Two-dimensional thermal map visualization system for construction quality of ultra-deep cutoff wall brushing

Abstract

The application discloses a kind of ultra-deep anti-infiltration wall wall brushing construction quality two-dimensional thermal map visualization system, belong to underground anti-infiltration wall construction quality detection technical field.System includes data acquisition and synchronization module, standardization pretreatment module, core calculation module and two-dimensional thermal map visualization module connected in turn.Communication data acquisition and synchronization module gathers arc array ultrasonic echo, carrier depth and attitude data and carries out time stamp synchronization;Standardization pretreatment module carries out original data noise reduction cleaning;Core calculation module identifies algorithm to calculate full-section mud thickness matrix;Two-dimensional thermal map visualization module converts thickness matrix into intuitive two-dimensional development thermal map, and automatically labels over-standard area.The present application solves the problem of mud distribution in the prior art, which is not intuitive and the defect positioning is not accurate, can quickly judge the quality of wall brushing, is suitable for detecting the concave arc side wall formed by joint pipe method construction, and the detection accuracy reaches millimeter level, and the positioning accuracy reaches centimeter level.

Description

Technical Field

[0001] This invention relates to the field of water conservancy and hydropower construction technology, and more particularly to the field of underground anti-seepage wall construction quality testing technology. Specifically, it relates to a two-dimensional thermal mapping visualization system for ultra-deep anti-seepage wall brushing construction quality, which is particularly suitable for the intuitive detection of the thickness and distribution of mud skin on the concave arc-shaped sidewall formed by the joint pipe method construction. Background Technology

[0002] Ultra-deep underground cutoff walls are core water-stopping structures in water conservancy projects, deep foundation pit support, and tailings dam seepage prevention. Their construction quality directly determines the seepage prevention effect and operational safety of the project. Currently, ultra-deep cutoff walls are mostly constructed in sections using the joint pipe method. After the joint pipe is removed, a concave cylindrical arc-shaped sidewall will form at the end of the section. The residual mud and sludge on the surface of this sidewall will directly affect the joint bonding strength of the subsequent concrete pouring, easily forming a leakage channel at the joint.

[0003] In existing technologies, mud cake detection in cutoff walls primarily employs methods such as manual probing, core drilling, and in-hole photography. These methods are unsuitable for detecting cutoff walls in ultra-deep strata, or suffer from limited detection depth, low efficiency, and poor accuracy. While ultrasonic testing technology, which has emerged in recent years, enables non-contact detection, it still has the following drawbacks: First, ultrasonic probes cannot meet the pressure requirements for ultra-deep strata detection and therefore cannot function properly. Second, the detection results are mostly discrete numerical values ​​or curves, failing to visually represent the distribution of mud cake thickness across the entire cross-section, making it difficult for on-site construction personnel to quickly assess the quality of the wall-brushing process. Third, there is a lack of dedicated visualization solutions for curved sidewalls, and planar coordinate mapping can easily lead to coordinate deviations, affecting the accuracy of defect location. Fourth, detection is disconnected from construction, failing to provide real-time and intuitive guidance for the wall-brushing operation.

[0004] Therefore, based on the above-mentioned technical problems, especially for high-altitude areas, special geological conditions, and the construction needs of ultra-deep cutoff walls, there is an urgent need to develop a visualization system for the construction quality of ultra-deep cutoff wall brushing that can intuitively display the distribution of mud cake thickness across the entire cross section, accurately locate the problem, and is easy to operate. Summary of the Invention

[0005] To address the limitations of existing technologies in performing full-depth inspections of ultra-deep cutoff walls or in accurately and intuitively visualizing inspection results, this application provides a two-dimensional thermal mapping visualization system for the construction quality of ultra-deep cutoff wall brushing. This system specifically addresses the issues of unclear distribution of mud cake thickness on the sidewalls of the preceding trench sections of ultra-deep cutoff walls, inaccurate defect location, and low efficiency in on-site judgment before subsequent trench construction.

[0006] To achieve the above objectives, the technical solution adopted in this application is as follows:

[0007] A two-dimensional heat map visualization system for the construction quality of ultra-deep cut-off wall brushing includes a data acquisition and synchronization module, a standardization preprocessing module, a core calculation module, a two-dimensional heat map visualization module, and a result export module, which are connected in sequence via communication. The data acquisition and synchronization module is used for multi-source data acquisition and timestamp synchronization binding to generate an original dataset with spatial reference. The multi-source data includes ultrasonic echo data, depth data, and attitude data collected by an arc-shaped array ultrasonic detection sensor installed on a carrier. The standardization preprocessing module is used to perform noise reduction, outlier removal, and bad channel shielding on the original dataset, and output a valid echo dataset. The core calculation module is used to perform global cylindrical coordinate system registration, overlapping area data fusion, and dual-interface recognition algorithms based on the valid echo dataset to calculate a two-dimensional matrix of mud cake thickness across the entire cross section. The two-dimensional heat map visualization module is used to convert the two-dimensional mud cake thickness matrix into a two-dimensional unfolded heat map, realizing an intuitive display of mud cake thickness and distribution.

[0008] Preferably, the data acquisition and synchronization module includes an arc-shaped array ultrasonic detection sensor, a multi-channel synchronous data acquisition card, a dual-axis tilt sensor, a depth encoder, a downhole explosion-proof acquisition circuit, and a ground terminal, all mounted on a carrier. The arc-shaped array ultrasonic detection sensor includes an arc-shaped circumferential tube, on which radial tubes are fixedly connected in a circumferential array. A valve plug for eliminating internal and external pressure differences is provided at the end of the circumferential tube. The multi-channel synchronous data acquisition card uses a time-division polling triggering algorithm to control each ultrasonic sensor to emit ultrasonic pulses sequentially, with a trigger interval of ≥200μs between adjacent ultrasonic sensors.

