Resistivity tomography-based mine tailing pond leakage detection method and system

Abstract

The invention belongs to the technical field of mine safety and environment monitoring, and particularly relates to a mine tailing pond leakage detection method and system based on resistivity tomography. The method comprises the steps that electrodes are pre-buried in a tailing pond dam body according to the layering and zoning principle to form a distributed three-dimensional electrode array, penetrating electrodes are arranged in the area below an impermeable layer in the mode that a sealing sleeve penetrates through the impermeable layer, benchmark measurement is carried out on the full-array electrodes, dam body three-dimensional background resistivity and background polarizability distribution is obtained through joint inversion, and the dam body three-dimensional background resistivity and background polarizability distribution is obtained. And inputting the background resistivity and the background polarizability into a database as a differential monitoring reference. According to the method, the spatial position, the thickness and the resistivity of the impermeable layer are used as strong prior constraints to be fused into the joint inversion target function, and the penetrating electrode is arranged below the impermeable layer, so that the detection capability of the area below the impermeable layer is enhanced from a collection end and an inversion end, and the problem of deep signal distortion caused by a high-resistance impermeable layer in the prior art is solved.

Description

Technical Field

[0001] This invention belongs to the field of mine safety and environmental monitoring technology, specifically a method and system for detecting leakage in mine tailings dams based on resistivity tomography. Background Technology

[0002] Mine tailings dams are specialized storage facilities for mineral processing waste and waste liquids. The dam body is subjected to hydrostatic and hydrodynamic pressures for a long time, making it prone to developing seepage channels inside. Seepage water softens the dam material, which can cause piping and even dam failure. Heavy metal ions such as lead, cadmium, and arsenic in the waste liquid will enter the soil and aquifer, causing long-term environmental pollution. Therefore, early detection and accurate monitoring are of great engineering significance.

[0003] Currently, tailings dam leakage detection mainly relies on traditional methods such as manual inspection, borehole sampling, and piezometer observation. These methods have limitations, including large blind spots, limited deployment, and the inability to perform three-dimensional analysis. While geophysical methods such as resistivity tomography have been applied to dam leakage detection with the development of non-destructive testing technology, their direct application to tailings dams still has significant limitations. The bottom of tailings dams is typically covered with high-resistivity geomembranes or other high-resistivity impermeable layers. The current from conventional resistivity methods cannot penetrate these layers, resulting in extremely weak signals from deep leakage channels below the impermeable layer and severely distorted inversion results. Existing methods do not specifically address this shielding effect. Furthermore, current electrical resistivity methods rely solely on resistivity, while tailings dam leakage water is rich in metal ions and sulfides, generating significant induced polarization under alternating electric fields. Although clay interlayers also exhibit low resistivity characteristics, they display drastically different polarizability responses. Resistivity alone cannot effectively distinguish between water-bearing leakage areas and low-resistivity geological anomalies, posing a significant risk of false interpretations. Summary of the Invention

[0004] (1) Technical problems to be solved The purpose of this invention is to provide a method and system for detecting leakage in mine tailings dams based on resistivity tomography, in order to solve the problem that relying solely on resistivity as a single parameter cannot effectively distinguish between water-bearing leakage areas and low-resistivity geological anomalies, leading to a large risk of false interpretations.

[0005] (2) Technical solution To achieve the above objectives, on the one hand, the present invention provides a method for detecting leakage in mine tailings dams based on resistivity tomography, comprising: Electrodes were pre-embedded in the tailings dam body according to the principle of layering and zoning to form a distributed three-dimensional electrode array. Penetrating electrodes were deployed through sealed sleeves to penetrate the seepage barrier layer in the area below the seepage barrier layer. Baseline measurements were performed on the entire array of electrodes, and the three-dimensional background resistivity of the dam body was obtained through joint inversion. and background polarization The distribution of background resistivity and background polarizability is recorded in the database as a differential monitoring benchmark.

[0006] A multi-frequency pseudo-random waveform current with a frequency range of 0.01Hz to 100Hz is input to the power supply electrode pairs via a multi-frequency current transmitter. The potential difference time series of each measurement electrode pair is synchronously acquired by a multi-channel receiver. The Fourier transform of the potential difference time series is performed to extract the complex impedance at each frequency, and the apparent resistivity of each measurement point is calculated based on the complex impedance. and apparent polarization This yields a multi-parameter observation dataset.

[0007] Based on the tailings dam design data, the spatial location, thickness, and resistivity range of the impermeable layer are determined. In the 3D inversion mesh, the grid cells containing the impermeable layer are assigned high initial resistivity and strong constraint weights are set. The logarithms of resistivity and polarizability are used as the joint model parameter vector. A joint inversion objective function containing prior constraints of the impermeable layer and induced polarization two-parameter fitting terms is constructed, and the three-dimensional resistivity of the dam body at the current moment is obtained by using an iterative algorithm. and polarizability distributed.

[0008] The resistivity and polarizability distributions obtained from the current inversion are compared with the background field to calculate the time-series differential resistivity. and temporal differential polarization The difference results are then subjected to a triple judgment using the amplitude criterion, spatial continuity criterion, and temporal persistence criterion. Areas that simultaneously meet all three criteria are identified as leakage anomaly zones.

[0009] The identified leakage anomaly area is reconstructed in three dimensions using the Kriging interpolation method to obtain the three-dimensional coordinate range and spatial morphology of the leakage channel. The diffusion range of the pollution plume is tracked based on the spatial distribution of the high polarization anomaly area. The leakage scale comprehensive assessment index is calculated by combining the decrease in differential resistivity, the increase in differential polarization, and the spatial range of the anomaly. The corresponding level of early warning is triggered based on the leakage scale comprehensive assessment index and the leakage expansion rate.

[0010] Furthermore, the step of pre-embedding electrodes within the tailings dam body according to a layered and zoned principle to form a distributed three-dimensional electrode array includes: Several monitoring layers are set along the height of the dam body. The burial depth of the electrodes in each layer is determined according to the design location of the dam body seepage line, the spatial location of the anti-seepage layer, and the historical leakage risk assessment. In each monitoring layer, electrodes are evenly distributed along the dam axis and perpendicular to the dam axis at a preset electrode spacing to form a two-dimensional grid. Electrodes are densified at the dam toe, the contact surface between the dam foundation and the dam body, and the anti-seepage layer damage risk area to obtain a three-dimensional electrode array network covering the entire cross section of the dam body.

