A comprehensive advanced three-dimensional geological modeling method for karst geological highway tunnel construction

Abstract

The present invention discloses a comprehensive advanced three-dimensional geological modeling method for karst geological highway tunnel construction, which includes drilling surface holes in the construction section to obtain hydrogeological information; using lidar to scan the three-dimensional point cloud data of the surface landform, and using the high-density resistivity method to detect the three-dimensional positions of the unfavorable geological structures of the tunnel; using methods such as three-dimensional imaging of ground penetrating radar to obtain data on unfavorable geology and lithology change zones in front of and around the heading face; using in-hole imaging of advanced horizontal drilling to obtain three-dimensional geological structural plane information such as the occurrence of rock strata and the distribution of joint surfaces; and fusing the detected data to establish point models, surface models, volume models and parameter models for describing geological bodies, so as to obtain an advanced three-dimensional geological model. The present invention can significantly reduce the safety risks of tunnel construction and at the same time save the economic construction cost.

Description

Technical Field

[0001] The present invention relates to the technical field of construction geological forecasting, and particularly relates to a comprehensive advanced three-dimensional geological modeling method for highway tunnel construction in karst geology. Background Art

[0002] During highway tunnel construction, adverse geological bodies such as mud and water inrush, fault fracture zones, etc. are often encountered. Practice shows that many major accidents are often caused by the inaccurate exploration of the distribution positions and development conditions of weak areas such as bedding planes, fissures, fracture structures, and weak structural planes existing inside the rock mass. If the understanding of the geological structure is insufficient, the engineering design cannot accurately provide a design plan, and the construction safety risk increases sharply. Especially during construction in karst landforms, disasters such as mud and water inrush and collapses often occur, causing huge losses to life and property. Therefore, how to accurately master and describe the three-dimensional internal structure of the rock mass in front of the tunnel face and accurately predict whether there are adverse geological conditions such as karst caves, fault fracture zones, and water-rich areas in the front is a key technical problem that urgently needs to be solved in tunnel construction. Summary of the Invention

[0003] The purpose of the present invention is to provide a comprehensive advanced three-dimensional geological modeling method for highway tunnel construction in karst geology, which can quickly and effectively perform advanced three-dimensional geological modeling and adverse geological forecasting, so as to master the three-dimensional geological information in front of the construction face and greatly reduce the construction safety risk.

[0004] The technical solution of the present invention is as follows:

[0005] A comprehensive advanced three-dimensional geological modeling method for highway tunnel construction in karst geology includes the following steps:

[0006] [1] Acquisition of data outside the tunnel

[0007] Perform surface drilling on the construction section to obtain hydrogeological information; use lidar to scan the three-dimensional point cloud data of the surface landform and establish a three-dimensional topographic map of the tunnel surface; use the high-density resistivity method to detect the three-dimensional position of the adverse geological structure of the tunnel.

[0008] [2] Acquisition of data inside the tunnel

[0009] Use geological radar three-dimensional imaging, advanced geological CT reflection imaging, and three-dimensional geological TGS detection to perform comprehensive geophysical exploration on the geological structure inside the tunnel to obtain data on adverse geology and lithology change zones in front of and around the tunnel face; use in-hole imaging of advanced horizontal drilling to obtain three-dimensional geological structural plane information such as rock stratum attitude and joint surface distribution.

[0010] [3] Data fusion

[0011] Fuse the detection data of steps [1] and [2], and successively establish the point model, surface model and volume model of the geological body in the construction section, and then construct the advanced three-dimensional geological model.

[0012] In the above comprehensive advanced three-dimensional geological prediction method for karst geological highway tunnel construction, the data fusion in step [3] includes the following steps:

[0013] 【3.1】 Take the three-dimensional coordinates of the center point of the tunnel face as the reference point for the data obtained by lidar, high-density resistivity method, 3D imaging of geological radar, advanced geological CT reflection imaging, and 3D geological TGS detection, form a geological point model under a unified coordinate, and perform surface fitting on different types of point data in the geological point model on different layers respectively. Finally, obtain an overall three-dimensional geological surface model;

[0014] 【3.2】 Perform discrete smooth interpolation on the three-dimensional geological surface model. According to the exposed line of the rock stratum and the attitude of the rock stratum, obtain the lithological interfaces including the terrain surface, completely weathered, and strongly weathered interfaces;

[0015] 【3.3】 Use the three-dimensional geological structural surface discrete data information of the attitude of the rock stratum and the distribution of joint surfaces obtained by the imaging in the advanced horizontal borehole to calibrate and correct the three-dimensional geological surface model and the parameters of the terrain surface, completely weathered, and strongly weathered interfaces obtained in step [3.2], obtain the calibrated fault and lithological interface information, and finally establish an advanced three-dimensional geological body model.

