A magnetic replacement device and method for a skip liner
By designing a magnetic replacement device for skip liner plates, and utilizing wall-climbing robots and image processing technology, intelligent identification and automatic replacement of skip liner plates are achieved. This solves the problems of low efficiency and high safety risks associated with manual replacement, and improves replacement efficiency and detection accuracy.
Patent Information
- Application Number
- CN202210734994.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-06-27
- Publication Date
- 2025-12-12
- Estimated Expiration
- 2042-06-27
AI Technical Summary
Manual replacement of skip liners is inefficient and dangerous, and existing technologies cannot effectively solve the problem of intelligent replacement of skip liners.
Design a magnetic suction replacement device for skip liner plates, including a wall-climbing robot, a gripping plate, a cleaning mechanism, and an image acquisition unit. The wall-climbing robot performs image acquisition and cleaning of the liner plates inside the skip, and the damaged areas are identified by combining image processing algorithms. The electromagnetic chuck is then used to automatically grip and replace the liner plates.
It enables intelligent identification and automatic replacement of skip liner plates, improving replacement efficiency, reducing safety risks, minimizing safety accidents caused by manual inspection, and enhancing detection accuracy and replacement reliability.
Smart Images

Figure CN115626573B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to a kind of bucket liner magnetic replacement device and method. BACKGROUND
[0002] In coal mine hoisting, often use bucket as hoisting container. When loading operation, the inner wall will be subjected to severe impact, under the severe impact and friction of coal and gangue, its service life is greatly shortened. To extend the service life of bucket, the main solution is to install wear-resistant lining at specific position of bucket wall, and according to the intensity of transfer work, the approximate replacement time of lining is estimated, and the damaged lining is often replaced. At present, the replacement method of bucket lining at home and abroad is mostly manual operation. However, due to the small space in the bucket, and there is no support platform, workers need to be suspended by rope for operation. Therefore, the manual replacement method has low work efficiency and high risk coefficient, and the working environment in the bucket is poor, especially in summer, which is seriously affected by high temperature. Therefore, it is urgent to seek an intelligent replacement method for bucket lining.
[0003] Application No. 202110330662.5 discloses a hoisting container lining wear prediction method and lining optimization layout method, which can effectively predict the wear of each block of lining of metal mine hoisting container, reduce the total weight of lining, improve its service life, reduce the replacement frequency of lining, has high processing efficiency and low processing cost;
[0004] Application No. 201811101517.4 discloses a mine large-tonnage coal falling buffer bucket; the large-tonnage coal falling buffer bucket can effectively alleviate the impact wear and damage of large-tonnage bucket lining under coal falling loading working condition, can realize quick disassembly and replacement of damaged lining, prolong the service life of large-tonnage bucket, and improve the production efficiency of coal mine.
[0005] However, the above two inventions mainly slow down the wear state of bucket lining from the aspects of lining wear prediction and improvement of bucket structure, and the replacement of bucket lining is still inevitable. SUMMARY
[0006] Technical problem: The purpose of the present application is to solve the problems of low efficiency and high risk coefficient of manual replacement of bucket lining, and provide a kind of bucket lining magnetic replacement device, which can realize intelligent replacement of damaged bucket lining.
[0007] Technical scheme: In order to achieve the above technical purpose, the present application adopts the following technical scheme:
[0008] A kind of bucket lining magnetic replacement device, comprising:
[0009] The wall climbing robot is adsorbed on the inner wall of any side of the bucket, comprising a support frame, a permanent magnet wheel and a cover plate, the permanent magnet wheel is connected to the two sides of the support frame, and the cover plate is fixed on the top of the support frame;
[0010] The grabbing plate is arranged on one side of the support frame near the lining plate through a grabbing plate position adjusting mechanism, and is used for replacing the damaged lining plate,
[0011] The cleaning mechanism is arranged on the support frame, and is used for cleaning the surface of the lining plate;
[0012] The image acquisition unit is arranged on the grabbing plate, and is used for acquiring the image of the surface of the lining plate; the image acquisition unit comprises three first cameras fixed on the three sides of the support frame without the grabbing plate and one second camera fixed on the grabbing plate;
[0013] The external traction device is arranged outside the bucket, is connected with the cover plate on the top of the wall climbing robot through a traction rope, and is used for sending or taking out the wall climbing robot from the bucket, and comprises:
[0014] The lifting platform and the lifting device, wherein,
[0015] The lifting platform is located at the bottom of the entire traction device;
[0016] The lifting device comprises a support frame, a winch, a lifting steel wire rope, a first fixed pulley and a second fixed pulley, wherein the support frame and the winch are fixed on the lifting platform;
[0017] The support frame comprises a vertical support rod and a horizontal cantilever rod, the first fixed pulley is installed on the top of the vertical support rod, and the second fixed pulley is installed on the overhanging end of the horizontal cantilever rod; the lifting steel wire rope released by the winch is connected with the lifting ring on the top of the wall climbing robot through the first fixed pulley and the second fixed pulley;
[0018] The horizontal laser positioning device is installed on the overhanging end of the horizontal cantilever rod, and the horizontal position of the wall climbing robot is judged by measuring the distance between the wall climbing robot and the bucket wall;
[0019] The vertical laser positioning device is installed on the overhanging end of the horizontal cantilever rod, and the vertical position of the wall climbing robot is judged by measuring the distance between the cover plate of the wall climbing robot and the wall climbing robot.
[0020] The grabbing plate position adjusting mechanism comprises:
[0021] A lead screw motor, a movable frame support plate, a lead screw, a nut block, a lead screw mechanism support frame, a sliding rail, a movable frame, a first servo electric cylinder and a second servo electric cylinder, wherein,
[0022] The output shaft of the lead screw motor is connected with the lead screw through a shaft coupling, the lead screw is connected with a nut block through a thread, the upper surface of the nut block is fixedly connected with a movable frame support plate, the lead screw motor is connected with a lead screw mechanism support frame, the lead screw is connected with the lead screw mechanism support frame through bearings at both ends, the lead screw mechanism support frame is fixed on the movable frame support plate, the movable frame support plate is fixed on a movable frame, the movable frame is slidably connected with the slide rail through a sliding block, and the slide rail is fixed on the support frame.
