Conveyor Belt Health Diagnosis Method, Device, Server, and Storage Medium

Abstract

The present application discloses a conveyor belt health diagnosis method, device, server, and storage medium, which relates to the field of belt detection. It receives belt data uploaded by an X-ray detector on the conveyor belt and performs classification and screening to obtain the coordinate information of abnormal detection areas; performs clustering detection on adjacent abnormal detection areas on the conveyor belt, calculates the regional spacing and regional overlap of each abnormal detection area; clusters and merges abnormal detection areas with a regional spacing less than a set threshold to obtain merged abnormal areas; calculates the abnormal defect area and quantity in the merged abnormal areas, determines the belt health level according to the historical abnormal defect area and quantity, and issues an alarm instruction when the belt wear is lower than the health level threshold. This solution uses a cloud server to uniformly detect all conveyor belts, realizes real-time monitoring of the worn areas of the conveyor belt, evaluates the belt health status according to the changes in the worn area and the number of blocks, gives an alarm in a timely manner, and realizes the maximization of resource utilization and the reduction of cost investment.

Description

Technical Field

[0001] This application relates to the field of belt detection, and particularly to a conveyor belt health diagnosis method, device, server, and storage medium. Background Art

[0002] The conveyor belt, also known as the transport belt, is a rubber, fiber, and metal composite product or a plastic and fabric composite product that plays a role in carrying and transporting materials in the belt conveyor. The conveyor belt is widely used in industries such as cement, coking, metallurgy, chemical industry, and steel in occasions with short conveying distances and small conveying volumes.

[0003] During the long-term use of the conveyor belt, it is inevitable to encounter longitudinal tearing, cracking, and surface damage of the conveyor belt, which are called conventional damages of the conveyor belt. Based on this, on-site maintenance personnel need to regularly check the conveyor belt situation, and the health diagnosis of the conveyor belt is particularly important. If any abnormality is found, the conveyor belt should be repaired and replaced according to technical requirements to ensure the normal operation of production operations.

[0004] In related technologies, in large industrial sites and harsh industrial and mining environments, a large amount of manpower needs to be invested in frequent inspections, and the inspection efficiency is not high, and it is also easy to miss inspections. Especially in some occasions with low bearing capacity and accuracy, if the conveyor belt is replaced as long as wear is found, it will surely increase the cost investment, and there is a lack of a reasonable wear degree detection and warning replacement mechanism. Summary of the Invention

[0005] This application provides a conveyor belt health diagnosis method, device, equipment, and storage medium to solve the problem of low detection efficiency of conveyor belt wear.

[0006] On the one hand, this application provides a conveyor belt health diagnosis method, and the method includes:

[0007] Receiving belt data collected and uploaded by a non-destructive X-ray detector on the conveyor belt through the MQTT protocol, and classifying and screening the belt data to obtain the coordinate information of the abnormal detection area; the abnormal detection area is a rectangular area calibrated according to the belt data;

[0008] Performing clustering detection on adjacent abnormal detection areas on the same conveyor belt, and calculating the area distance and area overlap of each abnormal detection area;

[0009] Clustering and merging abnormal detection areas with an area distance less than a set threshold to obtain a merged abnormal area;

[0010] Calculating the abnormal defect area and quantity in the merged abnormal area, and determining the belt health level according to the historical abnormal defect area and quantity. When the belt wear is lower than the health level threshold, a warning instruction is sent to the belt target detector.

[0011] On the other hand, the present application provides a conveyor belt health diagnosis device, which includes:

[0012] A screening module, configured to receive the belt data collected and uploaded by a non-destructive X-ray detector on the conveyor belt through the MQTT protocol, and classify and screen the belt data to obtain the coordinate information of the abnormal detection area; the abnormal detection area is a rectangular area calibrated according to the belt data;

[0013] A detection module, configured to perform clustering detection on adjacent abnormal detection areas on the same conveyor belt, and calculate the area distance and area overlap of each abnormal detection area;

[0014] A clustering module, configured to cluster and merge the abnormal detection areas with an area distance less than a set threshold to obtain a merged abnormal area;

[0015] A judgment module, configured to calculate the abnormal defect area and quantity in the merged abnormal area, determine the belt health level according to the historical abnormal defect area and quantity, and send an alarm instruction to the belt target detector when the belt wear is lower than the health level threshold.