[0011] Furthermore, the apparent resistivity of each measuring point is calculated based on the complex impedance. and apparent polarization include: According to the formula Calculate the apparent resistivity, where For device coefficients, The total potential difference, For the supply current; according to the formula Calculate the apparent polarizability, where The potential difference in the secondary field after power failure. The potential difference is the first field potential difference; the apparent resistivity and apparent polarizability at each frequency are combined to obtain a multi-parameter observation dataset.

[0012] Furthermore, the construction of the joint inversion objective function containing the prior constraint term of the impermeable layer and the induced polarization two-parameter fitting term includes: The initial resistivity of the grid cells containing the impermeable layer is set, and the prior variance of the impermeable layer region is set to a small value to form strong constraint weights. The joint inversion objective function is then constructed as follows: ; in and These are the observed apparent resistivity and apparent polarizability data vectors, respectively. and These are the forward operators for resistivity and polarizability, respectively. Here is the smoothness matrix. This is a priori model containing a high resistivity value of the impermeable layer. The prior weight matrix, A factor to balance the weighting of resistivity and polarizability data. The prior constraint strength factor, The regularization parameters are determined using the L-curve method; the Gauss-Newton method is used iteratively to solve for the parameters until the data fitting residuals are reduced to below a preset threshold, thus obtaining the three-dimensional resistivity of the dam body at the current moment. and polarizability distributed.

[0013] Furthermore, the calculation of time-series differential resistivity and temporal differential polarization include: According to the formula Calculate the time-series differential resistivity; according to the formula Calculate the temporal differential polarization; where and These are background resistivity and background polarizability, respectively. and These are the resistivity and polarizability obtained from the inversion at the current moment, respectively.

[0014] Furthermore, the triple discrimination of the difference results by sequentially applying the amplitude criterion, spatial continuity criterion, and temporal persistence criterion includes: Time-differential resistivity Temporal differential polarization rate below a preset negative threshold and within the same region Regions exceeding a preset positive threshold are marked as candidate regions satisfying the amplitude criterion; spatial continuity detection is performed on candidate regions satisfying the amplitude criterion, and regions with the smallest continuous volume are retained as regions satisfying the spatial continuity criterion, while isolated noise points that do not satisfy the minimum continuous volume are filtered out; temporal continuity detection is performed on regions satisfying the spatial continuity criterion, and regions that persist and show an expanding trend over several consecutive measurement cycles are identified as leakage anomaly zones, while candidate regions that appear only in a single measurement are excluded.

[0015] Furthermore, the triggering of corresponding level early warnings based on the comprehensive assessment index of leakage scale and leakage expansion rate includes: A three-tiered early warning system is established based on the comprehensive assessment index of leakage scale: when the abnormal scale is small and expands slowly, a blue Level 1 warning is triggered, monitoring frequency is increased and manual verification is carried out; when the abnormal scale reaches a certain level and shows a continuous expansion trend, a yellow Level 2 warning is triggered, a special investigation is initiated and an emergency plan is formulated; when the abnormal range is large and expands rapidly, posing a dam safety risk, a red Level 3 warning is triggered, and the emergency response procedure is initiated; all levels of early warning signals are pushed to the tailings dam safety management platform in real time to trigger the corresponding emergency response process.

[0016] Based on the same inventive concept, the present invention also provides a mine tailings dam leakage detection system based on resistivity tomography, including a pre-embedded distributed three-dimensional electrode array module, a multi-frequency data acquisition and transmission module, a resistivity-polarizability joint inversion module with prior constraints of the seepage prevention layer, a dynamic differential monitoring and multi-dimensional anomaly identification module, and a leakage three-dimensional location and graded early warning module connected in sequence. The pre-embedded distributed three-dimensional electrode array module includes corrosion-resistant electrodes and penetrating electrodes that are pre-embedded in the dam body according to the principle of layering and partitioning. The penetrating electrodes pass through the seepage prevention layer with sealed sleeves and are waterproof and sealed. The electrodes are connected to the dam top control box through multi-core cables.

[0017] The multi-frequency data acquisition and transmission module includes a multi-frequency current transmitter and a multi-channel receiver. The multi-frequency current transmitter inputs a multi-frequency pseudo-random waveform current with a frequency range of 0.01Hz to 100Hz to the power supply electrode. The multi-channel receiver synchronously acquires the multi-frequency complex potential difference response of each measurement electrode pair and transmits the data to the data processing center in real time.

[0018] The resistivity-polarizability joint inversion module with impermeable layer prior constraints is used to establish an initial geoelectric model based on the impermeable layer prior information. The joint inversion objective function, which includes strong impermeable layer prior constraints and induced polarization dual-parameter fitting terms, is used to perform three-dimensional joint inversion on the observation data to obtain the underground three-dimensional resistivity and polarizability distribution.

[0019] The dynamic differential monitoring and multidimensional anomaly identification module is used to calculate the time-series difference component between the current inversion result and the background field, and to make a triple judgment on the difference result by combining the amplitude criterion, spatial continuity criterion and temporal persistence criterion to identify leakage anomaly areas.

[0020] The 3D leakage location and graded early warning module is used to reconstruct the 3D space of the leakage anomaly area and track the pollution plume. Based on the comprehensive assessment index of the leakage scale, it triggers a level 3 early warning and pushes it to the safety management platform.

[0021] Furthermore, the corrosion-resistant electrode is made of titanium alloy or platinum-iridium alloy, and the multi-frequency current transmitter supports electrode combinations of Winner device, Schlumberger device and dipole-dipole device. The data acquisition is automatically executed according to a preset cycle under normal operating conditions, and automatically switches to high-frequency encrypted acquisition mode when the anomaly identification module triggers an early warning signal.

[0022] Furthermore, the three-dimensional leakage location and graded early warning module is also used to periodically and automatically generate a comprehensive detection report that includes a three-dimensional resistivity and polarizability distribution map, a three-dimensional rendering map of differential anomaly areas, the morphology and coordinates of leakage channels, the range of pollution plumes, and the early warning status.