[0023] The first servo cylinder and the second servo cylinder are respectively fixedly connected with the movable frame, and the telescopic rods of the first servo cylinder and the second servo cylinder are respectively connected with the same side of the grabbing plate.
[0024] The support frame is formed by profiled rod members and corners connected through bolts.
[0025] The grabbing plate comprises a grabbing plate body, a hook claw, a compression spring and an electromagnetic suction plate, wherein the grabbing plate body is connected with the grabbing plate position adjusting mechanism through a connecting piece and a pin shaft,
[0026] The electromagnetic suction plate is fixed on the grabbing plate and controls the adsorption capacity of the electromagnetic suction plate through an energized state.
[0027] The hook claw comprises four hook claws, and the four hook claws are respectively arranged at four corners of the grabbing plate through hinged parts.
[0028] The compression spring is arranged between the hook claw and the grabbing plate, when the grabbing plate body approaches the backing plate, the upper and lower ends of the backing plate push the hook claw, the hook claw rotates backward around the hinged part as a fulcrum to clamp the upper and lower ends of the backing plate, and meanwhile, the electromagnetic suction plate can be in contact with the backing plate.
[0029] The lifting platform is a scissor lifting platform, comprising a lockable universal wheel, a base, a third servo cylinder, a scissor support rod and a support plate, wherein the lockable universal wheel is fixed on the base through bolts, one end of the scissor support rod is hinged to the base, the other end is hinged to the support plate, the bottom ends of the two third servo cylinders are fixed on the base, and the telescopic rods are hinged to the scissor support rod.
[0030] The top of the cover plate is provided with a lifting ring for external traction.
[0031] The first camera is fixedly connected with the support frame through a camera support, and the second camera is fixedly connected with the middle inner side of the grabbing plate through a camera support.
[0032] The application further discloses a working method of the magnetic type replacing device for the ladle backing plate, and the working method comprises the following steps:
[0033] S1 adjustment position
[0034] The bucket replacement device is manually pushed to the vicinity of the shaft, the height of the lifting platform is adjusted according to the height of the bucket inlet, the external traction device drives the wall-climbing robot to enter the bucket from the center of the bucket inlet, the lateral distance is first adjusted, the position information fed back in real time by the lateral laser positioning device is used to confirm the distance between the wall-climbing robot and the rear wall of the bucket, so as to prevent the wall-climbing robot from colliding with the rear wall of the bucket and causing damage to the wall-climbing robot, and enable the three first cameras on the wall-climbing robot to simultaneously detect the bucket wall in three directions; after the lateral position of the wall-climbing robot is determined, the wall-climbing robot is driven to move longitudinally by the external traction device, so as to ensure that the permanent magnet wheel of the wall-climbing robot can be reliably adsorbed on the side wall of the bucket; the winch is started, and the wall-climbing robot moves downward under the action of its own gravity, and at the same time, the longitudinal laser positioning device monitors the downward movement distance of the wall-climbing robot,
[0035] S2 internal cleaning
[0036] The wall-climbing robot cleans the lining plate through the cleaning mechanism, after the first cleaning is completed, the lining plate is imaged by the second camera, the cleaning requirement is judged by analyzing the image information of the lining plate, if the cleaning requirement is met, step S3 is entered, if the cleaning requirement is not met, secondary cleaning is performed until the cleaning requirement is met;
[0037] S3 shooting S31, the wall-climbing robot is adsorbed on the side of the bucket, the cameras on the other three non-adsorbing surfaces are opened, and the respective shooting areas are vertically detected from top to bottom;
[0038] S32, after the wall-climbing robot reaches the bottom of the bucket, the permanent magnet wheel on the wall-climbing robot is switched to the horizontal direction, the wall-climbing robot is horizontally moved for a distance, the cameras for detecting the front and rear walls are closed, and then the opposite side of the adsorbed side of the bucket is shot from bottom to top;
[0039] S33, after the wall-climbing robot reaches the top end of the bucket, the permanent magnet wheel on the wall-climbing robot is switched to the horizontal direction, the wall-climbing robot is horizontally moved for a distance, and then the side of the adsorbed surface of the robot is shot from top to bottom;
[0040] S34, the wall-climbing robot is adsorbed to the opposite side, and steps S31-S33 are repeated to complete the shooting of all surfaces of the bucket.
[0041] S35, image processing, the images shot in step S34 are processed:
[0042] The geometric size and position coordinates of the defects on the surface of the bucket lining plate are obtained, if the defect degree meets the replacement requirement of the lining plate, a warning is sent to the client to remind the worker to replace the lining plate in time;
[0043] S4 secondary adjustment position
[0044] After scanning the liner plate on the entire inner wall of the bucket, the position of the liner plate to be replaced is determined, the wall climbing robot is first adjusted horizontally and then vertically, so that the permanent magnet wheel is adsorbed on the side wall of the bucket;
[0045] S5 grabbing and recycling damaged liner plate
[0046] Through the position adjustment mechanism of the grabbing plate, the grabbing plate is moved towards the liner plate, the position of the grabbing plate is adjusted to align the outline center of the grabbing plate with the outline center of the liner plate in combination with the picture collected by the second camera, and at the same time, the bolts between the liner plate and the bucket are manually disassembled outside. After alignment, the damaged liner plate is adsorbed by the electromagnetic chuck on the grabbing plate; after the above work is completed, the wall climbing robot is moved out of the bucket and placed in a safe position, the electromagnetic chuck switch is turned off, and the damaged liner plate is removed by the worker;
[0047] S6 replacing new liner plate
[0048] After removing the damaged liner plate, the worker installs the new liner plate on the grabbing plate, the bolts are pre-welded on the corresponding installation position on the new liner plate, so that the outline center of the new liner plate is aligned with the outline center of the grabbing plate, the electromagnetic chuck switch is started, and the electromagnetic chuck adsorbs the new liner plate. Then, the wall climbing robot is raised to the initial position by the external traction device, the wall climbing robot is lowered to the liner plate replacement position, the position of the grabbing plate is adjusted by the position adjustment mechanism of the grabbing plate, so that the new liner plate is aligned with the installation position, and the new liner plate is connected with the bucket with the assistance of the worker outside the bucket.
