Underground pipeline repairing method and device

Abstract

The invention discloses an underground pipeline repairing method and device, and relates to the technical field of underground pipeline repairing, and the method comprises the following steps: a robot collects and transmits an internal image of a pipeline and displacement information of the robot to an intelligent platform in real time; the intelligent platform processes the received pipeline internal image through an AI image algorithm and carries out comparative analysis on the pipeline internal image and a construction drawing so as to identify and mark abnormal point locations in the pipeline; the robot receives and analyzes the abnormal point location sent by the intelligent platform to determine the position and repair parameters of a target repair point; determining an optimal moving path of the robot based on the built pipeline internal environment model and the position of the target repairing point; according to the pipeline damage type, a corresponding hot melting repair parameter database is established through a hot melting control algorithm, and hot melting parameters are optimized according to repair effect information collected in real time. Through communication interconnection between the robot and the intelligent platform, image processing, path planning and hot melting control are completed, and rapid and low-cost pipeline damage point repairing is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of underground pipeline repair, and in particular to an underground pipeline repair method and device. Background Art

[0002] For underground pipelines that have been completed, when the pipelines are bent by heavy machinery or damaged and leaking, it is common to use ground excavation to deal with the pipeline damage problem. The specific operation is that the construction workers roughly determine the location of the pipeline based on the pipeline laying drawings or through some simple detection equipment, and then use large excavators to excavate the ground until the damaged pipeline is found. However, this traditional method has many disadvantages.

[0003] Firstly, when the direction of underground pipelines is unclear, the detection process may be inaccurate, resulting in deviation in the excavation position, which not only wastes time and manpower and material resources, but may also cause damage to other surrounding underground facilities.

[0004] Secondly, in the small area of ​​the construction site, large excavation equipment is difficult to use and its operation is restricted, which increases the difficulty of construction and safety risks.

[0005] Furthermore, when the geological conditions are complex, such as soft soil layers, rock layers or high groundwater levels, excavation can easily cause safety accidents such as ground collapse and landslides, and will also significantly increase the cost of repairs.

[0006] In addition, in old urban areas, the underground pipeline network is dense and complex, and the pipeline routes frequently conflict with numerous underground facilities. Excavation and repair can easily cause damage to other pipelines, affecting the normal living water, electricity, gas, etc. of surrounding residents. Moreover, the entire repair process is slow and cannot meet the needs of quickly restoring normal operation of the pipeline.

[0007] It can be seen that it is crucial to study a method for repairing underground pipelines that is safe, fast and low-cost. Summary of the invention

[0008] The technical problem to be solved by the present invention is how to use the trenchless repair technology to accurately find the damaged point and repair it when the direction of the underground pipeline is unclear.

[0009] In a first aspect of the present invention, in order to solve the above technical problems, a method for repairing an underground pipeline is provided, the method comprising the following steps: The robot collects and transmits the internal image of the pipeline and the displacement information of the robot to the intelligent platform in real time; The intelligent platform processes the received internal image of the pipeline through an AI image algorithm and compares and analyzes it with the construction drawings to identify and mark abnormal points in the pipeline; The robot receives and analyzes the abnormal point sent by the intelligent platform to determine the location and repair parameters of the target repair point; Determining an optimal path for the robot to move based on the constructed pipeline internal environment model and the position of the target repair point; According to the type of pipeline damage, a corresponding hot melt repair parameter database is established through the hot melt control algorithm, and the hot melt parameters are optimized based on the repair effect information collected in real time.

[0010] Furthermore, the method further includes a secondary detection, and the secondary detection specifically includes: After the hot-melt repair of the target repair point is completed, the robot performs a secondary scan of the entire pipeline system under the pipeline pressurization state, and transmits the data of the secondary scan to the intelligent platform; The intelligent platform receives and analyzes the data of the secondary scan to evaluate whether the target repair point after repair meets the requirements.

[0011] Furthermore, the displacement information of the robot includes position information and posture information of the robot in the pipeline.

[0012] Furthermore, the method for calculating the position information of the robot in the pipeline includes: Establish a three-dimensional coordinate system in the pipeline; The axial position of the robot along the pipeline and the position in the cross section of the pipeline are calculated respectively to obtain the current position information of the robot in the pipeline.

[0013] Further: The three-dimensional coordinate system includes: The central axis of the pipeline is the Z axis, any direction of the pipeline cross section is the X axis, and the direction perpendicular to the X axis and the Z axis is the Y axis; the origin of the coordinate system may be the initial position of the robot when it enters the pipeline; The step of calculating the axial position of the robot along the pipeline includes: The robot's built-in odometer records the number of revolutions of the robot's wheels, and the robot's axial position along the pipeline is Satisfies the expression:

[0014] in, is the circumference of the robot wheel, is the number of rotations of the robot's wheels; The step of calculating the position of the robot within the pipe cross section comprises: The robot is provided with at least three groups of laser distance measuring sensors at intervals of 120° from the X axis, and calculates the distances from the three groups of laser distance measuring sensors to the inner wall of the pipeline respectively. , , , then the position of the robot in the pipe cross section satisfies the expression:

[0015] Calculate the distance from the current robot to the centerline of the pipeline according to the Pythagorean theorem Satisfies the expression:

[0016] By comparison With pipe radius The size of the robot is used to correct the position coordinates of the robot in the pipe cross section.

