Robot automation platform for on-site intelligent maintenance of water cooling wall

Abstract

The invention discloses a robot automation platform for on-site intelligent maintenance of a water cooling wall, and a robot body function module comprises a laser vision automatic deviation correction module used for realizing path automatic deviation correction of a robot body by adopting a laser vision technology; the automatic lane changing module is used for realizing an automatic lane changing function so as to realize automatic switching of working paths of the robot; the ultrasonic obstacle avoidance module is used for realizing an obstacle avoidance function of the robot body; the pose recognition module is used for calculating and displaying the real-time pose of the robot body through a pose recognition function; the 360-degree image module is used for displaying surrounding environment information of the robot body in real time; the detection module comprises an electromagnetic ultrasonic thickness measurement module which is used for automatically measuring the thickness of the water wall tube by adopting an electromagnetic ultrasonic technology; and the defect detection module is used for automatically detecting surface defects and defect types of the water-cooled wall, realizes a remote automatic detection technology of the water-cooled wall, and has the advantages of high working efficiency, high detection precision, capability of reducing the operation cost, shortening the construction period and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of water wall detection, and more specifically, to a robotic automation platform for intelligent on-site maintenance of water walls. Background Art

[0002] Common types of water walls include membrane type and bare tube. With the continuous development of large boilers, membrane water walls are more commonly used. The number of safety accidents caused by water wall problems accounts for 40% of all accidents in thermal power plants. This number and proportion are still on the rise, bringing difficulties to people's electricity use and causing huge losses. Therefore, regular inspection and maintenance of water walls are of great significance for the safe operation of thermal power plants and the maintenance of national and corporate economic development.

[0003] Currently, the detection of water walls mainly relies on manual inspection. First, the boiler to be inspected is shut down. Then, the inspection workers approach the water wall tubes by building scaffolding or "lifting ladders" made of steel pipes, visually check for areas with potential safety hazards such as severe wear, and use the carried detection devices for inspection and elimination. This is time-consuming and laborious, with low detection efficiency and unreliable detection results.

[0004] Therefore, how to provide a robotic automation platform for intelligent on-site maintenance of water walls is an urgent problem to be solved by those skilled in the art. Summary of the Invention [[ID=1⑧]]

[0005] In view of this, the present invention provides a robotic automation platform for intelligent on-site maintenance of water walls, which develops a remote automation detection technology for water walls based on methods such as structured light detection, machine image recognition, and remote non-destructive detection. It has many advantages such as high working efficiency, high detection accuracy, reduced operation costs, and shortened construction period, and is an important means to solve the current problem of boiler tube explosion.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A robotic automation platform for intelligent on-site maintenance of water walls, comprising: a control system, a robotic body function module, and a detection module;

[0008] The control system includes:

[0009] A control main board, which is used to connect to an operation handle, a host computer, the robotic body function module, and the detection module, receive and upload detection data, and receive and issue operation instructions;

[0010] An operation handle, which controls the movement of the robotic body through the control main board and controls the opening and closing of the detection module and the robotic body function module;

[0011] The host computer is used to receive the detection data sent by the control main board and process it;

[0012] The robot body function modules include:

[0013] The laser vision automatic deviation correction module is used to realize the automatic path deviation correction of the robot body by using laser vision technology;

[0014] The automatic lane change module is used for the automatic lane change function to realize the automatic conversion of the robot operation path;

[0015] The ultrasonic obstacle avoidance module is used to realize the obstacle avoidance function of the robot body;

[0016] The pose recognition module is used for the pose recognition function to calculate and display the real-time pose of the robot body;

[0017] The 360-degree image module is used to display the surrounding environment information of the robot body in real time;

[0018] The detection module includes:

[0019] The electromagnetic ultrasonic thickness measurement module is used to automatically measure the thickness of the water-cooled wall tube by using electromagnetic ultrasonic technology;

[0020] The defect detection module is used to automatically detect the surface defects and defect types of the water-cooled wall.

