A rotary break-down cleaning mechanical arm

Abstract

The application discloses the technical fields of plugging cleaning technology, and relates to a rotary plugging breaking and cleaning mechanical arm, which comprises a chamber, a mine bin and an image recognition system, the mine bin is located at one side of the chamber, an installation seat is arranged in the inner cavity of the chamber, a breaking arm is arranged on the top of the installation seat, a chain plate type conveyor belt is arranged in the mine bin, a breaking bin is arranged at one end of the chamber close to the chain plate type conveyor belt, and the breaking bin is arranged below the chain plate type conveyor belt. Through the cooperation of the chamber, the image recognition system, the breaking arm, the breaking bin and the chain plate type conveyor belt, the chain plate type conveyor belt is used for conveying ore, and the image of the ore is collected by a high-speed camera when the ore falls, the images are then transmitted to a rear-end algorithm processing node, when the rear-end algorithm processing identifies that the size of the ore exceeds a preset threshold, a large ore alarm system is triggered, and the breaking arm is used for breaking the large ore, so that manual intervention is reduced, and the processing efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of blockage cleaning, in particular to a rotary blockage breaking and cleaning mechanical arm. BACKGROUND

[0002] In the mineral processing process, in order to ensure the working efficiency of the mill, there is a certain requirement for the particle size of the ore entering the mill, but large ore often blocks the ore mouth, which will directly affect the storage of the ore bin and seriously affect the working efficiency of the mill and the release of the mine capacity.

[0003] The existing cleaning method mainly relies on manual blockage cleaning on site, which is not conducive to the occupational health of workers due to high dust concentration and high temperature in the limited space on site, and manual cleaning efficiency is low, which will lead to production interruption. Based on this, a rotary blockage breaking and cleaning mechanical arm is proposed. SUMMARY

[0004] In order to improve the above-mentioned problem that the existing cleaning method mainly relies on manual blockage cleaning on site, which is not conducive to the occupational health of workers due to high dust concentration and high temperature in the limited space on site, and manual cleaning efficiency is low, which will lead to production interruption, the present application provides a rotary blockage breaking and cleaning mechanical arm.

[0005] The present application provides a rotary blockage breaking and cleaning mechanical arm, which adopts the following technical scheme:

[0006] A rotary blockage breaking and cleaning mechanical arm, comprising a chamber, an ore bin and an image recognition system, the ore bin is located on one side of the chamber, the inner cavity of the chamber is provided with a mounting seat, the top of the mounting seat is provided with a breaking arm, the inside of the ore bin is provided with a chain plate conveyor, and the left end of the chain plate conveyor extends into the inside of the chamber, the inside of the chamber is provided with a breaking bin near one end of the chain plate conveyor, and the breaking bin is located below the chain plate conveyor.

[0007] By adopting the above technical scheme, the chain plate conveyor works to convey the granite ore until the ore reaches the end of the chain plate conveyor, the image recognition system can capture the moment when the ore falls, and the collected image data is analyzed by algorithm to identify and measure the size of the ore. Under normal circumstances, if the size of the ore is small, it will be very smooth to roll into the inside of the conveying channel from top to bottom. When the image recognition system identifies that the size of the ore exceeds the preset threshold, the breaking arm works to divert the ore into the inside of the breaking bin, and the large ore is broken. After the large ore is broken, it enters the inside of the feeding bin through the conveying channel for storage or further processing. The whole process ensures the efficiency and safety of ore processing. Through automatic image recognition and breaking system, manual intervention is reduced, and processing efficiency is improved.

[0008] Optionally, the crushing arm comprises a slewing bearing, the top of the slewing bearing is provided with a support at the front and rear ends, the top of the two groups of supports is hinged with a large arm, a slewing motor is installed between the two groups of supports, the bottom of the slewing motor is connected with the top of the slewing bearing, the end of the large arm away from the support is hinged with a small arm, and the end of the small arm away from the large arm is hinged with a hydraulic hammer through an I-shaped frame link.

[0009] By adopting the above technical scheme, the oversized ore is crushed by the hydraulic hammer, so that the size is reduced to the standard size that can pass through the conveyor belt.

[0010] Optionally, the surface of the large arm is hinged with a first oil cylinder at the bottom of the support, the top of the small arm is hinged with a second oil cylinder at the middle of the large arm, and the top of the I-shaped frame link is hinged with a third oil cylinder between the small arm.

