Belt tear detection system, method, apparatus, storage medium and electronic equipment

Abstract

This invention relates to the field of belt inspection technology, and particularly to a belt tear detection system, method, apparatus, storage medium, and electronic device. The system includes: an image acquisition terminal disposed at a first end and a second end of the belt, for acquiring an image of the lower surface of the belt at the first end and an image of the upper surface of the belt at the second end; a tear detection module for detecting whether the belt has a through-tear based on the image of the lower surface of the belt at the first end and for detecting whether the belt has a non-through tear based on the image of the upper surface of the belt at the second end; and if the belt tears and the number of image frames of the tear exceeds a preset number, a belt stop signal is triggered; thereby enabling the detection of belt tears and the location of the tear.

Description

Technical Field

[0001] This invention relates to the field of belt inspection technology, and particularly to a belt tear detection system, method, apparatus, storage medium, and electronic device. Background Technology

[0002] Currently, material conveyor belts are crucial equipment in bulk cargo ports. The health of these belts severely restricts the normal operation of these ports, making tear detection a primary task in their maintenance. After a belt tears, the location of the tear is often difficult to determine, requiring considerable time and manual effort to locate the initial tear.

[0003] In the 1980s, foreign scholars proposed pre-embedding conductors in the belt and using electromagnetic induction to indirectly determine whether the belt had torn by detecting conductor breakage. However, this method was costly due to the need to pre-embed conductors, making it largely impractical for companies already operating belt conveyors. Therefore, accurately detecting belt tears and pinpointing the exact location of the tear after it occurs is of significant practical importance.

[0004] There is an urgent need in this field for a solution to the technical problem of how to accurately detect tears and locate their positions. Summary of the Invention

[0005] This invention provides a belt tear detection system, method, device, storage medium, and electronic device, which solves the technical problem of how to accurately detect tear occurrence and locate tear position.

[0006] In a first aspect, the present invention provides a belt tear detection system, comprising:

[0007] An image acquisition terminal is set at the first end point and the second end point of the belt to acquire an image of the lower surface of the cargo belt at the first end point and an image of the upper surface of the cargo belt at the second end point.

[0008] The tear detection module detects whether the belt has a through-tear based on an image of the lower surface of the cargo belt at the first end, and detects whether the belt has a non-through tear based on an image of the upper surface of the cargo belt at the second end.

[0009] In some embodiments, the first endpoint includes the unloading hopper of the port tippler room, and the second endpoint includes the return conveyor belt of the transfer room.

[0010] In some embodiments, the lower surface includes the surface of the belt that contacts the idler roller, and the upper surface includes the surface of the belt that contacts the cargo.

[0011] In some embodiments, the image acquisition terminal includes a laser emission sub-terminal and a visual inspection sub-terminal;

[0012] Two image acquisition terminals are set side by side at the first and second ends of the belt.

[0013] Secondly, the present invention provides a belt tear detection method, comprising:

[0014] The image acquisition terminal acquires an image of the belt based on the first aspect;

[0015] The tear detection module, based on the first aspect, detects whether the belt has been torn;

[0016] If the belt tears and the number of image frames showing the tear exceeds the preset number of frames, a belt stop signal will be triggered.

[0017] In some embodiments, the method further includes:

[0018] When the belt stop signal is triggered, the location of the belt tear is calculated based on the belt stop signal and belt speed data.

[0019] In some embodiments, the method further includes:

[0020] When the belt stop signal is triggered, the location of the belt tear is predicted based on the belt stop signal and a preset number of belt speed data.

[0021] In some embodiments, the method further includes: after detecting a tear, segmenting the torn image using an adaptive thresholding method, whereby the tear crack is set to white and the belt to black. Then, the width of the crack is calculated using morphological methods, and different levels of alarms are triggered by setting different thresholds.

[0022] Thirdly, the present invention provides a belt tear detection device, comprising:

[0023] Image acquisition device, used to acquire an image of the belt based on the image acquisition terminal of the first aspect;

[0024] A tear detection device is used to detect whether the belt has torn based on the tear detection module of the first aspect; if the belt is torn and the number of image frames of the belt tear is greater than a preset number of frames, a belt stop signal is triggered.

[0025] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method of the second aspect.

[0026] Fifthly, the present invention provides an apparatus including a processor and a memory, wherein a computer program is stored in the memory, and the processor executes the computer program to implement the method of the second aspect.

