Construction method of hole-tunnel induced polarization water exploration structure and inversion water exploration method

Abstract

The application provides a hole-tunnel excited polarization water exploration structure construction method and an inversion water exploration method, wherein the construction method comprises the following steps: a plurality of forward drill holes are arranged at intervals in the outer edge area of a tunnel face; a hole-in exploration cable is arranged in each forward drill hole, each hole-in exploration cable has a plurality of exploration electrodes, and each exploration electrode in each hole-in exploration cable is respectively configured as a power supply electrode A and a power supply electrode B; a plurality of hole-out exploration electrodes are arranged in an array on the area outside the area occupied by the forward drill holes of the tunnel face, each hole-out exploration electrode is configured as a measurement electrode M and a measurement electrode N; each power supply electrode forms a power supply closed loop, and each measurement electrode forms a measurement closed loop. Through hole excitation and face reception, three-dimensional effective information of the front water-conducting structure can be obtained, and the system interference caused by the polarization of the electrode and the coupling of the multi-core cable is eliminated.

Description

Technical Field

[0001] This invention belongs to the field of geophysical exploration technology, specifically relating to a construction method and inversion method for a borehole-tunnel induced polarization water exploration structure. Background Technology

[0002] Currently, my country is witnessing a surge in deep-buried tunnels in challenging mountainous areas and tunnels spanning rivers and seas, characterized by their long lengths and great depths. Various water-bearing geological features are more concealed and pose a greater risk of disaster, demanding higher resolution and making targets more difficult to detect. These features also make tunnel construction highly susceptible to sudden water inrushes, posing significant challenges to the accuracy of tunnel advance detection. The induced polarization method, an electrical detection method, utilizes the induced polarization effect of water-bearing geological bodies for detection. It is sensitive to water bodies, and several induced polarization-based tunnel advance water detection methods, such as Beam and TIP, have been applied in the advance prediction of water-bearing structures in tunnels, both domestically and internationally. The TIP method, in particular, can achieve the location and quantitative prediction of large water-bearing structures.

[0003] However, the above methods utilize tunnel sidewalls and the tunnel face for detection.

[0004] (1) Due to the confined environment of the tunnel, the detection resolution is usually on the order of meters, making it difficult to achieve fine detection of small water-bearing structures. Secondly, the excitation polarization field generated by the above method is distributed near the tunnel, resulting in poor response to water-bearing structures at a distance. Consequently, the detection distance of the above method is relatively short and cannot meet the needs of detecting water-bearing structures at a distance.

[0005] (2) Meanwhile, in terms of induced polarization data processing and imaging, traditional polarizability inversion is based on resistivity inversion results. The imaging process introduces many ambiguities from resistivity inversion, making it difficult for polarizability inversion to converge. Furthermore, polarizability itself has natural boundary conditions with values ​​within the (0,1] interval. Traditional inversion methods often ignore this condition, resulting in polarizability values ​​exceeding the limit and making it impossible to obtain reliable inversion results. At the same time, induced polarization is a potential field detection method, which is itself affected by volume effects and makes it difficult to perceive the boundary information of anomalies. This leads to the inability to perform fine inversion imaging of anomaly boundaries, resulting in the inability to perform accurate geological interpretation.

[0006] In addition, induced polarization detection requires high data accuracy. Therefore, it is necessary to focus on the initial observation data errors caused by electrode polarization and multi-core cable coupling. The former is caused by the continuous power supply and measurement of the metal electrode, while the latter is caused by the simultaneous power supply and measurement of multiple electrodes in a single multi-core cable. Summary of the Invention

[0007] Therefore, the technical problem to be solved by the present invention is to provide a construction method and an inversion method for a borehole-tunnel induced polarization water exploration structure, which improves the detection range of the induced polarization field for the target. By induced polarization in the borehole and received at the tunnel face, three-dimensional effective information of the water-conducting structure in front can be obtained. This avoids the problems of continuous power supply and measurement of a single metal electrode and the simultaneous power supply and measurement of multiple electrodes on the same cable. It fundamentally and naturally eliminates the system interference of electrode polarization and multi-core cable coupling, and is suitable for induced polarization fine exploration.

[0008] This invention provides a construction method for a borehole-tunnel induced polarization water exploration structure, comprising the following steps:

[0009] Multiple forward boreholes are spaced out at intervals along the outer edge of the tunnel face;

[0010] An in-hole detection cable is provided in each of the forward boreholes. Each in-hole detection cable has multiple detection electrodes, and each detection electrode in each in-hole detection cable is configured as a power supply electrode A and a power supply electrode B, respectively.

[0011] Multiple external detection electrodes are arrayed in the area outside the area occupied by each of the forward boreholes at the tunnel face, and each of the external detection electrodes is configured as a measuring electrode M and a measuring electrode N.