[0017] Furthermore, the method for calculating the posture information of the robot in the pipeline includes: The pitch angle and roll angle of the robot are calculated by measuring the gravity acceleration component of the robot's built-in accelerometer; the pitch angle Satisfies the expression:

[0018] The roll angle Satisfies the expression:

[0019] in, , , are respectively the gravitational acceleration components measured by the accelerometer on the three axes of the robot coordinate system; The yaw angle of the robot is calculated by measuring the magnetic field component of the built-in magnetometer of the robot. Satisfies the expression:

[0020] in, , are respectively the components of the magnetic field output by the magnetometer on the X-axis and Y-axis of the robot coordinate system.

[0021] Furthermore, the method for identifying and marking abnormal points in a pipeline includes: Preprocessing the collected internal image of the pipeline by using a Gaussian filtering method; Using an image feature extraction algorithm to extract key features of the preprocessed image inside the pipeline; According to the analysis results of the key features, identifying the abnormal points in the pipeline; The identified abnormal points are integrated and marked.

[0022] Furthermore, the key features include edge features and texture features.

[0023] Furthermore, the edge feature extraction method includes using an edge detection algorithm to determine the edge position by calculating the gradient of pixel points of the internal image of the pipeline.

[0024] Furthermore, the texture feature extraction method includes using a gray level co-occurrence matrix method to calculate gray level variation statistics of the internal image of the pipeline in different directions and distances to characterize the texture feature.

[0025] Further, the abnormal site includes a damaged area and / or a deformed area and / or a corroded area, wherein: For the damaged area: Separating the damaged area from the background by using an image segmentation algorithm, and determining the area of ​​the damaged area by calculating the area and perimeter of the damaged area; Determining the shape of the damaged area using the edge detection algorithm; Determining the position of the damaged area in the pipeline by an image coordinate system; For the deformation area: Analyze the changes in the inner wall profile of the pipeline and calculate the degree of pipeline deformation; For the corrosion area: The scope and severity of the pipeline corrosion site are determined by analyzing the color changes and material texture changes of the pipeline internal image.

[0026] Furthermore, the method for constructing the pipeline internal environment model includes: The interior of the pipeline is preliminarily scanned by an infrared camera, and a pipeline internal environment model is established based on the pipeline design drawings; wherein the pipeline internal environment model can use a grid map representation method to divide the internal space of the pipeline into a number of grids of equal area; a single grid has any one of the attributes of a passable area, an obstacle area, and an unknown area.

[0027] Furthermore, the obstacle avoidance strategy of the robot moving in the pipeline includes: Obstacle avoidance algorithm based on artificial potential field method, the current position of the robot is recorded as , the position of the target patch point is , the obstacle position is , the gravitational coefficient is , the repulsion coefficient is , then the distance from the robot to the target repair point Satisfies the expression:

[0028] The distance from the robot to the obstacle Satisfies the expression:

[0029] gravitational Satisfies the expression:

[0030] Repulsion Satisfies the expression:

[0031] in, is the safe distance between the robot and the obstacle; The total force on the robot Satisfies the expression:

[0032] The robot adjusts its moving direction according to the combined force.

[0033] Furthermore, the optimal path selection strategy for the robot to move in the pipeline includes: Using the evaluation function in AI algorithm To select the node expansion order until the target patch point is found; wherein the evaluation function Satisfies the expression:

[0034] in, From the starting position to the node The actual moving distance; is from the node A heuristic estimate of the cost to the target patch point.

[0035] Furthermore, the evaluation function in the AI ​​algorithm is used To select the node expansion order, specifically: The robot starts to calculate the surrounding traversable nodes from the starting position. value; Select the The traversable node with the smallest value is used as the next expansion node, while taking into account the obstacle avoidance strategy; Through continuous calculation and selection, the optimal path from the starting position to the target repair point is found.

[0036] Furthermore, the step of taking into account the obstacle avoidance strategy includes: In the grid map, the robot moves from the current grid Move to adjacent grid The cost Satisfies the expression:

[0037] in, For arrive The cost of movement; Then the heuristic estimated cost Satisfies the expression:

[0038] in, , , For Grid The center coordinates of .

[0039] Furthermore, the pipeline damage type includes at least hole-shaped damage, crack-shaped damage and area-shaped damage, and the characteristic parameters corresponding to the pipeline damage type are extracted through the pipeline internal image and the sensor, wherein: The characteristic parameters of the hole-shaped damage include at least the hole diameter and depth; The characteristic parameters of the crack-shaped damage include at least the length, width and direction of the crack; The characteristic parameters of the area damage at least include the area and shape of the damaged region.

[0040] Furthermore, the hot melt repair parameter database includes at least the optimal hot melt temperature, heating time and pressure corresponding to each type of pipeline damage.