[0011] By adopting the above technical scheme, the extension and retraction of the first oil cylinder can adjust the angle between the large arm and the support, the extension and retraction of the second oil cylinder can adjust the angle between the small arm and the large arm, and the extension and retraction of the third oil cylinder can adjust the angle between the hydraulic hammer and the small arm, so that the angle of the hydraulic hammer is adjustable during crushing, and the flexibility is higher, and the traditional mechanical crushing arm crushing ore from top to bottom along the vertical direction is changed.

[0012] Optionally, the bottom of the slewing bearing is provided with a slewing base, and a plurality of foundation bolts are arranged between the slewing base and the mounting seat.

[0013] By adopting the above technical scheme, the slewing base can be fixedly installed through the plurality of foundation bolts, and the whole crushing arm can be fixedly installed.

[0014] Optionally, the surface of the crushing arm is coated with a paint layer.

[0015] By adopting the above technical scheme, the whole crushing arm has certain corrosion resistance, and the service life of the crushing arm is prolonged.

[0016] Optionally, the top of the chain plate conveyor belt is symmetrically provided with a material blocking plate at the front and rear ends.

[0017] By adopting the above technical scheme, the ore on the chain plate conveyor belt can be protected by the material blocking plate, and the ore falling off the chain plate conveyor belt is avoided as much as possible.

[0018] Optionally, the first oil cylinder, the second oil cylinder and the third oil cylinder all adopt high-hardness chrome-plated piston rods, and the exposed parts of the piston rods are all provided with protective sleeves.

[0019] Through the above technical scheme, the strength of the first oil cylinder, the second oil cylinder and the third oil cylinder is high, and the service life of the first oil cylinder, the second oil cylinder and the third oil cylinder is prolonged.

[0020] Optionally, one side of the crushing chamber is provided with a conveying channel, a feeding port is arranged between the crushing chamber and the conveying channel, and a feeding chamber is mounted at the bottom of the conveying channel.

[0021] Through the above technical scheme, the ore can be conveyed to the inside of the feeding chamber through the conveying channel for storage or unified processing, and the ore can be conveniently collected.

[0022] Optionally, the image recognition system comprises a high-speed camera, a back-end algorithm processing and a large ore alarm, the output end of the high-speed camera is electrically connected with the input end of the back-end algorithm processing, the back-end algorithm processing is electrically connected with the input end of the large ore alarm, the output end of the large ore alarm is connected with the input end of an automatic system, and the output end of the automatic system is electrically connected with the input end of the crushing arm.

[0023] Through the above technical scheme, the high-speed camera can capture the moment when the ore falls, and provide original data for subsequent image processing. The image data from the high-speed camera is received by the back-end algorithm processing, and algorithms are used for analysis to identify and measure the size of the ore. When the back-end algorithm processing identifies that the size of the ore exceeds a preset threshold, the large ore alarm system is triggered to notify the operator or the automatic system to take measures, and the crushing arm is controlled by the automatic system to perform crushing work.

[0024] Optionally, the back-end algorithm processing adopts a MaskRCNN network architecture algorithm, and the network architecture of the MaskRCNN comprises a backbone network CNN (adopting ResNet101+FPN as a feature extractor), a region proposal network (RPN), an ROI classifier and a bounding box regressor, and a segmentation mask.

[0025] Through the above technical scheme, the target ore detection and segmentation results are fused to obtain a detection result consistent with the size of the original image and containing the target ore category and shape.

[0026] In summary, the present application has at least one of the following beneficial effects:

[0027] Through the cooperation of the chamber, the image recognition system, the crushing arm, the crushing bin and the chain plate conveyor, the chain plate conveyor performs the conveying work on the ore, and the high-speed camera collects the images of the ore when falling, and these images are then transmitted to the back-end algorithm processing node, which is responsible for analyzing the images and judging whether the ore is oversized, when the back-end algorithm processing identifies that the size of the ore exceeds the preset threshold, the large ore alarm system will be triggered, when the large ore alarm is triggered, the crushing arm will intervene to crush the large ore, ensuring the efficiency and safety of ore processing, through the automatic image recognition and crushing system, reducing manual intervention and improving processing efficiency.