[0027] This invention provides a belt tear detection system, method, apparatus, storage medium, and electronic device. The system uses image acquisition terminals located at the first and second ends of the belt to acquire images of the lower surface of the belt at the first end and the upper surface of the belt at the second end. A tear detection module detects whether there is a through-tear based on the lower surface image at the first end and whether there is a non-through tear based on the upper surface image at the second end. The tear width and severity are determined by cropping and segmenting the tear location image and performing morphological calculations. If the belt tears and the number of tear image frames exceeds a preset number, a belt stop signal is triggered. Based on the belt stop signal and the real-time belt speed, the specific location of the tear can be determined through integral calculation. Attached Figure Description

[0028] The invention will now be described in more detail with reference to embodiments and the accompanying drawings:

[0029] Figure 1 This is a schematic diagram of the structure of a belt tear detection system according to an embodiment of the present invention;

[0030] Figure 2 This is a schematic diagram of the installation of an image acquisition terminal according to an embodiment of the present invention;

[0031] Figure 3 This is a schematic diagram of the installation of another image acquisition terminal according to an embodiment of the present invention;

[0032] Figure 4 This is a schematic diagram of a belt tear detection method according to an embodiment of the present invention;

[0033] Figure 5 This is a schematic flowchart of another belt tear detection method according to an embodiment of the present invention;

[0034] Figure 6 This is a schematic diagram of a belt tear detection device according to an embodiment of the present invention.

[0035] In the accompanying drawings, the same parts are referred to by the same reference numerals, and the drawings are not drawn to scale. Detailed Implementation

[0036] To enable those skilled in the art to better understand the present invention and to fully understand and implement the process of how the present invention uses technical means to solve technical problems and achieve corresponding technical effects, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. The embodiments of the present invention and the various features therein can be combined with each other without conflict, and the resulting technical solutions are all within the protection scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0037] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0038] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0039] In the 1980s, foreign scholars proposed pre-embedding conductors in the belt and using electromagnetic induction to indirectly determine whether the belt was faulty by detecting whether the conductor had broken. However, this method required pre-embedding the conductors in the belt, which was costly and not very applicable to belt conveyor manufacturers already in production.

[0040] In 2011, Rong Feng et al. proposed X-ray fluoroscopy for detecting longitudinal tears in belts, indirectly determining whether a longitudinal tear had occurred by observing the difference in light intensity. This method is intuitive, reliable, and effective, but it did not consider the potential health hazards to operators from long-term use.

[0041] In 2008, Zhang Xi et al. used the difference image method to detect longitudinal tears in conveyor belts, detecting them by judging whether there is a tear in the material between two adjacent image frames. This technique is easily affected by dust in complex process environments, making it difficult to acquire high-quality images. Furthermore, it requires a huge amount of computation and has poor real-time performance. In addition, this method only detects material leakage after the belt tears; it cannot detect situations where the conveyor belt is not completely torn or where there is overlap at the tear point, where there is no leakage.

[0042] In 2015, Qiao Tiezhu et al. used visible light CCD and infrared CCD for binocular image acquisition. They proposed registering and fusing the corner and straight line feature images of conveyor belt tears acquired by infrared and visible light, and proposed a deep feature learning algorithm to classify and identify fault feature images. This method is a novel binocular vision online detection system that realizes real-time monitoring of longitudinal tears in conveyor belts and studies early warning strategies. However, this method requires extracting corner points from the images of the two cameras, then registering them, and then fusing features for learning, resulting in excessive computation and inaccurate detection results due to inaccurate corner point extraction.

[0043] Coal conveyor belts have high tensile strength, but their longitudinal tear resistance is only limited by the strength of the rubber itself, making them prone to longitudinal tears. Longitudinal tears can be caused by various factors, including detachment of equipment parts, loss of train car support rods, or the presence of hard objects such as chisels or I-beams. Belt conveyors are the lifeblood of coal transportation systems; longitudinal tears in these systems can result in significant direct and indirect losses.

[0044] Vision-based belt tear detection equipment is often installed between the upper and lower belts to detect whether there are penetrating tears on the cargo belt, but it cannot detect foreign objects scratching the belt or tear types that do not penetrate.

[0045] Traditional vision-based belt tear detection technology has limited analytical features and poor self-learning ability, making it difficult to adapt to the complex environment and characteristic changes in belt operation sites, such as changes in lighting, tear size and shape, and the influence of defects at belt joints and the belt itself.

