A channel full-area leakage detection method and system based on water-land electrical method combination

Abstract

The present application belongs to the technical field of channel leakage detection, in order to solve the problem of inaccurate and incomplete channel leakage detection, a channel full-area leakage detection method and system based on water-land electrical method combination is proposed, the water-based high-density resistivity method is used to determine the suspected leakage section of the channel bottom and the channel embankment; for the suspected leakage section of the channel embankment, the water-based static high-density resistivity method and the ground high-density resistivity method are used for detection, and the water-based static high-density resistivity method is used for detection for the suspected leakage section of the channel bottom; then, the apparent resistivity data obtained by the water-based moving high-density resistivity method, the water-based static high-density resistivity method and the ground high-density resistivity method are jointly inverted, the objective function of the joint inversion model is constructed based on the prior structure constraint, and the three-dimensional resistivity distribution model is solved, and then the leakage degree is determined. The method can realize high-precision leakage detection of the whole section of the channel, and can accurately determine the specific position of the leakage.

Description

Technical Field

[0001] This invention belongs to the technical field of channel leakage detection, and in particular relates to a method and system for detecting leakage in the entire channel area based on a combination of water, land and electricity methods. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] As the most common linear water conveyance structure in water diversion projects, canals are widely used in large-scale inter-basin water transfer projects and canal navigation projects. They typically employ a trapezoidal or rectangular cross-section design, with the water passage section formed by the canal bottom and side banks. They are characterized by long routes, large spans, complex geological conditions along the route, and difficulties in maintaining water supply during operation. Leakage is one of the most significant hidden dangers during canal operation, mainly including two types: bottom leakage and bank leakage. Severe leakage not only wastes water resources but can also lead to soil erosion of the canal banks, uneven structural settlement, and even collapse, seriously threatening water conveyance safety and the project's lifespan.

[0004] Currently, among geophysical detection technologies for potential seepage hazards in water conservancy projects, the high-density resistivity method has become one of the main methods for seepage detection due to its advantages such as sensitivity to aquifer structures, high resolution, and non-destructive testing. This method typically involves deploying electrode arrays on the slope or crest of the canal embankment to collect apparent resistivity data and invert the seepage field distribution inside the dam body. This type of method is mainly for water-retaining structures, where the electrodes are deployed on land surfaces without water cover, and the detection targets are changes in the phreatic line and seepage channels within the dam fill. For canal structures, which are in a state of continuous water flow, the canal bottom and the water-facing side of the canal embankment are constantly covered by water, and this unique engineering condition creates unique difficulties for seepage detection.

[0005] Some studies have attempted to combine high-density resistivity methods with other geophysical methods (such as ground-penetrating radar and temperature tracing methods). However, such combinations are mainly applied to the verification phase of canal embankment seepage surveys and detailed investigations, failing to address the unique challenge of simultaneous detection in both water and land environments characteristic of canal structures. Furthermore, they cannot achieve a progressive workflow from rapid long-distance surveys to precise detailed investigations of key sections. In addition, existing technologies lack the ability to quickly identify the location of canal seepage, making it impossible to directly determine whether the seepage occurs at the canal bottom or on the embankment based on detection results. The treatment methods for these two types of seepage are drastically different, and accurate identification of the seepage location is of significant practical importance for engineering decision-making. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, the present invention provides a method and system for detecting leakage in the entire channel area based on a combination of water, land and electricity methods, which can achieve high-precision leakage detection of the entire channel cross section and accurately identify the specific location of leakage.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for detecting leakage in a channel across the entire area based on a combination of hydro- and land-based electrical methods, comprising: The high-density resistivity method of water motion was used to obtain apparent resistivity data along multiple longitudinal survey lines of the channel. The possibility of leakage at each survey point on the survey line was quantitatively assessed to identify suspected leakage sections at the bottom of the channel and suspected leakage sections at the bank. For suspected seepage sections of the canal embankment, the static high-density resistivity method on water and the high-density resistivity method on the ground were used for detection. For suspected seepage sections of the canal bottom, the static high-density resistivity method on water was used for detection to obtain the corresponding apparent resistivity data. Apparent resistivity data obtained by the water-based moving high-density resistivity method, the water-based static high-density resistivity method, and the ground high-density resistivity method are input into the joint inversion model. The objective function of the joint inversion model is constructed based on prior structural constraints. The three-dimensional resistivity distribution model is obtained by solving the objective function. Based on a three-dimensional resistivity distribution model, the degree and type of leakage are determined by calculating the leakage index and combining it with abnormal spatial layers.

[0008] Secondly, the present invention provides a channel full-area leakage detection system based on a combination of water and land electrophysiological methods, comprising: The first module is configured to: use the high-density resistivity method of water motion to acquire apparent resistivity data along multiple longitudinal survey lines of the channel, quantitatively assess the possibility of leakage at each survey point on the survey line, and determine the suspected leakage sections at the bottom of the channel and the suspected leakage sections at the bank of the channel. The second module is configured to: detect suspected seepage sections of the canal embankment using the static high-density resistivity method on the water surface and the high-density resistivity method on the ground, and detect suspected seepage sections of the canal bottom using the static high-density resistivity method on the water surface, and obtain the corresponding apparent resistivity data. The third module is configured to: input the apparent resistivity data obtained by the water surface motion high-density resistivity method, the water surface static high-density resistivity method and the ground high-density resistivity method into the joint inversion model, construct the objective function of the joint inversion model based on the prior structural constraints, and solve the objective function to obtain the three-dimensional resistivity distribution model. The leakage determination module is configured to: determine the degree and type of leakage by calculating the leakage index based on a three-dimensional resistivity distribution model and combining it with the abnormal spatial layers.

[0009] Thirdly, the present invention provides an electronic device including a memory and a processor, and computer instructions stored in the memory and running on the processor, wherein the computer instructions, when executed by the processor, perform the method described in the first aspect.

[0010] Fourthly, the present invention provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the method described in the first aspect.

[0011] Fifthly, the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect.

[0012] The above one or more technical solutions have the following beneficial effects: In this invention, firstly, suspected seepage sections on the channel bottom and channel banks are identified using the water-based motion high-density resistivity method. For the suspected seepage sections on the channel banks, both the water-based static high-density resistivity method and the ground high-density resistivity method are used for detection; for the suspected seepage sections on the channel bottom, the water-based static high-density resistivity method is used for detection. Then, the apparent resistivity data obtained from the water-based motion high-density resistivity method, the water-based static high-density resistivity method, and the ground high-density resistivity method are jointly inverted. Based on prior structural constraints, an objective function for the joint inversion model is constructed, and the three-dimensional resistivity distribution model is solved to calculate the overall leakage index and the local leakage index, thereby determining the degree of channel leakage. This invention's method can achieve high-precision leakage detection across the entire channel cross-section and can accurately identify the specific location of leakage.

