A mine conveyor belt pre-tightening device and pre-tightening method based on visual detection

Abstract

The present invention discloses a pre-tightening device and pre-tightening method for a mine conveyor belt based on visual detection, comprising a pre-tightening adjustment part, symmetrically arranged on both sides of a non-driving end roller, comprising an adjustment base connected to the roller shaft end, a guide structure fixed on a truss, and a linear drive assembly; a visual detection unit, comprising a protective shell installed on a protective frame at an adjustable angle, an image collector and a fill light arranged in the protective shell, and a self-cleaning mechanism; the image collector processes the collected image, calculates the wear area ratio of the conveyor belt surface, and then determines that the conveyor belt needs to adjust the pre-tightening force through a set wear threshold; the image collector transmits a signal to the pre-tightening adjustment part, and tightens or releases the tension of the conveyor belt by starting the linear drive assembly to drive the adjustment base to support the roller. The present invention uses visual detection technology to accurately monitor the state of the conveyor belt, and accurately adjusts the pre-tightening force through the mechanical structure in the pre-tightening adjustment part, thereby overcoming the problem of inaccurate adjustment of traditional purely mechanical pre-tightening devices.

Description

Technical Field

[0001] The present invention belongs to the technical field of conveyor belt pre-tightening detection, and in particular relates to a mine conveyor belt pre-tightening device and a pre-tightening method based on visual detection. Background Art

[0002] In mining transportation operations, the pre-tightening of the conveyor belt is crucial for its normal operation. Traditional conveyor belt pre-tightening devices are mostly based on simple mechanical structures and manual regular inspection and adjustment, which lack accuracy and timeliness. For example, when adjusting the pre-tightening force solely by relying on a spiral tensioning device, it is difficult to accurately judge the actual tightness of the conveyor belt, and it is easy for insufficient pre-tightening force to cause the conveyor belt to slip, or excessive pre-tightening force to cause excessive wear and energy waste of the conveyor belt. Moreover, the traditional mechanical structure cannot intuitively obtain the dynamic changes of the conveyor belt during operation, and cannot adaptively adjust the pre-tightening force in time according to the wear, elongation and other conditions of the conveyor belt. Summary of the invention

[0003] The object of the present invention is to provide a mine conveyor belt pretensioning device and a pretensioning method based on visual detection which can adaptively adjust the conveyor belt pretensioning force.

[0004] The mining conveyor belt pre-tensioning device based on visual detection provided by the present invention comprises a conveying main structure, which comprises a truss, rollers arranged at both ends of the truss, a conveyor belt sleeved between the two rollers, a driving mechanism for driving the rollers on one side, rollers arranged in the truss, and a detection platform and a protective frame arranged on the side of the truss. It also comprises: a pre-tensioning adjustment part, which is symmetrically arranged on both sides of the non-driving end roller, comprising an adjustment base connected to the roller shaft end, a guide structure fixed on the truss, and a linear drive component, wherein the adjustment base is slidably arranged on the guide structure, and the output end of the linear drive component is transmission-connected with the adjustment base to drive it to move linearly along the guide structure; a visual detection unit, comprising a protective shell installed on the protective frame with an adjustable angle, an image collector and a fill light arranged in the protective shell, and a self-cleaning mechanism, wherein the protective shell is locked by an angle locking mechanism To achieve the fixation of the image acquisition angle, the self-cleaning mechanism includes a rotating brush assembly in contact with the surface of the protective shell, a cleaning cavity accommodating the rotating brush assembly, and a power assembly driving the rotating brush assembly to move. The inner wall of the cleaning cavity is provided with an elastic cleaning strip that generates contact vibration with the rotating brush assembly; the image collector performs grayscale processing on the collected image, uses an edge detection algorithm to detect the edge of the conveyor belt surface, performs morphological processing on the detected edge, and calculates the wear area ratio of the conveyor belt surface, and then determines whether the conveyor belt adjusts the preload force according to the set wear threshold; according to the mechanical principles and empirical formulas of the mine conveyor belt, a preload force calculation model is established with the elastic modulus, cross-sectional area, and friction coefficient of the conveyor belt; the image collector transmits a signal to the preload adjustment unit, and tightens or releases the tension of the conveyor belt by starting the linear drive component to drive the adjustment base support roller.

