A geophysical advanced detection method and system for underground engineering

Abstract

The present application relates to the field of geophysical technology, and more particularly to a geophysical advanced detection method and system for underground engineering; the present application introduces a dynamic boundary weight correction factor based on equivalent medium theory, combines an adaptive matching algorithm of a measured background field and a theoretical model, determines a dynamic boundary weight correction suitable for the region, corrects measured data based on the dynamic weight correction factor, thereby effectively separates distortion caused by a roadway geometric structure and a real geological anomaly response, the result is accurate, and the identification accuracy of a water-bearing structure is significantly improved.

Description

Technical Field

[0001] This invention relates to the field of geophysics, and in particular to a geophysical advanced detection method and system for underground engineering. Background Technology

[0002] Excavation operations in underground engineering projects such as coal mine roadways often face the threat of water hazards such as sudden water inrush and sand collapse. To address these water hazard issues, direct current resistivity detection technology is one of the commonly used methods. This technology is based on the difference in rock conductivity. By deploying an electrode array at the tunnel face and constructing a full-space direct current electric field, the electric field distribution characteristics are analyzed to invert the apparent resistivity of the rock strata ahead, thereby identifying water-bearing structures.

[0003] In advanced underground engineering detection, the electric field response detected by DC electrical resistivity tomography (EDT) not only contains information on the electrical differences in the water-bearing structure of the target geological body, but is also significantly affected by the unique full-space geometry formed by tunnel excavation. Existing technologies include schemes for correcting the unique full-space geometry formed by tunnel excavation. For example, a journal article on three-dimensional EDT advanced detection technology based on a comparative method to eliminate tunnel influence discloses a scheme for eliminating tunnel influence based on a comparative method. However, this scheme involves establishing four models for comparison, meshing, loading, and solving the models; and analyzing from the perspective of equipotential potential changes. The process is complex, the correction effect is poor, and the interpretability and portability are weak. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a geophysical advanced detection method and system for underground engineering, which solves the problems existing in the prior art.

[0005] This invention provides a geophysical advanced detection method for underground engineering, the method comprising the following steps: S1: A dynamically adjustable electrode array is deployed on the face of the tunnel to be tested; S2: Construct a stable DC electric field in the entire space, and collect the potential of the working face to be measured based on the dynamically adjustable electrode array; S3: Construct a dynamic boundary weight correction factor based on the equivalent medium theory, and perform optimization calculations on the hyperparameters of the dynamic boundary weight correction factor; Specifically, S3 is: S3.1: Construct dynamic boundary weight correction factors; Its expression is: ; In the formula, Let r be the three-dimensional full-space distortion correction factor for the observation point, and r be the full-space position vector of the observation point. θ is the equivalent radius of the tunnel, θ is the angle between the observation point position vector and the tunnel axis, α is the resistivity ratio of the tunnel wall and the surrounding rock, and λ is the correction coefficient for the geometric asymmetry of the tunnel. S3.2: Acquire the potential of the background field based on the dynamically adjustable electrode array; S3.3: Based on the aforementioned three-dimensional full-space theoretical model, input actual engineering parameters and calculate the electric potential of the theoretical normal field; S3.4: Optimize the hyperparameters in the dynamic boundary weight correction factor based on the relative deviation between the theoretical normal field potential and the background field potential; S4; The potential of the test face is corrected according to the dynamic boundary weight correction factor to obtain the corrected potential of the test face. S5: Calculate the apparent resistivity of each measuring point based on the corrected potential of the working face to be measured. S6: Establish a three-dimensional full-space theoretical model, and perform an inversion operation on the apparent resistivity of each measuring point based on the three-dimensional full-space theoretical model to reconstruct the resistivity distribution image of the rock strata in front of the tunnel face to be measured, thereby realizing advanced detection of underground engineering.

[0006] Preferably, in step S3.2, before the formal geological exploration, a background field measurement is performed without any geological targets. The specific data acquisition process is as follows: Maintaining the same electrode layout parameters as in the actual detection, and keeping the power supply current intensity I consistent, the potential difference of each measuring electrode pair is collected. , which is the electric potential of the background field, serves as a blank reference for the distortion of the entire space.