[0046] Current tear detection algorithms lack the ability to determine whether a belt has a through-and-through tear or a non-through tear, the extent of the tear, or the size of the tear gap. This makes it difficult to assess the degree of belt defect and determine whether appropriate preventative measures should be taken.

[0047] After a belt tears, the location of the tear is often difficult to determine, requiring a considerable amount of time for manual searching to pinpoint the initial tear. Therefore, accurately locating the tear after it has broken is of significant practical importance.

[0048] To address the aforementioned issues, developing a belt longitudinal tear protection method capable of simultaneously detecting both through-and-through tears and non-through tears, stably and accurately detecting whether a belt has torn, determining the tear severity based on the tear width, and accurately locating the specific location of the belt tear after the belt has been stopped is of significant practical importance.

[0049] This invention provides a method for detecting tears in coal conveyor belts at coal ports, belonging to the fields of belt conveyor defect detection and safety protection.

[0050] To maximize belt utilization and minimize operator workload, relevant designs were developed, encompassing equipment installation processes and the analysis and application of different testing results, in conjunction with the production process flow.

[0051] The technical solution and its effects of the present invention will be described below with reference to embodiments.

[0052] Example 1

[0053] Figure 1 This is a schematic diagram of a belt tear detection system according to an embodiment of the present invention. Figure 1 As shown, this embodiment provides a belt tear detection system, including:

[0054] An image acquisition terminal is set at the first end point and the second end point of the belt to acquire an image of the lower surface of the cargo belt at the first end point and an image of the upper surface of the cargo belt at the second end point.

[0055] The tear detection module detects whether the belt has a through-tear based on an image of the lower surface of the cargo belt at the first end, and detects whether the belt has a non-through tear based on an image of the upper surface of the cargo belt at the second end.

[0056] The technical solution of this embodiment includes an image acquisition terminal and a tear detection module. The image acquisition terminal is located at the first and second ends of the belt to acquire images of the lower surface of the belt at the first end and the upper surface of the belt at the second end. The tear detection module detects whether there is a through tear based on the image of the lower surface of the belt at the first end and whether there is a non-through tear based on the image of the upper surface of the belt at the second end. The solution can quickly calculate the location of through tears and non-through tears based on the positions of the first and second ends.

[0057] Example 2

[0058] Figure 2This is a schematic diagram of the installation of an image acquisition terminal according to an embodiment of the present invention, showing the installation of a tear detection device at the material discharge hopper of a tipper. Figure 3 This is a schematic diagram illustrating the installation of another image acquisition terminal according to an embodiment of the present invention, showing the installation of a tear detection device at the return belt in the transfer room. For example... Figure 2 and Figure 3 As shown, based on the above embodiments, the system provided in this embodiment has a first endpoint including the unloading hopper of the port tipper room and a second endpoint including the return belt of the transfer room.

[0059] In some embodiments, the lower surface includes the surface of the belt that contacts the pulley, and the upper surface includes the surface of the belt that contacts the cargo.

[0060] In some implementations, the image acquisition terminal includes a laser emitting sub-terminal and a visual inspection sub-terminal;

[0061] Two image acquisition terminals are set side by side at the first and second ends of the belt.

[0062] In the technical solution of this embodiment, both penetrating and non-penetrating tears are detected simultaneously.

[0063] This invention also installs a belt tear detection device at the unloading hopper of the port tipper room and at the return belt of the transfer room.

[0064] The belt tear detection device at the unloading hopper of the tipper room is installed between the cargo belt and the return belt. By shining an upward beam onto the lower surface of the cargo belt, it can detect through-type belt tear faults.

[0065] A through-type belt tear refers to a tear that extends from the top surface of the belt to the bottom surface. The top surface of the belt is the side that contacts the cargo. The bottom surface of the belt is the side that does not contact the cargo. The cargo belt is located above the return belt.

[0066] like Figure 2 As shown, the visual inspection terminal 1 and the laser emitting terminal 2 cooperate to detect belt tear features. Due to the wide belt, two sets of inspection equipment are arranged over two consecutive spans, one for inspecting the left side of the belt and the other for inspecting the right side.

[0067] The belt tear detection device at the return belt of the transfer room is installed below the return belt and shines upwards. It is used to detect the upper surface of the cargo belt and can be used to identify non-penetrating belt tears.

