Intelligent monitoring method for water seepage of pipe joint of immersed tunnel based on optimized direct current method

Abstract

The present application relates to a kind of intelligent monitoring method of sinking pipe tunnel pipe joint water seepage based on optimization direct current method, belong to leakage detection technical field.The method includes: S1: with sinking pipe tunnel joint as center symmetry axis, symmetrically and evenly arranged n vertical row electrodes on the both sides of joint, 2n electrodes are arranged between each vertical row;S2: electrode is divided into power supply electrode (X, Y) and measuring electrode (O, P), wherein X, Y, O, P are located on a straight line, O, P is always located near the arrangement center of X, Y;S3: the apparent resistivity of each measuring point on each measuring line is calculated, and the apparent resistivity of all measuring points is generated a three-dimensional data matrix;Actual resistivity is calculated using three-dimensional fast inversion algorithm based on neural network design, to determine water seepage condition.The present application realizes the real-time, high-precision monitoring of pipe joint water seepage condition, improves the safety and reliability of tunnel operation.

Description

Technical Field

[0001] This invention belongs to the field of leakage detection technology and relates to an intelligent monitoring method for water seepage at the joint of immersed tunnel pipe section based on optimized DC current method. Background Technology

[0002] Immersed tunnels, as core transportation infrastructure crossing waterways, are widely used in projects spanning rivers and seas. The tunnel segment joint is a critical component of immersed tunnels, and its waterproofing performance directly determines the safety and durability of the tunnel structure. In modern engineering practice, a single standard tunnel segment can reach 180 meters in length and weigh over 80,000 tons, requiring millimeter-level splicing accuracy through precision docking technology under complex hydrological conditions. However, the GINA waterstop at the tunnel segment joint is susceptible to ground settlement, water pressure fluctuations, and material aging during long-term operation, leading to a significant increase in the risk of leakage.

[0003] Limitations of existing seepage monitoring technologies:

[0004] (1) Deficiencies of manual inspection method:

[0005] Inefficient: Traditional manual inspections require periodic lane closures for visual inspections, with each inspection taking more than 48 hours, severely impacting traffic operation efficiency;

[0006] High rate of missed detection: Water seepage traces in hidden areas are difficult to detect with the naked eye, resulting in a high rate of missed detection;

[0007] Data lag: Unable to capture sudden leakage events, resulting in long maintenance response delays.

[0008] (2) Shortcomings of sensor monitoring technology:

[0009] 1) Pressure sensor: It is susceptible to interference from dynamic changes in water pressure, resulting in a high false alarm rate; cross-sea tunnels experience multiple false alarms due to tidal pressure fluctuations, increasing maintenance costs.

[0010] 2) Humidity sensor: Electrode corrosion failure occurs in high-salt environments, resulting in a service life of less than 3 years (compared to the laboratory nominal 5 years), requiring frequent replacement.

[0011] 3) Fiber optic sensing system: The wiring is complex and local breaks can cause blind spots in monitoring, resulting in high repair costs.

[0012] (3) The disconnect between structural improvements and monitoring technology

[0013] Current technological innovations are mostly focused on optimizing the structure of pipe sections (such as the steel shell reinforcement design in patent CN119956822A), while intelligent sensing technology for joint leakage is still stuck in the stage of single physical quantity detection and lacks the ability to fuse and analyze multimodal data.

[0014] As immersed tunnels develop towards ultra-long and deep-water directions, traditional monitoring methods can no longer meet three core requirements: 1) Real-time: It is necessary to achieve a response speed of seconds to prevent the spread of structural damage; 2) Accuracy: It is required that the leakage location error is less than 0.5 meters and the false alarm rate is less than 1%; 3) Adaptability: It is necessary to maintain stable operation for more than 10 years in extreme corrosive environments with pH values ​​of 2.8-10.5 and salinity of 35‰. Summary of the Invention

[0015] In view of this, the purpose of this invention is to provide an intelligent monitoring method for water seepage at the joint of immersed tunnel sections based on optimized DC current method. By increasing the number and density of electrodes, adjusting the electrode arrangement, adopting multi-mode current excitation, and introducing an inversion algorithm, the method can achieve real-time and high-precision monitoring of water seepage at the joint of the tunnel sections, thereby improving the safety and reliability of tunnel operation.

