A converter station valve hall water leakage detection robot and detection method

Abstract

The application discloses a water leakage detection robot and method for valve hall of a converter station, and has a mounting plate on the upper surface of a walking structure, and the mounting plate is provided with the following mechanisms: a roll paper conveying mechanism is mounted on the mounting plate and is used for conveying detection paper horizontally according to a predetermined time length and a step length, and the detection paper horizontally covers the upper surface of the mounting plate; an image acquisition mechanism is mounted at the middle position of the mounting plate and below the roll paper conveying mechanism, and is used for shooting the image of one or more watermarks on the surface of the detection paper; a processor is used for confirming the position of water drops according to the watermark image; wherein, the multiple watermark images are distinguished through the superposition of the images; and a controller is used for controlling the movement of the walking mechanism and the movement of the roll paper conveying mechanism. The application can realize intelligent inspection of the water leakage condition of the valve hall of the converter station, and realizes intelligent image transmission through mesh networking.

Description

Technical Field

[0001] This invention relates to the field of intelligent power grid detection, and specifically to a robot and method for detecting water leakage in the valve hall of a converter station. Background Technology

[0002] Converter valves are core equipment in DC power transmission, primarily relying on water-cooled piping for heat dissipation. During normal operation, if leakage occurs in the main circulation piping inside the valve hall, personnel cannot enter the valve hall and can only rely on infrared detection and visible light cameras for leak inspection. However, limitations in installation location and observation angle make it difficult to accurately pinpoint the leak location in the main pipeline, potentially forcing a shutdown of the DC system. In recent years, with the increasing demand for electrical energy, the reliability of DC power transmission projects has faced even higher requirements. Summary of the Invention

[0003] In view of this, the present invention provides a robot and method for detecting water leakage in the valve hall of a converter station, so as to solve the problem that water leakage is not easy to detect.

[0004] To address the above problems, this invention provides a robot and method for detecting water leakage in the valve hall of a converter station; comprising:

[0005] The upper surface of the walking structure has a mounting plate, and the mounting plate has the following mechanism:

[0006] A roll paper conveying mechanism is installed to horizontally convey test paper according to a predetermined time and step size, wherein the test paper horizontally covers the top of the mounting plate.

[0007] An image acquisition mechanism is installed in the middle of the mounting plate, below the roll paper conveying mechanism, for capturing images of one or more watermarks on the surface of the test paper.

[0008] The processor is used to determine the location of the water droplets based on the watermark image; where, for multiple watermark images, they are distinguished by overlapping the images.

[0009] A controller is used to control the movement of the walking mechanism and the movement of the paper winding mechanism.

[0010] Preferably, it also includes: a camera with its lens facing upwards mounted on the mounting plate, used to capture images of the location containing dripping water in the valve hall directly above the space.

[0011] Preferably, the roll paper conveying mechanism comprises three linearly arranged components;

[0012] Two of the paper roll conveying structures can extend or retract from the boundary of the mounting plate via guide rails;

[0013] Below each roll paper conveying mechanism is an image acquisition mechanism.

[0014] Preferably, the controller has a positioning module, and moves within the valve hall according to the coordinates of the positioning module and a predetermined trajectory.

[0015] Preferably, it also includes a mesh module to form a network system with a connected mesh-enabled router and / or another leak detection robot.

[0016] Embodiments of the present invention also provide a method for detecting water leakage in the valve hall of a converter station, which employs the aforementioned robot and includes the following detection steps:

[0017] The processor obtains a first image through an image acquisition mechanism below the paper roll conveying mechanism, and obtains a second image after the paper roll conveying mechanism operates according to the predetermined time and step size.

[0018] The processor removes a predetermined length of the second image along the traveling direction of the paper roll conveying mechanism to obtain a third image;

[0019] Align the first image and the third image. If the degree of overlap of the watermarks on the two images is related to the predetermined length, then they are considered to come from the same location; otherwise, the watermark is determined to come from an independent leak point.

