Intelligent coal bunker cleaning device based on OpenCV

Abstract

The invention discloses an intelligent coal bunker cleaning device based on OpenCV, and belongs to the technical field of coal bunker cleaning, the intelligent coal bunker cleaning device comprises a supporting plate arranged at the top of a coal bunker, a telescopic supporting beam and a driving device are installed at the top of the supporting plate, a rotating holder is installed at the output end of the telescopic supporting beam, and the rotating holder is located in the coal bunker; a double-spectrum camera is carried on the rotary holder and used for scanning the inner wall of the coal bunker, and a visible light and near-infrared dual-band fusion recognition algorithm is adopted to output a coal bunker adhesion thermodynamic diagram. Through scanning of the double-spectrum camera, analysis of the OpenCV processing module and operation of the automatic cleaning mechanism, remote control and unmanned cleaning are achieved, meanwhile, the double-spectrum camera outputs a coal bunker adhesion thermodynamic diagram by adopting a visible light and near-infrared double-waveband fusion recognition algorithm, accurate target information is provided for subsequent cleaning, and the coal bunker adhesion thermodynamic diagram is more accurate. Compared with traditional manual observation, caking coal can be found more comprehensively and accurately, and cleaning pertinence is improved.

Description

Technical Field

[0001] This invention belongs to the field of coal bunker cleaning technology, specifically relating to an intelligent coal bunker cleaning device based on OpenCV. Background Technology

[0002] In the coal production and storage process, coal bunkers serve as important transfer and storage facilities, and the problem of coal adhering to their inner walls has always been a key technical challenge for the industry. Traditional coal bunker cleaning mainly relies on manual operation, requiring workers to enter the bunker and use simple tools for cleaning. This is not only extremely labor-intensive, but the enclosed space, dim lighting, and presence of toxic and harmful gases and dust inside the coal bunker also greatly increase the risk of safety accidents such as oxygen deficiency, poisoning, and collapse, seriously threatening the lives of workers.

[0003] Meanwhile, manual cleaning suffers from low efficiency and incomplete cleaning. Due to the limitations of manual observation, it is difficult to fully grasp the distribution of coal adhering to the inner wall of the coal bunker, leading to the long-term accumulation of coal in some areas. This not only reduces the effective storage volume of the coal bunker and affects the normal transportation of coal, but may also cause blockages due to coal adhering to the bunker, resulting in production interruptions and huge economic losses to enterprises.

[0004] Furthermore, some existing semi-automatic cleaning devices lack accurate coal-adhering identification capabilities and often rely on indiscriminate cleaning, which not only consumes a lot of energy but also causes significant wear and tear on the inner walls of the coal bunker, shortening its service life. At the same time, these devices have a limited cleaning range and struggle to effectively handle coal adhering in complex locations, exhibiting poor adaptability. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent coal bunker cleaning device based on OpenCV to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent coal bunker cleaning device based on OpenCV, comprising a support plate set on the top of the coal bunker, wherein the support plate is cross-shaped or circular, a telescopic support beam and a driving device are installed on the top of the support plate, and the driving device is located on one side of the telescopic support beam. A rotating gimbal is installed at the output end of the telescopic support beam, and the rotating gimbal is located inside the coal bunker. A dual-spectrum camera is mounted on the rotating gimbal. The dual-spectrum camera is used to scan the inner wall of the coal bunker, and it uses a visible light and near-infrared dual-band fusion recognition algorithm to output a thermal map of coal adhesion.