[0007] Preferably, in S3.3, the actual engineering parameters include: tunnel cross-sectional dimensions, resistivity parameters of surrounding rock and tunnel wall, and power supply system parameters; the power supply system parameters include the power supply electrode distance AB, the measuring electrode distance MN, and the power supply current I.

[0008] Preferably, S3.4 specifically includes: Set initial values ​​for the hyperparameters; calculate the relative deviation between the theoretical normal field potential and the background field potential; use a genetic algorithm to find the hyperparameter combination that minimizes the relative deviation, which is then used as the exclusive hyperparameter for the current tunnel environment.

[0009] Preferably, a genetic algorithm is used to find the hyperparameter combination that minimizes the relative deviation, which is used as the exclusive hyperparameter for the current roadway environment. Specifically, the population is initialized, which is the value of the set hyperparameter combination. The fitness of each hyperparameter group is calculated. Selection, crossover, and mutation operations are performed to generate a new generation of population. The fitness calculation and crossover and mutation operations are repeated until the convergence condition is met. The convergence criteria are: root mean square relative error (RMSE) < 5%; number of iterations > 50.

[0010] Preferably, in step S4, the corrected potential at the observation point on the face to be measured is... for: ; In the formula, The corrected potential at the observation point of the tunnel face to be measured is used for subsequent inversion calculations. The dynamic boundary weight correction factor for the observation points on the tunnel face to be measured. The measured potential is the actual potential at the observation point of the tunnel face to be tested.

[0011] Preferably, in step S5, the apparent resistivity at each measuring point is... The calculation formula is: ; In the formula, K is the device coefficient and I is the power supply current.

[0012] Preferably, in step S6, the three-dimensional full-space theoretical model is described by basic governing equations, boundary conditions, and medium interface conditions; The basic governing equation is expressed as follows: ; In the formula, Here, σ is the gradient operator, and σ is the dielectric conductivity. The full-space position vector of the observation point The potential at the point is given by I, the DC current intensity between the supply electrodes, and δ is the Dirac delta function. Let A be the spatial position vector of the power supply electrode A. The spatial position vector of the power supply electrode B; The boundary conditions include infinity boundary conditions and power supply electrode boundary conditions; The expression for the boundary condition at infinity is: In the formula, The full-space position vector of the observation point The potential at the point; The boundary conditions for the power supply electrodes are as follows: at the locations of power supply electrodes A and B, the potentials are set to fixed values, and the expressions satisfying these conditions are: ; In the formula, and Let A and B be the potentials of the power supply electrodes, respectively. For a fixed voltage value; The medium interface condition is as follows: at the interface between the tunnel wall and the surrounding rock, the following electric field continuity condition is satisfied: ; In the formula, the superscripts (1) and (2) represent the two media, the tunnel wall and the surrounding rock, respectively. and These represent the electric potentials at the interface between the tunnel wall and the surrounding rock, respectively. and The electrical conductivity of the tunnel wall and the surrounding rock are two different media. This is the normal derivative of the interface.

[0013] Preferably, step S6 further includes: using a three-dimensional inversion algorithm with least squares method and regularization constraints to invert the apparent resistivity data of multiple measurement points and reconstruct the resistivity distribution image of the rock strata in front of the tunnel face.

[0014] According to another aspect of the present invention, a geophysical advanced detection system for underground engineering is provided, the system employing the above-described geophysical advanced detection method for underground engineering, the system comprising: The acquisition module is used to deploy a dynamically adjustable electrode array on the working face to be tested, construct a stable DC electric field in the whole space, and acquire the potential of the working face to be tested based on the dynamically adjustable electrode array. The module for constructing and solving dynamic boundary weight correction factors is used to construct dynamic boundary weight correction factors based on equivalent medium theory and to optimize the hyperparameters of the dynamic boundary weight correction factors. The correction module is used to correct the potential of the test face according to the dynamic boundary weight correction factor to obtain the corrected potential of the test face. Apparent resistivity calculation module, used to calculate the apparent resistivity of each measuring point based on the corrected potential of the working face to be measured; The inversion and detection module is used to establish a three-dimensional full-space theoretical model and perform an inversion operation on the apparent resistivity of each measuring point based on the three-dimensional full-space theoretical model to reconstruct the resistivity distribution image of the rock strata in front of the tunnel face to be measured, thereby realizing advanced detection of underground engineering.