[0023] (3) Beneficial effects Compared with existing technologies, this invention incorporates the spatial location, thickness, and resistivity of the impermeable layer as strong prior constraints into the joint inversion objective function, and deploys penetrating electrodes below the impermeable layer. This dual enhancement of the detection capability of the area below the impermeable layer from both the acquisition and inversion ends solves the problem of deep signal distortion caused by high-resistivity impermeable layers in existing technologies. By introducing polarizability parameters for resistivity-polarizability joint inversion, the fundamental difference in polarizability between the low-resistivity, high-polarity of the seepage fluid and the low-resistivity, low-polarity of the clay interlayer reduces the ambiguity of the inversion from a physical mechanism perspective. The pre-embedded corrosion-resistant electrode array enables long-term online monitoring without personnel entering the highly polluted area, and the dynamic background field difference mechanism and triple discrimination criteria achieve early identification and intelligent graded warning of leakage hazards, providing a complete technical solution for the safety management of mine tailings dams. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a block diagram of the system modules of the present invention; Figure 3 This is a schematic diagram of the pre-embedded distributed three-dimensional electrode array layout within the dam body according to an embodiment of the present invention; Figure 4 This is a flowchart illustrating the resistivity-polarizability joint inversion process with prior constraints on the impermeable layer in an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Throughout this specification, references to "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in connection with that embodiment or example is included in at least one embodiment of the invention. Therefore, the phrases "an embodiment," "an example," "an example," or "an example" appearing in various places throughout the specification do not necessarily refer to the same embodiment or example. Furthermore, specific features, structures, or characteristics can be combined in one or more embodiments or examples in any suitable combination and / or sub-combination. Moreover, those skilled in the art will understand that the illustrations provided herein are for illustrative purposes and are not necessarily drawn to scale. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0028] Example 1: This example uses a tailings dam in a lead-zinc mine as the application object. This tailings dam is an upstream tailings dam with a total height of 35m, a crest length of 280m, a trapezoidal cross-section (8m wide at the crest, 120m wide at the base), an inner slope ratio of 1:3, and an outer slope ratio of 1:2.5. A 1.5mm thick HDPE geomembrane is laid at the bottom of the dam as a seepage barrier, with a resistivity ranging from 10000Ω·m to 50000Ω·m. The tailings dam area stores a large amount of tailings slurry and waste liquid containing heavy metal ions such as lead, cadmium, and arsenic.

[0029] like Figure 1 As shown in the figure, the mine tailings dam leakage detection method based on resistivity tomography in this embodiment includes the following steps: Step S1: Electrode array layout like Figure 3 As shown, electrodes are pre-embedded in the tailings dam body according to the principle of layering and zoning to form a distributed three-dimensional electrode array. Specifically, four monitoring layers are set along the height of the dam body. The embedment depth of the electrodes in each layer is determined according to the design location of the dam body's phreatic line, the spatial location of the seepage prevention layer, and the historical leakage risk assessment. The specific parameters of each monitoring layer are as follows: The first monitoring layer is located 5m below the dam crest (elevation +95m), in the unsaturated region above the phreatic line, and is mainly used to monitor changes in the water content of the upper part of the dam body and the development of cracks on the dam crest. The second monitoring layer is located 15m below the dam crest (elevation +85m), near the design position of the phreatic line, and is the most sensitive area to changes in the seepage field. The third monitoring layer is located 25m below the dam crest (elevation +75m), in the saturated region below the phreatic line, and is used to monitor the development of deep seepage channels in the dam body. The fourth monitoring layer is located 33m below the dam crest (elevation +67m), about 2m directly above the anti-seepage layer, and is specifically used to monitor seepage anomalies in the area near the anti-seepage layer.

[0030] Electrodes are uniformly arranged along the dam axis (X-direction) and perpendicular to the dam axis (Y-direction) within each monitoring layer to form a two-dimensional grid at a preset electrode spacing. The electrode spacing for the first monitoring layer is set to 8m × 8m, with a total of 7 × 5 = 35 electrodes; the electrode spacing for the second monitoring layer is set to 6m × 6m, with a total of 8 × 6 = 48 electrodes; the electrode spacing for the third monitoring layer is set to 5m × 5m, with a total of 9 × 7 = 63 electrodes; and the electrode spacing for the fourth monitoring layer is set to 4m × 4m, with a total of 10 × 8 = 80 electrodes. The electrode spacing gradually decreases from top to bottom to improve the spatial resolution of the deep region.

[0031] Electrode densification was implemented in the dam toe area, the contact surface between the dam foundation and the dam body, and the historically damaged risk zone of the anti-seepage layer (the section from chainage K0+080 to K0+120 where seepage traces were observed in historical inspection records). The electrode spacing in the dam toe densification zone was reduced to 2m×2m, with a total of 36 additional electrodes added. All the aforementioned conventional electrodes are made of titanium alloy (Ti-6Al-4V), with an electrode diameter of 25mm and a length of 300mm, exhibiting excellent corrosion resistance and meeting the requirements for long-term use in the acidic environment of the tailings dam.

[0032] Penetrating electrodes are installed below the seepage barrier layer using sealed sleeves. In this embodiment, 12 penetrating electrodes are evenly distributed along the dam axis in the dam foundation below the seepage barrier layer. The penetrating electrodes are made of platinum-iridium alloy (Pt-10%Ir), with an electrode diameter of 20mm and a length of 250mm. The penetrating electrodes are penetrated through the HDPE seepage barrier layer using stainless steel sealed sleeves with an outer diameter of 50mm. A double-layer EPDM rubber sealing ring and polyurethane grouting are used for waterproofing and sealing between the sleeve and the seepage barrier layer to ensure the integrity of the seepage barrier layer is not compromised. The bottom of the penetrating electrodes is located 1.5m to 3m below the seepage barrier layer, embedded in the gravel layer of the dam foundation.

[0033] All electrodes are connected to the dam crest control box via polyethylene-insulated multi-core cables. Each layer of electrodes uses a 48-core cable, while the penetrating electrodes use a separate 16-core cable. The dam crest control box is equipped with an automatic electrode switching matrix and lightning protection devices. In this embodiment, a total of 226 conventional electrodes, 12 penetrating electrodes, and 36 densified electrodes at the dam toe are deployed, forming a distributed three-dimensional electrode array network covering the entire cross-section of the dam.

