Method for carrying out pipeline ultraviolet curing trenchless repair by combining robot

Abstract

The invention relates to a method for carrying out pipeline ultraviolet curing non-excavation repair by combining a robot, which comprises the following steps of: S1, entering a pipeline by using a detection robot, and collecting point cloud, a high-definition image and thermal imaging data; s2, constructing a three-dimensional grid model, completely restoring the internal geometrical shape and defect distribution state of the pipeline, and labeling a defect area in the model; s3, according to the type and positioning of the defect area, a repair task strategy is obtained, and the repair task strategy comprises division of repair segments and ultraviolet light curing parameters; s4, the repair task strategy is input into the three-dimensional grid model for virtual repair simulation, and a final repair scheme is obtained after parameters are optimized; and S5, based on the final repairing scheme, repairing is carried out based on a repairing robot, and the repairing state is detected in real time through a camera. The precision and efficiency of non-excavation repair of the pipeline are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of trenchless pipeline repair, and in particular to a method for trenchless pipeline repair by ultraviolet light curing combined with a robot. Background Art

[0002] With the acceleration of urbanization, underground pipelines, as an important part of urban infrastructure, have become increasingly important for maintenance and repair. However, traditional pipeline repair methods usually require large-scale excavation, with long construction periods, high costs, and are likely to have negative impacts on traffic, the environment, and surrounding facilities. Therefore, trenchless pipeline repair technology has been widely used in recent years, among which UV-CIPP (Ultraviolet Cured-in-Place Pipe) has become one of the mainstream technologies in the field of trenchless repair due to its high efficiency, environmental protection, and low interference.

[0003] UV curing repair uses special resin materials, which are exposed to ultraviolet light by robots, so that the materials can be quickly cured inside the pipe, forming a new lining structure to restore or enhance the strength of the pipe. This method avoids tedious excavation and replacement processes, greatly reducing the impact of construction on the environment. However, the internal defects of the pipeline are complex and diverse, such as cracks, corrosion, perforations and blockages, which are widely distributed and have great differences in characteristics. Therefore, traditional repair methods still face many challenges in how to accurately repair the target area, optimize the repair efficiency and ensure the quality of the repair. Summary of the invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide a method for performing UV-curing trenchless repair of pipelines in combination with a robot, which effectively improves the accuracy and efficiency of trenchless repair of pipelines.

[0005] To achieve the above object, the present invention adopts the following technical solutions: A method for performing UV-curing trenchless repair of pipelines in combination with a robot comprises the following steps: S1: Use the inspection robot to enter the pipeline, perform 3D laser scanning, panoramic photography and thermal imaging monitoring, and collect point cloud, high-definition image and thermal imaging data; S2: Based on the collected point cloud, high-definition images, and thermal images, a three-dimensional mesh model is constructed to completely restore the internal geometry and defect distribution of the pipeline, and the defect areas in the model are marked; S3: According to the defect area type and location, the repair task strategy is obtained, including the division of repair sections and UV curing parameters; S4: Input the repair task strategy into the three-dimensional grid model to perform virtual repair simulation, and obtain the final repair plan after optimizing the parameters; S5: Based on the final repair plan, the repair robot performs repairs and uses the camera to monitor the repair status in real time.

[0006] Furthermore, the S1 is specifically: The detection robot moves at a constant speed along the inner wall of the pipeline and performs real-time positioning: ; in, is the robot position at time t; is the robot position at time t-1; is the displacement between two points; The rotation matrix provided for the attitude sensor; The laser scanner collects point clouds on the pipeline surface at a data density of S points per second, and the laser ranges the i-th point L i Convert to point cloud midpoint (x i , y i , z i ) ; ; Where Z is the step length of the robot moving along the pipeline axis, θ i is the relative rotation angle, and finally a complete three-dimensional point cloud file P (x, y, z) is generated; The panoramic camera takes a picture once per step Δd, and the image data is stitched together to form an expandable panoramic inner wall map. The relative position is determined by the following formula: ; in, is the i-th picture; The thermal imaging sensor scans the infrared temperature data of the inner wall of the pipeline and records the temperature distribution matrix T.