[0009] More preferably, the data acquisition and synchronization module further includes a multi-source data timestamp synchronization unit and a data integrity verification unit; the multi-source data timestamp synchronization unit is used to add the same millisecond-level timestamp to echo data, depth data and tilt data at the same acquisition time to realize spatial reference binding of multi-source data; the data integrity verification unit uses the CRC32 check algorithm to verify the transmitted data packets.

[0010] Preferably, the standardized preprocessing module includes a data parsing unit and a preprocessing execution unit; the data parsing unit is used to parse the raw data packet and extract echo data, depth data, and tilt data of each channel; the preprocessing execution unit sequentially executes the following algorithms: a 100kHz-500kHz Butterworth bandpass filtering algorithm to filter out noise in the mud environment; a linear time gain compensation algorithm to compensate for the propagation attenuation of sound waves in the mud; an outlier removal algorithm to remove invalid data with mud skin thickness exceeding the range of 0-50mm; and a bad channel shielding algorithm to shield faulty sensor channels with an effective echo ratio lower than a preset threshold.

[0011] Preferably, the core calculation module includes a global cylindrical coordinate system registration unit. This unit establishes a global cylindrical coordinate system with the central axis of the measured arc-shaped sidewall as the Z-axis, the groove depth direction as the positive direction, the intersection of the axis and the top surface of the groove as the origin, and the direction of the curvature center as the radial direction. It maps the effective echo data to the arc-shaped sidewall surface and executes a cylindrical-to-planar coordinate transformation algorithm to convert the (θ,z) coordinates in the cylindrical coordinate system to (x,y) coordinates in a two-dimensional unfolded plane, where x=R×θ, y=z, R is the designed curvature radius of the arc-shaped sidewall, θ is the circumferential angle, and z is the depth. Simultaneously, based on the attitude data collected by the dual-axis tilt sensor, it corrects the normal distance deviation between the ultrasonic sensor and the arc-shaped sidewall. Specifically, based on the pitch and roll angle data of the carrier collected by the dual-axis tilt sensor, it calculates the spatial coordinate offset of each ultrasonic sensor in the global cylindrical coordinate system and corrects the sensor-by-sensor changes in the normal detection distance between each ultrasonic sensor and the arc-shaped sidewall caused by changes in carrier attitude.

[0012] More preferably, the core calculation module further includes a circumferential detection area division unit; the circumferential detection area division unit is used to calculate the effective coverage arc length W of a single ultrasonic sensor based on the -6dB half-diffusion angle θ, the rated normal detection distance L, and the radius of curvature R of the arc-shaped sidewall: when R / L≥5, W=2×L×tanθ; when R / L<5, the curvature correction formula W=2×R×arcsin((RL)×tanθ / R) is used; and the effective beam coverage arc segment is delineated according to the installation coordinates of each ultrasonic sensor, the overlapping coverage area of ​​adjacent sensors is identified, and the arc-shaped sidewall is divided into a single-probe coverage area and a multi-probe overlapping coverage area. It is worth noting that the original formula for the effective coverage arc length of the planar wall, W=2×L×tanθ, is based on the assumption that "the detection surface is a plane". When the detection surface is arc-shaped, the propagation path of the ultrasonic beam will undergo geometric distortion, resulting in the actual coverage arc length being less than the calculated value for the planar wall.

[0013] To facilitate understanding by those skilled in the art, the derivation of the formula W=2×L×tanθ is briefly described below. Before explaining the derivation, relevant parameters are defined, and the letters used for these parameters are solely for understanding the derivation process and are not intended for any other purpose or definition. Geometric model definition:

[0014] O is the center of curvature of the arc-shaped sidewall, and R is the radius of curvature of the arc-shaped sidewall; S is the position of the ultrasonic sensor, and the rated normal distance from the sensor to the arc-shaped sidewall is L, so OS=RL; the -6dB half-diffusion angle of the ultrasonic sensor is θ, and the angle between the beam edge and OS is θ; the intersection points of the beam edge and the arc-shaped sidewall are A and B, and arc AB is the actual effective coverage arc length of a single ultrasonic sensor.

[0015] In OSA, according to the law of sines: ,

[0016] Among them, OA=R, OS=RL, ∠OAS=180°-∠AOS-θ;

[0017] Since the beam center is perpendicular to the arc-shaped sidewall, ∠OAS≈90° (error <1°), therefore sin∠OAS≈1. Substituting this into the equation, we get: Since θ is usually small, sinθ≈tanθ, therefore:

[0018] Therefore, the effective coverage arc length W is the length of arc AB, i.e., the curvature correction formula is: ;

[0019] Regarding the applicable boundaries and error analysis, the following applies: When R / L ≥ 5, the curvature effect of the arc-shaped sidewall can be ignored. In this case: Therefore, the corrected formula degenerates into the planar wall formula W=2×L×tanθ, with an error of less than 2%, which is acceptable in actual anti-seepage wall construction projects.

[0020] Furthermore, preferably, the core computing module further includes a multi-source data fusion unit; the multi-source data fusion unit uses the beam center weighting coefficient method to fuse data from the overlapping coverage area of ​​multiple probes: calculating the normal vertical distance from the target coordinate point to the beam axis of each ultrasonic sensor covering that point. ; Calculate the original weight coefficients , ;in, Let be the half-arc length of the arc segment effectively covered by the i-th ultrasonic sensor; normalize the original weighting coefficients to obtain , ;in, is the normalized weighting coefficient, and n is the number of ultrasonic sensors covering the target coordinate point; the thickness value and echo amplitude of each sensor are weighted and summed using the normalized weighting coefficient to obtain the fused thickness value and echo amplitude.