[0068] like Figure 3 As shown, due to the wide belt, two sets of belt tear detection equipment are installed on two consecutive spans, one for detecting the left belt and the other for detecting the right belt.

[0069] In this embodiment, the advanced deep learning vision algorithm YOLOv5 is used to detect belt tears. This algorithm is robust, effectively overcoming the effects of changes in illumination, pose, and scale. Furthermore, it can continuously enrich and optimize the dataset to detect various types of tears. The YOLOv5 model is publicly released by Ultralytics. It is an improvement upon the YOLOv3 model, comprising four models: YOLOv5s, YOLOv5m, YOLOv5l, and YOLOv5x. The YOLOv5 model consists of a backbone network, a neck network, and a head network.

[0070] In this embodiment, upon detecting a belt tear, the image of the tear location is cropped, and then an adaptive threshold segmentation method is used to segment the torn image. The tear is marked in white, and the belt is marked in black. The width of the tear is represented by averaging the values ​​of 10 white pixels in the Y direction.

[0071] Two thresholds are set for different tear widths: a first threshold T1 and a second threshold T2. When the tear width Width < the first threshold T1, the tear is judged as a level 1 alarm; when the first threshold T1 < the tear width Width < the second threshold T2, the tear is judged as a level 2 alarm; and when the tear width Width > the second threshold T2, the tear is judged as a level 3 alarm.

[0072] In this embodiment, to accurately determine the location of a belt tear, the present invention employs a belt speed measuring wheel installed near the belt tear detection device to collect the belt's running speed v(t) at equal time intervals. During the process from when a belt tear is detected and the belt is stopped until it completely stops (its speed becomes zero), the belt speed collected during this period is integrated to obtain the precise location of the belt tear.

[0073] Equation (1) is the formula for calculating the location of the tear after the interlock belt stops when a belt tear is detected.

[0074]

[0075] Among them, S Teared S0 represents the initial position of the belt tear detection device after it stops, which is obtained from the initial measurement. t1 is the starting time of the belt stopping after the tear is detected. t2 is the time when the belt completely stops and its speed becomes zero. dt is the speed acquisition interval of the speed measuring wheel. By installing the belt tear detection device at the unloading hopper in the tippler room and at the return belt in the transfer room, the calculation of the tear position can be simplified.

[0076] The technical solution of this embodiment, by including a laser emission sub-terminal and a visual inspection sub-terminal in the image acquisition terminal, can acquire images based on laser, avoiding unreliable factors such as natural light and improving stability; by setting two image acquisition terminals side by side at the first end and the second end of the belt, it is possible to simultaneously detect whether there is a tear on the left and right sides of the belt, and locate the tear on the left and right sides respectively.

[0077] Example 3

[0078] Figure 4 This is a schematic flowchart of a belt tear detection method according to an embodiment of the present invention. Figure 4 As shown, based on the above embodiments, this embodiment provides a belt tear detection method, including:

[0079] The image acquisition terminal based on the above embodiment acquires an image of the belt;

[0080] The tear detection module based on the above embodiment detects whether the belt has been torn;

[0081] If the belt tears and the number of image frames showing the tear exceeds the preset number of frames, a belt stop signal will be triggered.

[0082] The technical solution of this embodiment triggers a belt stop signal by monitoring the number of image frames of belt tearing, which can promptly determine the occurrence of belt tearing and promptly start the location detection of belt tearing.

[0083] Example 4

[0084] Figure 5 This is a schematic flowchart of a belt tear detection method according to an embodiment of the present invention. Figure 5 As shown, based on the above embodiments, this embodiment provides a belt tear detection method, which further includes:

[0085] When the belt stop signal is triggered, the location of the belt tear is calculated based on the belt stop signal and belt speed data.

[0086] In this embodiment, the belt tear detection method includes the following steps:

[0087] (1) Read the belt start / stop signal from the PLC using the OPC communication protocol;

[0088] (2) If the belt is stopped, tear detection is not performed; when the belt is running, the YOLOV5 deep learning vision algorithm detects through and non-through belt tear faults in real time.

[0089] (3) If a tear is detected, crop the torn area image, perform adaptive threshold segmentation of the torn image, calculate the tear gap width by white pixel statistics, set different thresholds, and determine different tear levels.