[0016] In the above comprehensive advanced three-dimensional geological prediction method for karst geological highway tunnel construction, the steps for obtaining the lithological interface in step [3.2] are as follows: Determine the coordinates of the exposed line of the rock stratum in space according to the exposed line of the rock stratum in the plan view, store it in a text file in the form of point set data, and then import it into the GOCAD software as a curve object. Obtain the extension vector of the exposed line of the lithology through the attitude of the rock stratum, and extend the exposed line of the lithology in space to obtain the lithological interface.

[0017] In the above comprehensive advanced three-dimensional geological prediction method for karst geological highway tunnel construction, the calibration and correction steps are to identify the detailed morphology of the joint crack identification model based on the imaging in the drilling hole and adjust the model accuracy.

[0018] In the above comprehensive advanced three-dimensional geological prediction method for karst geological highway tunnel construction, the steps for obtaining hydrogeological information by surface drilling in step [1] include: Conduct surface drilling and coring along the tunnel alignment to obtain the stratigraphic geological parameters and mechanical parameters in the drilling hole; Measure the karst, attitude of the rock stratum, lithology and joint distribution surrounding rock information of the stratum where the tunnel is located.

[0019] In the above comprehensive advanced three-dimensional geological prediction method for karst geological highway tunnel construction, the advanced horizontal drilling distance is 150m - 200m, and the in-hole imaging angle is 360°.

[0020] In the above comprehensive advanced three-dimensional geological prediction method for karst geological highway tunnel construction, the three-dimensional geological structural plane information of the rock formation attitude and joint plane distribution obtained by in-hole imaging of advanced horizontal drilling includes the surrounding rock formation attitude, joint development plane, joint thickness, and three-dimensional space coordinate information of the karst cavity development area.

[0021] The present invention provides a method for advanced three-dimensional geological inversion modeling and prediction of karst tunnels that combines data outside and inside the tunnel, fuses multi-source geological data, intelligently identifies rock formation attitude images, and is based on the convolutional neural network algorithm. It fuses multi-source geological exploration data using point set data and coordinate information, establishes point models, surface models, volume models, and parameter models for describing geological bodies, obtains an advanced three-dimensional geological model embedded with multiple parameters and attributes, and verifies it based on the discrete data of advanced horizontal drilling and in-hole imaging. Finally, all-round three-dimensional geological data consistent with the actual situation is obtained, thus achieving accurate prediction of geological parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a schematic diagram of the comprehensive three-dimensional geological modeling and prediction process of the present invention;

[0023] Figure 2 is a fusion diagram of advanced three-dimensional geological data based on point set three-dimensional coordinates of the present invention;

[0024] Figure 3 is a schematic diagram of obtaining three-dimensional imaging data by advanced three-dimensional geological radar detection in the model of the present invention;

[0025] Figure 4 is a schematic diagram of digital multi-element tunnel advanced geological CT imaging data in the model of the present invention;

[0026] Figure 5 is a schematic diagram of the three-dimensional geological prediction system TGS and transient electromagnetic data in the model of the present invention;

[0027] Figure 6 is a schematic diagram of 360° in-hole imaging of advanced horizontal drilling in the model of the present invention.

[0028] Figure 7 is a schematic diagram of multiple interfaces in the model construction of the present invention;

[0029] Figure 8 is a schematic diagram of the lithological interface and fault in the model construction of the present invention;

[0030] Figure 9 is a schematic diagram of the interface between different weathered strata in the model construction of the present invention;

[0031] Figure 10 It is a schematic diagram of the interface between different lithologies in the model construction of the present invention.

[0032] Figure 11 It is a schematic diagram of the correction of borehole image recognition in the model construction of the present invention.

[0033] The reference signs are: 1 - the first interface; 2 - the second interface; 3 - the lithological interface; 4 - the fault; 5 - the interface between different weathered strata; 6 - different fault planes; 7 - image data. Detailed implementation manners

[0034] The present invention will be further described below with reference to the accompanying drawings.

[0035] The flow chart of the comprehensive three-dimensional geological prediction method is as Figure 1 shown, and the specific steps are as follows:

[0036] I. Acquisition of data outside the tunnel

[0037] Perform surface drilling on the construction section to obtain hydrogeological information; use lidar to scan the three-dimensional point cloud data of the surface topography and establish a three-dimensional topographic map of the tunnel surface; use the high-density resistivity method to detect the three-dimensional positions of the unfavorable geological structures of the tunnel.

[0038] (1) Use the surface vertical drilling data of geological exploration, the basic geological conditions along the tunnel, advanced horizontal drilling analysis, high-density resistivity three-dimensional data, and conventional two-dimensional advanced geological radar detection, etc. to conduct a preliminary exploration of advanced geological prediction. The specific implementation steps include excavating a highway tunnel under karst geological conditions. First, conduct surface drilling, and analyze the formation hydrogeology through the drilling data; secondly, use airborne lidar to scan the surface above the tunnel to obtain the three-dimensional point cloud data of the surface topography, and then establish an accurate three-dimensional topographic map of the tunnel surface; then use the high-density resistivity method to detect the mountain along the tunnel to obtain the three-dimensional positions of unfavorable geological structures such as large fracture zones, soft interlayers, and karst caves in the tunnel; finally, conduct geological sketching and observation records on the tunnel face after excavation, and conduct joint experimental water quality analysis on the water inrush and surface water.

