A method for preventing coal seam outbursts by controlling the stress field in deep coal seam backfilling mining based on digital twins.

Abstract

This invention discloses a method for stress field control and outburst prevention in deep coal seam backfilling mining based on digital twins. First, geological data of the target mining area is collected to construct a full-mine-scale digital twin model, reconstructing the non-uniform stress field distribution and identifying stress concentration areas. Then, a refined digital twin model at the mining area scale is established, and stress boundaries are accurately transferred between the two scales through coordinate interpolation. Subsequently, simulation is used to optimize mining layout parameters such as face width and pillar width, establishing a functional relationship between backfilling height, backfilling rate, and stress concentration. Next, the backfilling rate is dynamically adjusted based on time-series mining to reduce stress concentration from multiple working faces. Finally, the optimized strategy is applied back to the full-mine-scale model to achieve prediction and precise management of stress evolution in adjacent mining areas. This invention achieves precise control of the stress field in deep coal seams throughout the entire process, significantly reducing the risk of coal and gas outbursts, and is applicable to safe mining of deep coal seams under different geological conditions.

Description

Technical Field

[0001] This invention belongs to the field of coal and rock dynamic disaster prevention and control technology in mining, specifically involving a method for preventing outbursts by controlling the stress field during deep coal seam backfilling mining based on digital twins. It is applicable to the safe mining of coal seams under conditions of high ground stress, high gas, and strong mining disturbance. Background Technology

[0002] During continuous mining across multiple working faces, the superposition and disturbance effects of stress fields are significant, which can easily lead to non-uniform stress concentration and subsequently induce coal and gas outbursts and other coal and rock dynamic disasters. This has become a core bottleneck restricting the safe and efficient mining of deep coal resources.

[0003] Existing deep coal seam outburst prevention technologies and stress control methods have several shortcomings: First, existing technologies mostly focus on localized treatment after a disaster has occurred, such as using only localized methods like borehole decompression or hydraulic fracturing. These methods can only temporarily alleviate localized stress and cannot control the global imbalance of the stress field from its source, potentially even exacerbating surrounding rock fracturing and increasing the risk of disaster. Second, existing stress control methods for backfilling mining mostly adopt fixed backfill ratio designs, failing to consider the dynamic changes in stress superposition during multi-face sequential mining. This makes it impossible to achieve precise dynamic control of non-uniform stress fields, resulting in limited stress control effectiveness. Third, existing... The application of digital twin technology in the mining field is mostly concentrated on single-scale model construction, which can only realize stress monitoring and simple mining simulation. It cannot take into account the overall characterization of stress evolution in the entire mine and the high-precision simulation of local mining areas. Small-scale models have problems such as boundary condition distortion and significant boundary effects. Fourth, the existing technology has not formed a full-process processing system of "mining layout optimization - dynamic control of filling parameters - global stress verification - advanced disaster management". It treats coal and gas outbursts as independent disasters and ignores that stress concentration is the core cause of disasters. It cannot achieve advanced disaster prediction and overall prevention and control across mining areas.

[0004] To address the shortcomings of existing technologies, some research has been conducted in related fields. However, these studies all employ fixed filling ratio designs, failing to incorporate the dynamic characteristics of stress superposition across multiple working faces for precise control, and lacking the ability to predict global stress evolution using digital twin technology, thus hindering advanced management across mining areas. Therefore, there is an urgent need to develop a deep coal seam backfilling mining outburst prevention method capable of multi-source information fusion, dual-scale precise simulation, full-process closed-loop control, and advanced precise management, fundamentally improving the prevention and control of deep coal and rock dynamic disasters. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a method for stress field control and outburst prevention in deep coal seam backfilling mining based on digital twins. It achieves accurate global characterization of the stress field through dual-scale digital twin model coupling, achieves source control of the stress field through synergistic optimization of mining parameters and backfilling parameters, and achieves closed-loop outburst prevention and control throughout the entire process through dynamic backfilling and global backfilling verification. This effectively solves the technical problems of stress superposition and concentration and high outburst risk in multi-face mining of deep coal seams.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: a method for precise control of stress field in deep coal seam backfilling mining to prevent outbursts, based on digital twins, comprising the following steps: Step S1: Construction of a full-mine-scale digital twin model and characterization of the global stress field: Collect all geological and mining data of the target mining area and use FLAC... 3D Numerical simulation software is used to establish a full-scale digital twin model of the mine, and the measured ground stress, roof settlement and surrounding rock deformation data are assigned to the model and the self-weight balance calculation is completed. According to the actual mining succession sequence of the mine, the non-uniform stress field distribution in the deep mining area caused by shallow coal seam mining and disturbance of adjacent working faces is restored, the high-risk area of ​​stress concentration is identified, and the initial mining strategy and mining direction are formulated.