[0041] Furthermore, the step of optimizing the hot melt parameters according to the repair effect information collected in real time includes: Using the sensor carried by the hot melt terminal to monitor the temperature and pressure data in real time, and comparing the monitoring data with the standard data in the hot melt repair parameter database; According to the deviation between the monitoring data and the standard data, the robot decides whether to adjust the working state of the hot melt terminal.

[0042] A second aspect of the present invention provides an underground pipeline repair device, comprising a device body, wherein the device body comprises: A robot with moving wheels, wherein the robot is provided with a main board, a camera and a hot melt component; An intelligent platform is communicatively connected to the robot, and the intelligent platform can control the actions of the robot.

[0043] Furthermore, the mainboard is integrated with a control module, a power module and a communication module, wherein: The control module is used to process various data and instructions; The power module is used to provide power supply for the robot; The robot transmits data with the intelligent platform via the communication module.

[0044] Furthermore, the hot melt assembly includes a hot melt terminal and a mechanical arm connected to each other, and the hot melt terminal and the mechanical arm are both communicatively connected to the control module; wherein: The mechanical arm can be extended, rotated and moved to a specified position; The hot melt terminal heats the inner lining plastic material of the pipe to melt the inner lining plastic material and fill the target repair point; the hot melt terminal carries a sensor, which is used to monitor the temperature and pressure parameters in real time during the hot melt process.

[0045] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: The present invention completes image processing, path planning and hot-melt control through communication and interconnection between the robot and the intelligent platform, so as to accurately find the damaged point when the direction of the underground pipeline is unclear, avoid a series of problems caused by inaccurate detection, improve the accuracy of positioning, and do not need to rely on large-scale excavation equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0047] Figure 1 A flowchart of the disclosed embodiment of the present invention; Figure 2 A schematic diagram of a device disclosed in an embodiment of the present invention.

[0048] In the figure: 000, steel pipe; 001, plastic lining; 110. Camera; 120. Main board; 131. Robotic arm; 132. Hot melt terminal. DETAILED DESCRIPTION

[0049] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] The present invention aims to provide an underground pipeline repair method and device, the purpose of which is: on the one hand, to accurately find the damaged point when the direction of the underground pipeline is unclear, thereby avoiding a series of problems caused by inaccurate detection; on the other hand, there is no need to rely on large-scale excavation equipment to reduce construction difficulty and safety risks; furthermore, it can reduce damage to other pipelines.

[0051] First, the underground pipeline repair method disclosed in this embodiment is described.

[0052] See also Figure 1 , the method comprises the following steps: S1. The robot collects and transmits the internal image of the pipeline and the displacement information of the robot to the intelligent platform in real time.

[0053] The high-definition infrared camera on the micro robot in the underground pipeline is activated to collect comprehensive images of the interior of the pipeline. The camera 110 moves slowly with the robot in the pipeline to capture images of the inner wall of the pipeline at a certain speed and angle to ensure that the entire inner wall surface of the pipeline is covered to obtain detailed information on the interior of the pipeline, including whether there are abnormal conditions such as damage, deformation, corrosion, etc.

[0054] The image data collected by the camera 110 is transmitted to the intelligent platform in real time through the communication module on the robot's main board 120; at the same time, the displacement information of the robot during its movement in the pipeline is also transmitted so that the intelligent platform can accurately determine the specific location of the abnormal point in the pipeline.

[0055] The displacement information of the robot includes the position information and posture information of the robot in the pipeline.

[0056] The calculation of specific position information and posture information is shown below.

[0057] S11. Calculation of the robot's position information in the pipeline.

[0058] Establish a three-dimensional coordinate system inside the pipeline.

[0059] The three-dimensional coordinate system includes the central axis of the pipeline as the Z axis, any direction of the pipeline cross section as the X axis, and the direction perpendicular to the X axis and the Z axis as the Y axis; the origin of the coordinate system can be the initial position of the robot when it enters the pipeline.

[0060] The axial position of the robot along the pipeline and the position in the pipeline cross section are calculated respectively to obtain the current position information of the robot in the pipeline.

[0061] The steps for calculating the axial position of the robot along the pipeline include: The robot's built-in odometer records the number of revolutions of the robot's wheels, and the robot's axial position along the pipeline is Satisfies the expression:

[0062] in, is the circumference of the robot's wheels, is the number of rotations of the robot's wheels.

[0063] The steps for calculating the position of the robot within the pipe cross section include: The robot is equipped with at least three sets of laser distance measuring sensors at intervals of 120° from the X-axis. That is, there are three sensors on the robot, which are installed at 0°, 120°, and 240° from the X-axis. Calculate the distance from the laser distance measuring sensor to the inner wall of the pipe , , , then the position of the robot in the pipe cross section satisfies the expression:

[0064] Calculate the distance from the current robot to the centerline of the pipeline according to the Pythagorean theorem Satisfies the expression:

[0065] By comparison With pipe radius The size of the robot is used to correct the position coordinates of the robot in the pipe cross section to ensure the accuracy of the calculation.

[0066] S12. Calculation of the robot's posture information in the pipeline.