[0039] Among them, the analysis of the formation hydrogeology by the surface vertical drilling data includes: taking cores by surface drilling along the tunnel alignment to obtain the formation geological parameters and mechanical parameters in the borehole; measuring the surrounding rock information such as karst, rock occurrence, lithology, and joint distribution in the formation where the tunnel is located; through multiple drilling data and combined with the high-density resistivity method, conduct forward modeling and Bayesian inversion initial modeling of the formation, and establish the mountain model along the entire tunnel alignment, mainly including the mountain surface topographic map, and the three-dimensional coordinate positions of unfavorable geological structures such as large fracture zones, soft interlayers, and karst caves in the formation detected by inversion based on drilling formation data and the high-density resistivity method.

[0040] (2) For the adverse geological areas found in the preliminary results of advanced geological prediction, when the tunnel or inclined shaft heading face is excavated to a distance of 200 m from the adverse geological zone, conduct an airborne lidar scan of the tunnel surface in advance to obtain the 3D topographic point cloud data above the tunnel, and establish the 3D coordinate information of the surface topography in the adverse geological section to find the surface vegetation, gullies, water gathering areas, and water pits of the tunnel.

[0041] (3) For the sections where water gushes out from the advanced boreholes at the heading face, where mud and water inrusions have occurred, or where the risk of mud and water inrusions is preliminarily detected as high by the conventional 2D advanced geological radar in the tunnel, conduct geological sketching and observation records through the heading face after excavation; conduct a joint experimental water quality analysis of the water in the tunnel such as water gushing or water gushing in the borehole and the surrounding lakes, rivers, and streams on the surface, and preliminarily judge the connection path between the surface water and the water gushing in the tunnel by comparing the water quality components, such as the contents of total nitrogen, total phosphorus, fluoride, ammonia nitrogen, permanganate index, etc.; and conduct a statistical analysis of the reservoir water level to calculate the difference between the change in reservoir water volume and the water gushing volume in the tunnel.

[0042] II. Acquisition of Tunnel Data

[0043] Use comprehensive geophysical exploration methods such as 3D imaging of geological radar, advanced geological CT reflection imaging, and 3D geological TGS detection to obtain data on the geological structure in the tunnel, including the data of the 0 - 200 m ahead of the heading face and the surrounding adverse geology and lithology change zones; use in-hole imaging of advanced horizontal boreholes to obtain 3D geological structural plane information on the occurrence of rock strata and the distribution of joint planes.

[0044] (1) Utilize the directivity of the electromagnetic waves emitted by the geological radar antenna and adopt the directional radiation method to collect radar echo signals radially in front of the tunnel within a limited space. Qualitatively analyze the particle target media at different positions within the spatial area covered by the rays, invert the spatial position and scale form of the target media, and model and reconstruct the hydrogeological conditions in front of and around the tunnel.

[0045] (2) Adopt the digital multi-element advanced geological CT technology for 3D seismic wave advanced prediction detection.

[0046] (3) Adopt the 3D geological prediction system TGS technology and the transient electromagnetic method for advanced 3D geological ultra-long distance detection.

[0047] (4) Based on the specific location of the unfavorable geological body in front of the tunnel face obtained, advanced horizontal drilling with 360° in-hole imaging technology is used for advanced exploration. The borehole imaging data is utilized to verify the geophysical exploration results and, at the same time, provide specific and accurate geological images for the 3D geological body modeling. The occurrence of the surrounding rock strata, joint development planes, joint thicknesses, solution cavity development areas, etc. are located and identified, and converted into 3D spatial coordinate information such as lithological interfaces, solution cavity development surfaces, and joint spatial positions. Meanwhile, combined with the on-site borehole water inflow records, the water-containing volume of the solution cavity is predicted and analyzed to provide specific and accurate geological information for the 3D geological body modeling.

[0048] Through the rock mass identification of the images obtained from borehole imaging, the occurrence of the surrounding rock strata, joint development planes, joint thicknesses, solution cavity development areas, etc. are identified and converted into 3D spatial coordinate information such as lithological interfaces, solution cavity development surfaces, and joint spatial positions. The constructed surface model is verified and corrected.