[0005] In one embodiment of the above device, the top of the adjustment base is fixed with a bearing seat by fasteners to support the roller shaft end, and a clamping plate is welded at the bottom to slide and engage with the guide structure; the guide structure is composed of a pair of welded inclined plates and a slide groove is opened on the top edge, and the linear drive assembly is a servo motor, which drives the lead screw, which passes through the guide structure and cooperates with the slider thread, and the top of the slider is fixedly connected to the adjustment base and slides in the slide groove; the two ends of the guide structure are fixed to the truss bolts through triangular blocks, and the servo motor bolts are fixed to the truss end cover plate.

[0006] In one embodiment of the above device, the protective shell is connected to the protective frame via a support block hinged at the bottom, an inverted U-shaped clamping plate is provided at the bottom of the support block and is fixed to the protective frame by bolts; a pre-tightening screw and a threaded nut are provided at the top of the support block, a straight handle is fixed to the nut by bolts, and angle locking is achieved by rotating the nut to squeeze the support block.

[0007] In one embodiment of the above device, the rotating brush assembly is in a cross structure and is embedded with nylon bristles, and its central axis is connected to the power assembly via a pulley and a transmission belt; the cleaning chamber is composed of two semicircular plates connected as one body, wherein an elastic cleaning strip made of PP material is embedded in the inner wall of one semicircular plate, and a fixed sleeve is provided at the bottom of the other semicircular plate to be sleeved on the central axis of the rotating brush assembly; the power assembly is an electric motor, which is mounted on a flat plate on the bottom side wall of the cleaning chamber via fasteners.

[0008] In one embodiment of the above device, the side surface of the protective shell is made of transparent material, the image collector is a high-speed camera, and the fill light is a lighting lamp; the high-speed camera and the lighting lamp are arranged in the protective shell and aimed at the surface of the conveyor belt.

[0009] In one embodiment of the above device, the central axis of the pulley is fixed to the bottom of the protective shell by a sleeve bolt, and the bristles vibrate when they come into contact with the elastic cleaning strip.

[0010] A pre-tightening method using the above device, the specific steps are as follows:

[0011] S1. Collecting the conveyor belt surface image and preprocessing the collected image;

[0012] The image acquisition device is based on the frame rate Collect images of the conveyor belt surface, and the frame rate depends on the conveyor belt running speed Dynamic settings; fill light provides uniform lighting to ensure image brightness In the appropriate range, ;

[0013] Convert color images to grayscale images using weighted averaging , the formula is: Where: is the red channel pixel value; is the green channel pixel value; is the blue channel pixel value;

[0014] S2. performing edge detection and morphological processing on the preprocessed image obtained in step S1;

[0015] Performing Gaussian filtering on the grayscale image obtained in step S1 to remove noise;

[0016] Use the Sobel operator to calculate the pixel in the image Gradient in direction and Gradient in direction , gradient amplitude The formula is as follows: Then use 3×3 square structure elements for the edge image Perform expansion and erosion operations;

[0017] The expansion operation formula is as follows: The corrosion operation formula is as follows: Where: To expand the operating range, is the corrosion operating range, is the image brightness, represents the dilation operation, represents the erosion operation;

[0018] S3. Performing status analysis on the processed conveyor belt image obtained in step S2;

[0019] Calculate the wear area ratio of the conveyor belt surface , perform pixel statistics on the processed image; calculated by the following formula: Where: is the total number of pixels of the conveyor belt surface image, The number of pixels in the worn area; set the threshold , determine that the conveyor belt needs to adjust the preload;

[0020] Assume the conveyor belt image width is , center reference line Location is ; By analyzing the pixel position of the conveyor belt edge in the image , calculate the deviation : The threshold is set here as If the deviation , it is judged as deviation;