[0016] On the other hand, the present application provides a server, which includes a processor and a memory. At least one instruction, at least one program, a code set or an instruction set is stored in the memory, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the conveyor belt health diagnosis method described in the above aspect.

[0017] On the other hand, the present application provides a computer-readable storage medium, in which at least one instruction, at least one program, a code set or an instruction set is stored, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement the conveyor belt health diagnosis method described in the above aspect.

[0018] The beneficial effects brought by the technical solution provided by the embodiments of the present application at least include: installing a non-destructive X-ray detector on the conveyor belt to collect and upload belt data to the cloud server in real time. The cloud server stores and classifies the belt data of all conveyor belts in the system, screens the coordinate information of the abnormal detection areas, and then performs clustering detection on them. The abnormal detection areas that overlap or have a regional spacing less than the set threshold are clustered to achieve the classification and evaluation of wear blocks in the same wear area. Then, it compares with the wear situation in this wear area in the historical data, determines the belt health level according to the area and quantity, and triggers the warning mechanism in a timely manner. This operation method can achieve unified cloud processing, accurately measure the belt wear situation in the same wear area within a specific time, conduct health assessment according to the wear trend, without the need for workers to conduct inspections and replacements. Moreover, this warning mechanism can achieve directional monitoring, timely warning, slow down the belt replacement frequency, and reduce equipment downtime and labor costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a flowchart of the conveyor belt health diagnosis method provided by the embodiments of the present application;

[0020] Figure 2 is a schematic diagram of the interface after visualizing the abnormal detection area;

[0021] Figure 3 is a schematic diagram of the interface for clustering and merging overlapping abnormal detection areas to generate a merged abnormal area;

[0022] Figure 4 is a schematic diagram of calculating the regional distance in the coordinate system;

[0023] Figure 5 is a schematic diagram of matching the target abnormal defect area according to the merged abnormal detection area;

[0024] Figure 6 shows a schematic structural diagram of the conveyor belt health diagnosis device provided by the embodiments of the present application;

[0025] Figure 7 shows a structural block diagram of the server provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] To make the objectives, technical solutions, and advantages of the present application clearer, the following will further describe the embodiments of the present application in detail with reference to the accompanying drawings.

[0027] As used herein, "a plurality of" means two or more. "And / or" describes the relationship between associated objects and indicates that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally indicates an "or" relationship between the associated objects before and after.

[0028] Figure 1 It is a flowchart of the conveyor belt health diagnosis method provided by an embodiment of the present application, including the following steps:

[0029] Step 101: Receive the belt data collected and uploaded by the non-destructive X-ray detector on the conveyor belt through the MQTT protocol, and classify and screen the belt data to obtain the coordinate information of the abnormal detection area;

[0030] This method is applied to the server in the conveyor belt health diagnosis system, and the system includes several conveyor belt machines. A non-destructive X-ray detector is set on each conveyor belt machine. The detector is set with a unique ID identifier and establishes communication with the server through a wired or wireless communication module to upload the real-time collected conveyor belt data to the server.

[0031] The abnormal detection area is determined according to the belt thickness data measured by X-ray. When the thickness data does not match the preset value, it indicates wear. All the uploaded data is stored in the Apache Cassandra large-scale distributed database through the MQTT protocol.

[0032] The abnormal detection area is a rectangular area calibrated according to the actual belt defect data. Refer to Figure 2 , where o1, o2, o3, and o4 visually list 4 calibrated rectangular abnormal detection areas respectively. Its principle is to frame according to the image formed by the actual defect data. Since the worn areas are almost all irregular, in order to facilitate classification, direct framing processing is performed on them. Framing is the result of selecting from a series of defect points according to the coordinate value size, and the coordinate values of the four right-angled sides (points) of the rectangle are obtained. The present application only uses two coordinate points of the rectangular abnormal detection area, that is, the two end points on the same diagonal line. For example, from Figure 2 the two coordinates of point a at the upper left corner and point b at the lower right corner described in C1.

[0033] It should be noted that Figure 2 The visualized data presented for easy understanding. In fact, the server only processes the sensor thickness data and makes judgments based on the coordinate information of the thickness data. The coordinate system is established according to the structure of the conveyor belt and the position of the non-destructive X-ray detector. The server only performs defect identification and judgment on the coordinate information.

