Nanometric conductive coating in cooperation with an underwater detection device for the detection of a barrier

Abstract

The application discloses a nano-conductive coating and underwater detection device cooperative anti-seepage layer detection system and belongs to the technical field of underwater anti-seepage layer detection. The system is characterized in that a flexible nano-conductive coating is coated on the water-facing surface of the anti-seepage layer, and a distributed resistance sensing network is constructed. An underwater unmanned detection device is provided with a four-electrode excitation and collection module, and the surface of the coating is measured in terms of resistivity according to a preset scanning path. Acoustic positioning data and attitude inertial data are synchronously acquired, a resistivity measurement sequence with positioning marks is generated through space coordinate mapping and probe attitude drift compensation, a resistivity space distribution field is reconstructed, and an abnormal area is extracted, and finally, a seepage risk area distribution map is determined. The system can complete the positioning and identification of anti-seepage layer damage hidden dangers in a complex underwater environment, and improve the spatial resolution and data reliability of detection.

Description

Technical Field

[0001] This invention relates to the field of underwater seepage prevention layer detection technology, specifically to a seepage prevention layer detection system that combines a nano-conductive coating with an underwater detection device. Background Technology

[0002] After long-term operation, the impermeable layers in water conservancy projects such as reservoirs, dams, water conveyance tunnels, and water storage ponds may crack, detach, or deteriorate, forming hidden leakage channels. Traditional methods for inspecting the integrity of impermeable layers mainly rely on visual inspection by divers or single-point measurements using handheld probes. Visual inspection struggles to detect micro-cracks and damage that hasn't penetrated the surface, while handheld probe measurements can only obtain resistivity or potential values ​​at discrete points and are easily affected by differences in operator experience and low underwater visibility, making it difficult to guarantee detection coverage and consistency. Furthermore, while fixed electrode arrays can achieve continuous monitoring over a certain area, the installation process requires pre-embedding or drilling into the impermeable layer, which is not only difficult to construct but may also cause new damage to the impermeable layer due to the installation itself. Additionally, the spatial resolution of fixed arrays is limited by the number and spacing of electrodes, making it unable to flexibly adapt to impermeable layer surfaces with different curvatures and areas.

[0003] Underwater unmanned detection devices, coupled with electrical measurement probes, are increasingly used for underwater structure inspection. However, when scanning large-area impermeable layers, the underwater positioning accuracy of the device and changes in probe attitude can introduce significant errors into resistivity measurements. Acoustic positioning is susceptible to multipath effects and water temperature stratification, easily leading to coordinate shifts. When the device hovers at a stationary point, water flow disturbances and buoyancy changes cause the contact angle between the probe and the coating surface to continuously change, resulting in the effective electrode spacing deviating from the nominal value and causing drift in the resistivity calculation results. If these errors cannot be effectively compensated for, the boundaries of resistivity anomaly areas will become blurred, making it difficult to distinguish between actual leakage risk and measurement noise.

[0004] To address the aforementioned issues, it is necessary to solve how to achieve high-density measurement of the resistivity distribution of a large-area impermeable layer under complex underwater working conditions, and to accurately locate areas of abnormal resistivity through joint correction of spatial coordinates and attitude drift in order to identify leakage risks. Summary of the Invention

[0005] This invention provides a seepage prevention layer detection system that combines a nano-conductive coating with an underwater detection device. The purpose is to solve the problems of sparse measurement points, insufficient spatial resolution, and misjudgment of abnormal areas caused by positioning and attitude errors in existing underwater seepage prevention layer detection methods, and to achieve high-density resistivity distribution measurement and accurate determination of leakage risk areas.

[0006] To achieve the above objectives, the present invention provides the following technical solution: The present invention provides a seepage prevention layer detection system that combines a nano-conductive coating with an underwater detection device. The system includes a coating sensing module, a detection and acquisition module, a correction and mapping module, a field inversion module, and a risk identification module. The coating sensing module is used to coat the water-facing surface of the impermeable layer with a flexible nano-conductive coating and construct a distributed resistance sensing network based on the flexible nano-conductive coating. The detection and acquisition module is used to measure the resistivity of the surface of the flexible nano-conductive coating according to a preset scanning path through the four-electrode excitation and acquisition module mounted on the underwater unmanned detection device, and simultaneously acquire the acoustic positioning data and attitude inertial data of the underwater unmanned detection device. The correction and mapping module is used to perform spatial coordinate mapping and probe attitude drift compensation on the measured resistivity data according to the acoustic positioning data and the attitude inertial data, and generate a resistivity measurement sequence with positioning marks. The field inversion module is used to construct the resistivity spatial distribution field of the impermeable layer surface according to the resistivity measurement sequence with positioning marks, and extract the resistivity anomaly region in the resistivity spatial distribution field. The risk identification module is used to determine the leakage risk area distribution map of the impermeable layer according to the spatial location and resistivity change gradient of the resistivity anomaly region. This system works in conjunction with a flexible nano-conductive coating and an underwater unmanned detection device to achieve highly sensitive detection and precise positioning of micro-damage to the impermeable layer in complex underwater environments such as deep water and turbidity. It is highly automated and is not limited by underwater visibility.

[0007] As a preferred embodiment of the present invention, the process of constructing a distributed resistance sensing network based on the flexible nano-conductive coating includes: dividing the flexible nano-conductive coating into multiple independent resistance sensing units according to the geometric dimensions and surface curvature of the impermeable layer; connecting each resistance sensing unit to a corresponding measurement lead by embedding conductive busbars at the boundary of each resistance sensing unit; and converging all the measurement leads to a multi-channel switching switch installed on an underwater unmanned detection device to form a distributed resistance sensing network in which any resistance sensing unit can be selected by the multi-channel switching switch. Preferably, when the multi-channel switching switch selects the current resistance sensing unit for measurement, the unselected resistance sensing units are connected to a high-impedance state through the multi-channel switching switch to avoid current crosstalk between adjacent resistance sensing units. By selecting the measurement in a partitioned manner, the test current is concentrated in the target area, and signal crosstalk caused by parallel paths is effectively suppressed, significantly improving the spatial resolution and signal-to-noise ratio of resistivity measurement.

[0008] As a further preferred embodiment of the present invention, the process of measuring the resistivity of the surface of the flexible nano-conductive coating by means of the four-electrode excitation and acquisition module mounted on the underwater unmanned detection device according to a preset scanning path includes: generating a gridded scanning path covering the entire surface of the flexible nano-conductive coating based on a three-dimensional digital model of the impermeable layer surface, wherein the gridded scanning path includes multiple measurement stagnation points and the movement trajectory between adjacent measurement stagnation points; controlling the underwater unmanned detection device to move along the movement trajectory and stopping at each measurement stagnation point; at each measurement stagnation point, applying a constant current to the two outer electrodes through the four-electrode excitation and acquisition module, while simultaneously measuring the potential difference between the two inner electrodes, and calculating the surface resistivity value at the current measurement stagnation point based on the constant current and the potential difference. Simultaneously, the process of acquiring acoustic positioning data and attitude inertial data includes: each time the underwater unmanned probe arrives at the measurement station, triggering the acoustic positioning beacon on the underwater unmanned probe to emit an acoustic positioning signal, and receiving the time difference data of the acoustic positioning signal received by the surface array; calculating the three-dimensional spatial coordinates of the underwater unmanned probe at the current measurement station based on the time difference data; simultaneously activating the inertial measurement unit on the underwater unmanned probe to collect the three-axis acceleration and three-axis angular velocity data of the underwater unmanned probe at the current measurement station; performing attitude calculation on the three-axis acceleration and three-axis angular velocity data to obtain the roll angle, pitch angle, and yaw angle of the underwater unmanned probe at the current measurement station. This scheme synchronously triggers underwater positioning, attitude sensing, and resistivity measurement at the measurement station, ensuring that each resistivity measurement value obtains accurate spatial coordinates and probe attitude information, providing a reliable data foundation for subsequent high-precision spatial correction.

