Coal mine goaf karst tunnel construction method

Abstract

The invention relates to the technical field of tunnel construction, and discloses a coal mine goaf karst tunnel construction method which comprises the following steps: S1, performing three-dimensional scanning on the axis and the periphery of a tunnel by adopting a high-frequency geological radar in combination with aerial survey of an unmanned aerial vehicle, and accurately positioning the boundary of a goaf, the development form of a karst cave and filler distribution; and S2, integrating a seismic wave reflection method, drilling and coring data and a geological sketch result, constructing a three-dimensional geological model, and dynamically predicting a potential water inrush and mud inrush risk area. According to the method, a high-frequency geological radar is combined with aerial survey of an unmanned aerial vehicle, three-dimensional scanning is conducted on the axis and the periphery of a tunnel, geological radar data, a point cloud model and tunnel design BIM are integrated, the boundary of a goaf (the error is smaller than or equal to 0.3 m) and the karst cave form are marked, and the collapse risk level is predicted based on rock mass mechanical parameters (compressive strength and fracture density) and hydrological data (seepage paths). Therefore, the method has the advantages of advanced detection, multi-source data fusion and the like, and the resolution of geological forecast is improved.

Description

Technical Field

[0001] This invention relates to the field of tunnel construction technology, specifically a method for constructing karst tunnels in coal mine goaf areas. Background Technology

[0002] In the construction of transportation infrastructure in mountainous areas, tunnel projects often need to traverse coal-bearing strata and soluble rock areas, resulting in extremely unfavorable geological conditions where coal mine goafs and karst development zones overlap. Such complex geological environments pose extremely high construction risks, primarily manifested in the following ways: goafs may trigger water inrushes, mudslides, collapses, and gas accumulation, while karst provides well-hidden, irregularly shaped caves and water-rich channels. The combined effect of these two factors can easily lead to surrounding rock instability, support failure, and even major safety accidents.

[0003] Traditional tunnel construction methods have accumulated some technical expertise in dealing with single geological hazards, such as using advanced geological forecasting to identify faults or aquifers, reinforcing fractured surrounding rock with grouting, or ensuring the safety of high-gas tunnel operations through gas monitoring. However, when coal mine goafs and karst coexist, conventional methods reveal significant limitations: firstly, geological forecasting techniques largely rely on seismic reflection methods, which have low sensitivity to small karst caves, fissures, and other scattering bodies, making them prone to missed detections; secondly, construction and treatment measures are mostly based on experience, lacking a systematic, data-driven collaborative control process. Summary of the Invention

[0004] Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a method for constructing karst tunnels in coal mine goaf areas, which has advantages such as advanced detection and multi-source data fusion, thus solving the aforementioned problems.

[0005] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: a method for constructing karst tunnels in coal mine goaf areas, comprising the following steps: S1. High-frequency ground-penetrating radar (1000-1200MHz) combined with UAV aerial surveying is used to conduct a three-dimensional scan of the tunnel axis and a surrounding 50m range to accurately locate the boundary of the goaf, the development morphology of the karst cave, and the distribution of the filling material. S2. Integrating seismic reflection method (TSP), borehole core data and geological sketch results, a three-dimensional geological model is constructed to dynamically predict potential water and mud inrush risk areas. This scheme focuses on geological modeling rather than dynamic monitoring of the construction process. S3. For unfilled karst caves, high-pressure jet grouting piles (pile diameter 0.8-1.0m) are used to form a mesh skeleton to enhance the stability of the foundation. Nano-silicate-fly ash composite grout is injected into it to fill small cracks and block groundwater seepage (replacing the problem of easy crystallization of traditional cement grout). S4. After removing the loose filling material, backfill the graded sand and gravel mixture to 2 / 3 of the height of the karst cave, and pour lightweight foamed concrete on top to form a buffer layer. S5. The tunnel adopts a double-layer support system with an outer layer and an inner layer, including a U-shaped steel arch frame + double-layer steel mesh, an enlarged concrete foundation at the arch foot (embedded in bedrock ≥2m) to distribute the load to the stable rock layer, and the installation of deformable pressure relief anchors (4-6m in length) to allow moderate deformation of the surrounding rock to release stress and reduce the risk of sudden collapse. Closed-cell foam aluminum board (50mm thick) is laid between the primary lining and the secondary lining to absorb the impact energy of settlement in the goaf. S6. A distributed optical fiber sensing system (DTS) is used to monitor CH4 and CO concentrations in real time, and the intelligent ventilation system is linked to automatically adjust the air volume (replacing the traditional point sensor). Fiber grating strain gauges are pre-embedded in the arch and side walls, and combined with the microseismic monitoring system to capture rock mass fracture signals. When the threshold exceeds the limit, an audible and visual alarm is triggered. S7. Install a ring-shaped blind drain pipe and a water collection well, and regularly flush the pipe with high-pressure water jets. Integrate geological forecast, support stress, and environmental monitoring data based on the BIM platform, and dynamically adjust construction parameters using the fuzzy comprehensive evaluation method.