[0034] Step 102: Cluster and detect adjacent anomaly detection regions on the same conveyor belt, and calculate the regional spacing and regional overlap of each anomaly detection region;

[0035] The server receives the belt data uploaded by all non-destructive X-ray detectors in the system, determines the affiliated conveyor belt according to the ID identification, and performs clustering detection on adjacent anomaly detection regions of the same conveyor belt. For example Figure 2 the four rectangular detection regions in. The regional spacing is the distance between two rectangular regions. D14 represents the distance between regions C1 and C4. The overlap situation indicates whether two detection regions overlap, that is, whether the regional spacing is greater than 0. C1 and C2 belong to the overlapping regions.

[0036] Because the server needs to process a large amount of data, it will not establish a coordinate system and perform image rendering to judge the overlap situation in the cloud, but completely calculate and judge based on the coordinate information of the uploaded belt data.

[0037] Step 103: Cluster and merge the anomaly detection regions with regional spacing less than the set threshold to obtain merged anomaly regions;

[0038] The clustering and merging is the result of fully considering the belt wear situation. Because the worn areas are usually scattered, especially after one area is worn, the stress on the surrounding areas will change and cause a chain reaction, manifested as the continuous increase of wear in this area or the continuous generation of new wear blocks around it. When the distance between two wear blocks is large, they need to be aggregated, which is represented as aggregating into a larger wear block.

[0039] For this type of anomaly detection regions that need to be aggregated, it is judged by setting the regional spacing threshold. For example, when the anomaly detection regions already overlap or the regional spacing is less than 50 pixel units, they are aggregated. The aggregation is also judged by coordinate values, with the smallest rectangular range containing two small anomaly detection regions as the standard. In this application, the rectangular right-angle intersection points of the two anomaly detection regions are connected by lines, and the maximum value of the line connection distance is determined as the matrix diagonal of the merged anomaly region, and the merged anomaly region is constructed. Its essence is to select the maximum and minimum values from the two rectangular right-angle coordinate points to determine the diagonal endpoint coordinates of the merged anomaly region.

[0040] For example Figure 2 for the two intersecting anomaly defect regions C1 and C2 in, take the minimum values of the x-axis and y-axis of the upper-left pixel point coordinates of the two rectangular regions and the maximum values of the x-axis and y-axis of the lower-right pixel point coordinates;

[0041] X1 = MIN(x1, x2)

[0042] Y1 = MIN(y1, y2)

[0043] X1′ = MAX(x1′, x2′)

[0044] Y1′ = MAX(y1′, y2′)

[0045] The diagonal endpoint coordinates of the merged abnormal area after clustering are represented as (X1, Y1) and (X1′, Y1′). Figure 3 It is based on Figure 2 After clustering and merging the two overlapping abnormal detection areas C1 and C2 in [reference], the merged abnormal area is obtained. After merging, a new bounding rectangle will be formed, so the area distance between it and other surrounding abnormal detection areas will also change. Further aggregation is still required after overlap or the distance is less than the set threshold.

[0046] Step 104: Calculate the area and quantity of abnormal defects in the merged abnormal area, and determine the belt health level according to the historical area and quantity of abnormal defects. When the belt wear is lower than the health level threshold, send an alarm instruction to the belt target detector.

[0047] When the non-destructive X-ray detector collects belt data, it will upload data with coordinate information according to the belt transmission speed and time. This process involves cyclic collection. For example, in the belt data collected in two time periods, yesterday and today, the wear trend is determined by comparing the front and back of a wear block or the area of interest at the end to judge the belt health level. This process specifically includes judging based on the change trend of the area within the rectangular bounding area and the change trend of the number of wear blocks. Assuming that the health level is lower than the set threshold, the conveyor belt that needs to be alarmed and replaced is determined according to the ID identification of the target detector. When the wear area and the number of wear blocks do not endanger normal transportation, the server does not trigger the alarm mechanism.