[0009] Preferably, the process of performing spatial coordinate mapping and probe attitude drift compensation on the measured resistivity data to generate a resistivity measurement sequence with positioning marks includes: assigning the surface resistivity value measured at the current measurement station to the corresponding spatial coordinate position according to the three-dimensional spatial coordinates, generating initial resistivity coordinate mapping data; calculating the actual contact angle of the probe of the four-electrode excitation and acquisition module relative to the surface of the impermeable layer according to the roll angle, the pitch angle, and the yaw angle; applying electrode spacing projection correction to the surface resistivity value in the initial resistivity coordinate mapping data according to the actual contact angle, obtaining attitude-compensated resistivity coordinate mapping data; arranging the attitude-compensated resistivity coordinate mapping data in the order of measurement time to generate a resistivity measurement sequence with positioning marks containing spatial coordinates and compensated resistivity values. Through probe attitude drift compensation, the electrode spacing variation error caused by the swaying or wall-attaching attitude deviation of the underwater unmanned detection device is effectively eliminated, making the resistivity data more accurately reflect the conductivity characteristics of the impermeable layer surface.

[0010] For the field inversion module, the technical solution adopted in this invention is as follows: All measurement points in the resistivity measurement sequence with positioning marks are mapped onto a two-dimensional parameter plane on the surface of the impermeable layer according to spatial coordinates, resulting in a discrete set of measurement points. A radial basis function interpolation algorithm is used, with the discrete set of measurement points as interpolation nodes, to perform spatial interpolation calculations on the resistivity of the impermeable layer surface. The two-dimensional parameter plane is uniformly divided into multiple grid cells, and the radial basis function interpolation algorithm is used to calculate the resistivity interpolation at the center point of each grid cell. The resistivity interpolation of all grid cells constitutes the spatial distribution field of the resistivity on the impermeable layer surface. Radial basis function interpolation can fully adapt to the distribution characteristics of irregular measurement points on the impermeable layer surface, generating a smooth and continuous spatial distribution field, which facilitates the accurate identification of subsequent abnormal areas.

[0011] In extracting resistivity anomaly regions, the preferred embodiment of the present invention is as follows: Calculate the statistical median and absolute median difference of resistivity values ​​for all grid cells in the resistivity spatial distribution field; based on the statistical median and the absolute median difference, set a lower threshold and an upper threshold for resistivity anomaly determination; traverse each grid cell in the resistivity spatial distribution field, marking grid cells with resistivity values ​​below the lower threshold or above the upper threshold as anomalous grid cells; perform connected component clustering analysis on the anomalous grid cells, merging multiple spatially adjacent anomalous grid cells into the same resistivity anomaly region. Based on the adaptive threshold setting of the statistical median and absolute median difference, the influence of overall resistivity drift can be effectively eliminated, accurately separating the true resistivity anomaly regions caused by damage to the impermeable layer or changes in water content.

[0012] Further, the process of determining the leakage risk area distribution map of the impermeable layer based on the spatial location and resistivity change gradient of the resistivity anomaly region includes: calculating the geometric center coordinates and region boundary contour of each resistivity anomaly region; determining the spatial coverage of each resistivity anomaly region on the surface of the impermeable layer based on the geometric center coordinates and the region boundary contour; calculating the resistivity difference between adjacent grid cells in each grid cell within the resistivity anomaly region along a direction perpendicular to the region boundary contour; determining the radial resistivity change gradient of the resistivity anomaly region based on the resistivity difference; dividing the resistivity anomaly region into low-risk leakage areas, medium-risk leakage areas, and high-risk leakage areas based on the spatial coverage and the radial resistivity change gradient; and constructing the leakage risk area distribution map of the impermeable layer from all the divided resistivity anomaly regions. Specifically, for each resistivity anomaly region, the area value of its spatial coverage and the maximum gradient value of its radial resistivity change gradient are extracted. The area value is compared with a preset area threshold, and the maximum gradient value is also compared with a preset gradient threshold. If the area value is less than the area threshold and the maximum gradient value is less than the gradient threshold, the resistivity anomaly region is classified as a low-risk leakage region. If the area value is greater than or equal to the area threshold and the maximum gradient value is less than the gradient threshold, or if the area value is less than the area threshold and the maximum gradient value is greater than or equal to the gradient threshold, the resistivity anomaly region is classified as a medium-risk leakage region. If the area value is greater than or equal to the area threshold and the maximum gradient value is greater than or equal to the gradient threshold, the resistivity anomaly region is classified as a high-risk leakage region. This risk classification strategy comprehensively considers the spatial scale of the anomaly region and the severity of resistivity changes, effectively distinguishing between localized minor defects and extended leakage damage, providing a quantitative basis for the maintenance decision of the seepage prevention layer.

[0013] This invention organically integrates distributed sensing of flexible nano-conductive coating with acoustic positioning and inertial navigation technology of underwater unmanned detection devices, realizing full-coverage scanning of the resistivity distribution of the seepage prevention layer surface, adaptive anomaly identification, and leakage risk classification output. It solves the problems of inaccurate positioning, large interference from the environment, and inability to quantify risks in traditional underwater detection methods, and significantly improves the reliability and engineering practicality of health monitoring of seepage prevention structures in water conservancy projects.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows: A distributed resistance sensing network is constructed on the water-facing side of the geomembrane using a flexible nano-conductive coating. Each resistance sensing unit has a measurement lead extending from a conductive busbar and converging at a multi-channel switching switch, forming a measurement architecture accessible via selection logic. This method eliminates the need for pre-embedded electrodes or mechanical drilling in the geomembrane itself, avoiding secondary damage to the geomembrane during installation. Furthermore, the flexible coating can conform to geomembrane surfaces with varying curvatures, allowing the resistance sensing network to cover the entire area under test. Compared to fixed array electrodes, this significantly improves spatial sampling density and deployment flexibility. When the multi-channel switching switch selects a current resistance sensing unit, it places unselected units in a high-impedance state, cutting off current bypass between adjacent units, eliminating measurement errors caused by crosstalk, and ensuring the independence of resistivity data for each frame.

