Treatment method for building door tunnel penetrating karst underground river section

Abstract

The invention belongs to the technical field of tunnels, and particularly relates to a treatment method for a building door tunnel penetrating a karst underground river reach, which comprises the following steps: step 1, multi-source geophysical prospecting fusion investigation: aiming at the tunnel karst underground river reach and the range of 500m around the tunnel karst underground river reach; the method comprises the following steps of: delineating a karst abnormal area by adopting semi-aviation transient electromagnetism; screening a strong corrosion area by adopting a depression analysis method; verifying parameters of the abnormal area through surface drilling and coring; identifying surface hydrology, lithology and seepage data characteristics through unmanned aerial vehicle-mounted high-resolution multispectral and thermal infrared cooperative remote sensing; karst water dynamic conditions including underground water flow velocity, flow direction and permeability coefficient, corrosion evolution characteristics including cavern forming time and expansion rate, and key physical and mechanical parameters including cavern filler porosity, permeability coefficient, surrounding rock integrity coefficient, uniaxial compressive strength, elastic modulus and Poisson's ratio are obtained. According to the treatment method for the building door tunnel to penetrate through the karst underground river section, the karst space characteristics are clearly presented through the arrangement of the treatment method for the building door tunnel to penetrate through the karst underground river section.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunnel, in particular to a treatment method for a karst underground river section of a tunnel. BACKGROUND

[0002] In the field of tunnel engineering construction, crossing karst underground river section is a typical high-risk and high-difficulty working condition, especially in the area where the tunnel is located. Influenced by long-term karstification, the geological conditions present three characteristics: interlaced underground rivers, string-shaped dissolution cavity development, and significant difference in surrounding rock integrity. If the treatment scheme lacks scientificity, it is easy to cause engineering accidents such as collapse, gushing and mud inrush, and structure cracking and leakage in later stage, which directly threatens the construction safety and tunnel operation life.

[0003] The prior art generally relies on a single geophysical prospecting method to carry out preliminary investigation, lacks collaborative inversion and verification of multi-source data, and leads to large deviation in the measurement of karst parameters (hydrodynamic conditions, dissolution evolution characteristics, and mechanical parameters of surrounding rock), which cannot provide reliable basis for subsequent treatment. The prior art mostly uses two-dimensional geological profile or simplified three-dimensional model to represent the geological conditions of karst area, and does not construct a layered refined model matched with the actual working condition, which has large model error and weak guidance. The existing treatment scheme is mostly based on engineering experience and static design, and does not dynamically adapt and optimize the scheme in combination with the investigation data and three-dimensional model, which leads to "excessive treatment" or "insufficient treatment", imbalance between safety risk and economy, and the three links of investigation, modeling and treatment are disconnected, and the treatment measures do not predict long-term risks in combination with dissolution evolution characteristics (such as dissolution cavity expansion rate), which only meet the short-term stability, and the structure is prone to disease in later stage due to the expansion of dissolution cavity, which needs to be reworked multiple times, significantly increasing the engineering cost and construction period. SUMMARY

[0004] In view of the technical problems that the prior art has significant defects in the investigation accuracy, modeling level, treatment pertinence and collaboration in the treatment technology for the tunnel crossing karst underground river section, the present application provides a treatment method for a karst underground river section of a tunnel.

[0005] The treatment method for a karst underground river section of a tunnel provided by the present application comprises the following steps: step one: multi-source geophysical prospecting integrated investigation; for the tunnel karst underground river section and the surrounding 500m range, the following four data are jointly inverted to obtain the karst parameters: semi-airborne transient electromagnetic method is used to delineate the karst abnormal area, depression analysis method is used to screen the strong dissolution area, surface drilling core is taken to verify the abnormal area parameters, and unmanned aerial vehicle high-resolution multispectral and thermal infrared collaborative remote sensing is used to identify the surface hydrology, lithology and seepage data characteristics; the karst hydrodynamic conditions: groundwater flow rate, flow direction and rock mass permeability coefficient, dissolution evolution characteristics: dissolution cavity formation age and expansion rate, and key physical and mechanical parameters: dissolution cavity filling porosity, filling permeability coefficient, surrounding rock integrity coefficient, uniaxial compressive strength, elastic modulus and Poisson's ratio are obtained.

[0006] Step 2: 3D geological modeling: Based on lidar surface scanning with a point cloud density ≥100 points / m² and elevation error ≤5cm, using karst space data from ground-penetrating radar 3D imaging, micro-fracture data from advanced horizontal borehole imaging, and surface hydrological, lithological, and seepage data acquired by UAV-borne high-resolution multispectral and thermal infrared collaborative remote sensing, a layered 3D geological model of "surface layer-karst layer-surrounding rock layer" is constructed. The model parameters are verified and optimized through borehole data inversion, clarifying the direction of the underground river, the 3D coordinates and volume of the karst morphology, and the mechanical parameters of the surrounding rock.

[0007] Step 3: Adaptation of treatment measures: Based on the output data of Step 1 and Step 2, optimize the construction parameters of the pile foundation raft slab, use the positioning mechanism to complete the reinforcement scheme of the top of the karst cavity and the design of the drainage system, so as to achieve precise treatment of the karst underground river section.

[0008] Preferably, the semi-airborne transient electromagnetic data acquisition process in step one is as follows:

[0009] Instrument selection and parameter settings: The GDP-32Ⅱ multi-functional electrical resistivity meter is used, equipped with a 20m×20m rectangular transmitting coil with 8 turns. The receiving coil is mounted on the UAV. The flight path is zigzag. The horizontal distance between the receiving coil and the center of the transmitting coil is ≥30m. The flight altitude is 50m and the flight speed is 5m / s. The transmission frequency is 1Hz for deep detection depth ≥50m and 4Hz for shallow detection depth 10-30m. The peak current intensity is 5A and the sampling time window is 0.1ms-100ms.

[0010] Transmitting coil deployment: A total of 8 longitudinal survey lines and 12 transverse survey lines are set up. The transmitting coils are laid along the center of the survey lines and fixed to the ground surface.

[0011] Data recording: Three valid signals were collected at each measurement point. Outliers with a signal-to-noise ratio (SNR) < 3 were removed, and the transient electromagnetic response curve was recorded. The value is the instrument observation value, which is then calculated by inversely using the induced electromotive force from the receiving coil. ,in, To induce electromotive force;

[0012] Semi-airborne transient electromagnetic data processing: Data was processed using EMIGMA software, with the following steps: mean filtering, 5×5 window size, wavelet thresholding for noise reduction, db4 wavelet decomposition to 3 levels to eliminate high-frequency electromagnetic noise, and calculation of apparent resistivity using the late transient field formula. : ,in, The observation time after current cutoff is defined, with a sampling window of 0.1ms-100ms. ,in, This is the numerical value of the free magnetic permeability, taken as 4π × 10⁻⁶. -7 H / m, The equivalent radius of the transmitting coil ,constant The formula for calculation is: , The number of turns of the transmitting coil. The number of turns of the receiving coil. This represents the peak value of the step current in the transmitting coil. The area of ​​the transmitting coil, For the area of ​​the receiving coil, The horizontal distance from the center of the transmitting coil to the receiving coil. ;

[0013] Abnormal zone determination: This involves considering the apparent resistivity. Areas with an Ω·m value less than 50 Ω·m and lower than 30% of the average value of the surrounding rock are identified as karst anomaly zones.

[0014] Preferably, the process for screening strongly eroded areas using the depression analysis method in step one is as follows: DEM data acquisition and preprocessing: 5cm resolution elevation data is acquired using UAV aerial surveying, generating a 1m×1m raster DEM. Elevation anomalies are removed using ArcGIS software to eliminate micro depressions with an area <100㎡ and a depth <0.5m.

[0015] Calculate the area of ​​the depression : ,in, The first in the depression The area of ​​the grid, This represents the total number of grid cells within the depression.

[0016] Depth of depression : ,in The lowest elevation of the watershed surrounding the depression was extracted using ArcGIS flow direction analysis tools. The lowest elevation inside the depression was extracted using the ArcGIS raster calculator.

[0017] catchment area : ,in, For the first to flow into the depression The area of ​​each grid cell The total number of raster cells in the catchment area is calculated using ArcGIS's catchment analysis tools, with the D8 method used for flow direction.

[0018] Dissolution Intensity Index : ,when Areas larger than 5m² are marked as strong karst zones, corresponding to underground rivers or large solution cavities;

[0019] Criteria for determining key exploration areas: When the depression is... >5m² and corresponding to semi-aerospace transient electromagnetic apparent resistivity When the high conductivity anomaly zone of <30Ω・m, the surface seepage and lithological anomaly zone identified by UAV-borne high-resolution multispectral and thermal infrared collaborative remote sensing are superimposed, the superimposed area is marked as a key exploration area using ArcGIS spatial overlay analysis tools, and used as a subsequent borehole core sampling point.

[0020] Preferably, the process for verifying the parameters of the abnormal area by surface core drilling in step one is as follows:

[0021] Drilling layout: Three verification boreholes, numbered ZK1-ZK3, were set up in the key exploration area. Borehole ZK1 was located in the center of the depression, with a depth of 60m and a target stratum of exposing the beaded karst cave. Borehole ZK2 was located in the area of ​​apparent resistivity anomaly, with a depth of 55m and a target stratum of the top plate of the karst cave and a 2.6m thin rock layer. Borehole ZK3 was located in the area of ​​water catchment node, with a depth of 50m and a target stratum of the confluence of groundwater flows.

[0022] Drilling parameters: The XY-2 core drilling rig was used with a diameter of 110mm. The core recovery rate in the bedrock section was ≥90%, and the core recovery rate in the filling section was ≥85%. Lithology and karst cavities parameters were recorded every 1m, and the initial groundwater level and stable groundwater level were observed simultaneously.