[0007] Furthermore, the S2 is specifically: Pre-process the acquired point cloud, high-definition image, and thermal imaging data; Based on the preprocessed point cloud data, a three-dimensional mesh model of the pipeline surface is constructed using a triangular mesh reconstruction method; Identify defect areas and extract features through point cloud analysis, high-definition image edge detection, and thermal imaging temperature difference analysis; determine risk levels based on the extracted features; Mapping defect texture and temperature difference data to the surface of the 3D mesh model, and marking the defect position and category on the surface of the 3D mesh model; Establish pipeline axis reference and unify coordinates of point clouds, high-definition images and thermal images.

[0008] Furthermore, the preprocessing is specifically as follows: Point cloud data uses statistical filtering to remove isolated noise points: ; in, D(p i ) For point p i The density reference value, k is the standard deviation ratio; P filtered is the point cloud set retained after statistical filtering; mean(D) is the density reference value D(p i )’s mean; is the density reference value D(p i )’s standard deviation; Use the ICP algorithm to register the segmented point clouds and obtain complete three-dimensional information of the pipeline: ; Among them, T is the registration transformation matrix, R is the rotation matrix, is the offset vector, p i and q i Concentrate corresponding points on two point clouds; The high-definition image data is stitched together using a feature point matching algorithm to enhance contrast, filter out noise, fuse the high-definition image data into a three-dimensional mesh model, and generate a three-dimensional mesh model with surface texture through texture mapping corresponding to each point in the point cloud; Based on the thermal imaging data, the temperature distribution matrix is ​​extracted and the surface information of the point cloud is matched by interpolation.

[0009] Furthermore, the defect area is identified and features are extracted through point cloud analysis, high-definition image edge detection, and thermal imaging temperature difference analysis; based on the extracted features, the risk level is determined as follows: Using the geometric mutation of the point cloud surface and the texture edge detection of the image, the crack area is identified and the crack length L is obtained. crack and crack depth D crack : Corrosion is identified by matching the curvature depression of point cloud data with the high temperature area of ​​thermal imaging, and the corrosion depth is obtained. D corrosion and corrosion area A; The thermal imaging temperature difference is superimposed on the point cloud surface, the area and center position of the high temperature area are calculated, and the heat distribution is mapped to the surface of the three-dimensional grid model; According to the acquired abnormal features, the risk score P is calculated by the weighted model: ; Among them, w1, w2, w3, w4, w5 are defect risk contribution coefficients obtained through machine learning training; L crack is the crack length; D crack is the crack depth, and ΔT represents the temperature difference.

[0010] Furthermore, the geometric mutation of the point cloud surface and the texture edge detection of the image are used to identify the crack area and obtain the crack length L crack and crack depth D crack , as follows: By combining the geometric mutation characteristics of the point cloud and the edge information of the cloud surface texture, the cracks are located and the principal curvature of the point cloud is calculated: K=k1+k2; Among them, k1 and k2 are the two principal curvatures of the local surface of the point cloud, and K represents the degree of mutation of the local surface; If the curvature K>K threshold , marked as possible crack areas, where K threshold is the preset threshold; Calculate the change gradient ΔN of the normal vector N in the point neighborhood: ; Among them, N a and N b is the normal vector of the adjacent point; if ΔN>ΔN threshold , marked as the geometric mutation point, where ΔN threshold is the preset threshold; The Canny operator is used to detect the crack edge and obtain the crack edge binary image E(x,y); the skeleton S(x,y) is obtained by thinning method; Overlay the edge of the HD image onto the point cloud and match the point cloud and image by coordinate registration: ; in, is the three-dimensional coordinate of the point cloud; (u, v) is the image texture coordinate; I(u, v) is the pixel value corresponding to the image texture coordinate (u, v); It is the inverse mapping function from image texture coordinates to point cloud three-dimensional coordinates; Accumulate the distance along the skeleton points of the crack edge to obtain the crack length L crack ; Calculate the difference between the bottom point of the crack and the fitted smooth surface to obtain the crack depth D crack .