[0013] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0014] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0015] Figure 1 This is a flowchart of the canal full-area seepage detection method based on the combination of hydro-land-electric methods in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the spatial configuration of the combined water and land electrical resistivity detection array in Embodiment 1 of the present invention; wherein, a is a schematic diagram of the water-moving high-density resistivity method survey line, and b is a schematic diagram of the water-stationary high-density resistivity method survey line and the land-based high-density resistivity method survey line. Figure 3 This is a flowchart of the joint inversion of water and land electrical method results in Embodiment 1 of the present invention; Figure 4 This is a schematic diagram of the calculation results of a three-dimensional resistivity distribution model for a certain channel section in Embodiment 1 of the present invention; wherein, a is a 500m 3D view of the three-dimensional resistivity distribution model; b is a top view of a portion of the three-dimensional resistivity distribution model; c is a front view of a portion of the three-dimensional resistivity distribution model. Figure 5 The image shows the resistivity verification results of a certain channel section before and after repair in Embodiment 1 of the present invention; where a is the resistivity profile of the channel embankment test section from K3+200 to K3+500; b is the resistivity profile of the channel bottom test section from K8+700 to K9+100; and c is the resistivity profile of the channel bottom test section from K8+700 to K9+100. Detailed Implementation

[0016] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0017] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0018] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0019] Example 1 For canal structures, which are constantly in operation with water flowing through them, the canal bottom and the water-facing bank are perpetually covered by water. This unique engineering condition creates unique challenges for leakage detection. If only the ground-based high-density resistivity method is used to place electrodes on the back bank, the current field cannot effectively penetrate to the canal bottom and the interior of the water-facing bank, creating a blind spot for leakage detection. Furthermore, this method is costly and time-consuming for long-distance measurements. If only the static high-density resistivity method is used to place electrodes on the canal surface, although the internal structure of the canal at the water-facing end can be seen relatively clearly, the detection depth and length are limited by the coupling conditions of the electrodes and the electrode spacing. Measurements can only be taken at the water-facing side, which is close to the bank. It is difficult to operate at the far end, and the single data from the water surface cannot distinguish whether the anomaly originates from leakage at the canal bottom or the canal bank. If the moving high-density resistivity method is used, although large-scale measurements can be taken in a short time and the general condition of the canal can be roughly understood, the resistivity reflected is an estimate that needs further verification.

[0020] Therefore, this embodiment discloses a method for detecting leakage in the entire channel area based on a combination of water and land electrophysiological methods, including: The high-density resistivity method of water motion was used to obtain apparent resistivity data along multiple longitudinal survey lines of the channel. The possibility of leakage at each survey point on the survey line was quantitatively assessed to identify suspected leakage sections at the bottom of the channel and suspected leakage sections at the bank. For suspected seepage sections of the canal embankment, the static high-density resistivity method on water and the high-density resistivity method on the ground were used for detection. For suspected seepage sections of the canal bottom, the static high-density resistivity method on water was used for detection to obtain the corresponding apparent resistivity data. Apparent resistivity data obtained by the water-based moving high-density resistivity method, the water-based static high-density resistivity method, and the ground high-density resistivity method are input into the joint inversion model. The objective function of the joint inversion model is constructed based on prior structural constraints. The three-dimensional resistivity distribution model is obtained by solving the objective function. Based on a three-dimensional resistivity distribution model, the degree and type of leakage are determined by calculating the leakage index and combining it with abnormal spatial layers.

[0021] In this embodiment, firstly, suspected seepage sections on the channel bottom and channel banks are identified using the water-based motion high-density resistivity method. Then, in these sections, both the water-based static high-density resistivity method and the ground-based high-density resistivity method are used for detection. Next, the apparent resistivity data obtained from these three methods are jointly inverted. Based on prior structural constraints, an objective function for the joint inversion model is constructed, and the solution yields a three-dimensional resistivity distribution model, thereby determining the degree of leakage. This method enables high-precision leakage detection across the entire channel cross-section and can accurately identify the specific location of leakage.

[0022] This embodiment addresses the shortcomings of existing channel leakage detection technologies, such as detection blind spots caused by water cover, the difficulty in balancing accuracy and efficiency with single geophysical methods, and the inability to distinguish leakage locations on the channel bottom and bank. It provides a comprehensive channel leakage detection method based on combined hydro- and land-based electrical resistivity methods. Specifically, considering the structural characteristics of channels—long routes, narrow cross-sections, and the coexistence of water and land environments—this embodiment constructs a full-space electrical resistivity detection system covering the channel bottom, the water-facing bank, and the back bank. This method integrates the moving high-density resistivity method on water, the stationary high-density resistivity method on water, and the land-based high-density resistivity method to construct a three-level progressive detection system. By combining hydro- and land-based electrical resistivity data inversion and quantitative evaluation of the leakage index, it achieves high-precision leakage detection across the entire channel cross-section and can accurately identify the specific location of leakage.

[0023] The following is combined with Figure 1 This embodiment provides a detailed description of the technical solution designed for the geometric characteristics and operating conditions of a linear water conveyance structure like a canal: First, through on-site investigation, CAD drawings of the water conveyance channel are collected, and then on-site surveys are conducted to determine the channel sections to be inspected for leakage. Leakage areas are generally manifested in areas where there is large-scale water seepage and leakage on the downstream side of the channel, where the soil is relatively moist or sand boiling occurs.

[0024] The channel to be tested is divided into sections, and the length of each section is determined according to the channel's curvature and water depth variations. The design cross-sectional parameters of each section (including bottom width, slope ratio, and embankment crest width), water depth distribution along the channel, and geological data are collected.

[0025] The first stage is the rapid survey stage, which employs the high-density resistivity method using waterborne motion. A towed multi-channel electrode array is continuously dragged by the survey vessel along the longitudinal axis of the channel at a speed of 0.5–1.2 m / s. For example... Figure 2 As shown, D is the width of the channel water surface, d is the width of the channel bottom, α is the slope angle of the inner side of the channel embankment, H is the height of the channel embankment, h is the water depth, and Sq is the length of the channel section. The electrode spacing of the electrode array is selected based on 0.5 to 1.5 times the average water depth h of this channel section, and the measuring line spacing l is taken as... Furthermore, it is essential to ensure that there is at least one measuring line in the Dd region on both sides of the channel to guarantee that the current field can penetrate the water and reach the strata below the channel bottom and on the side of the channel bank. A single measuring line should not exceed 10km; measurements can be taken in segments or in a single 10km measurement. The measuring lines can be divided according to the actual situation. For example, there can be six measuring lines, L1 to L6. After measuring to the end point of measuring line L1, the measurement can be reversed from the end point of measuring line L2 back to the starting point of measuring line L2. The same applies to measuring lines L3 and L4, and L5 and L6. During the inversion process, the data can be mirrored, which can reduce the measurement time by half. Figure 2 As shown in (a).

[0026] Optionally, the surface high-density resistivity meter can be adapted to freshwater or even marine environments to detect geological bodies below the water surface. The working principle is to calculate the resistivity value of the underground medium by manually introducing current and measuring the voltage difference through the DC resistivity method.