[0007] In the above implementation process, the support plate can be cross-shaped or circular. The cross-shaped support plate can be continuously erected on the coal bunker without being removed during coal transportation, while the circular support plate can only be used during bunker cleaning. A telescopic support beam and a drive device are installed on the top of the support plate. The drive device is located on one side of the telescopic support beam and can be a winch or other device that can drive the robot to move up and down inside the coal bunker. A rotating gimbal is installed at the output end of the telescopic support beam and is located inside the coal bunker. A dual-spectrum camera mounted on it is used to scan the inner wall of the coal bunker. The telescopic support beam can flexibly adjust the depth of the rotating gimbal into the coal bunker. The rotating gimbal can drive the dual-spectrum camera to rotate at multiple angles to achieve a comprehensive scan of the inner wall of the coal bunker. The dual-spectrum camera uses a visible light and near-infrared dual-band fusion recognition algorithm to output a thermal map of coal adhesion in the coal bunker, providing accurate target information for subsequent cleaning. Compared with traditional manual observation, it can more comprehensively and accurately detect coal adhesion and improve the targeting of cleaning.

[0008] In one specific implementation, the dual-spectrum camera is equipped with an OpenCV processing module. The OpenCV processing module processes the coal adhesion heat map output by the dual-spectrum camera using the OpenCV algorithm to determine the number, outline, and geometric center of the coal to be cleaned.

[0009] In the above implementation process, the OpenCV algorithm uses an adaptive threshold segmentation algorithm to extract the coal-sticking area, determine the number of coal-sticking areas to be cleaned, and determine the number of coal-sticking areas to understand the cleaning workload. The contour detection function is used to extract the contour of the coal-sticking area, and the perimeter, area and circumscribed rectangle parameters of the contour are calculated. The contour information helps to plan the cleaning path. Then, the geometric center coordinates of the coal-sticking area are calculated based on the moment feature algorithm. The geometric center provides accurate positioning for cleaning and is output to the control system, which facilitates the selection of different cleaning methods to clean the coal bunker, making the cleaning work more targeted and efficient, avoiding blind cleaning and improving cleaning efficiency.

[0010] In one specific implementation, a hydraulic cylinder is provided on the outer wall of the coal bunker, and a push rod is installed at the output end of the hydraulic cylinder. One end of the push rod passes through the inner wall of the coal bunker and is connected to a push plate, and one side of the outer wall of the push plate is in contact with the inner wall of the coal bunker.

[0011] In the above process, a hydraulic cylinder is installed on the outer wall of the coal bunker, and a push rod is installed at the output end of the hydraulic cylinder. One end of the push rod passes through the inner wall of the coal bunker and connects to the push plate. The outer wall of one side of the push plate is in contact with the inner wall of the coal bunker. The hydraulic cylinder serves as a power source and transmits power through the push rod to drive the push plate to move, thereby loosening the coal pile blocking the coal bunker outlet and improving the overall cleaning efficiency.

[0012] In one specific implementation, the inner wall of the coal bunker is threaded with a limit rod, and the longitudinal section of the limit rod is T-shaped. A push plate is sleeved on the surface of the limit rod. When the hydraulic cylinder is started, the push plate can be pushed back and forth on the limit rod by the push rod.

[0013] In the above process, the T-shaped limiting rod plays a limiting and guiding role for the push plate, ensuring that the push plate always moves along the correct trajectory during the back and forth sliding process, avoiding the push plate from deviating, and ensuring the cleaning effect. The threaded connection of the limiting rod is easy to install and disassemble. When the limiting rod is worn or damaged, it can be easily replaced, making maintenance convenient.

[0014] In one specific implementation, a robotic suspension arm may be mounted on top of the coal bunker.

[0015] In the above process, the robot's suspension arm is connected to the robot body through a steel wire rope. The robot's suspension arm can move and adjust its position flexibly, and can accurately and stably deliver the cleaning robot body to different positions in the coal bunker, expanding the cleaning range. For some complex or high-positioned sticky coal, the suspension arm can provide effective support and movement path for the robot, improving the device's adaptability to the complex environment inside the coal bunker.

[0016] In one specific implementation, a support unit is mounted on top of the support plate.

[0017] In the above implementation process, when the support plate is a circular structure, a support unit is arranged on its top. The support unit is connected to a movable arm through a telescopic rod, and then the movable arm is connected to the robot body through a cable. The support unit can enhance the overall stability of the device. Especially when the support plate bears heavy components or the working environment is more complex, it can effectively distribute the load, prevent the support plate from being damaged due to excessive force, ensure the safe operation of the device, and extend its service life.