[0012] The power supply electrode A and power supply electrode B form a closed power supply circuit, and the measuring electrode M and measuring electrode N form a closed measurement circuit.

[0013] This invention also provides an inversion method for a borehole-tunnel induced polarization water exploration structure based on the above-mentioned construction method of the borehole-tunnel induced polarization water exploration structure, comprising:

[0014] The initial observation data acquisition step, used to obtain the induced polarization data of the borehole-tunnel induced polarization water exploration structure, adopts the following method:

[0015] Define any one of the plurality of forward boreholes as the first borehole, and select one of the probe electrodes in the borehole probe cable within the first borehole as power supply electrode A and the other as power supply electrode B, and supply power to make power supply electrode A and power supply electrode B form a power supply closed loop, wherein power supply electrode A is located between the tunnel face and power supply electrode B;

[0016] Select one of the multiple external probe electrodes as the measuring electrode M and the other as the measuring electrode N, and power supply so that the measuring electrode M and the measuring electrode N form a closed measurement circuit, thereby obtaining the initial observation data in this state;

[0017] Based on the aforementioned method, all the detection electrodes in the detection cables inside the first hole and the detection electrodes outside the hole are traversed and observed to obtain initial observation data based on the first hole;

[0018] Based on the observation method using the initial observation data from the first well, the corresponding initial observation data for each of the remaining wells are obtained sequentially.

[0019] In some embodiments, the inversion water exploration method further includes:

[0020] The initial observation data screening step is used to calculate the initial observation data based on each well obtained in the initial observation data acquisition step according to the following formula for calculating the geometric factor G, and to discard the corresponding initial observation data when G exceeds a preset value to form the probe observation data:

[0021]

[0022] Where AM, AN, BM to BN represent the distances between each power supply electrode and the measuring electrode, respectively; AM', AN', BM' to BN' represent the distances between each power supply electrode's virtual electrode point symmetrical about the tunnel face and the measuring electrode, respectively.

[0023] In some implementations...

[0024] The preset value is 3000.

[0025] In some embodiments, the inversion water exploration method further includes:

[0026] Based on the detection and observation data, establish corresponding initial resistivity and initial polarizability models, and perform conventional electrical inversion based on the initial resistivity models to obtain resistivity inversion models;

[0027] Establish a joint inversion objective function for tunnel resistivity polarizability clustering;

[0028] The minimum value of the joint inversion objective function for tunnel resistivity polarizability clustering is obtained by finding the minimum value of the joint inversion objective function for tunnel resistivity polarizability clustering, constructing the inversion equation based on the joint inversion objective function for tunnel resistivity polarizability clustering, updating the model parameter increment, and obtaining the final inversion result after iterative inversion.

[0029] In some implementations...

[0030] The objective function for the joint inversion of tunnel resistivity polarizability clustering is:

[0031] Φ=Φ σ +μΦ η +λ1Φ FCM +λ2Φ log Φσ =Φ dσ +β σ Φ mσ Φ η =Φ dη +β η Φ mη

[0032] Where, Φ σ Φ is the sum of the resistivity data terms and model terms obtained from the conventional inversion. η The sum of the polarizability data terms and model terms from the conventional inversion is given, μ is the weighting parameter for the two models of equilibrium resistivity and polarizability, and Φ is the polarizability. FCM For clustering terms, λ1 is a parameter that determines the weights of the applied clustering terms; Φ log β represents the boundary constraint term, λ2 represents the boundary constraint parameter; σ With β η Φ is the regularization parameter. dσ It is the resistivity data item, Φ mσ It is a resistivity model term, Φ dη It is the polarizability data item, Φ mη It is a polarizability model term.

[0033] In some implementations...

[0034]

[0035] Where, m σ Let m be the resistivity vector of the model. η v is the polarization vector of the model. k1 For the first cluster center (resistivity value) of the k-th class, v k2 The second cluster center (polarizability value) of the k-th class is added together according to the weight ξ; t represents the membership degree of the j-th unit relative to the k-th class; k The target cluster centers are provided by known prior rock physics information, and the vector ω = (ω1, ω2, ... ω k ) T In the example, each element parameter ω k The size of is determined by the reliability of the k-th prior cluster center; q is the fuzziness factor, which is a constant.

[0036] In some implementations...

[0037] u is the upper boundary of the constraint, M is the number of model parameters; and / or,

[0038] And / or,

[0039]

[0040] In some implementations...