[0016] To achieve the above objectives, the present invention provides the following technical solution:

[0017] Solution 1: A smart monitoring method for seepage in immersed tunnel joints based on optimized DC electrical resistivity tomography, specifically including the following steps:

[0018] S1: Electrode layout: With the immersed tunnel joint as the central axis of symmetry, n vertical rows of electrodes are symmetrically and evenly arranged on both sides of the joint, with 2n electrodes spaced apart in each vertical row, n≥5, and the horizontal and vertical spacing between the electrodes is the same.

[0019] S2: Multi-mode current excitation and reception: The electrodes are divided into power supply electrodes (X, Y) and measurement electrodes (O, P), where X, Y, O, and P are located on a straight line, and O and P are always located near the center of X and Y, and so on; each horizontal or vertical 2n electrode can be combined to form a measurement line; during each measurement, the power supply electrode excites and receives current I, and the two measurement electrodes measure the voltage, which is recorded as V;

[0020] S3: Signal Processing and Analysis: Calculate the apparent resistivity of each measuring point on each measuring line and generate a three-dimensional data matrix from the apparent resistivity of all measuring points; then use a three-dimensional fast inversion algorithm based on neural network design to calculate the actual resistivity; finally, based on the actual resistivity distribution obtained by inversion, compare it with the resistivity under normal conditions. If there is an obvious low resistivity area, it is determined that the area has abnormal water seepage.

[0021] Furthermore, in step S1, the electrode is made of a corrosion-resistant and highly conductive material and is securely installed on the tunnel wall using a special fixing device.

[0022] Furthermore, in step S3, the formula for calculating apparent resistivity is:

[0023] ρ=(π×((XY / 2) 2-(OP / 2) 2 ) / OP)×(V / I)

[0024] Where ρ represents the apparent resistivity of each measuring point, XY represents the distance between power supply electrodes X and Y, and OP represents the distance between power supply electrodes O and P.

[0025] Furthermore, in step S3, after each measurement line is calculated, each measurement point corresponds to an apparent resistivity ρ value. A single measurement line can obtain an inverted triangular two-dimensional profile, recording the apparent resistivity ρ at different positions and depths along the profile direction of the measurement line. Therefore, all the results of the 2n transverse measurement lines and the 2n longitudinal measurement lines of the array can be combined to generate a three-dimensional data matrix, which can be used for three-dimensional imaging.

[0026] Furthermore, in step S3, the 3D fast inversion algorithm based on neural network design specifically includes:

[0027] 1) Constructing a forward model: Under a homogeneous concrete background, generate apparent resistivity data d by meshing;

[0028] 2) Data scale: 10 generated 4 ~10 5 A three-dimensional apparent resistivity matrix, a three-dimensional resistivity model m, and corresponding apparent resistivity data d;

[0029] 3) Noise injection: Gaussian noise (signal-to-noise ratio ≥ 20dB) is added to the data d to simulate the actual measurement error;

[0030] 4) Neural Network Design:

[0031] Input: Three-dimensional apparent resistivity matrix (size: 10×10×10, consisting of 10 transverse and 10 longitudinal survey lines; for points without data, the average interpolation method is used to generate the matrix).

[0032] Output: Three-dimensional actual resistivity distribution matrix (same size);

[0033] 5) Network architecture: A 3D U-Net-based encoder-decoder structure is used to capture spatial features, and skip connections preserve details; a physical constraint layer is introduced (such as a non-negative activation function to ensure resistivity ≥ 0); a residual module is added to enhance the ability to extract deep features;

[0034] 6) Hyperparameters: The loss function is weighted MSE (focusing on outlier regions); the optimizer is Adam (initial learning rate 1e-4, dynamically decaying);

[0035] 7) Training and optimization: The training strategy is to use 80% of the data for training and 20% for validation;

[0036] 8) Inversion process:

[0037] Preprocessing: Normalize the measured apparent resistivity data ρ;

[0038] Obtaining actual resistivity: The normalized data is input into the trained network, which directly outputs a three-dimensional resistivity distribution matrix. Each data point represents the calculated actual resistivity ρ. 实 ;

[0039] Post-processing: Background filtering is applied to remove noise.

[0040] Option 2: An intelligent monitoring system for seepage in immersed tunnel joints based on optimized DC current method, comprising electrodes, a signal processing unit, a wireless communication module, and a storage module.