[0021] S4. Calculate the preload force using the data obtained in step S3; the specific steps are as follows:

[0022] The wear condition of the conveyor belt surface is collected from the image collector, and the wear area ratio is , the deviation amount of the deviation situation is And pre-set conveyor belt operating parameters, such as conveying material weight , running speed , Ambient temperature ;

[0023] Based on elastic modulus E, cross-sectional area S, friction coefficient Equal parameters to establish preload The calculation model is as follows: Where: are conventional coefficients obtained from experiments. ;

[0024] S5. Detect the data obtained in step S3 and step S4, and dynamically adjust the preload force;

[0025] When the conveyor belt surface is detected to have wear changes, or the conveyor belt is found to have deviation, it is detected or When , an adjustment signal is sent to the preload adjustment unit;

[0026] Set the optimal preload force. Compared with the optimal preload, if If it is greater than the optimal preload, reduce the preload; If the preload is less than the optimum preload, increase the preload. If it is equal to the optimum preload force, no adjustment is required;

[0027] When the preload force increases: the linear drive assembly drives the screw to rotate in the positive direction, and the slider moves along the guide structure slot, driving the adjustment base away from the drive end roller to tighten the conveyor belt;

[0028] When the preload force decreases: the linear drive assembly rotates in the opposite direction, the slider retreats, the base is adjusted to push the roller closer to the drive end, and the conveyor belt is loosened;

[0029] S6. The preload adjustment unit receives the conveyor belt status data updated by the visual detection unit in real time, and executes steps S1-S5 in a loop until the preload force reaches the optimal value.

[0030] The beneficial effects of the present invention are as follows:

[0031] 1. The state of the conveyor belt is accurately monitored by visual inspection technology, and then the preload force is accurately adjusted through the mechanical structure in the preload adjustment unit, which overcomes the problem of inaccurate adjustment of traditional purely mechanical preload devices; the preload force is automatically adjusted according to the real-time operating conditions of the conveyor belt, such as wear and deviation, without frequent manual intervention, which improves the stability and reliability of the conveyor belt operation;

[0032] 2. The self-cleaning mechanism is set up to sweep away the dust on the side of the protective shell through the rotating brush assembly, so that the image collector can keep the camera clear within the range of the fill light; the mechanical structure design of the whole device is simple, easy to manufacture and maintain, and has high durability in the harsh environment of the mine, reducing the maintenance cost of the visual inspection unit;

[0033] 3. By setting a number of elastic cleaning strips on the inner wall of the cleaning cavity, when the rotating brush assembly is turned into the dustproof box, the flexible cleaning strips are moved to generate vibrations to scatter the dust on the rotating brush assembly, which is beneficial to maintaining its cleanliness, thereby cleaning the side of the protective shell. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the overall assembly structure of the present invention;

[0035] Figure 2 For the present invention Figure 1 The main view of

[0036] Figure 3 It is a schematic diagram of the assembly structure of the pre-tightening adjustment part and the roller of the present invention;

[0037] Figure 4 This is a disassembled diagram of the preload adjustment part and the roller assembly of the present invention;

[0038] Figure 5 This is a schematic diagram of the assembly structure of the preload adjustment part of the present invention;

[0039] Figure 6 This is one of the schematic diagrams of the assembly structure of the visual inspection unit of the present invention;

[0040] Figure 7 This is a diagram of the protective housing assembly and disassembly of the present invention;

[0041] Figure 8 This is the second schematic diagram of the assembly structure of the visual inspection unit of the present invention;

[0042] Fig. 9 This is a schematic diagram of the assembly structure of the self-cleaning mechanism of the present invention;

[0043] Fig.10 It is a disassembled diagram of the rotary brush assembly of the present invention;

[0044] Fig.11 It is a schematic diagram of the cleaning chamber structure of the present invention;

[0045] The meaning of each number in the figure is:

[0046] 100, truss; 110, roller; 111, conveyor belt; 112, hopper; 113, bearing seat; 120, driving mechanism; 130, roller; 140, protective frame;