[0023] Groundwater observation: When the borehole exposes the aquifer, the initial water level is recorded using a measuring rope. Stable water level Calculate the drawdown : ;

[0024] Parameter calculation: For core samples taken from boreholes, one group of samples was taken every 5m, with 3 parallel samples in each group, and the test was conducted. The calculation formula is as follows:

[0025] Porosity : ,in, The total volume of the core was determined using the helium displacement method. The solid volume of the rock core is calculated using the following formula: , For core quality, the accuracy of electronic balance measurement, The density of the rock was determined using the hydrostatic bottle method for limestone. =2.65g / cm³, filling material =1.9-2.1 g / cm³;

[0026] Permeability coefficient : ,in, To determine the cross-sectional area of ​​the variable head pipe, we take 10 cm². The sample length is 5cm. The sample cross-sectional area is 78.5 cm², and the diameter is 10 cm. For the water head from Down to Time, The initial head, The final head;

[0027] Groundwater flow velocity was calculated using a borehole tracer test, in which NaC was injected into test borehole ZK1. l The tracer, at a concentration of 5%, was used to monitor concentration changes in downstream test well ZK3, and the time of peak occurrence was recorded. ;

[0028] groundwater flow rate : ,in, The horizontal distance between the two boreholes was measured using GPS. 5% NaC as tracer l Solution peak time, The angle between the water flow direction and the line connecting the borehole is determined by the apparent resistivity contour map.

[0029] Analysis of dissolution evolution characteristics: The dissolution pattern was clarified through dating and expansion rate calculation;

[0030] Age of formation of the cavity Uranium-series dating was used to determine the age of calcite deposits in the dissolution fractures of the ZK2 limestone sample. The unit is 10,000 years, representing the formation age of the cavity. Calcite precipitation and delayed dissolution take 20,000 years;

[0031] Cavity expansion rate : ,in, The initial cavity height is set to 0.5m. This represents the current cavity height. .

[0032] Preferably, the process of identifying surface hydrological, lithological, and seepage characteristics using UAV-borne high-resolution multispectral and thermal infrared collaborative remote sensing in step one is as follows:

[0033] Equipment selection and parameter settings: A multispectral camera with a resolution ≤0.1m and a wavelength band of 450-900nm is used; a thermal infrared imager with a temperature resolution ≤0.5℃ and a wavelength band of 8-14μm is used. The imager is mounted on a hexacopter UAV with a flight altitude of 30-50m, a forward overlap of 80%, a lateral overlap of 60%, and a sampling interval of 0.5s. The coordinates of three GPS control points that coincide with the lidar are collected simultaneously.

[0034] Data acquisition: One full-area aerial survey each during the dry season and the rainy season to acquire multispectral and thermal infrared images;

[0035] Multispectral data processing: Radiometric calibration was performed using ENVI software, with the following formula: ,in, A specific band in a multispectral image Earth's surface radiance, For this specific band The radiation gain coefficient, For this specific band The original digital quantization value, For this specific band radiative offset coefficient;

[0036] After atmospheric correction, surface runoff gullies and waterlogged areas are extracted using the normalized water index. The calculation formula is as follows: ,in, The normalized water index, This represents the surface reflectance at the 560nm band (green band) in the multispectral image. Surface reflectance in the 840nm band (near-infrared band) of multispectral imagery;

[0037] Lithology is identified using a spectral angle matching algorithm, calculated as follows: ,in, The cosine value of the spectral angle. For band number, The total number of bands participating in spectral matching. For the target pixel in the image at the th Surface reflectance in each band, For the standard lithological end-member in the first Surface reflectance in each band;

[0038] Limestone: 560nm characteristic reflection peak; Dolomite: 620nm characteristic reflection peak; Soil layer: 750nm reflection valley. Generate a lithological zoning map and calculate the surface dissolution rate. ,in, Surface dissolution rate, For the area of ​​karst landforms, The area covered by the image;

[0039] Thermal infrared data processing: Temperature inversion was performed using ArcGIS software, with the following formula: ,in, For surface temperature, , This refers to the factory calibration coefficients of the thermal infrared imager. It is the natural logarithm function. The radiance of thermal infrared images;

[0040] extract Low temperature anomaly zone ≤20℃ ,in, For temperature gradient, , They are respectively the first The, the The subsurface temperature at each sampling point , The first The, the Calculate the temperature gradient at the underground depth of each sampling point. >0.5℃ / m indicates a strong seepage zone. Low-temperature anomaly zones with temperatures ≤20℃ are extracted, including corresponding springs and seepage channels. Through temperature gradient analysis, areas with temperature gradients >0.5℃ / m are marked, corresponding to the upward discharge zone of underground seepage.

[0041] Feature recognition and output results: Overlaying multispectral lithological zoning maps, NDWI water body extraction maps, and thermal infrared low-temperature anomaly maps, according to... ,in, To calculate the overall score, For lithology score, To score points for the water body, Score for seepage;

[0042] when Areas with a score of ≥7 are designated as surface seepage-lithological anomaly zones. Output a map containing surface runoff, spring coordinates, and lithological boundaries, which will serve as the input for surface data inversion based on these four metrics.

[0043] Joint inversion is achieved using weight allocation and least squares fitting:

[0044] Data weights: Semi-airborne transient electromagnetic 40%, reflecting underground electrical anomalies; depression analysis 25%, reflecting surface dissolution intensity; core drilling 20%, directly verifying parameters; UAV remote sensing 15%, reflecting surface-subsurface correlation characteristics.

[0045] Inversion process: Using borehole core parameters as true values, the parameter deviations of other data sources are corrected by the least squares method, and finally a unified karst parameter dataset is output: hydrodynamic, dissolution evolution and mechanical parameters.

[0046] Preferably, the lidar scanning process in step two is as follows: using a RIEGLVZ-6000 device, a scanning distance of 500m, a point cloud density of 100 points / m², a scanning frequency of 500kHz, and generating a surface point cloud model after noise reduction using RiSCANPRO software and registration based on 3 GPS control points, with a planar accuracy of ≤5cm.

[0047] 3D imaging workflow of ground-penetrating radar: Using an SIR-4000 device with a 250MHz shielded antenna, a survey network is established with 5m spacing along the tunnel axis and 10m spacing perpendicular to the axis. The antenna moving speed is 0.5m / s, the sampling rate is 1024 points / scan, the recording length is 400ns, corresponding to a depth ≥30m. The electromagnetic wave velocity in limestone is 0.15m / ns. Processing steps:

[0048] Gain correction: Exponential gain correction is used: gain coefficient ,in, for Signal gain coefficient at time 10:00 Sampling time of ground-penetrating radar signal, gain attenuation compensation coefficient Initial gain coefficient To compensate for signal attenuation in deep areas;

[0049] Filtering: Frequency domain bandpass filtering, ranging from 50-250MHz, to eliminate high-frequency noise;

[0050] 3D offset imaging: The Kirchhoff offset algorithm is used with an offset speed of v=0.15m / ns to correct image distortion caused by signal delay and generate a 3D radar profile to identify cavity reflection features: phase axis interruption and strong reflection area;

[0051] Advanced horizontal borehole imaging: A borehole with a diameter of 90mm and a depth of 30m is laid in the tunnel, with an angle of 15° to the tunnel axis. An Olympus IPLEXFX endoscope with a resolution of 1280×720 is used to take images every 10cm at a speed of 5cm / s. The distance from the borehole opening and the diameter of the erosion cavities, as well as the orientation, dip angle and width of the fractures are recorded.

[0052] Preferably, in step two, the coordinate system of the three-dimensional model and the layered construction are as follows: Coordinate transformation: the three-dimensional geological model is converted from the WGS84 coordinate system. system, This refers to the tunnel axis mileage. This refers to the lateral distance along the vertical axis, ranging from -50 to 50 meters. The elevation ranges from -30 to 50 meters, with the tunnel track surface as ±0.

[0053] The three-dimensional geological model is constructed in layers, including: based on FLAC3D 7.0 software: surface layer. ≥0: A 1m×1m grid is generated based on the lidar point cloud through Delaunay triangulation, and the sinkhole is marked by superimposing the depression analysis results;

[0054] cavity layer -30≤ <0: Extract ≥500 cavity boundary points from the ground-penetrating radar profile. The three-dimensional surface of the cavity is generated using a quadratic polynomial surface: , The fitting coefficients are calculated using the least squares method, and the residuals are ≤0.5m.

[0055] Surrounding rock layer <-30: Classified by apparent resistivity ≥200Ω·m is considered intact surrounding rock, 100< <200Ω·m indicates relatively intact surrounding rock. For rock with an Ω·m value less than 100Ω·m, an interpolation algorithm is used to generate a rock integrity coefficient grid.

[0056] Calculation and Model Validation of Surrounding Rock Mechanical Parameters: Surrounding Rock Integrity Coefficient : ,in, Core recovery rate To measure the acoustic wave velocity during drilling, =5.5km / s, using an RS-ST01C acoustic wave detector, one point is measured every 2m. The standard wave velocity for fresh limestone is taken as 6.0 km / s;

[0057] Conversion of surrounding rock mechanical parameters: Uniaxial compressive strength : These are core test values;

[0058] elastic modulus : ;

[0059] Poisson's ratio ;

[0060] The three-dimensional geological model was validated using borehole data inversion: the elevations of the actual exposed top and bottom plates of the karst cavity in borehole ZK2 were compared with the model calculations. The error was ≤ ±0.2m. If the error was > 0.2m, the surface fitting coefficients were readjusted. Continue until the accuracy requirements are met;

[0061] Data fusion correction: integrating semi-airborne transient electromagnetic data. Abnormal area The superposition of <50Ω·m, the interruption of the phase axis of the cavity reflection zone from ground-penetrating radar, and the surface seepage and lithological anomaly zones detected by UAV remote sensing, results in an anomaly confidence level of ≥90% in the superimposed area. This is verified through borehole testing; for example, borehole ZK2 reveals a cavity that matches the superimposed area. Non-superimposed areas are corrected using borehole data, such as semi-airborne transient electromagnetic data. <50Ω·m, but with no radar reflection zone, it is determined to be a water-rich fissure rather than a karst cave.

[0062] Preferably, the pile foundation raft foundation construction in step three is optimized as follows: For the beaded karst caves in the bottom slab:

[0063] Pile location adjustment: Based on the distribution of beaded karst caves in a 3D model, and the filling area <30Ω·m, avoid areas with infill material, and arrange the pile locations in intact surrounding rock areas. ≥200Ω·m;

[0064] Pile length optimization: Increase the pile length by 1-2m according to the depth of the karst cavity. For example, if the maximum depth of the karst cavity is 29.1m, the pile length should be 31m to ensure that the pile tip penetrates ≥3m into the stable rock layer. The integrity coefficient of the surrounding rock layer is then used to determine the optimal length. ≥0.8 region determination;

[0065] Raft design: A 1.2m thick C30 reinforced concrete raft slab is poured on top of the piles. It is made of double-layer bidirectional Φ20×150 steel bars. The edge of the raft slab extends to the side wall of the karst cavity by ≥0.5m and connects with the shotcrete layer of the side wall of the karst cavity.