[0015] The underwater unmanned detection device travels along a gridded scanning path and stops at measurement stagnation points. A constant current is applied via a four-electrode excitation and acquisition module, and the potential difference is measured to obtain the surface resistivity value at the corresponding stagnation point. Simultaneously, acoustic positioning data and triaxial acceleration and angular velocity data output from the inertial measurement unit are acquired to calculate the three-dimensional spatial coordinates, roll angle, pitch angle, and yaw angle. The resistivity values ​​are assigned to the corresponding spatial positions on the impermeable layer surface using the three-dimensional spatial coordinates. Then, based on the attitude angles, the actual contact angle of the four-electrode probes relative to the impermeable layer surface is calculated. Electrode spacing projection corrections are applied to the resistivity values ​​to obtain attitude-compensated resistivity coordinate mapping data. This process eliminates resistivity calculation deviations caused by acoustic positioning errors and probe attitude drift, resulting in more accurate spatial markings for each measurement point. The resistivity values ​​reflect the true coating condition, and the generated resistivity measurement sequence with positioning marks provides reliable input for high-resolution field inversion. In the resistivity spatial distribution field, the abnormal threshold is dynamically set by statistical median and absolute median difference. Abnormal grid cells are clustered by connected domain and abnormal regions are extracted. Risk classification is carried out by combining area and radial resistivity change gradient. The resulting leakage risk area distribution map can directly reflect the severity of hidden dangers at different locations of the seepage prevention layer without relying on human experience judgment. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a schematic diagram of a seepage prevention layer detection system based on a nano-conductive coating and an underwater detection device. Figure 2 This is a flowchart of the surface resistivity gridded scanning measurement of flexible nano-conductive coatings; Figure 3 This is a flowchart of the acoustic positioning and attitude data synchronous acquisition and resistivity correction mapping process; Figure 4 This is a gridded scanning path diagram of the surface of the flexible nano-conductive coating of an underwater unmanned detection device; Figure 5 This is a comparison chart of surface resistivity before and after attitude compensation; Figure 6 This is a schematic diagram of the spatial distribution interpolation and abnormal threshold of the surface resistivity of the seepage barrier layer. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0019] See Figure 1 This invention provides a seepage-proof layer detection system that integrates a nano-conductive coating with an underwater detection device. The system includes a coating sensing module, a detection and acquisition module, a correction and mapping module, a field inversion module, and a risk identification module. The coating sensing module coats the water-facing side of the seepage-proof layer with a flexible nano-conductive coating and constructs a distributed resistance sensing network based on the coating. The detection and acquisition module uses a four-electrode excitation and acquisition module mounted on an underwater unmanned detection device to measure the resistivity of the flexible nano-conductive coating surface according to a preset scanning path, and simultaneously acquires the acoustic positioning data and attitude inertial data of the underwater unmanned detection device. The correction and mapping module performs spatial coordinate mapping and probe attitude drift compensation on the measured resistivity data based on the acoustic positioning data and attitude inertial data, generating a resistivity measurement sequence with positioning marks. The field inversion module constructs a resistivity spatial distribution field on the seepage-proof layer surface based on the resistivity measurement sequence with positioning marks and extracts resistivity anomaly regions from the resistivity spatial distribution field. The risk identification module determines the leakage risk area distribution map of the seepage-proof layer based on the spatial location and resistivity change gradient of the anomaly regions.

[0020] The specific implementation of the system will be described below with reference to several embodiments.

[0021] In practical implementation, the process of constructing a distributed resistance sensing network based on a flexible nano-conductive coating in the coating sensing module is as follows: First, the geometric dimensions and surface curvature of the impermeable layer are obtained. Using a pre-established 3D digital model of the impermeable layer, the surface contour data of the water-facing side is extracted. This surface contour data includes the surface coordinates and elevation values ​​at each discrete point on the water-facing side of the impermeable layer. Based on the surface contour data of the water-facing side of the impermeable layer, the surface curvature distribution at each point on the water-facing side is calculated. The surface curvature distribution includes the Gaussian curvature and the mean curvature at each point.

[0022] In one embodiment, the method for dividing the flexible nano-conductive coating into multiple independent resistance sensing units based on the geometry and surface curvature of the impermeable layer is as follows: A curvature change threshold is set, determined based on the uniform coating curvature range allowed by the adhesion process of the flexible nano-conductive coating, and the curvature change threshold is set to 0.05 mm⁻¹. Along the surface contour of the water-facing side of the impermeable layer, a segmented scan is performed starting from the initial boundary. When the surface curvature change between two adjacent locations exceeds the curvature change threshold, a unit boundary line is drawn between those two locations. Simultaneously, considering the overall geometry of the impermeable layer, the total surface area of ​​the water-facing side is divided into multiple independent regions ranging from 0.5 m² to 2 m². Within each independent region, the surface curvature change between any two points is less than the curvature change threshold, and each independent region forms a resistance sensing unit. The boundary lines of all resistance sensing units constitute a closed unit partitioning grid.

[0023] A conductive busbar is embedded at the boundary of each resistance sensing unit. The conductive busbar is printed onto a flexible nano-conductive coating using conductive silver paste via screen printing. The busbar is 2mm wide and 0.1mm thick. The conductive busbar extends continuously along the boundary line of the resistance sensing unit, forming lead-out terminals at both ends of the unit. At the lead-out terminals of the conductive busbar corresponding to each resistance sensing unit, the measuring leads are fixedly connected to the lead-out terminals using low-temperature soldering. The measuring leads are multi-strand copper core wires with insulating sheaths. One end of each measuring lead is connected to the lead-out terminal of the conductive busbar of a resistance sensing unit, and the other end is led into the underwater unmanned detection device's cabin along the wiring harness channel.

[0024] All measurement leads converge to a multi-channel switch mounted on the underwater unmanned detection device. The number of input channels of the multi-channel switch is the same as the total number of resistance sensing units, with each input channel connecting to two measurement leads of one resistance sensing unit. The control terminal of the multi-channel switch is electrically connected to the main control unit of the underwater unmanned detection device. The main control unit sends a channel selection command, causing the analog switch array inside the multi-channel switch to connect the two measurement leads of a designated resistance sensing unit to the measurement port of the four-electrode excitation and acquisition module, thereby selecting that resistance sensing unit. For the channels corresponding to unselected resistance sensing units, the multi-channel switch switches its input terminal to a high-impedance input buffer circuit. The input impedance of this high-impedance input buffer circuit is greater than 10MΩ, thus creating a high-impedance state between the measurement lead pairs of the unselected resistance sensing units. In this high-impedance state, the current flowing through the unselected resistance sensing units is limited to the nanoamp level, effectively avoiding current crosstalk between adjacent resistance sensing units. After the above steps, a distributed resistance sensing network is formed, in which any resistance sensing unit can be selected by a multi-channel switching switch.

[0025] In a distributed resistance sensing network, the resistance value of each resistance sensing unit is determined by the volume resistivity of the flexible nano-conductive coating, the length and width of the resistance sensing unit, and the coating thickness. Taking a rectangular resistance sensing unit as an example, the resistance value between its two ends... This can be expressed by the following formula:

[0026] in, This indicates the resistance value between the conductive busbars at both ends of the resistance sensing unit, in ohms; The volume resistivity of flexible nano-conductive coating materials is expressed in ohm-meters. The values ​​were obtained by measuring and calibrating the flexible nano-conductive coating sample before coating using the four-probe method. This indicates the distance between the conductive busbars along the current conduction direction of the resistance sensing unit, in meters. The value is obtained through geometric measurement within the resistance sensing unit area on the three-dimensional digital model of the impermeable layer; This indicates the width of the resistive sensing unit perpendicular to the direction of current conduction, in meters. The value was also obtained through geometric measurement of the resistance sensing unit area on the three-dimensional digital model of the impermeable layer; This indicates the average thickness of the flexible nano-conductive coating, in meters. The value is obtained by averaging multiple measurements using a coating thickness gauge after coating is completed. The above resistance expression verifies whether each resistance sensing unit is within its normal operating range and serves as the baseline resistance value for data analysis in subsequent measurements.

[0027] In specific implementation, please refer to Figure 2 The detection and acquisition module uses the four-electrode excitation and acquisition module mounted on the underwater unmanned detection device to measure the resistivity of the surface of the flexible nano-conductive coating according to the preset scanning path, as follows.