[0018] In one specific implementation, the drive unit, support unit, and robot suspension arm are all connected to the cleaning robot body that extends deep into the coal bunker.

[0019] In the above implementation process, multiple components can be connected to the cleaning robot body, which increases the ways and means for the robot to enter the coal bunker, improves the flexibility and applicability of the device, and can send the robot to the designated location through appropriate connecting components, ensuring effective cleaning of coal stuck in various parts of the coal bunker.

[0020] In one specific implementation, the cleaning robot body includes a support mechanism, a main frame, and a rotating mechanism. The support mechanism is located on both sides of the main frame and is used to fix the cleaning robot body inside the coal bunker. The rotating mechanism is located inside the main frame. A shelf is detachably installed on the output end of the rotating mechanism. An angle adjustment mechanism and a rocker arm mechanism are installed on the bottom of the shelf. The output end of the angle adjustment mechanism is connected to the outer surface of the rocker arm mechanism through a movable part.

[0021] In the above implementation process, the cleaning robot body includes a support mechanism, a main frame, and a rotating mechanism. The support mechanism is located on both sides of the main frame, which can firmly fix the robot in the coal bunker, ensuring that the robot does not shake during the cleaning process and improving the cleaning accuracy. The main frame provides the installation foundation for each component, making the structure compact and reasonable. The rotating mechanism is located in the main frame, and its output end can be detachably installed with a placement plate. An angle adjustment mechanism and a rocker arm mechanism are installed at the bottom of the placement plate, and the output end of the angle adjustment mechanism is connected to the outer surface of the rocker arm mechanism through a movable part. The rotating mechanism can drive the placement plate to rotate. The angle adjustment mechanism and the rocker arm mechanism work together to adjust the cleaning angle and range, allowing the robot to adapt to the cleaning of coal sticking at different positions and angles, improving the flexibility and comprehensiveness of the cleaning.

[0022] In one specific implementation, the output end of the rotary mechanism is detachably equipped with an actuator, and the actuator is an end effector.

[0023] In the above implementation process, the detachable actuator makes it easy to replace the end effector according to different cleaning needs. When it is necessary to clean different types or locations of sticky coal, the appropriate end effector can be quickly replaced, which improves the robot's versatility and adaptability and reduces equipment costs.

[0024] In one specific implementation, the end effector is provided with an adjustable scraper cleaning unit or a pneumatic rotary cleaning unit. The adjustable scraper cleaning unit is a scraper structure with adjustable tilt angle and length, and the tilt angle of the scraper is adjustable in the range of 0-60°. The pneumatic rotary cleaning unit is a specially made chain whip, driven by a pneumatic motor, with a rotation speed of 1000-2000 r / min.

[0025] In the above implementation process, the tilt angle and length of the adjustable scraper cleaning unit are adjustable, which can adapt to the cleaning of coal with different thicknesses and shapes. The angle range of 0-60° can meet the needs of various working conditions. The specially designed chain whip of the pneumatic rotary cleaning unit rotates at high speed under the drive of a pneumatic motor, which has a significant effect on cleaning harder or more firmly attached coal. The rotation speed of 1000-2000r / min can ensure cleaning efficiency. The setting of two cleaning units enables the robot to deal with various types of coal, improving the efficiency of coal bunker cleaning.

[0026] Compared with the prior art, the beneficial effects of the present invention are:

[0027] 1. This invention achieves remote control and unmanned cleaning through dual-spectrum camera scanning, OpenCV processing module analysis, and automated cleaning mechanism operation, fundamentally avoiding the safety hazards such as toxic and harmful gases, dust, and collapse that may be encountered during manual cleaning, and greatly improving the safety of coal bunker cleaning operations.