[0034] Step S2: Reference Measurement and Background Field Establishment Before the initial impoundment of the tailings dam or under normal operating conditions where there is no known leakage risk, baseline measurements were performed on all 274 electrodes of the array. Three electrode arrangements—dipole-dipole, Wenner, and Schlumberger—were used for full array scanning and data acquisition, yielding approximately 18,000 valid observations. The three-dimensional background resistivity of the dam body was obtained through joint inversion. and background polarization Distribution. Typical background resistivity values ​​are: approximately 5–20 Ω·m in saturated tailings sand areas, approximately 50–200 Ω·m in unsaturated tailings sand areas, approximately 10,000–50,000 Ω·m in HDPE geomembrane layers, and approximately 100–500 Ω·m in dam foundation gravel layers. Typical background polarizability values ​​are: approximately 3%–8% in tailings areas containing metal sulfides, approximately 0.5%–2% in clay interlayers, and approximately 0.1%–0.5% in clean gravel. Background resistivity and background polarizability are entered into a database as a benchmark for subsequent time-series differential monitoring.

[0035] Step S3: Multi-frequency data acquisition A multi-frequency pseudo-random waveform current with a frequency range of 0.01Hz to 100Hz is input to the power supply electrodes via a multi-frequency current transmitter. In this embodiment, five frequencies are selected: 0.01Hz, 0.1Hz, 1Hz, 10Hz, and 100Hz. An M-sequence pseudo-random encoded waveform is used, the effective value of the power supply current is 2A, and the maximum output power is 5kW. Within the 0.01Hz to 100Hz frequency range, the selection of frequency points should follow a logarithmic uniform distribution principle to ensure sufficient excitation of polarimeters of different scales. It is recommended to select no fewer than five frequency points. The lower limit of the frequency range, 0.01Hz, determines the detection capability for low-frequency induced polarization effects (related to larger polarimeters), while the upper limit, 100Hz, covers the relaxation frequency range of common metal sulfide minerals. In practical engineering, the optimal frequency combination within the above range can be selected based on the estimated size of the target anomaly and the on-site electromagnetic interference conditions.

[0036] The potential difference time series of each measurement electrode pair is synchronously acquired using a multi-channel receiver. This embodiment employs a 64-channel synchronous receiver with a sampling rate of 2400Hz, an input impedance greater than 100MΩ, and a resolution of 24 bits, capable of simultaneously acquiring 64 potential difference signals. The acquisition time window for each power supply-measurement combination is 120 seconds, encompassing multiple complete low-frequency current cycles.

[0037] A Fourier transform is performed on the potential difference time series to extract the complex impedance at each frequency. Specifically, a Discrete Fourier Transform (DFT) is performed on the 120-second time series signal, and amplitude and phase information are extracted at each frequency point of 0.01Hz, 0.1Hz, 1Hz, 10Hz, and 100Hz to obtain the complex potential difference at each frequency. .

[0038] Calculate the apparent resistivity at each measuring point based on the complex impedance. and apparent polarization Apparent resistivity is determined by the formula. Calculation, where This is the device coefficient, determined by the geometrical relationship between the power supply electrode and the measuring electrode; The total potential difference (using the DC component, i.e., the potential difference amplitude at a frequency of 0.01Hz); This is the effective value of the supply current. The apparent polarization is calculated using the formula... Calculation, where The secondary field decay potential difference after power failure (in time-domain induced polarization mode, it is the integral average value within the time window from 200ms to 2000ms after power failure). This represents the field potential difference. The apparent resistivity and apparent polarizability data from five frequency points are combined to obtain a multi-parameter observation dataset. A complete measurement cycle yields approximately 18000 × 5 = 90000 multi-frequency observation data points.

[0039] Step S4: Joint inversion with prior constraints of the impermeable layer like Figure 4 As shown, the spatial location, thickness, and resistivity range of the geomembrane are determined based on the tailings dam design data. In this embodiment, the HDPE geomembrane is located at an elevation of +65m, with a thickness of 1.5mm (included in the smallest grid cell with a thickness of 0.5m in the inversion grid), and a designed resistivity range of 10000Ω·m to 50000Ω·m.

[0040] A three-dimensional inversion mesh was established. The dam area was divided into 56 mesh cells in the X direction (dam axis direction), with a cell size of 5m; 24 mesh cells in the Y direction (perpendicular to the dam axis direction), with a cell size of 5m; and 35 mesh cells in the Z direction (depth direction), with a cell size of 1m. A total of 56 × 24 × 35 = 47040 mesh cells were created. The initial resistivity of 20000 Ω·m (logarithmic mean of the median resistivity range of the seepage barrier layer) was assigned to the mesh cell containing the seepage barrier layer (the 33rd layer in the Z direction, with a total of 56 × 24 = 1344 cells).

[0041] The logarithms of resistivity and polarizability are used as the parameter vector of the joint model. The parameter vector has a dimension of 47040 × 2 = 94080. The joint inversion objective function, including the prior constraint term of the impermeable layer and the induced polarization two-parameter fitting term, is constructed as follows: ; in and These are the observed apparent resistivity and apparent polarizability data vectors, respectively; and These are forward modeling operators for resistivity and polarizability, respectively. For resistivity forward modeling uses the finite element method to solve the three-dimensional Poisson equation, while polarizability forward modeling is based on the Cole-Cole model for dispersion calculation. It is a first-order difference smoothness matrix, applied between adjacent grid cells to ensure the spatial continuity of the inversion results; For the prior model containing the high resistivity value of the impermeable layer, the prior value of the impermeable layer mesh element is set to ln(20000)=9.9; Let be the prior weight matrix, be a diagonal matrix, and set the prior variance of the impermeable layer region as . (That is, the diagonal elements are 1 / 0.01 = 100) to form strong constraint weights, and the prior variance of the remaining regions is set to (Diagonal element is 1.0); To balance the weighting of resistivity and polarizability data, this embodiment takes... =0.5; As the prior constraint strength factor, this embodiment takes... =10; The regularization parameters, determined by the L-curve method, are automatically optimized in each iteration based on the trade-off curve between the data fitting term and the regularization term. The four parameters required by the Cole-Cole model (zero-frequency resistivity) Charging rate Relaxation time and frequency correlation coefficient The parameters can be obtained as follows: During the baseline measurement phase, representative tailings sand samples collected from the dam body (containing at least three typical media: saturated tailings sand, sulfide-containing tailings sand, and dam foundation gravel) are subjected to broadband induced polarization measurements under laboratory conditions. The frequency range is consistent with the field measurements (0.01Hz to 100Hz). Model parameters for each medium are obtained by fitting the measured dispersion curves to a Cole-Cole model. In this embodiment, the typical parameters for saturated tailings sand are... ≈0.15 ≈0.05s ≈0.5; Typical parameters for tailings areas containing metal sulfides are: ≈0.30、 ≈0.02s ≈0.6. During the inversion process, the above laboratory calibration values ​​are used as the initial model parameters, and the Cole-Cole parameters of each grid cell are updated through iterative inversion.