[0028] Based on the 3D digital model of the impermeable layer surface, a rasterized scanning path covering the entire flexible nano-conductive coating surface is generated. The 3D digital model of the impermeable layer surface is reconstructed from point cloud data obtained by pre-scanning the water-facing side of the impermeable layer using multibeam sonar. The 3D digital model of the impermeable layer surface contains complete surface shape information of the area covered by the flexible nano-conductive coating. The generation process of the rasterized scanning path is as follows: all triangular facets of the flexible nano-conductive coating area in the 3D digital model of the impermeable layer surface are extracted, and all extracted triangular facets are mapped to a 2D plane coordinate system to obtain a planar unfolded diagram of the coating area. On the planar unfolded diagram of the coating area, two mutually perpendicular raster directions are defined, where the first raster direction is along the main seepage direction of the water-facing side of the impermeable layer, and the second raster direction is perpendicular to the main seepage direction. A set of parallel straight lines is generated along the first raster direction at a first preset interval, and another set of parallel straight lines is generated along the second raster direction at a second preset interval. All intersections of the two sets of parallel straight lines constitute a planar raster intersection set. The values ​​of the first and second preset intervals are both set according to the coverage width of the four electrode arrays in the four-electrode excitation and acquisition module. The first preset interval is 80% of the coverage width of the four electrode arrays, and the second preset interval is 80% of the coverage width of the four electrode arrays. Each intersection point in the planar grid intersection set is back-projected onto the corresponding surface of the three-dimensional digital model of the impermeable layer surface to obtain the measurement stationary point in three-dimensional space. All measurement stationary points are connected sequentially in the order of first along the first grid direction and then along the second grid direction to generate a continuous polyline. The line segment between two adjacent measurement stationary points on the polyline is defined as the movement trajectory. All measurement stationary points and all movement trajectories together constitute the rasterized scanning path. The rasterized scanning path is stored in the form of a path file, which records the three-dimensional spatial coordinate sequence number, three-dimensional spatial coordinate value, and movement direction vector from the current measurement stationary point to the next measurement stationary point for each measurement stationary point.

[0029] After generating the rasterized scanning path, the control system of the underwater unmanned probe loads the path file and executes path-following control. The underwater unmanned probe is equipped with a four-electrode excitation and acquisition module, a thruster assembly, and an underwater inertial navigation system. Based on the current 3D position output by the underwater inertial navigation system, the control system calculates the position deviation vector between the current 3D position and the 3D spatial coordinates of the first measurement station point in the rasterized scanning path. Based on the position deviation vector, it generates thrust commands for each thruster in the thruster assembly, driving the underwater unmanned probe towards the first measurement station point. When the underwater unmanned probe reaches a preset distance threshold range from the 3D spatial coordinates of the first measurement station point, the thruster assembly enters position-holding mode, and the underwater unmanned probe stops moving at that measurement station point. The preset distance threshold is set to 0.05 meters.

[0030] After the underwater unmanned detection device stops moving at the measurement station, the four-electrode excitation and acquisition module performs resistivity measurement. The four-electrode excitation and acquisition module includes a retractable probe bracket with four electrodes mounted on its end face. The four electrodes are arranged in a straight line at equal intervals, with a fixed distance between adjacent electrodes. The probe bracket extends under the action of a first drive motor, pressing the tips of the four electrodes into the surface of the flexible nano-conductive coating, establishing reliable electrical contact between the four electrodes and the surface of the flexible nano-conductive coating. The excitation circuit inside the four-electrode excitation and acquisition module applies a constant current between the two outer electrodes. The amplitude of the constant current is set according to the resistance range of the flexible nano-conductive coating, and is set to 10 mA. Simultaneously, the acquisition circuit inside the four-electrode excitation and acquisition module measures the potential difference generated between the two inner electrodes. Based on the constant current and potential difference, the surface resistivity value at the current measurement station is calculated. The formula for calculating the surface resistivity value is:

[0031] in, This represents the surface resistivity value measured at the current measurement stagnation point, in ohm-meters. This indicates the distance between any two adjacent electrodes out of the four electrodes, in meters. The values ​​were obtained by directly measuring the electrode array on the end face of the probe holder on a coordinate measuring machine. The value is 0.02 meters; This represents the potential difference between the two inner electrodes, measured in volts. The value is obtained by sampling the potential difference signal through the analog-to-digital converter inside the four-electrode excitation and acquisition module; This represents the constant current value applied to the two outer electrodes, measured in amperes. The value is set by the current source inside the four-electrode excitation and acquisition module. The value is 0.01 amperes.

[0032] After the surface resistivity value is calculated, the probe bracket of the four-electrode excitation and acquisition module retracts under the action of the second drive motor, causing the four electrodes to detach from the surface of the flexible nano-conductive coating. The control system of the underwater unmanned detection device reads the three-dimensional spatial coordinates of the next measurement station in the path file, and moves to the next measurement station along the movement trajectory according to the aforementioned position control method. It repeats the steps of stopping movement, extending the probe, applying a constant current, measuring the potential difference, calculating the surface resistivity value, and retracting the probe until all measurement stations recorded in the rasterized scanning path have been traversed.

[0033] See Figure 4 The figure shows the two-dimensional planar distribution of the rasterized scanning path generated based on the three-dimensional digital model of the impermeable layer in one of the above embodiments. The horizontal axis represents the plane X coordinate (unit: meters), and the vertical axis represents the plane Y coordinate (unit: meters). In the legend, blue dots represent measurement station points, and black broken lines represent the movement trajectory of the underwater unmanned detection device along the preset scanning path.

[0034] As shown in the figure, the measurement points are evenly distributed within a 12m × 6m rectangular area, arranged in a regular grid. The spacing between the measurement points remains uniform, consistent with the setting in the above embodiment where both the first and second preset intervals are 80% of the coverage width of the four-electrode array. The movement trajectory follows a back-and-forth traversal pattern. First, it connects each measurement point sequentially along the X direction. After completing the resistivity measurement of one row, it moves along the Y direction to the next row and continues scanning in reverse along the X direction, sequentially covering all measurement points, ensuring comprehensive scanning coverage of the coating surface.

[0035] In specific implementation, please refer to Figure 3 The detection and acquisition module acquires the acoustic positioning data and attitude inertial data of the underwater unmanned detection device in the following way.

[0036] Each time the underwater unmanned probe reaches its measurement station, it triggers the acoustic positioning beacon on the probe to emit an acoustic positioning signal. The probe's control system sends a trigger pulse signal to the acoustic positioning beacon when it determines that the thruster assembly has entered position-holding mode and the four-electrode excitation and acquisition module has begun resistivity measurement. Upon receiving the trigger pulse signal, the acoustic positioning beacon emits an acoustic positioning pulse of a specific frequency, with a center frequency of 40 kHz and a pulse width of 2 milliseconds. The surface array consists of four hydrophones arranged in a rectangular layout on the water surface. The spatial coordinates of the four hydrophones are pre-calibrated using a differential GPS system. Each hydrophone receives the acoustic positioning pulse emitted by the beacon and records the time of receipt. The signal processing unit of the surface array subtracts the times of the acoustic positioning pulses received by the four hydrophones pairwise to obtain three independent sets of time-of-arrival (TOA) data. Based on three sets of independent time-of-arrival data and the spatial coordinates of four hydrophones, a hyperboloid intersection positioning algorithm is used to calculate the three-dimensional spatial coordinates of the acoustic positioning beacon at the time of launch. The calculated three-dimensional spatial coordinates are the three-dimensional spatial coordinates of the underwater unmanned detection device at its current measurement station. The three-dimensional spatial coordinates of the underwater unmanned detection device at its current measurement station are transmitted back to the control system of the underwater unmanned detection device via an underwater acoustic communication link and stored.

