hydrological index.

[0114] (2) Comprehensively analyze the influencing factors and calculate the quantitative values ​​of the prediction indicators:

[0115] a. Using the grid cover method to statistically calculate the fault similarity dimension D s An initial cover layer was established using a square grid with a base side length r = 1000m, and the grid was gradually refined at levels of 500×500m and 250×250m (see attached). Figure 2 The number of cell elements containing fault traces corresponding to the side length of each grid level is counted as N(r). Then, a regression fitting line is plotted in a coordinate system with lgN(r) as the ordinate and lgr as the abscissa. The absolute value of the slope is the similarity dimension, calculated using the following formula:

[0116]

[0117] In the formula, N(r) is the number of units; r is the side length.

[0118] The complexity of the fault structure was divided into four threshold intervals, among which D s ≤1.10 represents a simple construction; 1.10 < D s ≤1.35 indicates medium structure; 1.35 < D s ≤1.55 indicates a complex structure, D s >1.55 indicates an extremely complex structure.

[0119] b. Statistically calculate the fold plane deformation coefficient K using the sliding window method. p The calculation formula is:

[0120]

[0121] In the formula, h is the elevation difference between two adjacent contour lines, m; the horizontal distance between two contour lines, m; L1 is the actual length of the contour line closest to the unit point, m; L2 is the secant length of the contour line closest to the unit point, m.

[0122] Referring to DZ / T 0215—2002 "Specifications for Geological Exploration of Coal and Peat", based on the overall data distribution characteristics of the planar deformation coefficient of the roof folds in a certain mine's No. 2 coal seam, the complexity of the fold structure is divided into four threshold intervals, among which K p ≤0.15 represents a simple construction, and 0.15 < K p ≤0.35 indicates medium structure; 0.35 < K p ≤0.55 indicates a complex structure, K p A value greater than 0.55 indicates an extremely complex structure.

[0123] c. Referring to the K-value method, the ratio of the thickness of the roof sandstone layer and mudstone layer is selected as a quantitative indicator. The thicknesses of sandstone and mudstone in the roof within the study area of ​​the No. 2 coal seam in a certain mine are statistically analyzed, and the sandstone-mudstone ratio K is calculated. r The calculation formula is:

[0124]

[0125] In the formula, H s Thickness of sandstone strata, m; thickness of rock strata, m.

[0126] A high sandstone-mudstone ratio indicates a high sandstone content, and vice versa. The lithology of the roof is classified as follows: Sandstone Zone (K... r ≥1.5), sandstone-mudstone zone (1<K) r ≤1.0), mudstone-sandstone zone (0.5<K) r ≤1.0), mudstone area (K r ≤0.5).

[0127] d. The lithological assemblage L of the No. 2 coal seam in a certain mine is divided into 4 types: Type I: completely soft type (composed entirely of fine-grained sandy mudstone and mudstone), Type II: alternating soft and hard type (composed of relatively thin layers of mudstone and sandstone), Type III: hard at the bottom and soft at the top (the grain size of the rock strata becomes finer upwards), and Type IV: soft at the bottom and hard at the top (the grain size of the rock strata becomes coarser upwards).

[0128] e. The corrected rock mass quality coefficient for the roof was calculated using the Analytic Hierarchy Process (AHP) to determine the weights of the rock mass integrity index (31.6%), structural surface complexity index (47.2%), and rock mass strength index (21.2%). Referring to the rock mass quality coefficient method in GB / T 12719-2021 "Specifications for Hydrogeological and Engineering Geological Exploration in Mining Areas", a corrected rock mass quality coefficient Z was introduced as an index to quantitatively describe the structural strength of the coal seam roof rock mass. The roof rock mass structure of the No. 2 coal seam in a certain mine was divided into four threshold intervals: Z > 0.5 for integral blocky structure, 0.3 ≤ Z < 0.5 for layered structure, 0.1 ≤ Z < 0.3 for fractured structure, and Z < 0.1 for granular structure.

[0129] f. Maximum principal stress value S: Collect measured geostress data in the study area, and correct the geostress using the tectonic analytical method. The analytical formula is:

[0130]

[0131]

[0132]

[0133]

[0134]

[0135]

[0136] Where, σ v For vertical principal stress, MPa; σ H The maximum horizontal principal stress is σ, MPa; h σ1, σ2, and σ3 are the measured ground stress values, in MPa; l1, l2, and l3 are the products of the dip cosine and azimuth sine corresponding to the measured ground stress values; m1, m2, and m3 are the products of the dip cosine and azimuth sine corresponding to the measured ground stress values; n1, n2, and n3 are the dip cosine values ​​corresponding to the measured ground stress values.