[0047] 200, preload adjustment part; 210, adjustment base; 211, clamping plate; 220, guide structure; 221, slide groove; 230, linear drive assembly; 231, screw rod; 232, slider;

[0048] 300, visual detection unit; 310, image acquisition device; 320, fill light; 330, protective housing; 331, support block; 332, card board;

[0049] 340, self-cleaning mechanism; 341, rotating brush assembly; 3411, bristles; 3412, support sleeve; 3413, pulley; 3414, transmission belt; 342, cleaning chamber; 3421, elastic cleaning strip; 3422, fixed sleeve; 3423, fixed plate; 343, power assembly; 350, pre-tightening screw. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the relevant technical solutions. Obviously, the described embodiments are only part of the embodiments, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] like Figure 1 and Figure 2 As shown, the mine conveyor belt pretensioning device based on visual detection disclosed in this embodiment includes a pretensioning adjustment part 200 and a visual detection unit 300 arranged on the conveying main structure.

[0052] The main structure of the conveyor includes an inclined truss 100, with rollers 110 installed at both ends of the truss, and a conveyor belt 111 is sleeved between the two rollers. A driving mechanism 120 is coaxially connected to one side of the roller to drive the conveyor belt to run and transport materials. A plurality of rollers 130 are arranged inside the truss to support the conveyor belt. The driving mechanism is preferably a motor group.

[0053] like Figure 3 As shown, a hopper 112 is fixed at the top of the lower end of the truss 100 for feeding materials to the conveyor belt 111. An inspection table is provided on one side of the truss and a protective frame 140 is fixedly connected to the side of the inspection table for installing a plurality of visual inspection units 300 for triggering the operation of the preload adjustment unit 200. The protective frame is preferably a railing.

[0054] like Figure 4 and Figure 5 As shown, the preload adjustment unit 200 is located on both sides of the non-driving end roller 110 and is used to adjust the position of the corresponding roller. The preload adjustment unit includes an adjustment base 210, a guide structure 220 and a linear drive assembly 230.

[0055] The two ends of the central axis of the drum 110 are sleeved with bearing seats 113; the adjustment base is preferably an adjustment plate, the top of the adjustment plate is fixed with bolts to the bearing seat to support the drum shaft end, and the bottom of the adjustment plate is welded with a clamping plate 211 that is slidably engaged with the guide structure 220.

[0056] The guide structure 220 is formed by welding a pair of inclined plates, and both ends of the guide structure 220 are fixed to the truss 100 by bolts through triangular blocks.

[0057] The linear drive assembly 230 is preferably a servo motor, which is bolted to the end cover of the truss 100; the servo motor drives the screw 231 to rotate, the screw passes through the guide structure 220 and is threadedly matched with the slider 232, and the top of the slider is fixed to the adjustment base 210; the servo motor drives the screw to rotate, thereby driving the slider and the adjustment base to move, thereby realizing linear adjustment of the roller position.

[0058] A sliding groove 221 is formed on the top edge of the guide structure 220 , and the slider 232 is slidably engaged with the sliding groove, so that the slider does not interfere with the sliding and is restricted from rotating with the screw rod 231 .

[0059] like Figure 6 and Figure 7 As shown, the visual inspection unit 300 is installed on the protective frame 140 , and includes a protective shell 330 , an image collector 310 , a fill light 320 and a self-cleaning mechanism 340 .

[0060] The side of the protective shell 330 is made of transparent material, and an image collector 310 and a light filler 320 are arranged in the box to align with the surface of the conveyor belt. The protective shell is connected to the protective frame 140 through a support block 331 hinged at the middle of the bottom. The image collector is preferably a high-speed camera, and the light filler is preferably a lighting lamp.

[0061] An inverted U-shaped clamping plate 332 is provided at the bottom end of the support block 331 to be clamped with the protection frame 140 , and the clamping plate and the protection frame are fixed by through-connecting bolts.