[0028] Through the cooperation of the first oil cylinder, the second oil cylinder and the third oil cylinder, the first oil cylinder can adjust the angle of the large arm, the second oil cylinder can adjust the angle of the small arm, and the third oil cylinder can adjust the angle of the hydraulic hammer, so that the angle of the hydraulic hammer can be adjusted when crushing, and the flexibility is higher, which changes the traditional mechanical crushing arm crushing ore from top to bottom along the vertical direction. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced below, and obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0030] Figure 1 It is the overall structure schematic diagram of the present application;

[0031] Figure 2 It is the crushing arm structure schematic diagram of the present application;

[0032] Figure 3 It is the rotary bearing and rotary motor connection structure schematic diagram of the present application;

[0033] Figure 4 It is the system structure schematic diagram of the present application;

[0034] Figure 5 It is the stone crushing process flowchart of the present application;

[0035] Figure 6 It is the MaskRCNN network structure flowchart adopted by the present application;

[0036] Figure 7 It is the schematic diagram of the ore contour obtained by the MaskRCNN algorithm of the present application.

[0037] In the figure: 1, chamber; 2, high-speed camera; 3, breaking arm; 301, support; 302, rotary support; 303, large arm; 304, small arm; 305, I-beam connecting rod; 306, hydraulic hammer; 307, rotary motor; 308, first oil cylinder; 309, second oil cylinder; 310, third oil cylinder; 4, conveying channel; 5, feeding bin; 6, breaking bin; 7, material blocking plate; 8, chain plate conveyor; 9, mounting seat; 10, mine bin. DETAILED DESCRIPTION

[0038] The following will be described in detail in combination with the accompanying drawings Figures 1-7 The present application is further described in detail.

[0039] Please refer to the drawings in the specification Figure 1 An embodiment provided by the present application: a rotary breaking and cleaning mechanical arm for broken material, comprising a chamber 1, a mine bin 10 and an image recognition system, the mine bin 10 is located on one side of the chamber 1, the inner cavity of the chamber 1 is provided with a mounting seat 9, the top of the mounting seat 9 is provided with a breaking arm 3, and the surface of the breaking arm 3 is coated with a paint layer. So that the breaking arm 3 has certain corrosion resistance as a whole, thereby prolonging the service life of the breaking arm 3.

[0040] Please refer to the drawings in the specification Figure 1 The image recognition system comprises a high-speed camera 2, a back-end algorithm processing and a large ore alarm, the output end of the high-speed camera 2 is electrically connected with the input end of the back-end algorithm processing, the back-end algorithm processing is electrically connected with the input end of the large ore alarm, the output end of the large ore alarm is connected with the input end of an automatic system, and the output end of the automatic system is electrically connected with the input end of the breaking arm 3. The high-speed camera 2 can capture the moment when the ore falls, providing raw data for subsequent image processing. The image data from the high-speed camera 2 is received by the back-end algorithm processing and analyzed using algorithms to identify and measure the size of the ore. When the back-end algorithm processing identifies that the size of the ore exceeds a preset threshold, the large ore alarm system will be triggered to notify the operator or the automatic system to take measures, and then the breaking arm 3 is controlled by the automatic system to perform breaking work. The back-end algorithm processing adopts a MaskRCNN network architecture algorithm, and the network architecture of MaskRCNN includes a backbone network CNN (using ResNet101+FPN as a feature extractor, a region proposal network (RPN, a ROI classifier, a bounding box regressor and a segmentation mask. Thus, the target ore detection and segmentation results are fused to obtain a detection result consistent with the original image size and containing the target ore category and shape.

[0041] Please refer to the drawings in the specification Figure 1The inside of the ore bin 10 is provided with a chain plate conveyor 8, and the left end of the chain plate conveyor 8 extends into the inside of the chamber 1. The front and rear ends of the top of the chain plate conveyor 8 are symmetrically provided with a material blocking plate 7. The material blocking plate 7 can protect the ore on the chain plate conveyor 8, and try to avoid the ore from sliding off the chain plate conveyor 8. The inside of the chamber 1 is provided with a crushing bin 6 near one end of the chain plate conveyor 8, and the crushing bin 6 is located below the chain plate conveyor 8.

[0042] Please refer to the drawings in the specification Figure 1 The side of the crushing bin 6 is provided with a conveying passage 4, and the conveying passage 4 is provided with a feeding bin 5 at the bottom.

[0043] Please refer to the drawings in the specification Figure 1 、 Figure 2 and Figure 3 The crushing arm 3 includes a rotary bearing 302, and the top of the rotary bearing 302 is provided with a support 301 at the front and rear ends. The top of the two groups of supports 301 is hinged with a large arm 303, and the rotary motor 307 is installed between the two groups of supports 301. The bottom of the rotary motor 307 is connected with the top of the rotary bearing 302. The end of the large arm 303 away from the support 301 is hinged with a small arm 304, and the end of the small arm 304 away from the large arm 303 is hinged with a hydraulic hammer 306 through an I-shaped frame link 305. The hydraulic hammer 306 is used to crush the oversized ore, so that the size of the ore is reduced to the standard size that can pass through the conveyor belt.