[0049] The step S2 specifically comprises the following sub-steps:
[0050] S21, performing primary cleaning on the surface of the liner plate by the spray head;
[0051] S22, collecting the image of the surface of the bucket liner plate after primary cleaning by the second camera;
[0052] S23, performing gray scale conversion on the image of the surface of the liner plate, and performing fuzzy noise reduction to remove small noises in the image;
[0053] S24, detecting whether there is an area with a gray value less than 30 in the image, if there is, considering it as an area not cleaned ideally, recording the specific position of the area, and performing secondary cleaning on the area;
[0054] S25, collecting the image of the surface of the bucket liner plate after secondary cleaning, repeating steps S22-S23, and performing third cleaning on the surface of the liner plate;
[0055] S26, inspect the surface of the skip liner again. If there are no areas with a gray value less than 30 in the image, proceed to step S3. If there are still areas with unsatisfactory cleaning in the image, save the image with these areas, record the location of these areas, and send the image to the ground control room for manual judgment of the mud type and selection of the processing method.
[0056] Step S3 specifically includes the following sub-steps:
[0057] S31, the image is segmented using the global threshold segmentation method;
[0058] S32 uses a bilateral filtering method to filter out interference in the image and suppress high-frequency noise in the original image;
[0059] S33 uses morphological operations to refine and trim the burrs in the image, making it present a continuous and smooth edge;
[0060] S34, use the Canny edge detection operator to perform edge detection on the result image obtained in step D3;
[0061] S35, perform bounding rectangle fitting on the defect pattern on a single liner, and then filter out false markers in the image by the area of the bounding rectangle. This includes the following sub-steps:
[0062] S351, perform defect contour fitting on the segmented individual lining plates, using the circumscribed rectangle R. a For the detected defect profile C a To perform fitting, the circumscribed rectangle R a The parameters are defined as follows: First, determine the contour C to be fitted. a Find the point P1(x1,y1) with the largest ordinate, and draw a horizontal line L1 through this point; then, determine the contour C to be fitted. a Find the point P2(x2,y2) with the largest x-coordinate, and draw a horizontal line L2 through this point; then, calculate the intersection point P3 of L1 and L2 as a vertex of the circumscribed rectangle; finally, take the difference between the ordinate of the point with the smallest ordinate on the contour and the ordinate of P1 as the circumscribed rectangle R. a The first side length, and the difference between the x-coordinate of the point with the smallest x-coordinate on the contour and the x-coordinate of P2, is the circumscribed rectangle R. a The length of the other side determines the circumscribed rectangle R. a ;
[0063] S352, through the circumscribed rectangle R a area S a To remove false markers from the previous step, the specific method is to calculate the circumscribed rectangle R. a area S a The ratio k to the total area S of the image:
[0064]
[0065] A reasonable numerical interval Z is set according to the proportion of the liner defects in the map, and all the rectangles whose k values do not belong to the interval are considered as false marks and filtered out;
[0066] S36, the actual size of the surface defect area of the bucket liner is calculated, a quantification corresponding model of defects of different sizes and defect degrees is established, and the wall climbing robot performs different measures according to different defect degrees, specifically including the following steps:
[0067] S361, the perspective transformation correction is performed on the liner image with reference to the edge vertex of the liner;
[0068] S362, the Hough transformation algorithm is used to detect straight lines in the image, and the edges of the single bucket liner are determined in combination with the geometric structure of the surface of the bucket liner;
[0069] S363, the actual geometric size of the single liner and the numerical proportion of the pixel number of the single liner in the image are used as the parameter scale to calculate the actual size of the surface defects of the liner in the image;
[0070] S364, the defect condition is comprehensively judged according to the size of the size defect and the average gray value of the defect:
[0071] The defect with the maximum direction size siem∈[0,20]mm and the average gray value agv∈[100,150] is a small defect, the influence of the defect on the service effect of the bucket liner is small, the defect is not processed temporarily, the defect information is recorded, and the wall climbing robot continues to detect;
[0072] The defect with the maximum direction size siem∈[20,50]mm and the average gray value agv∈[70,100] is a medium defect, the defect has a certain influence on the service effect of the bucket liner, a warning should be sent to the ground working room to remind the worker to consider replacing or repairing the liner, meanwhile, the defect information is recorded, and the wall climbing robot continues to detect;
[0073] The defect with the maximum direction size siem>50mm and the average gray value agv<70 is identified as a large defect, a warning is sent to the ground working room to remind the worker to replace the liner in time, meanwhile, the defect information is recorded, and the wall climbing robot does not perform subsequent detection of the liner.
[0074] Beneficial effects:
[0075] The intelligent identification and replacement of the damaged bucket liner can be realized, and the main advantages are:
[0076] First: the present application is aimed at the technical problem of inconvenient damage detection of existing skip lining, proposes a wall climbing robot, realizes image collection of the inner wall of the skip through the camera installed on the wall climbing robot, and then processes the surface image of the lining through image processing means, instead of manual inspection into the skip, to avoid safety accidents caused by manual inspection.
[0077] Second: the present application proposes a mortar cleaning scheme combined with machine vision technology according to the characteristics that mortar may be attached to the surface of the lining, and the ideal area is cleaned multiple times by identifying the image features of the cleaned lining surface, effectively reducing the defect mis-detection phenomenon caused by the attachment of mortar on the surface of the lining, and making the adsorption effect of the permanent magnet wheel of the wall climbing robot more reliable.
[0078] Third: the present application proposes an image processing algorithm to solve the problem that the defects on the lining are not easy to detect and the background interference is more, which proposes an area screening method to filter out the false markers in the image, thereby completing the defect detection work of the lining of the vertical shaft lifting method, and improving the accuracy of the damage detection of the lining.