[0066] Optimization of cavity top reinforcement: for thin rock layers at the top: reinforcement range: calculated using a 3D model. <0.6 weak area;

[0067] Reinforcement parameters: The density of the shotcrete anchor rods is adjusted from the conventional 100cm×100cm to 80cm×80cm. The anchor rods are 600cm long Φ22 cement mortar anchor rods with an anchoring depth ≥1.5m. A 20cm×20cm Φ8 steel mesh is hung, and 10cm thick C25 concrete is shotcreted with a compressive strength ≥25MPa.

[0068] Impact buffer: A 50cm thick layer of waste tires is laid in the top suspended section. The tires are arranged laterally and the gaps are filled with C15 fine stone concrete to resist the impact of sporadic rockfalls from the top of the tunnel. The impact load is ≤50kN, which is verified by drop hammer test.

[0069] Drainage system design: Maximum inflow calculation: Atmospheric rainfall infiltration method, formula: ,in, The rainfall infiltration coefficient in karst areas This refers to the average annual rainfall over many years.

[0070] Inverted siphon design: Two 2m×2m reinforced concrete inverted siphons are used to increase flow capacity. ,in, The flow area of ​​the inverted siphon is =2m×2m reinforced concrete structure, 4m 2,, The average flow velocity of water within the inverted siphon pipe. This refers to the acceleration caused by gravity on the water flow, with a value of 9.8 m / s².2 , The difference in water level between the inverted siphon inlet and outlet;

[0071] Surface blind drains: Base blind drains are set up on the surface, with the main drain being 60cm×50cm and the branch drains being 40cm×30cm, spaced 10m apart. Φ100mm permeable pipes are laid in the blind drains and wrapped with geotextile to guide groundwater to the original drainage channel.

[0072] Optimization of biased support: For biased Class V surrounding rock, a complete pressure arch is formed:

[0073] Support parameters: The support adopts the biased pressure Class V surrounding rock support. The primary support consists of 25cm thick C25 shotcrete and Φ25 hollow grouting anchors with a spacing of 80cm×80cm and I20a type steel arch frame with a spacing of 60cm. The secondary lining is 50cm thick C35 reinforced concrete.

[0074] Eccentric pressure vent wall: A 2-3m thick C30 reinforced concrete vent wall shall be added to the open area on the right side of the tunnel. The vent wall foundation shall be an enlarged foundation with an enlarged dimension of ≥0.5m×0.5m. If necessary, Φ22 anchor bolts with a length of 4m shall be added.

[0075] Top backfill: 2-3m thick C20 concrete is backfilled in the exposed areas at the top of the tunnel to ensure the formation of a complete pressure arch. The arch stress is verified to be ≤20MPa through FLAC3D numerical simulation.

[0076] Preferably, the positioning mechanism in step three includes a positioning housing, a positioning frame with a rack slidably inserted into the inner wall of the positioning housing, a gear set rotatably connected to the inner wall of the positioning housing, one gear of the gear set meshing with the rack of the positioning frame, a drive gear ring rotatably connected to the inner wall of the positioning housing, the drive gear ring meshing with another gear of the gear set, and a drive gear rotatably connected to the outer surface of the positioning housing, the drive gear meshing with the drive gear ring.

[0077] Preferably, a collar is rotatably connected to the outer surface of the positioning housing, and a connecting block assembly is fixedly installed on the inner wall of the collar. The connecting block assembly is composed of multiple connecting blocks that are hinged end to end by pins. A magnetic block is fixedly installed on the outer surface of the connecting block assembly, and the outer surface of the magnetic block is magnetically connected to the outer surface of the collar. A universal tube is fixedly installed on the inner wall of the connecting block assembly. A measuring tape is fixedly installed on the inner wall of one connecting block of the connecting block assembly. One end of the measuring tape passes through the inner walls of the remaining connecting blocks of the connecting block assembly and extends to the outside. A transparent plate is fixedly installed on the outer surface of the connecting block assembly.

[0078] The beneficial effects of this invention are as follows:

[0079] 1. By setting up a treatment method for tunnels passing through karst underground rivers, the deep fusion of multi-source exploration data significantly improves the accuracy of karst parameter acquisition, providing a reliable geological basis for treatment design. This method combines four technologies: semi-airborne transient electromagnetic sampling to delineate anomaly areas, depression analysis to screen for strongly karstified areas, surface borehole core sampling for verification, and UAV-borne high-resolution multispectral and thermal infrared collaborative remote sensing to identify surface hydrological, lithological, and seepage data characteristics. This achieves comprehensive and high-precision acquisition of key karst parameters, effectively solving the problems of large parameter calculation deviations and high misjudgment rates of anomaly areas in existing technologies. It makes the parameter dimensions more complete, the anomaly area judgment more accurate, and the data verification more rigorous, thus solving the technical problem of exploration accuracy in existing treatment technologies for tunnels passing through karst underground rivers.

[0080] 2. By setting up a treatment method for tunnels passing through karst underground river sections, a layered three-dimensional geological model is accurately constructed, clearly presenting the spatial characteristics of karst, and realizing the visualization and refined design of treatment measures. This method adopts a refined three-dimensional geological model of surface layer, karst cavity layer, and surrounding rock layer, making the model input data more accurate, the layered structure more in line with reality, and the model verification and correction more rigorous, thus solving the technical problems of the existing modeling level for the treatment of tunnels passing through karst underground river sections.

[0081] 3. By setting up a treatment method for tunnels passing through karst underground rivers, the treatment measures are precisely adapted to the geological conditions, effectively resisting engineering risks in karst areas and ensuring the safety of tunnel construction and long-term stable operation. Based on the parameters of the previous exploration and the results of the three-dimensional model, this method optimizes the core treatment measures such as pile foundation raft slab, cavity reinforcement, and drainage system. It solves the problems of the existing technology, such as the measures being crude, poor adaptability, and easy to cause collapse and water inrush. The pile foundation raft slab design is safer and more reliable, the cavity top reinforcement is more targeted, and the drainage system design is more scientific and efficient. It solves the technical problems that the existing treatment technology for tunnels passing through karst underground rivers has significant defects in terms of treatment targeting and coordination. Attached Figure Description

[0082] Figure 1 This is a schematic diagram of a method for treating a karst underground river section through a building entrance tunnel, as proposed in this invention.

[0083] Figure 2 This is a perspective view of the positioning shell structure of a method for treating a tunnel through a karst underground river section proposed in this invention.

[0084] Figure 3 This is a perspective view of the drive gear ring structure of a method for treating a karst underground river section through a building tunnel proposed in this invention.

[0085] Figure 4 This is a perspective view of the ring structure of a method for treating a tunnel through a karst underground river section proposed in this invention.

[0086] Figure 5 This is a perspective view of the connecting block assembly structure of a method for treating a karst underground river section through a building tunnel proposed in this invention.

[0087] In the diagram: 1. Positioning housing; 2. Positioning frame; 21. Gear set; 22. Drive gear ring; 23. Drive gear; 3. Collar; 31. Connecting block assembly; 32. Magnetic block; 33. Universal tube; 34. Measuring tape; 35. Transparent plate. Detailed Implementation

[0088] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0089] Example 1

[0090] Reference Figure 1 A method for treating a tunnel through a karst underground river section includes the following steps: Step 1: Multi-source geophysical exploration: For the karst underground river section of the tunnel and the surrounding 500m range, the following methods are used: semi-airborne transient electromagnetic mapping to delineate karst anomaly zones, depression analysis to screen for strong dissolution zones, surface borehole core sampling to verify anomaly zone parameters, and UAV-borne high-resolution multispectral and thermal infrared collaborative remote sensing to identify surface hydrological, lithological, and seepage data characteristics. The four data are combined to obtain karst parameters, including karst hydrodynamic conditions: groundwater flow velocity, flow direction, and rock permeability coefficient; dissolution evolution characteristics: formation age and expansion rate of cavities; and key physical and mechanical parameters: porosity of cavity filling material, permeability coefficient of filling material, integrity coefficient of surrounding rock, uniaxial compressive strength, elastic modulus, and Poisson's ratio.

[0091] Step 1: Semi-airborne transient electromagnetic data acquisition process:

[0092] Instrument selection and parameter settings: The GDP-32Ⅱ multi-functional electrical resistivity tomography instrument is used, which can simultaneously meet the needs of deep (≥50m) and shallow (10-30m) karst exploration. It is equipped with a 20m×20m rectangular transmitting coil with 8 turns. The receiving coil is mounted on the UAV and a zigzag flight path is used to ensure full coverage of the exploration area without omission. The horizontal distance between the center of the receiving coil and the center of the transmitting coil is ≥30m, the flight altitude is 50m, the flight speed is 5m / s, the transmission frequency is 1Hz for deep exploration range ≥50m and 4Hz for shallow exploration range 10-30m, the peak current intensity is 5A, and the sampling time window is 0.1ms-100ms.

[0093] Transmitting coil deployment: A total of 8 longitudinal survey lines and 12 transverse survey lines are set up. The transmitting coils are laid along the center of the survey lines and fixed to the ground surface.

[0094] Data recording: Three valid signals were collected at each measurement point. Outliers with a signal-to-noise ratio (SNR) < 3 were removed, and the transient electromagnetic response curve was recorded. The value is the instrument observation value, which is then calculated by inversely using the induced electromotive force from the receiving coil. ,in, To induce electromotive force. The number of turns of the receiving coil. This represents the area of ​​the receiving coil.