[0044] Please refer to the drawings in the specification Figure 1 and Figure 2 A plurality of foundation bolts are arranged between the rotary base and the mounting seat 9. Thus, the rotary base can be fixedly installed through the plurality of foundation bolts, and the entire crushing arm 3 can be fixedly installed.

[0045] Please refer to the drawings in the specification Figure 1 and Figure 2 The surface of the large arm 303 and the bottom of the support 301 are hinged with a first oil cylinder 308. The top of the small arm 304 and the middle part of the large arm 303 are hinged with a second oil cylinder 309. The top of the I-shaped frame link 305 and the small arm 304 are hinged with a third oil cylinder 310. The extension and retraction of the first oil cylinder 308 can adjust the angle between the large arm 303 and the support 301. The extension and retraction of the second oil cylinder 309 can adjust the angle between the small arm 304 and the large arm 303. The extension and retraction of the third oil cylinder 310 can adjust the angle between the hydraulic hammer 306 and the small arm 304. Thus, the angle of the hydraulic hammer 306 can be adjusted during crushing, and the flexibility is higher. The traditional mechanical crushing arm is changed from crushing ore from top to bottom along the vertical direction.

[0046] Referring to the drawings of the specification Figure 2 The first oil cylinder 308, the second oil cylinder 309 and the third oil cylinder 310 all adopt high-hardness chromium-plated piston rods, and protective sleeves are installed on the exposed parts of the piston rods. The strength of the first oil cylinder 308, the second oil cylinder 309 and the third oil cylinder 310 is high, thereby prolonging the service life of the first oil cylinder 308, the second oil cylinder 309 and the third oil cylinder 310.

[0047] Referring to the drawings of the specification Figure 4 The system is composed of three main subsystems: the conveyor belt system, the image recognition system and the crushing system. The following is a detailed description of each subsystem and its components:

[0048] 1. Conveyor belt system:

[0049] Chain plate conveyor belt 8: This is the main part of the system, responsible for conveying ore from the mining site to the processing area. The ore moves on the chain plate conveyor belt 8 until it reaches the end of the chain plate conveyor belt 8.

[0050] 2. Image recognition system:

[0051] 1) High-speed image acquisition: Located at the end of the chain plate conveyor belt 8, it is used to capture images of the ore in real time. The high-speed camera 2 can capture the moment when the ore falls, providing raw data for subsequent image processing.

[0052] 2) Back-end algorithm processing: Receives image data from the high-speed camera 2 and uses algorithms for analysis to identify and measure the size of the ore.

[0053] 3) Large ore alarm: When the back-end algorithm processing identifies that the size of the ore exceeds the preset threshold, the large ore alarm system will be triggered to notify the operator or the automatic system to take measures.

[0054] Among them, the MaskRCNN target detection algorithm based on mask region convolutional neural network is used in high-speed image acquisition. MaskRCNN is a target detection and segmentation algorithm, which uses a two-stage framework: the first stage scans the image and generates proposals (regions that may contain a target); the second stage classifies the proposals and generates bounding boxes and masks. MaskRCNN adds a semantic segmentation branch based on FasterR-CNN, enabling semantic segmentation while performing target detection. The network architecture of MaskRCNN mainly consists of four parts: the backbone network CNN (using ResNet101+FPN as the feature extractor), the region proposal network (RPN), the ROI classifier and the bounding box regressor, and the segmentation mask. The network structure is as shown in Figure 6