[0015] The embodiments of the present invention have the following technical effects: This invention introduces a dynamic boundary weight correction factor based on the equivalent medium theory, and combines it with an adaptive matching algorithm between the measured background field and the theoretical model to determine a dynamic boundary weight calibration suitable for the region. Based on the dynamic weight calibration factor, the measured data is corrected, thereby effectively separating the normal distortion caused by the roadway geometry from the real geological anomaly response. The correction is more accurate and significantly improves the identification accuracy of water-bearing structures.

[0016] This invention introduces the concept of dynamic boundary weights to incorporate the distance dependence and orientation modulation of roadway geometric distortion into a three-dimensional full-space theoretical model, eliminating the influence of roadway geometric distortion and matching the theoretical distortion factor with the actual observed distortion characteristics. This corrects the distortion deviation of the three-dimensional full-space theoretical model, resulting in strong interpretability and a refined description of the electric field distortion in the entire space. Simultaneously, through multiple exponential decay functions of the three-dimensional spatial location, it achieves a refined characterization of the distortion effects at different orientations and depths of the roadway. For the first time, it considers the asymmetric influence of observation orientation and depth on distortion effects in a full-space DC electric field method. The calculation is simple, and it significantly improves the accuracy and adaptability of electric field distortion correction in complex roadway environments. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a geophysical advanced detection method for underground engineering provided by an embodiment of the present invention; Figure 2 This is a flowchart provided by an embodiment of the present invention for constructing a dynamic boundary weight correction factor based on the equivalent medium theory and optimizing the hyperparameters of the dynamic boundary weight correction factor. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0020] Example 1, Figure 1 A flowchart of a geophysical advanced detection method for underground engineering is shown, such as... Figure 1 As shown, a geophysical advanced detection method for underground engineering includes the following steps: S1: A dynamically adjustable electrode array is deployed on the face of the tunnel to be tested; The dynamically adjustable electrode array, through modular electrode groups and an electric adjustment device, enables real-time adjustment of the electrode distance between the power supply electrode and the measuring electrode, while optimizing the contact stability between the electrode and the working face, ensuring efficient acquisition of electric field signals throughout the entire space.

[0021] Specifically, the dynamically adjustable electrode array consists of power supply electrode groups (A, B) and measurement electrode groups (M, N). Each electrode group is modularly designed, including an electrode unit, an insulating support bracket, and an electric adjustment device. The electrode unit is made of a highly conductive material, with a hemispherical contact head at its front end to increase the contact area with the working face, and its rear end connected to the data acquisition system via an insulated wire. In this embodiment, the highly conductive material is a copper rod with a diameter of 10-15 mm and a silver-plated surface to prevent oxidation.

[0022] The electrode unit is fixed to a lightweight insulating support via a rotatable base. The bottom of the support is equipped with a magnetic adsorption layer or mechanical clamps to ensure that the electrode unit is in close contact with the tunnel face and does not shift due to tunneling vibrations. In this embodiment, the lightweight insulating support is made of ABS engineering plastic with an antistatic coating.

[0023] Each electrode unit is equipped with a telescopic rod driven by a micro stepper motor. The telescopic rod has a stroke range of 0~500mm and an adjustment accuracy of ±1mm. It receives instructions from controllers such as PLCs or embedded microprocessors to realize the dynamic adjustment of the electrode pitch.

[0024] S2: Construct a stable DC electric field in the entire space, and collect the potential of the working face to be measured based on the dynamically adjustable electrode array; The full-space DC electric field is formed in the tunnel and surrounding rock strata after a stable DC current (I) is introduced through the power supply electrode group (A, B). Because tunnel excavation disrupts the continuity of the original strata, its geometry significantly alters the natural distribution of the current field. The tunnel wall is a high-resistivity cavity, while the surrounding rock is a low-resistivity medium, thus forming an electrical abrupt boundary. This causes current lines to converge towards the tunnel wall, resulting in an abnormally high potential near the working face. Therefore, it is necessary to construct a stable DC electric field throughout the entire space.