[0062] The top center hole of the support block 331 is plugged with a pre-tightening screw 350, and nuts are threadedly connected at both ends of the pre-tightening screw; the rotating nut squeezes the support block to achieve angle locking, so that the protective housing 330 rotates backward and remains stable, ensuring that the image collector is aligned with the conveyor belt for shooting. A straight handle is fixedly connected to the nut by bolts to grasp the rotating nut to squeeze the support block.

[0063] like Figure 8 and Fig. 9 As shown, the self-cleaning mechanism 340 includes a rotating brush assembly 341, a cleaning cavity 342 and a power assembly 343. The rotating brush assembly sweeps away dust on the side of the protective housing 330, so that the image collector 310 can maintain clear photography within the range of the fill light 320.

[0064] like Fig.10 As shown, the rotating brush assembly 341 is in a cross structure, and a plurality of bristles 3411 made of nylon plastic are embedded on one side close to the protective shell 330; the central axis of the rotating brush assembly is bolted to the bottom of the protective shell through a supporting sleeve 3412 to support the stable rotation of the rotating brush assembly.

[0065] A pulley 3413 is sleeved on the end of the central axis of the rotating brush assembly 341. The power assembly 343 drives the pulley and the transmission belt 3414 to circulate, thereby driving the rotating brush assembly to rotate and continuously clean the dust on the side of the protective shell.

[0066] like Fig.11 As shown, the cleaning chamber 342 is composed of two semicircular plates connected in one piece, wherein the top of the semicircular plate is fixedly connected to the two side walls of the protective shell through a fixing plate 3423, and a fixing sleeve 3422 is provided at the bottom, and the central axis of the pulley 3413 located below is connected and rotated with the fixing sleeve 3422. The power component is preferably an electric motor 343, which is installed on the flat plate of the bottom side wall of the cleaning chamber 342 by bolts;

[0067] A thin elastic cleaning strip 3421 made of PP material is embedded in the inner wall of a semicircular plate of the cleaning cavity 342. When the rotating brush assembly 341 rotates into the cleaning cavity, the bristles move the elastic cleaning strip to generate vibration, causing dust to fall off, helping the rotating brush assembly to maintain cleanliness, thereby cleaning the side of the protective shell 330.

[0068] This embodiment also discloses a mine conveyor belt pre-tightening method based on visual detection, which realizes adaptive pre-tightening force adjustment through the following steps:

[0069] S1. Collect the conveyor belt surface image and pre-process the collected image; the specific steps are as follows:

[0070] A. Image acquisition parameter settings: The image collector uses a frame rate Collect images of the conveyor belt surface, and the frame rate depends on the conveyor belt running speed Dynamic settings; for example, when When ;

[0071] The fill light provides uniform lighting to ensure image brightness In the appropriate range, ;

[0072] B. Grayscale processing: Use weighted average method to convert color image into grayscale image , the formula is: Where: is the red channel pixel value; is the green channel pixel value; is the blue channel pixel value;

[0073] S2. Perform edge detection and morphological processing on the preprocessed image obtained in step S1; the specific steps are as follows:

[0074] a. Noise removal: Perform Gaussian filtering on the grayscale image obtained in step S1 to remove noise; set the Gaussian kernel size to , the standard deviation is , through convolution operation, weighted average is performed on each pixel in the image to reduce the impact of high-frequency noise;

[0075] b. Edge detection: Calculate the gradient magnitude and direction of the pixels in the image obtained in step a;

[0076] The Sobel operator is used to calculate the pixel Gradient in direction and Gradient in direction , gradient amplitude The formula is as follows: These gradients are obtained by convolving the image using a Sobel filter. In a convolution operation, a kernel is slid over the image, usually starting from the upper left corner and moving right and downward until the entire image is covered. Each time it moves, the area of ​​the image covered by the kernel is multiplied by the corresponding elements of the kernel, and then these products are added together to get a new pixel value. This process is repeated for every pixel in the image.

[0077] For edge detection, commonly used convolution kernels include Sobel operator, Prewitt operator and Roberts operator, etc. These operators detect edges by calculating the gradient of pixels in the image. For example, the Sobel operator contains two 3x3 convolution kernels, one for calculating the gradient in the x direction and the other for calculating the gradient in the y direction.