[0011] Furthermore, the curvature depression of the point cloud data is matched with the high temperature area of ​​the thermal imaging to identify corrosion and obtain the corrosion depth. D corrosion And the corrosion area A, as follows: Corrosion is identified by matching the curvature concavity of the point cloud data with the high temperature area of ​​the thermal imaging, the smooth surface is fitted by the least squares method, and the local surface height deviation of the point cloud is calculated: ; Among them, z actual is the height of the actual point in the point cloud; z smooth is the height value of the fitted smooth surface; if D corrosion >D threshold , defined as the corrosion pit, D threshold is the preset threshold; The high temperature area is extracted from the thermal imaging temperature difference matrix, and the high temperature area is matched with the concave area of ​​the point cloud to identify the corrosion concave point. The local surface height deviation of the point cloud corresponding to the corrosion concave point is the corrosion depth D of the corrosion concave point. corrosion , and calculate the corrosion area A.

[0012] Furthermore, the S3 is specifically as follows: Automatically divide the repair sections according to the defect type, distribution and location characteristics to determine whether overall repair or partial repair is required; According to the length of the crack L crack , depth D crack And the position calculation distribution density ρ crack ;like , divided into overall repair sections, where The preset threshold of distribution density; if ρ crack <ρ threshold , divided into local repair segments; According to the corrosion area A and corrosion depth D corrosion , calculate the corrosion ratio ,in is the pipe area; if , it is divided into the overall repair section, in which, Preset threshold for corrosion ratio, if , it is divided into local repair segments; For a local repair segment, calculate the starting position and length of the repair segment; For cracks, repair section length L segment for: ; in, The buffer repair length at both ends of the crack; For corrosion, repair section length L segment for: ; Formulate curing parameters according to defect characteristics, including light power, irradiation time and curing speed; Optical power P UV and crack depth D crack Proportional to: ; Among them, g1 is the influence coefficient of crack depth on optical power, and g2 is the basic power; like , need to increase power: ; Optical power and corrosion depth D corrosion Proportional to the corrosion area A: ; Among them, k3 and k4 influence coefficients; Irradiation time T UV By repairing the segment length L segment and curing speed v cure Decide: ; Adjust the curing speed according to the defect depth and distribution density: ; in, Is the basic curing speed, is the influence coefficient of crack density on solidification speed; is the influence coefficient of corrosion depth on solidification speed; Based on the risk score P and defect location, high-risk segments are repaired first.

[0013] Furthermore, the S4 is specifically as follows: Input the AI-generated repair task strategy into the digital twin environment; Simulate the repair process in a 3D mesh model and evaluate the repair effect; According to the simulation results, combined with the target repair effect and the constraints of consumables / energy consumption, the repair parameters are further optimized and the final optimized repair plan is output: the optimal repair segment division, light curing parameters and repair priority are obtained.

[0014] Furthermore, the S5 is specifically as follows: According to the optimized repair plan, control the repair robot to complete the repair tasks of cracks, corrosion, or blockage, including UV curing repair and resin filling operations; The repair robot's built-in high-definition camera and depth sensor can be used to collect the execution status of the repair task in real time; According to the detection results during the repair process, the repair parameters are dynamically adjusted to ensure the best repair effect.