[0021] Preferably, the laser vision automatic deviation correction module includes:

[0022] The laser sensor is used to emit line laser to the membrane water-cooled wall;

[0023] The camera is used to collect the laser image formed by the line laser irradiating on the membrane water-cooled wall tube;

[0024] The embedded controller module is used to process the laser image, generate the path deviation amount, send the path deviation amount to the control main board, and control the differential motion of the left and right walking motors of the robot through the control panel to realize the autonomous deviation correction walking along the axial direction of the water-cooled wall tube.

[0025] Preferably, the specific implementation process of the embedded controller module is:

[0026] The collected laser images are sequentially subjected to graying, filtering and denoising, and edge enhancement processing;

[0027] By combining Canny edge detection and Hough transform, the straight line segment corresponding to the laser line is extracted;

[0028] Calculate the intersection points between the straight line segments or the turning points of the laser line at the water-cooled wall tube, and define them as vertex feature points;

[0029] Screen the vertex feature points, and generate a path deviation amount based on the coordinate information of the optimized vertex feature points in the image.

[0030] Preferably, the automatic lane change module includes:

[0031] A counter for counting the water-cooled wall tubes swept.

[0032] Preferably, the specific implementation process of the automatic lane change module is as follows:

[0033] Input the tube number of the target water-cooled wall tube;

[0034] The control main board calculates the difference between the tube number of the water-cooled wall tube currently scanned by the line laser and the tube number of the target water-cooled wall tube, and controls the robot body to deflect towards the target water-cooled wall tube;

[0035] Each time the counter sweeps over a water-cooled wall tube, the value is incremented by 1. Determine whether the difference between the tube numbers is greater than 1. If it is not greater than 1 and when the value of the counter reaches 1, it means that the line laser has covered the next water-cooled wall tube. The control main board automatically activates the path automatic correction, and then the robot body walks along the axis of the target water-cooled wall tube, and the counter value is cleared;

[0036] When the difference between the tube numbers is greater than 1, and when the value of the counter reaches 1, that is, when the line laser scans to the next water-cooled wall tube, the control main board controls the robot body to stop deflecting and walk straight towards the target water-cooled wall tube; when the value of the counter reaches the difference between the tube numbers, that is, when the line laser scans to the target water-cooled wall tube, the control main board automatically activates the path automatic correction, and then the robot body walks along the axis of the target water-cooled wall tube, and the counter value is cleared.

[0037] Preferably, the defect detection module includes a lighting source, a CCD camera, and an image acquisition card;

[0038] The lighting source is used to provide a lighting source;

[0039] The CCD camera is used to collect the surface image of the water-cooled wall tube under the lighting source;

[0040] The image acquisition card is used to transmit the surface image of the water-cooled wall tube to the control main board and upload it to the host computer through the control main board;

[0041] The host computer accurately identifies the defect type and records the defect location through the fuzzy neural network algorithm.

[0042] Preferably, the 360 imaging module includes four cameras, which are respectively installed in the front, rear, left, and right directions of the robot body.

[0043] Preferably, the control main board is used to synthesize the four images collected by the four cameras to obtain a 360 image.

[0044] As can be seen from the above technical solutions, compared with the prior art, the present invention discloses and provides a robot automation platform for on-site intelligent maintenance of water walls, which has the following advantages:

[0045] 1) Develop automatic deviation correction applicable to the water wall surface to achieve rapid, precise movement and automatic lane change of the intelligent maintenance robot; install a non-destructive thickness measurement system, a vision detection system, etc. to achieve automatic identification of surface defects of water wall tubes and rapid measurement of the wall thickness of water walls;