[0021] Furthermore, preferably, the core calculation module further includes a dual-interface recognition and thickness calculation unit; the dual-interface recognition and thickness calculation unit employs a dual-interface feature peak recognition algorithm to identify the front interface echo peak corresponding to the mud-slurry and mud cake interface and the rear interface echo peak corresponding to the mud cake and concrete wall interface; and according to the formula Calculate the mud cake thickness, where t1 is the propagation time of the echo peak at the front interface, t2 is the propagation time of the echo peak at the rear interface, and c is the ultrasonic propagation velocity of the mud calibrated on site.

[0022] Preferably, the two-dimensional heatmap visualization module includes a coordinate mapping unit, a heatmap rendering unit, and an interactive annotation unit. The coordinate mapping unit is used to establish the correspondence between two-dimensional plane coordinates and mud thickness values. The heatmap rendering unit uses a bilinear interpolation algorithm to interpolate discrete thickness data into continuous grid data, and uses a linear gradient color mapping algorithm to render the heatmap, where blue corresponds to 0mm, green corresponds to 2.5mm, yellow corresponds to a 3mm warning threshold, and red corresponds to a ≥5mm unqualified threshold. The interactive annotation unit is used to automatically extract and annotate areas with thicknesses ≥3mm and ≥5mm that exceed the standard. The annotation content includes the maximum thickness of the area and its location coordinates, and supports hovering the mouse to view the precise thickness, depth, and circumferential position of any point, and supports zooming and panning operations.

[0023] Preferably, the system further includes a result export module, which is communicatively connected to the two-dimensional heat map visualization module. The result export module is used to export a high-resolution heat map image of the current view, a full-section mud cake thickness data matrix, and a detailed table of areas exceeding the standard. The exported high-resolution image has a print resolution of ≥300 DPI and a pixel size of not less than 2480×3508. The thickness data matrix is ​​in CSV format, and the detailed table of areas exceeding the standard contains the location, maximum thickness, average thickness, and area information of each area exceeding the standard.

[0024] Beneficial effects:

[0025] 1. This invention uses a two-dimensional unfolded heat map to intuitively display the thickness distribution of mud cake across the entire cross section. Different thicknesses are distinguished by different colors, and areas exceeding the standard are automatically marked. On-site construction personnel can judge the quality of the wall brushing at a glance, greatly improving the efficiency of judgment.

[0026] 2. This invention uses a global cylindrical coordinate system for data registration, which perfectly adapts to the curved sidewall surface, solves the coordinate deviation problem caused by the mapping of the planar coordinate system, and the defect positioning accuracy can reach the centimeter level.

[0027] 3. The system of this invention has a simple process and requires no complex setup. Construction personnel can use it after simple training. Furthermore, it can be directly integrated into existing ultrasonic testing systems for wall brush machines without replacing hardware, thus possessing high compatibility and promotional value. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1This is a structural block diagram of the two-dimensional thermal map visualization system for the construction quality of ultra-deep anti-seepage wall brushing, as described in this invention.

[0030] Figure 2 This is a two-dimensional thermogram of the arc-shaped sidewall of the seepage-proof wall with a joint pipe diameter of 1.2 meters after brushing, as shown in Embodiment 2 of the present invention.

[0031] In the diagram: 1-Data acquisition and synchronization module; 11-Arc array ultrasonic detection sensor; 12-Multi-channel synchronous data acquisition card; 13-Dual-axis tilt sensor; 14-Depth encoder; 15-Downhole explosion-proof acquisition circuit; 16-Ground terminal; 17-Multi-source data timestamp synchronization unit; 18-Data integrity verification unit; 2-Standardized preprocessing module; 21-Data parsing unit; 22-Preprocessing execution unit; 3-Core calculation module; 31-Global cylindrical coordinate system registration unit; 32-Circumferential detection area division unit; 33-Multi-source data fusion unit; 34-Dual interface recognition and thickness calculation unit; 4-Two-dimensional heat map visualization module; 41-Coordinate mapping unit; 42-Heat map rendering unit; 43-Interactive annotation unit; 5-Result export module. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0033] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0034] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0035] In the description of this application, it should be noted that the use of terms such as "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer" to indicate orientation or positional relationships is based on the orientation or positional relationships shown in the accompanying drawings, or the orientation or positional relationships commonly used when the product is in use. These terms are used solely for the convenience of describing this application and for 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, they should not be construed as limitations on this application. Furthermore, the use of terms such as "first" and "second" in the description of this application is only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0036] Furthermore, the use of terms such as "horizontal" and "vertical" in the description of this application does not imply that the component is required to be absolutely horizontal or suspended, but rather that it may be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal relative to "vertical," and does not mean that the structure must be completely horizontal, but rather that it may be slightly tilted.

[0037] In the description of this application, it should also be noted that, unless otherwise expressly specified and limited, the terms "set up," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0038] Example 1:

[0039] This embodiment provides a two-dimensional thermal mapping visualization system for the construction quality of ultra-deep anti-seepage walls, such as... Figure 1 As shown, the system includes, in sequence, a data acquisition and synchronization module 1, a standardization preprocessing module 2, a core calculation module 3, a two-dimensional heatmap visualization module 4, and a result export module 5. The composition and working principle of each module are explained below:

[0040] Data acquisition and synchronization module:

[0041] The data acquisition and synchronization module 1 includes an arc array ultrasonic detection sensor 11, a multi-channel synchronous data acquisition card 12, a dual-axis tilt sensor 13, a depth encoder 14, an explosion-proof acquisition circuit in the well 15, a ground terminal 16, a multi-source data timestamp synchronization unit 17, and a data integrity verification unit 18.

[0042] The arc-shaped array ultrasonic sensor 11 employs an arc-shaped circumferential tube combined with a circular array radial tube structure. The radius of curvature of the circumferential tube is the same as the radius of the connector tube, ensuring that in the detection state, the radial distance between any ultrasonic sensor and the sidewall of the seepage barrier is the same. A valve plug is provided at the end of the circumferential tube, and the tube is filled with insulating and sound-permeable dimethyl silicone oil to achieve adaptive pressure balance under ultra-deep and high-pressure environments. In this embodiment, the diameter of the connector tube is 1.2 meters, therefore the radius of curvature R of the arc-shaped circumferential tube is 0.6 meters.