[0006] Preferably, the high-frequency ground-penetrating radar in S1 identifies underground anomalies by utilizing differences in dielectric constant. The high-frequency antenna (≥1GHz) emits electromagnetic waves, and the underground structure is analyzed by analyzing the hyperbolic characteristics of the reflected signal (typical response of karst caves) and amplitude changes (indication of filling material). The resolution for shallow layers (0-30m) reaches the centimeter level, and it can distinguish cavities and water / mud fillings. The UAV flies along a gridded path (at an altitude of 50-100m), and the radar generates a high-precision DEM model (accuracy ±5cm). Simultaneously, it acquires surface cracks and signs of subsidence, and uses hyperspectral imaging to identify lithological alteration zones. It assists in delineating high-risk areas for karst cave development. The ground-penetrating radar data, point cloud model, and tunnel design BIM are integrated to mark the boundaries of the mining area (error ≤0.3m) and the morphology of karst caves. Based on rock mass mechanical parameters (compressive strength, fracture density) and hydrological data (seepage path), the collapse risk level is predicted.

[0007] Preferably, in step S2, by arranging a source and receiver array, the reflection signals of seismic waves in the rock mass are collected, the locations of faults, fracture zones, and water-rich areas are analyzed, and machine learning algorithms are combined to improve the efficiency of multi-source data interpretation. The accuracy of identifying adverse geological bodies can reach 90%. Core drilling is used to obtain the physical and mechanical parameters (strength, permeability) and fracture development characteristics of the rock core, the rock layer interfaces and aquifer distribution are calibrated, the TSP interpretation results are verified, and quantitative parameters such as the formation permeability coefficient are provided. The TSP point cloud, borehole coordinates, and sketch location are mapped to the same three-dimensional space. The seismic wave velocity field, rock core strength data, and sketch fracture distribution are aligned using a neural network. New borehole and TSP data added during construction automatically update the model, and the risk zoning is iteratively optimized.

[0008] Preferably, the high-pressure jet grouting pile in S3 uses a 20–40 MPa high-pressure jet to cut and break up the soil, forming cylindrical cement-soil consolidation bodies with a diameter of 0.8–1.0 m in the karst cave area. These bodies overlap to form a mesh-like skeleton, suitable for geological conditions such as gravelly soil and karst cavities. It can significantly improve the integrity and deformation resistance of the foundation. The nano-silicate-fly ash slurry can penetrate micro-cracks that the high-pressure jet grouting pile cannot completely treat. The nano-silicate particles are small (approximately 10–100 nm) and can effectively seal cracks <0.1 mm. The active components of fly ash (SiO2, Al2O3) react with calcium ions to generate cementitious substances, enhancing compactness and impermeability. After the slurry solidifies, it forms a low-permeability barrier (the permeability coefficient can be reduced to 10⁻). 7 (cm / s level) to block groundwater seepage. Through the synergistic effect of high-pressure jet grouting pile skeleton + nano-composite grout, the three-in-one goal of "structural reinforcement - crack sealing - water blocking and seepage prevention" is achieved in the karst cave area, which is especially suitable for large karst caves without filling.