[0027] Optionally, the power supply electrodes and measuring electrodes are arranged using Wenner or Schlumberger devices to continuously collect apparent resistivity data during the vessel's movement. Simultaneously, the location and actual water depth of each measuring point are recorded synchronously using a shipborne differential GPS and depth sounder. The detection time per kilometer of channel in this stage does not exceed 2 hours, achieving efficient surveying of long-distance water conveyance channels.

[0028] After data collection, set appropriate parameters and use data inversion software to invert the apparent resistivity data. If the inversion speed is too slow, the measured .dat file can be split into several smaller files, with each file containing an integer measurement line. For example, 10km can be split into three segments: 3km, 3km, and 4km. After inversion, the segments can be stitched together to improve the accuracy and speed of the inversion. The resistivity data of each measured profile of the channel can be obtained after the inversion.

[0029] The resistivity profile data of multiple measuring lines along the longitudinal direction of the channel are obtained. A leakage probability model is used to quantitatively assess the probability of leakage at each measuring point, replacing empirical threshold judgment and improving the objectivity and accuracy of identifying suspected leakage sections.

[0030] The leakage probability model is constructed based on a Bayesian framework, taking into account the magnitude of resistivity anomalies, abnormal vertical continuity, and prior information about the channel cross-section.

[0031] First, define the leakage probability at the i-th measuring point. The calculation formula is as follows:

[0032] in, The overall leakage score is determined by anomaly indicators of resistivity. Vertical connectivity index Abnormal water conductivity indicators and prior probability index The weighted sum of the four terms is given by the expression:

[0033] The weighting coefficients in the formula satisfy:

[0034] Recommended value based on experience Each weight is optimized and adjusted based on existing validation data in the engineering area through sensitivity analysis or machine learning methods (such as gradient boosting trees).

[0035] Abnormal resistivity indicators :

[0036] In the formula, It is the resistivity value obtained by inversion at the i-th measuring point. This value comes from the rapid inversion result of the first stage and represents the average resistivity of the strata at this measuring point. The unit is Ω·m. It is the background resistivity value of the layer where the i-th measuring point is located. The method of obtaining the value is to take the median of the resistivity of all measuring points in the same layer (such as the same depth range below the canal bottom) along the entire detection profile, or to take the average resistivity of the undisturbed section. If there is geoelectric stratification information, the profile can be divided into layers according to the sedimentary layers and then statistically analyzed. The unit is Ω·m. The resistivity is the lowest possible resistivity achievable at the stratum where the measuring point is located. It is generally determined based on experience or borehole data. For example, the resistivity of fully saturated clayey sediments can be as low as 1–5 Ω·m, or by referring to the measured minimum value in a known severely leaky area in the local area. The unit is Ω·m. Below the background resistivity value and approaching the minimum resistivity hour, Close to 1.

[0037] Vertical connectivity index :

[0038] In the formula, It is the total number of discrete grid layers along the depth direction, which is determined by the maximum detection depth of the first inversion model and the grid size, and is usually taken as 10 to 30 layers. Let be the leakage indicator function of the grid at the k-th depth at the i-th measuring point, and let the resistivity of the grid be . Less than the background resistivity at the same depth Furthermore, if, starting from this grid, three or more consecutive adjacent grids in the vertical direction all satisfy the condition that the resistivity is lower than the background value, then... =1; otherwise =0, this rule is used to identify leakage channels that penetrate formations; The thickness of the k-th mesh layer is determined during the inversion mesh generation. Typically, shallow mesh layers are thinner (0.5~1 m), while deeper mesh layers can be appropriately thickened. Unit: m. The maximum detection depth is determined by the maximum depth at which resistivity information is stably obtained through the inversion algorithm. It is usually 1 / 3 to 1 / 2 of the maximum electrode spacing of the electrode array, in meters.

[0039] Abnormal water conductivity indicators :

[0040] In the formula, The measured water conductivity at the i-th measuring point is obtained in real time by an automatic water sampling device added to the measuring vessel and connected to a multi-parameter water quality analyzer. The sampling frequency is synchronized with the resistivity data to ensure that each measuring point has a corresponding conductivity value. The unit is μS / cm. The background value of the water conductivity at the measuring point is taken as the low percentile (such as the 10th percentile) of the upstream background section or the entire detection profile that is not affected by leakage. It can also be set according to historical monitoring data or the average conductivity of normal water in the channel. The unit is μS / cm. The maximum possible conductivity caused by leakage is determined based on the measured conductivity of deep groundwater (usually higher than that of surface water) or the extreme conductivity after the known contaminants are mixed in. If this value cannot be obtained, it can be estimated by using 3 to 5 times the background value. The unit is μS / cm.

[0041] Abnormal water conductivity indicators The basic principle is that when highly mineralized seepage water (such as deep groundwater or polluted water) enters the channel, it will locally increase the electrical conductivity of the water body. Significantly higher than hour, A value close to 1 indicates the presence of active leakage input.

[0042] Prior probability index :

[0043] In the formula, Let j be the prior feature variables of the i-th measuring point. These variables are binary codes of 0 or 1, depending on the prior information before the measurement, such as: whether there is a construction joint in the canal bottom lining (yes=1, no=0), whether it is located above the permeable layer revealed by geological boreholes (yes=1, no=0), whether there is a historical leakage record (yes=1, no=0), and whether it is located in a low-lying area or an area with excessive soil moisture (yes=1, no=0). The regression coefficients for the corresponding characteristic variables can be obtained by collecting historical leakage detection data from this project or neighboring projects and training the system using the Logistic Regression method. If sufficient historical data is unavailable, values ​​can be assigned based on experience. For strongly correlated features (such as construction joints), the coefficients can be... =2, weak correlation feature A value of 0.5 can be taken. If no prior information is available, to ensure model stability, a value of 0.5 can be directly used. =0.5 indicates no prior bias.

[0044] Calculate the leakage probability at each measuring point. Subsequently, the regulations are as shown in Table 1: Table 1:

[0045] Alternatively, when using the high-density resistivity method for waterborne motion, given the narrow and long characteristics of the channel, the boat carrying the resistivity instrument can be designed to resemble an arrowhead to reduce water and air resistance. The electrode distance during measurement should be set between 1 m and 3 m, and the cable length between 20 m and 40 m, depending on the water depth. This ensures measurement accuracy while reaching the depth range of the channel. In addition to a vertical sonar for real-time measurement of the boat's water depth and a GPS device for real-time measurement of the boat's position, an automatic water sampling device can be added, connected to a multi-parameter water quality analyzer. This analyzer can analyze the conductivity of the surrounding water in real time. pH value, temperature and other water chemical parameters; install iron fence protective net at the propeller of the motor at the rear of the boat to prevent the propeller from getting entangled with fishing nets, troughs and other debris; increase lateral sonar to measure the distance between the boat and the shore in real time to ensure that the boat is parallel to the direction of the channel and travels on the established route.