[0046] 2) The automatic deviation correction technology applicable to the water wall surface is combined with laser vision technology. Through laser scanning recognition, the straight path of the water wall tube row is identified; through real-time calculation by the tracking control system, the differential motion adjustment of the robot is guided; based on this, it can ensure that the machine body travels along the axis of the water wall tube, thereby ensuring the accuracy of the robot's motion trajectory and providing a basis for the function of automatic thickness measurement and defect detection;

[0047] 3) The automatic detection technology for water wall surface defects uses a high-definition detection camera module to capture real-time images of the water wall tube row, providing raw data for rapid defect identification. Through the vision recognition module and the high-definition detection camera module, the deformation of the water wall is quickly detected remotely, automatically and efficiently, and local surface cracks, bruises, deformations, welding defects, etc. of the water wall are automatically identified and located. Moreover, this technology can detect water wall areas that workers cannot reach, greatly expanding the scope of water wall surface defect detection, saving manpower, improving detection efficiency and reducing the safety hazards of water walls.

[0048] 4) The automatic lane change technology of the water wall climbing robot enables the water wall intelligent robot to automatically transfer to the starting point of the next track after completing the detection of the current track, which will greatly shorten the time required for manually adjusting the robot track and ensure full coverage of the area to be detected, thereby improving the maintenance efficiency, shortening the maintenance cycle, and having extremely high economic value and promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0050] Figure 1 It is a schematic structural diagram of a robot automation platform for on-site intelligent maintenance of water walls provided by the present invention.

[0051] Figure 2Schematic diagram of the robot body provided by the present invention.

[0052] Figure 3 Schematic diagram of the line laser and vertex feature points provided by the present invention.

[0053] Figure 4 Flowchart of automatic lane change provided by the present invention.

[0054] Figure 5 Data flow diagram of the 360 imaging module provided by the present invention.

[0055] Figure 6 Schematic diagram of the control main board provided by the present invention.

[0056] Figure 7 Schematic diagram of the functions of the upper computer provided by the present invention.

[0057] Among them, 1. Laser vision automatic deviation correction module, 2. 360 imaging module, 3. Electromagnetic ultrasonic thickness measurement module, 4. Defect detection module, 5. Linear slide table. Specific implementation manners

[0058] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0059] An embodiment of the present invention discloses an intelligent on-site maintenance robot automation platform for water-cooled walls, as Figure 1 shown, including:

[0060] A control system, a robot body function module, and a detection module;

[0061] The control system includes:

[0062] A control main board, used to connect to an operation handle, an upper computer, a robot body function module, and a detection module, receive and upload detection data, and receive and issue operation instructions;

[0063] An operation handle, used to control the movement of the robot body through the control main board, and control the opening and closing of the detection module and the robot body function module;

[0064] An upper computer, used to receive the detection data sent by the control main board and process it; the detection data is the data collected by the robot body function module and the detection module.

[0065] The robot body function module includes:

[0066] The laser vision automatic deviation correction module 1 is used to achieve automatic path deviation correction of the robot body by using laser vision technology;

[0067] The automatic lane change module is used for the automatic lane change function to achieve automatic conversion of the robot operation path;

[0068] The ultrasonic obstacle avoidance module is used to achieve the obstacle avoidance function of the robot body;

[0069] The pose recognition module is used for the pose recognition function to calculate and display the real-time pose of the robot body;

[0070] The 360 imaging module 2 is used to display the surrounding environment information of the robot body in real time;

[0071] The detection module includes:

[0072] The electromagnetic ultrasonic thickness measurement module 3 is used to automatically measure the thickness of the water-cooled wall tube by using electromagnetic ultrasonic technology and upload it to the upper computer;

[0073] The defect detection module 4 is used to automatically detect the surface defects and defect types of the water-cooled wall and upload them to the upper computer.

[0074] The electromagnetic ultrasonic thickness measurement module 3, the defect detection module 4 and the robot body function module are all installed on the robot body, and the water-cooled wall is detected while the robot moves on the water-cooled wall surface. The robot body communicates remotely with the control system, as Figure 2 shown.