[0025] The construction of a stable DC electric field in the entire space is specifically as follows: before formal data acquisition, a small current is first introduced to allow the current field to gradually penetrate into the surrounding rock, avoiding local polarization caused by a sudden large current. Then, the target current is switched and maintained stably for more than 120 seconds to ensure that the current distribution in the entire space reaches a quasi-steady state, with potential difference fluctuations ≤ ±1mV.

[0026] The small current is 1~2A and lasts for 30s; the target current is 5~20A.

[0027] The specific steps for acquiring the potential of the working face based on the dynamically adjustable electrode array are as follows: The power supply electrode groups (A and B) are positioned in a stable surrounding rock area 2-3 meters behind the tunnel face to avoid disturbance near the working face. The contact surfaces between the power supply electrode groups (A and B) and the surrounding rock are coated with conductive paste to reduce contact resistance and ensure efficient current injection into the formation. The power supply for the power supply electrode groups (A and B) is a high-power DC constant current source with an output range of 5-20A and a power ≥5kW. The current intensity can be dynamically adjusted according to the background resistivity of the surrounding rock. Furthermore, electrode polarization effects, which could lead to potential signal distortion, can be avoided by extending the power supply time.

[0028] The measuring electrode group (M, N) is made of the same material as the power supply electrode group (A, B), and is distributed in a fan-shaped measuring line along the tunnel excavation direction, covering a range of 0-30m ahead. The specific parameters of the measuring electrode group (M, N) are as follows: Number of survey lines: 8 to 12, adjusted according to the width of the tunnel. When the tunnel width is ≤5m, take 8 lines; when the tunnel width is >5m, take 12 lines. Survey line angle: 30°~60° with the normal to the tunnel face, prioritizing coverage of potential anomalies directly in front of and to both sides of the tunnel face; Measurement point spacing: Each measurement line contains 5 to 10 measurement points, with a spacing of 0.5m between each measurement point. The spacing in the near-field 0 to 5m area is increased to 0.3m, and the total number of measurement points is ≥40 per detection.

[0029] S3: Construct a dynamic boundary weight correction factor based on the equivalent medium theory, and perform optimization calculations on the hyperparameters of the dynamic boundary weight correction factor; In advanced underground engineering exploration, the electric field response detected by direct current resistivity tomography (DCST) not only contains information on the electrical differences in the water-bearing structure of the target geological body, but is also significantly affected by the unique full-space geometry formed by tunnel excavation. Traditional methods do not fully consider this full-space effect, misinterpreting normal current field distortions caused by tunnel walls as geological anomalies, resulting in decreased reliability of the detection results.

[0030] Based on this, this embodiment introduces a dynamic boundary weight correction factor based on the equivalent medium theory. Combined with the adaptive matching algorithm of the measured background field and the theoretical model, a dynamic boundary weight calibration suitable for the region is determined. The measured data is corrected based on the dynamic weight calibration factor, thereby effectively separating the normal distortion caused by the roadway geometry from the real geological anomaly response, and significantly improving the identification accuracy of water-bearing structures.

[0031] Specifically, such as Figure 2 As shown, S3 specifically includes: S3.1: Constructing the dynamic boundary weight correction factor The finite low-resistivity cavities created during tunnel excavation are embedded within an infinitely large high-resistivity surrounding rock medium, disrupting the continuity of the natural strata. Current lines, due to the low-resistivity characteristics of the tunnel walls, deflect and concentrate towards the tunnel walls, resulting in an abnormally high potential near the working face. This geometric distortion masks the electric field response of the deep geological body. Traditional methods do not distinguish between tunnel geometric distortion and geological anomalies; therefore, a correction factor to quantify the impact of distortion is needed to separate the two.

[0032] The core of the equivalent medium theory is to replace complex geometric structures with equivalent media while retaining their electrical response characteristics. This embodiment, based on the equivalent medium theory, introduces the concept of dynamic boundary weight. This means that the influence of the tunnel's geometric boundary on the current field distortion is not uniform but dynamically changes with the location of the observation point. The closer the observation point is to the tunnel wall, the more severe the current line deflection distortion, and the greater the distortion weight should be. The current line distortion patterns along the tunnel axis and perpendicular to the axis differ, requiring consideration of the azimuth angle. To simultaneously capture the influence of distance and angle on distortion, this embodiment proposes a dynamic boundary weight correction factor, the expression of which is: ; In the formula, Let r be the three-dimensional full-space distortion correction factor for the observation point, and r be the full-space position vector of the observation point. θ is the equivalent radius of the tunnel, θ is the angle between the observation point position vector and the tunnel axis, α is the resistivity ratio of the tunnel wall and the surrounding rock, and λ is the tunnel geometric asymmetry correction coefficient, with a value range of [0, 1], reflecting the degree of deviation between the actual shape of the tunnel and the ideal circle / rectangle.