[0041] Electrode installation: During the construction of the pipe joint, pre-drill holes for electrode installation and use specialized clamps to install the electrodes, ensuring tight contact between the electrodes and the tunnel wall, and guaranteeing that the electrode spacing and positional accuracy meet the standards. Conduct conductivity tests after installation.

[0042] Intelligent alarm and data storage: When an abnormal water seepage is detected, an alarm is sent to the monitoring center via the wireless communication module, and at the same time, data such as the location, time and degree of abnormality of the seepage are saved in the local storage module.

[0043] Monitoring system debugging: Assemble and debug signal processing units and other equipment, set excitation current parameters, as well as the standard value range of received signals and alarm thresholds; simulate different degrees of water seepage to test the accuracy and reliability of the system.

[0044] Routine monitoring and maintenance: The monitoring system operates automatically during tunnel operation. Regularly inspect and maintain the monitoring equipment, analyze the stored data, and monitor seepage trends.

[0045] The beneficial effects of this invention are as follows:

[0046] (1) High-precision seepage monitoring capability: This invention, by employing a dense electrode array and multi-mode current excitation, combined with a three-dimensional fast inversion algorithm (3D U-Net network + physical constraint layer), can acquire more comprehensive and accurate current and potential difference signals, accurately invert the resistivity distribution of the pipe joint area, effectively capture minute seepage situations, greatly improve monitoring accuracy, and effectively overcome the problem of insufficient signal resolution caused by the low electrode density of traditional DC methods. Compared with the two-dimensional inversion error of conventional methods, this invention greatly improves the resistivity inversion accuracy of the seepage area through three-dimensional data matrix and noise modeling.

[0047] (2) Real-time early warning: The present invention processes and analyzes the monitoring data in real time, and immediately alarms once abnormal water seepage is detected, so as to promptly detect safety hazards and ensure the safe operation of the tunnel.

[0048] (3) Complete spatial coverage: This invention adopts a symmetrical electrode arrangement structure (with the joint as the axis of symmetry) + multi-dimensional measurement line combination to form a three-dimensional detection network. Compared with the conventional unidirectional electrode layout, the monitoring blind zone is reduced, and seepage cavities can be accurately identified.

[0049] (4) Convenient installation and maintenance: This invention improves electrode replacement efficiency and reduces contact resistance fluctuations through pre-fabricated electrode mounting holes and a dedicated fixture design. Compared to embedded electrode solutions, maintenance costs are reduced.

[0050] (5) Good anti-interference performance: The introduction of weighted MSE loss function + adaptive background filtering algorithm makes the false alarm rate of the system low in strong electromagnetic interference environment.

[0051] (6) High level of intelligence: It has automatic alarm and data storage functions, reducing manual intervention and improving monitoring and management efficiency.

[0052] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0053] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0054] Figure 1 This is a schematic diagram of motor installation according to an embodiment of the present invention;

[0055] Figure 2 This is a schematic diagram of the electrode test positions for multi-mode current excitation and reception in an embodiment of the present invention;

[0056] Figure 3 This is a schematic diagram of the formed three-dimensional resistivity matrix. Detailed Implementation

[0057] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0058] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0059] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present invention. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0060] Please see Figures 1-3 This invention provides an intelligent monitoring method for seepage at the joints of immersed tunnel sections based on an optimized direct current method, specifically including the following steps:

[0061] 1. Electrode Layout

[0062] (1) Quantity and density: Five vertical rows of electrodes are arranged on each side of the immersed tunnel joint, with 10 electrodes in each row and a spacing of 1 meter between each electrode (see...). Figure 1 The electrodes are made of corrosion-resistant and highly conductive materials, such as stainless steel and copper alloys, and are securely installed on the tunnel wall using special fixing devices.

[0063] (2) Arrangement: The electrodes are arranged symmetrically on both sides with the connector as the central axis of symmetry, so that the current distribution is more uniform, avoiding monitoring blind spots, and obtaining resistivity signals from different angles to more accurately reflect the resistivity changes at the connector.

[0064] 2. Multi-mode current excitation and reception

[0065] (1) It is divided into power supply electrodes (X, Y) and measuring electrodes (O, P), wherein X, Y, O, and P are located on a straight line, and O and P are always located near the center of the arrangement of X and Y (see Figure 2 And so on.

[0066] (2) Each of the 10 measuring points along a horizontal line or the 10 measuring points along a vertical line can be combined to form Figure 2 The survey line shown is a survey line.