[0007] Step S2: Construction of a refined digital twin model at the mining area scale and coupling of two scales: Taking the target mining area as the research object, a refined digital twin model at the mining area scale is established, which includes the production system, geological structure, and coal seam occurrence characteristics. The boundary space coordinates of the mining area scale model are derived, and the corresponding coordinates are matched and stress data are extracted in the full mine scale model. The stress data of discontinuities are converted into continuous stress data of the boundary surface of the mining area model through interpolation mathematical methods, and the stress boundary assignment and calculation balance of the mining area scale model are completed.

[0008] Step S3: Collaborative optimization of multi-face mining layout parameters: Based on the balanced digital twin model of the mining area, conduct simulation tests of stress superposition disturbance in multi-face mining to obtain the stress concentration coefficient and stress concentration range variation law under different combinations of face width and coal pillar width, establish the functional relationship between stress concentration characteristics and face width and coal pillar width, and determine the optimal mining layout parameters by combining stress evolution law and mining cost.

[0009] Step S4: Construction of a functional model of filling parameters and stress concentration characteristics: Based on the optimal stope layout parameters, different mining heights and filling rate gradients are set, and parallel numerical simulation tests under multiple working conditions are carried out to obtain the maximum disturbance stress concentration coefficient and stress concentration range variation law under different filling rates for continuous mining of multiple working faces, and to establish a quantitative functional relationship between filling height, filling rate and stress concentration degree.

[0010] Step S5: Dynamic control of filling rate under sequential mining: Conduct digital twin synchronous simulation of sequential filling mining of the working face. Based on the non-uniform stress distribution characteristics of different areas of the working face, and combined with the quantitative function relationship established in step S4, determine the optimal filling rate for the current mining stage. Update stress field data in real time as the working face advances, and dynamically adjust filling parameters to minimize stress superposition and concentration caused by continuous mining of multiple working faces, thereby reducing the risk of outburst in the mining area.

[0011] Step S6: Global Back-substitution Verification and Advanced Precision Outburst Prevention and Control: The optimized backfilling mining strategy of the target mining area is back-substituted into the digital twin model at the whole mine scale, and synchronous mining simulation of the whole mine is carried out to obtain the stress evolution and transfer law of adjacent and surrounding mining areas. Synchronous or advanced control measures are formulated for areas with abnormal stress concentration to reduce the potential outburst risk in subsequent mining areas and realize outburst prevention and control throughout the entire mining process.

[0012] Furthermore, in step S1, the collected full-volume geological and mining data includes geological exploration data of the mining area, stress monitoring data, microseismic and acoustic emission monitoring data, and mining history data; among which, the geological exploration data includes coal seam thickness, rock strata structure, distribution of faults and folds, and the stress monitoring data includes in-situ measured stress data, roof settlement data, and surrounding rock deformation data.

[0013] Furthermore, in step S1, after the digital twin model of the entire mine scale is constructed, a digital twin assimilation simulation analysis of the stress field is carried out, a stress field contour map is drawn, high-risk areas of stress concentration are marked, and the mining area and mining direction are optimized.

[0014] Furthermore, in step S2, the length, width, and height ratio difference of the digital twin model at the mining area scale does not exceed 30%, and the model expands outward on the basis of covering the target mining area to eliminate the boundary effect; the outward expansion distance of the model is adjusted according to the complexity of the geological structure of the mining area, with an outward expansion of 50-100m in areas with simple geological structures and 100-150m in areas with developed faults and complex structures.

[0015] Furthermore, in step S2, the interpolation mathematical method adopts Kriging interpolation or inverse distance weighted interpolation to convert the discrete coordinate stress data extracted from the whole mine scale model into continuous stress distribution data of the boundary surface of the mining area scale model.

[0016] Furthermore, in step S3, the simulation gradient of the face width is set to 50~300m, and the simulation gradient of the coal pillar width is set to 5~50m; the optimal mining layout parameters include the optimal face width and the coal pillar width, and the determination principle is: to prioritize ensuring that the stress concentration coefficient is lower than the critical threshold of coal seam outburst, while taking into account the optimal resource recovery rate and mining cost of the mining area.