[0043] The circumferential tubes adopt an arc-shaped structure design, which is adapted to the concave cylindrical arc-shaped anti-seepage wall sidewall formed after the joint tube is removed. This ensures that the axes of the arrayed radial tubes all point to the center of curvature of the arc-shaped sidewall, guaranteeing that the central beam axis of the ultrasonic sensor installed in each radial tube coincides with the normal of the anti-seepage wall sidewall at the corresponding position. Each sensor is arranged in an approximately perpendicular incident manner, and the beams of adjacent sensors form a predetermined overlapping coverage area on the sidewall surface, significantly improving the signal-to-noise ratio of the echo signal and reducing thickness measurement errors. The arc-shaped array of ultrasonic detection sensors is arranged at equal angular intervals along the circumferential tubes, and the angle between the beam axis of each ultrasonic sensor and the normal of the arc-shaped sidewall at the corresponding position is controlled within ±5°. The circumferential center arc length spacing S of adjacent sensors is determined based on the effective coverage arc length W of a single sensor and a predetermined overlap coefficient α, satisfying S=W×(1-α), where α=10%-30%. The circumferential and radial tubes are filled with insulating and sound-permeable dimethyl silicone oil, which connects the entire cavity through a through-hole. Combined with a valve plug at the end of the circumferential tube, when the sensor is lowered into the ultra-deep trench section by the wall-cleaning machine, the hydrostatic pressure of the external mud will cause the valve plug to deform adaptively, making the pressure of the silicone oil inside the tube equal to the external mud pressure in real time. This brings the internal and external pressure difference of the ultrasonic sensor close to zero, completely solving the problems of sensor sealing failure and piezoelectric vibrator damage in ultra-deep, high-pressure environments. Passive pressure balance can be achieved without additional power devices, relying solely on the adaptive deformation of the valve. This structure is simple and highly reliable. The sound-permeable layer at the end of the tube body prevents mud from entering the tube, and together with the insulating and sound-permeable dimethyl silicone oil, ensures lossless ultrasonic signal penetration and avoids signal attenuation. The detection ranges of adjacent ultrasonic sensors are designed with overlapping areas, completely eliminating detection blind spots between adjacent probes and achieving full cross-section coverage of the anti-seepage wall sidewall. The ultrasonic sensors adopt a detachable, sealed connection method, facilitating the inspection and replacement of individual sensors and reducing equipment maintenance costs.

[0044] The multi-channel synchronous data acquisition card 12 uses a 16-channel synchronous data acquisition card with a sampling rate of 20MS / s and a resolution of 12 bits. A time-division polling triggering algorithm is used to control each ultrasonic sensor to emit ultrasonic pulses sequentially, with a trigger interval of 250μs between adjacent ultrasonic sensors, completely avoiding acoustic crosstalk.

[0045] The dual-axis tilt sensor 13 is mounted on the carrier, with an accuracy of 0.05° and a sampling frequency of 20Hz, and is used to detect the pitch and roll angles of the carrier in real time.

[0046] The depth encoder 14 is installed on the drum shaft of the ground crane, with an accuracy of 0.5cm, and is used to detect the lowering depth of the carrier in real time.

[0047] The downhole explosion-proof acquisition circuit 15 adopts an intrinsically safe explosion-proof design and operates at 18V. It is used to power downhole sensors and condition signals.

[0048] The ground terminal 16 uses an industrial-grade laptop computer with an Intel Core i7-12700H CPU, 16GB of memory, a 1TB SSD hard drive, and runs the Windows 10 operating system.

[0049] The multi-source data timestamp synchronization unit 17 adds the same millisecond-level timestamp to each acquired echo data, depth data, and tilt data, binding the multi-source data together.

[0050] The data integrity verification unit 18 uses the CRC32 check algorithm to verify each data packet. If a data error is found, it requests a retransmission.

[0051] Standardized preprocessing module:

[0052] The standardized preprocessing module 2 runs on the ground terminal 16 and includes a data parsing unit 21 and a preprocessing execution unit 22.

[0053] The data parsing unit 21 is used to parse the raw data packet and extract the A-scan full-wavelength echo data of each channel, the corresponding depth value, tilt value, sensor number and acquisition time.

[0054] Preprocessing execution unit 22 executes the following algorithms sequentially:

[0055] A Butterworth bandpass filter algorithm for 100kHz-500kHz, with an order of 4, filters out low-frequency fluid noise and high-frequency electrical noise.

[0056] A linear time gain compensation algorithm with a gain coefficient of 0.1 dB / μs is used to compensate for the attenuation of sound waves propagating in mud.

[0057] Outlier removal algorithm: Remove invalid data with mud cake thickness exceeding the range of 0mm-50mm;

[0058] The bad channel shielding algorithm shields faulty sensor channels with an effective echo ratio of less than 30%.

[0059] Core computing module:

[0060] The core computing module 3 runs on the ground terminal 16 and includes a global cylindrical coordinate system registration unit 31, a circumferential detection area division unit 32, a multi-source data fusion unit 33, and a dual-interface recognition and thickness calculation unit 34.