[0049] III. Data Fusion

[0050] As Figure 2 shown, the above-mentioned out-of-tunnel and in-tunnel detection data are fused, and a point model, a surface model, and a volume model of the geological body in the construction section are established in sequence, and then an advanced 3D geological model is constructed. When constructing, the Python language is used to process the geological detection data, with the point set and coordinate information data in a unified format to achieve the fusion of multi-source geological data, and an advanced 3D geological model of the karst tunnel is established based on the point, surface, volume, and parametric models describing the geological body. The 3D data volume of the unfavorable geology is obtained by comprehensive geophysical exploration methods such as 3D geological radar imaging technology, digital multi-element advanced geological CT technology, and 3D geological prediction system TGS technology. Then, the 3D hydrogeological conditions are coordinate-reconstructed in the unified coordinate system in the form of a point set to obtain the 3D geological profiles of the surrounding rock fault fracture zones, mud-included water-bearing zones, and the 3D regional surfaces of the solution cavity development areas. The tunnel front dangerous situations such as water inrush, roof fall, water-bearing areas, and fracture zones can be identified through the processed images of the surrounding rock fault fracture zones and mud-included water-bearing zones.

[0051] (1) The data obtained from lidar, high-density resistivity method, 3D geological radar imaging, advanced geological CT reflection imaging, and 3D geological TGS detection are referenced by the 3D coordinates of the center point of the tunnel face to form a geological point model in the unified coordinate system, and the data in the geological point model is surface-fitted to obtain an overall 3D geological surface model.

[0052] During implementation, the three-dimensional coordinates of the center point of the tunnel face are used as the reference point and defined as the coordinate origin (0, 0, 0) of the three-dimensional space. The coordinates of the remaining points are relative coordinates with this point as the reference point. For example, in the point set data extracted from the two-dimensional contour map obtained by the high-density resistivity method, including the spatial coordinates and the corresponding resistivity, if the coordinates of the center point of the tunnel face are (x1, y1, z1), then in the new coordinate system, the new coordinates of the three-dimensional coordinate data points obtained by the inversion of the high-density resistivity method in the unified coordinate system are

[0053] (x′ i , y′ i , z′ i ) = (x i , y i , z i ) - (x1, y1, z1)

[0054] Meanwhile, along with the change of coordinates, the resistivity value remains unchanged. From λ i (x i , y i , z i ) it is changed to λ i (x′ i , y′ i , z′ i ).

[0055] In the unified coordinate system, the next step is to fit the point model into a surface model. Before fitting the surface for the constraint conditions, first, the triangular mesh of the surface is generated, and then the curve fitting is performed using the multi-function surface. When fitting a single surface, the following surface fitting formula is used:

[0056]

[0057] where a i , b i , c i , d are constant coefficients.

[0058] Select the degree n of the multi-function according to actual needs. Generally, the higher the n, the higher the accuracy of the fitted surface.

[0059] (2) Perform discrete smooth interpolation on the three-dimensional geological surface model. According to the outcrop line of the rock stratum and the occurrence of the rock stratum, obtain the lithological interfaces including the topographic surface, the completely weathered layer, and the strongly weathered interface.

[0060] When fitting multiple different surfaces, use the discrete smooth interpolation (DSI interpolation) function to perform interpolation fitting on the surface after dividing the grid.

[0061] Multiple different surfaces are due to obvious interfaces existing between different geological or stratigraphic formations. The main detection principle of the three-dimensional geological model obtained by each detection method is that different geological bodies or formations have different attribute characteristics, so as to distinguish the interfaces, and the data obtained are also the discrete coordinate points of the interfaces. Therefore, fitting out different surfaces is to find the interfaces. The discrete points of multiple surfaces obtained by detection are first connected by triangular meshes in combination with the topological relationship of grid nodes under the DSI interpolation method. The advantage is that it does not take spatial coordinates as parameters, so it is not restricted by dimensions. To establish a network of interconnected points among discretized data points, if the known node values on the network satisfy certain constraint conditions, the values on the unknown nodes can be obtained by solving linear equations.

[0062] The DSI method establishes an objective function for the optimal solution on the computational grid nodes, where is the global roughness function and is the linear constraint violation function. By minimizing the objective function, two objectives are achieved:

[0063] ① Minimize the global roughness function, so that the function value at any node approximates the mean value of the node values in the neighborhood of that point as much as possible, that is, make the value of each node as smooth as possible:

[0064] ② Convert the original sampled data into linear constraints defined on some nodes, and minimize the linear constraint violation function, that is, maximize the degree of compliance of the linear constraints, so that the values of related nodes approximate the sampled data as much as possible.

[0065] Mathematical description of the DSI interpolation algorithm: Inside the grid Ω formed by node connections, the known network node set is L, and the unknown network node set is I (I + L = Ω); f(*) is a piecewise continuous function inside Ω, and the function f(*) is assumed to be constant on the node set L. The purpose of the interpolation algorithm is to infer the expression of the interpolation function Φ(*) on the set I through f(*).

[0066] Obviously, the interpolation function can only approximate the unknown grid nodes infinitely. To select an "optimal" expression, a quadratic test function (global smoothness function) R(ψ) is used to test a possible interpolation function. The quadratic test function is shown as follows:

[0067] R(ψ) = ψ * [W] * ψ

[0068] Where [W] is a given positive definite symmetric matrix, and R(ψ) is determined by multiple local smoothness functions under linear constraints. Through the constraints of the test function, the optimal expression of the interpolation function can be obtained, and then the set of interpolation functions Φ(*) can be obtained.