[0015] The present invention has the following beneficial effects: 1. The present invention uses laser scanning, high-definition images and thermal imaging to achieve comprehensive perception of the interior of the pipeline, and builds a digital twin model to completely restore the three-dimensional geometric structure and defect distribution inside the pipeline. It organically combines point cloud, high-definition image and thermal imaging data to improve defect location and quantification accuracy, and conducts risk assessment based on defect characteristics, while achieving intuitive presentation through 3D visualization; 2. The present invention intelligently plans the local repair or overall repair section according to the defect type (crack / corrosion / blockage) and distribution status, avoids blind repair, improves the repair efficiency, and dynamically adjusts the UV light power, curing time and speed according to the defect characteristics (depth, area, etc.) to ensure the maximum curing effect and strength of the material; 3. Before actual repair, the present invention simulates the UV curing process, optimizes the repair parameters based on the physical model and simulation algorithm, can automatically divide the repair stages according to the defect characteristics and formulate targeted UV curing repair parameters, and optimizes the amount of repair materials and the repair sequence, thereby achieving efficient and accurate pipeline repair. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0017] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments: refer to Figure 1 In this embodiment, a method for performing UV-curing trenchless repair of pipelines in combination with a robot is provided, comprising the following steps: S1: Use the inspection robot to enter the pipeline, perform 3D laser scanning, panoramic photography and thermal imaging monitoring, and collect point cloud, high-definition image and thermal imaging data; S2: Based on the collected point cloud, high-definition images, and thermal images, a three-dimensional mesh model is constructed to completely restore the internal geometry and defect distribution of the pipeline, and the defect areas in the model are marked (such as high-risk points and crack distribution areas); S3: According to the defect area type and location, obtain the repair task strategy, including the division of repair sections (overall repair or local repair) and UV curing parameters (light power, irradiation time); S4: Input the repair task strategy into the three-dimensional grid model to perform virtual repair simulation, and obtain the final repair plan after optimizing the parameters; S5: Based on the final repair plan, the repair robot performs repairs and uses the camera to monitor the repair status in real time.

[0018] In this embodiment, S1 is specifically: The detection robot moves at a constant speed along the inner wall of the pipeline and performs real-time positioning: ; in, is the robot position at time t; is the robot position at time t-1; is the displacement between two points; The rotation matrix provided for the attitude sensor; The laser scanner collects point clouds on the pipeline surface at a data density of S points per second, and the laser ranges the i-th point L i Convert to point cloud midpoint (x i , y i , z i ) ; ; Where Z is the step length of the robot moving along the pipeline axis, θ i is the relative rotation angle, and finally a complete three-dimensional point cloud file P (x, y, z) is generated; The panoramic camera takes a picture once per step Δd, and the image data is stitched together to form an expandable panoramic inner wall map. The relative position is determined by the following formula: ; in, is the i-th picture; The thermal imaging sensor scans the infrared temperature data of the inner wall of the pipeline and records the temperature distribution matrix T.

[0019] In this embodiment, S2 is specifically: Pre-process the acquired point cloud, high-definition image, and thermal imaging data; Based on the preprocessed point cloud data, a three-dimensional mesh model of the pipeline surface is constructed using a triangular mesh reconstruction method (such as Delaunay triangulation); Identify defect areas and extract features through point cloud analysis, high-definition image edge detection, and thermal imaging temperature difference analysis; determine risk levels based on the extracted features; Map defect texture (cracks, corrosion, etc.) and temperature difference data to the surface of the 3D mesh model, and mark the defect location and category on the surface of the 3D mesh model: Establish pipeline axis reference and unify coordinates of point clouds, high-definition images and thermal images.

[0020] In this embodiment, the preprocessing is as follows: Point cloud data uses statistical filtering to remove isolated noise points: ; in, D(p i ) For point p i The density reference value, k is the standard deviation ratio; P filtered is the point cloud set retained after statistical filtering; mean(D) is the density reference value D(p i )’s mean; is the density reference value D(p i )’s standard deviation; Use the ICP algorithm to register the segmented point clouds and obtain complete three-dimensional information of the pipeline: ; Among them, T is the registration transformation matrix, R is the rotation matrix, is the offset vector, p i and q i Concentrate corresponding points on two point clouds; Use feature point matching algorithms (such as SIFT) to stitch the original single-frame images, enhance contrast, filter out noise, fuse the HD image data into a 3D mesh model, and generate a 3D mesh model with surface texture through texture mapping corresponding to each point in the point cloud; Based on the thermal imaging data, the temperature distribution matrix is ​​extracted and the surface information of the point cloud is matched by interpolation.