[0048] In summary, in this application, a non-destructive X-ray detector is installed on the conveyor belt to collect and upload belt data to the cloud server in real time. The cloud server stores and classifies and detects the belt data of all conveyor belts in the system, filters the coordinate information of the abnormal detection areas, and then performs clustering detection on them. The abnormal detection areas that overlap each other or have an area distance less than the set threshold are clustered to achieve the classification and evaluation of wear blocks in the same wear area. Then, the wear conditions in this wear area in the historical data are compared, and the belt health level is determined according to the area and quantity, and the warning mechanism is triggered in a timely manner. This operation method can achieve unified processing in the cloud, accurately measure the belt wear conditions in the same wear area within a specific time, conduct health assessment according to the wear trend, without the need for workers to conduct inspections and replacements, and this warning mechanism can achieve directional monitoring, timely warning, slow down the belt replacement frequency, and reduce equipment downtime and labor costs.

[0049] Although it is easy to judge whether there is an overlap in the anomaly detection area from a visualization perspective, for a cloud server, it is unrealistic to visualize a large amount of big data. Therefore, it is necessary to simply judge the overlap situation through coordinate information. Therefore, for calculating the distance between regions and the overlap situation of each anomaly detection area, it can be achieved through the following steps:

[0050] A. Obtain the two coordinate values of the diagonal endpoints in two adjacent anomaly detection areas, and calculate the coordinates of the center point of the defect according to the diagonal coordinate values;

[0051] Refer to Figure 2 and Figure 4 As shown in the content, the diagonal coordinate values of the first anomaly detection area c1 are a(x1, y1) and b(x1′, y1′); the center point coordinates o1 are expressed as follows:

[0052]

[0053] The diagonal coordinate values of the second anomaly detection area c2 are c(x2, y2) and d(x2′, y2′); the center point coordinates o2 are expressed as follows:

[0054]

[0055] Calculate the distances dx and dy between the center points of the two rectangles in the x-axis and y-axis directions respectively;

[0056] dx = |o2_x - o1_x|

[0057] dy = |o2_y - o1_y|

[0058] The calculation process for c3 and c4 is the same as above and will not be elaborated here.

[0059] B. Judge the distance between regions and the overlap situation according to the diagonal coordinate values of the two anomaly detection areas and the axial distance between the center points.

[0060] It should be emphasized that in the computer field, selection by the same direction and the same angle is mostly used. In this way, there will be various manifestations when calculating the distance between regions. That is, within a known coordinate system, when there is cross-coverage within the x-axis or y-axis range, the parallel distance can be directly calculated, but when there is no cross-coverage within the x-axis and y-axis ranges, the straight-line distance needs to be calculated according to the nearest right-angle point. Specifically as follows:

[0061] Calculate the average width width and average height height of the rectangles of the two anomaly detection areas respectively; expressed as follows:

[0062]

[0063]

[0064] width represents the average width of the two rectangular frames of the anomaly detection regions, and height represents the average height of the two rectangles. The introduction of the absolute value is mainly to eliminate the positive and negative value differences caused by different positions of the coordinate system establishment in different scenarios.

[0065] 1. When dx < width and dy >= height, it indicates that the anomaly detection regions c1 and c2 do not intersect or overlap, but there is an intersection in the x-axis direction. The region spacing min_dist is expressed as follows:

[0066] min_dist = dy - height

[0067] See Figure 2 The relationship between C3 and C4 in

[0068] is non-intersecting, but there is an intersection in the x-axis direction; D34 is min_dist.

[0069] min_dist = dx - width

[0070] See Figure 2 The relationships between C1 and C4, and C2 and C3 in

[0071] are non-intersecting, but there are intersections in the y-axis direction. The region spacings min_dist are D23 and D14 respectively.

[0072]

[0073] See Figure 2 The relationship between C1 and C3 in

[0074] is non-intersecting, and there is no intersection within the x and y axes. The region spacing is the straight-line distance between the lower-right and upper-left corner points. min_dist is D13, and this value can be calculated using the Pythagorean theorem.