[0028] 2. This invention uses an OpenCV processing module to determine the number, outline, and geometric center of the coal particles to be cleaned, thus enabling targeted cleaning of coal adhering to the inner wall of the coal bunker. Based on accurate coal information, the control system can rationally select either a pusher plate or a cleaning robot for cleaning, avoiding blind cleaning. The pusher plate can quickly clean large areas and regular locations of coal particles, while the cleaning robot can adjust its angle, length, etc., to specifically handle complex locations and different types of coal particles, significantly improving cleaning efficiency and thoroughness.

[0029] 3. This invention receives thermal images from a dual-spectrum camera and coal-adhering information from OpenCV through a control system. Based on the location and area of ​​the coal-adhering material, it automatically selects a push plate or the cleaning robot to perform the cleaning and controls the various actuators to work together, thereby improving the diversity and convenience of coal bunker cleaning. Attached Figure Description

[0030] Figure 1 This is a three-dimensional structural diagram of the present invention;

[0031] Figure 2 This is a plan view of the present invention;

[0032] Figure 3 This is a schematic diagram of the planar assembly structure of the hydraulic cylinder, push rod, and limit rod of the present invention;

[0033] Figure 4 This is a schematic diagram of the action of the hydraulic rod driving the push rod to push the push plate according to the present invention;

[0034] Figure 5 This is a layout diagram of the circular support plate and support unit of the present invention;

[0035] Figure 6 This is a plan view of the robot's suspension arm according to the present invention;

[0036] Figure 7 This is a planar structural diagram of the robot body of the present invention;

[0037] Figure 8 This is a schematic diagram of the planar assembly structure of the actuator and the adjustable scraper cleaning unit of the present invention;

[0038] Figure 9This is a schematic diagram of the planar assembly structure of the actuator and the pneumatic rotary cleaning unit of the present invention.

[0039] In the diagram: 1. Support plate; 2. Telescopic support beam; 3. Rotating gimbal; 4. Drive device; 5. Support unit; 6. Robot suspension arm; 7. Robot body; 8. Support mechanism; 9. Main frame; 10. Angle adjustment mechanism; 11. Rocker arm mechanism; 12. Rotation mechanism; 13. Actuator; 14. Adjustable scraper cleaning unit; 15. Pneumatic rotary cleaning unit; 16. Limit rod; 17. Push plate; 18. Hydraulic cylinder; 19. Push rod. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] Example 1,

[0042] Please see Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 7 and Figure 8This invention provides an intelligent coal bunker cleaning device based on OpenCV, including a support plate 1 set on the top of the coal bunker, the support plate 1 being cross-shaped. A telescopic support beam 2 and a drive device 4 are installed on the top of the support plate 1, with the drive device 4 located on one side of the telescopic support beam 2. A rotating gimbal 3 is installed at the output end of the telescopic support beam 2, and the rotating gimbal 3 is located inside the coal bunker. A dual-spectrum camera is mounted on the rotating gimbal 3, used to scan the inner wall of the coal bunker. It employs a visible light and near-infrared dual-band fusion recognition algorithm to output a coal adhesion thermal map. The dual-spectrum camera is equipped with an OpenCV processing module, which processes the coal adhesion thermal map output by the dual-spectrum camera using the OpenCV algorithm to determine the number, outline, and geometric center of the coal to be cleaned. A hydraulic cylinder 18 is installed on the outer wall of the coal bunker, and a push rod 19 is installed at the output end of the hydraulic cylinder 18, one end of which penetrates the coal. A push plate 17 is connected to the inner wall of the coal bunker, and one outer wall of the push plate 17 is in contact with the inner wall of the coal bunker. A limit rod 16 is threadedly connected to the inner wall of the coal bunker, and the longitudinal section of the limit rod 16 is T-shaped. The push plate 17 is sleeved on the surface of the limit rod 16. When the hydraulic cylinder 18 is started, it can push the push plate 17 to slide back and forth on the limit rod 16 through the push rod 19. The drive device 4 is connected to the cleaning robot body 7 that is inserted into the coal bunker. The cleaning robot body 7 includes a support mechanism 8, a main frame 9 and a rotary mechanism 12. The support mechanism 8 is set on both sides of the main frame 9 and is used to fix the cleaning robot body 7 in the coal bunker. The rotary mechanism 12 is set in the main frame 9. An actuator 13 is detachably installed at the output end of the rotary mechanism 12. The actuator 13 is an end effector. An adjustable scraper cleaning unit 14 is set on the end effector. The adjustable scraper cleaning unit 14 is a scraper structure with adjustable tilt angle and length. The tilt angle of the scraper is adjustable from 0 to 60°.