[0061] Global cylindrical coordinate system registration unit 31: Establishes a global cylindrical coordinate system with the central axis of the measured arc-shaped sidewall as the Z-axis, the depth direction of the groove segment as the positive Z-axis direction, the intersection of the axis and the top surface of the groove segment as the origin, and the direction of the curvature center as the radial direction. Also establishes a global cylindrical coordinate system with the circumferential direction of the arc-shaped sidewall as the angle axis θ. Effective echo data is mapped to the curved surface of the arc-shaped sidewall, completing the coordinate transformation from cylindrical to planar: x = R × θ, y = z, where x is the circumferential unfolded length, y is the depth, R = 0.6 meters is the radius of curvature of the arc-shaped sidewall, θ is the circumferential angle, and z is the depth. Based on the pitch and roll angle data of the carrier collected by the dual-axis tilt sensor 13, the actual spatial position of each ultrasonic sensor in the global cylindrical coordinate system is calculated through coordinate transformation. This corrects the measurement error of the normal detection distance between each sensor and the arc-shaped sidewall caused by the carrier's tilt, with a correction accuracy of ±0.5 mm.

[0062] Circumferential detection area division unit 32: The preset ultrasonic sensor operating frequency is 300kHz, -6dB half-diffusion angle θ=18°, rated normal detection distance L=100mm, and circumferential overlap coefficient α=20%. Calculate the effective coverage arc length of a single sensor: Since R / L=0.6 / 0.1=6≥5, the formula W=2×L×tanθ=2×0.1×tan18°≈0.0649m=64.9mm is used. Calculate the circumferential center arc length distance between adjacent sensors: S=W×(1-α)=64.9×0.8≈51.9mm. The designed arc length of the measured arc sidewall is B=π×1.2 / 2≈1.885m, therefore the required number of sensors N≥B / S=1.885 / 0.0519≈36.3, rounded up to 37 sensors. Based on the installation angle and effective coverage arc length of each sensor, the effective beam coverage arc segment of each sensor is defined, the overlapping coverage area of ​​adjacent sensors is identified, and the arc-shaped sidewall is divided into single-probe coverage area and multi-probe overlapping coverage area.

[0063] Multi-source data fusion unit 33: For coordinate points in the coverage area of ​​a single probe, the thickness value and echo amplitude measured by the sensor are directly used. For coordinate points in the overlapping coverage area of ​​multiple probes, the beam center weighting coefficient method is used for fusion.

[0064] Calculate the normal vertical distance d from the target coordinate point to the axis of each sensor beam covering that point. i ;

[0065] Calculate the original weight coefficients ,in In this embodiment, the effective coverage arc length of the i-th ultrasonic sensor is half the arc length. =64.9 / 2≈32.45mm;

[0066] The original weight coefficients are normalized: ;

[0067] The thickness value after fusion The amplitude of the fused echo H i Let A be the original value of the mud cake thickness measured by the i-th ultrasonic sensor. i This is the original value of the echo amplitude measured by the i-th ultrasonic sensor.

[0068] If the thickness value measured by a sensor exceeds the range of 0mm-50mm, or the echo signal-to-noise ratio is lower than 20dB, its weighting coefficient is set to 0 and it is not included in the fusion.

[0069] Dual-interface recognition and thickness calculation unit 34: Employs a dual-interface feature peak recognition algorithm to identify the front interface echo peak corresponding to the mud-mud interface and the rear interface echo peak corresponding to the mud-mud-concrete wall interface. Based on the on-site calibrated ultrasonic propagation velocity of mud c=1480m / s, the mud-mud thickness is calculated using the formula H=(t1-t2)×c / 2, where t1 is the propagation time of the front interface echo peak and t2 is the propagation time of the rear interface echo peak. A two-dimensional matrix of the full-section mud-mud thickness is generated by traversing all coordinate points, with a matrix resolution of 1cm×1cm.

[0070] The implementation process of the dual-interface feature peak recognition algorithm is described below:

[0071] Ultrasonic waves are reflected at the interface between media with different acoustic impedances. The amplitude, phase, and rise slope of the reflected wave are positively correlated with the acoustic impedance difference between the media on both sides of the interface. The working environment of this invention involves three media: mud, mud cake, and concrete wall, with acoustic impedance characteristics as follows: For ease of understanding and description, the acoustic impedances in mud, mud cake, and concrete wall are represented by Z1, Z2, and Z3, respectively, as follows:

[0072] Mud: Acoustic impedance Z1≈1.45×10 6 Pa・s / m

[0073] Mud slab: Acoustic impedance Z2≈2.2×10 6 Pa・s / m

[0074] Concrete wall: Acoustic impedance Z3≈8.0×10 6 Pa・s / m

[0075] Since Z3-Z2>>Z2-Z1, the reflected wave amplitude at the interface between the mud skin and the concrete is significantly greater than that at the interface between the mud slurry and the mud skin. This is the core basis for distinguishing the two interfaces.

[0076] Specific criteria for identifying characteristic peaks:

[0077] The "multi-feature joint determination" method is used to identify dual interfaces by sequentially applying the following four conditions:

[0078] Condition 1, Amplitude Threshold Condition: Front interface between mud and mud cake: Echo amplitude A1 ≥ 0.15 × A max , where A max This represents the maximum amplitude of the entire wave train in this channel; the back interface between the mud cake and concrete: echo amplitude A2 ≥ 0.4 × A max And A2≥1.5×A1; this threshold can achieve an accuracy of over 95% in mud environments with different specific gravities, such as mud with a specific gravity of 1.05-1.25 g / cm³, thus it is sufficient to clearly distinguish them.

[0079] Condition 2, Time Sequence Condition: The propagation time t1 of the echo peak at the front interface must be less than the propagation time t2 of the echo peak at the rear interface, and the time difference Δt = t2 - t1 must satisfy: 0 < Δt ≤ 67.6 μs, corresponding to a maximum mud cake thickness of 50 mm and a mud sound velocity c = 1480 m / s.

[0080] Condition 3, Rising Edge Slope Condition: The rising edge slope of the echo peak k = ΔA / Δt ≥ 0.05V / μs is used to distinguish between real interface reflections and random noise, where ΔA = A2 - A1.