[0137] Linear regression was used to fit the linear relationship between principal stress and burial depth in the study area, and the in-situ stress value was calculated back from the roof burial depth. According to the stress level judgment standard, the in-situ stress field of the roof of the No. 2 coal seam in a certain mine was divided into four threshold intervals: 0-10 MPa is the low stress zone, 10-18 MPa is the medium stress zone, 18-30 MPa is the high stress zone, and greater than 30 MPa is the ultra-high stress zone.

[0138] g. Unit inflow rate q, determined through on-site pumping tests, is calculated using the following formula:

[0139]

[0140] Where Q is the flow rate during pumping, in L / s; and S is the drawdown, in m.

[0141] According to the "Detailed Rules for Prevention and Control of Water in Coal Mines", the hydrogeological and geogeological types of the roof of the No. 2 coal seam in a certain mine are divided into four threshold ranges: q≤0.1 is simple, 0.1<K≤1.0 is medium, 1.0<K≤5.0 is complex, and K>5.0 is extremely complex.

[0142] Because the roof of the No. 2 coal seam in this mine lacks groundwater recharge and has extremely low water volume (0.00116 L / (s·m) to 0.00137 L / (s·m) per unit), and is generally classified as a weakly permeable roof, the unit water inflow is not included in the stability prediction index for the roof of the No. 2 coal seam. The specific index set is shown in Table 1.

[0143] Table 1. Set of prediction indicators for the roof stability of the No. 2 coal seam in a certain mine

[0144]

[0145] (3) Use Surfer software to generate a single-index partitioned raster network:

[0146] First, use Excel to complete the basic data editing, where X and Y are the spatial locations of the data points (geocentric coordinate system), and Z is the quantitative value of each prediction index. Import the basic data into Surfer for gridding, use Kriging interpolation to calculate, and whiten the part outside the boundary of the study area to generate the raster network of each of the 6 indices.

[0147] (4) Calculate the subjective and objective weights of various prediction indicators for coal seam roof stability:

[0148] a. Calculate the subjective weights, the specific steps are as follows:

[0149] First, a hierarchical model is established: the highest layer is set as the stability of the coal seam roof; the basic layers are set as various indicators affecting the stability of the coal seam, namely the roof sand-slurry ratio K. r Top lithological combination L, corrected rock mass quality coefficient Z, fault similarity dimension D s , Coefficient of deformation of the folded plane K p A total of 6 items, including the ground stress value S;

[0150] Secondly, construct the judgment matrix:

[0151]

[0152] In the formula, a ij The meaning is the degree of importance of the i-th indicator compared to the j-th indicator, such as a 12 This indicates the relative importance of the first indicator compared to the second indicator.ij The scoring criteria are as follows: 1 for two equally important indicators, 3 for indicator i being slightly more important than indicator j, 5 for indicator i being significantly more important than indicator j, 7 for indicator i being strongly more important than indicator j, and 9 for indicator i being extremely more important than indicator j. 2, 4, 6, and 8 are the values ​​when the above scale is directly compromised. The score can be any non-negative number.

[0153] The judgment matrix for six indicators of the No. 2 coal seam in a certain mine is as follows:

[0154]

[0155] Then, the root method is used to solve for the eigenvectors of the judgment matrix, i.e., the subjective weights of the indicators:

[0156] The first step is to calculate the nth root of the product of the elements in each row of the judgment matrix A, using the following formula:

[0157]

[0158] The second step is to M i The subjective weights of the indicators are obtained by proportional normalization, as shown in the following formula:

[0159]

[0160] In the formula, a ij Score the importance of the indicators; W i The subjective weight of the indicator.

[0161] Finally, a consistency check is performed.

[0162] The first step is to calculate the largest eigenvalue of the judgment matrix, using the following formula:

[0163]

[0164] The second step is to calculate the consistency index (CI), using the following formula:

[0165]

[0166] The third step is to perform a consistency check, the process of which is shown in the attached document. Figure 3 The formula is as follows:

[0167]

[0168] In the formula, n is the order of the judgment matrix; RI is the average random consistency index, and its values ​​are shown in Table 2:

[0169] Table 2. Values ​​of the average random consistency index

[0170]

[0171] When CR < 0.1, matrix A is considered to have passed the consistency test and the calculated weight values ​​are valid; otherwise, the judgment matrix needs to be readjusted until the consistency test is satisfied.