[0021] In this embodiment, the defect area is identified and features are extracted through point cloud analysis, high-definition image edge detection, and thermal imaging temperature difference analysis; based on the extracted features, the risk level is determined as follows: Using the geometric mutation of the point cloud surface and the texture edge detection of the image, the crack area is identified and the crack length L is obtained. crack and crack depth D crack : Corrosion is identified by matching the curvature depression of point cloud data with the high temperature area of ​​thermal imaging, and the corrosion depth is obtained. D corrosion and corrosion area A; The thermal imaging temperature difference is superimposed on the point cloud surface, the area and center position of the high temperature area are calculated, and the heat distribution is mapped to the surface of the three-dimensional grid model; According to the acquired abnormal features, the risk score P is calculated by the weighted model: ; Among them, w1, w2, w3, w4, w5 are defect risk contribution coefficients obtained through machine learning training; L crack is the crack length; D crack is the crack depth, and ΔT represents the temperature difference.

[0022] In this embodiment, the geometric mutation of the point cloud surface and the texture edge detection of the image are used to identify the crack area and obtain the crack length L crack and crack depth D crack , as follows: By combining the geometric mutation characteristics of the point cloud and the edge information of the cloud surface texture, the cracks are located and the principal curvature of the point cloud is calculated: K=k1+k2; Among them, k1 and k2 are the two principal curvatures of the local surface of the point cloud, and K represents the degree of mutation of the local surface; If the curvature K>K threshold , marked as possible crack areas, where K threshold is the preset threshold; Calculate the change gradient ΔN of the normal vector N in the point neighborhood: ; Among them, N a and N b is the normal vector of the adjacent point; if ΔN>ΔN threshold , marked as the geometric mutation point, where ΔN threshold is the preset threshold; The Canny operator is used to detect the crack edge and obtain the crack edge binary image E(x,y); the skeleton S(x,y) is obtained by thinning method; Overlay the edge of the HD image onto the point cloud and match the point cloud and image by coordinate registration: ; in, is the three-dimensional coordinate of the point cloud; (u, v) is the image texture coordinate; I(u, v) is the pixel value corresponding to the image texture coordinate (u, v); It is the inverse mapping function from image texture coordinates to point cloud three-dimensional coordinates; Accumulate the distance along the skeleton points of the crack edge to obtain the crack length L crack ; Calculate the difference between the bottom point of the crack and the fitted smooth surface to obtain the crack depth D crack .

[0023] In this embodiment, the corrosion is identified by matching the curvature concave of the point cloud data with the high temperature area of ​​the thermal imaging, and the corrosion depth is obtained. D corrosion And the corrosion area A, as follows: Corrosion is identified by matching the curvature concavity of the point cloud data with the high temperature area of ​​the thermal imaging, the smooth surface is fitted by the least squares method, and the local surface height deviation of the point cloud is calculated: ; Among them, z actual is the height of the actual point in the point cloud; z smooth is the height value of the fitted smooth surface; if D corrosion >D threshold , defined as the corrosion pit, D threshold is the preset threshold; The high temperature area is extracted from the thermal imaging temperature difference matrix, and the high temperature area is matched with the concave area of ​​the point cloud to identify the corrosion concave point. The local surface height deviation of the point cloud corresponding to the corrosion concave point is the corrosion depth D of the corrosion concave point. corrosion , and calculate the corrosion area A.