[0090] (4) Determination of consecutive frames of tear images: When the number of consecutive frames of belt tear detected is less than 5, it is considered that the tear is small or that it is a defect of the belt itself. When the number of consecutive frames of tear is greater than 5 but less than 10, an alarm signal is triggered and the tear image and tear information are saved; when the number of consecutive frames of belt tear detected is greater than or equal to 10, a belt stop signal is triggered, and the PLC is interlocked to control the belt to stop.

[0091] (5) When the belt stop signal is triggered, the belt stop signal and the belt speed data collected by the speed measuring wheel are combined to calculate the precise location of the belt tear by integration.

[0092] In this embodiment, to accurately determine the location of a belt tear, the present invention employs a belt speed measuring wheel installed near the belt tear detection device to collect the belt's running speed v(t) at equal time intervals. During the process from when a belt tear is detected and the belt is stopped until it completely stops (its speed becomes zero), the belt speed collected during this period is integrated to obtain the precise location of the belt tear.

[0093] Equation (1) is the formula for calculating the location of the tear after the interlock belt stops when a belt tear is detected.

[0094]

[0095] Among them, S Teared S0 represents the initial position of the belt tear detection device after it stops, which is obtained from the initial measurement. t1 is the starting time of the belt stopping after the tear is detected. t2 is the time when the belt completely stops and its speed becomes zero. dt is the speed acquisition interval of the speed measuring wheel. By installing the belt tear detection device at the unloading hopper in the tippler room and at the return belt in the transfer room, the calculation of the tear position can be simplified.

[0096] The technical solution of this embodiment obtains the precise location of the belt tear by integrating the belt speed collected during the process from detecting a belt tear to controlling the belt to stop until the belt comes to a complete stop (the speed becomes zero).

[0097] Example 5

[0098] Based on the above embodiments, this embodiment provides a belt tear detection method, which further includes:

[0099] When the belt stop signal is triggered, the location of the belt tear is predicted based on the belt stop signal and a preset number of belt speed data.

[0100] In this embodiment, the belt tear detection method includes the following steps:

[0101] (1) Read the belt start / stop signal from the PLC using the OPC communication protocol;

[0102] (2) If the belt is stopped, tear detection is not performed; when the belt is running, the YOLOV5 deep learning vision algorithm detects through and non-through belt tear faults in real time.

[0103] (3) If a tear is detected, crop the torn area image, perform adaptive threshold segmentation of the torn image, calculate the tear gap width by white pixel statistics, set different thresholds, and determine different tear levels.

[0104] (4) Determination of consecutive frames of tear images: When the number of consecutive frames of belt tear detected is less than 5, it is considered that the tear is small or that it is a defect of the belt itself. When the number of consecutive frames of tear is greater than 5 but less than 10, an alarm signal is triggered and the tear image and tear information are saved; when the number of consecutive frames of belt tear detected is greater than or equal to 10, a belt stop signal is triggered, and the PLC is interlocked to control the belt to stop.

[0105] (5) When the belt stop signal is triggered, after acquiring a preset number of belt speed data, the predicted location of the belt tear is calculated, and the maintenance personnel are reminded to prepare for repair. When the belt stop signal is triggered, the precise location of the belt tear is calculated by integration, combining the belt stop signal and the belt speed data collected by the speed measuring wheel.

[0106] The technical solution of this embodiment obtains a preset number of belt speed data points, such as 3 to 5, to estimate the tear location after the belt stops and alert maintenance personnel to prepare for repairs. While the maintenance personnel are preparing, the system continues to calculate the precise location of the belt tear, enabling them to quickly address the tear.

[0107] Example 6

[0108] Figure 6 This is a schematic diagram of a belt tear detection device according to an embodiment of the present invention. Figure 6 As shown, based on the above embodiments, this embodiment provides a belt tear detection device, including:

[0109] An image acquisition device is used to acquire an image of a belt based on the image acquisition terminal of the above embodiment;

[0110] The tear detection device is used to detect whether the belt has been torn based on the tear detection module of the above embodiment; if the belt is torn and the number of image frames of the belt tear is greater than a preset number of frames, a belt stop signal is triggered.

[0111] The technical solution of this embodiment includes an image acquisition terminal and a tear detection module. The image acquisition terminal is located at the first and second ends of the belt to acquire images of the lower surface of the belt at the first end and the upper surface of the belt at the second end. The tear detection module detects whether there is a through tear based on the image of the lower surface of the belt at the first end and whether there is a non-through tear based on the image of the upper surface of the belt at the second end. The solution can quickly calculate the location of through tears and non-through tears based on the positions of the first and second ends.