[0075] See Figure 2 The relationship between C1 and C2 in

[0076] After determining the regional distances of each abnormal defect area, it is necessary to determine the health condition of the belt. However, this process requires finding the historical data of the worn area from the database, that is, the historical abnormal defect area. Only the same area has the meaning of comparison before and after. In a real scenario, the sliding friction between the conveyor belt and the machine will cause a displacement difference in the data before and after, that is, the range represented by the coordinate values of the two worn areas is offset. At this time, it is necessary to calculate the similarity of the historical abnormal defect area by matching. For example, 10 wear points were recorded on the conveyor belt yesterday and stored in the database. When measuring today, the worn area located according to the uploaded data needs to be matched with the 10 wear points yesterday to find the target abnormal defect area. The specific operation of matching is determined by calculating the defect similarity. The following are included:

[0077] a. Obtain the diagonal coordinate values of the merged abnormal detection area as A(xa, ya) and B(xa′, ya′); the diagonal coordinate values of the target abnormal area to be matched are C(xb, yb) and D(xb′, yb′);

[0078] Figure 5 Figure 7 is a matching schematic diagram of the merged abnormal detection area CA (thick line) and the candidate abnormal defect area CB (thin line) of the data center. There will inevitably be an overlapping area during the matching process, otherwise it will be determined as a newly added worn area and added to the database.

[0079] b. Calculate the width-to-height similarity ratio aspectRatioSimilarity, position coincidence degree positionSimilarity, and angle difference angleDifference based on the diagonal coordinate values of the two rectangular areas;

[0080] The calculation of aspectRatioSimilarity is as follows:

[0081]

[0082]

[0083]

[0084] Among them, AR and BR respectively represent the width-to-height ratios of CA and CB, and aspectRatioSimilarity represents the width-to-height similarity ratio of the two rectangular areas; the width-to-height ratio of the rectangle is the cosine value of the trigonometric function (the included angle between the AB diagonal and the side length), and the numerical table of the width-to-height similarity ratio is used to characterize the similarity of the two rectangles. When the shapes of the two rectangles are closer, the value is closer to 1.

[0085] The calculation of positionSimilarity is as follows:

[0086] xOverlap = max(0, min(xa + |xa′ - xa|, xb′)) - max(xa, xb)

[0087] yOverlap = max(0, min(yb + 1|yb′ - yb1|, yb′)) - max(ya, yb)

[0088] overlapArea = xOverlap * yOverlap

[0089] aArea = |xa′ - xa| * |ya′ - ya|

[0090] bArea = |xb′ - xb| * |yb′ - yb|

[0091] minArea = min(aArea, bArea)

[0092]

[0093] Among them, xOverlap represents the width of the overlapping area, yOverlap represents the height of the overlapping area, overlapArea represents the area of the overlapping area (the area of the small rectangle of the BC diagonal), aArea and bArea respectively represent the areas of the CA and CB rectangles, and minArea represents the smaller part of the areas of the two rectangles ( Figure 5 In the case where the area of the CB rectangle is smaller than that of the CA rectangle, minArea is the CB rectangle), positionSimilarity represents the proportion of the overlapping area in the smaller area (the ratio of the area of the small rectangle of the BC diagonal to the area of the CB rectangle). positionSimilarity is used to characterize the degree of coincidence between the current overlap and the previous detection area. The higher the degree of coincidence, the closer the value is to 1.

[0094] The angle difference is calculated as follows:

[0095] angleDifference = |AR - BR|

[0096] The value of angleDifference is characterized as the difference in cosine values. The smaller the difference, the closer the angle between the diagonal and the side length of the rectangle.

[0097] c. The defect similarity similarity is calculated based on the mean of the width-to-height similarity ratio, position coincidence degree, and angle difference;

[0098] similarity is calculated as follows:

[0099]

[0100] The purpose of taking the average of similarity is mainly to improve accuracy and avoid excessive error in a single parameter.

[0101] When the similarity value is greater than the set threshold (for example, set to 0.8), it is determined that the merged abnormal detection area and the target abnormal area are the same wear defect block, and the merged abnormal detection area is numbered and stored in the database. In the subsequent matching process, the new CA rectangle is used as the matching area.

[0102] After determining the target abnormal area for matching, the belt health level can be further determined according to the abnormal defect area and quantity. It can include the following:

[0103] 1. Calculate the difference in the area of the block before and after aArea and bArea, and the number of defect blocks included in aArea and bArea:

[0104] 2. Determine the belt health level according to the change speed of the area difference of the block before and after, and the change speed of the number of defect blocks; the belt health level is negatively correlated with the change in wear area and the change in the number of defect blocks.