[0043] Furthermore, the support plate 1 adopts a cross shape, which can stably support the telescopic support beam 2 and the drive device 4. After the dual-spectrum camera scans the inner wall of the coal bunker to generate an adhesion heat map, the OpenCV processing module accurately determines the information of the sticky coal. For the sticky coal suitable for the push plate 17 to clean, the hydraulic cylinder 18 is activated and the push rod 19 pushes the push plate 17 to slide and clean on the T-shaped limit rod 16. When the sticky coal needs to be cleaned by the robot, the drive device 4 drives the cleaning robot body 7 into the coal bunker, expands outward and fits against the inner wall of the coal bunker, thereby fixing the main frame 9 in the coal bunker. The rotary mechanism 12 drives the end effector equipped with the adjustable scraper cleaning unit 14 to rotate, and pushes the scraper to move through the electric telescopic rod or hydraulic telescopic rod, so that the scraper is close to the inner wall of the coal bunker. Then, the tilt angle of the scraper is adjusted by the cooperation of the hydraulic rod and the moving parts until the scraper fits against the inner wall of the coal bunker. In the whole process, all components work together to achieve efficient and accurate cleaning of the sticky coal on the inner wall of the coal bunker.

[0044] This embodiment is applicable to small and medium-sized cylindrical coal bunkers, where the coal adhering to the inner wall is mainly a thin layer, a large area of ​​continuous adhesion, and the coal adhering location is concentrated in the lower part of the coal bunker. The specific workflow is as follows:

[0045] Support plate 1 adopts a cross shape and is fixed to the precast concrete beam at the top of the coal bunker with expansion bolts. Telescopic support beam 2 can be an electric push rod structure with a telescopic stroke of 0-3 meters. It can drive the rotating gimbal 3 to move radially in the coal bunker. The horizontal rotation angle of the rotating gimbal 3 is 0-360° and the vertical pitch angle is 30° to 90°, ensuring that the dual-spectrum camera scans the bunker wall without blind spots. After the thermal map output by the dual-spectrum camera is analyzed by the OpenCV processing module, if a large amount of coal accumulation is detected at the bottom of the coal bunker, the control system will first activate the hydraulic cylinder 18, which will drive the push rod 19 to push the push plate 17 back and forth. The push plate 17 is made of wear-resistant manganese steel plate. If there is blocky or strip-shaped sticky coal, the drive device 4 will drive the cleaning robot body 7 to move down. The support mechanism 8 adopts a three-section hydraulic outrigger, and rubber anti-slip pads are installed at the contact end with the bunker wall. The rotary mechanism 12 drives the adjustable scraper cleaning unit 14 to rotate. The scraper is made of high-chromium cast iron, and by adjusting the scraper angle and extension length, it can peel off the sticky coal layer by layer.