[0112] Other technical features and beneficial effects of this embodiment can be found in the above embodiments, and will not be repeated here.

[0113] Example 7

[0114] Based on the above embodiments, this embodiment provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method of the above embodiments.

[0115] The aforementioned storage media can be flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, server, app store, etc.

[0116] Example 8

[0117] Based on the above embodiments, this embodiment provides a device including a processor and a memory, wherein the memory stores a computer program, and the processor executes the method of the above embodiments.

[0118] The processor may be implemented as an Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor, or other electronic components, and is used to execute the methods in the above embodiments.

[0119] Memory can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0120] Example 9

[0121] Based on the above embodiments, this embodiment provides an application example.

[0122] In the 1980s, foreign scholars proposed pre-embedding conductors in the belt and using electromagnetic induction to indirectly determine whether the belt was faulty by detecting whether the conductor had broken. However, this method required pre-embedding the conductors in the belt, which was costly and not very applicable to belt conveyor manufacturers already in production.

[0123] In 2011, Rong Feng et al. proposed X-ray fluoroscopy for detecting longitudinal tears in belts, indirectly determining whether a longitudinal tear had occurred by observing the difference in light intensity. This method is intuitive, reliable, and effective, but it did not consider the potential health hazards to operators from long-term use.

[0124] In 2008, Zhang Xi et al. used the difference image method to detect longitudinal tears in conveyor belts, detecting them by judging whether there is a tear in the material between two adjacent image frames. This technique is easily affected by dust in complex process environments, making it difficult to acquire high-quality images. Furthermore, it requires a huge amount of computation and has poor real-time performance. In addition, this method only detects material leakage after the belt tears; it cannot detect situations where the conveyor belt is not completely torn or where there is overlap at the tear point, where there is no leakage.

[0125] In 2015, Qiao Tiezhu et al. used visible light CCD and infrared CCD for binocular image acquisition. They proposed registering and fusing the corner and straight line feature images of conveyor belt tears acquired by infrared and visible light, and proposed a deep feature learning algorithm to classify and identify fault feature images. This method is a novel binocular vision online detection system that realizes real-time monitoring of longitudinal tears in conveyor belts and studies early warning strategies. However, this method requires extracting corner points from the images of the two cameras, then registering them, and then fusing features for learning, resulting in excessive computation and inaccurate detection results due to inaccurate corner point extraction.

[0126] Coal conveyor belts have high tensile strength, but their longitudinal tear resistance is only limited by the strength of the rubber itself, making them prone to longitudinal tears. Longitudinal tears can be caused by various factors, including detachment of equipment parts, loss of train car support rods, or the presence of hard objects such as chisels or I-beams. Belt conveyors are the lifeblood of coal transportation systems; longitudinal tears in these systems can result in significant direct and indirect losses.

[0127] Vision-based belt tear detection equipment is often installed between the upper and lower belts to detect whether there are penetrating tears on the cargo belt, but it cannot detect foreign objects scratching the belt or tear types that do not penetrate.

[0128] Traditional vision-based belt tear detection technology has limited analytical features and poor self-learning ability, making it difficult to adapt to the complex environment and characteristic changes in belt operation sites, such as changes in lighting, tear size and shape, and the influence of defects at belt joints and the belt itself.

[0129] Current tear detection algorithms lack the ability to determine whether a belt has a through-and-through tear or a non-through tear, the extent of the tear, or the size of the tear gap. This makes it difficult to assess the degree of belt defect and determine whether appropriate preventative measures should be taken.

[0130] After a belt tears, the location of the tear is often difficult to determine, requiring a considerable amount of time for manual searching to pinpoint the initial tear. Therefore, accurately locating the tear after it has broken is of significant practical importance.

[0131] To address the aforementioned issues, developing a belt longitudinal tear protection method capable of simultaneously detecting both through-and-through tears and non-through tears, stably and accurately detecting whether a belt has torn, determining the tear severity based on the tear width, and accurately locating the specific location of the belt tear after the belt has been stopped is of significant practical importance.

[0132] This invention provides a method for detecting tears in coal conveyor belts at coal ports, belonging to the fields of belt conveyor defect detection and safety protection.