[0069] According to the outcrop lines of rock strata in the plan view, determine the coordinates of the outcrop lines of rock strata in space, store them in a text file in the form of point set data, and then import them into the GOCAD software as curve objects. Obtain the extension vector of the lithology outcrop line through the occurrence of the rock strata, and extend the lithology outcrop line in space to obtain the lithology interface.

[0070] (3) Use the three-dimensional geological structural plane discrete data information of the occurrence of rock strata and the distribution of joint planes obtained from the in-hole imaging of advanced horizontal boreholes to verify and correct the three-dimensional geological surface model, topographic surface, and parameters of the fully weathered and strongly weathered interfaces obtained in step [3.2], obtain the information of the verified faults and lithology interfaces, and establish a fine advanced three-dimensional geological body model.

[0071] The following gives specific embodiments:

[0072] This implementation case is illustrated by taking the excavation of a karst geological highway tunnel and inclined shaft as an example. The tunnel is a two-way six-lane double-hole highway tunnel with a total length of 7978 m. There is 1 construction inclined shaft with a length of 867 m and a slope of 12%. The cross-sectional area of the main tunnel is 163 ㎡. There are several main difficulties in the excavation of the main tunnel and the inclined shaft: ① The strata are mainly composed of dolomitic limestone intercalated with siltstone and shale, with developed fissures, fragmented rock mass, poor stability, and prone to collapse and spalling during construction; ② The tunnel and the inclined shaft pass through multiple karst development areas, fault fracture zones, water-rich areas and other areas with high geological disaster risks, and are prone to water inrush and mud gushing disasters; ③ The main tunnel of the tunnel is a large cross-section of three lanes on the expressway, and large deformations and collapses are likely to occur when encountering soft and fractured surrounding rocks; ④ The slope of the inclined shaft is 12%. When the amount of mud and water gushing is large, it will affect the construction progress and seriously threaten personal safety. At the same time, drainage will greatly increase the construction cost.

[0073] The comprehensive three-dimensional geological prediction method is as follows:

[0074] I. Acquisition of data outside the tunnel

[0075] (1) When excavating a highway tunnel under karst geological conditions, first drill holes on the surface, analyze the hydrogeology of the strata through the drilling data; secondly, use an airborne lidar to scan the surface above the tunnel to obtain three-dimensional point cloud data of the surface topography, and then establish an accurate three-dimensional topographic map of the tunnel surface; then use the high-density resistivity method to detect the mountain along the tunnel to obtain the three-dimensional positions of bad geological structures such as large fault fracture zones, soft interlayers, and karst caves along the tunnel; finally, conduct geological sketching and observation records on the tunnel face after excavation, and jointly conduct experimental water quality analysis on water inrush and other surface water. The preliminary exploration results show that the tunnel and the inclined shaft pass through multiple karst development areas, fault fracture zones, water-rich areas and other areas with high geological disaster risks, and are prone to water inrush and mud gushing disasters, and it is found that there are multiple reservoirs near the surface along the tunnel.

[0076] The analysis of the formation hydrogeology based on the surface vertical borehole data includes: taking core samples from surface boreholes along the tunnel alignment to obtain the formation geological parameters and mechanical parameters in the boreholes; measuring the surrounding rock information such as karst, rock occurrence, lithology and joint distribution in the formation where the tunnel is located; establishing the initial model of forward formation modeling and Bayesian inversion by using multiple borehole data combined with the high-density resistivity method to establish the mountain body model along the entire tunnel. After excavation, geological sketching and observation records are carried out on the tunnel face, and the experimental water quality analysis is jointly carried out on the water inrush and other water and surface water, including: timely carrying out water quality analysis and tests on the inrush water in the tunnel, the surrounding lakes, rivers, streams, etc. on the surface, and comparing the water quality components to preliminarily judge whether there is a connected path between the surface water and the inrush water in the tunnel.

[0077] (2) For the poor geological areas existing in the preliminary exploration results of the advanced geological prediction, when the tunnel or inclined shaft tunnel face is excavated to a distance of 200 m from the poor geological zone, the airborne lidar is used to scan to obtain the three-dimensional terrain point cloud data above the tunnel. It is found that the surface vegetation of the tunnel is dense, and there are multiple gullies, water collection areas and water pits, etc.

[0078] (3) For the sections with high risk of mud and water inrush detected by the advanced borehole water inrush at the tunnel face, the existing mud and water inrush or the conventional two-dimensional advanced geological radar in the tunnel, through geological sketching and observation records on the tunnel face after excavation, the experimental water quality analysis is jointly carried out on the water inrush in the tunnel or borehole and the surrounding lakes, rivers, streams, etc. on the surface. By comparing the water quality components, such as the content of total nitrogen, total phosphorus, fluoride, ammonia nitrogen, permanganate index, etc., it is judged that the chemical composition difference between the reservoir water and the tunnel water inrush is large, and through the statistical analysis of the reservoir water level, it is found that the change amount of the reservoir water is quite different from the difference in the inrush water volume in the tunnel. It is speculated that the current hydraulic connection between the two is not close, and it is preliminarily judged that there is no connected path between the surface water and the inrush water in the tunnel.