[0009] Preferably, in step S4, silt, quicksand, and other loose materials inside the karst cave are thoroughly removed to prevent settlement after backfilling. If groundwater leakage is present, lightweight wellpoint dewatering or cofferdam drainage should be used first to keep the working surface dry. The sand and gravel mixture must be hard, with a mud content of <5% and an organic matter content of <5%. A mixture of medium-coarse sand and pebbles / crushed stone is recommended. Backfilling should be done in layers, with each layer 20-30cm thick, and mechanically compacted (e.g., with a frog-type rammer) until the compaction coefficient is ≥0.97. The backfill height should be strictly controlled to 2 / 3 of the karst cave height. Overfilling may compress the top foamed concrete. Lightweight foamed concrete is made of cementitious cementitious material + foaming agent (physical foaming), with a density of 200-1200 kg / m³. 3 (Recommended 400-500 kg / m²) 3 (Insulation buffer layer), foaming agent aqueous solution and cement slurry weight ratio 1:10~16, stir and mix for 1-4 minutes to form a fluid slurry, pump and pour, discharge port 0.8-1.5m from the pouring surface, single pour thickness 15-20cm, reserve vent holes at the top (spaced 1-2m) to prevent gas accumulation and avoid the concentration of oversized stones, sieve and adjust gradation before backfilling, if groundwater is encountered, leakage needs to be detected after backfilling, and grouting should be added to seal if necessary, when the height of the karst cave is >3m, it is recommended to pre-embed tie bars in the sand and gravel layer to enhance the integrity, vent holes need to be set at the top of the formwork before pouring to avoid obstruction of slurry flow, graded sand and gravel provide skeleton support, foamed concrete fills the gaps and distributes the load, the two need to be tightly connected, and the foamed concrete is leveled with a screed before final setting to ensure that it fits with the superstructure.

[0010] Preferably, the double-layer support (outer initial support + inner reinforced support) in S5 is a "strong support" measure for weak surrounding rock (such as strongly weathered marl, fractured rock mass) and large-span sections (such as large-span sections of subway tunnels). The outer support (U-shaped steel arch frame + double-layer steel mesh + shotcrete) serves as the first line of defense, quickly sealing the surrounding rock and controlling initial deformation. The inner support (usually reinforced concrete or composite structure) serves as the second line of defense, further improving the structural rigidity and preventing excessive deformation from causing surface subsidence or arch collapse.

[0011] Preferably, in step S6, the absorption characteristics of CH4 / CO gas in a specific infrared band are utilized. A gas-sensitive material (such as a metal-organic framework material) is coated on the sensing fiber. The gas concentration is inverted by analyzing the wavelength shift of the Raman scattered light, replacing the traditional point sensor and achieving continuous spatial coverage. Gas leakage is accompanied by a sudden drop in local temperature (CH4 expansion and heat absorption) or an abnormal temperature rise (CO accumulation point oxidation). The DTS captures the temperature field change in real time through Raman scattering to assist in the location of gas leakage. Fiber grating strain gauges are pre-embedded in the tunnel arch and sidewalls to sense the micro-strain of the rock mass in real time. The strain value is calculated by wavelength shift to locate potential delamination zones.

[0012] Preferably, in S7, the annular blind drain and the sump form a drainage framework. High-pressure water jets ensure smooth pipe flow and prevent drainage failure caused by siltation. By integrating real-time data such as geological forecasts (e.g., rock permeability), support stress (anchor bolt stress), and environmental monitoring (water level / temperature and humidity) through BIM, a dynamic data stream is formed. Multi-source heterogeneous data (e.g., geological risk level, support structure stability, environmental fluctuations) are transformed into fuzzy indicators. Through weight allocation and membership calculation, suggestions for adjusting construction parameters are output.