[0133] To maximize belt utilization and minimize operator workload, relevant designs were developed, encompassing equipment installation processes and the analysis and application of different testing results, in conjunction with the production process flow.

[0134] (1) Simultaneous detection of penetrating and non-penetrating tears

[0135] This invention also installs a belt tear detection device at the unloading hopper of the port tipper room and at the return belt of the transfer room.

[0136] The belt tear detection device at the unloading hopper of the tipper room is installed between the cargo belt and the return belt. By shining an upward beam onto the lower surface of the cargo belt, it can detect through-type belt tear faults.

[0137] A through-type belt tear refers to a tear that extends from the top surface of the belt to the bottom surface. The top surface of the belt is the side that contacts the cargo. The bottom surface of the belt is the side that does not contact the cargo. The cargo belt is located above the return belt.

[0138] like Figure 2 As shown, the visual inspection terminal and the laser emission terminal work together to detect belt tear features. Due to the wide belt, two sets of inspection equipment are arranged over two consecutive spans, one for inspecting the left side of the belt and the other for inspecting the right side.

[0139] The belt tear detection device at the return belt of the transfer room is installed below the return belt and shines upwards. It is used to detect the upper surface of the cargo belt and can be used to identify non-penetrating belt tears.

[0140] like Figure 3 As shown, due to the wide belt, two sets of belt tear detection equipment are installed on two consecutive spans, one for detecting the left belt and the other for detecting the right belt.

[0141] (2) Deep learning algorithm YOLOv5

[0142] This study employs the advanced deep learning vision algorithm YOLOv5 for belt tear detection. This algorithm is robust, effectively overcoming the effects of changes in lighting, pose, and scale. Furthermore, it can be continuously enriched and optimized with the dataset to detect various tear types. The YOLOv5 model is publicly released by Ultralytics. Based on the YOLOv3 model, the YOLOv5 model comprises four versions: YOLOv5s, YOLOv5m, YOLOv5l, and YOLOv5x. The YOLOv5 model consists of a backbone network, a neck network, and a head network.

[0143] (3) Determination of tear width threshold

[0144] If a belt tear is detected, the image at the tear location is cropped, and then an adaptive thresholding segmentation method is used to segment the torn image. The tear is set to white, and the belt is set to black. Then, the values ​​of 10 white pixels in the Y direction are counted, and the average value is used to represent the width of the tear.

[0145] Two thresholds are set for different tear widths: a first threshold T1 and a second threshold T2. When the tear width Width < the first threshold T1, the tear is judged as a level 1 alarm; when the first threshold T1 < the tear width Width < the second threshold T2, the tear is judged as a level 2 alarm; and when the tear width Width > the second threshold T2, the tear is judged as a level 3 alarm.

[0146] (4) Precise location of tear

[0147] To accurately determine the location of a belt tear, this invention employs a belt speed measuring wheel installed near the belt tear detection device to collect the belt's running speed v(t) at equal time intervals. During the process from when a belt tear is detected and the belt is stopped until it completely stops (its speed becomes zero), the collected belt speed data is integrated to obtain the precise location of the belt tear.

[0148] Equation (1) is the formula for calculating the location of the tear after the interlock belt stops when a belt tear is detected.

[0149]

[0150] Among them, S Teared S0 represents the initial position of the belt tear detection device after it stops, which is obtained from the initial measurement. t1 is the starting time of the belt stopping after the tear is detected. t2 is the time when the belt completely stops and its speed becomes zero. dt is the speed acquisition interval of the speed measuring wheel. By installing the belt tear detection device at the unloading hopper in the tippler room and at the return belt in the transfer room, the calculation of the tear position can be simplified.

[0151] The belt tear detection method of the present invention includes the following steps:

[0152] (1) Read the belt start / stop signal from the PLC using the OPC communication protocol;

[0153] (2) If the belt is stopped, tear detection is not performed; when the belt is running, the YOLOV5 deep learning vision algorithm detects through and non-through belt tear faults in real time.

[0154] (3) If a tear is detected, crop the torn area image, perform adaptive threshold segmentation of the torn image, calculate the tear gap width by white pixel statistics, set different thresholds, and determine different tear levels.