[0037] Simultaneously with triggering the acoustic positioning beacon to emit acoustic positioning signals, the control system of the underwater unmanned probe activates the inertial measurement unit (IMU) mounted on the probe to collect triaxial acceleration and triaxial angular velocity data at the current measurement station. The IMU comprises three orthogonally mounted microelectromechanical accelerometers (MEMS) and three orthogonally mounted microelectromechanical gyroscopes (MEMS). The MEMS measures triaxial acceleration data with a range of ±16g; the MEMS measures triaxial angular velocity data with a range of ±2000 degrees per second. The IMU continuously collects triaxial acceleration and angular velocity data over a period of 500 milliseconds at a sampling frequency of 200 Hz. The arithmetic mean of all triaxial acceleration and angular velocity data collected within this 500-millisecond period is then calculated to obtain an average triaxial acceleration data set and an average triaxial angular velocity data set for the current measurement station.

[0038] Attitude calculations were performed on the average triaxial acceleration and angular velocity data to obtain the roll, pitch, and yaw angles of the underwater unmanned probe at the current measurement station. A complementary filtering algorithm was used for attitude calculation, with a proportional gain of 0.02 and an integral gain of 0.001. In the complementary filtering algorithm, the initial roll and pitch angles of the underwater unmanned probe were calculated using the gravitational acceleration component in the average triaxial acceleration data; the attitude angle changes were calculated by time integration of the average triaxial angular velocity data; and the initial roll and pitch angles were complementaryly fused with the attitude angle changes to obtain the roll and pitch angles of the underwater unmanned probe at the current measurement station. The yaw angle of the underwater unmanned detection device is obtained by accumulating the heading angular velocity component in the three-axis angular velocity data obtained by integrating the yaw angle of the underwater unmanned detection device at the previous measurement station and the current measurement station. The yaw angle at the first measurement station is assigned by the initial heading of the underwater unmanned detection device calculated by the surface array.

[0039] The calibration mapping module performs spatial coordinate mapping and probe attitude drift compensation on the measured resistivity data to generate a resistivity measurement sequence with positioning markers, as follows.

[0040] Based on the three-dimensional spatial coordinates, the surface resistivity value measured at the current measurement station is assigned to the corresponding spatial coordinate position, generating initial resistivity coordinate mapping data. In the specific implementation, the calibration mapping module establishes a set of resistivity coordinate records. Each resistivity coordinate record contains a spatial coordinate field and a resistivity value field. The three-dimensional spatial coordinate value of the current measurement station is written into the spatial coordinate field, and the surface resistivity value measured at the current measurement station is written into the resistivity value field, forming an initial resistivity coordinate mapping record.

[0041] Based on the roll angle, pitch angle, and yaw angle, the actual contact angle of the probe of the four-electrode excitation and acquisition module relative to the surface of the impermeable layer is calculated. The probe of the four-electrode excitation and acquisition module is mounted on the bottom of the underwater unmanned detection device, with the probe's central axis coinciding with the normal vector of the bottom surface of the underwater unmanned detection device. The direction vector of the bottom surface normal vector of the underwater unmanned detection device in the carrier coordinate system is denoted as... , The value is A rotation matrix from the carrier coordinate system to the geodetic coordinate system is constructed using roll angle, pitch angle, and yaw angle. Transforming to the geodetic coordinate system, the probe's direction vector in the geodetic coordinate system is obtained. In the three-dimensional digital model of the impermeable layer surface, the surface normal vector at the current measurement point is acquired. This surface normal vector is obtained by weighted averaging of the normal vectors of the triangular facets in the neighborhood of the current measurement point. The angle between the probe's direction vector in the geodetic coordinate system and the surface normal vector at the current measurement point is calculated; this angle is the actual contact angle of the probe of the four-electrode excitation and acquisition module relative to the impermeable layer surface.

[0042] Based on the actual contact tilt angle, the surface resistivity values ​​in the initial resistivity coordinate mapping data are corrected by electrode spacing projection to obtain the attitude-compensated resistivity coordinate mapping data. The electrode spacing projection correction method is as follows: the nominal spacing of the four electrodes is... Projecting this onto the normal plane containing the actual contact angle yields the projected equivalent electrode spacing. The calculation formula is:

[0043] in, This represents the equivalent electrode spacing after projection, in meters. This refers to the nominal distance between any two adjacent electrodes out of the four electrodes, in meters. The value is 0.02 meters; This indicates the actual contact angle of the probe of the four-electrode excitation and acquisition module relative to the surface of the impermeable layer, expressed in radians. The value is calculated from the angle between the aforementioned direction vectors. The equivalent electrode spacing after projection is used. Replace the surface resistivity calculation formula The surface resistivity value is recalculated to obtain the attitude-compensated surface resistivity value. The surface resistivity value in the initial resistivity coordinate mapping record is updated with the attitude-compensated surface resistivity value to obtain the attitude-compensated resistivity coordinate mapping record.

[0044] The resistivity coordinate mapping records after attitude compensation are arranged in chronological order of measurement to generate a resistivity measurement sequence with location markers, containing spatial coordinates and compensated resistivity values. The resistivity measurement sequence with location markers is stored in a linear list structure. Each element in the linear list contains the three-dimensional spatial coordinates of a measurement station, the attitude-compensated surface resistivity value, and a measurement timestamp.

[0045] See Figure 5 In the figure, the horizontal axis represents the measurement station number, ranging from 0 to 600, and the vertical axis represents the surface resistivity value, in ohm-meters, ranging from approximately 110 to 185. The figure contains two curves: the orange solid line represents the surface resistivity value before attitude compensation, and the blue dashed line represents the surface resistivity value after attitude compensation.

[0046] The curve trends show that the surface resistivity before attitude compensation fluctuates significantly, especially in the measurement station numbers approximately 200-300 and 400-500, where obvious peak-to-trough variations occur. The highest peak value approaches 185 ohm-meters, while the lowest valley value drops to approximately 115 ohm-meters, exhibiting a wide fluctuation range. In contrast, the surface resistivity curve after attitude compensation shows a significantly reduced fluctuation range and a more stable overall trend. The resistivity values ​​are mainly concentrated in the 140-160 ohm-meter range, and there are no obvious abnormal peaks or sudden drops throughout the entire measurement station range.

[0047] Upon closer inspection, the resistivity curve after attitude compensation exhibits periodic fluctuations in multiple stagnation point regions, and these fluctuation trends partially overlap with the curve before attitude compensation, but the amplitude is smaller. This indicates that attitude compensation effectively eliminates measurement errors and signal drift caused by changes in the attitude of the underwater unmanned detection device, while maintaining the spatial distribution characteristics of resistivity.

[0048] In practice, the process by which the field inversion module constructs the spatial distribution field of resistivity on the surface of the impermeable layer based on the resistivity measurement sequence with positioning marks is as follows.

[0049] All measurement points in the resistivity measurement sequence with positioning marks are mapped to a two-dimensional parameter plane on the surface of the impermeable layer according to their spatial coordinates, resulting in a discrete measurement point set. The field inversion module extracts the three-dimensional spatial coordinates and attitude-compensated surface resistivity values ​​of each measurement point from the resistivity measurement sequence with positioning marks. The three-dimensional digital model of the impermeable layer surface has a pre-established isoparametric mapping relationship between the three-dimensional surface of the impermeable layer and the two-dimensional parameter plane. The two-dimensional parameter plane is a normalized parameter domain, and the two coordinate axes of the parameter domain range from 0 to 1. For the three-dimensional spatial coordinates of each measurement point, the field inversion module finds the surface triangular facet closest to that three-dimensional spatial coordinate value in the three-dimensional digital model of the impermeable layer surface, obtains the three-dimensional spatial coordinates of the three vertices of the surface triangular facet and the corresponding two-dimensional parameter coordinates, and calculates the two-dimensional parameter coordinates corresponding to the measurement point using the centroid coordinate interpolation method. The attitude-compensated surface resistivity value of each measurement point is combined with the calculated two-dimensional parameter coordinates to form a two-dimensional parameter point with a resistivity value. All two-dimensional parameter points with resistivity values ​​constitute the discrete measurement point set.