[0079] II. Acquisition of tunnel data

[0080] (1) Utilizing the directivity of the electromagnetic wave emitted by the geological radar antenna, adopting the directional radiation method, collecting the radar echo signals radially in front of the tunnel in a limited space, qualitatively analyzing the particle target media at different positions in the spatial area covered by the ray, and inversely calculating the spatial position and scale form of the target media to reconstruct the hydrogeological conditions in front of and around the tunnel by modeling.

[0081] The three-dimensional imaging data is obtained by the advanced three-dimensional geological radar detection in the tunnel. As Figure 3 shown, it can be seen that most of the measurement and control area is within the control range of the fault fracture zone, the surrounding rock of the tunnel body is rich in water, the joint fissures are developed, and the runoff direction of the fissure water is mainly from top to bottom; the mud and water inrush at the current tunnel face has formed a fissure water runoff channel in the surrounding rock of the tunnel body, and the runoff channel mainly extends from top to bottom along the left side of the tunnel body, and the influence scale of the channel is more than 3000 m 3, which is the main channel for sudden mud and water gushing; the surrounding rock of the tunnel section XK0+040~XK0+020 is rich in water and is the main source of sudden water gushing; the left side of XK0+036 mileage is the main outlet point of sudden mud and water gushing.

[0082] (2) Use digital multi-element tunnel advanced geological CT technology to conduct three-dimensional seismic wave advance prediction detection results, such as Figure 4 As shown, it can be seen that within the scope of this detection, the longitudinal wave velocity of the rock mass is between 1500 and 3500 m / s, mainly dolomite, with developed joints and fissures, relatively broken rock mass, developed groundwater, and rich water. Among them, the XK0+038~XK0+008 section is karst, and there is an obvious through-type cavities on the left side of the centerline of the face, which are filled with a small amount of water and partially filled with mud. The construction is very likely to cause landslides or mud bursts, and the risk level is high; the main speculation is that the surrounding rock level within the detection range is V-level, the rock strength is low, and the self-stabilization ability is poor, which belongs to the complex geological conditions of the karst section.

[0083] (3) Using the 3D geological prediction system TGS technology and transient electromagnetic method to obtain advanced 3D geological ultra-long distance detection data results, such as Figure 5 As shown, it is concluded that: in the tunnel mileage section XK0+031.9~XK0+028.0, Vp, density and static Young's modulus are low, and it is speculated that the surrounding rock in this section is extremely unstable, the surrounding rock is broken, and the rough hole is prone to falling blocks and collapse. In the tunnel mileage section XK0+012.5~XK0-008.4, Vp / Vs and Poisson's ratio are high, and it is speculated that the surrounding rock in this section is extremely unstable and the possibility of water inrush is high.

[0084] (4) Based on the specific location of the unfavorable geological body in front of the tunnel face, the 360° imaging technology in the advanced horizontal borehole is used for advanced drilling. The horizontal drilling distance is up to 150m. The borehole imaging data is used to verify the geophysical exploration results and provide specific and accurate geological images for 3D geological body modeling. Figure 6 By performing rock mass image recognition on the borehole imaging data, the occurrence of the surrounding rock strata, joint development surface, joint thickness, dissolution cavity development area, etc. are identified and converted into three-dimensional spatial coordinate information such as lithology interface, dissolution cavity development surface, joint spatial position, etc., and the surface model constructed above is verified and corrected.

[0085] Advance horizontal drilling and borehole imaging show that there is a cavity 0-15 meters in front of the existing heading face, which is formed by mud and water inrush. The surrounding rock is loose from 15-35 meters, and the risk of water and mud inrush is relatively high. The dolomite core is broken from 35-60 meters, showing a fragmented structure, with poor cementation effect, calcareous filling, and local strong water-rich, and the risk of mud and water inrush is high. Among them, during the advance horizontal drilling process, many problems such as the outburst of confined water in the borehole, borehole collapse, a large amount of loose rock debris, and blocked boreholes were encountered. For the problem of confined water, a method of binding a thin steel bar with a diameter of 6 mm and a length of 2 m on the flexible push rod of the borehole camera was adopted, and the borehole camera was gradually pushed into the borehole against the pressure of the confined water for detection; for the problem of damaging the borehole camera during borehole collapse, a simple steel bar protective shell was made and put outside the borehole camera for protection, and at the same time, it did not affect the imaging quality; for the problems of a large amount of loose rock debris and blocked boreholes, after drilling, a fan was used to blow air into the borehole to clean the loose rock debris and obstacles such as silt and fine sand blocking the borehole. Finally, the imaging problem of borehole camera was successfully realized under the problems of outburst of confined water in the borehole, borehole collapse, a large amount of loose rock debris, and blocked boreholes.