[0020] Preferably, the detection paper horizontally covers the area directly above the mounting plate, forming an area with more than two logical partitions, and below each logical partition is an image acquisition mechanism;

[0021] The processor performs the detection steps for each logical partition.

[0022] Embodiments of the present invention also provide a method for detecting water leakage in the valve hall of a converter station, employing the aforementioned robot, including:

[0023] After obtaining the first image, the processor controls the robot to move a predetermined step length.

[0024] The processor obtains the second image;

[0025] Align the first image and the second image. If the degree of overlap of the watermarks on the two images is related to the distance of the predetermined step size, then they are considered to come from the same location; otherwise, the watermark is determined to come from an independent leak point.

[0026] This invention, through the aforementioned detection method and robot, can detect the number of leak locations by comparing images, and can acquire images of the leak locations using a zoomable camera. This achieves intelligent inspection. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the equipment structure above the walking mechanism in the embodiment;

[0028] Figure 2 , Figure 3 These are comparison images from the embodiments;

[0029] Figure 4 This is a structural diagram of the test strip with logical partitions in the embodiment;

[0030] Figure 5 This is a schematic diagram of the device structure with a retractable mechanism in the embodiment;

[0031] Figure 6 These are comparison images from the embodiments;

[0032] Figure 7 This is a schematic diagram of the inspection trajectory in the embodiment;

[0033] Figure 8 This is a schematic diagram of the mesh network of multiple robots in the embodiment.

[0034] 1. Robot shell, 2. Left telescopic water receiving plate, 3. Right telescopic water receiving plate, 4. Top fixed water receiving plate, 5. Variable zoom HD camera bracket, 6. Variable zoom HD camera, 7. Telescopic arm controller, 8. Track, 9. Track base, 10. Track, 11. Controller, 12. Battery, 13. Wireless charger, 14. Power supply cable, 15. Induction coil. Detailed Implementation

[0035] To clearly illustrate the solutions in this invention, preferred embodiments are given below and described in detail with reference to the accompanying drawings.

[0036] Embodiments of the present invention provide a leakage detection robot for the valve hall of a converter station, comprising: as follows Figure 1 As shown,

[0037] The upper surface of the walking structure has a mounting plate, and the mounting plate has the following mechanism:

[0038] The traveling mechanism is not shown in the diagram. It can be a tracked vehicle or a motor-controlled tire axle. Multiple sets of wheels enable forward and backward movement. For example, a tracked chassis with remote control functionality could be used. Mounting plates are installed on its surface, such as... Figure 1 The bottom plate of the middle.

[0039] The device is equipped with a paper conveying mechanism that horizontally conveys test paper according to a predetermined time and step size, with the test paper horizontally covering the top of the mounting plate; wherein, the two ends of the paper conveying mechanism are rolls, and the middle has a motor-controlled rotating shaft, which can be controlled to rotate or stop from one end.

[0040] The preset duration here can be 2 seconds, 3 seconds, or 5 seconds, etc. The step size usually refers to the step size of the roll movement. This step size is just enough to remove most of the test strip between the two rolls from the surface of the mounting plate, so that new test strips that have not been photographed are blocked directly above the mounting plate.

[0041] An image acquisition mechanism is installed in the middle of the mounting plate, below the roll paper conveying mechanism, for capturing images of one or more watermarks on the surface of the test paper; the image acquisition mechanism may be a miniature camera with a camera lens.

[0042] The processor is used to determine the location of the water droplets based on the watermark image; wherein, for multiple watermark images, the number of water droplet locations is distinguished by the superposition of the images.

[0043] The miniature camera transmits multiple captured images to a processor, which then overlays the images and determines the number of water droplets based on the overlap of the watermarks.

[0044] The controller is used to control the movement of the walking mechanism and the paper-winding mechanism. This allows the robot to move within the valve hall according to a pre-set trajectory, thereby enabling the inspection of dripping water.

[0045] Preferably, in the embodiment, it further includes: a camera that extends outward and has an upward-facing lens on the mounting plate, used to capture images of the location containing dripping water in the valve hall directly above the space.