[0105] In some other possible implementation manners, the health level can be divided in the following way:

[0106] 1. The number of abnormal skeleton wires;

[0107] Slight: The number <= 1;

[0108] Warning: 1 < the number < 3;

[0109] Danger: The number >= 3;

[0110] 2. The wear length of the cover rubber (DPI):

[0111] Slight: The length <= 100;

[0112] Warning: 100 < the length < 200;

[0113] Danger: The number >= 200;

[0114] 3. Classify the level according to the growth trend of the number or length of regional abnormalities between the previous and current dates;

[0115] Slight: The growth <= 20%;

[0116] Warning: 20% < the growth < 50%;

[0117] Danger: The growth > 50%;

[0118] The above several judgment methods can reasonably evaluate the health degree of belt wear and trigger the warning mechanism in a timely manner. For example, when the number of a certain joint of a conveyor belt / a certain section of the conveyor belt reaches 3 dangerous levels, it is determined that the health status of the entire conveyor belt is poor, and manual intervention for maintenance or replacement of a new conveyor belt is required.

[0119] Through the examples of the present invention, the conveyor operation and maintenance personnel can timely understand the health status of the conveyor belt, and there is no need for on-site inspections. When the replacement conditions are met, the machine is shut down for replacement, leaving more response time for the operation and maintenance personnel and reducing the economic losses caused by the conveyor belt.

[0120] Figure 6 The structural schematic diagram of the conveyor belt health diagnosis device provided by the embodiment of the present application is shown. The device includes:

[0121] A screening module 610, configured to receive the belt data collected and uploaded by the non-destructive X-ray detector on the conveyor belt through the MQTT protocol, and classify and screen the belt data to obtain the coordinate information of the abnormal detection area; the abnormal detection area is a rectangular area calibrated according to the belt data;

[0122] A detection module 620, configured to perform clustering detection on adjacent abnormal detection areas on the same conveyor belt, and calculate the area distance and area overlap of each abnormal detection area;

[0123] A clustering module 630, configured to cluster and merge the abnormal detection areas with the area distance less than the set threshold to obtain a merged abnormal area;

[0124] A judgment module 640, configured to calculate the abnormal defect area and quantity in the merged abnormal area, determine the belt health level according to the historical abnormal defect area and quantity, and send an alarm instruction to the belt target detector when the belt wear is lower than the health level threshold.

[0125] In addition, the present application also provides a server, which includes a processor and a memory. At least one instruction, at least one program, a code set or an instruction set is stored in the memory. The at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the conveyor belt health diagnosis method described in the above aspects.

[0126] In addition, the present application also provides a computer-readable storage medium. At least one instruction, at least one program, a code set or an instruction set is stored in the readable storage medium. The at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the conveyor belt health diagnosis method described in the above aspects.

[0127] The conveyor belt health diagnosis device provided by the embodiments of the present application can be applied to the conveyor belt health diagnosis method provided in the above embodiments. For relevant details, refer to the above method embodiments. The implementation principles and technical effects are similar and will not be elaborated here.

[0128] It should be noted that when the conveyor belt health diagnosis device provided by the embodiments of the present application performs the snap gauge operation, only the above-mentioned division of each functional module / functional unit is used for illustration. In actual applications, the above functions can be allocated to different functional modules / functional units according to needs, that is, the internal structure of the conveyor belt health diagnosis device is divided into different functional modules / functional units to complete all or part of the functions described above. In addition, the implementation manner of the conveyor belt health diagnosis method provided by the above method embodiments and the implementation manner of the conveyor belt health diagnosis device provided by this embodiment belong to the same concept. For the specific implementation process of the conveyor belt health diagnosis device provided by this embodiment, refer to the above method embodiments and will not be elaborated here.

[0129] Figure 7 The block diagram of the server provided by an exemplary embodiment of the present application is shown. The server may include, but is not limited to, a processor and a memory. Among them, the processor and the memory can be connected through a bus or other means. Among them, the processor can be a central processing unit (CPU). The processor can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, graphics processing units (GPUs), embedded neural network processors (NPUs) or other dedicated deep learning co-processors, discrete gate or transistor logic devices, discrete hardware components and other chips, or a combination of the above types of chips.

[0130] The processor may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. The processor may be implemented in at least one of the following hardware forms: DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 1701 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the wake state, also known as the CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor may be integrated with a GPU (Graphics Processing Unit), and the GPU is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor may further include an AI (Artificial Intelligence) processor, which is used to process computational operations related to machine learning.