[0046] The second stage is the fine detection stage, which distinguishes the leakage location based on the suspected leakage sections marked in the first stage. If the leakage occurs in the channel embankment section (including but not limited to the water conveyance structures connected to the channel such as dikes, aqueducts, intake gates, and inverted siphons) and causes lateral leakage, the leakage channel embankment is measured by the above-water static high-density resistivity method and the ground high-density resistivity method.

[0047] The static high-density resistivity method on water involves laying multiple static measuring lines perpendicular to the longitudinal axis of the canal on the water-facing side of the canal embankment, in conjunction with the suspected seepage area. The number of measuring lines, n, satisfies n ≥ The spacing between the survey lines is 0.5m to 2m, and the distance between the survey lines and the shore for the static high-density resistivity method on water is ≤ To ensure the measuring line is only above the canal embankment, effectively detecting the embankment lining structure, the electrode array is kept completely still after being submerged in water. Employing a multi-channel, multi-overlay measurement mode, hundreds to thousands of sets of apparent resistivity data are acquired within a 30-60 minute acquisition cycle. Long-term overlay effectively suppresses water flow disturbance and environmental noise, significantly improving the signal-to-noise ratio and vertical resolution, enabling clear differentiation of the canal bottom structure and shallow geological details.

[0048] On the corresponding downstream side of the canal embankment, high-density resistivity survey lines are laid along the embankment crest or slope, with the survey lines parallel to the longitudinal axis of the canal. Stainless steel electrodes are inserted into the unwatered soil. If the ground is hardened (asphalt, masonry), non-polarized electrodes with bentonite are used instead, ensuring that the electrode grounding resistance meets the measurement requirements. The ground survey lines are also collected using a static acquisition method, with the measurement time varying from approximately 0.5 to 2 hours depending on the measurement method, to obtain high-resolution resistivity data of the canal embankment interior.

[0049] If the survey line is too long and the cable length is insufficient, the roller method can be used to extend the survey line. By collecting data simultaneously or at different times on the water and on the land, seamless coverage of the water-covered area and the land area at the same cross-section of the channel is achieved.

[0050] If leakage occurs at the bottom of the canal, resulting in vertical seepage that bypasses the embankment and reaches the downstream side, multiple parallel survey lines can be laid out at the leakage site to form a rectangular grid. Measurements can then be taken using the static high-density resistivity method. If possible, the canal bottom can be drained, and a survey line plan can be developed above the sediment layer at the suspected leakage location for surface high-density resistivity measurements to collect apparent resistivity data. For both types of leakage, if possible, an underwater robot equipped with an optical camera and imaging sonar can be used for underwater verification to confirm the location, shape, and size of the underwater seepage channels at the bottom of the canal and the downstream end of the embankment.

[0051] Alternatively, when using the hydrostatic high-density resistivity method, the wiring method can be selected according to the channel conditions: First, the hydrostatic high-density resistivity method can be used parallel to the channel direction. If the channel width is less than 10m, two people can pull two ropes, standing at opposite ends of the river. The ends of the cable are fixed to the ropes via rails for measurement. If the channel width is greater than 10m, it is difficult to extend the ropes to the opposite side of the river. In this case, the anchoring method can be used. One person connects the cable to the measuring instrument at the beginning of the cable, while the other person holds the end of the cable and walks away from the channel bank on the same side, straightening the cable. A weight is then tied to the end of the cable, tilted at a certain angle in the direction indicated by the end of the cable, and thrown. The weight will fall into the water and sink to the bottom, acting as an anchor. The weight provides a force to the end of the cable away from the channel bank. Since the water flow is along the channel, the cable will naturally be pulled from the end to the beginning. The distance from which the weight is thrown is the distance S between the cable and the channel bank. Figure 2 As shown in (b). If the anchoring distance is too far in this method, the distance can be adjusted by pulling the cable; if the distance is too short, the weight can be pulled back and re-thrown. Second, the static high-density resistivity method for water in the vertical channel direction. Two people, each holding one end of the cable, complete the measurement at both banks of the channel embankment.

[0052] The third stage is the combined land and water constraint inversion. The inversion process is as follows: Figure 3 As shown, the apparent resistivity data of water movement obtained in the first stage, the apparent resistivity data of water stationary obtained in the second stage, and the high-density apparent resistivity data of the ground are jointly input into the joint inversion model.

[0053] The joint inversion model utilizes the ResIPy software platform. After importing the two-dimensional survey lines, coordinates are assigned to all survey points using data such as water depth and GPS. A priori structural cognitive model is constructed based on channel design drawings and geological data. The canal bottom lining layer, canal bottom foundation, water-facing bank fill, backwater bank fill, and dam foundation are each designated as different layers, with a priori resistivity range assigned to each layer (e.g., 7–14 Ω·m for water-supported layers, 20–50 Ω·m for saturated fill, and 100–500 Ω·m for dry or compacted fill). The joint inversion model is discretized into a regular grid, with the resistivity value of each grid denoted as... In the inversion, the actual solution is its logarithm. This is to ensure that the result is positive.

[0054] The goal of joint inversion is to find an optimal model vector. This allows it to simultaneously fit three types of observation data while satisfying prior structural constraints.

[0055] Therefore, an objective function is defined. as follows:

[0056] In the formula, For data fitting terms; For model smoothing constraints; This is a penalty term for deviation from a priori resistivity; This is a penalty term for deviations from the geometric structure. , , is the regularization coefficient, used to balance the contributions of each term; empirically, it is taken as . =0.1, =0.5= .

[0057] The inversion process not only seeks to minimize the error between the forward response of the joint inversion model and the measured data, but also... Minimize, while also being constrained by the smoothness of the model, the physical range of resistivity, and the actual engineering geometry, so as to obtain a reliable solution that conforms to both geophysical data and prior engineering knowledge.

[0058] Data fitting term :

[0059] In the formula, =1, 2, 3 represent the high-density resistivity method of water motion, the high-density resistivity method of water stationary, and the high-density resistivity method of land, respectively. For the first The total number of measurement points for this type of data; This is the measured apparent resistivity; For the current model Theoretical apparent resistivity calculated using forward modeling; For the first Standard error of class data; This is the average of all measured apparent resistivity values. Due to the low signal-to-noise ratio of water motion data, we take... =0.05 The signal-to-noise ratio of static water data and ground data is high, so we choose... = =0.02 This formula measures the model's fit to the observed data using a weighted sum of squared residuals, with the denominator... The weight of low-quality data is automatically reduced, making the inversion more reliant on high-quality data.

[0060] Model smoothing constraint term :

[0061] In the formula, This represents the total number of grid cells. Representation and grid Adjacent grid sets (in six directions: up, down, left, right, front, and back). and Adjacent grids and grid Logarithmic resistivity.

[0062] This penalty for resistivity differences between adjacent grids forces the model to be spatially continuous and smooth, preventing isolated spurious anomalies from being generated during inversion.