[0046] After the processor detects the watermark in the image, it further triggers the upward-facing camera to capture images of the locations containing dripping water within the valve hall. These dripping locations could be pipes used for cooling deionized water.

[0047] Detecting the location of water droplets from images requires distinguishing between droplets at multiple locations. The solutions described in this embodiment can be implemented in several ways, including the following detection steps:

[0048] S10: The processor obtains the first image, and after the paper roll conveying mechanism operates according to the predetermined duration and step size, the processor obtains the second image;

[0049] S11: The processor removes a predetermined length of the second image along the traveling direction of the paper roll conveying mechanism to obtain a third image; Figure 2 In the example, the length of the removed image is b.

[0050] S12: Align the first image and the third image. If the degree of overlap of the watermarks on the two images is related to the predetermined length, then it is considered to come from the same location; otherwise, the watermark is determined to come from an independent leak point.

[0051] For details, please refer to [link / reference]. Figure 2 ,exist Figure 2 In the image, after aligning the two images, the difference between the two watermarked images is 'a'. Comparison reveals that the distance difference 'a' between the two images and the length 'b' of the removed image are essentially identical in pixel size. This confirms that the watermarks in both images originate from the same droplet location.

[0052] Furthermore, such as Figure 3 As shown, after aligning the two images, the difference between the two watermarked images is 'a'. Comparison reveals that the distance difference 'a' between the two images and the length 'b' of the removed image differ significantly in pixel size. If the number of pixels in the distance difference 'a' is much greater than the number of pixels in the image length 'b', or vice versa, then it is considered that the two images originate from two independent water droplet locations.

[0053] Through multiple experiments, it was found that when a water droplet falls onto the test strip, the shape of the watermark varies each time, but the boundary range remains basically consistent. Therefore, by comparing the superimposed images and determining the difference in boundary distance, it can be confirmed whether the watermarks from the two images originate from the same droplet or from two different droplets. This method is more accurate than simply comparing shapes.

[0054] Preferably, in order to achieve more accurate detection of the location and number of water droplets, the number of cameras under the paper roll can be increased, i.e., the number of image acquisition mechanisms on the mounting plate.

[0055] For example, such as Figure 4 As shown, the detection paper horizontally covers the area directly above the mounting plate, forming an area with more than two logical partitions, and below each logical partition is an image acquisition mechanism.

[0056] The processor performs the detection steps S10 to S12 described above on each logical partition, thereby achieving a wider detection range and more accurate detection.

[0057] Preferably, in the embodiments, the above-mentioned leak detection robot, in order to facilitate the acquisition of images of dripping water over a wider range under the same specifications, employs multiple sets of image acquisition mechanisms and a paper roll conveying mechanism.

[0058] For example, in Figure 5The illustrated scheme shows that the paper conveying mechanism includes two linearly arranged, telescopic units: a left telescopic water receiving plate 2 and a right telescopic water receiving plate 3. The telescopic extension and retraction of the water receiving plates on both sides are achieved via a telescopic arm controller 7 and a track 8. The surface of the water receiving plates can serve as a mounting plate. The paper conveying structure is mounted on the surface, and the mounting plate can extend or retract beyond its boundary via a guide rail frame. Below each paper conveying mechanism, an image acquisition mechanism is placed on the surface of the water receiving plate.

[0059] The telescopic arm controller 7 can be a control component for the track 8; the track 8 can be a rigid telescopic guide rail to extend or retract the telescopic water receiving plate. Alternatively, it can be a linear telescopic module, a linear robotic arm, or a screw guide mechanism slide rail, etc.

[0060] The above solution enables robot miniaturization. A paper roll conveying mechanism and a bottom image acquisition mechanism can be placed in the middle of the robot's mounting plate, specifically at the top water receiving plate 4. This expands the watermark detection range. The three paper roll conveying mechanisms can be arranged linearly.