[0131] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the methods in the above embodiments of the present application. By running the non-transitory software programs, instructions, and modules stored in the memory, the processor can execute various functional applications and data processing of the processor, that is, implement the methods in the above method embodiments. The memory may include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created by the processor, etc. In addition, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely set relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above networks include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0132] The embodiments of the present application also disclose a computer-readable storage medium. Specifically, the computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the methods in the above method embodiments are implemented. Those skilled in the art can understand that to implement all or part of the processes in the above method embodiments of the present application, it can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the above method embodiments. Among them, the storage medium can be a magnetic disk, an optical disc, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above types of memories.

[0133] This specific embodiment is only an interpretation of the present invention and is not a limitation thereof. After reading this specification, those skilled in the art can make modifications to this embodiment that do not contribute creatively as needed, but as long as they are within the scope of the claims of the present invention, they are protected by the patent law.

Claims

1. A conveyor belt health diagnosis method, characterized in that, the method includes: Receiving the belt data collected and uploaded by the non-destructive X-ray detector on the conveyor belt through the MQTT protocol, and classifying and screening the belt data to obtain the coordinate information of the abnormal detection area; the abnormal detection area is a rectangular area calibrated according to the belt data; Performing clustering detection on adjacent abnormal detection areas on the same conveyor belt, and calculating the area spacing and area overlap of each abnormal detection area; Clustering and merging the abnormal detection areas with the area spacing less than the set threshold to obtain a merged abnormal area; Calculating the abnormal defect area and quantity in the merged abnormal area, and determining the belt health level according to the historical abnormal defect area and quantity. When the belt wear is lower than the health level threshold, sending an alarm instruction to the belt target detector.

2. The conveyor belt health diagnosis method according to claim 1, characterized in that, the calculation of the area spacing and area overlap of each abnormal detection area includes: Obtaining two coordinate values of the diagonal endpoints of the rectangle in two adjacent abnormal detection areas, and calculating the coordinates of the defect center point according to the diagonal coordinates; Among them, the diagonal coordinate values of the first abnormal detection area c1 are a(x1, y1) and b(x1′, y1′); the center point coordinates o1 are expressed as follows: The diagonal coordinate values of the second abnormal detection area c2 are c(x2, y2) and d(x2′, y2′); the center point coordinates o2 are expressed as follows: Calculating the distances dx and dy of the centers of the two rectangles in the x-axis and y-axis directions respectively; dx = |o2_x - o1_x| dy = |o2_y - o1_y| Judging the area spacing and overlap situation according to the diagonal coordinate values of the two abnormal detection areas and the axial distance between the center points.

3. The conveyor belt health diagnosis method according to claim 2, characterized in that, the judgment of the overlap situation according to the diagonal coordinate values of the two abnormal detection areas and the axial distance between the center points includes: Calculating the average width width and average height height of the rectangles of the two abnormal detection areas respectively; expressed as follows: When dx < width and dy >= height, it indicates that c1 and c2 do not intersect and overlap, but there is an intersection in the x-axis direction, and the area spacing min_dist is expressed as follows: min_dist = dy - height When dx >= width and dy < height, it indicates that c1 and c2 do not intersect and overlap, but there is an intersection in the y-axis direction, and the area spacing min_dist is expressed as follows: min_dist = dx - width When dx >= width and dy >= height, it indicates that c1 and c2 do not intersect and overlap, and there is no intersection in the x and y axis directions, and the area spacing min_dist is expressed as follows: When dx < width and dy < height, it indicates that c1 and c2 intersect, and the area spacing min_dist is 0.

4. The conveyor belt health diagnosis method according to claim 1, characterized in that, Clustering and merging the abnormal detection regions with a regional spacing less than a set threshold to obtain a merged abnormal region, including: When two abnormal detection regions intersect and overlap, or the regional spacing is less than 50 pixel unit values, connect the rectangular right-angle intersection points in the two abnormal detection regions, determine the maximum value of the connection distance as the matrix diagonal of the merged abnormal region, and construct the merged abnormal region; the merged abnormal region includes two abnormal detection regions with a regional spacing less than the set threshold.