[0081] Condition 4: The full width at half maximum (FWHM) of the echo peak is ≤2μs, which is used to distinguish between interface reflections and scattered echoes from bubbles and impurities in the mud.

[0082] When the mud cake thickness is ≤1mm, the echo peaks of the front and back interfaces partially or completely overlap. In this case, a separation algorithm combining wavelet transform and Gaussian fitting is used, and the specific steps are as follows:

[0083] The overlapping echo signals are subjected to 3-level db4 wavelet decomposition to extract the second-level detail coefficients and enhance the high-frequency features of the signal. Then, a double Gaussian function is used to fit the enhanced signal; the fitting formula is as follows:

[0084] Where A1, t1, and σ1 are the amplitude, center time, and standard deviation of the peak at the front interface, respectively; and A2, t2, and σ2 are the corresponding parameters of the peak at the rear interface. The fitting parameters are solved using the least squares method to obtain the separated t1 and t2, and then the mud cake thickness is calculated. Following this method, mud cake interfaces with a thickness ≥0.3 mm can be accurately separated, meeting millimeter-level detection accuracy requirements and providing a reliable basis for the construction of the next trench section.

[0085] 2D heatmap visualization module:

[0086] The two-dimensional heat map visualization module 4 runs on the ground terminal 16 and includes a coordinate mapping unit 41, a heat map rendering unit 42, and an interactive annotation unit 43.

[0087] Coordinate mapping unit 41: Establishes a one-to-one correspondence between two-dimensional plane coordinates x, y and mud thickness H.

[0088] Heatmap rendering unit 42: Employs bilinear interpolation to interpolate discrete thickness data points into continuous 1cm × 1cm grid data. Uses a linear gradient color mapping algorithm.

[0089] Thickness 0mm: RGB(0,0,255) (Blue)

[0090] Thickness 2.5mm: RGB(0,255,0) (Green)

[0091] 3mm thickness: RGB(255,255,0) (yellow, warning threshold)

[0092] Thickness ≥ 5mm: RGB(255,0,0) (Red, unacceptable threshold)

[0093] A two-dimensional unfolded heatmap is generated by rendering pixel by pixel based on the interpolated grid data.

[0094] Interactive annotation unit 43: Employs the Canny edge detection algorithm to extract the contours of areas with thicknesses ≥3mm and ≥5mm. A rectangular annotation box is generated for each exceeding the standard area: a yellow dashed box corresponds to the 3mm warning area, and a red solid box corresponds to the 5mm non-compliant area. The annotation box displays the maximum thickness and location coordinates of the area. Global statistical information is displayed at the top of the heatmap: maximum mud cake thickness across the entire cross-section, average mud cake thickness, and percentage of the exceeding area. Mouse hover is supported to view the precise thickness, depth, and circumferential position of any point, and scroll wheel zoom and mouse drag pan operations are supported.

[0095] The results export module 5 runs on the ground terminal 16 and supports: exporting PNG format heat map images with a resolution of 300 DPI, which automatically include the project name, trench number, detection time and legend;

[0096] Export the full-section thickness data matrix in CSV format; export the out-of-range area details table in Excel format, including the location, maximum thickness, average thickness, and area of ​​each out-of-range area. In this embodiment, algorithms not specifically described are existing technologies and are not key technical improvements of this invention, and therefore will not be detailed.

[0097] The technical advantages of this embodiment are as follows: This embodiment converts the full-section mud cake thickness data into a two-dimensional unfolded heat map, uses color gradients to intuitively display the mud cake thickness distribution and automatically marks areas exceeding the standard. Combined with interactive functions such as mouse hover parameter checking, zooming and panning, it enables on-site construction personnel to quickly judge the wall brushing quality. At the same time, it uses a global cylindrical coordinate system for data registration and cylindrical-plane coordinate transformation, perfectly adapting to the concave arc-shaped sidewall surface formed after the joint pipe is removed, solving the coordinate deviation problem caused by the plane coordinate system mapping. The defect positioning accuracy can reach the centimeter level, and it supports one-click export of high-definition heat maps, thickness data matrices and detailed tables of areas exceeding the standard, meeting the data requirements for on-site construction records and project acceptance.

[0098] Meanwhile, by using an arc-shaped array ultrasonic sensor with a valve plug in conjunction with the insulated and acoustically permeable dimethyl silicone oil filling inside the pipe, passive differential pressure adaptive balance is achieved in ultra-deep and high-pressure environments, ensuring stable operation of the sensor under the conditions of an ultra-deep seepage barrier wall at the hundred-meter level. A time-division polling triggering algorithm is adopted to avoid acoustic crosstalk between multiple sensors, and a reliable spatial reference is provided by synchronous binding of millisecond-level timestamps of multi-source data. Combined with a dual-interface recognition algorithm with multi-feature joint judgment and overlapping peak separation technology, millimeter-level detection accuracy of ≥0.3mm thin mud skin is achieved. At the same time, a curvature correction formula is introduced to ensure accurate calculation of the coverage range of sidewalls with different curvatures. The beam center weighting coefficient method is used to fuse multi-probe data to suppress random errors, and a standardized preprocessing process is used to filter out clutter interference, comprehensively improving the accuracy, consistency and robustness of the detection data.

[0099] Example 2:

[0100] This embodiment describes the specific process of visual inspection after brushing the curved sidewall of the seepage barrier with a joint pipe diameter of 1.2 meters using the system described in Embodiment 1. The depth of the inspection trench is 80 meters.