[0013] Compared with the prior art, the present invention provides a method for constructing karst tunnels in coal mine goaf areas, which has the following beneficial effects: 1. This invention employs high-frequency ground-penetrating radar combined with UAV aerial surveying to perform a three-dimensional scan of the tunnel axis and its surroundings. It identifies underground anomalies by utilizing differences in dielectric constant. A high-frequency antenna (≥1GHz) emits electromagnetic waves, and by analyzing the hyperbolic characteristics of the reflected signals (typical response of karst caves) and amplitude changes (indicating infill material), the underground structure is analyzed. This allows for precise location of the goaf boundary, karst cave development morphology, and infill material distribution. By integrating the Transient Seismic Wave Reflection Method (TSP), borehole core data, and geological sketch results, a three-dimensional geological model is constructed to dynamically predict potential water and mud inrush risk areas. The ground-penetrating radar data, point cloud model, and tunnel design BIM are integrated to mark the goaf boundary (error ≤0.3m) and karst cave morphology. Based on rock mechanics parameters (compressive strength, fracture density) and hydrological data (seepage path), the collapse risk level is predicted. Therefore, this method possesses advantages such as advanced detection and multi-source data fusion, improving the resolution of geological prediction.

[0014] 2. This invention uses high-pressure jet grouting piles (pile diameter 0.8-1.0m) to form a mesh-like skeleton for unfilled karst caves, enhancing the stability of the base. Nano-silicate-fly ash composite grout is injected into it to fill small cracks and block groundwater seepage, thus replacing the problem of easy crystallization of traditional cement grout.

[0015] 3. This invention uses a distributed optical fiber sensing system (DTS) to monitor CH4 and CO concentrations in real time, and links an intelligent ventilation system to automatically adjust the air volume, replacing traditional point sensors. Fiber grating strain gauges are pre-embedded in the arch and side walls, and combined with a microseismic monitoring system to capture rock mass fracture signals. When the threshold is exceeded, an audible and visual alarm is triggered. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the method steps of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figure 1 A method for constructing karst tunnels in coal mine goaf areas includes the following steps: S1. High-frequency ground-penetrating radar (1000-1200MHz) combined with UAV aerial surveying is used to conduct a three-dimensional scan of the tunnel axis and a surrounding 50m range to accurately locate the boundary of the goaf, the development morphology of the karst cave, and the distribution of the filling material. S2. Integrating seismic reflection method (TSP), borehole core data and geological sketch results, a three-dimensional geological model is constructed to dynamically predict potential water and mud inrush risk areas. This scheme focuses on geological modeling rather than dynamic monitoring of the construction process. S3. For unfilled karst caves, high-pressure jet grouting piles (pile diameter 0.8-1.0m) are used to form a mesh skeleton to enhance the stability of the foundation. Nano-silicate-fly ash composite grout is injected into it to fill small cracks and block groundwater seepage (replacing the problem of easy crystallization of traditional cement grout). S4. After removing the loose filling material, backfill the graded sand and gravel mixture to 2 / 3 of the height of the karst cave, and pour lightweight foamed concrete on top to form a buffer layer. S5. The tunnel adopts a double-layer support system with an outer layer and an inner layer, including a U-shaped steel arch frame + double-layer steel mesh, an enlarged concrete foundation at the arch foot (embedded in bedrock ≥2m) to distribute the load to the stable rock layer, and the installation of deformable pressure relief anchors (4-6m in length) to allow moderate deformation of the surrounding rock to release stress and reduce the risk of sudden collapse. Closed-cell foam aluminum board (50mm thick) is laid between the primary lining and the secondary lining to absorb the impact energy of settlement in the goaf. S6. A distributed optical fiber sensing system (DTS) is used to monitor CH4 and CO concentrations in real time, and the intelligent ventilation system is linked to automatically adjust the air volume (replacing the traditional point sensor). Fiber grating strain gauges are pre-embedded in the arch and side walls, and combined with the microseismic monitoring system to capture rock mass fracture signals. When the threshold exceeds the limit, an audible and visual alarm is triggered. S7. Install a ring-shaped blind drain pipe and a water collection well, and regularly flush the pipe with high-pressure water jets. Integrate geological forecast, support stress, and environmental monitoring data based on the BIM platform, and dynamically adjust construction parameters using the fuzzy comprehensive evaluation method.