[0078] These two gradients can then be used to calculate the gradient magnitude and direction, and subsequently determine the location and strength of the edge;

[0079] c. Morphological processing: dilation and erosion operations are performed on the edge image, using a 3×3 square structure element ; The structural element defines the local shape and size of the operation and can be designed and adjusted according to specific needs; The expansion operation formula is as follows: The corrosion operation formula is as follows: Where: To expand the operating range, is the corrosion operating range, is the image brightness, represents the dilation operation, represents the erosion operation;

[0080] The dilation operation increases the size of objects in the image, and is usually used to fill small holes or connect adjacent objects. The dilation operation is defined as sliding the structural element over the image and calculating the maximum value in the area covered by the structural element. For binary images, this is usually the maximum value of 1 or True. The erosion operation reduces the size of objects in the image, and is usually used to remove small objects or split connected objects. The erosion operation is defined as sliding the structural element over the image and calculating the minimum value in the area covered by the structural element. For binary images, this is usually the minimum value of 0 or False.

[0081] That is, the dilation operation increases the size of the object in the image by calculating the maximum value in the area covered by the structural element; the erosion operation reduces the size of the object in the image by calculating the minimum value in the area covered by the structural element;

[0082] S3. Performing status analysis on the processed conveyor belt image obtained in step S2; the specific steps are as follows:

[0083] 1. Wear area calculation: Calculate the wear area ratio of the conveyor belt surface , by counting the pixels of the processed image; it is calculated by the following formula: Where: is the total number of pixels of the conveyor belt surface image, The number of pixels in the worn area; set the threshold , determine that the conveyor belt needs to adjust the preload;

[0084] 2. Deviation detection: Assume the conveyor belt image width is , center reference line Location is ; By analyzing the pixel position of the conveyor belt edge in the image , calculate the deviation : The threshold is set here as If the deviation , it is judged as deviation;

[0085] S4. Calculate the preload force using the data obtained in step S3; the specific steps are as follows:

[0086] The wear condition of the conveyor belt surface is collected from the image collector, and the wear area ratio is , the deviation amount of the deviation situation is And pre-set conveyor belt operating parameters, such as conveying material weight , running speed , Ambient temperature ;

[0087] Based on elastic modulus E, cross-sectional area S, friction coefficient Equal parameters to establish preload The calculation model is as follows: Where: are conventional coefficients obtained from experiments. ;

[0088] The collected data here is assumed to be: Then, by bringing the preload calculation model into the calculation, we can get:

[0089]

[0090] S5. Detect the data obtained in step S3 and step S4, and dynamically adjust the preload force; the specific steps are as follows:

[0091] Adjustment signal trigger: When the wear change of the conveyor belt surface is detected, or the conveyor belt is found to be deviated, it is detected or When the preload adjustment signal is sent to the preload adjustment unit;

[0092] Set the optimal preload force. Compared with the optimal preload, if If it is greater than the optimal preload, reduce the preload; If the preload is less than the optimum preload, increase the preload; If it is equal to the optimum preload, no adjustment is required;

[0093] Preload force increases: the linear drive assembly drives the screw to rotate in the positive direction, and the slider moves along the guide structure slot, driving the adjustment base away from the drive end roller and tightening the conveyor belt;

[0094] Preload force decreases: the linear drive assembly rotates in the opposite direction, the slider moves back, the base is adjusted to push the roller closer to the drive end, and the conveyor belt is loosened;

[0095] S6. The preload adjustment unit receives the conveyor belt status data updated by the visual detection unit in real time, and executes steps S1-S5 in a loop until the preload force reaches the optimal value.