[0172] The subjective weights of the six predictive indicators for the roof stability of the No. 2 coal seam in a certain mine are shown in Table 3:

[0173] Table 3. Subjective Weight Analysis Results of Roof Stability Evaluation Indicators for No. 2 Coal Seam in a Certain Mine

[0174]

[0175] The largest eigenvalue is 6.441. According to the RI average random consistency index value table, the corresponding RI value is 1.24. Since CR=CI / RI=0.07<0.1, it passes the one-time test.

[0176] b. Calculate the objective weights, the specific steps are as follows:

[0177] First, a sample evaluation matrix is ​​constructed, which is the set of quantitative values ​​calculated for each of the six predictive indicators, and the matrix is ​​normalized according to the positive and negative indicators:

[0178] Positive indicators, i.e., indicators that are beneficial to roof stability:

[0179]

[0180] Negative indicators, i.e., indicators that are detrimental to roof stability:

[0181]

[0182] In the formula, r max r represents the maximum value under the same index. min It is the minimum value under the same indicator.

[0183] Secondly, calculate the information entropy value of the indicator:

[0184]

[0185]

[0186] In the formula, b ij The index is the normalized value; H j This refers to the entropy of the indicator information.

[0187] Finally, calculate the index entropy weights:

[0188]

[0189] In the formula, H j This refers to the entropy of the indicator information.

[0190] The subjective weights of the six prediction indicators for the roof stability of the No. 2 coal seam in a certain mine are shown in Table 4:

[0191] Table 4. Objective Weight Analysis Results of Roof Stability Evaluation Indicators for No. 2 Coal Seam in a Certain Mine

[0192]

[0193] (5) Calculate the comprehensive weight of the forecasting indicators:

[0194] The multiplier synthesis and normalization method is used to combine subjective and objective weights to form a combined weight value W. k The calculation formula is as follows:

[0195]

[0196] In the formula, W Aj W represents the weight of the j-th index in the analytic hierarchy process. Ej Let be the weight of the j-th index in the entropy weight method.

[0197] The subjective weights of the six predictive indicators for the roof stability of the No. 2 coal seam in a certain mine are shown in Table 5:

[0198] Table 5 Calculation Results of Subjective and Objective Combined Weights

[0199]

[0200] (6) Conduct comprehensive zoning prediction:

[0201] Referring to DZ / T 0215—2002 "Specifications for Geological Exploration of Coal and Peat", MT / T 1091—2008 "Evaluation Standards for Hydrogeological, Engineering Geological and Environmental Geological Exploration of Coal Mining Areas", GB / T 12719-2021 "Specifications for Hydrogeological and Engineering Geological Exploration of Mining Areas", "Detailed Rules for Geological Work in Coal Mines", and "Interpretation of Detailed Rules for Geological Work in Coal Mines", and based on the maximum, minimum and overall distribution patterns of six prediction indicators, three thresholds were determined (Table 1). These were then reclassified into four levels. A single-indicator zoning map is attached. Figures 4-9 .

[0202] The specific steps for zoning prediction using ArcGIS software are as follows:

[0203] a. Import the single-index partitioned raster network generated by Surfer software in (3) into ArcGIS. Use the spatial analysis tool in ArcToolbox, select reclassification, and perform four-class reclassification of the raster network of the six prediction indicators according to the threshold range in Table 1.

[0204] b. Using the spatial analysis tools in ArcToolbox, select the weighted sum under the overlay analysis, input the raster data that has been reclassified and attach its corresponding subjective and objective combination weights to generate a weighted overlay raster network;

[0205] c. The weighted stacked raster network is divided into four categories using the equal-spacing method. The comprehensive zoning prediction is shown in the appendix. Figure 10 .

[0206] After completing the zoning prediction, to verify its accuracy, two typical working faces were selected in the study area, and the variation of hydraulic support resistance during the mining process was recorded: Working face 1 is located in a stable to relatively stable roof area, with a monolithic block structure, simple lithology, high sand content in the overlying strata, poorly developed structural planes, and strong bonding, resulting in a relatively simple geological structure; Working face 2 is located in a general roof area, with a layered roof structure, soft strata at the bottom and hard strata at the top, and relatively well-developed minor geological structures. By recording the hydraulic support resistance during the mining process, line graphs of the advance distance and support resistance for working faces 1 and 2 were plotted, as shown in the attached figure. Figure 11 The resistance of the support at working face 1 fluctuates between 25 and 30 MPa, and the pressure is not obvious; as shown in the attached figure. Figure 12 The resistance of the support at working face 2 fluctuated between 25 and 35 MPa, indicating significant pressure. The measured data reflects that the self-stabilizing capacity of the roof of working face 2 is weaker than that of the roof of working face 1, which is consistent with the predicted results.