[0061] The robot using this embodiment can also detect leak locations using the following steps: An image acquisition mechanism continuously acquires images at the location of the robot's telescopic or top water-receiving plate and analyzes whether a watermark exists. An image of the detection paper on any of the detection paper's roll conveying mechanisms is captured and sent to the processor to execute the following steps:

[0062] S20: After the processor obtains the first image, it controls the robot to move a predetermined step interval distance b;

[0063] S21: The processor obtains the second image;

[0064] The interval 'b' between the two images can be the movement after the first image is taken, or the positional interval between the coordinates of the shooting positions after the next inspection.

[0065] S22: Align the first image and the second image. If the degree of overlap of the watermarks on the two images is related to the predetermined length, then it is considered to come from the same location; otherwise, the watermark is determined to come from an independent leak point.

[0066] See Figure 6 By using the pixel difference 'a' between the first image and the second image, a mapping relationship can be established with the distance change 'b' between the actual shooting positions, thus obtaining a linear relationship between the deviation of each pixel and the distance unit change (e.g., N centimeters) of the actual shooting position.

[0067] When the difference 'a' in the number of pixels between two images is used to determine the actual distance difference between the shooting positions as 'a1' through a linear relationship, if 'a1' = 'b', then the two watermarks are considered to be at the same leaking location, with only one leaking point; otherwise, if they are different, then they are considered to be two leaking points.

[0068] See Figure 7 The robot in the above embodiments also has a positioning module, and the controller moves within the valve hall according to the coordinates of the positioning module and a predetermined trajectory.

[0069] The robot can perform inspections within the valve hall according to a predetermined trajectory. Since the valve hall equipment is parallel to the ground at the top of the space, common leakage locations of the equipment can be determined on a virtual plane mapped at the top during installation. The inspection trajectory formed by these locations is then used as the inspection trajectory on the ground. The robot then performs the inspections according to this trajectory.

[0070] The processor is used to determine the coordinates of the water droplet that forms the watermark on the valve hall device based on the current trajectory coordinates and the pixel coordinates of the watermark on the detection paper.

[0071] During inspection, the robot uses its own center position, which is also the center position of the detection paper, as the coordinate point of its trajectory. When a water droplet is detected, the pixel coordinates of the watermark at this point can be determined to be different from the pixel coordinates of the center of the detection paper. This pixel difference can be used to calculate the offset between the actual physical coordinates and the robot's current trajectory position, plus the offset difference, to determine the coordinate position of the water droplet on the top of the valve hall. The plane at the top of the valve hall is parallel to the ground and can be considered as the same coordinate system.

[0072] like Figure 7 As shown, it includes the following steps:

[0073] (a) Set up two detection endpoints for the robot, which also serve as the robot's charging points. The endpoints are located on either side below the valve tower water pipes in the hall to be inspected, forming a rectangular area on the ground along a diagonal line. This area is where the robot will perform leak scanning.

[0074] (b) After starting from one side, the robot begins to move forward, moving a distance of 2 / 3 of the telescopic arm length of the telescopic water receiving plate, and stays for a period of time to take pictures and detect whether there is a watermark. This cycle can be set manually.

[0075] (c) After stopping at the current point, the robot moves forward a corresponding distance, with 1 / 3 overlap, which can ensure the integrity of the detection coverage.

[0076] (d) After each dwell time at a point, the image acquisition device records no fewer than three images of the dripping pattern. Utilizing the principle of rotation, the device dwells at each point, thus acquiring test photos of all points. By analyzing the order and overlap of the dripping photos, the leak location can be accurately pinpointed.

[0077] In this embodiment, when the processor detects a watermark in the image, such as a discolored test strip, it can also determine this through image recognition. This triggers the variable-focus high-definition camera 6, which is fixed on the variable-focus camera bracket 5 and faces directly upwards to capture an image of the area containing the watermark. This further pinpoints the exact location of the leak.

[0078] In this embodiment, the robot's walking mechanism can be tracked, with the bottom of the mounting plate having a walking component track 10. A controller 11 is used to control the robot's direction of travel.