[0050] A radial basis function (RBF) interpolation algorithm is employed, using a discrete set of measurement points as interpolation nodes to perform spatial interpolation calculations on the resistivity of the impermeable layer surface. The kernel function selected for the RBF interpolation algorithm is the Gaussian kernel function, which has the following form: ,in Represents the radial basis function values. This represents the Euclidean distance between two two-dimensional parameter points. This represents the shape parameter of the Gaussian kernel function. The value is determined based on the reciprocal of the minimum Euclidean distance between all measurement points in the discrete measurement point set, specifically: ,in This represents the minimum Euclidean distance between all measurement points in the discrete measurement point set. The numerical values ​​are obtained by traversing all point pairs in the discrete measurement point set and calculating the minimum Euclidean distance. The field inversion module constructs a system of linear equations with the same number of measurement points as the discrete measurement point set. The coefficient matrix of the linear equations consists of the kernel function values ​​of the Gaussian kernel function between each interpolation node, and the right-hand side term vector consists of the surface resistivity values ​​of each interpolation node. Solving this system of linear equations yields the weight coefficients corresponding to each interpolation node.

[0051] The two-dimensional parametric plane is uniformly divided into multiple grid cells. Each grid cell has a uniform number of subdivisions along both coordinate axes of the two-dimensional parametric plane: 200 subdivisions along the first coordinate axis and 200 subdivisions along the second coordinate axis. The center point coordinates of each grid cell are obtained by equally dividing the normalized coordinates of the two-dimensional parametric plane. The resistivity interpolation at the center point of each grid cell is calculated using a radial basis function interpolation algorithm. For each grid cell center point, the field inversion module calculates the Euclidean distance between the center point and all interpolation nodes. These Euclidean distances are then substituted into a Gaussian kernel function to obtain a set of kernel function values. This set of kernel function values ​​is then weighted and summed with the weighting coefficients corresponding to each interpolation node to obtain the resistivity interpolation at the center point of the grid cell. The resistivity interpolations of all grid cells are arranged according to their row and column positions on the two-dimensional parametric plane, forming the resistivity spatial distribution field of the impermeable layer surface. The resistivity spatial distribution field is stored in the form of a two-dimensional array, where each element of the array is the resistivity interpolation of a grid cell.

[0052] The process of extracting resistivity anomaly regions in the resistivity spatial distribution field by the field inversion module is as follows.

[0053] Calculate the statistical median and absolute median difference of resistivity values ​​for all grid cells in the resistivity spatial distribution field. The field inversion module traverses the two-dimensional array corresponding to the resistivity spatial distribution field, extracts the resistivity interpolation values ​​for all grid cells, sorts these resistivity interpolation values ​​according to their numerical values, and takes the median value of the sorted sequence as the statistical median. If the total number of grid cells is even, the average of the two median values ​​is taken. The statistical median is denoted as . The absolute median difference is calculated as follows: for each grid cell, calculate its resistivity interpolation and statistical median. The absolute value of the difference is used to obtain the absolute deviation value of each grid cell. Then, the statistical median of the absolute deviation values ​​of all grid cells is calculated to obtain the absolute median difference. The calculation formula is:

[0054] in, This represents the absolute median difference, expressed in ohms-meters. Represents the resistivity interpolation of the grid cell located in the m-th row and n-th column in the resistivity spatial distribution field, in ohm-meters; This represents the statistical median of the resistivity interpolation of all grid cells in the resistivity spatial distribution field, in ohm-meters; symbol This indicates the operation of taking the median.

[0055] Based on the statistical median and absolute median difference, lower and upper thresholds are set for resistivity anomaly detection. Lower threshold Take as Subtract three times Upper limit threshold Take as Add three times The coefficient three is set based on the fact that when resistivity data follows an approximately normal distribution, the three-times absolute median difference criterion can detect extreme values ​​that deviate from the median by more than four and a half times the standard deviation, thus meeting the sensitivity and false positive rate balance requirements for detecting local anomalies in the resistivity of flexible nano-conductive coating surfaces.

[0056] Iterate through each grid cell in the resistivity spatial distribution field, interpolating the resistivity below a lower threshold. or above the upper limit threshold The grid cells are marked as anomalous grid cells. The field inversion module sets a Boolean flag value for each grid cell in the resistivity spatial distribution field, initially all Boolean flag values ​​are set to false. When the resistivity interpolation of a grid cell meets the condition of being below the lower threshold or above the upper threshold, the Boolean flag value of that grid cell is set to true, indicating that the grid cell is an anomalous grid cell.

[0057] Connectivity clustering analysis is performed on anomalous grid cells to merge multiple spatially adjacent anomalous grid cells into a single resistivity anomaly region. The connectivity clustering analysis employs an eight-connected region labeling algorithm. The field inversion module traverses each grid cell in the resistivity spatial distribution field in row-major order. When a grid cell with a Boolean label value of true and no region number has been assigned, a new region number is assigned to that grid cell. Using this grid cell as the seed point, eight-neighbor region growth is performed, assigning the same region number to all grid cells with eight-neighbor connections and all Boolean label values ​​of true. After traversal, grid cells with the same region number form a connected component, and each connected component is a resistivity anomaly region. Each resistivity anomaly region is identified by its region number, and stores the row and column positions of all grid cells contained within the region, along with their corresponding resistivity interpolations.

[0058] See Figure 6 In the figure, the vertical axis represents the resistivity interpolation in ohm-meters, and the horizontal axis represents the arc length along the surface of the impermeable layer in meters. The figure shows the resistivity interpolation curve as a function of arc length, plotted as a solid black line. The blue and red dashed lines represent the lower and upper thresholds for resistivity anomaly detection, respectively. This data corresponds to the results of the field inversion module in the above embodiment for constructing the spatial distribution field of resistivity and extracting anomaly regions from the resistivity measurement sequence with positioning markers.

[0059] The overall resistivity value fluctuates between 140 and 150 ohm-meters, falling between the upper and lower thresholds, indicating that the resistivity of most areas of the geomembrane is within the normal range. In the arc lengths of approximately 9.5 to 12 meters and 20 to 23 meters, the resistivity interpolation shows a significant decrease, with the lowest value approaching 115 ohm-meters, below the lower threshold, indicating the presence of resistivity anomalies on the geomembrane surface in these two ranges. Within these anomalies, the resistivity value exhibits two distinct troughs with arc length, concentrated at approximately 10 meters and 20 meters respectively, suggesting that this anomaly region may consist of two relatively independent resistivity anomaly sub-regions.

[0060] The upper and lower thresholds in this figure are set based on the statistical median and absolute median difference of the resistivity interpolation of all grid cells in the resistivity spatial distribution field, and are determined using the three-times absolute median difference criterion to ensure a balance between the sensitivity and false positive rate of anomaly detection. Regions where the curve is significantly lower than the lower threshold are resistivity anomaly regions, conforming to the anomaly grid cell detection rules in the above embodiment.