[0086] III. Data Fusion

[0087] (1) The data obtained from lidar, high-density resistivity method, 3D imaging of geological radar, advance geological CT reflection imaging, and 3D geological TGS detection are referenced to the 3D coordinates of the center point of the heading face to form a geological point model under a unified coordinate, and the data in the geological point model are surface-fitted to obtain an overall 3D geological surface model.

[0088] (2) As Figures 7 to 10 shown, discrete smooth interpolation is performed on the 3D geological surface model. According to the outcrop line of the rock stratum and the occurrence of the rock stratum, the lithological interfaces including the terrain surface, completely weathered, and strongly weathered interfaces are obtained, and the discrete smooth interpolation (DSI interpolation) function is used for interpolation fitting on the surface after dividing the grid.

[0089] Among them Figure 7 shows a model schematic diagram of multiple interfaces including the first interface 1 and the second interface 2; Figure 8 shows a model schematic diagram of the lithological interface 3 and the fault 4; Figure 9 shows a model schematic diagram of the different weathered stratum interfaces 5; Figure 10 shows a model schematic diagram of the lithological interface of different fault planes 6.

[0090] (3) As Figure 11 shown, based on the discrete 3D geological structural plane image data 7 such as the occurrence of the rock stratum and the distribution of joint planes obtained from the in-hole imaging of the advance horizontal drilling, the 3D geological surface model and the parameters of the terrain surface, completely weathered, and strongly weathered interfaces are calibrated and corrected to obtain the calibrated fault and lithological interface information, and a fine advance 3D geological body model is established.

[0091] Extend the lithology outcrop line along the calculated direction vector, and then trim the surface according to the modeling area to obtain the macroscopic morphology of the lithology interface. Further, add constraints to the extended surface (the constraints here are discrete data obtained through boreholes), perform DSI fitting interpolation, and the resulting fault surface is consistent with the lithology outcrop line and the bedding attitude in macroscopic morphology and is consistent with the discrete data determined by boreholes locally. Finally, multiple fault surfaces are formed.

[0092] After all the topographic surface, various weathering interfaces, lithology interfaces, etc. are all edited, construct the volume model through the surface model. Input the stratigraphic information in combination with the geological exploration report, and determine the integration relationship between the strata according to the actual situation. Check the morphology of the volume model at any time through the profile, and adjust the morphology and accuracy of the model according to the joint fracture identification in the borehole imaging of the drilling holes. Finally, establish an advanced three-dimensional geological body model with multiple parameters and attributes embedded, which can intuitively reflect the advanced three-dimensional geological conditions during the construction of the karst geological highway tunnel.

[0093] During specific implementation, integrate all the above-mentioned advanced geological exploration data, such as the surface vertical borehole data of geological exploration, the borehole imaging data of the long-distance advanced horizontal boreholes in the tunnel, the three-dimensional topographic point cloud data of the tunnel scanned by the airborne lidar of the unmanned aerial vehicle, the three-dimensional data of high-density resistivity, the three-dimensional imaging data of the geological radar, the digital multi-element advanced three-dimensional geological CT data, the three-dimensional geological prediction system TGS data, etc., convert them into the three-dimensional coordinate information of the point set under the unified coordinate system for advanced three-dimensional geological modeling, so as to accurately model and predict the three-dimensional geology 100m - 200m in front of the tunnel face on the basis of realizing the integration of multi-source geological data. It is concluded that the inclined shaft section belongs to the middle-low mountain landform area of tectonic denudation and corrosion, and the engineering geological and hydrogeological conditions are complex; according to the comprehensive analysis of the surrounding rock during the geological survey, drilling and the original inclined shaft construction process, the karst development degree of the tunnel body is moderately developed, but the possibility of karst development in the fault fracture zone cannot be excluded; the difficulty of the original inclined shaft entering the left line plan is high, and the risk of water inrush and mud burst occurring again is high. It is recommended to optimize the tunnel entry plan.

[0094] Using the combined advanced 3D geological model to accurately guide the construction of mud and water inrush disasters and take the most favorable prevention and control measures. From the detection results, it is necessary to avoid high-risk poor geological sections, optimize the alignment of the inclined shaft. Through advanced detection, it is known that there is a wide fissure water outlet near ZK35+265 on the left line of the optimized plane position of the inclined shaft, and the alignment should pay attention to avoidance; from ZK35+265 to the small mileage direction, the core integrity gradually improves and shows a blocky shape with short columns, the tunnel wall is relatively smooth, and no obvious fissure water outlet is seen; the right side of the existing inclined shaft face is better than the left side. According to the results of geophysical exploration and advanced horizontal drilling, it is recommended to set the optimized plane starting point between XK0+065 and XK0+080 on the right side of the existing inclined shaft face, and the end point between ZK35+240 and ZK35+255 on the left line of the tunnel; the surrounding rock of the optimized plane of the inclined shaft is dolomite, dolomitic limestone intercalated with marl of the Longwangmiao Formation, the rock mass is relatively fractured - fractured, and the surrounding rock grade is all grade V. During the construction of the inclined shaft, comprehensive advanced prediction methods such as TSP, TGS, TCT, transient electromagnetic, 3D geological radar and in-hole imaging of advanced horizontal drilling are used.