[0046] Example 2,

[0047] Please see Figure 1 , Figure 2 , Figure 7 and Figure 9 This invention provides an intelligent coal bunker cleaning device based on OpenCV, comprising a support plate 1 set on the top of the coal bunker, the support plate 1 being cross-shaped, a telescopic support beam 2 and a drive device 4 mounted on the top of the support plate 1, the drive device 4 being located on one side of the telescopic support beam 2, a rotating gimbal 3 mounted on the output end of the telescopic support beam 2, and the rotating gimbal 3 being located inside the coal bunker, the rotating gimbal 3 being equipped with a dual-spectrum camera, the dual-spectrum camera being used to scan the inner wall of the coal bunker, and employing a visible light and near-infrared dual-band fusion recognition algorithm to output a coal adhesion thermal map of the bunker, the dual-spectrum camera being equipped with an OpenCV processing module, the OpenCV processing module processing the coal image output by the dual-spectrum camera using the OpenCV algorithm. The heat map of coal adhesion in the coal bunker is processed to determine the number, outline, and geometric center of the coal to be cleaned. The drive device 4 is connected to the cleaning robot body 7, which is inserted into the coal bunker. The cleaning robot body 7 includes a support mechanism 8, a main frame 9, and a rotary mechanism 12. The support mechanism 8 is located on both sides of the main frame 9 and is used to fix the cleaning robot body 7 in the coal bunker. The rotary mechanism 12 is located in the main frame 9. The output end of the rotary mechanism 12 is detachably equipped with an actuator 13, which is an end effector. The end effector is equipped with a pneumatic rotary cleaning unit 15, which is a specially designed chain whip driven by a pneumatic motor with a rotation speed of 1000-2000 r / min.

[0048] Furthermore, the dual-spectrum camera and OpenCV processing module work together to determine the information of sticky coal. For sticky coal that is difficult to clean, the drive device 4 is connected to the cleaning robot body 7. After the robot is sent into the coal bunker, the support mechanism 8 ensures the stability of the robot. The rotary mechanism 12 drives the end effector equipped with the pneumatic rotary cleaning unit 15 to rotate. The chain-type special whip rotates at a high speed of 1000-2000 r / min under the drive of the pneumatic motor. The chain-type special whip collides with the material hanging on the inner wall of the coal bunker, thereby sweeping off the coal sludge sticking to the inner wall. It is especially suitable for cleaning firmly attached sticky coal.

[0049] This embodiment is applicable to large conical coal bunkers where the inner wall contains hard, strongly adhesive lumps of coal, and the bunker wall is inclined at an angle of 30°-60°. The specific workflow is as follows:

[0050] The support plate 1 is cross-shaped and welded from Q355B steel, with reinforcing ribs welded to the bottom to ensure no significant deformation when the robot is suspended. When scanning with a dual-spectrum camera, the near-infrared band focuses on identifying high-temperature coal accumulation areas, while the visible light band captures the edges of clumps. The OpenCV algorithm extracts the circumscribed rectangle parameters of the sticky coal through a contour detection function. The cleaning robot body 7 is driven by the drive device 4 to move up and down in the coal bunker. The rotary mechanism 12 uses a hollow shaft reducer to drive the end effector to rotate at high speed. The chain-type special whip of the pneumatic rotary cleaning unit 15 uses 8mm steel wire rope wrapped with polyurethane. During operation, after each clump area is cleaned, the robot's real-time position can be adjusted by the robotic arm or the drive device 4 to achieve continuous operation.

[0051] Example 3,

[0052] Please see Figure 6 and Figure 7 This invention provides an intelligent coal bunker cleaning device based on OpenCV, including a robot suspension arm 6 mounted on the top of the coal bunker. The robot suspension arm 6 is connected to a cleaning robot body 7 that extends into the coal bunker. The cleaning robot body 7 includes a support mechanism 8, a main frame 9, and a rotating mechanism 12. The support mechanism 8 is located on both sides of the main frame 9 and is used to fix the cleaning robot body 7 inside the coal bunker. The rotating mechanism 12 is located inside the main frame 9. A shelf is detachably installed at the output end of the rotating mechanism 12. An angle adjustment mechanism 10 and a rocker arm mechanism 11 are installed at the bottom of the shelf. The output end of the angle adjustment mechanism 10 is connected to the outer surface of the rocker arm mechanism 11 through a movable part.

[0053] Furthermore, a robot suspension arm 6 is erected on the top of the coal bunker, and the robot is connected by a steel wire rope. When cleaning the coal bunker, the dual-spectrum camera scans and the data is processed by the OpenCV processing module. The robot suspension arm 6 accurately delivers the cleaning robot body 7 to the location where coal is stuck. The support mechanism 8 fixes the robot, the rotation mechanism 12 drives the placement plate to rotate, and the angle adjustment mechanism 10 and the rocker arm mechanism 11 work together to adjust the tilt angle and extension length of the rocker arm mechanism 11 according to actual needs. This allows for flexible cleaning of coal stuck in different locations, expands the cleaning range, and is especially suitable for large or complex coal bunkers.