[0041] Resistivity model increment Δm σ The inversion equation is:

[0042]

[0043] Among them, J σ It is the resistivity sensitivity matrix, β σ U is the damping factor (a constant, set empirically) of the resistivity model constraint terms, C is the number of clusters, and U is the number of clusters. k Let Δd be the membership matrix of the grid belonging to the k-th cluster. σ Let m be the data difference vector of resistivity, which is the difference between the measured data and the predicted data. σ Resistivity As a resistivity reference model, v k1 The resistivity cluster center of the k-th cluster; and / or,

[0044] The iterative update equation for the polarizability boundary constraint parameter λ2 is:

[0045]

[0046] Where, ζ - and ζ + These are two intermediate variables, and the final calculated ζ is also an intermediate coefficient used to update λ2, while γ is a constant.

[0047] In some implementations...

[0048] Polarizability model increment Δm η The inversion equation is:

[0049] In the formula, X = diag(m1, m2, ... m M ), Y = uI - X,

[0050] Among them, J η Let Δd be the sensitivity matrix for polarizability. η Let v be the difference vector between the observed and predicted polarizability data. k2 Let be the polarizability cluster center of the k-th cluster, and e be an M-dimensional column vector;

[0051]

[0052] Among them, W d Weighted matrix of data, For polarizability observation data, d η For polarizability simulation data, βη W is the damping factor for the constraint term of the polarizability model. m Let m be the smoothness constraint matrix. η Polarizability This is the reference model for polarizability, where u is the upper limit of polarizability.

[0053] The present invention provides a construction method and an inversion method for a borehole-tunnel induced polarization water exploration structure, which have the following beneficial effects:

[0054] By setting multiple forward boreholes on the tunnel face and placing detection electrodes in each borehole, the limitations of traditional detection methods, which are restricted by the limited space of the tunnel, are overcome. By arranging the detection elements (i.e., detection electrodes) in the forward boreholes of the tunnel, close-range detection is achieved, which effectively improves the response of the excitation polarization field to the detection target, increases the detection distance, and solves the previous problem of not being able to detect or clearly detect small water-conducting structures in front of the tunnel. In this invention, the borehole-tunnel excitation polarization is a four-electrode observation method. Using a four-electrode observation array, excitation is performed inside the borehole and received at the tunnel face, which can obtain three-dimensional effective information of water-conducting structures in front. By distributing the power supply electrode pairs and the measurement electrode pairs to the forward boreholes and the tunnel face, the problems of continuous power supply and measurement by a single metal electrode and the simultaneous power supply and measurement of multiple electrodes on the same cable are avoided. The system interference of electrode polarization and multi-core cable coupling is fundamentally eliminated, making it suitable for fine excitation polarization detection.

[0055] This invention establishes clustering constraints for the inversion process, performs cluster analysis on resistivity and polarizability, and incorporates regularization into the inversion process. This leverages the advantages of cluster analysis in boundary characterization, enabling a detailed characterization of the water body boundary ahead of the tunnel. A joint inversion framework for resistivity and polarizability is established, achieving simultaneous inversion of resistivity and polarizability within a single inversion objective function. This reduces the dependence of polarizability inversion on resistivity inversion results and ensures convergence of the inversion process. Boundary constraints are added to the inversion of the three-dimensional polarizability model, integrating these constraints into the joint inversion framework. This ensures that the polarizability meets the limitations of its natural boundaries, thereby avoiding the defects of excessively small polarizability inversion model values ​​and data deviation from the actual model. Attached Figure Description

[0056] Figure 1 This is a side view schematic diagram of the hole-tunnel induced polarization water exploration structure according to an embodiment of the present invention;

[0057] Figure 2 This is a front view of the borehole-tunnel induced polarization water exploration structure according to an embodiment of the present invention (i.e., viewed from the tunnel face side);

[0058] Figure 3 A schematic diagram of the hole-tunnel-induced polarization quadrupole observation method;

[0059] Figure 4This is a schematic diagram illustrating the calculation of the geometric factor G;

[0060] Figure 5 This is a flowchart illustrating the joint inversion of tunnel-induced polarization resistivity and polarizability clustering in an embodiment of the present invention.

[0061] Figure 6 This is a schematic diagram of the geoelectric model used in the numerical simulation of this invention.

[0062] Figure 7 This is a comparison of the apparent resistivity imaging results of two methods in the embodiments of the present invention. In the figure, a) is the apparent polarizability imaging result of the clustering joint inversion method, and b) is the apparent polarizability imaging result of the traditional sequential inversion method.

[0063] Figure 8 The figures show a comparison of the apparent polarizability imaging results of two methods in the embodiments of the present invention. In the figures, a) is the apparent resistivity imaging result of the clustering joint inversion method, and b) is the apparent resistivity imaging result of the traditional sequential inversion method.