[0061] In practical implementation, the risk identification module determines the distribution map of leakage risk areas of the anti-seepage layer based on the spatial location of the resistivity anomaly area and the resistivity change gradient as follows.

[0062] The geometric center coordinates and boundary contour of each resistivity anomaly region are calculated. The risk identification module traverses each resistivity anomaly region obtained through connected component clustering analysis in the above embodiment, acquiring the row and column positions of all grid cells contained within that region. For the currently processed resistivity anomaly region, the arithmetic mean of the coordinates of the center points of all grid cells on the two-dimensional parameter plane is calculated to obtain the coordinates of the geometric center on the two-dimensional parameter plane. Then, through the mapping relationship between the two-dimensional parameter plane and the three-dimensional surface established in the three-dimensional digital model of the impermeable layer surface, the two-dimensional parameter coordinates of the geometric center are converted into three-dimensional spatial coordinates. These three-dimensional spatial coordinates are the geometric center coordinates of the resistivity anomaly region. The boundary contour is determined as follows: within the currently processed resistivity anomaly region, the risk identification module extracts all grid cells located at the region edge. The criterion for determining an edge grid cell is that at least one grid cell in its eight neighborhoods does not belong to the same resistivity anomaly region. The center points of the extracted edge mesh cells are connected in a clockwise or counterclockwise order to form a closed polygonal polyline. The coordinate sequence of this closed polygonal polyline on the two-dimensional parameter plane constitutes the regional boundary profile of the resistivity anomaly region. The two-dimensional parameter coordinate sequence of the regional boundary profile is also converted into a three-dimensional spatial coordinate sequence through a mapping relationship, obtaining the three-dimensional boundary profile of the resistivity anomaly region on the surface of the impermeable layer. Based on the geometric center coordinates and the three-dimensional spatial coordinate sequence of the regional boundary profile, the spatial coverage of each resistivity anomaly region on the surface of the impermeable layer is determined. The spatial coverage is represented by the curved surface region enclosed by the regional boundary profile.

[0063] For each grid cell within a resistivity anomaly region, the resistivity difference between adjacent grid cells is calculated along a direction perpendicular to the region's boundary contour. The radial resistivity gradient of the anomaly region is then determined based on this resistivity difference. In the specific implementation, the risk identification module, for the currently processed resistivity anomaly region, starts from each edge grid cell on the region's boundary contour and traces the grid cells layer by layer into the interior of the anomaly region along the inner normal direction of the boundary contour at the edge grid cell's location, forming several radial tracing paths. For each radial tracing path, starting from the first grid cell at the region boundary, the resistivity interpolation difference between two adjacent grid cells on the path is calculated sequentially. This difference is divided by the spatial distance between the center points of these two adjacent grid cells to obtain the local resistivity gradient between them. The spatial distance is obtained by converting the coordinates of the center points of the two adjacent grid cells on the two-dimensional parameter plane to three-dimensional spatial coordinates and then calculating the Euclidean distance. The maximum absolute value of all local resistivity change gradients along a radial tracing path is taken as the representative gradient value for that radial tracing path. The maximum value among all representative gradient values ​​of radial tracing paths is then taken as the maximum gradient value of the radial resistivity change gradient in the resistivity anomaly region. The calculation expression is:

[0064] in, This represents the maximum gradient value of the radial resistivity change gradient in the currently processed resistivity anomaly region, in ohm-meters per meter. This represents the total number of radial tracing paths constructed from the region boundary contour of the currently processed resistivity anomaly region. The value is equal to the total number of grid cells on the upper edge of the region boundary contour; Indicates the first The total number of grid cells contained in a radial tracing path; Indicates the first The first radial tracing path Resistivity interpolation for each grid cell, in ohm-meters; Indicates the first The first radial tracing path Resistivity interpolation for each grid cell, in ohm-meters; Indicates the first The first radial tracing path The center point of the first grid cell and the... The three-dimensional spatial distance between the center points of each grid cell, in meters. The value is calculated from the three-dimensional spatial coordinates corresponding to these two grid cells using the Euclidean distance formula.

[0065] Based on the spatial coverage area and the maximum gradient value of the radial resistivity change gradient, the resistivity anomaly areas are divided into low-risk, medium-risk, and high-risk leakage areas. A leakage risk area distribution map of the impermeable layer is constructed from all these divided resistivity anomaly areas. The specific division rule is as follows: for each resistivity anomaly area, extract the area value of its spatial coverage area and the maximum gradient value of the radial resistivity change gradient. The area value of the spatial coverage is obtained by integrating the surface area of ​​the curved region enclosed by the boundary contour on a 3D digital model of the impermeable layer surface. The area value of the spatial coverage is compared with a preset area threshold, which is set as a percentage of the total surface area of ​​the impermeable layer, specifically 0.5% of the total surface area of ​​the water-facing side. Simultaneously, the maximum gradient value is... The gradient threshold is compared with the preset gradient threshold, which is set according to the normal resistivity fluctuation range of the flexible nano-conductive coating under non-leakage conditions. The gradient threshold is taken as the 95th percentile of the statistical value of the spatial change gradient of resistivity under normal working conditions of the flexible nano-conductive coating. This 95th percentile is obtained by sorting the resistivity gradient data of the non-leakage area in historical tests and taking the percentile.

[0066] If the area value is less than the area threshold and the maximum gradient value If the resistivity is less than the gradient threshold, the corresponding area of ​​abnormal resistivity is classified as a low-risk leakage area; if the area value is greater than or equal to the area threshold and the maximum gradient value is... Less than the gradient threshold, or the area value is less than the area threshold and the maximum gradient value is... If the resistivity anomaly is greater than or equal to the gradient threshold, the corresponding area is classified as a medium-risk leakage area; if the area value is greater than or equal to the area threshold and the maximum gradient value is... If the resistivity is greater than or equal to the gradient threshold, the corresponding area of ​​abnormal resistivity is identified as a high-risk leakage area.

[0067] After determining the risk level of each resistivity anomaly region, the risk identification module compiles the geometric center coordinates, the three-dimensional spatial coordinate sequence of the region's boundary contour, the area value of the spatial coverage, the maximum gradient value of the radial resistivity change gradient, and the determined risk level of each region into a dataset. Based on this dataset, a leakage risk area distribution map of the impermeable layer is generated on a three-dimensional digital model of the impermeable layer surface. The map is rendered using a first color to represent low-risk leakage areas, a second color to represent medium-risk leakage areas, and a third color to represent high-risk leakage areas. This leakage risk area distribution map is stored and displayed as a three-dimensional visualization model file, which includes the overall curved surface model of the impermeable layer surface and risk area markers superimposed on the curved surface model.

[0068] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A seepage-proof layer detection system that integrates a nano-conductive coating with an underwater detection device, characterized in that, The system includes: A coating sensing module is used to coat a flexible nano-conductive coating on the water-facing side of the impermeable layer, and to construct a distributed resistance sensing network based on the flexible nano-conductive coating. The detection and acquisition module is used to measure the resistivity of the surface of the flexible nano-conductive coating according to a preset scanning path through the four-electrode excitation and acquisition module mounted on the underwater unmanned detection device, and simultaneously acquire the acoustic positioning data and attitude inertial data of the underwater unmanned detection device. The calibration mapping module is used to perform spatial coordinate mapping and probe attitude drift compensation on the measured resistivity data based on the acoustic positioning data and the attitude inertial data, and generate a resistivity measurement sequence with positioning marks. The field inversion module is used to construct the resistivity spatial distribution field of the impermeable layer surface based on the resistivity measurement sequence with positioning marks, and to extract the resistivity anomaly region in the resistivity spatial distribution field. The risk identification module is used to determine the distribution map of leakage risk areas of the impermeable layer based on the spatial location and resistivity change gradient of the resistivity anomaly area.