5. The conveyor belt health diagnosis method according to claim 4, characterized in that determining the belt health level according to the historical abnormal defect area and quantity includes: matching the target abnormal defect region of the conveyor belt from the database according to the coordinate information included in the merged abnormal region; the database includes the coordinate information and region numbers of historical abnormal defect regions, and one region number corresponds to one wear defect block of the conveyor belt; calculating the defect similarity according to the overlapping region between the merged abnormal region and the historical abnormal defect region; When the defect similarity is greater than the set threshold, indicating that the merged abnormal region and the historical abnormal defect region are the same wear defect block on the conveyor belt, and determining the belt health level according to the difference in their defect areas and the number of included defects.

6. The conveyor belt health diagnosis method according to claim 5, characterized in that calculating the defect similarity according to the overlapping region between the merged abnormal region and the historical abnormal defect region includes: Obtain the diagonal coordinate values of the merged abnormal detection region as A(xa, xa) and B(xa′, ya′); the diagonal coordinate values of the matching target abnormal region are C(xb, yb) and D(xb′, yb′); Calculate the width-to-height similarity ratio aspectRatioSimilarity, position coincidence degree positionSimilarity, and angle difference angleDifference based on the diagonal coordinate values of the two rectangular regions; Calculate the defect similarity similarity according to the mean value of the width-to-height similarity ratio, position coincidence degree, and angle difference; The calculation of aspectRatioSimilarity is as follows: where AR and BR respectively represent the rectangular width-to-height ratios of the merged abnormal detection region and the target abnormal region, and aspectRatioSimilarity represents the width-to-height similarity ratio of the two rectangular regions; The calculation of positionSimilarity is as follows: xOverlap = max(0, min(xa + |xa′ - xa|, xb′)) - max(xa, xb) yOverlap = max(0, min(yb + |yb′ - yb|, yb′)) - max(ya, yb) overlapArea = xOverlap * yOverlap aArea = |xa′ - xa| * |ya′ - ya| bArea = |xb′ - xb| * |yb′ - yb| minArea = min(aArea, bArea) Among them, xOverlap represents the width of the overlapping area, yOverlap represents the height of the overlapping area, overlapArea represents the area of the overlapping area, aArea and bArea respectively represent the areas of the merged abnormal detection area and the target abnormal area, minArea represents the smaller part of the areas of the two rectangles, and positionSimilarity represents the proportion of the overlapping area in the smaller area of them.

7. The conveyor belt health diagnosis method according to claim 6, characterized in that angleDifference is calculated as follows: angleDifference = |AR - BR| similarity is calculated as follows: When the value of similarity is greater than the set threshold, it is determined that the merged abnormal detection area and the target abnormal area are the same wear defect block, and the merged abnormal detection area is numbered and stored in the database; The determining the belt health level according to the historical abnormal defect area and quantity includes: Calculating the difference in the area of the aArea and bArea between the previous and current blocks, as well as the number of defect blocks included in the aArea and bArea; Determining the belt health level according to the change rate of the area difference between the previous and current blocks and the change rate of the number of defect blocks; among them, the belt health level is negatively correlated with the change in wear area and the change in the number of defect blocks.

8. A conveyor belt health diagnosis device, characterized in that the device includes: A screening module, configured to receive belt data collected and uploaded by a non-destructive X-ray detector on the conveyor belt through the MQTT protocol, and classify and screen the belt data to obtain the coordinate information of the abnormal detection area; the abnormal detection area is a rectangular area calibrated according to the belt data; A detection module, configured to perform clustering detection on adjacent abnormal detection areas on the same conveyor belt, and calculate the regional distance and regional overlap of each abnormal detection area; A clustering module, configured to cluster and merge abnormal detection areas with a regional distance less than the set threshold to obtain a merged abnormal area; A judgment module, configured to calculate the abnormal defect area and quantity in the merged abnormal area, determine the belt health level according to the historical abnormal defect area and quantity, and send an alarm instruction to the belt target detector when the belt wear is lower than the health level threshold.

9. A server, characterized in that the server includes a processor and a memory, and at least one instruction, at least one program, a code set or an instruction set is stored in the memory, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the conveyor belt health diagnosis method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that at least one instruction, at least one program, a code set or an instruction set is stored in the readable storage medium, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement the conveyor belt health diagnosis method according to any one of claims 1 to 7.

Images