[0064] The reference numerals in the attached figures are as follows:

[0065] 1. Tunnel cavity; 2. Forward borehole; 5. Power supply electrode A; 6. Power supply electrode B; 7. Measuring electrode M; 8. Measuring electrode N. Detailed Implementation

[0066] See also Figures 1 to 8 As shown in the embodiment of the present invention, a construction method for a borehole-tunnel induced polarization water exploration structure is provided, comprising the following steps:

[0067] Multiple forward boreholes 2 are arranged at intervals in the outer edge area of ​​the tunnel face;

[0068] An in-hole detection cable is provided in each of the forward boreholes 2. Each in-hole detection cable has multiple detection electrodes, and each detection electrode in each in-hole detection cable is configured as a power supply electrode A5 and a power supply electrode B6, respectively.

[0069] Multiple external detection electrodes are arrayed in the area outside the area occupied by each of the forward boreholes 2 at the tunnel face. Each of the external detection electrodes is configured as a measuring electrode M7 and a measuring electrode N8.

[0070] The power supply electrode A5 and power supply electrode B6 form a closed power supply circuit, and the measuring electrode M7 and measuring electrode N8 form a closed measurement circuit.

[0071] In this technical solution, by setting multiple forward boreholes 2 on the tunnel face and setting detection electrodes in each forward borehole 2, the limitations of traditional detection methods, which are restricted by the limited space of the tunnel, are overcome. By arranging the detection elements (i.e., detection electrodes) in the forward boreholes 2 of the tunnel, close-range detection is achieved, which effectively improves the response of the excitation polarization field to the detection target, increases the detection distance, and solves the problem that small water-conducting structures in front of the tunnel cannot be detected or clearly detected in the past. In this invention, the borehole-tunnel excitation polarization is a four-electrode observation method. Using a four-electrode observation array, excitation is performed in the borehole and received at the tunnel face, which can obtain three-dimensional effective information of water-conducting structures in front. By distributing the power supply electrode pairs and the measurement electrode pairs to the forward boreholes and the tunnel face, the problems of continuous power supply and measurement by a single metal electrode and the simultaneous power supply and measurement of multiple electrodes on the same cable are avoided. The system interference of electrode polarization and multi-core cable coupling is fundamentally eliminated, which is suitable for fine excitation polarization detection.

[0072] According to an embodiment of the present invention, an inversion method for a borehole-tunnel induced polarization water exploration structure based on the above-described construction method for the borehole-tunnel induced polarization water exploration structure is also provided, comprising:

[0073] The initial observation data acquisition step, used to obtain the induced polarization data of the borehole-tunnel induced polarization water exploration structure, adopts the following method:

[0074] Define any one of the plurality of forward boreholes as the first borehole, and select one of the probe electrodes in the borehole probe cable within the first borehole as power supply electrode A and the other as power supply electrode B, and supply power to make power supply electrode A and power supply electrode B form a power supply closed loop, wherein power supply electrode A is located between the tunnel face and power supply electrode B;

[0075] Select one of the multiple external probe electrodes as the measuring electrode M and the other as the measuring electrode N, and power supply so that the measuring electrode M and the measuring electrode N form a closed measurement circuit, thereby obtaining the initial observation data in this state;

[0076] Based on the aforementioned method, all the detection electrodes in the detection cables inside the first hole and the detection electrodes outside the hole are traversed and observed to obtain initial observation data based on the first hole;

[0077] Based on the observation method using the initial observation data from the first well, the corresponding initial observation data for each of the remaining wells are obtained sequentially.

[0078] Specifically, by traversing and coordinating each detection electrode (power supply electrode pair) in each forward borehole 2 with each detection electrode (measurement electrode pair) on the tunnel face, it is possible to fully acquire three-dimensional effective information over a large distance in front of the tunnel face, achieve refined detection, and ensure the accuracy of the detection results.

[0079] In some embodiments, the inversion water exploration method further includes:

[0080] The initial observation data screening step is used to calculate the initial observation data based on each well obtained in the initial observation data acquisition step according to the following formula for calculating the geometric factor G, and to discard the corresponding initial observation data when G exceeds a preset value to form the probe observation data:

[0081]

[0082] Where AM, AN, BM to BN represent the distances between each power supply electrode and the measuring electrode; AM', AN', BM' to BN' represent the distances between the virtual electrode points of each power supply electrode symmetrical about the tunnel face and the measuring electrode. Figure 4 Taking the illustration in the figure as an example, only the relative distance information between the power supply electrode A and the measuring electrode M on the tunnel face is shown. A' in the figure is the symmetrical virtual electrode point of the power supply electrode A with respect to the tunnel face. This yields the corresponding distances AM and AM'. The relevant distances of other detection electrodes follow the same principle and will not be elaborated further. In a specific embodiment, the preset value is 3000. That is, observation methods with G greater than 3000 are eliminated, ultimately establishing an optimized hole-tunnel excitation polarization quadrupole effective observation mode. In this technical solution, data exceeding the preset value are eliminated through geometric factors, which can reduce a large amount of low signal-to-noise ratio or invalid data, ensuring the accuracy and validity of subsequent data.