[0079] Fourth: the present application designs a targeted detection effect evaluation method combined with the detection characteristics of the wall climbing robot: first, the image of the lining is corrected through perspective transformation, which can effectively reduce the influence of inclination and shaking of the robot during operation on the detection effect, then the actual size of the defect part is directly obtained through the size conversion of the single lining and the known parameters, avoiding the cumbersome camera calibration operation, and finally different feedbacks are made according to the characteristics of the lining defects, which can enable workers to quickly identify the defect type and take corresponding measures in time. BRIEF DESCRIPTION OF DRAWINGS
[0080] Figure 1 It is a front view of the invention of the magnetic type replacement device for the lining of the skip;
[0081] Figure 2 It is a top view of the invention of the magnetic type replacement device for the lining of the skip;
[0082] Figure 3 It is a schematic diagram of the external traction device base mechanism of the invention of the magnetic type replacement device for the lining of the skip;
[0083] Figure 4 It is a schematic diagram of the traction type wall climbing robot structure of the invention of the magnetic type replacement device for the lining of the skip;
[0084] Figure 5 It is a front view of the traction type wall climbing robot of the invention of the magnetic type replacement device for the lining of the skip;
[0085] Figure 6 It is a schematic diagram of the position adjusting device structure of the traction type wall climbing robot of the invention of the magnetic type replacement device for the lining of the skip;
[0086] Figure 7 Permanent magnet wheel structure diagram of the invention;
[0087] Wherein: 1, lockable universal wheel; 2, base; 3, third servo electric cylinder; 4, scissor support rod; 5, support plate; 6, winch; 7, lifting steel wire rope; 8, support frame; 9, counterweight; 10, first fixed pulley; 11, second fixed pulley; 12, transverse laser positioning device; 13, longitudinal laser positioning device; 14, tension sensor; 15, high-pressure water pipe; 16, spray head; 17, second servo electric cylinder; 18, electromagnetic suction cup; 19, linear module; 20, second camera; 21, wall climbing robot support frame; 22, hook claw; 23, support frame fixing plate; 24, permanent magnet wheel; 25, wall climbing robot cover plate; 26, slide rail; 27, first camera; 28, pull ring; 29, movable frame support plate; 30, movable frame; 31, second servo electric cylinder support plate; 32, sliding block; 33, spring; 34, first servo electric cylinder support plate; 35, grabbing plate; 36, adjusting spring.
[0088] Figure 8 It is a detection schematic diagram of the wall climbing robot located in the middle of the skip;
[0089] Figure 9 It is a detection schematic diagram of the wall climbing robot located on the left side of the skip;
[0090] Figure 10 It is a wall climbing robot running flowchart;
[0091] Figure 11 It is a liner image gray effect of the invention;
[0092] Figure 12 It is a liner image bilateral filtering effect of the invention;
[0093] Figure 13 It is a liner image global threshold processing result of the invention;
[0094] Figure 14 It is a liner defect feature Canny edge detection effect of the invention;
[0095] Figure 15 It is a liner image straight line segmentation filtering effect and liner defect feature marking screening effect of the invention. DETAILED DESCRIPTION
[0096] The invention will be described in detail below in conjunction with specific embodiments.
[0097] As Figure 1 , Figure 2As shown, the skip lining magnetic replacement device includes an external traction device and a wall climbing robot.
[0098] In this embodiment, the external traction device includes a scissor lift platform and a lifting device.
[0099] The scissor lift platform includes lockable universal wheels 1, a base 2, servo electric cylinders 3, scissor support rods 4 and support plates 5. Four lockable universal wheels 1 are fixed on the base 2 by bolts, the scissor support rods 4 are hinged to the base 2 and the support plates 5, and the two servo electric cylinders 3 are fixed on the base 2 at the base end and hinged to the scissor support rods 4 at the telescopic rod end.
[0100] The lifting device includes a winch 6, a lifting steel wire rope 7, a tension sensor 14 and fixed pulleys 10 and 11. The winch 6 is fixed on the support plate 5 by bolts, the two fixed pulleys 10 and 11 are fixed on the support frame 8 by bolts, the lifting steel wire rope 7 released by the winch 6 is connected to the tension sensor 14 and the wall climbing robot through the fixed pulleys.
[0101] The support mechanism includes a support frame 8, a counterweight 9, a transverse laser positioning device 12, a longitudinal laser positioning device 13 and a support frame fixing plate 23. The support frame 8 is connected to the support plate 5 by the support frame fixing plate 23, the counterweight 9 is fixed on the support frame 8, the weight of which is determined according to the weight of the wall climbing robot and the lining plate, the transverse laser positioning device 12 and the longitudinal laser positioning device 13 are fixed on the support frame 8 by bolts, the transverse laser positioning device 12 judges the transverse position of the wall climbing robot by measuring the distance between the wall climbing robot and the skip wall, and the longitudinal laser positioning device 13 judges the longitudinal position of the wall climbing robot by measuring the distance between the wall climbing robot and the cover plate 25.
[0102] In this embodiment, the wall climbing robot includes a cleaning mechanism, a grabbing plate position adjusting mechanism, a grabbing plate mechanism, an external support mechanism and an image recognition device.
[0103] The cleaning mechanism includes a high-pressure water pipe 15 and a spray head 16. The high-pressure water pipe 15 is connected to the water inlet of the spray head 16 by an external water pump, and the spray head 16 is fixed above the cover plate 25 of the wall climbing robot by bolts.
[0104] The grabbing plate position adjusting mechanism includes a grabbing plate extension and retraction mechanism and a transverse position adjusting mechanism. The grabbing plate extension and retraction mechanism includes two servo electric cylinders and two servo electric cylinder support plates. The two servo electric cylinders are connected to the two servo electric cylinder support plates by bolts, the two servo electric cylinder support plates are connected to the movable frame 30 by bolts, the telescopic rods of the two servo electric cylinders are connected to the grabbing plate 35 by bolts, and the extension and retraction of the grabbing plate 35 are controlled by the extension and retraction of the two servo electric cylinders.
[0105] The transverse position adjusting mechanism of the grabbing plate 35 comprises a linear module 19, a movable frame support plate 29, a screw mechanism support frame, sliding rails 26, a movable frame 30 and sliding blocks 32. The linear module 19 is connected to the movable frame support plate 29 by screw fastening, the movable frame support plate 29 is fixed on the movable frame 30 by screw fastening, the movable frame 30 is connected to the four sliding blocks 32 by screw fastening, and the four sliding rails 26 are fixed on the wall climbing robot support frame 21 by screw fastening. The movable frame 30 controls the transverse movement of the grabbing plate 35 by controlling the transverse movement of the sliding blocks 32 on the linear module 19.