[0033] This embodiment introduces the concept of dynamic boundary weights to incorporate the distance dependence and azimuth modulation of roadway geometric distortion into the three-dimensional full-space theoretical model, eliminating the influence of roadway geometric distortion and matching the theoretical distortion factor with the actual observed distortion characteristics. This corrects the distortion deviation of the three-dimensional full-space theoretical model, achieving a refined description of the full-space electric field distortion. Simultaneously, through multiple exponential decay functions of three-dimensional spatial position, it achieves a refined characterization of the distortion effects at different azimuths and depths of the roadway. For the first time, it considers the asymmetric influence of observation azimuth and depth on the distortion effect in the full-space DC electric method, significantly improving the accuracy and adaptability of electric field distortion correction in complex roadway environments.

[0034] S3.2: Acquire the potential of the background field based on the dynamically adjustable electrode array; Before formal geological exploration, background field measurements are conducted to determine the absence of geological targets. When determining the background field, a completed tunnel with the same stratum and lithology as the tunnel face to be explored, or a rear area at a distance of ≥3 times the tunnel diameter / width from the tunnel face, is selected as a reference area. Based on the previous geological survey data and the surrounding rock conditions of the exposed tunnels, it is confirmed that there are no known geological targets in the reference area, and that the surrounding rock lithology and structure are relatively uniform.

[0035] The specific data collection process is as follows: Maintaining the same electrode layout parameters as in the actual detection, and keeping the power supply current intensity I consistent, the potential difference of each measuring electrode pair is collected. , which is the electric potential of the background field, serves as a blank reference for the distortion of the entire space.

[0036] S3.3: Based on the aforementioned three-dimensional full-space theoretical model, input actual engineering parameters and calculate the electric potential of the theoretical normal field; The actual engineering parameters include: tunnel cross-sectional dimensions, resistivity parameters of surrounding rock and tunnel wall, and power supply system parameters; the power supply system parameters include the electrode spacing AB of the power supply electrodes, the electrode spacing MN of the measuring electrodes, and the power supply current I, etc.

[0037] S3.4: Optimize the hyperparameters in the dynamic boundary weight correction factor based on the relative deviation between the theoretical normal field potential and the background field potential; To ensure that the dynamic boundary weight correction factor is applicable to the current tunnel environment, this step uses a measured-simulation iterative matching algorithm to optimize the hyperparameters of the dynamic boundary weight correction factor; wherein, the hyperparameters are the resistivity ratio of the tunnel wall to the surrounding rock, the related parameter α, and the tunnel geometric asymmetry correction coefficient λ.

[0038] Specifically, S3.4 is as follows: Set the initial values ​​for the hyperparameters; Generally, the resistivity of the tunnel wall is about 55% of that of the surrounding rock. The resistivity ratio of the tunnel wall to the surrounding rock is set to 1.8, and the tunnel geometric asymmetry correction coefficient λ is set to 0.5 to control the influence of angle on smoothness.

[0039] Calculate the relative deviation between the theoretical normal field potential and the background field potential; Wherein, the relative deviation The calculation formula is: ; In the formula, Let be the electric potential of the background field. This represents the electric potential of a theoretical normal field.

[0040] A genetic algorithm is used to find the hyperparameter combination that minimizes the relative deviation, which is then used as the specific hyperparameter for the current roadway environment. Specifically, the process involves: initializing the population, which is defined by a combination of hyperparameters; calculating the fitness of each set of hyperparameters; performing selection, crossover, and mutation operations to generate a new generation of population; and repeating the fitness calculation and crossover / mutation operations until the convergence condition is met.

[0041] In this embodiment, iteration stops when any of the following conditions are met; that is, the convergence condition is: Root mean square relative error (RMSE) < 5%; number of iterations > 50.