[0024] In this embodiment, S3 is as follows: Automatically divide the repair sections according to the defect type, distribution and location characteristics to determine whether overall repair or partial repair is required; According to the length of the crack L crack , depth D crack And the position calculation distribution density ρ crack ;like , divided into overall repair sections, where The preset threshold of distribution density; if ρ crack <ρ threshold , divided into local repair segments; According to the corrosion area A and corrosion depth D corrosion , calculate the corrosion ratio ,in is the pipe area; if , it is divided into the overall repair section, in which, Preset threshold for corrosion ratio, if , it is divided into local repair segments; For a local repair segment, calculate the starting position and length of the repair segment; For cracks, repair section length L segment for: ; in, The buffer repair length at both ends of the crack; For corrosion, repair section length L segment for: ; Formulate curing parameters according to defect characteristics, including light power, irradiation time and curing speed; Optical power P UV and crack depth D crack Proportional to: ; Among them, g1 is the influence coefficient of crack depth on optical power, and g2 is the basic power; like , need to increase power: ; Optical power and corrosion depth D corrosion Proportional to the corrosion area A: ; Among them, k3 and k4 influence coefficients; Irradiation time T UV By repairing the segment length L segment and curing speed v cure Decide: ; Adjust the curing speed according to the defect depth and distribution density: ; in, Is the basic curing speed, is the influence coefficient of crack density on solidification speed; is the influence coefficient of corrosion depth on solidification speed; Based on the risk score P and defect location, high-risk segments are repaired first.

[0025] In this embodiment, S4 is specifically as follows: Input the AI-generated repair task strategy (repair segment division, UV curing parameters, etc.) into the digital twin environment; Simulate the repair process in a 3D mesh model to evaluate the repair effect (such as crack filling, corrosion coverage, curing time, etc.); According to the simulation results, combined with the target repair effect and the constraints of consumables / energy consumption, the repair parameters are further optimized and the final optimized repair plan is output: the optimal repair segment division, light curing parameters and repair priority are obtained.

[0026] In this embodiment, S5 is specifically as follows: According to the optimized repair plan, control the repair robot to complete the repair tasks of cracks, corrosion, or blockage, including UV curing repair and resin filling operations; The repair robot's built-in high-definition camera and depth sensor can be used to collect the execution status of the repair task in real time (such as whether the crack is completely filled, the quality of material curing, the coverage rate of the repair area, etc.); According to the detection results during the repair process, the repair parameters are dynamically adjusted to ensure the best repair effect.

[0027] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0028] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0029] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0030] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0031] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

Claims

1. A method for non-excavation repair of pipelines by UV curing combined with a robot, characterized in that: The following steps are involved: S1: Use the inspection robot to enter the pipeline, perform 3D laser scanning, panoramic photography and thermal imaging monitoring, and collect point cloud, high-definition image and thermal imaging data; S2: Based on the collected point cloud, high-definition images, and thermal images, a three-dimensional mesh model is constructed to completely restore the internal geometry and defect distribution of the pipeline, and the defect areas in the model are marked; S3: According to the defect area type and location, the repair task strategy is obtained, including the division of repair sections and UV curing parameters; S4: Input the repair task strategy into the three-dimensional grid model to perform virtual repair simulation, and obtain the final repair plan after optimizing the parameters; S5: Based on the final repair plan, the repair robot performs repairs and uses the camera to monitor the repair status in real time.

2. The method for performing UV-curing trenchless repair of pipelines in combination with a robot according to claim 1, characterized in that: The S1 is specifically: The detection robot moves at a constant speed along the inner wall of the pipeline and performs real-time positioning: ； in, is the robot position at time t; is the robot position at time t-1; is the displacement between two points; The rotation matrix provided for the attitude sensor; The laser scanner collects point clouds on the pipeline surface at a data density of S points per second, and the laser ranges the i-th point L i Convert to point cloud midpoint (x i , y i , z i ) ; ； Where Z is the step length of the robot moving along the pipeline axis, θ i is the relative rotation angle, and finally a complete three-dimensional point cloud file P (x, y, z) is generated; The panoramic camera takes a picture once per step Δd, and the image data is stitched together to form an expandable panoramic inner wall map. The relative position is determined by the following formula: ； in, is the i-th picture; The thermal imaging sensor scans the infrared temperature data of the inner wall of the pipeline and records the temperature distribution matrix T.