[0207] Specific implementation examples show that this prediction method is simple to operate, effective, and highly applicable to engineering projects. Based on the prediction results, it can effectively guide the refined management of the roof, significantly reduce the incidence of roof disasters, and transform roof disasters from passive prevention to active control.

Claims

1. A method for predicting the stability of a coal seam roof based on multiple structure factors, characterized in that, Comprise the following steps: (1) determine the tectonic physical factors affecting the stability of coal seam roof and the quantitative prediction index of each factor, including: a. Geological structure quantification indicators, including fault similarity dimension D s and fold plane deformation coefficient K p ; b. Roof lithology and rock mass structure quantitative indicators, including roof sand-silt ratio K r , roof lithology combination L and modified rock mass quality coefficient Z; c. quantitative index of ground stress, including the maximum principal stress value S of the coal seam roof area; d. quantitative index of hydrogeological quality, including unit inflow q; (2) comprehensive analysis of each influencing factor, calculate the quantitative value of prediction index; (3) use Surfer software to generate grid network of each single index; (4) determine the subjective and objective weights of each single index, including: a. objective weight: determined by using entropy weight method, including determining sample evaluation matrix, normalizing positive and negative index matrix, calculating index information entropy value, calculating index entropy weight value; b. subjective weight: determined by using AHP hierarchical analysis method, including establishing hierarchical model, constructing judgment matrix, solving characteristic vector, and performing consistency test; (5) Calculate the comprehensive weight, adopt the multiplier synthesis normalization method, combine the subjective weight and the objective weight to determine the combination weight value W j , the calculation formula is as follows: In the formula, W Aj is the weight of the jth index of the analytic hierarchy process; W Ej is the weight of the jth index of the entropy weight method; (6) use ArcGIS software for zoning prediction, including: a. import single index zoning grid network generated by Surfer software into ArcGIS, use spatial analysis tool in ArcToolbox, select reclassification, and perform four-class reclassification on the grid network of 7 prediction indexes; b. use spatial analysis tool in ArcToolbox, select weighted sum of superposition analysis, input the reclassified grid data and its corresponding subjective and objective combined weights, and generate weighted superposition grid network of regional prediction; c. select equal interval method to divide the weighted superposition grid network into four classes, namely class I stable type, class II relatively stable type, class III general stable type and class IV unstable type.

2. The method of claim 1, wherein, In step (1), the 4 types of roof lithologic combination L include: a. full soft type, which is composed of fine-grained sandy mudstone and mudstone; b. soft and hard interbedded type, which is composed of multiple thin layers of mudstone and sandstone interbedded; c. lower hard and upper soft type, the particle size of which decreases from bottom to top; d. lower soft and upper hard type, the particle size of which increases from bottom to top.

3. The method of claim 1, wherein, In step (2), the specific quantitative calculation method of multiple tectonic factors includes: (1) Calculation of fault similar dimension D s , the specific steps are as follows: using variable side length grid coverage, establishing an initial coverage layer with square grid of reference side length r = 1000 m, and gradually encrypting the grid according to 500 × 500 m, 250 × 250 m; counting the number of fault traces of each level grid unit, recording the number of grid units N(r) corresponding to each level grid side length containing faults, and then making a regression fitting straight line in the coordinate system with lgN(r) as the vertical coordinate and lgr as the horizontal coordinate, and the absolute value of the slope is the fault similar dimension D s , the calculation formula is: In the formula, N(r) is the number of units; r is the side length, m; (2) The sliding window method is used to calculate the deformation coefficient K of the fold plane p The calculation formula is: In the formula, h is the elevation difference between the two adjacent contour lines, m; L0 is the horizontal distance between the two adjacent contour lines, m; L1 is the actual length of the nearest contour line to the unit point, m; L2 is the length of the secant line from the unit point to the nearest contour line, m; (3) Calculate the roof sand mud ratio K r The calculation formula is: In the formula, H s is the thickness of sandstone formation, m; H r is the thickness of mudstone formation, m; (4) divide the roof lithologic combination L into four types: class I full soft type, class II soft and hard interbedded type, class III lower hard and upper soft type, and class IV lower soft and upper hard type; (5) calculate the roof modified rock mass quality coefficient Z, the calculation formula is: In the formula, a is the weight of rock mass integrity; C is the core sampling rate; b is the weight of structure surface complexity; f is the structure surface complexity; c is the weight of rock mass strength; K is the rock mass strength; (6) collect the measured data of ground stress in the study area and correct it by using tectonic analysis method, use linear regression to fit the linear relationship between vertical principal stress, maximum horizontal principal stress and minimum horizontal principal stress and burial depth, and use burial depth conversion to calculate the maximum principal stress value S. (7) The calculation formula of the roof unit water inflow is: In the formula, Q is the flow rate during pumping, L / s; S is the water level drawdown, m.