[0095] Semi-airborne transient electromagnetic data processing: Data was processed using EMIGMA software, with the following steps: mean filtering, 5×5 window size, wavelet thresholding for noise reduction, db4 wavelet decomposition to 3 levels to eliminate high-frequency electromagnetic noise, and calculation of apparent resistivity using the late transient field formula. : , In practical applications, fine-tuning is required based on the calibration values ​​of the instruments on site, typically within the range of 0.004 to 0.006. The observation time after current cutoff is defined, with a sampling window of 0.1ms-100ms. ,in, This is the numerical value of the free magnetic permeability, taken as 4π × 10⁻⁶. -7 H / m, The equivalent radius of the transmitting coil ,constant The formula for calculation is: , The number of turns of the transmitting coil. This represents the peak value of the step current in the transmitting coil. The area of ​​the transmitting coil, The horizontal distance from the center of the transmitting coil to the receiving coil. .

[0096] Abnormal zone determination: This involves considering the apparent resistivity. Areas with resistivity <50 Ω·m and below 30% of the average resistivity of the surrounding rock are identified as karst anomaly zones. Karst zones are rich in groundwater or filled with loose materials, and their resistivity is much lower than that of intact limestone. The apparent resistivity of intact limestone is usually ≥200 Ω·m. 50 Ω·m is a commonly used critical value for karst anomalies in the industry. 30% lower than the average resistivity of the surrounding rock: to avoid misjudgment by a single threshold (such as local surrounding rock having low resistivity but not karst), further screening is done through relative difference to ensure that the anomaly zone is "low resistivity caused by karst" rather than a naturally low-resistivity stratum.

[0097] Step 1: Depression Analysis Method for Screening Strongly Aligned Areas: DEM Data Acquisition and Preprocessing: 5cm resolution elevation data is acquired using UAV aerial surveying, generating a 1m×1m raster DEM. Anomalies are removed using the ArcGIS software's vertices editing tool to ensure the authenticity of the DEM data and eliminate tiny depressions with an area <100㎡ and a depth <0.5m.

[0098] Calculate the area of ​​the depression : ,in, The first in the depression The area of ​​the grid, The area represents the total number of grids within the depression. The larger the area, the wider the range of dissolution, which may correspond to a large cavity.

[0099] Depth of depression : ,in The lowest elevation of the watershed surrounding the depression was extracted using ArcGIS flow direction analysis tools. The lowest elevation inside the depression was extracted using the ArcGIS raster calculator. The greater the depth, the stronger the dissolution, the greater the depth of water infiltration, and the faster the development of underground cavities.

[0100] catchment area : ,in, For the first to flow into the depression The area of ​​each grid cell The total number of grid cells in the catchment area is calculated using ArcGIS's catchment analysis tools. The flow direction algorithm uses the D8 method. The larger the catchment area, the more concentrated the surface water flow, the stronger the infiltration and erosion, and the higher the probability of underground rivers.

[0101] Dissolution Intensity Index : ,when Areas larger than 5 m² are marked as areas of strong karstification, corresponding to underground rivers or large solution cavities. The higher the value, the stronger the dissolution effect. A depth greater than 5 m² indicates concentrated water flow and deep dissolution, corresponding to underground rivers or large karst caves, such as... A depression of 6 m² may correspond to a cavity with a diameter of ≥ 10 m.

[0102] Criteria for determining key exploration areas: When the depression is... >5m², but it is necessary to verify whether there is a cavity underground; apparent resistivity of airborne transient electromagnetic radiation <30Ω·m: underground high conductivity anomaly (water-rich or infilled), but it is necessary to confirm whether it is related to dissolution.

[0103] Drone remote sensing of seepage - lithological anomaly zone: There is seepage on the surface, such as springs and sinkholes, and the lithology is easily soluble limestone, indicating that the dissolution channels are connected.

[0104] By using ArcGIS spatial overlay analysis and intersection tools, the vector boundaries of the three types of anomalies are overlaid. The overlapping area is the key exploration area with strong surface dissolution, high underground conductivity, and surface seepage, which serves as the core sampling point for boreholes (e.g., ZK1 is placed in the center of the depression to directly verify the underground cavities), thus avoiding the blind drilling that has a low hit rate, as traditional random boreholes do not have.

[0105] Step 1 involves verifying the parameters of the abnormal area using surface core drilling:

[0106] Drilling layout: Three verification boreholes, numbered ZK1-ZK3, were set up in the key exploration area. Borehole ZK1 was located in the center of the depression, with a depth of 60m and a target stratum of exposing the beaded karst cave. Borehole ZK2 was located in the area of ​​apparent resistivity anomaly, with a depth of 55m and a target stratum of the top plate of the karst cave and a 2.6m thin rock layer. Borehole ZK3 was located in the area of ​​water catchment node, with a depth of 50m and a target stratum of the confluence of groundwater flows.

[0107] Drilling parameters: The XY-2 core drilling rig was used with a diameter of 110mm. The core recovery rate of the bedrock section was ≥90%. The core recovery rate of intact limestone was high. If the core recovery rate was low, such as <80%, it indicated that the bedrock was broken and there might be dissolution fissures. The core recovery rate of the filling material section was ≥85%. The lithology and dissolution pore parameters were recorded every 1m. The initial groundwater level and stable groundwater level were observed simultaneously.

[0108] Groundwater observation: When the borehole exposes the aquifer, the initial water level is recorded using a measuring rope. Stable water level Calculate the drawdown : This reflects the water-bearing capacity of the aquifer. The larger the aquifer, the stronger its water level recovery ability and the better its water-bearing capacity. =5m, possibly corresponding to an underground river.

[0109] Parameter calculation: For core samples taken from boreholes, one group of samples was taken every 5m, with 3 parallel samples in each group, and the test was conducted. The calculation formula is as follows:

[0110] Porosity : ,in, The total volume of the core was determined using the helium displacement method. The solid volume of the rock core is calculated using the following formula: , For core quality, the accuracy of electronic balance measurement, The density of the rock was determined using the hydrostatic bottle method for limestone. =2.65g / cm³, filling material =1.9-2.1 g / cm³, porosity The larger the size, the looser the filling material, and the stronger the permeability.

[0111] Permeability coefficient : ,,in, To determine the cross-sectional area of ​​the variable head pipe, we take 10 cm². The sample length is 5cm. The sample cross-sectional area is 78.5 cm², and the diameter is 10 cm. For the water head from Down to Time, The initial head, The final head;

[0112] Groundwater flow velocity was calculated using a borehole tracer test, in which NaC was injected into test borehole ZK1. l The tracer, at a concentration of 5%, was used to monitor concentration changes in downstream test well ZK3, and the time of peak occurrence was recorded. .

[0113] groundwater flow rate : ,in, The horizontal distance between the two boreholes was measured using GPS. 5% NaC as tracer l Solution peak time, The angle between the water flow direction and the line connecting the borehole is determined by the apparent resistivity contour map; the direction in which the contour lines are densest indicates the water flow direction, and the flow velocity reflects the degree of unobstructed flow in the underground river. =0.5m / d indicates that the underground river has a large flow rate and requires increased drainage capacity.

[0114] Analysis of dissolution evolution characteristics: The dissolution pattern was clarified by dating and expansion rate calculation.

[0115] Age of formation of the cavity Uranium-series dating was used to determine the age of calcite deposits in the dissolution fractures of the ZK2 limestone sample. The unit is 10,000 years, representing the formation age of the cavity. Calcite precipitation and retardation take 20,000 years to dissolve.

[0116] Cavity expansion rate : ,in, This represents the initial cavity height, taken as 0.5m, an industry reference value. This represents the current cavity height. The current cavity height revealed by ZK1, such as =100,000 years, then =2.86m / 10,000 years, which can predict the future expansion trend of the cavity. For example, if the height increases by 0.0286m after 100 years, no short-term rework is required, thus verifying the long-term effectiveness of the treatment plan.

[0117] Step 1 describes the process of using UAVs to perform high-resolution multispectral and thermal infrared collaborative remote sensing to identify surface hydrological, lithological, and seepage characteristics:

[0118] Equipment selection and parameter settings: A multispectral camera with a resolution ≤0.1m and a wavelength range of 450-900nm is used, capable of identifying surface details down to 10cm (such as small runoff gullies and lithological boundaries); the wavelength range of 450-900nm covers blue (450-500nm), green (500-560nm), red (620-680nm), and near-infrared (780-900nm), with the green band (560nm) identifying limestone and the near-infrared band (840nm) identifying water bodies (water bodies have low reflectivity in the near-infrared range);

[0119] The thermal infrared imager has a temperature resolution of ≤0.5℃, which can distinguish minute temperature differences (such as a seepage zone being 0.5℃ lower than the surrounding area). It operates in the 8-14μm band, within the atmospheric window band (where atmospheric absorption is low and thermal radiation signals have strong penetration), making it suitable for surface temperature inversion. It is mounted on a hexacopter UAV, flying at an altitude of 30-50m, with a forward overlap of 80% and a lateral overlap of 60%. The sampling interval is 0.5s, and it simultaneously acquires the coordinates of three GPS control points that coincide with the lidar. This ensures the consistency of coordinates between the multi-source data (remote sensing and lidar) and facilitates subsequent overlay analysis.

[0120] Data acquisition: One full-area aerial survey was conducted during the dry season and one during the rainy season to acquire multispectral and thermal infrared images.

[0121] Dry season data: There is little surface runoff, and the lithological boundaries can be clearly identified (such as the color difference between limestone and soil layers), and permanent springs (water still seeps out during the dry season).

[0122] Rainy season data: abundant surface runoff, which can identify temporary runoff valleys (potential seepage channels) and post-rain seepage areas (traces of underground seepage flowing upwards).

[0123] Comparative analysis: By comparing images from the dry and rainy seasons, temporary water accumulation (non-seepage) during the rainy season is removed, while seepage areas (real dissolution channels) that still exist during the dry season are retained, thus improving the accuracy of anomaly identification.

[0124] Multispectral data processing: Radiometric calibration was performed using ENVI software, with the following formula: ,in, A specific band in a multispectral image Earth's surface radiance, For this specific band The radiation gain coefficient, For this specific band The original digital quantization value, For this specific band The radiation offset coefficient eliminates the DN value deviation caused by differences in camera response (such as different pixel sensitivities), ensuring that the radiation brightness truly reflects the surface characteristics.

[0125] Using ENVI's FLAASH module, the effects of atmospheric scattering and absorption on radiance are eliminated. For example, water vapor in the atmosphere absorbs near-infrared bands, resulting in low reflectivity of water bodies. After correction, water bodies can be accurately extracted.