[0075] The adsorption method of the present invention adopts permanent magnet adsorption. The magnetic adsorption wall-climbing robot mechanism is relatively simple, has high reliability, and will not reduce the adsorption force due to the unevenness of the wall surface, and its adsorption force is also much greater than that of vacuum negative pressure adsorption. The characteristics of permanent magnet adsorption and electromagnetic adsorption are obvious. Electromagnetic adsorption is easy to control, and permanent magnet adsorption has strong load capacity. However, during the working process, once a power failure occurs, the electromagnetic adsorption wall-climbing robot will fall from the wall surface and cause danger, while the permanent magnet adsorption wall-climbing robot will not be affected.

[0076] The driving method adopts motor drive. The motor drive control is simple and flexible, has strong load capacity, and can quickly achieve forward and reverse rotation. When two motors rotate forward or reverse at the same speed, forward or backward movement can be completed. When the motors on both sides of the robot rotate at different speeds, a turning action is completed, as Figure 2 shown.

[0077] Specifically, the laser vision automatic deviation correction module 1 includes:

[0078] The laser sensor is used to emit line laser to the membrane water-cooled wall;

[0079] A camera, which is used to collect the laser image formed by the line laser irradiating on the membrane water-cooled wall tube;

[0080] An embedded controller module, which is used to process the laser image, generate a path deviation amount, send the path deviation amount to the control main board, and control the differential motion of the left and right walking motors of the robot through the control panel to achieve autonomous deviation correction walking along the axial direction of the water-cooled wall tube.

[0081] The specific implementation process of the embedded controller module is as follows:

[0082] The collected laser image is sequentially subjected to grayscale conversion, filtering and denoising, and edge enhancement processing;

[0083] By combining the Canny edge detection and the Hough transform, the straight line segment corresponding to the laser line is extracted;

[0084] Calculate the intersection points between the straight line segments or the turning points of the laser line at the water-cooled wall tube, and define them as vertex feature points, as Figure 3 shown;

[0085] Screen the vertex feature points, and generate a path deviation amount based on the coordinate information of the optimized vertex feature points in the image to ensure that the wall-climbing robot body walks along the axis of the water-cooled wall tube.

[0086] The laser vision automatic deviation correction module 1 combines laser vision technology, and through laser scanning recognition, it recognizes the straight path of the water-cooled wall tube row, guides the differential motion adjustment of the robot. The present invention can recognize the special structure feature points of the water-cooled wall and guide the differential walking of the robot walking motor to ensure that the robot can walk along the axial direction of the water-cooled wall tube with autonomous deviation correction.

[0087] In this embodiment, the specific process of the automatic lane change of the robot body is as follows: when the robot body completes the defect detection and thickness measurement operations along a certain water-cooled wall tube, the wall-climbing robot is deflected to the left or right so that the line laser covers the water-cooled wall tube corresponding to the next operation path, and then the path deviation correction function is turned on, and the wall-climbing robot can be slowly transferred to the next detection path and run along the axis of the water-cooled wall.

[0088] The realization of the automatic lane change function of the robot body requires the help of a counter, which is installed directly in front of the laser sensor, as Figure 4 shown, and the specific control logic is as follows:

[0089] Step 1: Number the water-cooled wall tubes in the detected area;

[0090] Step 2: Before the automatic lane change is required, the operator inputs the tube number of the water-cooled wall tube that the line laser needs to scan in the next detection trajectory;

[0091] Step 3: Click the "Lane Change" button on the operating handle to control the main board to calculate the difference between the tube number of the water-cooled wall tube scanned by the current line laser and the target water-cooled wall tube number, and then control the robot body to deflect towards the target water-cooled wall tube;

[0092] Step 4: Each time the counter scans a water-cooled wall tube, the value is incremented by 1. Determine whether the difference between the tube numbers is greater than 1. If it is not greater than 1 and when the value of the counter reaches 1, it means that the line laser emitted by the laser sensor has covered the next water-cooled wall tube. The main board automatically activates the trajectory correction, and then the robot walks along the axis of the target water-cooled wall tube, and the counter value is cleared;