[0106] The grabbing plate mechanism comprises a grabbing plate 35, a hook 22, a spring 33 and electromagnetic suction cups 18. The grabbing plate 35 is connected to two servo electric cylinders 17 through connecting pieces and pin shafts, and four electromagnetic suction cups 35 are uniformly fixed on the grabbing plate 35 by screw fastening. The suction capacity of the electromagnetic suction cups is controlled by the energization state. When the grabbing mechanism is extended to grab the lining plate, the hook 22 is folded inward by the action of the spring 33. When the suction is completed, the grabbing mechanism is retracted, and the hook 22 is reset by the action of the spring 33, thereby preventing the lining plate from sliding off.
[0107] The external support mechanism comprises a wall climbing robot support frame 21, permanent magnet wheels 24, first cameras 27 and a wall climbing robot cover plate 25. Sixteen permanent magnet wheels 24 are connected to the wall climbing robot support frame 21 by wheel seat bolts on both sides, the wall climbing robot cover plate 25 is fixed on the wall climbing robot support frame 21 by screw fastening, and three first cameras 27 are fixed on the three sides (except the side of the grabbing plate) of the wall climbing robot support frame 21 by camera seats. The wall climbing robot support frame 21 is connected by profiled bar members and corners by screw fastening.
[0108] The image recognition device comprises a second camera 20 and a control feedback algorithm. The second camera 20 is fixed in the middle of the grabbing plate 35 by a screw, and the lens is aligned with the position of the lining plate to be photographed and grabbed. The control feedback algorithm mainly uses the second camera 20 to collect image information of the lining plate, analyzes the image information, and feeds back the automatic cleaning work of the bucket interior, the lining plate damage identification work and the lining plate grabbing and replacement work using the spray head 16. After the lining plate cleaning work is performed, the second camera 20 photographs the lining plate to identify whether the lining plate meets the cleanliness requirement. If the lining plate meets the cleanliness requirement, the damage identification is performed, and the positioning and grabbing action of the lining plate that needs to be replaced are performed. If the lining plate does not meet the cleanliness requirement, the lining plate is cleaned again until the cleanliness requirement is met, and then the damage identification is performed. In this way, the influence of the coal ash adhered to the lining plate in the bucket on the precision of the damage identification is avoided, and the safety and effectiveness of the magnetic attraction method in grabbing the lining plate are ensured, and the electromagnetic attraction force of the electromagnet is not affected by the adhesion of the coal ash.
[0109] The permanent magnet wheels are distributed on both sides of the support frame of the wall-climbing robot, and the wheel bodies of the permanent magnet wheels are adsorbed on the inner wall of the skip. When uneven conditions are encountered, the extension of the adjusting spring ensures good adsorption between the permanent magnet wheel body and the skip.
[0110] The operation steps of the skip lining plate magnetic attraction type replacement device are as follows:
[0111] S1 adjust position
[0112] The skip replacement device is manually pushed to the vicinity of the shaft, the height of the lifting platform is adjusted according to the height of the skip entrance, the external traction device drives the wall-climbing robot to enter the skip from the center of the skip entrance, the lateral distance is first adjusted, the position information fed back in real time by the lateral laser positioning device is used to confirm the distance between the wall-climbing robot and the rear wall of the skip, so as to prevent the wall-climbing robot from impacting the rear wall of the skip and causing damage to the wall-climbing robot, and enable the three first cameras on the wall-climbing robot to simultaneously detect the skip wall in three directions; after the lateral position of the wall-climbing robot is determined, the wall-climbing robot is driven to move longitudinally by the external traction device, so as to ensure that the permanent magnet wheels of the wall-climbing robot can be reliably adsorbed on the side wall of the skip; the winch is started, and the wall-climbing robot moves downward under the action of its own gravity, and at the same time, the longitudinal laser positioning device monitors the downward movement distance of the wall-climbing robot,
[0113] S2 internal cleaning
[0114] The wall-climbing robot cleans the lining plate through the cleaning mechanism. The first cleaning is directly performed after the first camera and the second camera reach the corresponding position, after the first cleaning is completed, the lining plate is imaged by the second camera, the lining plate image information is analyzed to determine whether the cleaning requirement is met, if the cleaning requirement is met, damage identification is performed, if the cleaning requirement is not met, secondary cleaning is performed until the cleaning requirement is met.
[0115] S3 shooting
[0116] S31, the wall-climbing robot is adsorbed on the side of the skip, the cameras on the other three non-adsorbed surfaces are turned on, and vertical damage detection is performed on the respective shooting areas from top to bottom;
[0117] S32, after the wall-climbing robot reaches the bottom of the skip, the permanent magnet wheels on the wall-climbing robot are switched to the horizontal direction, the wall-climbing robot is horizontally operated for a distance, the cameras for detecting the front and rear walls are turned off, and then the opposite side of the adsorbed side of the skip is shot from bottom to top;
[0118] S33, after the wall-climbing robot reaches the top end of the skip, the permanent magnet wheels on the wall-climbing robot are switched to the horizontal direction, the wall-climbing robot is horizontally operated for a distance, and then the side of the adsorbed surface of the robot is shot from top to bottom;
[0119] S34, the wall-climbing robot is adsorbed to the opposite side, and steps S31-S33 are repeated to complete the shooting of all the surfaces of the bucket.
[0120] S35, image processing, processing the images shot in step S34:
[0121] The geometric size and position coordinates of the surface defects of the bucket liner are obtained, and if the defect degree reaches the replacement requirement of the liner, a warning is sent to the client to remind the worker to replace the liner in time;
[0122] S4 secondary position adjustment
[0123] After scanning the liner on the entire inner wall of the bucket, the position of the liner to be replaced is determined, the wall-climbing robot is adjusted horizontally and then longitudinally, and the permanent magnet wheel is adsorbed on the side wall of the bucket;
[0124] S5, grabbing and recycling damaged liner
[0125] The grabbing plate position adjusting mechanism is used to drive the grabbing plate to move towards the liner, the position of the grabbing plate is adjusted to align the outline center of the grabbing plate with the outline center of the liner according to the picture collected by the second camera, and the damaged liner is adsorbed by the electromagnetic suction cup on the grabbing plate after alignment; after the above work is completed, the wall-climbing robot is moved out of the bucket and placed in a safe position, the electromagnetic suction cup switch is turned off, and the damaged liner is removed by the worker;
[0126] S6, replacing a new liner
[0127] After removing the damaged liner, the worker installs a new liner on the grabbing plate, the new liner is pre-welded with bolts at the corresponding installation position, the outline center of the new liner is aligned with the outline center of the grabbing plate, the electromagnetic suction cup switch is started, and the electromagnetic suction cup adsorbs the new liner; then the wall-climbing robot is raised to the initial position by the external traction device, the wall-climbing robot is lowered to the liner replacement position, the position of the grabbing plate is adjusted by the grabbing plate position adjusting mechanism, the new liner is aligned with the installation position, and the new liner is connected with the bucket with the assistance of the worker outside the bucket.