[0067] (3) During each measurement, the power supply electrode is excited and received by the current I, and the two measuring electrodes are used to measure the voltage. The measured voltage is recorded as V.

[0068] 3. Signal Processing and Analysis

[0069] (1) Formula for calculating apparent resistivity ρ:

[0070] ρ=(π×((XY / 2) 2 -(OP / 2) 2 ) / OP)×(V / I)

[0071] (2) Calculation steps:

[0072] 1) Data acquisition: Record V and I at different XY intervals;

[0073] 2) Calculate apparent resistivity: Substitute into the apparent resistivity calculation formula to obtain ρ at each measuring point.

[0074] (3) Data Matrix

[0075] refer to Figure 3 After calculation for each survey line, each red dot represents a obtained ρ value. A single survey line can yield an inverted triangular two-dimensional profile, recording the apparent resistivity ρ at different positions and depths along the profile direction of that survey line. Therefore, by combining all the results from the 10 transverse survey lines and the 10 longitudinal survey lines of this array, a three-dimensional data matrix can be generated, enabling three-dimensional imaging.

[0076] (4) Three-dimensional fast inversion algorithm based on neural network design

[0077] 1) Constructing a forward model: Under the background of homogeneous concrete, generate apparent resistivity data d by meshing.

[0078] 2) Data scale: 10 generated 4 ~10 5 Group (three-dimensional resistivity model m, corresponding to apparent resistivity data d).

[0079] 3) Noise injection: Gaussian noise (signal-to-noise ratio ≥ 20dB) is added to the data d to simulate the actual measurement error.

[0080] 4) Neural Network Design

[0081] Input: Three-dimensional apparent resistivity matrix (size: 10×10×10, consisting of 10 transverse and 10 longitudinal survey lines; for points without data, the average interpolation method is used to generate the matrix).

[0082] Output: 3D true resistivity distribution matrix (same size).

[0083] 5) Network Architecture: 3D U-Net: Utilizes an encoder-decoder structure to capture spatial features, with skip connections preserving details. Introduces physically constrained layers (e.g., non-negative activation functions to ensure resistivity ≥ 0). Adds residual modules to enhance deep feature extraction capabilities.

[0084] 6) Hyperparameters: The loss function is weighted MSE (focusing on outlier regions); the optimizer is Adam (initial learning rate 1e-4, dynamically decaying).

[0085] 7) Training and optimization: The training strategy is to use 80% of the data for training and 20% for validation.

[0086] 8) Inversion process

[0087] Preprocessing: Normalize the measured apparent resistivity data ρ.

[0088] Obtaining actual resistivity: Input the data into the trained network, and it directly outputs a three-dimensional resistivity matrix, where each data point represents the calculated actual resistivity ρ. 实 .

[0089] Post-processing: Background filtering is applied to remove noise.

[0090] (5) Determine water seepage: Based on the actual resistivity ρ obtained from the inversion. 实 By comparing the resistivity distribution with that under normal conditions, if a region with significantly low resistivity is observed, it is determined that there may be water seepage in that region.

[0091] 4. Intelligent alarm and data storage

[0092] When an abnormal water seepage is detected, an alarm is sent to the monitoring center via a wireless communication module, and data containing information such as the location, time, and degree of abnormality of the seepage is stored locally.

[0093] 5. Electrode Installation

[0094] During the construction of the pipe joint, pre-drill holes for electrode installation and use specialized clamps to install the electrodes, ensuring close contact between the electrodes and the tunnel wall and guaranteeing that the electrode spacing and positional accuracy meet the standards. Conduct conductivity tests are performed after installation.

[0095] 6. Monitoring system debugging

[0096] Assemble and debug signal processing units and other equipment, and set excitation current parameters, standard range of received signal values, and alarm thresholds. Simulate different degrees of water seepage to test the accuracy and reliability of the system.