[0126] The vector boundaries of surface runoff valleys and waterlogged depressions are extracted using the normalized water index. The calculation formula is as follows: ,in, The normalized water index, This represents the surface reflectance at the 560nm band (green band) in the multispectral image. Surface reflectance in the 840nm band (near-infrared band) of multispectral imagery.

[0127] Lithology is identified using a spectral angle matching algorithm, calculated as follows: ,in, The cosine value of the spectral angle. For band number, The total number of bands participating in spectral matching. For the target pixel in the image at the th Surface reflectance in each band, For the standard lithological end-member in the first Surface reflectance in each band, The smaller ( The closer to 1), the more similar the pixel and end-member lithology;

[0128] Limestone: 560nm characteristic reflection peak; Dolomite: 620nm characteristic reflection peak; Soil layer: 750nm reflection valley. Generate a lithological zoning map and calculate the surface dissolution rate. ,in, Surface dissolution rate, For the area of ​​karst landforms, The area covered by the image. >30% is a strongly karstified surface.

[0129] Thermal infrared data processing: Temperature inversion was performed using ArcGIS software, with the following formula: ,in, For surface temperature, , This refers to the factory calibration coefficients of the thermal infrared imager. It is the natural logarithm function. The radiance of thermal infrared images.

[0130] extract Low temperature anomaly zones ≤20℃, assuming an ambient temperature of 25℃, are mostly characterized by groundwater seepage, such as springs and seepage channels. The groundwater temperature is lower than the surface temperature, leading to surface cooling.

[0131] Temperature gradient: ,,in, For temperature gradient, , They are respectively the first The, the The subsurface temperature at each sampling point , The first The, the Calculate the temperature gradient at the underground depth of each sampling point. A temperature gradient >0.5℃ / m indicates a strong seepage zone. Low-temperature anomaly zones with temperatures ≤20℃ are extracted, including corresponding springs and seepage channels. Through temperature gradient analysis, areas with temperature gradients >0.5℃ / m are marked, corresponding to the upward discharge zone of underground seepage.

[0132] Feature recognition and output results: Overlaying multispectral lithological zoning maps, NDWI water body extraction maps, and thermal infrared low-temperature anomaly maps, according to... ,in, To calculate the overall score, For lithology score, To score points for the water body, The weighting for seepage scores is based on the following: lithology is the basis of dissolution, and only easily soluble limestone can form solution cavities, with a maximum weight of 0.4; water is the driving force of dissolution, and water flow carries CO2, with a weight of 0.3; seepage is the indicator of dissolution channels, with a weight of 0.3.

[0133] Limestone = 10 points, Dolomite = 7 points, Soil layer = 3 points;

[0134] S_water: Runoff / spring = 10 points, standing water = 7 points, no water body = 0 points;

[0135] Strong seepage zone = 10 points, low temperature anomaly zone = 7 points, no anomaly = 0 points;

[0136] when Surfaces with a score of ≥7 are characterized by easily soluble lithology, water flow, and seepage. The region is a surface seepage-lithological anomaly zone. The output map includes surface runoff, spring coordinates, and lithological boundaries, which serves as the input for surface data inversion based on the four factors.

[0137] Joint inversion is achieved using weight allocation and least squares fitting:

[0138] Data weights: Semi-airborne transient electromagnetic 40%, reflecting underground electrical anomalies; depression analysis 25%, reflecting surface dissolution intensity; core drilling 20%, directly verifying parameters; UAV remote sensing 15%, reflecting surface-subsurface correlation characteristics.

[0139] Inversion process: Using borehole core parameters as true values, the parameter deviations of other data sources are corrected by the least squares method, and finally a unified karst parameter dataset is output: hydrodynamic, dissolution evolution and mechanical parameters.

[0140] Determining True Values ​​(Surface Borehole Core Data): Core karst parameters were directly measured from three verification boreholes (ZK1-ZK3) and used as the sole true values. ,in, Porosity (measured by helium displacement method). Permeability coefficient (measured by variable head test). Groundwater flow velocity (measured by tracer test). The height of the cavity (directly exposed by drilling). The surrounding rock integrity coefficient (sound wave velocity calculation);

[0141] Organizing parameters to be corrected (three types of indirect data sources): Collecting parameters from semi-airborne transient electromagnetic data, depression analysis, and UAV remote sensing to form a matrix to be corrected. : ,in, As a semi-aerospace transient electromagnetic data source, As a data source for depression analysis, This serves as a data source for UAV remote sensing.

[0142] Weighting: Based on the correlation between the data source and underground karst, weights are allocated: Semi-airborne transient electromagnetic (covering the entire underground area): =0.4; Depression analysis (surface-subsurface correlation):

[0143] =0.25; UAV remote sensing (surface-assisted): =0.25; Weight matrix (simplified to coefficients): =[0.4,0.25,0.15].

[0144] Least squares fitting (core correction): Clearly define the objective: to make the corrected parameters: As close to the true value as possible That is, minimizing the weighted sum of squared errors;

[0145] Solve for the optimal parameters directly using a simplified formula: ;

[0146] Verification of correction effect: Compare the deviation of the corrected parameters with the true values. All parameter deviations are ≤5%, indicating that the correction is effective.

[0147] Step 2: 3D Geological Modeling: Based on lidar surface scanning with a point cloud density ≥100 points / m² and elevation error ≤5cm, and using karst space data from ground-penetrating radar 3D imaging, micro-fracture data from advanced horizontal borehole imaging, and surface hydrological, lithological, and seepage data acquired by UAV-borne high-resolution multispectral and thermal infrared collaborative remote sensing, a layered 3D geological model of "surface layer-karst layer-surrounding rock layer" is constructed. The model parameters are verified and optimized through borehole data inversion, clarifying the direction of the underground river, the 3D coordinates and volume of the karst morphology, and the mechanical parameters of the surrounding rock.

[0148] Step 2, the lidar scanning process: using the RIEGLVZ-6000 device, the scanning distance is 500m, the point cloud density is 100 points / ㎡, the scanning frequency is 500kHz, and the surface point cloud model is generated after noise reduction by RiSCANPRO software and registration based on 3 GPS control points with a planar accuracy of ≤5cm.

[0149] 3D imaging workflow of ground-penetrating radar: Using an SIR-4000 device with a 250MHz shielded antenna, a survey network is established with 5m spacing along the tunnel axis and 10m spacing perpendicular to the axis. The antenna moving speed is 0.5m / s, the sampling rate is 1024 points / scan, the recording length is 400ns, corresponding to a depth ≥30m. The electromagnetic wave velocity in limestone is 0.15m / ns. Processing steps:

[0150] Gain correction: Exponential gain correction is used: gain coefficient ,in, for Signal gain coefficient at time 10:00 Sampling time of ground-penetrating radar signal, gain attenuation compensation coefficient Initial gain coefficient This compensates for signal attenuation in deep areas.

[0151] Filtering: Frequency domain bandpass filtering, ranging from 50 to 250 MHz, to eliminate high-frequency noise.

[0152] 3D offset imaging: The Kirchhoff offset algorithm is used with an offset speed of v=0.15m / ns to correct image distortion caused by signal delay and generate a 3D radar profile to identify cavity reflection characteristics: phase axis interruption and strong reflection area.

[0153] Advanced horizontal borehole imaging: A borehole with a diameter of 90mm and a depth of 30m is laid in the tunnel, at an angle of 15° to the tunnel axis. An Olympus IPLEXFX endoscope with a resolution of 1280×720 is used to take images every 10cm at a speed of 5cm / s. The distance and diameter of the karst cavities from the borehole opening, as well as the orientation, dip angle and width of the fractures, are recorded. This verifies the karst cavities identified by the ground-penetrating radar (e.g., if the ground-penetrating radar shows a karst cavity at 15m, borehole imaging confirms that there is a karst cavity with a diameter of 2m at that location). It also supplements the micro fracture data (ground-penetrating radar cannot identify fractures <1m).

[0154] Step Two: 3D Model Coordinate System and Layered Construction: Using ArcGIS's projection coordinate system transformation tool, the 3D geological model is converted to the WGS84 coordinate system. "system, This refers to the tunnel axis mileage. This refers to the lateral distance along the vertical axis, ranging from -50 to 50 meters. The elevation ranges from -30 to 50 meters, with the tunnel track surface as ±0.

[0155] The three-dimensional geological model is constructed in layers, including: based on FLAC3D 7.0 software: surface layer ( ≥0): A 1m×1m grid is generated based on the lidar point cloud through Delaunay triangulation, and the sinkhole is marked by superimposing the depression analysis results.

[0156] cavity layer (-30≤ <0): Extract ≥500 cavity boundary points from the ground-penetrating radar profile. The three-dimensional surface of the cavity is generated using a quadratic polynomial surface: , The fitting coefficients are calculated using the least squares method, and the residuals are ≤0.5m.

[0157] Surrounding rock strata ( <-30): Classified by apparent resistivity ≥200Ω·m is considered intact surrounding rock, 100< <200Ω·m indicates relatively intact surrounding rock. For rock with an Ω·m value less than 100Ω·m, an interpolation algorithm is used to generate a rock integrity coefficient grid.

[0158] Calculation and Model Validation of Surrounding Rock Mechanical Parameters: Surrounding Rock Integrity Coefficient : ,in, Core recovery rate To measure the acoustic wave velocity during drilling, =5.5km / s, using an RS-ST01C acoustic wave detector, one point is measured every 2m. The standard wave velocity for fresh limestone is taken as 6.0 km / s. ≥0.8 indicates intact surrounding rock, 0.6 < <0.8 indicates a relatively complete result. <0.6 indicates breakage.

[0159] Conversion of surrounding rock mechanical parameters: Uniaxial compressive strength : These are values ​​from core tests.

[0160] elastic modulus : The linear relationship between the elastic modulus of limestone and its uniaxial compressive strength is shown in the example. For instance, if Rc = 24 MPa, then E = 480 MPa, which is consistent with the mechanical properties of limestone (E is usually 400-600 MPa).