[0083] In some embodiments, the inversion water exploration method further includes:

[0084] Based on the detection and observation data, establish corresponding initial resistivity and initial polarizability models, and perform conventional electrical inversion based on the initial resistivity models to obtain resistivity inversion models;

[0085] Establish a joint inversion objective function for tunnel resistivity polarizability clustering;

[0086] The minimum value of the joint inversion objective function for tunnel resistivity polarizability clustering is obtained by finding the minimum value of the joint inversion objective function for tunnel resistivity polarizability clustering, constructing the inversion equation based on the joint inversion objective function for tunnel resistivity polarizability clustering, updating the model parameter increment, and obtaining the final inversion result after iterative inversion.

[0087] Specifically, the objective function for the joint inversion of tunnel resistivity polarizability clustering is:

[0088] Φ=Φ σ +μΦ η +λ1Φ FCM +λ2Φlog Φ σ =Φ dσ +β σ Φ nσ Φ η =Φ dη +β η Φ mη

[0089] Where, Φ σ Φ is the sum of the resistivity data terms and model terms obtained from the conventional inversion. η Φ is the sum of the polarizability data and model terms in the conventional inversion. μ is the weighting parameter for the two models of resistivity and polarizability. In this invention, μ = 1, indicating that resistivity and polarizability have equal weight in the inversion data and model terms. FCM λ1 is a clustering term that influences the clustering behavior of the two sets of rock physics models for resistivity and polarizability obtained from the inversion. Φ is a parameter that determines the weight of the applied clustering term. log λ2 represents the boundary constraint term, and λ2 represents the boundary constraint parameter; these parameters constrain the three-dimensional polarizability model during the inversion process; β σ With β η Φ is a regularization parameter used to balance the weights of data terms and model terms in the resistivity and polarizability models, respectively. dσ It is the resistivity data item, Φ mσ It is a resistivity model term, Φ dη It is the polarizability data item, Φ mη It is a polarizability model term.

[0090] This technical solution establishes clustering constraints for the inversion process, performs cluster analysis on resistivity and polarizability, and incorporates regularization into the inversion process. This leverages the advantages of cluster analysis in boundary characterization, enabling a detailed characterization of the water body boundary ahead of the tunnel. A joint inversion framework for resistivity and polarizability is established, achieving simultaneous inversion of resistivity and polarizability within a single inversion objective function. This reduces the dependence of polarizability inversion on resistivity inversion results and ensures convergence of the inversion process. Boundary constraints are added to the inversion of the three-dimensional polarizability model, integrating these constraints into the joint inversion framework. This ensures that the polarizability meets the limitations of its natural boundaries, thus avoiding the defects of excessively small polarizability inversion model values ​​and data deviation from the actual model.

[0091] In the clustering items,

[0092]

[0093] Where, m σ Let m be the resistivity vector of the model. η v is the polarization vector of the model. k1 For the first cluster center (resistivity value) of the k-th class, v k2The second cluster center (polarizability value) of the kth class is added together with the weight ξ. According to statistical principles, when this term is minimized, all units of the model are considered to have been correctly classified. t represents the membership degree of the j-th unit relative to the k-th class; k The target cluster centers are provided by known prior rock physics information, and the vector ω = (ω1, ω2, ... ω k ) T In the example, each element parameter ω k The size of ω depends on the reliability of the k-th prior cluster center. If the reliability of the cluster center provided by the k-th prior information is low, then ω should be appropriately reduced. k The value of ; q is the fuzzification factor, which is a constant, usually 2.

[0094] In some implementations...

[0095] u is the upper boundary of the constraint. Considering that the polarizability of the water body in front of the tunnel is generally 0.3, in this invention, for the polarization detection of the tunnel borehole, u = 0.3 is set, and M is the number of model parameters. Specifically, in practical applications, forward and inverse models are performed by discretization methods such as finite element method. The model is divided into grids, and M is the number of all grids, which is also the total number of model parameters for each grid.

[0096] The initial value of λ2 is:

[0097]

[0098] Among them, W d Weighted matrix of data, For polarizability observation data, d η For polarizability simulation data (also known as prediction data), β η W is the damping factor (a constant, set empirically) for the constraint terms of the polarizability model. m Let m be the smoothness constraint matrix. η Polarizability This serves as a reference model for polarizability, where u is the upper limit of polarizability;

[0099] Membership degree and cluster center v k1 v k2 The inversion equations are as follows:

[0100] And / or,

[0101]

[0102] And / or,

[0103] Resistivity model increment Δmσ The inversion equation is:

[0104]

[0105] Among them, J σ It is the resistivity sensitivity matrix, β σ U is the damping factor (a constant, set empirically) of the resistivity model constraint terms, C is the number of clusters, and U is the number of clusters. k Let Δd be the membership matrix of the grid belonging to the k-th cluster. σ Let m be the data difference vector of resistivity, which is the difference between the measured data and the predicted data. σ Resistivity As a resistivity reference model, v k1 The resistivity cluster center of the k-th cluster; and / or,

[0106] In some implementations...