[0063] Prior resistivity constraint :

[0064] In the formula, , ; and They are the first The upper and lower limits of resistivity of the layer where each grid is located are given in the prior intervals above.

[0065] This formula generates a positive penalty term only when the resistivity of the model exceeds the prior interval, and the penalty increases as the resistivity exceeds the prior interval; the penalty is zero when the resistivity falls within the prior interval. This constraint ensures that the inversion results conform to physical common sense. For example, the canal water layer will not invert an abnormally high resistivity >100Ω·m, and saturated fill will not be mistakenly transformed into a dry high-resistivity body.

[0066] Geometric constraints :

[0067] In the formula, The number of prior geometric interfaces (such as the channel bottom interface, the water-facing channel bank interface, the backwater channel bank interface, etc.). To start from the current model The extracted first The actual depth of an interface is generally automatically identified by the location of the maximum gradient of resistivity in the vertical direction. For the first The design depth of each interface is derived from design drawings or geological profiles; To allow for a depth deviation tolerance, this embodiment uses 0.2~0.5 m.

[0068] The deviation of the penalty inversion interface from the design structure makes the resistivity stratification consistent with the actual geometry of the channel, thereby effectively eliminating data jumps caused by differences in measurement methods at the water-land interface.

[0069] The minimum value of the objective function is found using the Gauss-Newton iterative method. In each iteration, the Jacobian matrix and forward response are calculated based on the current model, and the model update is obtained by solving the system of linear equations. Then with step size Update the model:

[0070] Where t represents the number of iterations. Indicates the step size.

[0071] After each iteration, two checks are performed: first, the resistivity is checked grid by grid to see if it exceeds the prior range; if it does, a forced correction to the range boundary is performed; second, the interface depth of each layer is extracted, and if the deviation from the design depth exceeds the specified value, the error is corrected. Then, the design depth is adjusted towards the prior structural cognitive model. The iterative convergence condition is the relative change of the objective function. Less than 0.001 or the maximum number of iterations of 50 has been reached.

[0072] Convergence criterion: △ <0.001, or the maximum number of iterations of 50 is reached.

[0073] After convergence, the logarithmic resistivity is converted back to resistivity. This model outputs a three-dimensional resistivity distribution model in a unified coordinate system. Within the same coordinate frame, this model integrates resistivity information from the water body coverage area obtained through both moving and stationary high-density resistivity data from the water surface, as well as resistivity information from the land area obtained through the ground high-density resistivity method. Due to the introduction of geometric constraints, the resistivity contour lines at the water-land interface transition smoothly, eliminating data jumps and providing a reliable foundation for subsequent leakage index calculations. Multiple engineering practices have shown that the three-dimensional resistivity distribution model proposed in this embodiment improves leakage identification efficiency by approximately 20% compared to the traditional two-dimensional resistivity distribution model, and its structural intuitiveness is greatly enhanced.

[0074] Finally, based on the inverted three-dimensional resistivity distribution model, the leakage index is calculated. First, the three-dimensional resistivity distribution model is discretized into a regular grid: the grid side length is 5 m along the channel longitudinal direction (x-direction), 2 m perpendicular to the channel direction (y-direction), and 1 m in the depth direction (z-direction), with each grid having a volume of 10 m³. 3 .

[0075] Suppose the three-dimensional resistivity distribution model has N grids, for each grid Define their relative resistivity difference for:

[0076] In the formula, For this grid Resistivity, in Ω·m; For this grid The background resistivity of the layer, in Ω·m, is usually taken as the median of the resistivity of all grids in the same layer.

[0077] To account for the vertical connectivity of the leakage channels, a connectivity correction factor is introduced. Check along the vertical direction: If there are three or more consecutive grids... That is, the difference in relative resistivity If the percentage exceeds 15%, these grid lines are considered to constitute leakage channels, and samples are taken from them. ,otherwise .

[0078] Define the local leakage index The leakage intensity at a single grid location reflects the severity of leakage at that local point, expressed as a percentage:

[0079] Define the overall leakage index wTo characterize the average leakage severity of the entire detection area (such as a detection section or the entire channel), it is the arithmetic mean of the local leakage indices of all grids, expressed as a percentage:

[0080] In the formula, N is the total number of grid cells; For the first The connectivity correction factor for each grid; For the first The relative resistivity difference of each grid (dimensionless).

[0081] This formula reflects the average decrease in resistivity caused by leakage in the entire detection area (amplified by connectivity). The higher the value, the more severe the overall leakage.

[0082] According to the local leakage index Value and overall leakage index By comparing the leakage level with the corresponding preset threshold, the leakage level can be divided into four levels, as shown in Tables 2 and 3.

[0083] Table 2:

[0084] Table 3:

[0085] Based on actual engineering needs, firstly, based on the overall leakage index... The numerical value determines whether repair is needed. Generally speaking... If the leakage rate is >15%, repairs are required. An underwater robot (ROV) can be used to assess the overall leakage index. Direct observation was conducted to verify the severe leakage area (>15%). Secondly, the local leakage index was analyzed grid by grid. Repair areas exceeding 15% of the leakage rate. After repair, the resistivity testing process described above can be repeated. The repair effectiveness can be evaluated using the inversion results. If the overall leakage index... If the overall leakage index is less than 5%, the repair effect is good; if the overall leakage index is less than 5%, the repair effect is good. If the result is greater than 5%, the repair effect is generally poor. This embodiment forms a complete technical closed loop of "geophysical exploration - joint inversion - index discrimination - repair verification".

[0086] The repair methods for seepage at the bottom of the canal and seepage at the canal bank differ in some ways: Canal bottom leakage is typically caused by the following factors: hydraulic fracturing due to water pressure differences between the inside and outside of the canal; damage to the geomembrane at the canal bottom due to aging, mechanical damage, or construction defects during long-term operation; and deterioration and cracking of the cement-soil impermeable layer due to freeze-thaw cycles, wet-dry cycles, or chemical erosion. These causes result in leakage channels developing downwards from the canal bottom, bypassing the bottom of the canal bank, and forming a leakage outlet on the other side of the bank from bottom to top. Canal bottom leakage must be repaired under waterless conditions. Specific measures include: shutting off water and draining the canal section, removing sediment from the canal bottom, repairing or re-casting the damaged cement-soil layer, replacing or repairing the geomembrane, and, if necessary, using underwater repair materials for targeted grouting. In emergency situations where water cannot be shut off, special underwater grouting materials can be used with the assistance of divers or underwater robots.

[0087] Canal embankment leakage is mainly caused by the following factors: exposure of expansion joints or construction joints in the canal embankment; aging and failure of sealing materials; damage or poor overlap of the geomembrane during embankment slope laying; and insufficient compaction of the embankment backfill, creating preferential flow channels under water level changes. These factors cause leakage channels to form on the water-facing side of the canal embankment, extending laterally through the canal embankment and forming leakage outlets on the back slope or at the toe of the slope. Canal embankment leakage can be repaired under conditions of canal operation with or without partial water supply. Specific measures include: grouting the leakage channels (cement grout, chemical grout, or polyurethane grout); constructing vertical anti-seepage walls on the water-facing or back slope of the canal embankment; reinforcing the embankment slope with masonry or concrete panels; and resealing the expansion joints. Usually, it is not necessary to drain the entire canal, and the construction has minimal impact on water transport.

