A belt conveyor operation abnormality detection method, device and electronic equipment

Abstract

The present application provides a belt conveyor operation abnormality detection method, device and electronic device, the method comprising: obtaining point cloud data of the conveyor belt of the belt conveyor in a fully loaded state, the point cloud data comprising: the cross-section of the conveyor belt at different times; identifying the cross-section curve from the cross-section; and determining the belt conveyor operation abnormality detection result according to the material cross-section curve. In the implementation process of the above scheme, by obtaining the point cloud data of the conveyor belt of the belt conveyor in a fully loaded state, and identifying the material cross-section curve from the cross-section of the point cloud data at different times, and then determining the belt conveyor operation abnormality detection result according to the cross-section curve, the point cloud data can be used to achieve accurate material contour and volume measurement, which makes the accuracy of abnormality detection much higher than traditional manual inspection and image detection-based methods, and effectively improves the accuracy of detecting abnormalities of the conveyor belt of the belt conveyor.

Description

Technical Field

[0001] The present application relates to the technical field of industrial manufacturing and belt conveyors, and in particular, to a method, device and electronic equipment for detecting abnormal operation of a belt conveyor. Background Art

[0002] At present, the detection of whether there is an abnormality in the conveyor belt of the belt conveyor is usually done through manual inspection. In the actual practice, it is found that some conveyor belts are very long, and the inspectors need to run back and forth during the inspection, which makes it impossible to find the abnormality of the belt conveyor in time. In addition, the human eye may not be able to see the relatively small differences in the manual inspection method. Therefore, the accuracy of detecting the abnormality of the conveyor belt of the belt conveyor through manual inspection is low. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a method, device and electronic equipment for detecting abnormal operation of a belt conveyor, so as to improve the problem of low accuracy in detecting abnormalities of a conveyor belt of a belt conveyor.

[0004] The embodiment of the present application provides a method for detecting abnormal operation of a belt conveyor, including: obtaining point cloud data of the conveyor belt of the belt conveyor in a fully loaded state, the point cloud data including: the cross-section of the conveyor belt at different times; identifying the material cross-section curve from the cross-section; and determining the abnormal operation detection result of the belt conveyor according to the material cross-section curve. In the implementation process of the above scheme, by obtaining the point cloud data of the conveyor belt of the belt conveyor in a fully loaded state, and identifying the material cross-section curve from the cross-section of the point cloud data at different times, and then determining the abnormal operation detection result of the belt conveyor according to the cross-section curve, the point cloud data can be used to achieve accurate material contour and volume measurement, which makes the accuracy of abnormality detection much higher than traditional manual inspection and image detection-based methods, and effectively improves the accuracy of detecting abnormalities of the conveyor belt of the belt conveyor.

[0005] Optionally, in an embodiment of the present application, the abnormal operation detection result of the belt conveyor is determined according to the material cross-section curve, including: obtaining historical data of the conveyor belt within a preset time length, and calculating the point cloud average value according to the historical data within the preset time length; dividing the point cloud average value along the cross-section direction of the conveyor belt to obtain the average line of the conveyor belt; subtracting the point value on the average line of the conveyor belt from the point value on the material cross-section curve to obtain a characteristic data value; judging whether there is a data value in the characteristic data value whose curvature exceeds a preset curvature threshold; if so, determining the abnormal detection result as a large block abnormality. In the implementation process of the above scheme, by obtaining historical data within a preset time length and calculating the point cloud average value, subtle changes in material distribution can be captured, which helps to timely discover abnormal situations, especially those large block abnormalities that have potential impacts on production efficiency and safety. At the same time, the average line of the conveyor belt is calculated, and then the point value on the material cross-section curve is subtracted from the point value on the average line. This method can more accurately identify abnormal peaks in material distribution, that is, the location of large blocks of material, and improve the sensitivity and accuracy of detection.