[0093] Step 5: When the difference between the tube numbers is greater than 1, when the value of the counter reaches 1, that is, when the line laser scans the next water-cooled wall tube, the main board controls the trolley to stop deflecting and walk straight towards the target water-cooled wall tube;

[0094] Step 6: When the value of the counter reaches the difference between the tube numbers, that is, when the line laser scans the target water-cooled wall tube, the main board automatically activates the trajectory correction function and then the counter value is cleared, thus realizing that the wall-climbing robot automatically moves to the next detection trajectory.

[0095] In this embodiment, the defect detection module 4 can quickly carry out remote automatic and efficient detection of the deformation of the water-cooled wall through visual recognition and high-definition detection cameras, and automatically identify and locate local surface cracks, dents, deformations, welding defects, etc. on the water-cooled wall. Establish a defect model database based on the intelligent defect recognition system and the water-cooled wall, use computer learning to expand, generate simulation samples, and enrich the defect database; and through neural network training, improve the defect recognition ability and reduce the misjudgment rate.

[0096] The defect detection module includes a lighting source, a CCD camera, and an image acquisition card;

[0097] The lighting source is used to provide a lighting source;

[0098] The CCD camera is used to collect the surface image of the water-cooled wall tube under the lighting source;

[0099] The image acquisition card is used to transmit the surface image of the water-cooled wall tube to the main board and upload it to the host computer through the main board;

[0100] The host computer processes the brightness, color, etc. of the surface image of the water-cooled wall tube and converts it into a digital signal; accurately reflects the actual situation of the water-cooled wall surface; by developing a fuzzy neural network algorithm for image recognition and using the typical defect images in the database to repeatedly train this algorithm, so that this algorithm can accurately identify the defect type and record the defect location.

[0101] (1) Selection of light source: The illumination light source is an important part of the machine vision system. The light source should have high brightness, adjustable brightness, good uniformity, and high stability to suppress the large impact of various external lights on the image quality, which may cause faults or misjudgment behaviors in the machine vision system. In this system, an LED lamp is used as the CCD light source.

[0102] (2) Selection of CCD camera: CCD (charge coupled device) is a semiconductor device that can convert optical images into digital signals. The image of the object to be photographed is focused on the CCD chip through a lens. The CCD accumulates corresponding charges according to the intensity of light. Under the control of the video timing, the charges accumulated in each pixel are shifted out point by point. After filtering and amplification, a video signal is output. It has the advantages of high sensitivity, resistance to strong light, small distortion, small size, long life, and anti-vibration.

[0103] The present invention can detect obvious defects such as local cracks, bruised deformation welding, and high-temperature corrosion; the defect detection algorithm can achieve rapid identification.

[0104] In this embodiment, the electromagnetic ultrasonic thickness measurement module 3 is installed in the front of the robot body through a linear slide 5, and is used to automatically measure the thickness of the water-cooled wall tube by using electromagnetic ultrasonic technology, which uses the electromagnetic coupling method to excite and receive ultrasonic waves.

[0105] In this embodiment, as Figure 5 shown, 1080P, large-angle high-definition cameras are respectively configured at the front, rear, left, and right positions of the vehicle body to build a 360-degree imaging device. Its function is to monitor the surrounding environment during the detection process of the intelligent robot. The four high-definition cameras in the front, rear, left, and right capture the environmental information images in the four directions of the front, rear, left, and right of the robot body and upload them to the control main board; the control main board synthesizes the four images and uploads them to the upper computer to realize remote imaging of the working surface and assist remote control or path planning.