[0107] The iterative update equation for the polarizability boundary constraint parameter λ2 is:

[0108]

[0109] Where, ζ - and ζ + These are two intermediate variables, and the final calculated ζ is also an intermediate coefficient used to update λ2. γ is a constant, typically with a value of 0.925.

[0110] Polarizability model increment Δm η The inversion equation is:

[0111]

[0112] In the formula,

[0113] X = diag(m1, m2, ... m M ), Y = uI - X,

[0114] Among them, J η Let Δd be the sensitivity matrix for polarizability. η Let v be the difference vector between the observed and predicted polarizability data. k2 Let be the polarizability cluster center of the k-th cluster, and e be an M-dimensional column vector.

[0115] The technical solution of the present invention will be further described below with reference to a specific embodiment:

[0116] The water exploration method of the present invention can generally include two parts: data measurement and data processing and imaging. The main contents are as follows:

[0117] (1) Data measurement

[0118] In specific examples, such as Figure 1 and Figure 2 As shown, three 65m long forward boreholes 2 are drilled at the tunnel face. Horizontal cables (i.e., the borehole detection cables mentioned earlier, hereinafter the same) are laid inside the boreholes, with 32 electrodes (i.e., the aforementioned detection electrodes) spaced 2m apart on each cable. 25 electrode potentials (i.e., the borehole detection electrodes mentioned earlier) are laid at the tunnel face (i.e., the tunnel face mentioned earlier), with each electrode point connected to the cable. This operation is called "wire laying". It should be noted that the electrode spacing inside the boreholes can be adjusted according to actual detection needs. Using a smaller spacing, such as 1m, will correspondingly improve the detection resolution, and vice versa. The number of electrodes laid in each borehole and at the tunnel face can be adjusted according to actual needs, but generally should not exceed 32.

[0119] After the wiring is completed, an observation mode is generated according to the established observation method. The detection employs a hole-tunnel excited polarization quadrupole observation method, such as... Figure 3 As shown. Taking one data acquisition as an example, power supply electrodes A5 and B6 are arranged in the same borehole, with power supply electrode A5 positioned closer to the working face than power supply electrode B6. Measurement electrodes M7 and N8 are arranged at the working face, with their relative positions remaining unchanged. When generating the observation mode, all electrode potentials are traversed according to the above relative position rules. Considering that there are three boreholes at the working face, a total of 446,400 observation data points can be generated, forming a preliminary borehole-tunnel excitation polarization observation mode (the obtained data is also the initial observation and detection data). However, a large number of these data points have low signal-to-noise ratios or are invalid, so further filtering is required.

[0120] For each data point in the generated preliminary observation model, a geometric factor is calculated, using the formula (1) above. Figure 4 Then, the data in the observation mode were filtered according to the geometric factor, and the observation data with a geometric factor greater than 3000 were removed. The remaining observation data constituted the final hole-tunnel induced polarization observation mode.

[0121] After completing the wiring, observation mode generation, and data filtering, connect the cable to the host, power on the host and connect it to the main control computer to begin data acquisition. Once the data measurement is completed, data processing and imaging can begin.

[0122] (2) Data processing imaging

[0123] (1) Establish a geoelectric model of the low-resistivity body in front of the tunnel, and perform forward modeling on the geoelectric model to obtain two sets of observation data. (resistivity observation data) (Polarizability observation data) Establish two initial models corresponding to the two sets of observation data spaces. The initial resistivity model was inverted using conventional electrical resistivity methods. Conventional electrical resistivity forward and inverse methods are well-known in the industry and will not be elaborated here.

[0124] (2) Calculate the polarizability boundary constraint parameters The initial values ​​are set based on experience, with the initial cluster center vectors determined accordingly.

[0125] (3) Calculate the membership degree of each unit in the model. Calculate the cluster center vector for each cluster. In this invention, ξ = 1, meaning that resistivity and polarizability have equal weight in the membership calculation. In this invention, q = 2; parameter ω... k Based on the prior cluster center t k Depends on the reliability, let ω k =1.

[0126] (4) Calculate the sensitivity matrix of resistivity inversion in the nth iteration. Sensitivity matrix derived from polarizability inversion Sensitivity calculations are primarily based on the reciprocity criterion, yielding the increment of the resistivity model in the nth iteration. and polarizability model increment Update resistivity model In this invention, β is set σ =0.05, indicating that the weight of the model term in the resistivity inversion term is 0.05; λ1 =0.01, indicating that the weight of the applied clustering term is 0.01.