[0096] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although detailed descriptions have been made with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein with equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A mine conveyor belt pre-tensioning device based on visual detection, comprising a conveyor main structure, the conveyor main structure comprising a truss (100), rollers (110) arranged at both ends of the truss, a conveyor belt (111) sleeved between the two rollers, a driving mechanism (120) for driving the rollers on one side, a roller (130) arranged in the truss, and a detection platform and a protective frame (140) arranged on the side of the truss, characterized in that: Also includes: The preload adjustment part (200) is symmetrically arranged on both sides of the non-driving end roller (110), and comprises an adjustment base (210) connected to the roller shaft end, a guide structure (220) fixed on the truss (100), and a linear drive assembly (230), wherein the adjustment base (210) is slidably arranged on the guide structure (220), and the output end of the linear drive assembly (230) is drivingly connected to the adjustment base (210) to drive it to move linearly along the guide structure (220); A visual inspection unit (300) comprises a protective shell (330) mounted on a protective frame (140) with an adjustable angle, an image collector (310) and a light filler (320) arranged in the protective shell, and a self-cleaning mechanism (340), wherein the protective shell (330) fixes the image collection angle through an angle locking mechanism, and the self-cleaning mechanism (340) comprises a rotating brush assembly (341) in contact with the surface of the protective shell, a cleaning cavity (342) accommodating the rotating brush assembly, and a power assembly (343) driving the rotating brush assembly to move, and an elastic cleaning strip (3421) is provided on the inner wall of the cleaning cavity (342) to generate contact vibration with the rotating brush assembly (341); The image collector (310) performs grayscale processing on the collected image, uses an edge detection algorithm to detect the edge of the surface of the conveyor belt (111), performs morphological processing on the detected edge, and calculates the wear area ratio of the surface of the conveyor belt (111), and then determines whether the preload force of the conveyor belt (111) is adjusted according to a set wear threshold; a preload force calculation model is established based on the mechanical principles and empirical formulas of mine conveyor belts and the elastic modulus, cross-sectional area, and friction coefficient of the conveyor belt (111); the image collector (310) transmits a signal to the preload adjustment unit (200), and the linear drive component (230) is started to drive the support roller (110) of the adjustment base (210), thereby tightening or releasing the tension of the conveyor belt (111).

2. The mine conveyor belt pre-tensioning device based on visual detection according to claim 1 is characterized in that: The top of the adjustment base (210) is fixed with a bearing seat (113) by means of fasteners to support the shaft end of the roller, and the bottom thereof is welded with a clamping plate (211) to be slidably engaged with the guide structure (220); The guide structure (220) is composed of a pair of welded inclined plates and a slide groove (221) is provided on the top edge. The linear drive assembly (230) is a servo motor. The servo motor drives a screw rod (231). The screw rod (231) penetrates the guide structure (220) and is threadedly engaged with a slider (232). The top of the slider (232) is fixedly connected to the adjustment base (210) and is engaged and slidably engaged with the slide groove (221). The two ends of the guide structure (220) are bolted to the truss (100) via triangular blocks, and the servo motor is bolted to the end cover plate of the truss.

3. The mine conveyor belt pre-tensioning device based on visual detection according to claim 1 is characterized in that: The protective shell (330) is connected to the protective frame via a support block (331) hinged at the bottom; an inverted U-shaped clamping plate (332) is provided at the bottom end of the support block (331) and is fixed to the protective frame via bolts; A pre-tightening screw rod (350) and a threaded nut are provided on the top of the support block (331); a straight handle is fixedly connected to the nut via a bolt, and the angle is locked by rotating the nut to squeeze the support block.

4. The mine conveyor belt pre-tensioning device based on visual detection according to claim 1 is characterized in that: The rotating brush assembly (341) is in a cross structure and is embedded with nylon bristles (3411), and its central axis is connected to the power assembly (343) via a pulley (3413) and a transmission belt (3414); The cleaning cavity (342) is composed of two semicircular plates connected in one piece, wherein an elastic cleaning strip (3421) made of PP material is embedded in the inner wall of one semicircular plate, and a fixed sleeve (3422) is provided at the bottom of the other semicircular plate to be sleeved on the central axis of the rotating brush assembly; The power assembly (343) is an electric motor, which is mounted on a flat plate on the bottom side wall of the cleaning chamber (342) via fasteners.