[0161] Poisson's ratio , The larger (the more intact the surrounding rock). The smaller the value (μ≈0.25 for intact limestone, μ≈0.35 for fractured surrounding rock), the greater the Poisson's ratio of the rock mass becomes as the integrity decreases, because fractured surrounding rock undergoes greater lateral deformation under pressure.

[0162] The three-dimensional geological model was validated using borehole data inversion: the elevations of the actual exposed top and bottom plates of the karst cavity in borehole ZK2 were compared with the model calculations. The error was ≤ ±0.2m. If the error was > 0.2m, the surface fitting coefficients were readjusted. (e.g., increase) (This reduces the surface elevation) until the error meets the requirements.

[0163] Data fusion correction: integrating semi-airborne transient electromagnetic data. Abnormal area <50Ω·m and the cavity reflection zone of ground-penetrating radar (interruption of the same phase axis) and the surface seepage and lithological anomaly zone of UAV remote sensing ( (≥7 points) The three are superimposed, and the anomaly confidence of the superimposed area is ≥90% (e.g., the cavity revealed by the ZK2 hole is located in the superimposed area, verifying that the area is a real cavity). Verification is performed by drilling, such as the cavity revealed by the ZK2 hole matching the superimposed area.

[0164] Non-overlay region correction:

[0165] Only semi-aerospace transient electromagnetic anomalies ( <50Ω·m), no radar reflection or remote sensing anomaly: determined to be a water-rich fissure (non-cavity), the fissure is low resistance due to water abundance, but there is no obvious cavity space.

[0166] Only ground-penetrating radar reflection, no electromagnetic or remote sensing anomalies: determined to be a dry karst cavity (no groundwater), because there is no water body in the cavity, the electromagnetic resistivity is not low, and there is no seepage on the surface.

[0167] Avoid misidentifying water-rich fissures as solution cavities (leading to overtreatment) or underidentifying dry solution cavities (leading to insufficient treatment).

[0168] Step 3: Adaptation of treatment measures: Based on the output data of Step 1 and Step 2, optimize the construction parameters of the pile foundation raft slab, use the positioning mechanism to complete the reinforcement scheme of the top of the karst cavity and the design of the drainage system, so as to achieve precise treatment of the karst underground river section.

[0169] Step 3 optimization of pile foundation raft slab construction: Addressing the beaded karst caves in the foundation slab:

[0170] Pile location adjustment: Based on: the distribution of cavities and the grading of surrounding rock in the 3D model (intact surrounding rock) ≥200Ω·m ≥0.8).

[0171] Measures: Avoid areas containing filling material ( <30Ω·m), the pile positions are arranged in a complete surrounding rock area (such as the pile position at K1+300, where 30m below is complete limestone).

[0172] Pile length optimization: Increase the pile length by 1-2m according to the depth of the karst cavity. For example, if the maximum depth of the karst cavity is 29.1m, the pile length should be 31m to ensure that the pile tip penetrates ≥3m into the stable rock layer. The integrity coefficient of the surrounding rock layer is then used to determine the optimal length. ≥0.8 region determination.

[0173] Raft foundation design: 1.2m thick: to withstand the pile foundation reaction force and the load of the superstructure (such as the weight of the tunnel lining), C30 concrete compressive strength ≥30MPa, to meet the bearing capacity requirements.

[0174] Reinforcing steel configuration: double-layer, bidirectional Φ20×150 (diameter 20mm, spacing 150mm) to resist raft slab bending stress (to prevent raft slab cracking).

[0175] Edge extension: Extends to the side wall of the cavity ≥0.5m, connecting with the shotcrete layer (C25 shotcrete) on the side wall of the cavity to form an integral load-bearing structure (to prevent the side wall of the cavity from collapsing and squeezing the raft).

[0176] Optimization of cavity top reinforcement: for thin rock layers at the top: reinforcement range: calculated using a 3D model. <0.6 Weak area.

[0177] Reinforcement parameters: The density of the shotcrete anchor rods is adjusted from the conventional 100cm×100cm to 80cm×80cm. The anchor rods are 600cm long Φ22 cement mortar anchor rods with an anchoring depth ≥1.5m. A 20cm×20cm Φ8 steel mesh is hung, and 10cm thick C25 concrete is shotcreted with a compressive strength ≥25MPa.

[0178] Impact buffer: A 50cm thick layer of waste tires is laid in the top suspended section. The tires are arranged laterally and the gaps are filled with C15 fine stone concrete to resist the impact of sporadic rockfalls from the top of the tunnel. The impact load is ≤50kN, which has been verified by drop hammer test.

[0179] Drainage system design: Maximum inflow calculation: Atmospheric rainfall infiltration method, formula: ,in, The rainfall infiltration coefficient in karst areas This represents the average annual rainfall over many years.

[0180] Inverted siphon design: Two 2m×2m reinforced concrete inverted siphons are used to increase flow capacity. ,in, The flow area of ​​the inverted siphon is =2m×2m reinforced concrete structure, 4m 2,, The average flow velocity of water within the inverted siphon pipe. This refers to the acceleration caused by gravity on the water flow, with a value of 9.8 m / s². 2 , The difference in water level between the inlet and outlet of the inverted siphon.

[0181] Surface blind drains: Base blind drains are set up on the surface, with the main drain being 60cm×50cm and the branch drains being 40cm×30cm, spaced 10m apart. Φ100mm permeable pipes are laid in the blind drains and wrapped with geotextile to guide groundwater to the original drainage channel.

[0182] Optimization of biased support: For biased Class V surrounding rock, a complete pressure arch is formed:

[0183] Support parameters: The support adopts the biased pressure Class V surrounding rock support. The primary support consists of 25cm thick C25 shotcrete and Φ25 hollow grouting anchors with a spacing of 80cm×80cm and I20a type steel arch frames with a spacing of 60cm. The secondary lining is 50cm thick C35 reinforced concrete.

[0184] Biased support wall: A 2-3m thick C30 reinforced concrete support wall is added to the open area on the right side of the tunnel. The support wall foundation is set with an enlarged foundation with an enlarged dimension of ≥0.5m×0.5m. If necessary, a Φ22 anchor bolt with a length of 4m is added.

[0185] Top backfill: 2-3m thick C20 concrete is backfilled in the exposed areas at the top of the tunnel to ensure the formation of a complete pressure arch. The arch stress is verified to be ≤20MPa through FLAC3D numerical simulation.

[0186] Example 2

[0187] Reference Figures 2-5 As shown, the positioning mechanism includes a positioning housing 1. A positioning frame 2 with a rack is slidably inserted into the inner wall of the positioning housing 1. One end of the positioning frame 2 is V-shaped to facilitate the fixing of the anchor rod. A gear set 21 is rotatably connected to the inner wall of the positioning housing 1. The gear set 21 includes two gears. One gear of the gear set 21 meshes with the rack of the positioning frame 2. A drive gear ring 22 is rotatably connected to the inner wall of the positioning housing 1. The rotation of the drive gear ring 22 can drive the movement of the positioning frame 2 through the transmission of the gear set 21. The drive gear ring 22 meshes with the other gear of the gear set 21. A drive gear 23 is rotatably connected to the outer surface of the positioning housing 1. The drive gear 23 meshes with the drive gear ring 22.

[0188] A collar 3 is rotatably connected to the outer surface of the positioning housing 1. A connecting block assembly 31 is fixedly installed on the inner wall of the collar 3. The rotation of the collar 3 facilitates the adjustment of the direction of the connecting block assembly 31 to complete the positioning of adjacent anchor rods. The connecting block assembly 31 is composed of multiple connecting blocks that are hinged end to end by pins. A magnetic block 32 is fixedly installed on the outer surface of the connecting block assembly 31. The outer surface of the magnetic block 32 is magnetically connected to the outer surface of the collar 3. By bending the connecting block assembly 31, the magnetism of the connecting block assembly 31 can be easily stored on the outer surface of the collar 3. A universal tube 33 is fixedly installed on the inner wall of the connecting block assembly 31. The universal tube 33 can deform but will not move. A measuring tape 34 is fixedly installed on the inner wall of one connecting block of the connecting block assembly 31. One end of the measuring tape 34 passes through the inner walls of the remaining connecting blocks of the connecting block assembly 31 and extends to the outside. A transparent plate 35 is fixedly installed on the outer surface of the connecting block assembly 31 to facilitate the observation of the data of the measuring tape 34.

[0189] Working principle: When anchor bolt positioning is required, the positioning housing 1 is fitted onto the outer surface of the first anchor bolt that has been fixed. The drive gear 23 is rotated, which drives the drive gear ring 22 to rotate, which in turn drives the internal gear set 21 to rotate. The gear set 21 drives the positioning frame 2 that meshes with it to move, so that the positioning frame 2 clamps one end of the anchor bolt. According to the direction in which the next anchor bolt needs to be driven in, the position of the connecting block assembly 31 is adjusted by rotating the collar 3, so that the connecting block assembly 31 faces the direction of the next anchor bolt. The positioning point of the next fixed anchor bolt is marked according to the dimension on the measuring tape 34 through the transparent plate 35. When fixing the anchor bolt on the arch surface of the tunnel, the universal tube 33 is bent so that the universal tube 33 can drive the hinged connecting block assembly 31 to bend and fit against the inner wall of the tunnel, causing the measuring tape 34 to bend into the arch shape of the tunnel, thus completing the positioning.

[0190] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for treating a tunnel through a karst underground river section, characterized in that: Step 1: Multi-source geophysical exploration: For the karst underground river section of the tunnel and the surrounding 500m range, the following methods are used: semi-airborne transient electromagnetic mapping to delineate karst anomaly areas, depression analysis to screen for strong karstification areas, surface borehole core sampling to verify anomaly parameters, and UAV-borne high-resolution multispectral and thermal infrared collaborative remote sensing to identify surface hydrological, lithological, and seepage data characteristics. These four methods are combined to obtain karst parameters, including karst hydrodynamic conditions (groundwater flow velocity, flow direction, and rock permeability coefficient), karst evolution characteristics (cavity formation age and expansion rate), and key physical and mechanical parameters (cavity filling porosity, filling permeability coefficient, surrounding rock integrity coefficient, uniaxial compressive strength, elastic modulus, and Poisson's ratio). Step 2: 3D geological modeling: Based on surface scanning by lidar, spatial data of karst caverns obtained by 3D imaging of ground-penetrating radar, micro-fracture data obtained by imaging inside advanced horizontal boreholes, and surface hydrological, lithological, and seepage data obtained by UAV-borne high-resolution multispectral and thermal infrared collaborative remote sensing, a layered 3D geological model of surface layer-karst cavern layer-surrounding rock layer is constructed. The model parameters are verified and optimized through borehole data inversion, and the 3D coordinates and volume of underground river direction, karst cavern morphology, and surrounding rock mechanical parameters are clarified. Step 3: Adaptation of treatment measures: Based on the output data of Step 1 and Step 2, optimize the construction parameters of the pile foundation raft slab, use the positioning mechanism to complete the reinforcement scheme of the top of the karst cavity and the design of the drainage system, so as to achieve precise treatment of the karst underground river section.