[0127] (5) Update the polarizability boundary constraint parameter λ2. The parameter γ is used to prevent the value of each element in the model from reaching the boundary precisely so that the logarithmic constraint iteration can continue. In this invention, γ is set to 0.925.

[0128] (6) Obtain the polarizability model increment And update the polarizability model This invention sets β η =0.1, indicating that the weight of the model term in the polarizability inversion term is 0.1; X = diag(m1,m2,...m M ), Y = uI - X.

[0129] As a typical implementation method:

[0130] (1) Establish a geoelectric model of the low-resistivity body in front of the tunnel, such as... Figure 5As shown in the model, the low-resistivity water-bearing body is located 5m in front of the tunnel, with a size of 2*3*2. Four forward boreholes were drilled in front of the tunnel, each with 20 electrodes, and the boreholes were spaced 1m apart. A four-hole joint observation method was used to collect two sets of observation data: apparent resistivity and apparent polarizability within a 20m range in front of the tunnel. The inversion region is 8*8*20m.

[0131] (2) Establish a three-dimensional coordinate system with the center of the tunnel face as the origin, and establish an initial resistivity model. and initial model of polarizability and the corresponding reference model Calculate the smoothness matrix W of the model m And data weighted matrix W d ;

[0132] (3) Initial resistivity model Perform several linear resistivity inversions and use the inversion results as the initial resistivity model for the next step of resistivity-polarizability clustering joint inversion.

[0133] (4) Initial model of polarizability Forward modeling yields initial prediction data for polarizability. Substitute these parameters into formula (1) to calculate the polarizability boundary constraint parameters. The initial value;

[0134] (5) Set the initial cluster center vector Let the first initial cluster center (resistivity value) of the k-th class be denoted as . This is the second initial cluster center (polarizability value) for the k-th class.

[0135] (6) Begin clustering joint inversion iteration and calculate the membership degree of each unit in the model. The cluster center vector of each cluster is calculated using formula (3).

[0136] (7) Calculate the sensitivity matrix of resistivity inversion in the nth iteration. Sensitivity matrix derived from polarizability inversion J σ The sensitivity matrix corresponding to the apparent resistivity data and resistivity model is calculated using traditional methods. ηi,j For matrix J η The element in the i-th row and j-th column, σ bj Let J be the background conductivity of the j-th unit. The potential prediction data is for the i-th unit when there is no excitation polarization effect. This represents the total potential prediction data under the excitation polarization effect of the i-th unit.

[0137] (8) Obtain the resistivity model increment in the nth iteration. Update resistivity model

[0138] (9) Update polarizability boundary constraint parameters The value of is obtained. The polarizability model increment is then obtained. And update the polarizability model

[0139] (10) Update the resistivity model and polarizability model By performing forward modeling separately, the resistivity prediction data for the nth iteration is obtained. and polarizability prediction data The residuals are calculated by subtracting the observed data of resistivity and polarizability from their corresponding predicted data.

[0140] (11) Calculate the fitting error RMS and RMS_IP of the nth iteration. If the convergence condition is not met, continue to cycle through steps (6) to (11). If the convergence condition is met, output the imaging result.

[0141] This embodiment compares the imaging results of the clustering joint inversion method with those of the traditional sequential inversion method. After performing 40 resistivity inversions on the initial model separately, the clustering joint inversion is performed 60 times. The inversion results are as follows. Figure 7 a) Figure 8 As shown in a), the traditional sequential inversion was followed by 60 separate resistivity and polarizability inversions, and the inversion results are as follows. Figure 7 b) Figure 8 As shown in b).

[0142] Figure 7 , Figure 8 The apparent polarizability and apparent resistivity of the y=0 profile from the two inversion results are output and imaged. By comparing the imaging results of the two methods, it can be seen that the induced polarization clustering joint inversion method of the present invention can accurately characterize the boundary of the water-bearing body in front of the tunnel.