[0128] Further, as a preferred technical solution of the present application, the step S2 internal cleaning specifically includes the following sub-steps:
[0129] S21, the liner surface is initially cleaned by the nozzle;
[0130] S22, the second camera is used to collect the images of the liner surface after initial cleaning;
[0131] S23, the liner surface image is converted to grayscale and blurred and denoised to remove small noises in the image;
[0132] S24, detecting whether there is a region with a gray value less than 30 in the image, if there is, considering it as an area which is not cleaned ideally, recording the specific position of the area, and performing secondary cleaning on the area;
[0133] S25, collecting the image of the surface of the ladle liner after secondary cleaning, repeating steps S22-S23, and performing third cleaning on the surface of the liner;
[0134] S26, detecting the surface of the ladle liner again, if there is no region with a gray value less than 30 in the image, entering step S3; if there is still an area which is not cleaned ideally in the image, saving the image with the area, recording the position of the area, and sending the image to the ground control room for manual judgment of the type of the lime and selection of the processing mode.
[0135] Further, as a preferred technical scheme of the present application, the step S3 of detecting and identifying the damaged liner specifically comprises the following sub-steps:
[0136] S31, performing segmentation on the image by using a global threshold segmentation method;
[0137] S32, filtering the interference existing in the image by using a bilateral filtering method, and suppressing the high-frequency noise in the original image;
[0138] S33, refining and pruning the burrs existing in the image by using a morphological operation method, so that the burrs present continuous and smooth edges;
[0139] S34, performing edge detection on the result image obtained in step S33 by using a Canny edge detection operator;
[0140] S35, fitting an inscribed rectangle to the defect pattern on a single liner, and then filtering the pseudo marks in the image through the area of the inscribed rectangle, specifically comprising the following sub-steps:
[0141] S351, fitting a defect contour to the segmented single liner, and adopting an inscribed rectangle R a to the detected defect contour C a , fitting an inscribed rectangle R a The definition of each parameter is as follows: first, determining a point P1(x1, y1) with the maximum vertical coordinate on the contour C a to be fitted, and drawing a horizontal line L1 through the point; then, determining a point P2(x2, y2) with the maximum horizontal coordinate on the contour C a to be fitted, and drawing a horizontal line L2 through the point; next, calculating the intersection point P3 of L1 and L2 as a vertex of the inscribed rectangle; finally, taking the difference between the vertical coordinate of the point with the minimum vertical coordinate on the contour and the vertical coordinate of P1 as the height of the inscribed rectangle R athe first side length of the circumscribed rectangle R a the other side length of the circumscribed rectangle R a ;
[0142] S352, the area S a of the circumscribed rectangle R a is calculated a ; a The ratio k of the area S of the circumscribed rectangle R to the total area S of the image is calculated:
[0143]
[0144] A reasonable value interval Z is set according to the proportion of the liner defects in the image, and all rectangles whose k values do not belong to the interval are considered as false markers and filtered out;
[0145] S36, the actual size of the surface defect area of the bucket liner is calculated, and a quantitative corresponding model of defects of different sizes and their defect degrees is established, and the wall climbing robot performs different measures according to the different defect degrees, which specifically includes the following steps:
[0146] S361, the perspective transformation correction is performed on the liner image with the edge vertex of the liner as the reference;
[0147] S362, the Hough transform algorithm is used to detect straight lines in the image, and the edges of the single bucket liner are determined in combination with the geometric structure of the surface of the bucket liner;
[0148] S363, the actual geometric size of the single liner and the numerical ratio of the number of pixels occupied by the single liner in the image are used as the parameter scale to calculate the actual size of the surface defects of the liner in the image;
[0149] S364, the defect condition is comprehensively judged according to the size of the dimensional defect and the average gray value at the defect:
[0150] The defect with the maximum direction size sizem∈[0,20]mm and the average gray value agv∈[100,150] is a small defect, which has little effect on the service effect of the bucket liner, and is not processed temporarily, the defect information is recorded, and the wall climbing robot continues to detect;
[0151] The defect with the maximum direction size sizem∈[20,50]mm and the average gray value agv∈[70,100] is a medium-sized defect, which has a certain influence on the service effect of the bucket liner, and a warning should be sent to the ground working chamber to remind the worker to consider replacing or repairing the liner, and the defect information is recorded, and the wall climbing robot continues to detect;
[0152] The defect whose maximum direction size sizem is greater than 50 mm and whose average gray value agv is less than 70 is identified as a large defect, a warning is sent to the ground working room to remind the staff to replace the liner in time, and the defect information is recorded, and the wall climbing robot no longer performs subsequent detection of this block of liner.
[0153] It should be understood that, for those skilled in the art, modifications or changes can be made according to the above description, and all the modifications and changes shall belong to the protection scope of the appended claims of the present application.