[0088] The following is an example of a channel full-area seepage detection system according to this embodiment, which includes: The high-density resistivity method data acquisition subsystem for waterborne operations includes an electrical resistivity meter, a towed 13-electrode cable, an arrow-shaped inflatable boat, differential GPS (positioning error <0.1 m), a dual-frequency sonar depth sounder (accuracy 1 cm), an automatic water sampling device, and a multi-parameter water quality analyzer. This system is used for long-distance rapid surveys at speeds of 0.5 m / s to 1.2 m / s, and for real-time acquisition of apparent resistivity, water depth, location, and water conductivity data.

[0089] The underwater static high-density resistivity method acquisition subsystem includes the same instruments as the kinematic resistivity method, but adds an anchoring device (such as a weight, rope, or rail) to lay out static survey lines perpendicular to the channel direction in suspected sections. The acquisition cycle is 30-60 minutes, and high signal-to-noise ratio data is obtained.

[0090] Ground high-density resistivity method acquisition subsystem: includes an electrical resistivity meter, cables, and >96 stainless steel electrodes. This system is used for land area measurements on the backwater side of canal embankments, with survey line lengths of 100~500 m and acquisition time of 0.5~2 hours per survey line.

[0091] Optionally, the cable has 32 electrode interfaces.

[0092] Data processing and inversion subsystem: including but not limited to ResIPy, Res2Dinvx software, etc., used to import three types of data, construct a priori structural models, perform Gauss-Newton joint inversion, and output a three-dimensional resistivity distribution model.

[0093] Leakage index calculation and discrimination module: Built into the processor, it automatically calculates the relative resistivity difference of each grid based on the three-dimensional resistivity distribution model. Connectivity correction factor The overall leakage index (SI) is calculated using the formula for the overall leakage index (SI), and leakage levels are classified according to a set threshold. At the same time, the leakage type is automatically determined based on the abnormal layer.

[0094] Underwater verification subsystem: includes an underwater robot (ROV) equipped with an optical camera with a resolution of ≥1080p, an image sonar with a detection range of 50 m, and a positioning system, used to assess the overall leakage index. Direct observation was conducted to verify the severe leakage area of ​​>15%.

[0095] Central control and communication module: It adopts an industrial-grade tablet computer or laptop and connects to each subsystem via WiFi or wired connection. It displays the measurement trajectory, resistivity profile, conductivity curve and leakage probability Pi in real time, and stores the raw data and inversion results.

[0096] This embodiment uses a water diversion project channel as an example to describe in detail a method for detecting seepage in the entire channel area based on a combination of hydro-land-electric methods. The project section is 16.9 km long. Before the renovation, the bottom width was 45 m, the design water depth was 3.2 m, and the slope ratio was 1:3. After the renovation, the channel bottom width is 60 m, the design water depth is 4.2 m, and the slope ratio is 1:3. The slope below the abutment uses a composite geomembrane for seepage prevention, and the channel bottom uses a 40 cm thick cement-soil seepage prevention method.

[0097] First, design drawings and geological survey data (riverbed elevation 32.20 m, strata mainly composed of clay and sandy loam, permeability coefficient 0.18 m / d~1.38 m / d) and the distribution of structures along the river (25 drainage pumping stations, 17 culverts, etc.) were collected. On-site reconnaissance confirmed no large-scale seepage and wetting areas on the backwater side; however, considering that the widening and deepening of the river channel had damaged the original seepage-proof layer, a full-section inspection was necessary. The 16.9 km long section was divided into four inspection segments based on the locations of bridges, culverts, etc.: K2+250~K6+500, K6+500~K10+800, K10+800~K15+000, and K15+000~K19+250, each segment approximately 4 km long.

[0098] A high-density electrical resistivity tomography (EDT) instrument was used, equipped with a towed cable consisting of 13 graphite electrodes. The electrode spacing was 3 m based on an average water depth of 4.2 m (meeting a depth requirement of 0.5 to 1.5 times the water depth). The survey vessel was an arrow-shaped inflatable boat to reduce resistance, with a speed controlled at 0.8 m / s. The survey lines were laid out as follows: four parallel survey lines were laid longitudinally along the channel, two along the channel bottom and two near the bank (approximately 9 m apart), for a total length of 4 × 16.9 ≈ 68 km. During the measurement, a vessel-mounted differential GPS with an accuracy of <0.1 m recorded the location in real time, a depth sounder recorded the water depth synchronously, and an automatic water sampling device connected to a multi-parameter water quality analyzer collected water conductivity, pH, and temperature data every 10 seconds. The power supply and measurement electrodes were arranged using a Winner device, with a sampling rate of 1000 samples / minute. Each measurement segment employed a reciprocating survey method: for example, survey line L1 was measured from the starting point to the end point, and then L2 was measured from the end point back to the starting point to save time. The entire inspection process takes approximately 8 hours, with each kilometer taking about 0.5 hours to inspect.

[0099] The collected apparent resistivity data were inverted using data inversion software. The 16.9 km data was divided into four segments (approximately 4 km each), inverted separately, and then stitched together to obtain a 16.9 km longitudinal resistivity profile. Then, a leakage probability model was used to calculate the probability value for each measuring point. The background resistivity of this channel section is based on statistics from the undisturbed section: the measured resistivity of the channel water layer is approximately 11 Ω·m, the background resistivity of the saturated fill is taken as 35 Ω·m, and the lowest resistivity... Take 5 Ω·m as the background value for water conductivity. Take 450 μS / cm (upstream clean section). 1500 μS / cm was used (based on actual measurements of deep groundwater). Prior characteristic variable x i,j According to the design drawings, the value is set to 1 at construction joints and 0 elsewhere. Calculations show there are 3 sections. ≥0.7, located at K3+200~K3+500, K8+700~K9+100, and K14+200~K14+600 respectively, are marked as suspected leakage sections.

[0100] For the suspected seepage section from K3+200 to K3+500, the seepage type was first determined: localized dampness was observed on the backwater side of the canal embankment, indicating lateral seepage. Therefore, a combined approach of static high-density resistivity measurement on water and ground was employed. On one side of the water surface, three static high-density resistivity measurement lines were laid parallel to the canal in the suspected section, spaced 1 m apart (1 m, 2 m, and 3 m from the bank, respectively), each line 62 m long (extending from the bank into the canal), with an electrode spacing of 2 m. Anchoring was used to place the cable in the water, and the cable was pulled to maintain complete stillness. Data was collected for 60 minutes using a multi-channel overlay mode, yielding approximately 3600 sets of apparent resistivity data. Correspondingly, two ground high-density resistivity measurement lines were laid on the top of the backwater side of the canal embankment along the parallel canal direction (0 m and 2 m from the dam crest), each 62 m long, with an electrode spacing of 1 m. Stainless steel electrodes were inserted into the soil, and data was collected using an electrical resistivity meter, with a measurement time of approximately 1 hour per line.