[0042] The final optimal parameter combination is the exclusive hyperparameter of the dynamic boundary weight correction factor for the current roadway environment, ensuring the accuracy of subsequent data correction.

[0043] S4; The potential of the test face is corrected according to the dynamic boundary weight correction factor to obtain the corrected potential of the test face. After considering the distortion caused by the tunnel geometry, the corrected potential at the observation point of the working face to be measured is... for: ; In the formula, The corrected potential at the observation point of the tunnel face to be measured is used for subsequent inversion calculations. The dynamic boundary weight correction factor for the observation points on the tunnel face to be measured. The measured potential is the actual potential at the observation point of the tunnel face to be measured.

[0044] S5: Calculate the apparent resistivity of each measuring point based on the corrected potential of the working face to be measured. Apparent resistivity at each measuring point The calculation formula is: ; In the formula, K is the device coefficient and I is the power supply current.

[0045] S6: Establish a three-dimensional full-space theoretical model, and perform an inversion operation on the apparent resistivity of each measuring point based on the three-dimensional full-space theoretical model to reconstruct the resistivity distribution image of the rock strata in front of the tunnel face to be measured, so as to realize the advanced detection of underground engineering. To establish a computable three-dimensional full-space theoretical model, the following reasonable assumptions are introduced: Homogeneity of the medium: The resistivity of the surrounding rock and the tunnel wall is uniform; Isotropic: The electrical properties of a dielectric are the same in all directions, ignoring the anisotropy caused by microstructures such as cracks; Steady-state DC field: Without considering electromagnetic induction and polarization effects, the current field reaches a steady-state distribution; Infinite boundary: The computational domain of the three-dimensional full-space theoretical model is large enough that the boundary conditions have a negligible impact on the target region; Geometric regularity: The tunnel cross-section is simplified to a standard geometric shape, which facilitates numerical discretization calculation.

[0046] Based on the above assumptions, the three-dimensional full-space theoretical model is described by basic governing equations, boundary conditions, and medium interface conditions. In the entire space medium, the steady-state DC electric field satisfies the Laplace equation; therefore, the fundamental governing equation is expressed as: ; In the formula, Here, σ is the gradient operator, and σ is the dielectric conductivity. The full-space position vector of the observation point The potential at the electrode is given by I, the DC current intensity between the supply electrodes is given by δ, and δ is the Dirac delta function, which describes the concentrated effect of the current source at the electrode location. Let A be the spatial position vector of the power supply electrode A. The spatial position vector of the power supply electrode B; The boundary conditions include infinity boundary conditions and power supply electrode boundary conditions; The boundary condition at infinity is as follows: at the boundary of the computational domain, i.e., at a sufficiently far distance from the roadway, the potential tends to zero, and the expression is: .

[0047] The boundary conditions for the power supply electrodes are as follows: at the locations of power supply electrodes A and B, the potentials are set to fixed values, and the expressions satisfying these conditions are: ; In the formula, and Let A and B be the potentials of the power supply electrodes, respectively. It is a fixed voltage value.

[0048] The medium interface condition is as follows: at the interface between the tunnel wall and the surrounding rock, the following electric field continuity condition is satisfied: ; In the formula, the superscripts (1) and (2) represent the two media, the tunnel wall and the surrounding rock, respectively. and These represent the electric potentials at the interface between the tunnel wall and the surrounding rock, respectively. and The electrical conductivity of the tunnel wall and the surrounding rock are two different media. This is the normal derivative of the interface.

[0049] A three-dimensional inversion algorithm using the least squares method combined with regularization constraints is employed to invert apparent resistivity data from multiple measurement points and reconstruct the resistivity distribution image of the rock strata ahead of the tunnel face.

[0050] In the resistivity distribution image of the rock strata in front of the tunnel face, the focus is on the area where the apparent resistivity is significantly lower than that of the surrounding rock. Based on the area where the apparent resistivity is significantly lower than that of the surrounding rock, combined with geological data, the possibility and degree of water inflow are further judged.