[0042] The Gauss-Newton method is used for iterative solution. Each iteration requires calculating the Jacobian sensitivity matrix J (90000×94080 dimensions), which is efficiently calculated using the adjoint state method; then the normal equations are solved. Get model update amount Update the model The convergence condition is set as follows: the root mean square (RMS) of the data fitting residuals falls below 1.0, or the rate of change of the residuals between two consecutive iterations is less than 1%. In this embodiment, convergence is typically achieved after 8–12 iterations, yielding the three-dimensional resistivity of the dam body at the current moment. and polarizability distributed.

[0043] Step S5: Timing Difference Calculation The resistivity and polarizability distributions obtained from the current inversion are compared with the background field to calculate the time-series differential resistivity. and temporal differential polarization .

[0044] The time-series differential resistivity is based on the formula Calculation. When leakage occurs in a certain area, the leaking water fills the pores, causing a decrease in the resistivity of that area. The value is negative and abnormal; the more severe the leakage, the greater the negative value.

[0045] The temporal differential polarization is based on the formula Calculations show that when leachate containing heavy metal ions enters the dam body, the metal ions and sulfide particles undergo interfacial polarization under an alternating electric field, increasing the polarizability. It exhibits a positive anomaly. While the clay interlayer also exhibits low resistivity, its polarizability does not change significantly due to leakage. Approaching zero or showing slight changes can help distinguish between genuine leakage and low-resistivity geological anomalies such as clay interlayers.

[0046] Step S6: Triple discrimination The difference results are then subjected to a triple judgment using the amplitude criterion, spatial continuity criterion, and temporal duration criterion, as detailed below: Amplitude criterion judgment, using time-series differential resistivity Temporal differential polarization rate below a preset negative threshold of -20% and within the same region Regions exceeding a preset positive threshold by 30% are marked as candidate regions satisfying the amplitude criterion. The basis for setting this dual-parameter joint threshold is that, for genuine metal ion-containing leakage channels, the resistivity decrease typically exceeds 20% (because the resistivity of leakage water is approximately 1–10 Ω·m, far lower than the 50–200 Ω·m of dry tailings sand), and the polarizability increase typically exceeds 30% (because the increased metal ion concentration significantly enhances the interfacial polarization effect). Resistivity changes caused by simple rainfall infiltration or temperature variations generally do not exceed ±15% and are not accompanied by significant polarizability changes, thus they can be effectively excluded.

[0047] Spatial continuity criteria are used to determine the spatial continuity of candidate regions that meet the amplitude criteria. A three-dimensional connected region labeling algorithm (26-neighbor connectivity criterion) is employed to calculate the volume of each connected region. Regions with a minimum continuous volume of not less than 8 m³ (i.e., containing at least 2 × 2 × 2 = 8 adjacent grid cells) are retained as regions that meet the spatial continuity criterion, while isolated noise points that do not meet the minimum continuous volume requirement are filtered out. This criterion can effectively eliminate scattered outliers caused by random noise.

[0048] The temporal continuity criterion is used to determine the leakage anomaly. Areas that meet the spatial continuity criterion are monitored for temporal continuity. Areas that persist for three or more consecutive measurement cycles (in this embodiment, the standard measurement cycle is once a week, i.e., more than three consecutive weeks) and whose area or volume shows an expanding trend (the abnormal volume growth rate is greater than zero between adjacent weeks) are identified as leakage anomaly zones. Candidate zones that appear only in one or two measurements and subsequently disappear are excluded; these are usually false anomalies caused by transient disturbances (such as transient seepage fluctuations within the dam or measurement errors).

[0049] Areas that simultaneously meet the above three criteria are identified as areas with abnormal leakage. The above amplitude discrimination threshold ( Reference value -20%, (Reference value + 30%) is the recommended initial value for the composition of seepage fluid in lead-zinc mine tailings dams in this embodiment. In practical engineering applications, the threshold can be calibrated on-site using the following method: In the initial stage of tailings dam construction or during the benchmark measurement phase, the standard deviations of differential resistivity and differential polarizability are statistically analyzed for N consecutive measurements (N ≥ 12 recommended) in known non-seepage areas within the dam body. and The amplitude threshold is set to three times the respective standard deviation (i.e., the 3σ criterion). Threshold = -3 , Threshold = +3 This is to adapt to different mineral types and strata background noise levels. For leachate with low metal ion concentrations (such as neutral mine wastewater), the noise level can be appropriately reduced. The threshold is set to +15% to +20%; for sites with high background noise, the two thresholds can be appropriately increased to reduce the false alarm rate.

[0050] Step S7: 3D Reconstruction and Hierarchical Early Warning The identified leakage anomaly areas were reconstructed in three dimensions using Kriging interpolation. Differential resistivity was used as the basis for the reconstruction. and differential polarization The spatial distribution data is used as input, in Three-dimensional spatial interpolation is performed within the isosurface range of <-20% to obtain the three-dimensional coordinate range (X start and end coordinates, Y start and end coordinates, Z start and end depth) and spatial morphology (strip-like, clum-like, or dendritic) of the leakage channel.

[0051] According to the high polarizability anomaly region ( The spatial distribution of the pollution plume (>+30%) can be used to track its diffusion range. As heavy metal ions in the leachate diffuse, they form a concentration gradient at the leading edge and sides of the leakage channel. The corresponding polarizability anomaly also shows a spatial distribution characteristic of gradually weakening from the center to the periphery. Based on this, the spatial range and diffusion direction of the pollution plume can be delineated.