2. The seepage prevention layer detection system based on the nano-conductive coating and underwater detection device according to claim 1, characterized in that, The construction of a distributed resistance sensing network based on the flexible nano-conductive coating includes: Based on the geometric dimensions and surface curvature of the impermeable layer, the flexible nano-conductive coating is divided into multiple independent resistance sensing units. By embedding a conductive busbar at the boundary of each resistance sensing unit, each resistance sensing unit is connected to the corresponding measurement lead. All the measurement leads are brought together to a multi-channel switching switch installed on the underwater unmanned detection device, forming a distributed resistance sensing network in which any one of the resistance sensing units can be selected by the multi-channel switching switch.

3. The seepage prevention layer detection system based on the nano-conductive coating and underwater detection device according to claim 1, characterized in that, The four-electrode excitation and acquisition module mounted on the underwater unmanned detection device performs resistivity measurement on the surface of the flexible nano-conductive coating according to a preset scanning path, including: Based on the three-dimensional digital model of the impermeable layer surface, a gridded scanning path covering the entire surface of the flexible nano-conductive coating is generated. The gridded scanning path includes multiple measurement stagnation points and the movement trajectory between adjacent measurement stagnation points. Control the underwater unmanned detection device to move along the movement trajectory and stop moving at each of the measurement points; At the measurement stagnation point, a constant current is applied to the two outer electrodes through the four-electrode excitation and acquisition module, while the potential difference between the two inner electrodes is measured. The surface resistivity value at the current measurement stagnation point is calculated based on the constant current and the potential difference.

4. The seepage prevention layer detection system based on the nano-conductive coating and underwater detection device according to claim 3, characterized in that, The synchronous acquisition of acoustic positioning data and attitude inertial data of the underwater unmanned detection device includes: Each time the underwater unmanned detection device reaches the measurement station, it triggers the acoustic positioning beacon set on the underwater unmanned detection device to emit an acoustic positioning signal and receives the time difference data of the acoustic positioning signal received by the surface array. Based on the time difference data, it calculates the three-dimensional spatial coordinates of the underwater unmanned detection device at the current measurement station. Simultaneously, the inertial measurement unit mounted on the underwater unmanned detection device is activated to collect the three-axis acceleration and three-axis angular velocity data of the underwater unmanned detection device at the current measurement station point; The attitude calculation is performed on the triaxial acceleration and triaxial angular velocity data to obtain the roll angle, pitch angle and yaw angle of the underwater unmanned detection device at the current measurement station.

5. The seepage prevention layer detection system based on the nano-conductive coating and underwater detection device according to claim 4, characterized in that, The process of performing spatial coordinate mapping and probe attitude drift compensation on the measured resistivity data to generate a resistivity measurement sequence with positioning markers includes: Based on the three-dimensional spatial coordinates, the surface resistivity value measured at the current measurement station is assigned to the corresponding spatial coordinate position, generating initial resistivity coordinate mapping data; Based on the roll angle, pitch angle, and yaw angle, calculate the actual contact angle of the probe of the four-electrode excitation and acquisition module relative to the surface of the impermeable layer; Based on the actual contact tilt angle, the surface resistivity value in the initial resistivity coordinate mapping data is corrected by electrode spacing projection to obtain the resistivity coordinate mapping data after attitude compensation. The attitude-compensated resistivity coordinate mapping data are arranged in the order of measurement time to generate a resistivity measurement sequence with positioning marks, which includes spatial coordinates and compensated resistivity values.

6. The seepage prevention layer detection system based on the nano-conductive coating and underwater detection device according to claim 1, characterized in that, The step of constructing the spatial distribution field of resistivity on the surface of the impermeable layer based on the resistivity measurement sequence with positioning marks includes: All measurement points in the resistivity measurement sequence with positioning marks are mapped onto the two-dimensional parameter plane of the impermeable layer surface according to spatial coordinates to obtain a discrete measurement point set; The radial basis function interpolation algorithm is used to perform spatial interpolation calculation on the resistivity of the impermeable layer surface, with the discrete measurement point set as the interpolation node; The two-dimensional parameter plane is uniformly divided into multiple grid cells, and the resistivity interpolation value at the center point of each grid cell is calculated using the radial basis function interpolation algorithm. The resistivity interpolation values ​​of all grid cells constitute the resistivity spatial distribution field of the impermeable layer surface.

7. The seepage prevention layer detection system based on the nano-conductive coating and underwater detection device according to claim 6, characterized in that, The extraction of resistivity anomaly regions in the resistivity spatial distribution field includes: Calculate the statistical median and absolute median difference of resistivity values ​​for all grid cells in the resistivity spatial distribution field; Based on the statistical median and the absolute median difference, a lower threshold and an upper threshold for determining resistivity anomalies are set. Traverse each grid cell in the resistivity spatial distribution field and mark grid cells with resistivity values ​​below the lower threshold or above the upper threshold as abnormal grid cells; Connectivity clustering analysis is performed on the abnormal grid cells to merge multiple spatially adjacent abnormal grid cells into the same resistivity anomaly region.

8. The seepage prevention layer detection system in collaboration with the nano-conductive coating and underwater detection device according to claim 7, characterized in that, The step of determining the leakage risk area distribution map of the impermeable layer based on the spatial location and resistivity change gradient of the resistivity anomaly region includes: Calculate the geometric center coordinates and region boundary contour of each resistivity anomaly region, and determine the spatial coverage of each resistivity anomaly region on the surface of the impermeable layer based on the geometric center coordinates and the region boundary contour. For each grid cell within the resistivity anomaly region, the resistivity difference between adjacent grid cells is calculated along a direction perpendicular to the boundary profile of the region, and the radial resistivity change gradient of the resistivity anomaly region is determined based on the resistivity difference. Based on the spatial coverage and the radial resistivity change gradient, the resistivity anomaly area is divided into low-risk leakage area, medium-risk leakage area and high-risk leakage area, and all the divided resistivity anomaly areas constitute the leakage risk area distribution map of the seepage prevention layer.

9. The seepage prevention layer detection system based on the nano-conductive coating and underwater detection device according to claim 8, characterized in that, The method of dividing the resistivity anomaly region into low-risk leakage region, medium-risk leakage region, and high-risk leakage region based on the spatial coverage area and the radial resistivity change gradient includes: For each resistivity anomaly region, extract the area value of its spatial coverage and the maximum gradient value of the radial resistivity change gradient; The area value is compared with a preset area threshold, and the maximum gradient value is compared with a preset gradient threshold. If the area value is less than the area threshold and the maximum gradient value is less than the gradient threshold, the resistivity abnormal area is determined to be a low-risk leakage area. If the area value is greater than or equal to the area threshold and the maximum gradient value is less than the gradient threshold, or if the area value is less than the area threshold and the maximum gradient value is greater than or equal to the gradient threshold, the resistivity abnormal area is determined to be a medium-risk leakage area. If the area value is greater than or equal to the area threshold and the maximum gradient value is greater than or equal to the gradient threshold, the area with abnormal resistivity is identified as a high-risk leakage area.

10. The seepage prevention layer detection system based on the nano-conductive coating and underwater detection device according to claim 2, characterized in that, When the multi-channel switching switch selects the current resistance sensing unit for measurement, the unselected resistance sensing units are connected to a high-impedance state through the multi-channel switching switch to avoid current crosstalk between adjacent resistance sensing units.

Images