3. The method for performing UV-curing trenchless repair of pipelines in combination with a robot according to claim 2, characterized in that: The S2 is specifically: Pre-process the acquired point cloud, high-definition image, and thermal imaging data; Based on the preprocessed point cloud data, a three-dimensional mesh model of the pipeline surface is constructed using a triangular mesh reconstruction method; Identify defect areas and extract features through point cloud analysis, HD image edge detection, and thermal imaging temperature difference analysis; Based on the extracted features, determine the risk level; Mapping defect texture and temperature difference data to the surface of the 3D mesh model, and marking the defect position and category on the surface of the 3D mesh model; Establish pipeline axis reference and unify coordinates of point clouds, high-definition images and thermal images.

4. The method for performing UV-curing trenchless repair of pipelines in combination with a robot according to claim 3 is characterized in that: The preprocessing is specifically as follows: Point cloud data uses statistical filtering to remove isolated noise points: ； in, D(p i ) For point p i The density reference value, k is the standard deviation ratio; P filtered is the point cloud set retained after statistical filtering; mean(D) is the density reference value D(p i )’s mean; is the density reference value D(p i )’s standard deviation; Use the ICP algorithm to register the segmented point clouds and obtain complete three-dimensional information of the pipeline: ; Among them, T is the registration transformation matrix, R is the rotation matrix, is the offset vector, p i and q i Concentrate corresponding points on two point clouds; The high-definition image data is stitched together using a feature point matching algorithm to enhance contrast, filter out noise, fuse the high-definition image data into a three-dimensional mesh model, and generate a three-dimensional mesh model with surface texture through texture mapping corresponding to each point in the point cloud; Based on the thermal imaging data, the temperature distribution matrix is ​​extracted and the surface information of the point cloud is matched by interpolation.

5. The method for performing UV-curing trenchless repair of pipelines in combination with a robot according to claim 3 is characterized in that: The point cloud analysis, high-definition image edge detection, and thermal imaging temperature difference analysis are used to identify defect areas and extract features; based on the extracted features, the risk level is determined as follows: Using the geometric mutation of the point cloud surface and the texture edge detection of the image, the crack area is identified and the crack length L is obtained. crack and crack depth D crack : Corrosion is identified by matching the curvature depression of point cloud data with the high temperature area of ​​thermal imaging, and the corrosion depth is obtained. D corrosion and corrosion area A; The thermal imaging temperature difference is superimposed on the point cloud surface, the area and center position of the high temperature area are calculated, and the heat distribution is mapped to the surface of the three-dimensional grid model; According to the acquired abnormal features, the risk score P is calculated by the weighted model: ； Among them, w1, w2, w3, w4, w5 are defect risk contribution coefficients obtained through machine learning training; L crack is the crack length; D crack is the crack depth, and ΔT represents the temperature difference.

6. The method for performing UV-curing trenchless repair of pipelines in combination with a robot according to claim 5, characterized in that: The method uses the geometric mutation of the point cloud surface and the texture edge detection of the image to identify the crack area and obtain the crack length L crack and crack depth D crack , as follows: By combining the geometric mutation characteristics of the point cloud and the edge information of the cloud surface texture, the cracks are located and the principal curvature of the point cloud is calculated: K=k1+k2; Among them, k1 and k2 are the two principal curvatures of the local surface of the point cloud, and K represents the degree of mutation of the local surface; If the curvature K>K threshold , marked as possible crack areas, where K threshold is the preset threshold; Calculate the change gradient ΔN of the normal vector N in the point neighborhood: ； Among them, N a and N b is the normal vector of the adjacent point; if ΔN>ΔN threshold , marked as the geometric mutation point, where ΔN threshold is the preset threshold; The Canny operator is used to detect the crack edge and obtain the crack edge binary image E(x,y); the skeleton S(x,y) is obtained by thinning method; Overlay the edge of the HD image onto the point cloud and match the point cloud and image by coordinate registration: ； in, is the three-dimensional coordinate of the point cloud; (u, v) is the image texture coordinate; I(u, v) is the pixel value corresponding to the image texture coordinate (u, v); It is the inverse mapping function from image texture coordinates to point cloud three-dimensional coordinates; Accumulate the distance along the skeleton points of the crack edge to obtain the crack length L crack ; Calculate the difference between the bottom point of the crack and the fitted smooth surface to obtain the crack depth D crack .