2. The method for treating a tunnel through a karst underground river section according to claim 1, characterized in that: The semi-airborne transient electromagnetic data acquisition process in step one is as follows: Instrument selection and parameter settings: The GDP-32Ⅱ multi-functional electrical resistivity meter is used, equipped with a 20m×20m rectangular transmitting coil with 8 turns. The receiving coil is mounted on the UAV. The flight path is "zigzag". The horizontal distance between the receiving coil and the center of the transmitting coil is ≥30m. The flight altitude is 50m and the flight speed is 5m / s. The transmission frequency is 1Hz for deep detection at a depth of ≥50m and 4Hz for shallow detection at a depth of 10-30m. The peak current intensity is 5A and the sampling time window is 0.1ms-100ms. Transmitting coil deployment: A total of 8 longitudinal survey lines and 12 transverse survey lines are set up. The transmitting coils are laid along the center of the survey lines and fixed to the ground surface. Data recording: Three valid signals were collected at each measurement point. Outliers with a signal-to-noise ratio (SNR) < 3 were removed, and the transient electromagnetic response curve was recorded. The value is the instrument observation value, which is then calculated by inversely using the induced electromotive force from the receiving coil. ,in, To induce electromotive force. The number of turns of the receiving coil. The area of ​​the receiving coil; Semi-airborne transient electromagnetic data processing: Data was processed using EMIGMA software, with the following steps: mean filtering, 5×5 window size, wavelet thresholding for noise reduction, db4 wavelet decomposition to 3 levels to eliminate high-frequency electromagnetic noise, and calculation of apparent resistivity using the late transient field formula. : ,in, The observation time after current cutoff is defined, with a sampling window of 0.1ms-100ms. ,in, This is the numerical value of the free magnetic permeability, taken as 4π × 10⁻⁶. -7 H / m, The equivalent radius of the transmitting coil ,constant The formula for calculation is: , The number of turns of the transmitting coil. This represents the peak value of the step current in the transmitting coil. The area of ​​the transmitting coil, The horizontal distance from the center of the transmitting coil to the receiving coil. ; Abnormal zone determination: This involves considering the apparent resistivity. Areas with an Ω·m value less than 50 Ω·m and 30% lower than the average value of the surrounding rock are identified as karst anomaly zones.

3. The method for treating a tunnel through a karst underground river section according to claim 2, characterized in that: The process of screening strong karst areas using the depression analysis method in step one is as follows: DEM data acquisition and preprocessing: 5cm resolution elevation data is acquired using UAV aerial surveying, generating a 1m×1m raster DEM. Elevation outliers are removed using ArcGIS software to eliminate small depressions with an area <100㎡ and a depth <0.5m. Calculate the area of ​​the depression : ,in, The first in the depression The area of ​​the grid, This represents the total number of grid cells within the depression. Depth of depression : ,in The lowest elevation of the watershed surrounding the depression was extracted using ArcGIS flow direction analysis tools. The lowest elevation inside the depression was extracted using the ArcGIS raster calculator. Catchment area : ,in, For the first to flow into the depression The area of ​​each grid cell The total number of raster cells in the catchment area is calculated using ArcGIS's catchment analysis tools, with the D8 method used for flow direction. Dissolution Intensity Index : ,when Areas larger than 5m² are marked as strong karst zones, corresponding to underground rivers or large solution cavities; Criteria for determining key exploration areas: When the depression is... >5m² and corresponding to semi-aerospace transient electromagnetic apparent resistivity When the high conductivity anomaly zone of <30Ω·m, the surface seepage and lithological anomaly zone identified by UAV-borne high-resolution multispectral and thermal infrared collaborative remote sensing are superimposed, the superimposed area is marked as a key exploration area using ArcGIS spatial overlay analysis tools, and used as a subsequent borehole core sampling point.

4. The method for treating a tunnel through a karst underground river section according to claim 3, characterized in that: The procedure for verifying the parameters of the abnormal area by drilling cores on the surface in step one is as follows: Drilling layout: Three verification boreholes, numbered ZK1-ZK3, were set up in the key exploration area. Borehole ZK1 was located in the center of the depression, with a depth of 60m and a target stratum of exposing the beaded karst cave. Borehole ZK2 was located in the area of ​​apparent resistivity anomaly, with a depth of 55m and a target stratum of the top plate of the karst cave and a 2.6m thin rock layer. Borehole ZK3 was located in the area of ​​water catchment node, with a depth of 50m and a target stratum of the confluence of groundwater flows. Drilling parameters: The XY-2 core drilling rig was used with a diameter of 110mm. The core recovery rate in the bedrock section was ≥90%, and the core recovery rate in the filling section was ≥85%. Lithology and karst cavities parameters were recorded every 1m, and the initial groundwater level and stable groundwater level were observed simultaneously. Groundwater observation: When the borehole exposes the aquifer, the initial water level is recorded using a measuring rope. Stable water level Calculate the drawdown : ; Parameter calculation: For core samples taken from boreholes, one group of samples was taken every 5m, with 3 parallel samples in each group, and the test was conducted. The calculation formula is as follows: Porosity : ,in, The total volume of the core was determined using the helium displacement method. The solid volume of the rock core is calculated using the following formula: , For core quality, the accuracy of electronic balance measurement, The density of the rock was determined using the hydrostatic bottle method for limestone. =2.65g / cm³, filling material =1.9-2.1 g / cm³; Permeability coefficient : ,in, To determine the cross-sectional area of ​​the variable head pipe, we take 10 cm². The sample length is 5cm. The sample cross-sectional area is 78.5 cm², and the diameter is 10 cm. For the water head from Down to Time, The initial head, The final head; Groundwater flow velocity was calculated using a borehole tracer test, in which NaC was injected into test borehole ZK1. l The tracer, at a concentration of 5%, was used to monitor concentration changes in downstream test well ZK3, and the time of peak occurrence was recorded. ; groundwater flow rate : ,in, The horizontal distance between the two boreholes was measured using GPS. 5% NaCl tracer l Solution peak time, The angle between the water flow direction and the line connecting the borehole is determined by the apparent resistivity contour map. Analysis of dissolution evolution characteristics: The dissolution pattern was clarified through dating and expansion rate calculation; Age of formation of the cavity Uranium-series dating was used to determine the age of calcite deposits in the dissolution fractures of the ZK2 limestone sample. The unit is 10,000 years, representing the formation age of the cavity. Calcite precipitation and delayed dissolution take 20,000 years; Cavity expansion rate : ,in, The initial cavity height is set to 0.5m. This represents the current cavity height. .

5. The method for treating a tunnel through a karst underground river section according to claim 4, characterized in that: The process of identifying surface hydrological, lithological, and seepage characteristics using UAV-borne high-resolution multispectral and thermal infrared coordinated remote sensing in step one is as follows: Equipment selection and parameter settings: A multispectral camera with a resolution ≤0.1m and a wavelength band of 450-900nm is used; a thermal infrared imager with a temperature resolution ≤0.5℃ and a wavelength band of 8-14μm is used. The imager is mounted on a hexacopter UAV with a flight altitude of 30-50m, a forward overlap of 80%, a lateral overlap of 60%, and a sampling interval of 0.5s. The coordinates of three GPS control points that coincide with the lidar are collected simultaneously. Data acquisition: One full-area aerial survey each during the dry season and the rainy season to acquire multispectral and thermal infrared images; Multispectral data processing: Radiometric calibration was performed using ENVI software, with the following formula: ,in, A specific band in a multispectral image Earth's surface radiance, For this specific band The radiation gain coefficient, For this specific band The original digital quantization value, For this specific band radiative offset coefficient; After atmospheric correction, surface runoff gullies and waterlogged areas are extracted using the normalized water index. The calculation formula is as follows: ,in, The normalized water index, The surface reflectance in the 560nm band and green band of the multispectral image; Surface reflectance in the 840nm and near-infrared bands of multispectral images; Lithology is identified using a spectral angle matching algorithm, calculated as follows: ,in, The cosine value of the spectral angle. For band number, The total number of bands participating in spectral matching. For the target pixel in the image at the th Surface reflectance in each band, For the standard lithological end-member in the first Surface reflectance in each band; Limestone: 560nm characteristic reflection peak; Dolomite: 620nm characteristic reflection peak; Soil layer: 750nm reflection valley. Generate a lithological zoning map and calculate the surface dissolution rate. ,in, The surface erosion rate, For the area of ​​karst landforms, The area covered by the image; Thermal infrared data processing: Temperature inversion was performed using ArcGIS software, with the following formula: ,in, For surface temperature, , This refers to the factory calibration coefficients of the thermal infrared imager. It is the natural logarithm function. The radiance of thermal infrared images; extract Low temperature anomaly zone ≤20℃ ,in, For temperature gradient, , They are respectively the first The, the The subsurface temperature at each sampling point , The first The, the Calculate the temperature gradient at the underground depth of each sampling point. >0.5℃ / m indicates a strong seepage zone. Low-temperature anomaly zones with temperatures ≤20℃ are extracted, including corresponding springs and seepage channels. Through temperature gradient analysis, areas with temperature gradients >0.5℃ / m are marked, corresponding to the upward discharge zone of underground seepage. Feature recognition and output results: Overlaying multispectral lithological zoning maps, NDWI water body extraction maps, and thermal infrared low-temperature anomaly maps, according to... ,in, To calculate the overall score, For lithology score, To score points for the water body, Score for seepage; when Areas with a score of ≥7 are designated as surface seepage-lithological anomaly zones. Output a map containing surface runoff, spring coordinates, and lithological boundaries, which will serve as the input for surface data inversion based on these four metrics. Joint inversion is achieved using weight allocation and least squares fitting: Data weights: Semi-airborne transient electromagnetic 40%, reflecting underground electrical anomalies; depression analysis 25%, reflecting surface dissolution intensity; core drilling 20%, directly verifying parameters; UAV remote sensing 15%, reflecting surface-subsurface correlation characteristics. Inversion process: Using borehole core parameters as true values, the parameter deviation of the data source is corrected by the least squares method, and finally a unified karst parameter dataset is output: hydrodynamic, dissolution evolution and mechanical parameters.