[0055] ​The ore image is transmitted to the CNN network for feature extraction. Due to the complex environment background of the ore image, a deep residual network ResNet101 with stronger feature extraction capability is combined with FPN as a feature extraction module to extract features from the ore image. ResNet101 can extract low-level features (ore edges, corner points) and high-level features (ore types and shapes, background, etc.), forming 5 layers of features with different sizes and dimensions. The addition of FPN is to make up for the inability of ResNet101 to detect small ore features, and to better fuse the features of each layer. The features extracted by ResNet101 are fully utilized, and the feature maps extracted by the CNN network are transmitted to the region proposal network. The region proposal network classifies the image into target ore and background, and uses a box that fits the target ore to frame the ore. At this time, only the approximate area of the target ore can be obtained, and detailed classification and target ore segmentation cannot be performed. After the candidate region network, multiple candidate boxes containing the target ore are obtained, and the ROIAlign is pooled into a fixed size feature map. The obtained feature map is input into the subsequent two branches, one of which is a branch network that uses a region of interest (ROI) classifier and a bounding box regressor to identify the target ore. Both the region of interest (ROI) classifier and the bounding box regressor are composed of a fully connected layer. The region of interest (ROI) classifier obtains the accurate target ore category through the fully connected layer, and the bounding box regressor selects the appropriate center point coordinates and aspect ratio of the region of interest through the fully connected layer, so that the region of interest can fit the edge of the target ore as much as possible. The other branch is a segmentation mask generation network composed of a fully convolutional network. This network generates a mask consistent with the size and shape of the target ore for instance segmentation of the target ore. Finally, the target ore detection and segmentation results are fused to obtain a detection result consistent with the original image size and containing the target ore category and shape. The specific situation is as follows Figure 7 As an example, the algorithm is used to obtain the ore contour diagram.

[0056] 3. Breaking system:

[0057] Breaking arm 3: When the large ore alarm is triggered, the breaking arm 3 intervenes to break the oversized ore to reduce its size to the standard size that can pass through the conveyor belt.

[0058] In this system, the chain plate conveyor belt 8 conveys the ore to the high-speed image acquisition node, which detects the falling of the ore and captures images. These images are then transmitted to the back-end algorithm processing node, which is responsible for analyzing the images and determining whether the ore is oversized. If a large ore is detected, the system will trigger the breaking arm 3 for processing through the large ore alarm node. The whole process aims to ensure the efficiency and safety of ore processing, and through the automatic image recognition and breaking system, manual intervention is reduced and processing efficiency is improved.

[0059] Referring to the attached drawings Figure 5 , the ore handling process involves the transportation, detection, crushing (if necessary), and final destination of the ore. The following is a detailed explanation of this process:

[0060] 1. Conveyor belt transports ore: The process begins with a chain conveyor belt 8, which is responsible for transporting the ore from the mining site or storage area to the processing area.

[0061] 2. Ore drops off the conveyor belt: The ore drops off the end of the chain conveyor belt 8, which triggers the subsequent detection process.

[0062] 3. High-speed camera 2 and backend algorithm processing: At the moment the ore drops off, the high-speed camera 2 captures images of the ore, which are then transmitted to the backend algorithm processing system. This system is responsible for analyzing the size of the ore to determine if it exceeds the established size threshold.

[0063] 4. Identifying and marking oversized ore: If the backend algorithm processing system identifies that the ore exceeds the size threshold, the system will mark these oversized ores.

[0064] 5. Ore size determination: The system determines whether the ore exceeds the size threshold. If the ore exceeds the size threshold, the process will branch to "yes"; if the ore size is within the normal range, it will branch to "no".

[0065] 6. Yes branch: If the ore exceeds the size threshold, the system will issue a large ore alert, indicating that it needs to be processed. Then, the crushing arm 3 (external system) will be activated to crush the large ore to reduce its size. After processing, the system will check the size of the ore again to ensure it meets the standard.

[0066] 7. No branch: If the ore size is within the normal range, the ore will smoothly roll into the feed inlet and does not require further processing.

[0067] 8. Ore enters the silo: Regardless of whether the ore has been crushed, it will eventually enter the silo 5 for storage or further processing.

[0068] The entire process aims to automate the handling of ore, ensuring that only ore that meets size requirements can enter the silo, thereby improving the efficiency and safety of ore handling. Through the combination of high-speed camera 2 and backend algorithms, the system can effectively identify and process oversized ore, reducing manual intervention and reducing operating costs.

[0069] Working principle: in use, open the high-speed camera 2 and the chain plate conveyor 8, the chain plate conveyor 8 works to convey the granite ore, in the process of conveying, the ore is blocked by two groups of material blocking plates 7 to prevent the ore from falling off the chain plate conveyor 8, until the ore reaches the end of the chain plate conveyor 8, the high-speed camera 2 can capture the moment of the ore falling, and the collected image data is transmitted to the back-end algorithm processing, the back-end algorithm processing receives the image data from the high-speed camera 2, and analyzes the image data by using the algorithm to identify and measure the size of the ore, under normal circumstances, if the size of the ore is small, it will be very smooth from top to bottom into the inside of the conveying channel 4, when the back-end algorithm processing identifies that the size of the ore exceeds the preset threshold, the large ore alarm system will be triggered, when the large ore alarm is triggered, the operator or the automatic system will be notified to take measures, at this time, the first oil cylinder 308, the second oil cylinder 309 and the third oil cylinder 310 are controlled to work, so that the ore can be allocated to the inside of the crushing bin 6, and the large ore is crushed by the hydraulic hammer 306, after the large ore is crushed, it enters the conveying channel 4 through the inlet of the crushing bin 6, and finally enters the inside of the feeding bin 5 for storage or further processing, the whole process ensures the efficiency and safety of the ore processing, and reduces the manual intervention and improves the processing efficiency through the automatic image recognition and crushing system.