[0017] Furthermore, in step S4, different filling heights are used to simulate different filling rate gradients to establish corresponding quantization function relationships.

[0018] Furthermore, in step S5, the synchronous simulation of time-series filling mining adopts a dynamic cyclic mode of sequential mining, simulation, and control. The filling rate is increased in the peak pressure zone and stress anomaly zone of the working face, and the filling rate is optimized in the stress stable zone to control mining costs.

[0019] Furthermore, in step S6, the advanced treatment measures include one or more combinations of advanced borehole decompression, hydraulic fracturing, local enhanced filling, and coal body reinforcement. Based on the stress evolution prediction results of the full-mine scale model, the advanced treatment is completed before the stress concentration area is formed.

[0020] Compared with the prior art, the present invention has the following advantages: 1. This invention constructs a dual-scale digital twin coupled model at the full mine scale and the mining area scale, solving the industry pain point that existing single-scale models cannot simultaneously represent the stress evolution of the entire domain and accurately simulate the local mining area. By using coordinate interpolation, the stress boundary of the large-scale model is accurately transferred to the small-scale model, effectively eliminating the boundary effect of the small-scale model, improving the simulation accuracy of the stress field at the mining area scale by more than 30%, and laying a data foundation for the precise control of the stress field.

[0021] 2. This invention establishes a collaborative optimization system that integrates stope layout parameter optimization, filling parameter function modeling, and time-series dynamic filling control, overcoming the limitations of existing technologies that rely on single parameter optimization and fixed filling rate design. By establishing a quantitative functional relationship between stress concentration characteristics and face width, coal pillar width, and filling rate, it reduces the risk of stress superposition from the source of stope layout. Simultaneously, it adjusts the filling rate in real time for the dynamic stress field of time-series mining, which can reduce the maximum stress concentration coefficient of multi-face mining by 25% to 40%, fundamentally weakening the driving force of coal and gas outbursts.

[0022] 3. This invention establishes a full-process control and outburst prevention system that integrates local mining area optimization, global mine verification, and advanced and precise management. It substitutes the mining area optimization strategy back into the full-mine scale model to accurately predict the stress evolution and transfer patterns of adjacent and surrounding mining areas. This realizes the transformation of coal and rock dynamic disasters from "post-event handling" to "advanced prediction and source control," solving the shortcomings of existing technologies that only treat the symptoms and not the root cause. It can realize full-process outburst prevention and control for continuous mining of multiple working faces in the entire mine.

[0023] 4. The method of the present invention has strong applicability and can be adapted to deep coal seams with different occurrence states such as near-horizontal, inclined and steeply inclined. It can meet the anti-outburst requirements of different scenarios such as complex structural areas and multiple working face superimposed disturbance areas. It has both safety and economy and has broad engineering application value. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating the overall process of the present invention.

[0025] Figure 2 This is a stress distribution cloud map of the full-mine-scale digital twin model of Embodiment 1 of the present invention.

[0026] Figure 3 This is a schematic diagram of the backfilling mining simulation along the AB section of the 3317 working face in Embodiment 1 of the present invention. Detailed Implementation

[0027] The present invention will be further described below.

[0028] Example 1: The application scenario of this example is a deep coal mine in western China. The main coal seam mined in this mine is No. 2 coal seam, with an average thickness of 6.2m, a dip angle of 2°~5°, and is a near-horizontal coal seam. The mining depth is 850~920m, which belongs to deep mining. The mine has typical characteristics of "high temperature, high humidity, high temperature, high risk of gas, and high risk of methane pollution," with a measured maximum horizontal principal stress of 32.7MPa and a coal seam gas content of 8.6m³. 3 / t, the outburst type is stress-dominated coal and gas outburst. During continuous mining of multiple working faces, the stress superposition disturbance is significant. The original anti-outburst methods cannot achieve precise control of the stress field, and the outburst risk is high.