Claims
1. A magnetic replacement device for a skip liner, characterized in that include: A wall-climbing robot, which adheres to the inner wall of any side of a basket, includes: a support frame, permanent magnet wheels, and a cover plate. The permanent magnet wheels are connected to both sides of the support frame, and the cover plate is fixed to the top of the support frame. A gripping plate, with a gripping plate position adjustment mechanism, is located on the side of the support frame near the liner plate, and is used to replace damaged liner plates. A cleaning mechanism, mounted on a support frame, is used to clean the surface of the lining plate; An image acquisition unit, mounted on the gripping plate, is used to acquire images of the liner surface; it includes three first cameras fixed to the three sides of the support frame where the gripping plate is not mounted, and one second camera fixed to the gripping plate. An external traction device, installed outside the basket, is connected to the cover plate on top of the wall-climbing robot via a traction rope, and is used to send the wall-climbing robot into or out of the basket, including: Lifting platform and lifting device, among which, The lifting platform is located at the very bottom of the entire traction device; The lifting device includes a support frame, a winch, a lifting wire rope, a first fixed pulley, and a second fixed pulley, wherein the support frame and the winch are fixed on the lifting platform; The support frame includes a vertical support rod and a horizontal cantilever rod. A first fixed pulley is installed at the top of the vertical support rod, and a second fixed pulley is installed at the cantilever end of the horizontal cantilever rod. The lifting wire rope released by the winch is connected to the hanging ring at the top of the wall-climbing robot through the first and second fixed pulleys. A lateral laser positioning device is installed on the cantilever end of the lateral cantilever rod to determine the lateral position of the wall-climbing robot by measuring the distance between it and the bucket wall. A longitudinal laser positioning device is installed on the cantilever end of the transverse cantilever rod to determine the longitudinal position of the wall-climbing robot by measuring the distance to the cover plate of the wall-climbing robot. The gripper plate position adjustment mechanism includes: The components include a lead screw motor, a movable frame support plate, a lead screw, a nut block, a lead screw mechanism support frame, a slide rail, a movable frame, a first servo electric cylinder, and a second servo electric cylinder. The output shaft of the lead screw motor is connected to the lead screw via a coupling. The lead screw and the nut block are connected by threads. The upper surface of the nut block is fastened to the movable frame support plate. The lead screw motor is connected to the lead screw mechanism support frame. The lead screw is connected to the lead screw mechanism support frame via bearings at both ends. The lead screw mechanism support frame is fixed to the movable frame support plate. The movable frame support plate is fixed to the movable frame. The movable frame is slidably connected to the slide rail via a slider. The slide rail is fixed to the support frame. The first servo electric cylinder and the second servo electric cylinder are respectively fixedly connected to the movable frame, and the telescopic rods of the first servo electric cylinder and the second servo electric cylinder are respectively connected to the same side of the gripping plate. The gripping plate includes a gripping plate body, hooks, compression springs, and an electromagnetic chuck. The gripping plate body is connected to the gripping plate position adjustment mechanism via a connector and a pin. The electromagnetic chuck is fixed to the gripping plate, and its adsorption capacity is controlled by the energized state. The hooks include four, and the four hooks are respectively set at the four corners of the gripping plate through hinges; The compression spring is arranged between the hook claw and the grabbing plate, when the grabbing plate body approaches the lining plate, the upper and lower ends of the lining plate push the hook claw, the hook claw overcomes the elastic force of the compression spring and rotates backward around the hinge part until the upper and lower ends of the lining plate are clamped, meanwhile, the electromagnetic chuck can contact the lining plate.
2. The skip liner magnetic changeover device of claim 1, wherein, The support frame is formed by profiled rod members and corners connected by bolts.
3. The skip liner magnetic changeover device of claim 1, wherein, The lifting platform is a scissor lifting platform, comprising lockable casters, a base, third servo electric cylinders, a scissor support rod and a support plate, wherein the lockable casters are fixed on the base by bolts, one end of the scissor support rod is hinged to the base, the other end is hinged to the support plate, the bottom ends of the two third servo electric cylinders are fixed on the base, and the telescopic rod end is hinged to the scissor support rod.
4. The magnetic skip lining replacement device according to claim 1, characterized in that, The top of the cover plate is provided with a lifting ring for external traction device to pull.
5. The magnetic skip lining replacement device according to claim 1, characterized in that, The first camera is fixedly connected with the support frame through a camera support; and the second camera is fixed in the middle of the grabbing plate through a camera support.
6. A method of operating a magnetic bucket liner changing device according to any one of claims 1 to 5, characterized in that, The method comprises the following steps: S1 adjusting position The bucket changing device is manually pushed to the vicinity of the shaft, the height of the lifting platform is adjusted according to the height of the bucket entrance, the external traction device drives the wall-climbing robot to enter the bucket from the center of the bucket entrance, the horizontal distance is first adjusted, the position information fed back in real time by the horizontal laser positioning device is used to confirm the distance between the wall-climbing robot and the rear wall of the bucket, so as to prevent the wall-climbing robot from colliding with the rear wall of the bucket and causing damage to the wall-climbing robot, and the three first cameras on the wall-climbing robot can simultaneously detect the walls of the bucket in three directions; after the horizontal position of the wall-climbing robot is determined, the wall-climbing robot is longitudinally moved by the external traction device, so as to ensure that the permanent magnet wheel of the wall-climbing robot can be reliably adsorbed on the side wall of the bucket; the winch is started, the wall-climbing robot is moved downward under the action of its own gravity, and at the same time, the longitudinal laser positioning device monitors the moving distance of the wall-climbing robot, S2 internal cleaning The wall-climbing robot cleans the lining plate through the cleaning mechanism, after the first cleaning is completed, the second camera is used to collect images of the lining plate, the cleaning requirement is determined by analyzing the image information of the lining plate, if the cleaning requirement is met, step S3 is entered, if the cleaning requirement is not met, secondary cleaning is performed until the cleaning requirement is met; S31, the wall-climbing robot is adsorbed on the side of the bucket, the cameras on the other three non-adsorbed surfaces are opened, and vertical damage detection is performed on the respective shooting areas from top to bottom; S32, after the wall-climbing robot reaches the bottom of the bucket, the permanent magnet wheel on the wall-climbing robot is switched to the horizontal direction, the wall-climbing robot is horizontally moved for a distance, the cameras for detecting the front and rear walls are closed, and then the opposite side of the adsorbed side of the bucket is shot from bottom to top; S33, after the wall-climbing robot reaches the top end of the bucket, the permanent magnet wheel on the wall-climbing robot is switched to the horizontal direction, the wall-climbing robot is horizontally moved for a distance, and then the side of the adsorbed surface of the robot is shot from top to bottom again; S34, the wall-climbing robot is adsorbed to the opposite side, and steps S31-S33 are repeated to complete the shooting of all the surfaces of the bucket; S35, image processing, processing the images shot in step S34: The geometric size and position coordinates of the surface defects of the bucket liner are obtained, and if the defect degree reaches the replacement requirement of the liner, a warning is sent to the client to remind the worker to replace the liner in time; S4 secondary position adjustment After the scanning of the liner on the inner wall of the whole bucket is completed, the position of the liner to be replaced is determined, the wall-climbing robot is first adjusted horizontally and then adjusted vertically, so that the permanent magnet wheel is adsorbed on the side wall of the bucket; S5, grabbing and recycling damaged liner Through the grabbing plate position adjusting mechanism, the grabbing plate is driven to move towards the liner, the position of the grabbing plate is adjusted to make the outline center of the grabbing plate aligned with the outline center of the liner, and at the same time, the worker assists in dismounting the bolts between the liner and the bucket. After alignment, the damaged liner is adsorbed by the electromagnetic suction cup on the grabbing plate; after the above work is completed, the wall-climbing robot is moved out of the bucket and placed in a safe position, the electromagnetic suction cup switch is turned off, and the damaged liner is taken off by the worker; S6, replacing a new liner After the damaged liner is removed, the worker installs a new liner on the grabbing plate, the new liner is pre-welded with bolts at the corresponding installation position, so that the outline center of the new liner is aligned with the outline center of the grabbing plate, the electromagnetic suction cup switch is started, and the electromagnetic suction cup adsorbs the new liner. Then, the external traction device is used to lift the wall-climbing robot to the initial position, the wall-climbing robot is lowered to the position for replacing the liner, the position of the grabbing plate is adjusted by the grabbing plate position adjusting mechanism, so that the new liner is aligned with the installation position, and the worker assists in connecting the new liner with the bucket outside the bucket.