[0052] Overall differential resistivity reduction The increase in differential polarization and the volume of the abnormal body space The Leakage Scale Index (LSI) is calculated using the following formula: in =0.4、 =0.3、 =0.3 is the weight factor for each item; =50% =100% =1000m³ is the normalized reference value for each component. The above weighting factors... =0.4、 =0.3、 =0.3 is determined based on pairwise comparisons of the relative importance of the three indicators using the Analytic Hierarchy Process (AHP). The decrease in resistivity reflects the water saturation level of the leakage channel and has the highest weight; the increase in polarizability reflects the pollutant concentration and is second; abnormal volume reflects the scale of leakage and is considered on par with the first two. In practical applications, each mine can adjust the weighting based on the distance to sensitive downstream targets (such as residential areas and water sources). The weighting should be adjusted when there are important protection targets downstream. Increase to above 0.4. Normalized reference value. =50% =100% =1000m³ is an empirical value for engineering, corresponding to the typical differential resistivity reduction, differential polarizability increase, and spatial volume of a medium-sized seepage channel. Each mine can adjust it according to the dam size and historical seepage case statistics.

[0053] Simultaneously, the leakage expansion rate VER (VolumeExpansionRate) is calculated, defined as the rate of change of abnormal volume between two adjacent measurement weeks: ,in The measurement period interval (week).

[0054] A comprehensive assessment index for leakage scale is calculated by combining the decrease in differential resistivity, the increase in differential polarizability, and the spatial range of the anomaly. This comprehensive assessment index is generated through a pre-defined weighted assessment model, the core idea of ​​which is to quantify the "intensity" and "scale" of the leakage anomaly. Specifically, it can be calculated based on the maximum decrease in differential resistivity, which characterizes the leakage intensity. The maximum increase in differential polarization rate, which characterizes the degree of pollution. and the spatial volume of the anomaly representing the extent of destruction. As input parameters, quantitative indices are obtained through normalization and weighted summation. The normalized baseline values ​​for each parameter can be preset and adjusted based on the tailings dam's design level and the safety factor of the dam structure. The weights between parameters can be calibrated using the Analytic Hierarchy Process (AHP) or based on historical leakage case data to reflect the differences in the contribution of different parameters to dam safety. Simultaneously, the rate of change of abnormal volume over time, i.e., the leakage propagation rate (VER), is combined to form the criteria for graded early warning.

[0055] A three-level early warning system is established based on the Leakage Scale Index (LSI) and the Leakage Expansion Rate (VER). The system pushes early warning signals at all levels to the tailings dam safety management platform in real time. The platform automatically triggers the corresponding emergency response procedures, including notifying relevant responsible persons of early warning information via SMS / WeChat, automatically generating leakage anomaly reports, and activating video monitoring linkage.

[0056] In an actual monitoring exercise of this embodiment, a low resistivity anomaly with ΔR = -35% to -42% was observed at chainage K0+100 of the dam body, in the area from +72m to +78m. In the same area, a high polarization anomaly with ΔP = +45% to +62% was also observed, with an anomaly volume of approximately 85m³. This anomaly persisted and gradually expanded over four consecutive measurement cycles. After triple discrimination, it was determined to be a seepage anomaly zone, with calculated LSI = 0.48 and VER = 12% / week, triggering a yellow level-two warning. Subsequent borehole sampling confirmed the presence of a hot-melt weld defect of approximately 0.3m in length in the seepage barrier layer at this location. The lead concentration in the seepage fluid was 2.3mg / L, completely consistent with the physical explanation of the high polarization anomaly.

[0057] Example 2, as Figure 2 As shown, based on the same inventive concept, this embodiment also provides a mine tailings dam leakage detection system based on resistivity tomography. This system uses the method described in Embodiment 1 and includes the following five functional modules: (1) Pre-embedded distributed three-dimensional electrode array module This module includes corrosion-resistant electrodes pre-embedded within the dam body according to a layered and zoned principle, and penetrating electrodes that penetrate the seepage barrier. The corrosion-resistant electrodes are made of titanium alloy (Ti-6Al-4V), with a diameter of 25mm and a length of 300mm. Their surface is anodized to enhance acid and alkali corrosion resistance, and their designed service life is no less than 30 years. The penetrating electrodes are made of platinum-iridium alloy (Pt-10%Ir), with a diameter of 20mm and a length of 250mm. The penetrating electrodes pass through the HDPE seepage barrier using a stainless steel sealing sleeve (outer diameter 50mm, wall thickness 3mm, SUS316L material). The contact surface between the sleeve and the seepage barrier is statically sealed with a double-layer EPDM rubber sealing ring. The annular gap between the sleeve and the electrode is filled and sealed with polyurethane grout to ensure a sealing pressure of at least 0.5MPa. All electrodes are connected to the dam crest control box via polyethylene insulated multi-core cables (cross-sectional area 1.5mm² / core, insulation resistance >100MΩ / km). The dam crest control box is equipped with an automatic electrode switching matrix (maximum 512 channels), lightning protection devices (nominal discharge current 40kA), and an uninterruptible power supply (UPS, backup time 8 hours). To verify the integrity of the penetrating electrode seal, the system automatically executes a self-check procedure for the seal status before each data acquisition: by measuring the DC resistance between the outer wall of the penetrating electrode sleeve and the dam body reference electrode, if the resistance value decreases by more than 20% compared to the reference value at the initial installation, it is determined that the seal may be deteriorating, triggering a seal check prompt; if the apparent resistivity at the penetrating electrode location shows a local abnormal increase (more than 50% above the background value), it should also be investigated whether it is a false anomaly caused by a local short circuit in the seepage prevention layer due to sleeve damage, so as to distinguish the seal failure signal from the real leakage signal.

[0058] (2) Multi-frequency data acquisition and transmission module This module includes a multi-frequency current transmitter and a multi-channel receiver. The multi-frequency current transmitter has an output frequency range of 0.01Hz to 100Hz, a maximum output current of 5A, and a maximum output power of 10kW. It supports M-sequence pseudo-random coded waveforms and conventional rectangular wave waveforms. The multi-frequency current transmitter supports three electrode combinations: Wenner, Schlumberger, and dipole-dipole. It can automatically select the optimal electrode combination based on the monitoring target (e.g., deep detection priority or shallow resolution priority). The multi-channel receiver is a 64-channel synchronous acquisition system with a sampling rate of 2400Hz, an input impedance greater than 100MΩ, a resolution of 24 bits (approximately 21 effective bits), and inter-channel crosstalk less than -120dB. Data acquisition is automatically performed at a preset cycle (once a week) under normal operating conditions, with a single complete acquisition taking approximately 6 hours. When the anomaly detection module triggers an early warning signal, it automatically switches to a high-frequency encrypted data acquisition mode, increasing the acquisition frequency to once a day or once every 12 hours. The acquired raw data is preprocessed (denoising, detrending, and quality factor calculation) by the industrial control computer in the control box, and then transmitted in real time to the data processing center via a 4G / 5G wireless communication network.