[0097] 7. Daily monitoring and maintenance

[0098] The monitoring system operates automatically during tunnel operation. Regular inspections and maintenance of the monitoring equipment, analysis of stored data, and monitoring of seepage trends are conducted.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for intelligent monitoring of seepage at joints of immersed tunnel sections based on optimized direct current method, characterized in that, The method specifically includes the following steps: S1: Electrode Layout: With the immersed tunnel joint as the central axis of symmetry, n vertical rows of electrodes are symmetrically and evenly arranged on both sides of the joint, with 2n electrodes spaced apart in each vertical row. The horizontal and vertical spacing between the electrodes is the same; S2: Multi-mode current excitation and reception: The electrodes are divided into power supply electrodes X and Y and measurement electrodes O and P, where X, Y, O, and P are located on a straight line, and O and P are always located near the center of X and Y, and so on; each horizontal or vertical 2n electrode can be combined to form a measurement line; during each measurement, the power supply electrode excites and receives current I, and the two measurement electrodes measure the voltage, which is recorded as V; S3: Signal Processing and Analysis: Calculate the apparent resistivity of each measuring point on each measuring line and generate a three-dimensional data matrix from the apparent resistivity of all measuring points; then use a three-dimensional fast inversion algorithm based on neural network design to calculate the actual resistivity; finally, based on the actual resistivity distribution obtained by inversion, compare it with the resistivity under normal conditions. If there is an obvious low resistivity area, it is determined that the area has abnormal water seepage.

2. The intelligent monitoring method for water seepage at the joint of immersed tunnel sections according to claim 1, characterized in that, In step S1, the electrode is securely installed on the tunnel wall using a special fixing device.

3. The intelligent monitoring method for water seepage at the joint of immersed tunnel sections according to claim 1, characterized in that, In step S3, the formula for calculating apparent resistivity is: ρ=(π×((XY / 2)²-(OP / 2)²) / OP)×(V / I) Where ρ represents the apparent resistivity of each measuring point, XY represents the distance between power supply electrodes X and Y, and OP represents the distance between power supply electrodes O and P.

4. The intelligent monitoring method for water seepage at the joint of immersed tunnel sections according to claim 1, characterized in that, In step S3, after each measurement line is calculated, each measurement point corresponds to an apparent resistivity ρ value. A single measurement line can obtain an inverted triangular two-dimensional profile, recording the apparent resistivity ρ at different positions and depths in the profile direction of the measurement line. Therefore, all the results of the 2n transverse measurement lines and the 2n longitudinal measurement lines are combined together to generate a three-dimensional data matrix and perform three-dimensional imaging.

5. The intelligent monitoring method for water seepage at the joint of immersed tunnel sections according to claim 1, characterized in that, In step S3, the three-dimensional fast inversion algorithm based on neural network design specifically includes: 1) Constructing a forward model: Under a homogeneous concrete background, generate apparent resistivity data d by meshing; 2) Data scale: 10 generated 4 ~10 5 A three-dimensional apparent resistivity matrix, a three-dimensional resistivity model m, and corresponding apparent resistivity data d; 3) Noise injection: Gaussian noise is added to the data d to simulate actual measurement errors; 4) Neural Network Design: Input: Three-dimensional apparent resistivity matrix; Output: Three-dimensional actual resistivity distribution matrix; 5) Network architecture: Utilizes an encoder-decoder structure based on 3D U-Net to capture spatial features, with skip connections preserving details; introduces a physical constraint layer; and adds a residual module to enhance deep feature extraction capabilities; 6) Hyperparameters: The loss function is weighted MSE; the optimizer is Adam. 7) Training and optimization: The training strategy is to use 80% of the data for training and 20% for validation; 8) Inversion process: Preprocessing: Normalize the measured apparent resistivity data ρ; Obtaining actual resistivity: The normalized data is input into the trained network, which directly outputs a three-dimensional resistivity distribution matrix. Each data point represents the calculated actual resistivity ρ. 实 ; Post-processing: Background filtering is applied to remove noise.

6. The intelligent monitoring method for seepage at the joint of immersed tunnel sections according to any one of claims 1 to 5, characterized in that, The monitoring system applicable to this method includes electrodes, a signal processing unit, a wireless communication module, and a storage module.

7. The intelligent monitoring method for water seepage at the joint of immersed tunnel sections according to claim 6, characterized in that, When an abnormal water seepage is detected, an alarm is sent to the monitoring center via the wireless communication module, and the location, time and severity of the seepage are saved in the local storage module.

8. The intelligent monitoring method for water seepage at the joint of immersed tunnel sections according to claim 6, characterized in that, During the commissioning phase of the monitoring system, the signal processing unit was assembled and debugged, the excitation current parameters, the standard value range of the received signal, and the alarm threshold were set; different degrees of water seepage were simulated to test the accuracy and reliability of the system.

Images