[0006] Optionally, in an embodiment of the present application, the abnormal operation detection result of the belt conveyor is determined according to the material cross-section curve, including: judging whether the distance between the line center point of the material cross-section curve and the surface center point of the cross-section of the transport belt exceeds a preset distance threshold; if so, the abnormal detection result is determined as a deviation abnormality. In the implementation process of the above scheme, by continuously monitoring the distance between the two center points, an early warning can be issued at the early stage of deviation, allowing the operator or control system to respond in time to avoid further deviation of the belt, thereby reducing equipment damage and downtime. At the same time, the distance between the center points is used as a basis for judgment, providing a quantitative index, making the identification of deviation abnormality more objective and accurate, reducing the error of human judgment, and thus improving the accuracy of the detection of belt conveyor operation abnormality.

[0007] Optionally, in an embodiment of the present application, the abnormal operation detection result of the belt conveyor is determined according to the material cross-section curve, including: obtaining point cloud data of the transport belt in an unloaded state, and identifying the belt scanning curve from the scanning surface of the point cloud data in the unloaded state; determining whether the abnormal detection result is a material blocking abnormality according to the material cross-section curve and the belt scanning curve. In the implementation process of the above scheme, by acquiring and processing point cloud data in real time, it is possible to monitor the material blocking abnormality in real time and respond quickly to prevent equipment damage or production interruption caused by aggravated material blocking. At the same time, by comparing the material cross-section curve and the belt scanning curve in the unloaded state, it is possible to clearly distinguish between normal material distribution and abnormal accumulation of material blocking, thereby improving the accuracy and reliability of identifying material blocking abnormalities.

[0008] Optionally, in an embodiment of the present application, the material blocking anomaly includes: accumulated material blocking; determining whether the abnormal detection result is a material blocking anomaly based on the material cross-section curve and the belt scanning curve, including: subtracting the lowest point value on the belt scanning curve from the highest point value on the material cross-section curve to obtain the actual height of the cross section; if the actual height of the cross section exceeds a preset height threshold, the abnormal detection result is determined to be accumulated material blocking. In the implementation process of the above scheme, by calculating the actual height of the cross section in real time, the system can immediately detect abnormal conditions that exceed the preset threshold, allowing operators or automated systems to take prompt action to prevent further deterioration of the material blocking, and automatically calculating the actual height of the cross section, eliminating errors in manual measurement, and ensuring the accuracy and consistency of the detection results.

[0009] Optionally, in an embodiment of the present application, the material blocking anomaly includes: complete material blocking; determining whether the anomaly detection result is a material blocking anomaly based on the material cross-section curve and the belt scanning curve, including: calculating the upstream material flow of the transport belt and the downstream material flow of the transport belt based on the material cross-section curve and the belt scanning curve; judging whether the difference between the upstream material flow and the downstream material flow is greater than a preset difference threshold; if so, determining the anomaly detection result as complete material blocking. In the implementation process of the above scheme, by directly comparing the upstream and downstream material flows, the system can accurately identify whether the material flow is blocked, which is more reliable than relying solely on a single sensor or visual inspection. Accurate flow control helps to balance the material flow of the entire production line, avoid upstream overload and downstream idleness, thereby improving overall production efficiency and resource utilization.

[0010] Optionally, in an embodiment of the present application, after the belt scanning curve is identified from the scanning surface of the point cloud data in the no-load state, it also includes: using a neural network model to predict the material cross-sectional curve and the belt scanning curve to obtain an abnormal prediction result of the belt conveyor, wherein the neural network model is obtained by training the neural network using the material cross-sectional curve and the belt scanning curve as training data and the abnormal detection results as training labels. In the implementation process of the above scheme, by predicting future abnormal conditions, enterprises can take measures in advance to perform preventive maintenance and reduce sudden failures. The neural network model can learn and recognize complex patterns, and can identify abnormal signals even in small changes that are difficult for humans to detect, thereby improving the intelligence and accuracy of abnormal detection.