7. The method for performing UV-curing trenchless repair of pipelines in combination with a robot according to claim 5, characterized in that: The method uses the curvature concave of the point cloud data to match the high temperature area of ​​the thermal imaging to identify corrosion and obtain the corrosion depth D corrosion And the corrosion area A, as follows: Corrosion is identified by matching the curvature concavity of the point cloud data with the high temperature area of ​​the thermal imaging, the smooth surface is fitted by the least squares method, and the local surface height deviation of the point cloud is calculated: ； Among them, z actual is the height of the actual point in the point cloud; z smooth is the height value of the fitted smooth surface; if D corrosion >D threshold , defined as the corrosion pit, D threshold is the preset threshold; The high temperature area is extracted from the thermal imaging temperature difference matrix, and the high temperature area is matched with the concave area of ​​the point cloud to identify the corrosion concave point. The local surface height deviation of the point cloud corresponding to the corrosion concave point is the corrosion depth D of the corrosion concave point. corrosion , and calculate the corrosion area A.

8. The method for performing UV-curing trenchless repair of pipelines in combination with a robot according to claim 1, characterized in that: The S3 is specifically as follows: Automatically divide the repair sections according to the defect type, distribution and location characteristics to determine whether overall repair or partial repair is required; According to the length of the crack L crack , depth D crack And the position calculation distribution density ρ crack ;like , divided into overall repair sections, where The preset threshold of distribution density; if ρ crack <ρ threshold , divided into local repair segments; According to the corrosion area A and corrosion depth D corrosion , calculate the corrosion ratio ,in is the pipe area; if , it is divided into the overall repair section, in which, Preset threshold for corrosion ratio, if , it is divided into local repair segments; For a local repair segment, calculate the starting position and length of the repair segment; For cracks, repair section length L segment for: ； in, The buffer repair length at both ends of the crack; For corrosion, repair section length L segment for: ; Formulate curing parameters according to defect characteristics, including light power, irradiation time and curing speed; Optical power P UV and crack depth D crack Proportional to: ； Among them, g1 is the influence coefficient of crack depth on optical power, and g2 is the basic power; like , need to increase power: ； Optical power and corrosion depth D corrosion Proportional to the corrosion area A: ； Among them, k3 and k4 influence coefficients; Irradiation time T UV By repairing the segment length L segment and curing speed v cure Decide: ； Adjust the curing speed according to the defect depth and distribution density: ； in, For the basic curing speed, is the influence coefficient of crack density on solidification speed; is the influence coefficient of corrosion depth on solidification speed; Based on the risk score P and defect location, high-risk segments are repaired first.

9. The method for performing UV-curing trenchless repair of pipelines in combination with a robot according to claim 1, characterized in that: The S4 is specifically as follows: Input the AI-generated repair task strategy into the digital twin environment; Simulate the repair process in a 3D mesh model and evaluate the repair effect; According to the simulation results, combined with the target repair effect and the constraints of consumables / energy consumption, the repair parameters are further optimized and the final optimized repair plan is output: the optimal repair segment division, light curing parameters and repair priority are obtained.

10. The method for performing UV-curing trenchless repair of pipelines in combination with a robot according to claim 1, characterized in that: The S5 is specifically as follows: According to the optimized repair plan, control the repair robot to complete the repair tasks of cracks, corrosion, or blockage, including UV curing repair and resin filling operations; The repair robot's built-in high-definition camera and depth sensor can be used to collect the execution status of the repair task in real time; According to the detection results during the repair process, the repair parameters are dynamically adjusted to ensure the best repair effect.

Images