[0067] For pitch angle: use the robot's built-in accelerometer to measure.

[0068] The accelerometer can measure the components of gravity acceleration on the three axes of the robot coordinate system and calculate the pitch angle through the inverse tangent function. , then the pitch angle Satisfies the expression:

[0069] in, , , They are the gravitational acceleration components measured by the accelerometer on the three axes of the robot coordinate system.

[0070] For the yaw angle: the robot's built-in magnetometer is used to measure the angle of the earth's magnetic field in the robot's coordinate system to obtain the yaw angle. Satisfies the expression:

[0071] in, , They are respectively the components of the magnetic field output by the magnetometer on the X-axis and Y-axis of the robot coordinate system.

[0072] For the roll angle: the gravitational acceleration component is also measured by the accelerometer to calculate the roll angle Satisfies the expression:

[0073] S2. The intelligent platform processes the received internal images of the pipeline through AI image algorithms and compares and analyzes them with the construction drawings to identify and mark abnormal points in the pipeline.

[0074] After the intelligent platform receives the image data from the robot, its data processing module compares and analyzes the image data with the pre-stored normal pipeline inner wall image template.

[0075] The image data is processed through AI algorithms to identify abnormal points in the pipeline, such as the size, shape, location and other information of the damaged points, and the distribution of these abnormal points in the pipeline system is integrated and marked.

[0076] After confirming that all abnormal points in the pipeline section have been scanned and the data has been accurately transmitted, proceed to the next step.

[0077] Specifically, the method for identifying and marking abnormal points in the pipeline includes: S21. Preprocess the collected internal image of the pipeline using a Gaussian filtering method.

[0078] The normal pipeline inner wall image template is a standard image collected after the pipeline construction is completed or when it is known that there is no damage, and contains characteristic information such as the texture and color of the pipeline inner wall.

[0079] The collected images are preprocessed using the Gaussian filtering method, including grayscale processing and image filtering. Grayscale processing is used to convert color images into grayscale images to reduce the amount of data and highlight the texture features of the image; image filtering is used to remove noise interference to improve the clarity of the image.

[0080] S22. Use an image feature extraction algorithm to extract key features of the preprocessed internal image of the pipeline.

[0081] The image feature extraction algorithm is used to extract the key features in the pipeline inner wall image.

[0082] Among them, the key features include edge features and texture features.

[0083] The edge feature extraction method includes using an edge detection algorithm to determine the edge position by calculating the gradient of the pixel points of the image inside the pipeline.

[0084] The texture feature extraction method includes using the gray level co-occurrence matrix method to calculate the gray level change statistics of the internal image of the pipeline in different directions and distances to characterize the texture features.

[0085] The extracted key features are analyzed and compared with the features of the normal pipe inner wall.

[0086] For example, regarding edge features, the edge of a normal pipe inner wall should be continuous and smooth. If it is damaged, the edge will be interrupted or irregular.

[0087] Regarding texture features, the texture of the damaged area may change, such as texture disorder or loss.

[0088] S23. According to the analysis results of the key features, identify abnormal points in the pipeline.

[0089] The abnormal points in the pipeline are identified based on the key feature analysis results, wherein the abnormal points include damaged areas and / or deformed areas and / or corroded areas.

[0090] For damaged areas: Use image segmentation algorithms to separate damaged areas from the background, and determine the area of ​​the damaged area by calculating the area and perimeter of the damaged area. Use edge detection algorithms to determine the shape of the damaged area, such as circular, rectangular, irregular, etc. Use the image coordinate system to determine the location of the damaged area in the pipeline.

[0091] For the deformation area: analyze the changes in the inner wall profile of the pipeline and calculate the degree of pipeline deformation; such as the depth and length of the depression or bulge.

[0092] For corrosion areas: determine the scope and severity of pipeline corrosion by analyzing the color changes and material texture changes of the pipeline internal image, and calculate the corresponding parameters.

[0093] S24, integrating and marking the identified abnormal points.

[0094] The identified abnormal point information is integrated, including the type (damage, deformation, corrosion), location (axial position and cross-sectional position in the pipeline), size, shape and other information of the abnormal point. The abnormal points are marked on the 3D model or 2D drawing of the pipeline system to intuitively display the distribution of the abnormal points in the entire pipeline system.

[0095] Among them, the marking method can use different colors and shapes to distinguish different types of abnormal points. For example, red circles are used to mark damaged areas, green rectangles are used to mark deformed areas, and yellow triangles are used to mark corroded areas. At the same time, detailed information of the abnormal points, such as size, severity, etc., are marked next to the mark to facilitate subsequent viewing and analysis.

[0096] S3. The robot receives and analyzes the abnormal points sent by the intelligent platform to determine the location and repair parameters of the target repair point.

[0097] The intelligent platform sends the processed data to be repaired to the robot terminal through the wireless communication module. After receiving the data, the robot's main board 120 analyzes the instructions and determines the specific points to be repaired and the corresponding repair parameters, such as hot melt temperature and time.

[0098] S4. Determine the optimal path for the robot to move based on the constructed pipeline internal environment model and the location of the target repair point.