[0054] This embodiment is applicable to vertical long coal bunkers, where it is necessary to clean the dispersed sticky coal on the bunker walls. The specific workflow is as follows:

[0055] The robotic suspension arm 6, mounted on the top of the coal bunker, is a three-section telescopic arm equipped with a winch driven by a servo motor. This allows the cleaning robot body 7 to be suspended to any depth via cables. During scanning by a dual-spectrum camera, a laser ranging module is added to simultaneously obtain the distance between the sticky coal and the top. The OpenCV processing module marks and sorts the coordinates, and the control system plans the cleaning path from top to bottom. After the cleaning robot body 7 moves down to the target position, pressure sensors are installed at the top of the outriggers of the support mechanism 8 to detect the tension and ensure stability during high-altitude operations. The slewing mechanism 12 drives the placement plate to rotate 360°. The angle adjustment mechanism 10 and the rocker arm mechanism 11 form a two-degree-of-freedom robotic arm with a working radius of 2m. A high-definition camera can be installed at the end as needed to transmit the cleaning image in real time. Operators can manually adjust the posture via a remote control console.

[0056] Example 4,

[0057] Please see Figure 5 and Figure 9 This invention provides an intelligent coal bunker cleaning device based on OpenCV, including a support plate 1 set on the top of the coal bunker, the support plate 1 being circular, a support unit 5 installed on the top of the support plate 1, a telescopic support beam 2 installed on the bottom of the support plate 1, a rotating gimbal 3 installed at the output end of the telescopic support beam 2, and the rotating gimbal 3 being located inside the coal bunker, the rotating gimbal 3 being equipped with a dual-spectrum camera, the dual-spectrum camera being used to scan the inner wall of the coal bunker, it using a visible light and near-infrared dual-band fusion recognition algorithm to output a coal bunker adhesion heat map, the dual-spectrum camera being equipped with an OpenCV processing module, the OpenCV processing module processing the coal bunker adhesion heat map output by the dual-spectrum camera through the OpenCV algorithm, thereby determining the number, outline and geometric center of the coal to be cleaned, and the support unit 5 being connected to a cleaning robot body 7 that extends into the coal bunker.

[0058] Furthermore, the support plate 1 is circular, and the support unit 5 installed on the top enhances the stability of the overall structure. After the dual-spectrum camera and OpenCV processing module determine the coal sticking information, the support unit 5 sends the cleaning robot body 7 into the coal bunker through the movable arm and cable. Different cleaning methods can be selected to treat the coal sticking on the inner wall of the coal bunker according to the actual use scenario. It is suitable for coal bunker cleaning scenarios with high stability requirements.

[0059] The working principle and usage process of this invention: The telescopic support beam 2 can adjust the position of the rotating gimbal 3. The rotating gimbal 3 drives the dual-spectrum camera to perform an all-round scan of the inner wall of the coal bunker. The dual-spectrum camera uses a visible light and near-infrared dual-band fusion recognition algorithm to generate a coal bunker adhesion heat map. The OpenCV processing module processes the heat map to accurately determine the relevant information of the sticky coal. Subsequently, the control system selects an appropriate cleaning method according to the sticky coal information. If the sticky coal is suitable for cleaning by the push plate 17, the hydraulic cylinder 18 is activated, and the push plate 17 is pushed to slide on the limit rod 16 for cleaning through the push rod 19. If robot cleaning is required, the drive device 4, support unit 5 or robot suspension arm 6 can be selected according to different actual scenarios to drive the cleaning robot body 7 to move to the designated position. The cleaning is carried out by the rocker arm mechanism 11 or the cleaning unit of the end effector. During the cleaning process, the support mechanism 8 can use a hydraulic rod to support the inner wall of the coal bunker to ensure the stability of the robot, and the multi-angle and all-round cleaning is achieved through the rotary mechanism 12.