[0155] (4) Determination of consecutive frames of tear images: When the number of consecutive frames of belt tear detected is less than 5, it is considered that the tear is small or that it is a defect of the belt itself. When the number of consecutive frames of tear is greater than 5 but less than 10, an alarm signal is triggered and the tear image and tear information are saved; when the number of consecutive frames of belt tear detected is greater than or equal to 10, a belt stop signal is triggered, and the PLC is interlocked to control the belt to stop.

[0156] (5) When the belt stop signal is triggered, the belt stop signal and the belt speed data collected by the speed measuring wheel are combined to calculate the precise location of the belt tear by integration.

[0157] (6) When the belt stop signal is triggered, after obtaining a preset number of belt speed data, the predicted location of the belt tear is calculated and the maintenance personnel are reminded to prepare for repair.

[0158] The technical solution of this embodiment has the following beneficial effects.

[0159] (1) YOLOv5's deep learning algorithm is more robust than traditional vision algorithms, can identify various types of tearing features, and can continuously expand the dataset to improve the algorithm's accuracy and precision.

[0160] (2) Install a belt tear detection device at the unloading hopper in the tipper room and at the return belt in the transfer room. This installation and detection method can detect both through-tears and non-through-tears of the belt at the same time, so as to achieve comprehensive protection against longitudinal belt tears and timely shutdown after tearing.

[0161] (3) It can identify and judge the width of the tear. By setting different thresholds, the tear level can be evaluated, which helps staff to make the correct handling measures.

[0162] (4) Accurate judgment of the belt tear location helps to quickly find the specific location of the tear after the machine stops due to the tear, and to eliminate the fault and resume production more quickly.

[0163] In the embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and flowcharts, as well as combinations of blocks in the block diagrams and flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0164] It should be noted that, in this invention, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element limited by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0165] While the embodiments disclosed in this invention are as described above, the above content is merely for the purpose of facilitating understanding of this invention and is not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and changes in form and detail of the implementation without departing from the spirit and scope disclosed in this invention; however, the scope of patent protection of this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A belt tear detection method based on a belt tear detection system, characterized in that, The belt tear detection system includes: An image acquisition terminal is set at the first end point and the second end point of the belt, wherein an image of the lower surface of the cargo belt is acquired at the first end point and an image of the upper surface of the cargo belt is acquired at the second end point; The tear detection module detects whether there is a through-tear on the belt based on the image of the lower surface of the cargo belt acquired at the first end of the belt, and detects whether there is a non-through-tear on the belt based on the image of the upper surface of the cargo belt acquired at the second end of the belt. The belt tear detection method includes: The image acquisition terminal acquires an image of the belt; The tear detection module is used to detect whether the belt has torn. If the belt tears and the number of image frames of the belt tear exceeds the preset number of frames, a belt stop signal will be triggered. When the belt stop signal is triggered, based on the belt stop signal and a preset number of belt speed data, the tear location after the belt stops is predicted, and the maintenance personnel are reminded to prepare for repair. During the preparation process, the belt speed data is integrated to calculate the precise location of the belt tear. The preset number is greater than or equal to 3 and less than or equal to 5. The first endpoint includes the unloading hopper of the port tippler room, and the second endpoint includes the return conveyor belt of the transfer room. The belt includes a cargo belt, the lower surface includes the surface of the cargo belt in contact with the idler roller, and the upper surface includes the surface of the cargo belt in contact with the cargo. The image acquisition terminal includes a laser emission sub-terminal and a visual inspection sub-terminal; Two image acquisition terminals are arranged side by side at the first and second ends of the belt.

2. The belt tear detection method based on a belt tear detection system according to claim 1, characterized in that, The method further includes: When a belt tear is detected, an adaptive threshold method is used to segment the torn area of ​​the belt, and a morphological method is used to calculate the size of the tear and determine the tear level.

3. A belt tear detection device based on the belt tear detection method of any one of claims 1 to 2, characterized in that, include: Image acquisition device, used to acquire images of the belt based on an image acquisition terminal; A tear detection device is used to detect whether a belt has torn, based on a tear detection module. If the belt tears and the number of image frames showing the tear exceeds the preset number of frames, a belt stop signal will be triggered.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the belt tear detection method based on the belt tear detection system as described in any one of claims 1 to 2.

5. An electronic device comprising a processor and a memory, characterized in that, The memory stores a computer program, and when the processor executes the computer program, it implements the belt tear detection method based on the belt tear detection system as described in any one of claims 1 to 2.

Images