[0079] The robot can be wirelessly charged, as shown in the figure, consisting of battery 12 and wireless charger 13.

[0080] Battery 12 can supply power to surrounding electrical equipment, such as image acquisition mechanism, walking mechanism, and variable zoom high-definition camera that captures images of the water leakage location above the valve hall, through power supply cable 14.

[0081] In this embodiment, an induction coil 15 is also provided on the top of the robot. When a strong magnetic field signal is sensed, the robot can avoid the path and thus prevent interference from the strong magnetic field.

[0082] Preferably, the leak detection robot also includes a mesh wireless module, forming a network system with a connected mesh-enabled router and / or another leak detection robot. See also Figure 8 The system employs a robot control network and strategy based on low-frequency signal penetration technology for converter valve halls. This network is based on the characteristic that the valve hall walls have good low-frequency penetration but poor high-frequency signal penetration. It uses low-frequency 2.4GHz Mesh self-organizing network technology for robot networking and control, combined with real-time image transmission technology. This can accurately locate the water pipe leak and enable close-range real-time control of the robot and emergency video viewing and confirmation functions.

[0083] Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention for the solutions described in the various embodiments of the present invention should be included within the protection scope of the present invention.

Claims

1. A robot for detecting water leakage in the valve hall of a converter station, characterized in that, include: The upper surface of the walking structure has a mounting plate, and the mounting plate has the following mechanism: A roll paper conveying mechanism is installed to horizontally convey test paper according to a predetermined time and step size, wherein the test paper horizontally covers the top of the mounting plate. An image acquisition mechanism is installed in the middle of the mounting plate, below the roll paper conveying mechanism, for capturing images of one or more watermarks on the surface of the test paper. The processor is used to determine the location of the water droplets based on the watermark image; where, for multiple watermark images, they are distinguished by overlapping the images. A controller is used to control the movement of the walking mechanism and the movement of the paper winding mechanism; The processor obtains a first image through an image acquisition mechanism below the paper roll conveying mechanism, and obtains a second image after the paper roll conveying mechanism operates according to the predetermined time and step size. The processor removes a predetermined length of the second image along the traveling direction of the paper roll conveying mechanism to obtain a third image; Align the first image and the third image. If the degree of overlap of the watermarks on the two images is related to the predetermined length, then they are considered to come from the same location; otherwise, the watermark is determined to come from an independent leak point.

2. The leak detection robot according to claim 1, characterized in that, Also includes: The mounting plate is equipped with an upward-facing camera to capture images of the dripping water locations within the valve hall directly above the space.

3. The leak detection robot according to claim 1, characterized in that, The paper roll conveying mechanism comprises three linearly arranged components; Two of the paper roll conveying structures can extend or retract from the boundary of the mounting plate via guide rails; Below each roll paper conveying mechanism is an image acquisition mechanism.

4. The leak detection robot according to claim 1, characterized in that, It has a positioning module, and the controller moves within the valve hall according to the coordinates of the positioning module and a predetermined trajectory; The processor is used to determine the coordinates of the water droplet that forms the watermark on the valve hall device based on the current trajectory coordinates and the pixel coordinates of the watermark on the detection paper.

5. The leak detection robot according to claim 1, characterized in that, It also includes a mesh module, which forms a network system with a connected mesh-enabled router and / or another leak detection robot.

6. The leak detection robot according to claim 1, wherein the detection paper horizontally covers the area directly above the mounting plate, forming an area with two or more logical partitions, and below each logical partition is an image acquisition mechanism; The processor performs a detection step for each logical partition.

7. A method for detecting water leakage in the valve hall of a converter station, employing the robot described in claim 1 or 3, characterized in that, include: After obtaining the first image, the processor controls the robot to move a predetermined step length. The processor obtains the second image; Align the first image and the second image. If the degree of overlap of the watermarks on the two images is related to the distance of the predetermined step size, then they are considered to come from the same location; otherwise, the watermark is determined to come from an independent leak point.

Images