[0029] The method for precise control of stress field in deep coal seam backfilling mining based on digital twins to prevent outbursts, as described in this embodiment, has the following specific steps: Step S1: Construction of a full-scale digital twin model and global stress field characterization: Collect all geological and mining data of the mine, including geological exploration data of the mining area (coal seam thickness, rock strata structure, fault distribution), in-situ measured geostress data, roof settlement and surrounding rock deformation monitoring data, microseismic and acoustic emission monitoring data, and all mining history and succession data since the mine's construction; using FLAC... 3D The software establishes a full-scale digital twin model of the mine, with dimensions of 6000m × 5000m × 1000m, covering all mineable coal seams and mining areas. Measured ground stress and roof settlement data are assigned to the model, and the model's self-weight balance is calculated. Following the actual mining succession sequence, the software reconstructs the non-uniform stress field distribution in deep mining areas caused by shallow coal seam mining and disturbance of already mined working faces. Figure 2 As shown, a digital twin assimilation simulation of the stress field was carried out, and a stress field contour map was drawn. It was determined that mining areas 33 and 35 are high-risk areas of stress concentration. Mining area 33 was identified as the priority mining area, and the mining direction was to advance from south to north along the direction of the maximum principal stress to avoid the high stress concentration caused by working face mining and coal roadway excavation.

[0030] Step S2: Construction of a refined digital twin model at the mining area scale and coupling between two scales: Taking mining area 33 as the research object, a refined digital twin model at the mining area scale is established. The model includes elements such as the mining area production system, coal seam occurrence characteristics, and small fault structures. The geological structure of this mining area is simple, and the model size is set to 2000m×1800m×1000m. While covering the entire range of mining area 33, it is expanded outward by 80m on all sides to eliminate boundary effects. The difference in the length, width, and height ratios of the model is less than 20%, which meets the simulation requirements. The boundary space coordinates of the mining area scale model are exported, and the corresponding coordinates are matched in the full mine scale model. Stress data of 1200 sets of discrete coordinate points are extracted. Kriging interpolation is used to convert the discontinuous point stress data into continuous stress distribution data of the boundary surface of the mining area model. This continuous stress data is used as the boundary condition of the mining area model to complete the model calculation balance and achieve precise coupling of stress boundaries between the two scale models.

[0031] Step S3: Collaborative optimization of multi-face mining layout parameters: Based on the balanced digital twin model of mining area 33, a simulation test of stress superposition disturbance in multi-face mining was carried out. The simulation gradients for the face width were set to 80m, 100m, 120m, 150m, 180m, and 200m, and the simulation gradients for the coal pillar width were set to 10m, 15m, 20m, 25m, 30m, and 40m, for a total of 36 sets of combined working conditions. The stress concentration coefficient and stress concentration range variation law under different working conditions were obtained, and the functional relationship between stress concentration characteristics and face width and coal pillar width was established.

[0032] Step S4: Construction of a functional model for filling parameters and stress concentration characteristics: Based on the optimal mining face width of 150m and coal pillar width of 20m, the mining height is set to 2.0m~6.2m. The corresponding filling height is determined according to the mining height, and different filling heights are used to simulate different filling rate gradients. Parallel numerical simulation tests under multiple working conditions are carried out. Figure 3 As shown, the variation law of the maximum disturbance stress concentration coefficient and stress concentration range in the continuous mining of multiple working faces in the 33 mining area under different filling ratios was obtained. A quantitative function relationship between filling height, filling ratio and stress concentration degree was established by fitting. The results show that: for every 10% increase in filling ratio, the maximum stress concentration coefficient decreases by 0.08~0.12 and the maximum stress concentration coefficient decreases by 0.05~0.07.

[0033] Step S5: Dynamic Control of Filling Rate under Sequential Mining: For the three continuous working faces (3315, 3317, and 3319) in the 33 mining area, a digital twin synchronous simulation of sequential filling mining was conducted, adopting a dynamic cyclic mode of "mining 50m, simulating once, and adjusting once". Based on the non-uniform stress distribution characteristics of different areas of the working face, combined with the quantitative function relationship established in Step S4, the optimal filling rate for the current mining stage was determined: for stress concentration areas such as the opening cut and the stop line of the working face, the filling rate was set to 95%~100%; for the peak pressure area of ​​the advance support of the working face, the filling rate was set to 90%; for the stress stable area in the middle of the working face, the filling rate was optimized to 80%, taking into account both safety and economy. The stress field data was updated in real time as the working face advanced, and the filling parameters were dynamically adjusted. Finally, the maximum stress concentration coefficient of continuous mining of multiple working faces in the 33 mining area was controlled within 1.5. Compared with the original fixed 80% filling rate mining method, the stress concentration coefficient was reduced by 21.3%, effectively eliminating the outburst risk caused by stress superposition.