[0099] S41. Construction of pipeline internal environment model Before entering the pipeline, the robot conducts a preliminary scan of the inside of the pipeline using a high-definition infrared camera to obtain the approximate outline of the interior of the pipeline and information on possible obstacles.

[0100] At the same time, combined with the pre-stored normal pipeline inner wall image template (including information such as pipe diameter, direction, bend position, etc.), an environmental model of the interior of the pipeline is established.

[0101] Among them, the internal environment model of the pipeline can adopt the grid map representation method to divide the internal space of the pipeline into multiple grids of equal size, each grid has different attributes, such as passable, obstacle, unknown, etc.

[0102] According to the internal image of the pipeline collected by the camera 110 and the normal pipeline inner wall image template, the pipeline wall, known obstacles, etc. are marked as inaccessible areas, that is, obstacle areas, and the remaining space is marked as accessible areas. The areas that have not yet been detected are marked as unknown areas.

[0103] S42. Obstacle avoidance strategy for robots moving in pipelines When the robot moves in the pipeline, it uses high-definition infrared cameras and laser ranging sensors to monitor the surrounding environment in real time. When an obstacle is detected in front of it, it takes corresponding obstacle avoidance measures based on the distance information and obstacle location information fed back by the sensor.

[0104] Among them, the obstacle avoidance algorithm is based on the artificial potential field method. The principle is: the robot is regarded as moving in a virtual force field. The target point generates gravity on the robot, causing the robot to move toward the target point; while the obstacle generates repulsion on the robot, preventing the robot from approaching the obstacle. The robot adjusts its movement direction according to the combined force direction of gravity and repulsion to achieve obstacle avoidance.

[0105] The current position of the robot is , the location of the target patch point is , the obstacle position is , the gravitational coefficient is , the repulsion coefficient is , then the distance from the robot to the target repair point Satisfies the expression:

[0106] The distance from the robot to the obstacle Satisfies the expression:

[0107] gravitational Satisfies the expression:

[0108] Repulsion Satisfies the expression:

[0109] in, is the safe distance between the robot and the obstacle; The total force on the robot Satisfies the expression:

[0110] The robot adjusts its movement direction according to the combined force.

[0111] S43. Optimal path selection strategy for robot moving in pipeline.

[0112] Given a known pipeline environment model and the location of the target repair point, the robot needs to select an optimal path to reach the target point based on the AI ​​algorithm.

[0113] Using the evaluation function in AI algorithm to select the node expansion order until the target patch point is found.

[0114] Among them, the evaluation function Satisfies the expression:

[0115] in, From the starting position to the node The actual moving distance; It is a slave node Heuristic estimate of the cost to the target patch point.

[0116] The robot starts to calculate the surrounding traversable nodes from the starting position value.

[0117] choose The traversable node with the smallest value is selected as the next expansion node, while taking the obstacle avoidance strategy into consideration.

[0118] Through continuous calculation and selection, the optimal path from the starting position to the target patch point is found.

[0119] In the process of expanding nodes, the obstacle avoidance strategy is taken into consideration to avoid choosing paths that pass through obstacles. Specifically: In a grid map, the robot moves from the current grid Move to adjacent grid The cost Satisfies the expression:

[0120] in, For arrive The cost of movement; The heuristic cost estimate is Satisfies the expression:

[0121] in, , , For Grid The center coordinates of .

[0122] Through continuous calculation and selection, the optimal path from the starting position to the target repair point is found.

[0123] When the robot reaches the target position through the optimal path, it drives the retractable rotating mechanical arm 131 to move to the repair point.

[0124] After the robot arm 131 reaches the designated position, the robot arm 131 adjusts the position and angle of the hot melt terminal 132 to align it with the damaged point.

[0125] The hot melt terminal 132 heats the inner lining plastic 001 material to a corresponding precise temperature according to the size, shape, position and other parameters of the repair point, and starts the hot melt repair operation.

[0126] During the repair process, the sensor attached to the hot melt terminal 132 monitors parameters such as temperature and pressure in real time, and feeds the data back to the main board 120. The main board 120 adjusts the heating power and time according to the preset process requirements to ensure the quality of the repair.

[0127] The sensor data type is synchronized with the acquisition frequency, and the sensors attached to the hot melt terminal 132 include a temperature sensor and a pressure sensor.

[0128] Among them, the temperature sensor is used to monitor the temperature of the lining plastic 001 material during the hot melting process. The pressure sensor is used to monitor the pressure change during the hot melting process. In order to ensure that the acquisition frequency of the temperature sensor and the pressure sensor is the same, the data is collected every 0.1s so that the temperature and pressure information at the same time can be obtained during data fusion. Before data fusion, the data collected by the temperature sensor and pressure sensor are calibrated.

[0129] Among them, for the temperature sensor, by calibrating it in a known temperature environment, the relationship between the temperature measurement value and the actual temperature value is obtained, so as to correct the collected temperature data. The data is preprocessed, including removing abnormal values, filtering and other operations. Among them: Outliers may be erroneous data caused by sensor failure or interference, which can be judged and removed by setting reasonable thresholds.