[0106] Among them, the ultrasonic obstacle avoidance module is used to realize the obstacle avoidance function of the robot body. If the distance is too close, the upper computer issues an alarm or an automatic stop command to avoid collisions; the pose recognition module is used for pose recognition to calculate and display the real-time pose of the robot body. The above two modules are implemented by existing technologies and will not be elaborated here.

[0107] In this embodiment, as Figure 6As shown in the figure, the control mainboard uses an embedded industrial control module as the main control unit, and adopts a hardware structure based on the Ethernet fieldbus system EtherCAT or other similar industrial control platforms. Its function is to realize the motion control of the intelligent robot body and the control of other functional modules, such as the opening and closing of functional modules, the reception of thickness measurement data and the upload to the upper computer, and the execution of the robot walking deviation correction algorithm, etc.

[0108] The operation handle is connected to the control mainboard, and the motion of the intelligent robot can be controlled through the control mainboard, including setting the crawling speed, controlling the robot to move forward, backward and turn. In addition, it can also control the start and stop of the thickness measurement and defect detection modules, the start and stop of the robot body walking deviation correction and automatic lane change functions, etc.; all detection data (thickness, defects, images, path deviation, etc.) are uniformly uploaded to the upper computer by the control mainboard.

[0109] As Figure 7 shown in the figure, the upper computer is connected to the control mainboard and is equipped with intelligent robot data analysis software. Its main functions are:

[0110] Equipped with a data analysis system, which receives the water-cooled wall tube thickness data and displays, stores and analyzes the data, displays the analysis results in the form of line charts, pie charts, etc., and can export the data in the form of Excel;

[0111] Receives the surface image of the water-cooled wall tube, processes the image to identify whether there are defects on the surface, the type and location of the defects, and outputs and displays and saves the defect data;

[0112] Displays the 360° real-time image around the robot;

[0113] Integrates the information of each sensor of the robot, such as the obstacle distance information measured by the ultrasonic obstacle avoidance sensor;

[0114] Through status warning and display, it can record and judge the operation situation and historical information;

[0115] Realtime displays the system and robot real-time device status information such as the collision, speed, positioning attitude, equipment, task status, communication, synchronization, etc. of the robot body.

[0116] The main functions of the data analysis system include data analysis and comparison of water-cooled wall detection, graphical display, historical task query, database management, etc. The specific functions are as follows:

[0117] 1) Multidimensional analysis platform: Supports configuring charts and dashboards in a configurable manner; supports various charts such as bar charts, pie charts, line charts, radar charts, etc.; Historical task viewing: Supports visualizing the task results and exporting reports. The refined granularity reaches the data comparison display at 25 mm intervals for each water-cooled wall height.

[0118] 2) Export of thickness measurement data: Support for exporting the thickness data of water wall tubes as an Excel file.

[0119] 3) Have the function of portal dashboard:

[0120] Be able to display the statistical comparison of the number of inspections, inspection duration, and defect information for each boiler;

[0121] Display the inspection duration, defect classification and proportion, detailed defect information, etc. for the most recent inspection task;

[0122] Display the statistical curve of the thickness of each inspection of the selected water wall tube.

[0123] In this specification, each embodiment is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0124] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A robotic automation platform for on-site intelligent maintenance of water walls, characterized in that, Including: A control system, a robot body function module, and a detection module; The control system includes: A control mainboard, which is used to connect to an operation handle, a host computer, the robot body function module, and the detection module, receive and upload detection data, and receive and issue operation instructions; An operation handle, which controls the movement of the robot body through the control mainboard, and controls the opening and closing of the detection module and the robot body function module; A host computer, which is used to receive the detection data sent by the control mainboard and process it; The robot body function module includes: A laser vision automatic deviation correction module, which is used to realize the automatic deviation correction of the robot body path by using laser vision technology; An automatic lane change module, which is used for the automatic lane change function to realize the automatic conversion of the robot operation path; An ultrasonic obstacle avoidance module, which is used to realize the obstacle avoidance function of the robot body; A pose recognition module, which is used for the pose recognition function to calculate and display the real-time pose of the robot body; A 360-degree imaging module, which is used to display the surrounding environment information of the robot body in real time; The detection module includes: An electromagnetic ultrasonic thickness measurement module, which is used to realize the automatic measurement of the water-cooled wall tube thickness by using electromagnetic ultrasonic technology; A defect detection module, which is used to automatically detect the surface defects and defect types of the water-cooled wall.