[0034] Step S6: Global Backfilling Verification and Precise Outburst Prevention and Control: The optimized backfilling mining strategy of mining area 33 was backfilled into the digital twin model at the whole mine scale. Simulation of synchronous mining throughout the whole mine was carried out to simulate the stress evolution and transfer patterns of adjacent mining areas 35 and 31 during the mining of mining area 33. It was predicted that stress anomaly concentration areas would appear in the south of mining area 35 and the east of mining area 31, with a maximum stress concentration coefficient of up to 1.92. Based on this prediction, advanced control measures were formulated. Before mining of mining area 35, combined treatment of advanced drilling decompression and local enhanced backfilling was carried out in the southern stress anomaly area. Before mining of mining area 31, hydraulic fracturing decompression was carried out in the eastern stress concentration area. Finally, the stress concentration coefficient of the above areas was controlled within 1.1, reducing the potential outburst risk and achieving precise outburst prevention and control throughout the entire mining process.

[0035] Example 2: The application scenario of this example is a deep coal mine in eastern China. The main coal seam mined in this mine is No. 8 coal seam, with an average thickness of 12.5m and a dip angle of 42°~48°, making it a steeply dipping thick coal seam. The mining depth is 980~1100m, which belongs to ultra-deep mining. The mine has well-developed fault structures, with a measured maximum horizontal principal stress of 41.2MPa and a coal seam gas content of 12.3m³. 3 / t, the outburst type is gas-stress coupled coal and gas outburst. During the multi-section joint mining process, the stress superposition and gas desorption effect are significant. The original anti-outburst method cannot achieve precise control of the stress field in the whole area. Outburst warnings have occurred many times, and the risk of safe mining is extremely high.

[0036] The method for precise control of stress field in deep coal seam backfilling mining based on digital twins to prevent outbursts, as described in this embodiment, has the following specific steps: Step S1: Construction of a full-scale digital twin model and global stress field characterization: Collect all geological and mining data of the mine, focusing on fault structure distribution, steeply dipping coal seam occurrence characteristics, measured geostress data, gas pressure monitoring data, and historical mining and water and gas inrush accident data; FLAC was used. 3D The software establishes a full-scale digital twin model of the mine, with dimensions of 5500m×4500m×1200m, fully covering all mineable coal seams, major fault structures, and mining areas. Measured ground stress, surrounding rock deformation, and gas pressure data are assigned to the model, and the model's self-weight balance is calculated. Following the actual mining succession sequence, the software reconstructs the non-uniform stress field distribution in deep mining areas caused by shallow coal seam mining and disturbance of mined sections. A digital twin assimilation simulation of the stress field is conducted, and stress field contour maps are drawn. The analysis identifies mining areas II6 and II8 as high-risk areas for stress-gas coupling outbursts. Mining area II6 is designated as the priority mining area, with the mining direction being from top to bottom along the direction of maximum principal stress, i.e., along the coal seam dip, to avoid high stress concentrations caused by working face mining and coal roadway excavation, while also avoiding the stress anomaly concentration areas around faults F12 and F15.

[0037] Step S2: Construction of a refined digital twin model at the mining area scale and coupling between two scales: Taking the II6 mining area as the research object, a refined digital twin model at the mining area scale is established. The model includes elements such as the inclined longwall production system of the mining area, the steeply inclined occurrence characteristics of the coal seam, and the branch structure of the F12 fault. The mining area has developed faults and complex geological structures. The model size is set to 1800m×1500m×1000m. Based on covering the entire II6 mining area, it is expanded outward by 120m on all four sides to eliminate boundary effects. The difference in the length, width, and height ratios of the model is less than 25%, which meets the simulation requirements. The boundary space coordinates of the mining area scale model are exported, and the corresponding coordinates are matched in the full mine scale model. Stress data of 1500 sets of discrete coordinate points are extracted. The inverse distance weighted interpolation method is used to convert the discontinuous point stress data into continuous stress distribution data of the boundary surface of the mining area model. This continuous stress data is used as the boundary condition of the mining area model to complete the model calculation balance and achieve precise coupling of stress boundaries between the two scale models.