[0019] The S1 medium-high frequency ground-penetrating radar identifies underground anomalies by utilizing differences in dielectric constant. A high-frequency antenna (≥1GHz) emits electromagnetic waves, and the underground structure is analyzed by examining the hyperbolic characteristics of the reflected signals (typical response of karst caves) and amplitude changes (indicating infill material). It achieves centimeter-level resolution for shallow layers (0-30m) and can distinguish between cavities and water / mud fillings. A drone flies along a gridded path (50-100m altitude), and the radar generates a high-precision DEM model (accuracy ±5cm). Simultaneously, it acquires surface cracks and signs of subsidence, and uses hyperspectral imaging to identify lithological alteration zones, assisting in delineating high-risk areas for karst cave development. By integrating ground-penetrating radar data, point cloud models, and tunnel design BIM, it marks the boundaries of mined-out areas (error ≤0.3m) and karst cave morphology. Based on rock mechanics parameters (compressive strength, fracture density) and hydrological data (seepage path), it predicts the collapse risk level.

[0020] In S2, seismic source and receiver arrays are deployed to collect reflection signals of seismic waves in the rock mass, analyze the location of faults, fracture zones, and water-rich areas, and improve the efficiency of multi-source data interpretation by combining machine learning algorithms. The accuracy of identifying adverse geological bodies can reach 90%. Borehole core sampling is used to obtain the physical and mechanical parameters (strength, permeability) and fracture development characteristics of the rock core, calibrate the rock layer interfaces and aquifer distribution, verify the TSP interpretation results, and provide quantitative parameters such as formation permeability coefficient. The TSP point cloud, borehole coordinates, and sketch location are mapped to the same three-dimensional space. The seismic wave velocity field, rock core strength data, and sketch fracture distribution are aligned using neural networks. New borehole and TSP data added during construction automatically update the model and iteratively optimize the risk zoning.

[0021] S3 medium-high pressure jet grouting piles use 20–40 MPa high-pressure jets to cut and break up the soil, forming cylindrical cement-soil consolidation bodies with a diameter of 0.8–1.0 m in karst areas. These bodies interlock to form a mesh-like framework, suitable for geological conditions such as gravelly soil and karst cavities. This significantly improves the integrity and deformation resistance of the foundation. The nano-silicate-fly ash grout can penetrate micro-cracks that high-pressure jet grouting piles cannot completely treat. The nano-silicate particles are extremely small (approximately 10–100 nm), effectively sealing cracks <0.1 mm. The active components of fly ash (SiO2, Al2O3) react with calcium ions to form cementitious substances, enhancing density and impermeability. After solidification, the grout forms a low-permeability barrier (permeability coefficient can be reduced to 10⁻). 7 (cm / s level) to block groundwater seepage. Through the synergistic effect of high-pressure jet grouting pile skeleton + nano-composite grout, the three-in-one goal of "structural reinforcement - crack sealing - water blocking and seepage prevention" is achieved in the karst cave area, which is especially suitable for large karst caves without filling.

[0022] In section S4, thoroughly remove loose materials such as silt and quicksand from the karst cave to prevent settlement after backfilling. If groundwater leakage is present, use lightweight wellpoint dewatering or cofferdam drainage first, keeping the work surface dry. The sand and gravel mixture must be hard, with a mud content of <5% and an organic matter content of <5%. A mixture of medium-coarse sand and pebbles / crushed stone is recommended. Backfill in layers, each 20-30cm thick, and mechanically compact (e.g., with a frog-type rammer) until the compaction coefficient is ≥0.97. The backfill height should be strictly controlled to 2 / 3 of the karst cave height. Overfilling may compress the top foamed concrete. Lightweight foamed concrete uses cementitious cementitious materials and foaming agents (physical foaming), with a density of 200-1200 kg / m³. 3 (Recommended 400-500 kg / m²) 3 (Insulation buffer layer), foaming agent aqueous solution and cement slurry weight ratio 1:10~16, stir and mix for 1-4 minutes to form a fluid slurry, pump and pour, discharge port 0.8-1.5m from the pouring surface, single pour thickness 15-20cm, reserve vent holes at the top (spaced 1-2m) to prevent gas accumulation and avoid the concentration of oversized stones, sieve and adjust gradation before backfilling, if groundwater is encountered, leakage needs to be detected after backfilling, and grouting should be added to seal if necessary, when the height of the karst cave is >3m, it is recommended to pre-embed tie bars in the sand and gravel layer to enhance the integrity, vent holes need to be set at the top of the formwork before pouring to avoid obstruction of slurry flow, graded sand and gravel provide skeleton support, foamed concrete fills the gaps and distributes the load, the two need to be tightly connected, and the foamed concrete is leveled with a screed before final setting to ensure that it fits with the superstructure.