[0101] For the section from K8+700 to K9+100, preliminary exploration using an underwater robot (ROV) revealed cracks in the cement-soil layer at the bottom of the canal, indicating leakage. Three parallel measuring lines (perpendicular to the canal direction) were laid directly above the leakage area, with a line spacing of 2 m and an electrode spacing of 2 m, using the static high-density resistivity method. Simultaneously, after draining the water from this section of the canal, a thick layer of riverbed sediment was found on the surface of the cement-soil layer. Therefore, three ground-based high-density measuring lines were laid above the sediment, using stainless steel electrodes. The electrode and measuring line spacing was the same as in the static high-density resistivity method, collecting high-precision resistivity data.

[0102] The apparent resistivity data of the four waterborne motion survey lines obtained in the first stage, the data of the six waterborne static survey lines obtained in the second stage, and the data of the five ground survey lines were imported into ResIPy software. A priori structural cognitive model was constructed: the layers are divided into channel water layer (resistivity 7Ω·m~14 Ω·m), channel bottom cement soil layer (40Ω·m~100 Ω·m), saturated fill layer (20Ω·m~50Ω·m), dry fill layer (100Ω·m~500 Ω·m), and geomembrane layer (>1000 Ω·m).

[0103] Discretized mesh sizes: 5 m in the x-direction, 2 m in the y-direction, and 1 m in the z-direction. The objective function regularization parameters are set to μ1=0.1, μ2=0.5, and μ3=0.5. The model converges after 32 Gauss-Newton iterations with an RMS of 4.2%. A three-dimensional resistivity distribution model is output, and a 500 m section is selected as an example. Figure 4 As shown in (a), a partial top view of the three-dimensional resistivity distribution model is extracted as follows. Figure 4 As shown in (b), a partial front view of the three-dimensional resistivity distribution model is extracted as follows. Figure 4As shown in (c), the overall leakage of the channel can be analyzed by cutting the three-dimensional resistivity distribution model into blocks, sections, and setting thresholds.

[0104] Analysis of the entire three-dimensional resistivity distribution model revealed a low-resistivity zone running through the water-facing side of the canal embankment from K3+200 to K3+500. The underlying stratum exhibited an anomaly of low resistivity less than 200 Ω·m (background value 400 Ω·m), suggesting potential damage to the geomembrane / concrete lining. Referring to the two-dimensional cross-section of this tested section, as shown... Figure 5 As shown in (a), and connected to the wet area on the back side, it was confirmed as a seepage channel of the canal embankment; the resistivity of the cement-soil layer at the bottom of the canal from K8+700 to K9+100 decreased from the normal value of 80 Ω·m to 12~25 Ω·m, and the backfill below showed low resistance anomaly, which was judged to be caused by the damage of the cement-soil at the bottom of the canal, resulting in vertical seepage at the bottom of the canal.

[0105] Calculate the overall leakage index For the K3+200~K3+500 segment, the total number of grids N=12000, and the average relative resistivity difference... =0.32, connectivity correction factor =1.5 (channel length > 3 m), =(1 / 12000)×Σ(1.5×0.32)×100%=48%. According to the grading standard ( (Leakage exceeding 30% is considered extremely serious), and is therefore classified as extremely serious leakage. For the section from K8+700 to K9+100, =22%, judged as serious leakage.

[0106] High-pressure jet grouting was used to treat the seepage points on the water-facing side of the canal embankment from K3+200 to K3+500, while the backwater side was reinforced with masonry. A follow-up measurement was conducted after the repairs. A decrease to 4% indicates a good repair effect.

[0107] The resistivity profile of the channel section from K8+700 to K9+100 before repair is as follows: Figure 5 As shown in (b), it can be seen that the resistivity at the bottom of the canal is intermittently lower than the background resistivity at a depth of 4 m. Measures were taken to completely drain and re-line the entire canal section using cement and soil. Six months after the completion of this work, the entire 68 km section of the canal was re-measured under the same measurement scheme. The resistivity profile of the repaired section from K8+700 to K9+100 is shown below. Figure 5 As shown in (c), the canal bottom is flat and intact, and the water resistivity is stable. The leakage rate dropped to 0.3%, indicating that the leakage level was reduced to zero or slight. This demonstrates the significant effectiveness of the repair solution and verifies the validity of this embodiment.

[0108] This embodiment addresses the characteristics of long canals and dual water and land environments by constructing a full-space electrical resistivity detection system covering the canal bottom and both banks. It integrates the moving high-density resistivity method on water, the stationary high-density resistivity method on water, and the ground high-density resistivity method, forming a three-stage progressive detection process: rapid general survey using a ship-borne towed electrode array for long-distance data acquisition; fine detection by deploying stationary survey lines in suspected sections to obtain high-resolution data; joint inversion based on a priori structural model to construct an objective function and output a three-dimensional resistivity distribution model; and finally, calculation of the leakage index and automatic identification of the leakage type at the canal bottom or banks. This embodiment achieves high-precision leakage detection across the entire canal cross-section, providing direct evidence for engineering repair and can be widely applied to leakage detection in water diversion projects, canals, and other linear water conveyance structures.

[0109] Example 2 The purpose of this embodiment is to provide a channel full-area leakage detection system based on a combination of water and land electrophysiological methods, including: The first module is configured to: use the high-density resistivity method of water motion to acquire apparent resistivity data along multiple longitudinal survey lines of the channel, quantitatively assess the possibility of leakage at each survey point on the survey line, and determine the suspected leakage sections at the bottom of the channel and the suspected leakage sections at the bank of the channel. The second module is configured to: detect suspected seepage sections of the canal embankment using the static high-density resistivity method on the water surface and the high-density resistivity method on the ground, and detect suspected seepage sections of the canal bottom using the static high-density resistivity method on the water surface, and obtain the corresponding apparent resistivity data. The third module is configured to: input the apparent resistivity data obtained by the water surface motion high-density resistivity method, the water surface static high-density resistivity method and the ground high-density resistivity method into the joint inversion model, construct the objective function of the joint inversion model based on the prior structural constraints, and solve the objective function to obtain the three-dimensional resistivity distribution model. The leakage determination module is configured to: determine the degree and type of leakage by calculating the leakage index based on a three-dimensional resistivity distribution model and combining it with the abnormal spatial layers.

[0110] In further embodiments, the following is also provided: An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When executed by the processor, the computer instructions perform the method described in Embodiment 1. For brevity, further details are omitted here.

[0111] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0112] Memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of memory may also include non-volatile random access memory. For example, memory may also store information about the device type.

[0113] A computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the method described in Embodiment 1.

[0114] The method in Embodiment 1 can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not provided here.