5. The mine conveyor belt pre-tensioning device based on visual detection according to claim 1 is characterized in that: The side surface of the protective shell (330) is made of a transparent material, the image collector (310) is a high-speed camera, and the fill light (320) is an illumination lamp; The high-speed camera and the lighting lamp are arranged in a protective housing and are aimed at the surface of the conveyor belt (111).

6. The mine conveyor belt pre-tensioning device based on visual detection according to claim 4 is characterized in that: The central axis of the pulley (3413) is bolted to the bottom of the protective housing via a support sleeve (3412), and the bristles (3411) generate vibrations when in contact with the elastic cleaning strip (3421).

7. A pre-tightening method for a mine conveyor belt pre-tightening device using the visual detection-based pre-tightening device according to any one of claims 1 to 6, characterized in that: The specific steps are as follows: S1. Collecting the conveyor belt surface image and preprocessing the collected image; The image collector (310) is configured at a frame rate Collect the surface image of the conveyor belt (111), the frame rate is based on the running speed of the conveyor belt Dynamic settings; fill light provides uniform lighting to ensure image brightness In the appropriate range, ; Convert color images to grayscale images using weighted averaging , the formula is: Where: is the red channel pixel value; is the green channel pixel value; is the blue channel pixel value; S2. performing edge detection and morphological processing on the preprocessed image obtained in step S1; Performing Gaussian filtering on the grayscale image obtained in step S1 to remove noise; Use the Sobel operator to calculate the pixel in the image Gradient in direction and Gradient in direction , gradient amplitude The formula is as follows: Then use 3×3 square structure elements for the edge image Perform expansion and erosion operations; The expansion operation formula is as follows: The corrosion operation formula is as follows: Where: To expand the operating range, is the corrosion operating range, is the image brightness, represents the dilation operation, represents the erosion operation; S3. Performing status analysis on the processed conveyor belt image obtained in step S2; Calculate the wear area ratio of the conveyor belt surface , perform pixel statistics on the processed image; calculated by the following formula: Where: is the total number of pixels of the conveyor belt surface image, The number of pixels in the worn area; set the threshold , determine that the conveyor belt needs to adjust the preload; Assume the conveyor belt image width is , center reference line Location is ; By analyzing the pixel position of the conveyor belt edge in the image , calculate the deviation : The threshold is set here as If the deviation , it is judged as deviation; S4. Calculate the preload force using the data obtained in step S3; the specific steps are as follows: The wear condition of the conveyor belt surface is collected from the image collector, and the wear area ratio is , the deviation amount of the deviation situation is And pre-set conveyor belt operating parameters, such as conveying material weight , running speed , Ambient temperature ; Based on elastic modulus E, cross-sectional area S, friction coefficient Equal parameters to establish preload The calculation model is as follows: Where: are conventional coefficients obtained from experiments. ; S5. Detect the data obtained in step S3 and step S4, and dynamically adjust the preload force; When the conveyor belt surface is detected to have wear changes, or the conveyor belt is found to have deviation, it is detected or When , an adjustment signal is sent to the preload adjustment unit; Set the optimal preload force. Compared with the optimal preload, if If it is greater than the optimal preload, reduce the preload; If the preload is less than the optimum preload, increase the preload. If it is equal to the optimum preload force, no adjustment is required; When the preload force increases: the linear drive assembly (230) drives the screw rod (231) to rotate in the positive direction, and the slider (232) moves along the slide groove of the guide structure (220), driving the adjustment base (210) away from the drive end roller (110), thereby tightening the conveyor belt (111); When the preload force decreases: the linear drive assembly (230) rotates in the opposite direction, the slider (232) retreats, the adjustment base (210) pushes the roller (110) closer to the drive end, and the conveyor belt (111) is loosened; S6. The preload adjustment unit receives the conveyor belt status data updated by the visual detection unit in real time, and executes steps S1-S5 in a loop until the preload force reaches the optimal value.

Images