[0060] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An OpenCV-based intelligent cleaning device for coal bunker, comprising a support plate (1) arranged at the top of the coal bunker, and the support plate (1) is cross-shaped or circular, characterized in that, The top of the support plate (1) is provided with a telescopic support beam (2) and a driving device (4), and the driving device (4) is located on one side of the telescopic support beam (2), the output end of the telescopic support beam (2) is provided with a rotating holder (3), and the rotating holder (3) is located in the inside of the coal bunker, a dual-spectrum camera is mounted on the rotating holder (3), and the dual-spectrum camera is used for scanning the inner wall of the coal bunker, and a visible light and near-infrared dual-band fusion recognition algorithm is used to output a coal adhesion thermal map of the coal bunker.

2. The OpenCV-based intelligent cleaning device for coal bunkers according to claim 1, characterized in that, The dual-spectrum camera is provided with an OpenCV processing module, and the OpenCV processing module processes the coal adhesion thermal map output by the dual-spectrum camera through an OpenCV algorithm, so as to determine the number, contour and geometric center of the coal to be cleaned.

3. The OpenCV-based intelligent cleaning device for coal bunkers according to claim 1, characterized in that, The outer wall of the coal bunker is provided with a hydraulic cylinder (18), the output end of the hydraulic cylinder (18) is provided with a push rod (19), one end of the push rod (19) penetrates through the inner wall of the coal bunker and is connected with a push plate (17), and the outer wall of one side of the push plate (17) is attached to the inner wall of the coal bunker.

4. The OpenCV-based intelligent cleaning device for coal bunkers according to claim 3, characterized in that, The inner wall of the coal bunker is threadedly connected with a limiting rod (16), the longitudinal section of the limiting rod (16) is in a T-shaped structure, the push plate (17) is sleeved on the surface of the limiting rod (16), and when the hydraulic cylinder (18) is started, the push plate (17) can slide back and forth on the limiting rod (16) through the push rod (19).

5. The OpenCV-based intelligent cleaning device for coal bunkers of claim 1, characterized in that, The top of the coal bunker is provided with a robot suspension arm (6).

6. The OpenCV-based intelligent cleaning device for coal bunkers of claim 1, characterized in that, The top of the support plate (1) is provided with a support unit (5).

7. The OpenCV-based intelligent cleaning device for coal bunkers according to any one of claims 1, 5 or 6, characterized in that, The driving device (4), the support unit (5) and the robot suspension arm (6) are all connected with a cleaning robot body (7) which penetrates into the coal bunker.

8. The OpenCV-based intelligent cleaning device for coal bunkers according to claim 7, characterized in that, The cleaning robot body (7) comprises a support mechanism (8), a main frame (9) and a rotary mechanism (12), the support mechanism (8) is arranged on both sides of the main frame (9) and is used for fixing the cleaning robot body (7) in the coal bunker, the rotary mechanism (12) is arranged in the main frame (9), the output end of the rotary mechanism (12) is detachably provided with a storage plate, the bottom of the storage plate is provided with an angle adjusting mechanism (10) and a rocker mechanism (11), and the output end of the angle adjusting mechanism (10) is connected with the outer surface of the rocker mechanism (11) through a movable part.

9. The OpenCV-based intelligent cleaning device for coal bunkers according to claim 8, characterized in that, The output end of the rotary mechanism (12) is detachably provided with an execution mechanism (13), and the execution mechanism (13) is an end effector.

10. The OpenCV-based intelligent cleaning device for coal bunkers according to claim 9, characterized in that, An adjustable scraper cleaning unit (14) or a pneumatic rotary cleaning unit (15) is arranged on the end effector, the adjustable scraper cleaning unit (14) is a scraper structure with adjustable inclination angle and length, the inclination angle of the scraper is adjustable in the range of 0-60°, and the pneumatic rotary cleaning unit (15) is a chain type special whip driven by a pneumatic motor, and the rotating speed is 1000-2000r / min.

Images