7. The working method of the bucket liner magnetic replacement device according to claim 6, wherein The step S2 specifically comprises the following sub-steps: S21, the surface of the liner is initially cleaned by the cleaning mechanism; S22, the second camera is used to collect the image of the surface of the bucket liner after initial cleaning; S23, the image of the surface of the liner is converted to grayscale, and the image is blurred and denoised to remove small noises in the image; S24, whether there is an area with a gray value less than 30 in the image is detected, if there is, it is considered that the area is not cleaned ideally, the specific position of the area is recorded, and the area is cleaned again; S25, the image of the surface of the bucket liner after secondary cleaning is collected, and steps S22-S23 are repeated to clean the surface of the liner for the third time; S26, the surface of the bucket liner is detected again, if there is no area with a gray value less than 30 in the image, step S3 is entered; if there is still an area not cleaned ideally in the image, the image with the area is saved, the position of the area is recorded, and the image is sent to the ground control room for manual judgment of the type of mud and selection of the treatment method.
8. The method of operating a magnetic skip liner changing device according to claim 7, wherein, The step S3 specifically comprises the following sub-steps: S31, the image is segmented by using a global threshold segmentation method; S32, the interference existing in the image is filtered by using a bilateral filtering method to suppress high-frequency noise in the original image; S33, the method of morphological operation is adopted to refine and trim the burrs existing in the image, so that the image presents continuous and smooth edges; S34, the Canny edge detection operator is adopted to perform edge detection on the result image obtained in step S33; S35, the rectangle fitting is performed on the defect pattern on the single block lining plate, and then the pseudo marks in the image are filtered through the area of the circumscribed rectangle, which specifically includes the following sub-steps: S351, fitting defect contour to the segmented single block lining, adopting circumscribed rectangle to the detected defect contour , fitting, circumscribed rectangle The definition of each parameter is as follows: first, determine the contour to be fitted The point with the maximum vertical coordinate , and draw a horizontal line through this point; then, determine the contour to be fitted The point with the maximum horizontal coordinate , and draw a horizontal line through this point; next, calculate the intersection of the horizontal lines as a vertex of the circumscribed rectangle; finally, take the difference between the vertical coordinate of the point with the minimum vertical coordinate on the contour and the vertical coordinate of the intersection as the first side length of the circumscribed rectangle , and take the difference between the horizontal coordinate of the point with the minimum horizontal coordinate on the contour and the horizontal coordinate of the intersection as the other side length of the circumscribed rectangle , thereby determining the circumscribed rectangle ; S352, the area of the circumscribed rectangle S352, the area of the circumscribed rectangle S352, the area of the circumscribed rectangle S352, the area of the circumscribed rectangle S352, the area of the circumscribed rectangle S352, the area of the circumscribed rectangle S352, the area of the circumscribed rectangle ; A reasonable value interval is set according to the proportion of the liner defects in the map and the rectangles whose values do not belong to the interval are considered as false marks and filtered out. S36, the actual size of the surface defect area of the bucket lining plate is calculated, a quantitative corresponding model of different sizes of defects and their defect degrees is established, and the wall climbing robot performs different measures according to the different defect degrees, which specifically includes the following steps: S361, the perspective transformation correction is performed on the lining plate image with reference to the edge vertex of the lining plate; S362, the Hough transformation algorithm is adopted to detect the straight lines in the image, and the edges of the single block bucket lining plate are determined in combination with the geometric structure of the surface of the bucket lining plate; S363, the actual size of the lining plate surface defect in the image is calculated by taking the numerical ratio of the actual geometric size of the single block lining plate to the number of pixels occupied by the single block lining plate in the image as a parameter scale; S364, the defect condition is comprehensively judged according to the size of the size defect and the average gray value at the defect: Maximum size of defect Average gray value The defect is a small defect, and the influence of such defect on the service effect of the ladle liner is small. The defect information is recorded, and the wall climbing robot continues to detect. Maximum size of the defect in the direction Average gray value The defect is a medium-sized defect, which has a certain impact on the service effect of the bucket lining plate. A warning should be sent to the ground working room to remind the staff to consider replacing or repairing the lining plate, and the defect information should be recorded. The wall climbing robot continues to detect. Maximum size of the defect in the direction Average gray value The defect is identified as a large defect, a warning is sent to the ground studio, reminding the staff to replace the liner in time, recording the defect information, and the wall climbing robot no longer performs subsequent detection of this block of liner.
Citation Information
Patent Citations
A type of large-tonnage coal dropping buffer skip in mines
CN109160410B
Data driving-based wear prediction and lining plate optimization layout of lifting container lining plate
CN112949002A
Pull-type metal wall surface working wall-climbing robot
CN106240664A
Ferromagnetism wall climbing working robot
CN107127770A