[0023] The double-layer support (outer initial support + inner reinforced support) in S5 is a "strong support" measure for weak surrounding rock (such as strongly weathered marl and fractured rock mass) and large-span sections (such as large-span sections of subway tunnels). The outer support (U-shaped steel arch frame + double-layer steel mesh + shotcrete) serves as the first line of defense, quickly sealing the surrounding rock and controlling initial deformation. The inner support (usually reinforced concrete or composite structure) serves as the second line of defense, further improving the structural rigidity and preventing excessive deformation from causing surface subsidence or arch collapse.

[0024] In S6, the absorption characteristics of CH4 / CO gases in specific infrared bands are utilized. Gas-sensitive materials (such as metal-organic framework materials) are coated on the sensing optical fiber. By analyzing the wavelength shift of Raman scattered light, the gas concentration is inverted, replacing the traditional point sensor and achieving continuous spatial coverage. Gas leakage is accompanied by a sudden drop in local temperature (CH4 expansion and heat absorption) or an abnormal temperature rise (CO accumulation point oxidation). DTS captures the temperature field changes in real time through Raman scattering to assist in the location of gas leakage. Fiber optic grating strain gauges are pre-embedded in the tunnel arch and sidewalls to sense the micro-strain of the rock mass in real time. The strain value is calculated by wavelength shift to locate potential delamination zones.

[0025] In S7, the ring-shaped blind drain and the sump form a drainage framework. High-pressure water jets ensure smooth pipe flow and prevent drainage failure caused by siltation. By integrating real-time data such as geological forecasts (e.g., rock permeability), support stress (anchor bolt stress), and environmental monitoring (water level / temperature and humidity) through BIM, a dynamic data stream is formed. Multi-source heterogeneous data (e.g., geological risk level, support structure stability, environmental fluctuations) are transformed into fuzzy indicators. Through weight allocation and membership calculation, suggestions for adjusting construction parameters are output.

[0026] During operation, a high-frequency ground-penetrating radar (1000-1200MHz) combined with UAV aerial surveying is used to perform a three-dimensional scan of the tunnel axis and a 50m radius around it. This accurately locates the boundaries of the goaf, the development morphology of karst caves, and the distribution of filling materials. By integrating seismic reflection (TSP) data, core drilling data, and geological sketching results, a three-dimensional geological model is constructed to dynamically predict potential water and mud inrush risk areas. For unfilled karst caves, high-pressure jet grouting piles (0.8-1.0m diameter) are used to form a mesh-like framework to enhance the stability of the foundation. Nano-silicate-fly ash composite grout is injected into these caves to fill small cracks and block groundwater seepage. After removing loose filling materials, graded sand and gravel mixture is backfilled to 2 / 3 of the karst cave height. Lightweight foamed concrete is poured on top to form a buffer layer. The tunnel adopts a double-layer support system, including a U-shaped steel arch frame and a double-layer steel mesh. An enlarged concrete foundation (embedded in bedrock ≥2m) is set at the arch foot to distribute the load to the stable rock layer. Deformable pressure-relieving anchors (4-6m in length) are installed to allow moderate deformation of the surrounding rock to release stress. A distributed fiber optic sensing system (DTS) is used to monitor CH4 and CO concentrations in real time, and the intelligent ventilation system is linked to automatically adjust the air volume. Fiber optic strain gauges are pre-embedded in the arch crown and side walls, and combined with a microseismic monitoring system to capture rock fracture signals. When the threshold is exceeded, an audible and visual alarm is triggered. A ring-shaped blind drain and a water collection well are set up, and the pipes are regularly flushed with high-pressure water jets. Geological forecasts, support stress, and environmental monitoring data are integrated based on the BIM platform, and the construction parameters are dynamically adjusted using a fuzzy comprehensive evaluation method.