[0130] The filtering operation adopts the sliding average filtering method to smooth the continuously collected data, reduce data fluctuations, and improve data stability and reliability.

[0131] S5. According to the type of pipeline damage, a corresponding hot melt repair parameter database is established through the hot melt control algorithm, and the hot melt parameters are optimized based on the repair effect information collected in real time.

[0132] S51. Data Fusion Different weights are assigned to temperature data and pressure data according to the degree of influence of temperature and pressure on the quality of hot melt repair. Assume that the weight of the influence of temperature on the repair quality is , the weight of pressure on repair quality is ,and + =1. Assume that the data sequence collected by the temperature sensor is , the data sequence collected by the pressure sensor is ( is the number of collected data), the data after weighted average fusion satisfies the expression:

[0133] Determine the optimal weight values ​​of temperature and pressure based on actual hot melt repair process requirements.

[0134] In the current hot melt repair process, the influence of temperature on the repair quality is more critical, which can be set , Through this data fusion method, the temperature and pressure information are comprehensively considered, providing a more accurate basis for the mainboard 120 to adjust the working state of the hot melt terminal 132.

[0135] S52, hot melt control The pipeline damage types include at least hole-shaped damage, crack-shaped damage and area-shaped damage. The characteristic parameters of the corresponding pipeline damage types are extracted through the internal image of the pipeline and the sensor, where: The characteristic parameters of hole-shaped damage include at least hole diameter and depth.

[0136] The characteristic parameters of crack-like damage include at least the length, width and direction of the crack.

[0137] The characteristic parameters of area damage include at least the area and shape of the damaged region.

[0138] Establish a hot melt repair parameter database. Based on actual engineering experience, establish a corresponding hot melt repair parameter database for different types of pipeline damage.

[0139] The hot melt repair parameter database includes at least the best hot melt temperature, heating time, pressure and other parameters corresponding to each type of pipeline damage. For example: For diameters smaller than The best hot melt temperature is , heating time is , the pressure is .

[0140] For length The cracks between them are broken, and the best hot melt temperature is obtained from the experiment. , heating time is , the pressure is .

[0141] In a further solution, the sensor carried by the hot melt terminal 132 is used to monitor the temperature and pressure data in real time, and the monitored data is compared with the standard data in the hot melt repair parameter database.

[0142] According to the deviation between the monitoring data and the standard data, the robot decides whether to adjust the working state of the hot melt terminal 132. For example: If the temperature is lower than the standard value, increase the heating power.

[0143] If the pressure is too high, reduce the heating speed appropriately.

[0144] At the same time, the repair effect is observed in real time through high-definition infrared cameras, and the hot melt parameters are further optimized according to the repair situation to ensure accurate repairs for various types of damage.

[0145] S6. Secondary detection After the hot-melt repair of the target repair point is completed, the entire pipeline is flushed to remove any debris and debris that may remain in the pipeline, and then a pressure test is carried out to simulate the pressure conditions during normal operation of the pipeline.

[0146] When the pipeline is pressurized, the robot starts the high-definition infrared camera again to scan the entire pipeline system for the second time, focusing on checking whether the repair points are firm and whether there are leakages and other problems, and transmits the data from the second scan to the intelligent platform.

[0147] The intelligent platform receives and analyzes the data from the second scan to evaluate whether the target repair point meets the requirements. If the repair point does not meet the requirements, it will be repaired again or other measures will be taken according to the situation until the pipeline system returns to normal operation.

[0148] See also Figure 2 The present invention also discloses an underground pipeline repair device, including a device body, wherein the device body includes: a robot with moving wheels and an intelligent platform connected to the robot in communication, wherein the intelligent platform can control the movement of the robot. The robot is provided with a main board 120, a camera 110 and a hot melt component. The main board 120 integrates a control module, a power module and a communication module.

[0149] Camera 110 uses a high-definition infrared camera to capture images of the internal environment of the pipeline. Its high-definition imaging function can clearly capture the details inside the pipeline, and its infrared function enables it to work effectively in dimly lit or complex pipeline environments, providing accurate visual information for subsequent analysis and repair operations.

[0150] The power module provides a stable power supply for the entire device to ensure the normal operation of each component. It can be in the form of a rechargeable battery to meet the needs of long-term work in underground pipelines.

[0151] The control module is an integrated chip, which is responsible for processing various data and instructions, such as processing image data collected by the camera 110, generating control instructions for the robot arm 131 and the hot melt terminal 132, etc.

[0152] The communication module realizes data transmission between the device and external equipment (such as an intelligent platform), including uploading image data collected by the camera 110 and the robot's own status information, and receiving control instructions and repair data from the intelligent platform.

[0153] The hot melt assembly includes a hot melt terminal 132 and a mechanical arm 131 connected to each other, and the hot melt terminal 132 and the mechanical arm 131 are both communicatively connected to the control module; wherein: The robot arm 131 has telescopic and rotating functions, and can flexibly move in the pipeline to reach the designated position for operation. The design of the robot arm 131 enables it to adapt to the requirements of different pipe diameters and pipe shapes, accurately locate the damaged point under intelligent control, and provide accurate position adjustment for hot melt repair operations.