2. The robotic automation platform for on-site intelligent maintenance of water walls according to claim 1, characterized in that The laser vision automatic deviation correction module includes: A laser sensor, which is used to emit line laser to the membrane water-cooled wall; A camera, which is used to collect the laser image formed by the line laser irradiating on the membrane water-cooled wall tube; An embedded controller module, which is used to process the laser image, generate a path deviation amount, send the path deviation amount to the control mainboard, and control the differential movement of the left and right walking motors of the robot through the control panel to realize the autonomous deviation correction walking along the axial direction of the water-cooled wall tube.

3. The robotic automation platform for on-site intelligent maintenance of water walls according to claim 2, characterized in that, The specific implementation process of the embedded controller module is: Successively perform grayscale conversion, filtering and denoising, and edge enhancement processing on the collected laser image; Extract the straight line segments corresponding to the laser lines by combining Canny edge detection and Hough transform; Calculate the intersection points between the straight line segments or the turning points of the laser lines at the water-cooled wall tube, and define them as vertex feature points; Screen the vertex feature points, and generate a path deviation amount based on the coordinate information of the optimized vertex feature points in the image.

4. The robotic automation platform for on-site intelligent maintenance of water walls according to claim 2, characterized in that, The automatic lane change module includes: A counter, which is used to count the water-cooled wall tubes scanned.

5. The robotic automation platform for on-site intelligent maintenance of water walls according to claim 4, characterized in that, The specific implementation process of the automatic lane change module is: Input the target water-cooled wall tube number; The control mainboard calculates the difference between the water-cooled wall tube number scanned by the current line laser and the target water-cooled wall tube number, and controls the robot body to deflect towards the target water-cooled wall tube; When the counter scans each water-cooled wall tube, the value is incremented by 1. Determine whether the difference between the tube numbers is greater than 1. If it is not greater than 1 and when the value of the counter reaches 1, it means that the line laser has covered the next water-cooled wall tube. The control mainboard automatically activates the path automatic deviation correction, and then the robot body walks along the axis of the target water-cooled wall tube, and the counter value is cleared; When the difference between the tube numbers is greater than 1, and when the value of the counter reaches 1, that is, when the line laser scans to the next water-cooled wall tube, the control main board controls the robot body to stop deflecting and walk straight towards the target water-cooled wall tube; when the value of the counter reaches the difference between the tube numbers, that is, when the line laser scans to the target water-cooled wall tube, the control main board automatically activates the path automatic correction, and then the robot body walks along the axis of the target water-cooled wall tube, and the counter value is cleared to zero.

6. The robotic automation platform for on-site intelligent maintenance of a water wall, according to claim 1, is characterized in that The defect detection module includes a lighting source, a CCD camera and an image acquisition card; The lighting source is used to provide a lighting source; The CCD camera is used to collect the surface image of the water-cooled wall tube under the lighting source; The image acquisition card is used to transmit the surface image of the water-cooled wall tube to the control main board and upload it to the upper computer through the control main board; The upper computer accurately identifies the defect type and records the defect position through the fuzzy neural network algorithm.

7. The robotic automation platform for on-site intelligent maintenance of a water wall, as described in claim 1, is characterized in that, The 360 imaging module includes four cameras, which are respectively installed in the front, rear, left and right directions of the robot body.

8. The robotic automation platform for on-site intelligent maintenance of a water wall, according to claim 7, wherein The control main board is used to synthesize the four images collected by the four cameras to obtain a 360 image.

Images