[0101] The testing steps are as follows: System calibration and parameter preset: The on-site calibration of the ultrasonic propagation velocity in the mud is 1480 m / s, the preset rated normal detection distance is 100 mm, the circumferential overlap coefficient is 20%, and the input parameters are: radius of curvature R of the arc-shaped sidewall = 0.6 m, design arc length B = 1.885 m, and design depth of the trench section = 80 m. Those skilled in the art can perform the calibration according to existing technical manuals, or they can complete the calibration using the following steps. Specifically, the on-site calibration method for the ultrasonic propagation velocity c in the mud is as follows: Take a mud sample from the trench section to be tested and measure the sound velocity using a sound velocity meter at the mud temperature; or place a standard reflector of known thickness in the mud section and calculate the mud sound velocity by measuring the time difference of ultrasonic wave propagation at the interface before and after the standard reflector; the calibration frequency should be no less than once per trench section, and recalibration should be performed when the mud specific gravity changes by more than ±0.05 g / cm³.

[0102] System installation: Fix the arc array ultrasonic detection sensor 11 on the carrier of the wall brush machine, so that the arc circumferential tube of the sensor is coaxially arranged with the arc side wall to be measured, and attach the limiting rotary bodies at the upper and lower ends of the carrier to the side wall surface.

[0103] Data acquisition: The carrier is lowered at a constant speed of 0.5 m / s by the ground crane. The multi-channel synchronous data acquisition card 12 triggers 37 ultrasonic sensors to collect echo data in sequence at 250 μs intervals. The depth encoder 14 and the dual-axis tilt sensor 13 synchronously collect depth and attitude data. All data are transmitted to the ground terminal 16 through the underground explosion-proof acquisition circuit 15.

[0104] Data preprocessing: Ground terminal 16 performs parsing, bandpass filtering, time gain compensation, outlier removal, and bad channel shielding on the raw data to obtain an effective echo dataset.

[0105] Core Calculation: Core calculation module 3 performs global cylindrical coordinate system registration, circumferential detection area division, overlapping area data fusion, and dual interface recognition algorithm to calculate the two-dimensional matrix of mud skin thickness across the entire cross section within a depth range of 80 meters.

[0106] Visualization: The 2D heatmap visualization module 4 renders the thickness matrix into a 2D unfolded heatmap, automatically marks the areas exceeding the standard, and displays global statistical information.

[0107] Results Export: Export heat map image, thickness data matrix, and detailed list of areas exceeding the standard.

[0108] Test results: The two-dimensional heat map of this embodiment is as follows: Figure 2 As shown, the horizontal axis of the heat map represents the circumferential unfolded length of 0-2 meters, and the vertical axis represents the depth of the trench section of 0-80 meters.

[0109] Overall statistics: The maximum mud cake thickness across the entire cross section is 4.2 mm, the average thickness is 1.8 mm, and the combined proportion of the warning area (≥3 mm) and the non-compliant area (≥5 mm) is 2.1%.

[0110] Areas exceeding standards: A total of 3 yellow warning areas with a thickness ≥3mm were detected, and no red non-compliant areas with a thickness ≥5mm were detected. The 3 warning areas are located at:

[0111] Depth 12-15 meters, circumference 0.6-0.8 meters, maximum thickness 3.7 mm;

[0112] Depth 38-40 meters, circumference 1.2-1.4 meters, maximum thickness 4.2 mm;

[0113] Depth 65-67 meters, circumference 0.3-0.5 meters, maximum thickness 3.5 mm.

[0114] According to the test results, the wall brushing quality of this section is basically qualified. Only the three warning areas mentioned above need to be brushed at specific points before the construction of the next section can proceed.

[0115] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A two-dimensional thermal mapping visualization system for the construction quality of ultra-deep seepage-proof walls, characterized in that, The system includes a data acquisition and synchronization module (1), a standardization preprocessing module (2), a core calculation module (3), a two-dimensional heat map visualization module (4), and a result export module (5) connected in sequence. The data acquisition and synchronization module (1) is used for multi-source data acquisition and timestamp synchronization binding to generate an original dataset with a spatial reference. The multi-source data includes ultrasonic echo data, depth data, and attitude data collected by an arc-shaped array ultrasonic detection sensor (11) installed on the carrier. The standardization preprocessing module (2) is used to perform noise reduction, outlier removal, and bad channel shielding on the original dataset to output an effective echo dataset. The core calculation module (3) is used to perform global cylindrical coordinate system registration, overlapping area data fusion, and dual-interface recognition algorithms based on the effective echo dataset to calculate the two-dimensional matrix of mud skin thickness across the entire cross section. The two-dimensional heat map visualization module (4) is used to convert the two-dimensional mud skin thickness matrix into a two-dimensional unfolded heat map to achieve an intuitive display of mud skin thickness and distribution.

2. The two-dimensional thermal mapping visualization system for the construction quality of ultra-deep seepage-proof walls according to claim 1, characterized in that, The data acquisition and synchronization module (1) includes an arc-shaped array ultrasonic detection sensor (11), a multi-channel synchronous data acquisition card (12), a dual-axis tilt sensor (13), a depth encoder (14), an explosion-proof acquisition circuit (15), and a ground terminal (16) mounted on a carrier. The arc-shaped array ultrasonic detection sensor (11) includes an arc-shaped circumferential tube, on which radial tubes are fixedly connected in a circumferential array. A valve plug for eliminating the internal and external pressure difference is provided at the end of the circumferential tube. The multi-channel synchronous data acquisition card (12) uses a time-division polling triggering algorithm to control each ultrasonic sensor to emit ultrasonic pulses in sequence. The triggering interval between adjacent ultrasonic sensors is ≥200μs.

3. The two-dimensional thermal mapping visualization system for the construction quality of ultra-deep seepage-proof walls according to claim 2, characterized in that, The data acquisition and synchronization module (1) further includes a multi-source data timestamp synchronization unit (17) and a data integrity verification unit (18); the multi-source data timestamp synchronization unit (17) is used to add the same millisecond-level timestamp to the echo data, depth data and tilt data at the same acquisition time to realize the spatial reference binding of multi-source data; the data integrity verification unit (18) uses the CRC32 verification algorithm to verify the transmitted data packets.