[0143] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0144] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0145] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0146] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0147] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0148] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A water exploration inversion method for a borehole-tunnel induced polarization water exploration structure formed by a construction method of the borehole-tunnel induced polarization water exploration structure, characterized in that, The construction method of the borehole-tunnel induced polarization water exploration structure includes the following steps: Multiple forward boreholes are spaced out at intervals along the outer edge of the tunnel face; An in-hole detection cable is provided in each of the forward boreholes. Each in-hole detection cable has multiple detection electrodes, and each detection electrode in each in-hole detection cable is configured as a power supply electrode A and a power supply electrode B, respectively. Multiple external detection electrodes are arrayed in the area outside the area occupied by each of the forward boreholes at the tunnel face, and each of the external detection electrodes is configured as a measuring electrode M and a measuring electrode N. The power supply electrode A and power supply electrode B form a closed power supply circuit, and the measuring electrode M and measuring electrode N form a closed measurement circuit. The inversion water exploration method includes: The initial observation data acquisition step, used to obtain the induced polarization data of the borehole-tunnel induced polarization water exploration structure, adopts the following method: Define any one of the plurality of forward boreholes as the first borehole, and select one of the probe electrodes in the borehole probe cable within the first borehole as power supply electrode A and the other as power supply electrode B, and supply power to make power supply electrode A and power supply electrode B form a power supply closed loop, wherein power supply electrode A is located between the tunnel face and power supply electrode B; Select one of the multiple external probe electrodes as the measuring electrode M and the other as the measuring electrode N, and power supply so that the measuring electrode M and the measuring electrode N form a closed measurement circuit, thereby obtaining the initial observation data in this state; Based on the aforementioned method, all the detection electrodes in the detection cables inside the first hole and the detection electrodes outside the hole are observed to obtain initial observation data based on the first hole. Based on the observation method using the initial observation data from the first well, the corresponding initial observation data for each of the remaining wells are obtained sequentially. Also includes: The initial observation data screening step is used to calculate the initial observation data based on each well obtained in the initial observation data acquisition step according to the following formula for calculating the geometric factor G, and to discard the corresponding initial observation data when G exceeds a preset value to form the probe observation data: ， Where AM, AN, BM to BN represent the distances between each power supply electrode and the measuring electrode, respectively; AM', AN', BM' to BN' represent the distances between each power supply electrode and the virtual electrode point symmetrical about the tunnel face, respectively; Also includes: Based on the detection and observation data, establish corresponding initial resistivity and initial polarizability models, and perform conventional electrical inversion based on the initial resistivity models to obtain resistivity inversion models; Establish a joint inversion objective function for tunnel resistivity polarizability clustering; The minimum value of the joint inversion objective function for tunnel resistivity polarizability clustering is obtained based on the tunnel resistivity polarizability clustering joint inversion objective function. An inversion equation based on the tunnel resistivity polarizability clustering joint inversion objective function is constructed, the model parameter increment is updated, and the final inversion result is obtained after iterative inversion. The objective function for the joint inversion of tunnel resistivity polarizability clustering is: ， ， in, This is the sum of the resistivity data terms and model terms obtained from a conventional inversion. This is the sum of the polarizability data term and the model term obtained from the conventional inversion. To balance the weighting parameters of the two models for resistivity and polarizability, For clustering items, These are parameters that determine the weights of the clustering items applied. For boundary constraint terms, These are the boundary constraint parameters; and For regularization parameters; It is a resistivity data item. It is a resistivity model term. It is a polarizability data item. It is a polarizability model term.

2. The water inversion method according to claim 1, characterized in that, The preset value is 3000.

3. The water inversion method according to claim 1, characterized in that, ， in, Let be the resistivity vector of the model. This is the polarization vector of the model. The first cluster center of the k-th class. The second cluster center of the k-th class is determined by their weights. Add; This represents the membership degree of the j-th unit relative to the k-th class; The target cluster center provided by the known prior rock physics information, vector In, each element parameter The size depends on the reliability of the k-th prior cluster center; q is the fuzziness factor, which is a constant; M is the number of model parameters; C is the number of clusters; v k This is the cluster center vector.

4. The water inversion method according to claim 3, characterized in that, , The upper boundary of the constraint is M, where M is the number of model parameters; and / or, ; and / or, ， ， Where C is the number of clusters, v i1 v is the first cluster center of the i-th class. i2 It is the second cluster center of the i-th class.

5. The water inversion method according to claim 4, characterized in that, Incremental resistivity model The inversion equation is: ， in, It is the resistivity sensitivity matrix. Here, C is the damping factor of the resistivity model constraint term, and C is the number of clusters. Let be the membership matrix of the grid belonging to the k-th cluster. This is the data difference vector of resistivity, representing the difference between measured and predicted data. Resistivity As a resistivity reference model, is the resistivity cluster center of the k-th cluster; Wd is the data weighting matrix, and Wm is the smoothness constraint matrix; And / or, Polarizability boundary constraint parameters The iterative update equation is: ， ， in, and These are two intermediate variables, and the final calculation is... It is also an intermediate coefficient used for updating ; It is a constant.

6. The water inversion method according to claim 5, characterized in that, Incremental polarizability model The inversion equation is: In the formula, ， ， ， Where I is the identity matrix; This is the sensitivity matrix for polarizability. Let be the difference vector between the observed polarizability data and be the difference between the observed and predicted polarizability data. Let be the polarizability cluster center of the k-th cluster, and e be an M-dimensional column vector; ， in, Weighted matrix of data, For polarizability observation data, For polarizability simulation data, The damping factor is a constraint term in the polarizability model. The smooth constraint matrix is, Polarizability Here is the reference model for polarizability, and u is the upper boundary of the constraint.

Images