[0095] According to the results of advanced geological prediction, timely optimize and adjust the design to avoid collapses and roof falls. Finally, the inclined shaft safely and smoothly enters the main tunnel, reducing the construction safety risk and greatly saving the economic construction cost at the same time.

Claims

1. A comprehensive advanced three-dimensional geological modeling method for karst geological highway tunnel construction, characterized in that , including the following steps: 【1】 Acquisition of data outside the tunnel Drill holes on the ground surface of the construction section to obtain hydrogeological information; use lidar to scan the three-dimensional point cloud data of the ground surface topography and establish a three-dimensional topographic map of the tunnel ground surface; use the high-density resistivity method to detect the three-dimensional position of the unfavorable geological structure of the tunnel. 【2】 Acquisition of data inside the tunnel Use comprehensive geophysical exploration methods such as geological radar three-dimensional imaging, advanced geological CT reflection imaging, and three-dimensional geological TGS detection to obtain data on unfavorable geology and lithology change zones in front of and around the tunnel face; use in-hole imaging of advanced horizontal drill holes to obtain three-dimensional geological structural plane information on the occurrence of rock strata and the distribution of joint planes. 【3】 Data fusion Fuse the detection data in steps 【1】 and 【2】, and successively establish a point model, a surface model, and a volume model of the geological body in the construction section, and then construct an advanced three-dimensional geological model. The data fusion in step 【3】 includes the following steps: 【3.1】 Take the three-dimensional coordinates of the center point of the tunnel face as the reference point for the data obtained by lidar, high-density resistivity method, geological radar three-dimensional imaging, advanced geological CT reflection imaging, and three-dimensional geological TGS detection, form a geological point model under a unified coordinate system, and perform surface fitting on different types of point data in the geological point model on different layers respectively. Finally, obtain an overall three-dimensional geological surface model. 【3.2】 Perform discrete smooth interpolation on the three-dimensional geological surface model, and obtain lithological interfaces including the terrain surface, completely weathered, and strongly weathered interfaces according to the outcrop line of the rock strata and the occurrence of the rock strata. Use the discrete data information of the three-dimensional geological structural plane of the occurrence of the rock strata and the distribution of joint planes obtained by in-hole imaging of advanced horizontal drill holes to calibrate and correct the parameters of the three-dimensional geological surface model and the terrain surface, completely weathered, and strongly weathered interfaces obtained in step 【3.2】, obtain the calibrated fault and lithological interface information, and finally establish an advanced three-dimensional geological body model.

2. The comprehensive advanced three-dimensional geological modeling method for karst geological highway tunnel construction according to claim 1, wherein: The steps for obtaining the lithological interface in step 【3.2】 are as follows: Determine the coordinates of the outcrop line of the rock strata in space according to the outcrop line of the rock strata in the plan view, store it in a text file in the form of point set data, and then import it into the GOCAD software as a curve object. Obtain the extension vector of the lithological outcrop line through the occurrence of the rock strata, and extend the lithological outcrop line in space to obtain the lithological interface.

3. The comprehensive advanced three-dimensional geological modeling method for karst geological highway tunnel construction according to claim 1, characterized in that: The calibration and correction in step 【3.3】 are based on the detailed morphology of the joint fracture identification model of in-hole imaging of drill holes and adjust the model accuracy.

4. The comprehensive advanced three-dimensional geological modeling method for karst geological highway tunnel construction according to claim 1, characterized in that: The steps for obtaining hydrogeological information by surface drilling in step 【1】 include: Take core samples by surface drilling along the tunnel alignment to obtain the stratigraphic geological parameters and mechanical parameters inside the drill holes; measure the karst, occurrence of rock strata, lithology, and joint distribution surrounding rock information of the strata where the tunnel is located.

5. The comprehensive advanced three-dimensional geological modeling method for karst geological highway tunnel construction according to claim 1, characterized in that: The distance of the advanced horizontal drill hole is 150m - 200m, and the in-hole imaging angle is 360°.

6. The comprehensive advanced three-dimensional geological modeling method for karst geological highway tunnel construction according to claim 1, characterized in that: The three-dimensional geological structural plane information of the occurrence of the rock strata and the distribution of joint planes obtained by in-hole imaging of the advanced horizontal drill hole includes the occurrence of the surrounding rock strata, the joint development surface, the joint thickness, and the three-dimensional space coordinate information of the solution cavity development area.

Images