4. The two-dimensional thermal mapping visualization system for the construction quality of ultra-deep seepage-proof wall brushing as described in claim 1, characterized in that, The standardized preprocessing module (2) includes a data parsing unit (21) and a preprocessing execution unit (22). The data parsing unit (21) is used to parse the raw data packet and extract echo data, depth data and tilt data of each channel. The preprocessing execution unit (22) executes the following algorithms in sequence: 100kHz-500kHz Butterworth bandpass filtering algorithm to filter out noise in the mud environment; linear time gain compensation algorithm to compensate for the propagation attenuation of sound waves in the mud; outlier removal algorithm to remove invalid data with mud skin thickness exceeding the range of 0-50mm; and bad channel shielding algorithm to shield faulty sensor channels with an effective echo ratio lower than a preset threshold.

5. The two-dimensional thermal mapping visualization system for the construction quality of ultra-deep seepage-proof walls according to claim 1, characterized in that, The core calculation module (3) includes a global cylindrical coordinate system registration unit (31); the global cylindrical coordinate system registration unit (31) is used to establish a global cylindrical coordinate system with the central axis of the measured arc sidewall as the Z axis, the depth direction of the groove as the positive direction, the intersection of the axis and the top surface of the groove as the coordinate origin, and the direction of the curvature center as the radial direction, to map the effective echo data to the arc sidewall surface; and to execute the cylindrical and planar coordinate transformation algorithm to convert the (θ,z) coordinates in the cylindrical coordinate system into (x,y) coordinates in the two-dimensional unfolded plane, where x=R×θ, y=z, R is the design curvature radius of the arc sidewall, θ is the circumferential angle, and z is the depth; at the same time, according to the attitude data collected by the dual-axis tilt sensor (13), the normal distance deviation between the ultrasonic sensor and the arc sidewall is corrected.

6. The two-dimensional thermal mapping visualization system for the construction quality of ultra-deep anti-seepage walls according to claim 5, characterized in that, The core calculation module (3) also includes a circumferential detection area division unit (32); the circumferential detection area division unit (32) is used to calculate the effective coverage arc length W of a single ultrasonic sensor based on the -6dB half-diffusion angle θ, the rated normal detection distance L and the radius of curvature R of the arc sidewall of the ultrasonic sensor: when R / L≥5, W=2×L×tanθ; when R / L<5, the curvature correction formula W=2×R×arcsin((RL)×tanθ / R) is used; and the effective beam coverage arc segment is delineated according to the installation coordinates of each ultrasonic sensor, the overlapping coverage area of ​​adjacent sensors is identified, and the arc sidewall is divided into a single probe coverage area and a multi-probe overlapping coverage area.

7. The two-dimensional thermal mapping visualization system for the construction quality of ultra-deep seepage-proof walls according to claim 6, characterized in that, The core computing module (3) also includes a multi-source data fusion unit (33); the multi-source data fusion unit (33) uses the beam center weighting coefficient method to fuse the data of the overlapping coverage area of ​​multiple probes: calculating the normal vertical distance from the target coordinate point to the beam axis of each ultrasonic sensor covering that point. ; Calculate the original weight coefficients , ;in, Let be the half-arc length of the arc segment effectively covered by the i-th ultrasonic sensor; normalize the original weighting coefficients to obtain , ;in, is the normalized weighting coefficient, and n is the number of ultrasonic sensors covering the target coordinate point; the thickness value and echo amplitude of each sensor are weighted and summed using the normalized weighting coefficient to obtain the fused thickness value and echo amplitude.

8. The two-dimensional thermal mapping visualization system for the construction quality of ultra-deep seepage-proof walls according to claim 7, characterized in that, The core calculation module (3) also includes a dual-interface recognition and thickness calculation unit (34); the dual-interface recognition and thickness calculation unit (34) adopts a dual-interface feature peak recognition algorithm to identify the front interface echo peak corresponding to the mud-slurry and mud cake interface and the rear interface echo peak corresponding to the mud cake and concrete wall interface; and according to the formula Calculate the mud cake thickness, where t1 is the propagation time of the echo peak at the front interface, t2 is the propagation time of the echo peak at the rear interface, and c is the ultrasonic propagation velocity of the mud calibrated on site.

9. The two-dimensional thermal mapping visualization system for the construction quality of ultra-deep seepage-proof walls according to claim 1, characterized in that, The two-dimensional heat map visualization module (4) includes a coordinate mapping unit (41), a heat map rendering unit (42), and an interactive annotation unit (43). The coordinate mapping unit (41) is used to establish the correspondence between two-dimensional plane coordinates and mud thickness values. The heat map rendering unit (42) uses a bilinear interpolation algorithm to interpolate discrete thickness data into continuous grid data, and uses a linear gradient color mapping algorithm to render the heat map. It also uses a color mapping algorithm to convert mud thickness values ​​into different colors and automatically annotates warning and unqualified areas. The interactive annotation unit (43) is used to automatically extract and annotate areas with thicknesses ≥3mm and ≥5mm that exceed the standard. The annotation content includes the maximum thickness of the area and its location coordinates. It also supports hovering the mouse to view the precise thickness, depth, and circumferential position of any point and supports zooming and panning operations.

10. The two-dimensional thermal mapping visualization system for the construction quality of ultra-deep seepage-proof wall brushing according to claim 1, characterized in that, It also includes a result export module (5), which is communicatively connected to the two-dimensional heat map visualization module (4); the result export module (5) is used to export the high-definition heat map image of the current view, the full-section mud skin thickness data matrix and the list of areas exceeding the standard; wherein, the exported high-definition image has a printing resolution of ≥300DPI and a pixel size of not less than 2480×3508, the thickness data matrix is ​​in CSV format, and the list of areas exceeding the standard contains the location, maximum thickness, average thickness and area information of each area exceeding the standard.

Images