[0059] (3) Resistivity-polarizability joint inversion module with prior constraints of impermeable layer This module is used to establish an initial geoelectric model based on prior information about the impermeable layer. Specifically, the three-dimensional spatial coordinates, thickness (1.5mm HDPE membrane), and resistivity range (10000–50000 Ω·m) of the impermeable layer are extracted from the tailings dam design drawings and completion acceptance data. The grid cells containing the impermeable layer are calibrated in the three-dimensional inversion grid, assigned a high initial resistivity (20000 Ω·m), and the prior variance of this region is set to 0.01 to form a strong constraint weight. A joint inversion objective function containing strong prior constraints on the impermeable layer and induced polarization two-parameter fitting terms is used to perform a three-dimensional joint inversion on the observation data. The Gauss-Newton method is used for iterative solution to obtain the underground three-dimensional resistivity and polarizability distribution. This module is deployed on the GPU computing server in the data processing center, utilizing CUDA parallel computing to accelerate forward modeling and sensitivity matrix calculations. A single inversion calculation takes approximately 2–4 hours.

[0060] (4) Dynamic differential monitoring and multidimensional anomaly identification module This module is used to calculate the time-series difference component between the current inversion result and the background field, and then uses a triple discrimination based on the amplitude criterion, spatial continuity criterion, and temporal persistence criterion to identify leakage anomaly zones. The threshold for the amplitude criterion is set as follows: <-20% and The spatial continuity criterion requires that the connected volume of the anomalous region be no less than 8 m³; the temporal persistence criterion requires that the anomalous region persist for three or more consecutive measurement periods and show an expanding trend. This module runs automatically after each inversion calculation, and the calculation takes approximately 5 minutes.

[0061] (5) Leakage three-dimensional location and graded early warning module This module is used for 3D spatial reconstruction and plume tracking of leakage anomaly areas. Based on the LSI (Leakage Scale Index), it triggers three levels of early warning (blue - Level 1, yellow - Level 2, red - Level 3) and pushes the alerts to the safety management platform. This module also generates comprehensive monitoring reports periodically (once a month or automatically after each early warning). The reports include: 3D resistivity and polarizability distribution cloud maps, 3D renderings of differential anomaly areas (displaying the spatial morphology of the anomaly area using isosurface and volume rendering), the 3D coordinate range and geometric parameters of the leakage channels, the spatial range and diffusion direction of the plume, historical anomaly evolution trend curves, and the current early warning status. The reports are automatically archived in PDF format and pushed to the tailings dam safety management platform.

[0062] It should be noted that the specific methods by which each module performs operations in the system described in the above embodiments have been described in detail in Embodiment 1 of the method, and will not be repeated here.

[0063] It should be noted that the specific methods by which each module performs operations in the system described in the above embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0064] Finally, it should be noted that although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for detecting leakage in mine tailings dams based on resistivity tomography, characterized in that, The method includes the following steps: Electrodes are pre-embedded in the tailings dam body according to the principle of layering and zoning to form a distributed three-dimensional electrode array. Penetrating electrodes are deployed below the seepage barrier layer through sealed sleeves. A benchmark measurement is performed on the entire array of electrodes. Both the benchmark measurement and subsequent periodic monitoring measurements use the same acquisition method: a multi-frequency current transmitter inputs multi-frequency pseudo-random waveform current, and a multi-channel receiver synchronously acquires the potential difference time series. The three-dimensional background resistivity of the dam body is obtained through joint inversion. and background polarization The distribution of background resistivity and background polarizability is recorded in the database as a differential monitoring benchmark. A multi-frequency pseudo-random waveform current with a frequency range of 0.01Hz to 100Hz is input to the power supply electrode pairs via a multi-frequency current transmitter. The potential difference time series of each measurement electrode pair is synchronously acquired by a multi-channel receiver. The Fourier transform of the potential difference time series is performed to extract the complex impedance at each frequency, and the apparent resistivity of each measurement point is calculated based on the complex impedance. and apparent polarization This yields a multi-parameter observation dataset; Based on the tailings dam design data, the spatial location, thickness, and resistivity range of the impermeable layer are determined. In the 3D inversion mesh, the grid cells containing the impermeable layer are assigned high initial resistivity and strong constraint weights are set. The logarithms of resistivity and polarizability are used as the joint model parameter vector. A joint inversion objective function containing prior constraints of the impermeable layer and induced polarization two-parameter fitting terms is constructed, and the three-dimensional resistivity of the dam body at the current moment is obtained by using an iterative algorithm. and polarizability distributed; The resistivity and polarizability distributions obtained from the current inversion are compared with the background field to calculate the time-series differential resistivity. and temporal differential polarization The difference results are then subjected to a triple judgment using the amplitude criterion, spatial continuity criterion, and temporal persistence criterion. Areas that simultaneously meet all three criteria are identified as areas with abnormal leakage. The identified leakage anomaly area is reconstructed in three dimensions using the Kriging interpolation method to obtain the three-dimensional coordinate range and spatial morphology of the leakage channel. The diffusion range of the pollution plume is tracked based on the spatial distribution of the high polarization anomaly area. The leakage scale comprehensive assessment index is calculated by combining the decrease in differential resistivity, the increase in differential polarization, and the spatial range of the anomaly. The corresponding level of early warning is triggered based on the leakage scale comprehensive assessment index and the leakage expansion rate.

2. The method for detecting leakage in mine tailings dams based on resistivity tomography according to claim 1, characterized in that, The method of pre-embedding electrodes in the tailings dam body according to the principle of layering and zoning to form a distributed three-dimensional electrode array includes: Several monitoring layers are set along the height of the dam body. The burial depth of the electrodes in each layer is determined according to the design location of the dam body seepage line, the spatial location of the anti-seepage layer, and the historical leakage risk assessment. In each monitoring layer, electrodes are evenly distributed along the dam axis and perpendicular to the dam axis at a preset electrode spacing to form a two-dimensional grid. Electrodes are densified at the dam toe, the contact surface between the dam foundation and the dam body, and the anti-seepage layer damage risk area to obtain a three-dimensional electrode array network covering the entire cross section of the dam body.