[0051] Example 2: The present invention also provides a geophysical advanced detection system for underground engineering. The system employs a geophysical advanced detection method for underground engineering as described in Example 1. The system includes: The acquisition module is used to deploy a dynamically adjustable electrode array on the working face to be tested, construct a stable DC electric field in the whole space, and acquire the potential of the working face to be tested based on the dynamically adjustable electrode array. The module for constructing and solving dynamic boundary weight correction factors is used to construct dynamic boundary weight correction factors based on equivalent medium theory and to optimize the hyperparameters of the dynamic boundary weight correction factors. The correction module is used to correct the potential of the test face according to the dynamic boundary weight correction factor to obtain the corrected potential of the test face. Apparent resistivity calculation module, used to calculate the apparent resistivity of each measuring point based on the corrected potential of the working face to be measured; The inversion and detection module is used to establish a three-dimensional full-space theoretical model and perform an inversion operation on the apparent resistivity of each measuring point based on the three-dimensional full-space theoretical model to reconstruct the resistivity distribution image of the rock strata in front of the tunnel face to be measured, thereby realizing advanced detection of underground engineering.

[0052] Example 3: The present invention also provides an electronic device, including one or more processors and a memory.

[0053] A processor can be a central processing unit (CPU) or other form of processing unit with data processing and / or instruction execution capabilities, and can control other components in an electronic device to perform desired functions.

[0054] The memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and a processor may execute the program instructions to implement a geophysical advanced detection method for underground engineering as described above in any embodiment of this application, and / or other desired functions. Various contents such as initial extrinsic parameters and thresholds may also be stored in the computer-readable storage medium.

[0055] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A geophysical advanced detection method for underground engineering, characterized in that, The method includes the following steps: S1: A dynamically adjustable electrode array is deployed on the face of the tunnel to be tested; S2: Construct a stable DC electric field in the entire space, and collect the potential of the working face to be measured based on the dynamically adjustable electrode array; S3: Construct a dynamic boundary weight correction factor based on the equivalent medium theory, and perform optimization calculations on the hyperparameters of the dynamic boundary weight correction factor; Specifically, S3 is: S3.1: Construct dynamic boundary weight correction factors; Its expression is: ; In the formula, Let r be the three-dimensional full-space distortion correction factor for the observation point, and r be the full-space position vector of the observation point. θ is the equivalent radius of the tunnel, θ is the angle between the observation point position vector and the tunnel axis, α is the resistivity ratio of the tunnel wall and the surrounding rock, and λ is the correction coefficient for the geometric asymmetry of the tunnel. S3.2: Acquire the potential of the background field based on the dynamically adjustable electrode array; S3.3: Based on the three-dimensional full-space theoretical model, input actual engineering parameters to calculate the electric potential of the theoretical normal field; S3.4: Optimize the hyperparameters in the dynamic boundary weight correction factor based on the relative deviation between the theoretical normal field potential and the background field potential; S4; The potential of the test face is corrected according to the dynamic boundary weight correction factor to obtain the corrected potential of the test face. S5: Calculate the apparent resistivity of each measuring point based on the corrected potential of the working face to be measured. S6: Establish a three-dimensional full-space theoretical model, and perform an inversion operation on the apparent resistivity of each measuring point based on the three-dimensional full-space theoretical model to reconstruct the resistivity distribution image of the rock strata in front of the tunnel face to be measured, thereby realizing advanced detection of underground engineering.

2. The geophysical advanced detection method for underground engineering according to claim 1, characterized in that, In section S3.2, before the formal geological exploration, a background field measurement is conducted where there are no geological targets. The specific data acquisition process is as follows: Maintaining the same electrode layout parameters as in the actual detection, and keeping the power supply current intensity I consistent, the potential difference of each measuring electrode pair is collected. , which is the electric potential of the background field, serves as a blank reference for the distortion of the entire space.

3. The geophysical advanced detection method for underground engineering according to claim 1, characterized in that, In S3.3, the actual engineering parameters include: tunnel cross-sectional dimensions, resistivity parameters of surrounding rock and tunnel wall, and power supply system parameters; the power supply system parameters include the power supply electrode distance AB, the measuring electrode distance MN, and the power supply current I.

4. The geophysical advanced detection method for underground engineering according to claim 1, characterized in that, Specifically, S3.4 is as follows: Set initial values ​​for the hyperparameters; calculate the relative deviation between the theoretical normal field potential and the background field potential; use a genetic algorithm to find the hyperparameter combination that minimizes the relative deviation, which is then used as the exclusive hyperparameter for the current tunnel environment.