[0070] The above are preferred embodiments of the present application, and are not intended to limit the protection scope of the present application, therefore: any equivalent changes made according to the structure, shape, principle of the present application should be covered within the protection scope of the present application.

Claims

1. A rotary plug breaking and cleaning mechanical arm, comprising a chamber (1), a bin (10), and an image recognition system, wherein the bin (10) is located on one side of the chamber (1), characterized in that: The inner cavity of the chamber (1) is provided with a mounting seat (9), the top of the mounting seat (9) is provided with a crushing arm (3), the inside of the bunker (10) is provided with a chain plate conveyor (8), and the left end of the chain plate conveyor (8) extends into the inside of the chamber (1), and the end of the chamber (1) close to the chain plate conveyor (8) is provided with a crushing bin (6), and the crushing bin (6) is located below the chain plate conveyor (8). The image recognition system comprises a high-speed camera (2), a rear-end algorithm processing and a large ore alarm, the output end of the high-speed camera (2) is electrically connected with the input end of the rear-end algorithm processing, the rear-end algorithm processing is electrically connected with the input end of the large ore alarm, the output end of the large ore alarm is connected with the input end of the automatic system, and the output end of the automatic system is electrically connected with the input end of the crushing arm (3). When the rear-end algorithm processing identifies that the size of the ore exceeds the preset threshold, the large ore alarm system is triggered, the crushing arm intervenes, and the oversized ore is crushed.

2. The mechanical arm for cleaning the jammed material according to claim 1, wherein: The crushing arm (3) comprises a rotary support (302), the front and rear ends of the top of the rotary support (302) are provided with supports (301), the top of the two groups of supports (301) are hingedly connected with large arms (303), a rotary motor (307) is arranged between the two groups of supports (301), and the bottom of the rotary motor (307) is connected with the top of the rotary support (302), the end, away from the support (301), of the large arm (303) is hingedly connected with a small arm (304), and the end, away from the large arm (303), of the small arm (304) is hingedly connected with a hydraulic hammer (306) through an I-shaped frame connecting rod (305).

3. The mechanical arm for cleaning the jammed material according to claim 2, wherein: The surface of the large arm (303) and the bottom of the support (301) are hingedly connected with a first oil cylinder (308), the top of the small arm (304) and the middle part of the large arm (303) are hingedly connected with a second oil cylinder (309), and the top of the I-shaped frame connecting rod (305) and the small arm (304) are hingedly connected with a third oil cylinder (310).

4. The mechanical arm for cleaning the jammed material according to claim 2, wherein: The bottom of the rotary support (302) is provided with a rotary base, and a plurality of foundation bolts are arranged between the rotary base and the mounting seat (9).

5. The mechanical arm for cleaning the jammed material according to claim 1, wherein: The surface of the crushing arm (3) is coated with a paint layer.

6. The mechanical arm for cleaning the jammed material according to claim 1, wherein: The front and rear ends of the top of the chain plate conveyor (8) are symmetrically provided with material blocking plates (7).

7. The mechanical arm for cleaning the jammed material according to claim 3, wherein: The first oil cylinder (308), the second oil cylinder (309) and the third oil cylinder (310) all adopt high-hardness chromium-plated piston rods, and the exposed parts of the piston rods are all provided with protective sleeves.

8. The mechanical arm for cleaning the jammed material according to claim 1, wherein: One side of the crushing bin (6) is provided with a conveying channel (4), a feeding port is arranged between the crushing bin (6) and the conveying channel (4), and the bottom of the conveying channel (4) is provided with a feeding bin (5).

9. The mechanical arm for cleaning the jammed material according to claim 1, wherein: The rear-end algorithm processing adopts a MaskRCNN network architecture algorithm, and the network architecture of the MaskRCNN comprises a backbone network CNN, a region proposal network, an ROI classifier, a bounding box regressor and a segmentation mask.

Images