3. The method for detecting leakage in mine tailings dams based on resistivity tomography according to claim 1, characterized in that, The apparent resistivity of each measuring point is calculated based on the complex impedance. and apparent polarization include: According to the formula Calculate the apparent resistivity, where For device coefficients, The total potential difference, For the supply current; according to the formula Calculate the apparent polarizability, where The potential difference in the secondary field after power failure. The potential difference is the first field potential difference; the apparent resistivity and apparent polarizability at each frequency are combined to obtain a multi-parameter observation dataset.

4. The method for detecting leakage in mine tailings dams based on resistivity tomography according to claim 1, characterized in that, The joint inversion objective function, which includes the prior constraint term of the impermeable layer and the induced polarization two-parameter fitting term, includes: The initial resistivity of the grid cells containing the impermeable layer is set, and the prior variance of the impermeable layer region is set to a small value to form strong constraint weights. The joint inversion objective function is then constructed as follows: ； in and These are the observed apparent resistivity and apparent polarizability data vectors, respectively. and These are the forward operators for resistivity and polarizability, respectively. Here is the smoothness matrix. This is a priori model containing a high resistivity value of the impermeable layer. The prior weight matrix, A factor to balance the weighting of resistivity and polarizability data. The prior constraint strength factor, The regularization parameters are determined using the L-curve method; the Gauss-Newton method is used iteratively to solve for the parameters until the data fitting residuals are reduced to below a preset threshold, thus obtaining the three-dimensional resistivity of the dam body at the current moment. and polarizability distributed.

5. The method for detecting leakage in mine tailings dams based on resistivity tomography according to claim 1, characterized in that, The calculation of time-series differential resistivity and temporal differential polarization include: According to the formula Calculate the time-series differential resistivity; according to the formula Calculate the temporal differential polarization; where and These are background resistivity and background polarizability, respectively. and These are the resistivity and polarizability obtained from the inversion at the current moment, respectively.

6. The method for detecting leakage in mine tailings dams based on resistivity tomography according to claim 1, characterized in that, The triple discrimination of the difference results by sequentially applying the amplitude criterion, spatial continuity criterion, and temporal persistence criterion includes: Time-differential resistivity Temporal differential polarization rate below a preset negative threshold and within the same region Regions exceeding a preset positive threshold are marked as candidate regions satisfying the amplitude criterion; spatial continuity detection is performed on candidate regions satisfying the amplitude criterion, and connected regions with continuous volumes greater than or equal to a preset volume threshold are retained as regions satisfying the spatial continuity criterion, while isolated noise points that do not meet the minimum continuous volume are filtered out; temporal persistence detection is performed on regions satisfying the spatial continuity criterion, and regions that persist and show an expanding trend over several consecutive measurement cycles are identified as leakage anomaly zones, while candidate regions that appear only in a single measurement are excluded.

7. The method for detecting leakage in mine tailings dams based on resistivity tomography according to claim 1, characterized in that, The triggering of corresponding level early warnings based on the comprehensive assessment index of leakage scale and leakage expansion rate includes: A three-tiered early warning system is established based on the comprehensive assessment index of leakage scale: when the abnormal scale is small and expands slowly, a blue Level 1 warning is triggered, monitoring frequency is increased and manual verification is carried out; when the abnormal scale reaches a certain level and shows a continuous expansion trend, a yellow Level 2 warning is triggered, a special investigation is initiated and an emergency plan is formulated; when the abnormal range is large and expands rapidly, posing a dam safety risk, a red Level 3 warning is triggered, and the emergency response procedure is initiated; all levels of early warning signals are pushed to the tailings dam safety management platform in real time to trigger the corresponding emergency response process.

8. A mine tailings dam leakage detection system based on resistivity tomography, characterized in that, The system uses the method described in any one of claims 1 to 7, comprising a pre-embedded distributed three-dimensional electrode array module, a multi-frequency data acquisition and transmission module, a resistivity-polarizability joint inversion module with prior constraints of the seepage prevention layer, a dynamic differential monitoring and multi-dimensional anomaly identification module, and a three-dimensional leakage location and graded early warning module connected in sequence. The pre-embedded distributed three-dimensional electrode array module includes corrosion-resistant electrodes and penetrating electrodes that are pre-embedded in the dam body according to the principle of layering and partitioning. The penetrating electrodes pass through the seepage prevention layer with sealed sleeves and are waterproof and sealed. The electrodes are connected to the dam top control box through a multi-core cable. The multi-frequency data acquisition and transmission module includes a multi-frequency current transmitter and a multi-channel receiver. The multi-frequency current transmitter inputs a multi-frequency pseudo-random waveform current with a frequency range of 0.01Hz to 100Hz to the power supply electrode. The multi-channel receiver synchronously acquires the multi-frequency complex potential difference response of each measurement electrode pair and transmits the data to the data processing center in real time. The resistivity-polarizability joint inversion module with impermeable layer prior constraints is used to establish an initial geoelectric model based on the impermeable layer prior information. The joint inversion objective function with strong impermeable layer prior constraints and induced polarization dual parameter fitting terms is used to perform three-dimensional joint inversion on the observation data to obtain the underground three-dimensional resistivity and polarizability distribution. The dynamic differential monitoring and multidimensional anomaly identification module is used to calculate the time-series difference component between the current inversion result and the background field, and to make a triple judgment on the difference result by combining the amplitude criterion, spatial continuity criterion and temporal persistence criterion to identify the leakage anomaly area; The 3D leakage location and graded early warning module is used to reconstruct the 3D space of the leakage anomaly area and track the pollution plume. Based on the comprehensive assessment index of the leakage scale, it triggers a level 3 early warning and pushes it to the safety management platform.

9. The mine tailings dam leakage detection system based on resistivity tomography according to claim 8, characterized in that, The corrosion-resistant electrode is made of titanium alloy or platinum-iridium alloy. The multi-frequency current transmitter supports electrode combinations of Winner device, Schlumberger device and dipole-dipole device. The data acquisition is automatically executed according to a preset cycle under normal operating conditions. When the anomaly identification module triggers an early warning signal, it automatically switches to high-frequency encrypted acquisition mode.

10. The mine tailings dam leakage detection system based on resistivity tomography according to claim 8, characterized in that, The three-dimensional leakage location and graded early warning module is also used to periodically and automatically generate a comprehensive detection report that includes three-dimensional resistivity and polarizability distribution maps, three-dimensional rendering maps of differential anomaly areas, leakage channel morphology and coordinates, pollution plume range and early warning status.

Images