[0154] The hot melt terminal 132 heats the inner lining plastic 001 material of the galvanized steel pipe 000 to melt the inner lining plastic 001 material and fill it to the target repair point.

[0155] The hot melt terminal 132 carries a sensor, which is used to monitor the temperature and pressure parameters in the hot melt process in real time to ensure the safety and quality of the repair process. The sensor feeds back the data to the main board 120, and the main board 120 adjusts the working state of the hot melt terminal 132 according to the data to ensure that the repair operation meets the process requirements.

[0156] By heating the inner lining plastic 001 material to the corresponding temperature, it melts and fills the damaged part to achieve the repair of the pipeline. The heating temperature and time can be precisely controlled according to the size, shape and position of the repair point to achieve the best repair effect.

[0157] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for repairing an underground pipeline, characterized in that: The method comprises the following steps: The robot collects and transmits the internal image of the pipeline and the displacement information of the robot to the intelligent platform in real time; The intelligent platform processes the received internal image of the pipeline through an AI image algorithm and compares and analyzes it with the construction drawings to identify and mark abnormal points in the pipeline; The robot receives and analyzes the abnormal point sent by the intelligent platform to determine the location and repair parameters of the target repair point; Determining an optimal path for the robot to move based on the constructed pipeline internal environment model and the position of the target repair point; According to the type of pipeline damage, a corresponding hot melt repair parameter database is established through the hot melt control algorithm, and the hot melt parameters are optimized based on the repair effect information collected in real time.

2. The underground pipeline repair method according to claim 1, characterized in that: The method further includes a secondary detection, which specifically includes: After the hot-melt repair of the target repair point is completed, the robot performs a secondary scan of the entire pipeline system under the pipeline pressurization state, and transmits the data of the secondary scan to the intelligent platform; The intelligent platform receives and analyzes the data of the secondary scan to evaluate whether the target repair point after repair meets the requirements.

3. The underground pipeline repair method according to claim 1, characterized in that: The displacement information of the robot includes the position information and posture information of the robot in the pipeline.

4. The underground pipeline repair method according to claim 3, characterized in that: The method for calculating the position information of the robot in the pipeline includes: Establish a three-dimensional coordinate system in the pipeline; The axial position of the robot along the pipeline and the position in the cross section of the pipeline are calculated respectively to obtain the current position information of the robot in the pipeline.

5. The underground pipeline repair method according to claim 4, characterized in that: The three-dimensional coordinate system includes: The central axis of the pipeline is the Z axis, any direction of the pipeline cross section is the X axis, and the direction perpendicular to the X axis and the Z axis is the Y axis; the origin of the coordinate system may be the initial position of the robot when it enters the pipeline; The step of calculating the axial position of the robot along the pipeline includes: The robot's built-in odometer records the number of revolutions of the robot's wheels, and the robot's axial position along the pipeline is Satisfies the expression: in, is the circumference of the robot wheel, is the number of rotations of the robot's wheels; The step of calculating the position of the robot within the pipe cross section comprises: The robot is provided with at least three groups of laser distance measuring sensors at intervals of 120° from the X axis, and calculates the distances from the three groups of laser distance measuring sensors to the inner wall of the pipeline respectively. , , , then the position of the robot in the pipe cross section satisfies the expression: Calculate the distance from the current robot to the centerline of the pipeline according to the Pythagorean theorem Satisfies the expression: By comparison With pipe radius The size of the robot is used to correct the position coordinates of the robot in the pipe cross section.

6. The underground pipeline repair method according to claim 3, characterized in that: The method for calculating the posture information of the robot in the pipeline includes: The pitch angle and roll angle of the robot are calculated by measuring the gravity acceleration component of the robot's built-in accelerometer; the pitch angle Satisfies the expression: The roll angle Satisfies the expression: in, , , are respectively the gravitational acceleration components measured by the accelerometer on the three axes of the robot coordinate system; The yaw angle of the robot is calculated by measuring the magnetic field component of the built-in magnetometer of the robot. Satisfies the expression: in, , are respectively the components of the magnetic field output by the magnetometer on the X-axis and Y-axis of the robot coordinate system.

7. The underground pipeline repair method according to claim 1, characterized in that: The method for identifying and marking abnormal points in a pipeline comprises: Preprocessing the collected internal image of the pipeline by using a Gaussian filtering method; Using an image feature extraction algorithm to extract key features of the preprocessed image inside the pipeline; According to the analysis results of the key features, identifying the abnormal points in the pipeline; The identified abnormal points are integrated and marked.

8. The underground pipeline repair method according to claim 7, characterized in that: The key features include edge features and texture features.

9. The underground pipeline repair method according to claim 8, characterized in that: The edge feature extraction method includes using an edge detection algorithm to determine the edge position by calculating the gradient of the pixel points of the internal image of the pipeline.

10. The underground pipeline repair method according to claim 8, characterized in that: The texture feature extraction method includes using a gray level co-occurrence matrix method to calculate gray level change statistics of the pipeline internal image in different directions and distances to characterize the texture feature.