[0115] A computer program product includes a computer program that, when executed by a processor, implements the method described in Embodiment 1.

[0116] The present invention also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which execute in a device on a target real or virtual processor to perform the processes / methods described above. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided among program modules as needed. The machine-executable instructions for the program modules can execute within a local or distributed device. In a distributed device, the program modules can reside in both local and remote storage media.

[0117] The computer program code used to implement the methods of the present invention may be written in one or more programming languages. This computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the computer or other programmable data processing device, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a computer, partially on a computer, as a stand-alone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.

[0118] In the context of this invention, computer program code or related data may be carried by any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and the like. Examples of signals may include electrical, optical, radio, sound, or other forms of propagation signals, such as carrier waves, infrared signals, etc.

[0119] Those skilled in the art will recognize that the units and algorithm steps described in conjunction with the embodiments herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0120] 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 method for detecting leakage in a channel across the entire area based on a combination of hydro- and land-based electro-optical methods, characterized in that, include: The high-density resistivity method of water motion was used to obtain apparent resistivity data along multiple longitudinal survey lines of the channel. The possibility of leakage at each survey point on the survey line was quantitatively assessed to identify suspected leakage sections at the bottom of the channel and suspected leakage sections at the bank. For suspected seepage sections of the canal embankment, the static high-density resistivity method on water and the high-density resistivity method on the ground were used for detection. For suspected seepage sections of the canal bottom, the static high-density resistivity method on water was used for detection to obtain the corresponding apparent resistivity data. Apparent resistivity data obtained by the water-based moving high-density resistivity method, the water-based static high-density resistivity method, and the ground high-density resistivity method are input into the joint inversion model. The objective function of the joint inversion model is constructed based on prior structural constraints. The three-dimensional resistivity distribution model is obtained by solving the objective function. Based on the three-dimensional resistivity distribution model, the degree and type of leakage are determined by calculating the leakage index and combining it with the abnormal spatial strata. The probability of leakage at each measuring point on the measuring line is quantitatively assessed to identify suspected leakage sections at the bottom of the canal and at the banks of the canal. Specifically, the leakage probability at each measuring point is determined based on anomaly in resistivity, vertical connectivity, anomaly in water conductivity, and prior probability. The resistivity anomaly index for: in, It is the average resistivity of the stratum at the i-th measuring point; It is the background resistivity of the layer where the i-th measuring point is located; This represents the lowest resistivity reached at the layer where the measurement point is located. The vertical connectivity index for: in, It is the total number of discrete mesh layers along the depth direction; It is the leakage indication function of the k-th depth grid at the i-th measuring point; Let be the thickness of the k-th mesh layer; Maximum detection depth; The abnormal water conductivity index for: in, Let be the measured water conductivity at the i-th measuring point; The maximum possible conductivity due to leakage; This is the background value of the water conductivity at this measuring point; The prior probability index for: in, Let j be the prior feature variable of the i-th measurement point; Let be the regression coefficient of the j-th prior feature variable; The objective function for: in, For data fitting terms; For model smoothing constraints; This is a penalty term for deviation from a priori resistivity; This is a penalty term for deviations from the geometric structure. , , The regularization coefficient is used. =1, 2, 3 represent the high-density resistivity method of water motion, the high-density resistivity method of water stationary, and the high-density resistivity method of land, respectively. For the first The total number of measurement points for this type of data; This is the measured apparent resistivity; For the current joint inversion model Theoretical apparent resistivity calculated using forward modeling; For the first Standard error of class data; This is the average of all measured apparent resistivity; This represents the total number of grid cells. Representation and grid Adjacent grid sets, and Adjacent grids and grid Logarithmic resistivity; , ; and They are the first The upper and lower limits of resistivity of the layer in which each grid is located; The number of prior geometric interfaces; To use the current joint inversion model The extracted first The actual depth of each interface; For the first The design depth of each interface; This refers to the allowable depth deviation tolerance.

2. The method for detecting leakage in a channel across the entire area based on a combination of hydro-land and electro-optic methods as described in claim 1, characterized in that, Based on a three-dimensional resistivity distribution model, the degree and type of leakage are determined by calculating the leakage index and combining it with anomalous spatial strata. Specifically: The three-dimensional resistivity distribution model is discretized into a regular grid, and the relative resistivity difference is calculated for each grid. Based on the connectivity correction factor and the corresponding relative resistivity difference of each grid, the local leakage index of the grid and the overall leakage index of all grids are calculated; wherein, the value of the connectivity correction factor is determined according to whether the relative resistivity difference of at least three consecutive grids is greater than a set relative resistivity difference threshold. Based on the local leakage index and the overall leakage index, the corresponding preset thresholds are compared to determine the channel penetration level, and the leakage type is determined based on the abnormal spatial layer.

3. A method for detecting leakage in a channel across the entire area based on a combination of hydro-land and electro-optic methods, as described in any one of claims 1-2, characterized in that... Also includes: After each iteration of the joint inversion model, an inspection is performed, specifically: the apparent resistivity is checked grid by grid to see if it exceeds the prior interval. If it does, it is forcibly corrected to the interval boundary; the interface depth of each layer is extracted. If the deviation from the design depth exceeds the allowable depth deviation tolerance, it is adjusted to the design depth direction in the prior structural cognitive model.

4. The method for detecting leakage in a channel across the entire area based on a combination of hydro- and land-based electro-optical methods as described in claim 2, characterized in that, The overall leakage index is: in, The overall leakage index is N; N is the total number of grid cells. For the first The connectivity correction factor for each grid; For the first The relative resistivity difference of each grid.

5. A channel full-area leakage detection system based on a combination of water-land-electric methods, employing the channel full-area leakage detection method based on any one of claims 1-4, characterized in that, include: The first module is configured to: use the high-density resistivity method of water motion to acquire apparent resistivity data along multiple longitudinal survey lines of the channel, quantitatively assess the possibility of leakage at each survey point on the survey line, and determine the suspected leakage sections at the bottom of the channel and the suspected leakage sections at the bank of the channel. The second module is configured to: detect suspected seepage sections of the canal embankment using the static high-density resistivity method on the water surface and the high-density resistivity method on the ground, and detect suspected seepage sections of the canal bottom using the static high-density resistivity method on the water surface, and obtain the corresponding apparent resistivity data. The third module is configured to: input the apparent resistivity data obtained by the water surface motion high-density resistivity method, the water surface static high-density resistivity method and the ground high-density resistivity method into the joint inversion model, construct the objective function of the joint inversion model based on the prior structural constraints, and solve the objective function to obtain the three-dimensional resistivity distribution model. The leakage determination module is configured to: determine the degree and type of leakage by calculating the leakage index based on a three-dimensional resistivity distribution model and combining it with the abnormal spatial layers.

6. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the method according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, perform the method described in any one of claims 1-4.

8. A computer program product, characterized in that, Includes a computer program, which, when executed by a processor, implements the method described in any one of claims 1-4.

Images