[0027] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for constructing karst tunnels in coal mine goaf areas, characterized in that: Includes the following steps: S1. High-frequency ground-penetrating radar combined with UAV aerial surveying is used to conduct a three-dimensional scan of the tunnel axis and a surrounding 50m range to accurately locate the boundary of the goaf, the development morphology of the karst cave and the distribution of the filling material. S2. Integrate seismic reflection method (TSP), borehole core data and geological sketch results to construct a three-dimensional geological model and dynamically predict potential water and mud inrush risk areas; S3. For unfilled karst caves, high-pressure jet grouting piles (pile diameter 0.8-1.0m) are used to form a mesh skeleton to enhance the stability of the foundation. Nano-silicate-fly ash composite grout is injected into it to fill small cracks and block groundwater seepage. S4. After removing the loose filling material, backfill the graded sand and gravel mixture to 2 / 3 of the height of the karst cave, and pour lightweight foamed concrete on top to form a buffer layer. S5. The tunnel adopts a double-layer support system with an outer layer and an inner layer, including a U-shaped steel arch frame + double-layer steel mesh, an enlarged concrete foundation (embedded in bedrock ≥2m) at the arch foot to distribute the load to the stable rock layer, and the installation of deformable pressure relief anchors (4-6m in length) to allow moderate deformation of the surrounding rock to release stress and reduce the risk of sudden collapse. S6. A distributed optical fiber sensing system (DTS) is used to monitor CH4 and CO concentrations in real time. Fiber grating strain gauges are pre-embedded in the arch and sidewalls. Combined with a microseismic monitoring system, rock mass fracture signals are captured. When the threshold is exceeded, an audible and visual alarm is triggered. S7. Install a ring-shaped blind drain pipe and a water collection well, and regularly flush the pipe with high-pressure water jets. Integrate geological forecast, support stress, and environmental monitoring data based on the BIM platform, and dynamically adjust construction parameters using the fuzzy comprehensive evaluation method.

2. The method for constructing karst tunnels in coal mine goaf areas according to claim 1, characterized in that: The S1 high-frequency ground-penetrating radar identifies underground anomalies by utilizing differences in dielectric constant. The high-frequency antenna (≥1GHz) emits electromagnetic waves, and the underground structure is analyzed by analyzing the hyperbolic characteristics of the reflected signal (typical response of karst caves) and amplitude changes (indicating filling material). It achieves centimeter-level resolution for shallow layers (0-30m) and can distinguish cavities and water / mud fillings. The UAV flies along a gridded path (at an altitude of 50-100m), and the radar generates a high-precision DEM model (accuracy ±5cm). Simultaneously, it acquires surface cracks and signs of subsidence, and uses hyperspectral imaging to identify lithological alteration zones. It also helps to delineate high-risk areas for karst cave development. The ground-penetrating radar data, point cloud model, and tunnel design BIM are integrated to mark the boundaries of mined-out areas (error ≤0.3m) and the morphology of karst caves. Based on rock mass mechanical parameters (compressive strength, fracture density) and hydrological data (seepage path), the collapse risk level is predicted.

3. The method for constructing karst tunnels in coal mine goaf areas according to claim 1, characterized in that: In S2, seismic wave reflection signals in the rock mass are collected by arranging seismic source and receiver arrays, analyzing the location of faults, fracture zones and water-rich areas, and improving the efficiency of multi-source data interpretation by combining machine learning algorithms. The accuracy of identifying adverse geological bodies can reach 90%. Core drilling is used to obtain the physical and mechanical parameters (strength, permeability) and fracture development characteristics of the rock core, calibrate the rock layer interface and aquifer distribution, verify the TSP interpretation results, and provide quantitative parameters such as formation permeability coefficient. The TSP point cloud, borehole coordinates and sketch location are mapped to the same three-dimensional space, and the seismic wave velocity field, rock core strength data and sketch fracture distribution are aligned using neural networks.