11. The underground pipeline repair method according to claim 9, characterized in that: The abnormal sites include damaged areas and / or deformed areas and / or corroded areas, wherein: For the damaged area: Separating the damaged area from the background by using an image segmentation algorithm, and determining the area of ​​the damaged area by calculating the area and perimeter of the damaged area; Determining the shape of the damaged area using the edge detection algorithm; Determining the position of the damaged area in the pipeline by an image coordinate system; For the deformation area: Analyze the changes in the inner wall profile of the pipeline and calculate the degree of pipeline deformation; For the corrosion area: The scope and severity of the pipeline corrosion site are determined by analyzing the color changes and material texture changes of the pipeline internal image.

12. The underground pipeline repair method according to claim 1, characterized in that: The method for constructing the pipeline internal environment model comprises: The interior of the pipeline is preliminarily scanned by an infrared camera, and a pipeline internal environment model is established based on the pipeline design drawings; wherein the pipeline internal environment model can use a grid map representation method to divide the internal space of the pipeline into a number of grids of equal area; a single grid has any one of the attributes of a passable area, an obstacle area, and an unknown area.

13. The underground pipeline repair method according to claim 12, characterized in that: The obstacle avoidance strategy of the robot moving in the pipeline includes: Obstacle avoidance algorithm based on artificial potential field method, the current position of the robot is recorded as , the position of the target patch point is , the obstacle position is , the gravitational coefficient is , the repulsion coefficient is , then the distance from the robot to the target repair point Satisfies the expression: The distance from the robot to the obstacle Satisfies the expression: gravitational Satisfies the expression: Repulsion Satisfies the expression: in, is the safe distance between the robot and the obstacle; The total force on the robot Satisfies the expression: The robot adjusts its moving direction according to the combined force.

14. The underground pipeline repair method according to claim 13, characterized in that: The optimal path selection strategy for the robot to move in the pipeline includes: Using the evaluation function in AI algorithm To select the node expansion order until the target patch point is found; wherein the evaluation function Satisfies the expression: in, From the starting position to the node The actual moving distance; is from the node A heuristic estimate of the cost to the target patch point.

15. The underground pipeline repair method according to claim 14, characterized in that: The evaluation function in the AI ​​algorithm is used To select the node expansion order, specifically: The robot starts to calculate the surrounding traversable nodes from the starting position. value; Select the The traversable node with the smallest value is used as the next expansion node, while taking into account the obstacle avoidance strategy; Through continuous calculation and selection, the optimal path from the starting position to the target repair point is found.

16. The underground pipeline repair method according to claim 15, characterized in that: The steps of taking into account the obstacle avoidance strategy include: In the grid map, the robot moves from the current grid Move to adjacent grid The cost Satisfies the expression: in, For arrive The cost of movement; Then the heuristic estimated cost Satisfies the expression: in, , , For Grid The center coordinates of .

17. The underground pipeline repair method according to claim 1, characterized in that: The pipeline damage types include at least hole-shaped damage, crack-shaped damage and area-shaped damage. The characteristic parameters corresponding to the pipeline damage types are extracted through the pipeline internal image and the sensor, wherein: The characteristic parameters of the hole-shaped damage include at least the hole diameter and depth; The characteristic parameters of the crack-shaped damage include at least the length, width and direction of the crack; The characteristic parameters of the area damage at least include the area and shape of the damaged region.

18. The underground pipeline repair method according to claim 1, characterized in that: The hot melt repair parameter database at least includes the optimal hot melt temperature, heating time and pressure corresponding to each type of pipeline damage.

19. The underground pipeline repair method according to claim 1, characterized in that: The step of optimizing the hot melt parameters based on the repair effect information collected in real time includes: Using the sensor carried by the hot melt terminal to monitor the temperature and pressure data in real time, and comparing the monitoring data with the standard data in the hot melt repair parameter database; According to the deviation between the monitoring data and the standard data, the robot decides whether to adjust the working state of the hot melt terminal.

20. An underground pipeline repair device, characterized in that: The device comprises a device body, wherein the device body comprises: A robot with moving wheels, wherein the robot is provided with a main board, a camera and a hot melt component; An intelligent platform is communicatively connected to the robot, and the intelligent platform can control the actions of the robot.

21. The underground pipeline repair device according to claim 20, characterized in that: The mainboard integrates a control module, a power module and a communication module, wherein: The control module is used to process various data and instructions; The power module is used to provide power supply for the robot; The robot transmits data with the intelligent platform via the communication module.

22. The underground pipeline repair device according to claim 21, characterized in that: The hot melt assembly includes a hot melt terminal and a mechanical arm connected to each other, and the hot melt terminal and the mechanical arm are both in communication connection with the control module; wherein: The mechanical arm can be extended, rotated and moved to a specified position; The hot melt terminal heats the inner lining plastic material of the pipe to melt the inner lining plastic material and fill the target repair point; the hot melt terminal carries a sensor, which is used to monitor the temperature and pressure parameters in real time during the hot melt process.

Images