[0038] Step S3: Collaborative optimization of multi-face mining layout parameters: Based on the balanced digital twin model of the II6 mining area, a simulation test of stress superposition disturbance in multi-section mining of steeply inclined coal seams was carried out. The simulation gradients for the working face slope length (corresponding to the width of the near-horizontal coal seam mining face) were set to 60m, 80m, 100m, 120m, and 150m, and the simulation gradients for the section coal pillar width were set to 10m, 15m, 20m, 25m, and 30m, for a total of 25 sets of combined working conditions. The stress concentration coefficient, stress concentration range, and development law of the coal pillar plastic zone under different working conditions were obtained, and the functional relationship between stress concentration characteristics and working face slope length and section coal pillar width was established.

[0039] Step S4: Construction of a functional model for the relationship between filling parameters and stress concentration characteristics: Based on the optimal working face slope length of 100m and section coal pillar width of 20m, for the layered mining process of steeply inclined thick coal seams, the layered mining height is set to 3.0m~12.5m, corresponding to filling rate gradients of 60%, 70%, 80%, 85%, 90%, 95%, and 100%. The influence of the early strength performance of the filling body on the stress concentration characteristics is simulated simultaneously, and parallel numerical simulation tests under multiple working conditions are carried out. The maximum disturbance stress concentration coefficient, stress concentration range, and development law of the plastic zone of the surrounding rock are obtained under different filling parameters for continuous mining of multi-section working faces in the II6 mining area. A quantitative functional relationship between filling height, filling rate, and stress concentration degree is established by fitting. The results show that for steeply inclined coal seams, when the filling rate reaches more than 85%, the decrease in stress concentration coefficient is significantly improved. For every 5% increase in filling rate, the maximum stress concentration coefficient decreases by 0.10~0.15.

[0040] Step S5: Dynamic Control of Filling Ratio under Sequential Mining: For the three continuous working faces of 6101, 6103, and 6105 in the II6 mining area, a digital twin synchronous simulation of layered sequential filling mining is carried out, adopting a cyclical mode of "layered mining, segment-by-segment simulation, and dynamic control". Based on the non-uniform stress distribution characteristics of different sections and layers of the working face, combined with the quantitative function relationship established in Step S4, the optimal filling ratio for the current mining stage is determined: For stress concentration areas such as the periphery of fault structures and the upper and lower ends of the working face, the filling ratio is set... The filling rate is 95%~100%, using high early strength filling materials; for the peak pressure zone of the working face, the filling rate is set at 90%; for the stress stability zone in the middle of the working face, the filling rate is optimized to 85%, effectively controlling the filling cost; the stress field and gas pressure data are updated in real time as the working face advances, and the filling parameters are dynamically adjusted, ultimately achieving that the maximum stress concentration coefficient of multi-section continuous mining in the II6 mining area is controlled within 1.15. Compared with the original fixed 90% filling rate mining method, this effectively reduces the risk of gas-stress coupling outburst.

[0041] Step S6: Global Back-Substitution Verification and Advanced Precision Outburst Prevention and Control: The optimized backfilling mining strategy of the II6 mining area was back-substituted into the full-mine-scale digital twin model to conduct synchronous mining simulation of the entire mine. The simulation simulated the stress evolution and transfer patterns of the adjacent II8 and II4 mining areas during the mining process of the II6 mining area. It was predicted that an abnormal stress concentration area would appear around the F15 fault in the north of the II8 mining area and in the deep part of the II4 mining area, with a maximum stress concentration coefficient of up to 1.95, accompanied by an abnormal increase in gas pressure. Based on this prediction, advanced control measures were formulated. Before the mining of the II8 mining area, advanced hydraulic fracturing and gas pre-drainage were combined to control the stress anomaly area around the northern fault. Before the deep mining of the II4 mining area, advanced drilling and enhanced backfilling along the goaf were implemented. Ultimately, the stress concentration coefficient in the above areas was controlled within 1.15, and the gas pressure was reduced to below 0.54 MPa, reducing the risk of gas-stress coupling outburst and realizing the full-process precise control of outburst prevention and control in the ultra-deep mining of steeply inclined thick coal seams.