4. The method for constructing karst tunnels in coal mine goaf areas according to claim 1, characterized in that: The S3 high-pressure jet grouting pile uses a 20–40 MPa high-pressure jet to cut and break up the soil, forming cylindrical cement-soil consolidation bodies with a diameter of 0.8–1.0 m in karst areas. These bodies overlap to form a mesh-like framework, suitable for geological conditions such as gravelly soil and karst cavities. It can significantly improve the integrity and deformation resistance of the foundation. The nano-silicate-fly ash slurry can penetrate micro-cracks that high-pressure jet grouting piles cannot completely treat. The nano-silicate particles are very small (approximately 10–100 nm), effectively sealing cracks <0.1 mm. The active components of fly ash (SiO2, Al2O3) react with calcium ions to form cementitious substances, enhancing compactness and impermeability. After the slurry solidifies, it forms a low-permeability barrier (permeability coefficient can be reduced to 10⁻). 7 (cm / s level) to block groundwater seepage. Through the synergistic effect of high-pressure jet grouting pile skeleton + nano-composite grout, the three-in-one goal of "structural reinforcement - crack sealing - water blocking and seepage prevention" is achieved in the karst cave area, which is especially suitable for large karst caves without filling.

5. A method for constructing karst tunnels in coal mine goaf areas according to claim 1, characterized in that: In step S4, thoroughly remove loose materials such as silt and quicksand from the karst cave to prevent settlement after backfilling. If groundwater leakage is present, use lightweight wellpoint dewatering or cofferdam drainage first, keeping the work surface dry. The sand and gravel mixture must be hard, with a mud content of <5% and an organic matter content of <5%. A mixture of medium-coarse sand and pebbles / crushed stone is recommended. Backfill in layers, each 20-30cm thick, and mechanically compact (e.g., with a frog-type rammer) until the compaction coefficient is ≥0.

97. The backfill height should be strictly controlled to 2 / 3 of the karst cave height. Overfilling may compress the top foamed concrete. Lightweight foamed concrete uses cementitious cementitious materials and foaming agents (physical foaming), with a density of 200-1200 kg / m³. 3 (Recommended 400-500 kg / m²) 3 (Insulation buffer layer), foaming agent aqueous solution and cement slurry weight ratio 1:10~16, stir and mix for 1-4 minutes to form a fluid slurry.

6. The method for constructing karst tunnels in coal mine goaf areas according to claim 1, characterized in that: The double-layer support (outer initial support + inner reinforced support) in S5 is a "strong support" measure for weak surrounding rock (such as strongly weathered marl, fractured rock mass) and large-span sections (such as large-span sections of subway tunnels). The outer support (U-shaped steel arch frame + double-layer steel mesh + shotcrete) serves as the first line of defense, quickly sealing the surrounding rock and controlling initial deformation. The inner support (usually reinforced concrete or composite structure) serves as the second line of defense, further improving the structural rigidity and preventing excessive deformation from causing surface subsidence or arch collapse.

7. A method for constructing karst tunnels in coal mine goaf areas according to claim 1, characterized in that: In the S6, the absorption characteristics of CH4 / CO gas in a specific infrared band are utilized. A gas-sensitive material (such as a metal-organic framework material) is coated on the sensing fiber. The gas concentration is inverted by analyzing the wavelength shift of the Raman scattered light, replacing the traditional point sensor and achieving continuous spatial coverage. Gas leakage is accompanied by a sudden drop in local temperature (CH4 expansion and heat absorption) or an abnormal temperature rise (CO accumulation point oxidation). The DTS captures the temperature field change in real time through Raman scattering to assist in the location of gas leakage. Fiber grating strain gauges are pre-embedded in the tunnel arch and sidewalls to sense the micro-strain of the rock mass in real time. The strain value is calculated by wavelength shift to locate potential delamination zones.

8. A method for constructing karst tunnels in coal mine goaf areas according to claim 1, characterized in that: In the S7, the annular blind drain and the collection well form a drainage framework. High-pressure water jets ensure smooth pipe flow and prevent drainage failure caused by siltation. By integrating real-time data such as geological forecasts (e.g., rock permeability), support stress (anchor bolt stress), and environmental monitoring (water level / temperature and humidity) through BIM, a dynamic data stream is formed. Multi-source heterogeneous data (e.g., geological risk level, support structure stability, environmental fluctuations) are transformed into fuzzy indicators. Through weight allocation and membership calculation, suggestions for adjusting construction parameters are output.

Images