4. The method of claim 1, wherein, In step (3), the specific steps for generating a single-index partition grid network are as follows: a. Use Excel to complete the basic data editing, in which X and Y define the spatial position of the data points, i.e., the geodetic coordinate system, and the Z value is the quantitative value of each prediction index; b. Import the basic data into Surfer for gridding, select the Kriging interpolation method for calculation, and whiten the part outside the boundary of the study area to generate a grid network for each index.

5. The method of claim 1, wherein, In step (4), the specific calculation method of the objective weight is as follows: (1) Determine the sample evaluation matrix and perform Min-Max normalization processing on the matrix for positive and negative indicators: Positive indicators: Negative indicators: wherein r max is the maximum value under the same index, r min is the minimum value under the same index; (2) Calculate the index information entropy value: (3) Calculate the index entropy weight value, i.e., the objective index: In the formula, b ij is the normalized index; H j is the index information entropy.

6. The method of claim 1, wherein, In step (6), the threshold values for the four types of grid reclassification are as follows: the sand-mud ratio with thresholds of 0.5, 1, and 1.5, the rock mass quality coefficient with thresholds of 0.1, 0.3, and 0.5, the stress level with thresholds of 10, 18, and 30, the fault similarity dimension with thresholds of 1.10, 1.35, and 1.55, the fold plane deformation coefficient with thresholds of 0.15, 0.35, and 0.55, and the unit water inflow with thresholds of 0.1, 1, and 5.

7. The method of claim 3, wherein, In step (5), the rock mass integrity weight, structure surface complexity weight, and rock mass strength weight involved in the calculation of the corrected rock mass quality coefficient Z of the roof are determined using the AHP method.

8. The method of claim 3, wherein, In step (5), the quantitative indicators of rock mass integrity are the roof core recovery rate; the quantitative indicators of structure surface complexity are the product of structure surface density and structure surface strength; and the quantitative indicators of rock mass strength are the percentage of hard rock thickness in the total thickness of the roof.

9. The method of claim 3, wherein, In step (6), the formula for analyzing ground stress is: In the formula, σ v is the vertical principal stress, MPa; σ H is the maximum horizontal principal stress, MPa; σ h is the minimum horizontal principal stress, MPa; σ1, σ2, σ3 are measured in-situ stress values, MPa; l1, l2, l3 are the products of the cosine of the inclination angle and the sine of the azimuth angle corresponding to the measured in-situ stress values; m1, m2, m3 are the products of the cosine of the inclination angle and the cosine of the azimuth angle corresponding to the measured in-situ stress values; n1, n2, n3 are the cosine values of the inclination angle corresponding to the measured in-situ stress values.

10. The method of claim 1 or 7, wherein, The method for solving the subjective weight using the AHP method is as follows: (1) Establish hierarchical model: the highest layer is set as the stability of coal seam roof, the basic layer is set as each index affecting the stability of coal seam, i.e. the sand-mud ratio K of roof r , the roof lithology combination L, the modified rock mass quality coefficient Z, the fault similar dimension D s , the fold plane deformation coefficient K p , the ground stress value S and the roof unit water inflow q totally 7 indexes; (2) Construct the judgment matrix: wherein a ij The meaning of a 12 i.e. the importance of the 1st indicator compared to the 2nd indicator, with a value of 1-9. (3) Solve the characteristic vector of the judgment matrix, i.e., the subjective weight value: a. Calculate the nth root of the product of the elements in each row of the judgment matrix A, as follows: b. M i The subjective weight value is obtained by performing the proportional normalization, and the formula is as follows: (4) Test the consistency of the judgment matrix: a. Calculate the maximum eigenvalue of the judgment matrix, as follows: b. Calculate the consistency check index CI, as follows: c. Perform consistency check judgment, as follows: In the formula, n is the order of the judgment matrix; RI is the average random consistency index, which can be obtained by referring to the table; When CR < 0.1, it is considered that the matrix A passes the consistency test, and the subjective weight value obtained is valid, otherwise, the judgment matrix needs to be adjusted until it meets the consistency test.

Images