[0042] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for preventing coal seam outbursts by controlling the stress field during deep coal seam backfilling mining based on digital twins, characterized in that, Includes the following steps: S1. Collect all geological and mining data of the target mining area, establish a full-scale digital twin model of the mine, assign the measured data to the model and complete the self-weight balance calculation; according to the actual mining succession sequence of the mine, restore the non-uniform stress field distribution of the deep mining area, analyze the high-risk areas of stress concentration, and formulate the initial mining strategy and mining direction. S2. Taking the target mining area as the research object, establish a refined digital twin model of the mining area scale that includes the production system, geological structure, and coal seam occurrence characteristics. Derive the boundary space coordinates of the mining area scale model, match the corresponding coordinates in the whole mine scale model and extract stress data. Use interpolation mathematical methods to convert the discontinuity stress data into continuous stress data of the boundary surface of the mining area model, and complete the stress boundary assignment and calculation balance of the mining area scale model. S3. Based on the balanced digital twin model of the mining area scale, conduct simulation tests of stress superposition disturbance in multi-working-face mining, obtain the stress concentration coefficient and stress concentration range variation law under different combinations of mining face width and coal pillar width, establish the functional relationship between stress concentration characteristics and mining face width and coal pillar width, and determine the optimal mining layout parameters. S4. Based on the optimal stope layout parameters, set different mining heights and filling rate gradients, conduct parallel numerical simulation tests under multiple working conditions, obtain the maximum disturbance stress concentration coefficient and stress concentration range variation law of continuous mining of multiple working faces under different filling rates, and establish a quantitative function relationship between filling height, filling rate and stress concentration degree. S5. Conduct synchronous digital twin simulation of time-series backfilling mining at the working face. Based on the non-uniform stress distribution characteristics of different regions of the working face and combined with the quantitative function relationship established in step S4, determine the optimal backfilling rate for the current mining stage. Update the stress field data in real time as the working face advances and dynamically adjust the backfilling parameters. S6. Substitute the optimized backfilling mining strategy of the target mining area back into the digital twin model at the whole mine scale, carry out synchronous mining simulation of the whole mine, obtain the stress evolution and transfer law of adjacent and surrounding mining areas, formulate synchronous or advanced treatment measures for areas with abnormal stress concentration, and realize the anti-outburst control of the whole mine mining process.

2. The method according to claim 1, characterized in that, In step S1, the collected geological and mining data includes geological exploration data of the mining area, stress monitoring data, microseismic and acoustic emission monitoring data, and mining history data. Among them, the geological exploration data includes coal seam thickness, rock strata structure, distribution of faults and folds, and the stress monitoring data includes in-situ measured stress data, roof settlement data, and surrounding rock deformation data.

3. The method according to claim 1, characterized in that, In step S1, after the digital twin model of the entire mine scale is constructed, the stress field digital twin assimilation simulation analysis is carried out, the stress field contour map is drawn, the high-risk areas of stress concentration are marked, and the mining area and mining direction are optimized.

4. The method according to claim 1, characterized in that, In step S2, the difference in the length, width and height ratio of the digital twin model at the mining area scale does not exceed 30%, and the model expands outward on the basis of covering the target mining area to eliminate the boundary effect. The model extension distance is adjusted according to the complexity of the geological structure of the mining area. In areas with simple geological structures, the extension distance is 50-100m, while in areas with developed faults and complex structures, the extension distance is 100-150m.

5. The method according to claim 1, characterized in that, In step S2, the interpolation mathematical method adopts Kriging interpolation or inverse distance weighted interpolation to convert the discrete coordinate stress data extracted from the whole mine scale model into continuous stress distribution data of the boundary surface of the mining area scale model.

6. The method according to claim 1, characterized in that, In step S3, the simulation gradient of the face width is set to 50~300m, and the simulation gradient of the coal pillar width is set to 5~50m. The optimal mining layout parameters include the optimal face width and the coal pillar width. The determination principle is to prioritize ensuring that the stress concentration coefficient is lower than the critical threshold of coal seam outburst, while taking into account the optimal resource recovery rate and mining cost of the mining area.

7. The method according to claim 1, characterized in that, In step S4, different filling heights are used to simulate different filling rate gradients to establish the corresponding quantization function relationship.

8. The method according to claim 1, characterized in that, In step S5, the synchronous simulation of time-series filling mining adopts a dynamic cyclic mode of sequential mining, simulation, and control. It improves the filling rate in the peak pressure zone and stress anomaly zone of the working face and optimizes the filling rate in the stress stable zone to control mining costs.

9. The method according to claim 1, characterized in that, In step S6, the advanced treatment measures include one or more combinations of advanced borehole decompression, hydraulic fracturing, local enhanced filling, and coal body reinforcement. Based on the stress evolution prediction results of the full-mine scale model, the advanced treatment is completed before the stress concentration area is formed.

Images