6. The method for treating a tunnel through a karst underground river section according to claim 5, characterized in that: The laser radar scanning process in step two is as follows: using a RIEGLVZ-6000 device, scanning distance of 500m, point cloud density of 100 points / ㎡, scanning frequency of 500kHz, noise reduction by RiSCANPRO software and registration based on 3 GPS control points, and generating a surface point cloud model with a planar accuracy of ≤5cm. 3D imaging workflow of ground-penetrating radar: Using an SIR-4000 device with a 250MHz shielded antenna, a survey network is established with 5m spacing along the tunnel axis and 10m spacing perpendicular to the axis. The antenna moving speed is 0.5m / s, the sampling rate is 1024 points / scan, the recording length is 400ns, corresponding to a depth ≥30m. The electromagnetic wave velocity in limestone is 0.15m / ns. Processing steps: Gain correction: Exponential gain correction is used: gain coefficient ,in, for Signal gain coefficient at time 10:00 Sampling time of ground-penetrating radar signal, gain attenuation compensation coefficient Initial gain coefficient To compensate for signal attenuation in deep areas; Filtering: Frequency domain bandpass filtering, ranging from 50-250MHz, to eliminate high-frequency noise; 3D offset imaging: The Kirchhoff offset algorithm is used with an offset speed of v=0.15m / ns to correct image distortion caused by signal delay and generate a 3D radar profile to identify cavity reflection features: phase axis interruption and strong reflection area; Advanced horizontal borehole imaging: A borehole with a diameter of 90mm and a depth of 30m is laid in the tunnel, with an angle of 15° to the tunnel axis. An Olympus IPLEXFX endoscope with a resolution of 1280×720 is used to take images every 10cm at a speed of 5cm / s. The distance from the borehole opening and the diameter of the erosion cavities, as well as the orientation, dip angle and width of the fractures are recorded.

7. The method for treating a tunnel through a karst underground river section according to claim 6, characterized in that: In step two, the coordinate system and layered construction of the three-dimensional model are as follows: Coordinate transformation: The three-dimensional geological model is transformed from the WGS84 coordinate system. system, This refers to the tunnel axis mileage. This refers to the lateral distance along the vertical axis, ranging from -50 to 50 meters. The elevation ranges from -30 to 50 meters, with the tunnel track surface as ±0. The three-dimensional geological model is constructed in layers, including: based on FLAC3D 7.0 software: surface layer. ≥0: A 1m×1m grid is generated based on the lidar point cloud through Delaunay triangulation, and the sinkhole is marked by superimposing the depression analysis results; cavity layer -30≤ <0: Extract ≥500 cavity boundary points from the ground-penetrating radar profile. The three-dimensional surface of the cavity is generated using a quadratic polynomial surface: , The fitting coefficients are calculated using the least squares method, and the residuals are ≤0.5m. Surrounding rock layer <-30: Classified by apparent resistivity ≥200Ω·m is considered intact surrounding rock, 100< <200Ω·m indicates relatively intact surrounding rock. For rock with an Ω·m value less than 100Ω·m, an interpolation algorithm is used to generate a rock integrity coefficient grid. Calculation and Model Validation of Surrounding Rock Mechanical Parameters: Surrounding Rock Integrity Coefficient : ,in, Core recovery rate To measure the acoustic wave velocity during drilling, =5.5km / s, using an RS-ST01C acoustic wave detector, one point is measured every 2m. The standard wave velocity for fresh limestone is taken as 6.0 km / s; Conversion of surrounding rock mechanical parameters: Uniaxial compressive strength : These are core test values; elastic modulus : ; Poisson's ratio : ; The three-dimensional geological model was validated using borehole data inversion: the elevations of the actual exposed top and bottom plates of the karst cavity in borehole ZK2 were compared with the model calculations. The error was ≤ ±0.2m. If the error was > 0.2m, the surface fitting coefficients were readjusted. Continue until the accuracy requirements are met; Data fusion correction: integrating semi-airborne transient electromagnetic data. Abnormal area The superposition of <50Ω·m, the interruption of the phase axis of the cavity reflection zone from ground-penetrating radar, and the surface seepage and lithological anomaly zones detected by UAV remote sensing, results in an anomaly confidence level of ≥90% in the superimposed area. This is verified through borehole testing; for example, borehole ZK2 reveals a cavity that matches the superimposed area. Non-superimposed areas are corrected using borehole data, such as semi-airborne transient electromagnetic data. <50Ω·m, but with no radar reflection zone, it is determined to be a water-rich fissure rather than a karst cave.

8. The method for treating a tunnel through a karst underground river section according to claim 7, characterized in that: The optimization of pile foundation raft slab construction in step three: Addressing the beaded karst caves in the bottom slab: Pile location adjustment: Based on the distribution of beaded karst caves in a 3D model, and the filling area <30Ω·m, avoid areas with infill material, and arrange the pile locations in intact surrounding rock areas. ≥200Ω·m; Pile length optimization: Increase the pile length by 1-2m according to the depth of the karst cavity. For example, if the maximum depth of the karst cavity is 29.1m, the pile length should be 31m to ensure that the pile tip penetrates ≥3m into the stable rock layer. The integrity coefficient of the surrounding rock layer is then used to determine the optimal length. ≥0.8 region determination; Raft design: A 1.2m thick C30 reinforced concrete raft slab is poured on top of the piles. It is made of double-layer bidirectional Φ20×150 steel bars. The edge of the raft slab extends to the side wall of the karst cavity by ≥0.5m and connects with the shotcrete layer of the side wall of the karst cavity. Optimization of cavity top reinforcement: Reinforcement range for thin rock layers at the top: calculated using a 3D model. <0.6 weak area; Reinforcement parameters: The density of the shotcrete anchor rods is adjusted from the conventional 100cm×100cm to 80cm×80cm. The anchor rods are 600cm long Φ22 cement mortar anchor rods with an anchoring depth ≥1.5m. A 20cm×20cm Φ8 steel mesh is hung, and 10cm thick C25 concrete is shotcreted with a compressive strength ≥25MPa. Impact buffer: A 50cm thick layer of waste tires is laid in the top suspended section. The tires are arranged laterally and the gaps are filled with C15 fine stone concrete to resist the impact of sporadic rockfalls from the top of the tunnel. The impact load is ≤50kN, which is verified by drop hammer test. Drainage system design: Maximum inflow calculation: Atmospheric rainfall infiltration method, formula: ,in, The rainfall infiltration coefficient in karst areas This refers to the multi-year average rainfall. Inverted siphon design: Two 2m×2m reinforced concrete inverted siphons are used to increase flow capacity. ,in, The flow area of ​​the inverted siphon is =2m×2m reinforced concrete structure, 4m 2,, The average flow velocity of water within the inverted siphon pipe. This refers to the acceleration caused by gravity on the water flow, with a value of 9.8 m / s². 2 , The difference in water level between the inverted siphon inlet and outlet; Surface blind drains: Base blind drains are set up on the surface, with the main drain being 60cm×50cm and the branch drains being 40cm×30cm, spaced 10m apart. Φ100mm permeable pipes are laid in the blind drains and wrapped with geotextile to guide groundwater to the original drainage channel. Optimization of biased support: For biased Class V surrounding rock, a complete pressure arch is formed: Support parameters: The support adopts the biased pressure Class V surrounding rock support. The primary support consists of 25cm thick C25 shotcrete and Φ25 hollow grouting anchors with a spacing of 80cm×80cm and I20a type steel arch frame with a spacing of 60cm. The secondary lining is 50cm thick C35 reinforced concrete. Eccentric pressure vent wall: A 2-3m thick C30 reinforced concrete vent wall shall be added to the open area on the right side of the tunnel. The vent wall foundation shall be an enlarged foundation with an enlarged dimension of ≥0.5m×0.5m, or a Φ22 anchor bolt with a length of 4m shall be added. Top backfill: 2-3m thick C20 concrete is backfilled in the exposed areas at the top of the tunnel to ensure the formation of a complete pressure arch. The arch stress is verified to be ≤20MPa through FLAC3D numerical simulation.

9. A method for treating a tunnel through a karst underground river section according to claim 8, characterized in that: The positioning mechanism in step three includes a positioning housing (1), a positioning frame (2) with a rack slidably inserted into the inner wall of the positioning housing (1), a gear set (21) rotatably connected to the inner wall of the positioning housing (1), one gear of the gear set (21) meshing with the rack of the positioning frame (2), a drive gear ring (22) rotatably connected to the inner wall of the positioning housing (1), the drive gear ring (22) meshing with another gear of the gear set (21), and a drive gear (23) rotatably connected to the outer surface of the positioning housing (1), the drive gear (23) meshing with the drive gear ring (22).

10. A method for treating a tunnel through a karst underground river section according to claim 9, characterized in that: The outer surface of the positioning housing (1) is rotatably connected to a collar (3). A connecting block assembly (31) is fixedly installed on the inner wall of the collar (3). The connecting block assembly (31) is composed of multiple connecting blocks that are hinged end to end by pins. A magnetic block (32) is fixedly installed on the outer surface of the connecting block assembly (31). The outer surface of the magnetic block (32) is magnetically connected to the outer surface of the collar (3). A universal tube (33) is fixedly installed on the inner wall of the connecting block assembly (31). A measuring soft ruler (34) is fixedly installed on the inner wall of one of the connecting blocks of the connecting block assembly (31). One end of the measuring soft ruler (34) passes through the inner walls of the remaining connecting blocks of the connecting block assembly (31) and extends to the outside. A transparent plate (35) is fixedly installed on the outer surface of the connecting block assembly (31).

Images