5. A geophysical advanced detection method for underground engineering according to claim 4, characterized in that, The genetic algorithm is used to find the hyperparameter combination that minimizes the relative deviation, which is used as the exclusive hyperparameter for the current roadway environment. Specifically, the population is initialized, which is the value of the set hyperparameter combination. The fitness of each hyperparameter combination is calculated. Selection, crossover, and mutation operations are performed to generate a new generation of population. The fitness calculation and crossover and mutation operations are repeated until the convergence condition is met. The convergence criteria are: root mean square relative error (RMSE) < 5%; number of iterations > 50.

6. The geophysical advanced detection method for underground engineering according to claim 1, characterized in that, In S4, the corrected potential at the observation point of the tunnel face to be measured. for: ； In the formula, The corrected potential at the observation point of the tunnel face to be measured is used for subsequent inversion calculations. The dynamic boundary weight correction factor for the observation points on the tunnel face to be measured. The measured potential is the actual potential at the observation point of the tunnel face to be tested.

7. A geophysical advanced detection method for underground engineering according to claim 6, characterized in that, In S5, the apparent resistivity of each measuring point The calculation formula is: ； In the formula, K is the device coefficient and I is the power supply current.

8. A geophysical advanced detection method for underground engineering according to claim 1, characterized in that, In S6, the three-dimensional full-space theoretical model is described by basic governing equations, boundary conditions, and medium interface conditions. The basic governing equation is expressed as follows: ； In the formula, ∇ is the gradient operator, and σ is the dielectric conductivity. The full-space position vector of the observation point The potential at the point is given by I, the current intensity at the supply electrode is given by δ, and the Dirac delta function is given by δ. Let A be the spatial position vector of the power supply electrode A. The spatial position vector of the power supply electrode B; The boundary conditions include infinity boundary conditions and power supply electrode boundary conditions; The expression for the boundary condition at infinity is: In the formula, The full-space position vector of the observation point The potential at the point; The boundary conditions for the power supply electrodes are as follows: at the locations of power supply electrodes A and B, the potentials are set to fixed values, and the expressions satisfying these conditions are: ； In the formula, and Let A and B be the potentials of the power supply electrodes, respectively. For a fixed voltage value; The medium interface condition is as follows: at the interface between the tunnel wall and the surrounding rock, the following electric field continuity condition is satisfied: ； In the formula, the superscripts (1) and (2) represent the two media, the tunnel wall and the surrounding rock, respectively. and These represent the electric potentials at the interface between the tunnel wall and the surrounding rock, respectively. and The electrical conductivity of the tunnel wall and the surrounding rock are two different media. This is the normal derivative of the interface.

9. A geophysical advanced detection method for underground engineering according to claim 8, characterized in that, S6 further includes: using a three-dimensional inversion algorithm with least squares method and regularization constraints to invert the apparent resistivity data of multiple measurement points and reconstruct the resistivity distribution image of the rock strata in front of the tunnel face.

10. A geophysical advanced detection system for underground engineering, characterized in that, The system employs a geophysical advanced detection method for underground engineering as described in any one of claims 1-9, and the system comprises: The acquisition module is used to deploy a dynamically adjustable electrode array on the working face to be tested, construct a stable DC electric field in the whole space, and acquire the potential of the working face to be tested based on the dynamically adjustable electrode array. The module for constructing and solving dynamic boundary weight correction factors is used to construct dynamic boundary weight correction factors based on equivalent medium theory and to optimize the hyperparameters of the dynamic boundary weight correction factors. The correction module is used to correct the potential of the test face according to the dynamic boundary weight correction factor to obtain the corrected potential of the test face. Apparent resistivity calculation module, used to calculate the apparent resistivity of each measuring point based on the corrected potential of the working face to be measured; The inversion and detection module is used to establish a three-dimensional full-space theoretical model and perform an inversion operation on the apparent resistivity of each measuring point based on the three-dimensional full-space theoretical model to reconstruct the resistivity distribution image of the rock strata in front of the tunnel face to be measured, thereby realizing advanced detection of underground engineering.

Images