[0011] Optionally, in an embodiment of the present application, after obtaining the abnormal prediction result of the belt conveyor, it also includes: if the abnormal prediction result of the belt conveyor is different from the abnormal operation detection result of the belt conveyor, and the abnormal prediction result of the belt conveyor is verified to be an erroneous result, then the material cross-section curve and the belt scanning curve are used as incremental training data, and the abnormal operation detection result of the belt conveyor is used as an incremental training label to perform incremental training on the neural network model. In the implementation process of the above scheme, the model can adapt to changes in the production environment in real time through incremental training, such as changes in material properties, equipment aging, or changes in operating conditions, to ensure that the prediction results are always in line with the actual situation. Compared with retraining the entire model from scratch, incremental training only needs to update part of the model's weights, which greatly reduces computing resources and time costs. The model can continuously learn the latest data, gradually correct errors, and improve the accuracy and reliability of predictions.

[0012] The embodiment of the present application also provides a method for detecting abnormal operation of a belt conveyor, including: obtaining point cloud data of the conveying belt of the belt conveyor in a fully loaded state, and identifying a material cross-section curve from the cross section of the point cloud data in the fully loaded state; obtaining point cloud data of the conveying belt in an unloaded state, and identifying a belt scanning curve from the scanning surface of the point cloud data in the unloaded state; determining the abnormal operation detection result of the belt conveyor according to the material cross-section curve and the belt scanning curve through a rule algorithm; using a neural network model to predict the material cross-section curve and the belt scanning curve to obtain an abnormal prediction result of the belt conveyor; if the abnormal prediction result of the belt conveyor is the same as the abnormal operation detection result of the belt conveyor, the abnormal prediction result of the belt conveyor is determined as the final abnormal result. In the implementation process of the above scheme, the combination of the rule algorithm and the neural network model provides double verification, enhances the accuracy and reliability of abnormal detection, and reduces the false positives or omissions that may exist in a single technology. The rule algorithm can process data in real time and provide immediate feedback, while the neural network model focuses on the prediction of long-term trends. The two complement each other and realize the effective combination of immediate response and future prediction.

[0013] Optionally, in an embodiment of the present application, it also includes: if the abnormal prediction result of the belt conveyor is different from the abnormal operation detection result of the belt conveyor, then after the abnormal prediction result of the belt conveyor is verified to be an erroneous result, the material cross-section curve and the belt scanning curve are used as incremental training data, and the abnormal operation detection result of the belt conveyor is used as an incremental training label to perform incremental training on the neural network model. In the implementation process of the above scheme, the model can adapt to changes in the production environment in real time through incremental training, such as changes in material properties, equipment aging, or changes in operating conditions, to ensure that the prediction results are always in line with the actual situation. Compared with retraining the entire model from scratch, incremental training only needs to update part of the model's weights, which greatly reduces computing resources and time costs. The model can continuously learn the latest data, gradually correct errors, and improve the accuracy and reliability of predictions.

[0014] An embodiment of the present application also provides a belt conveyor operation abnormality detection device, including: a point cloud data acquisition module, used to obtain point cloud data of the conveyor belt of the belt conveyor in a fully loaded state, the point cloud data including: a cross-section of the conveyor belt at different times; a material curve recognition module, used to identify the material cross-section curve from the cross-section; an abnormal result determination module, used to determine the belt conveyor operation abnormality detection result based on the material cross-section curve.

[0015] Optionally, in an embodiment of the present application, an abnormal result determination module includes: a historical data processing submodule, used to obtain historical data of the conveyor belt within a preset time length, and calculate the point cloud average value based on the historical data within the preset time length; a cloud average value slicing submodule, used to slice the point cloud average value along the transverse direction of the conveyor belt to obtain the average line of the conveyor belt; a characteristic data acquisition submodule, used to subtract the point value on the average line of the conveyor belt from the point value on the material transverse curve to obtain a characteristic data value; a characteristic data judgment submodule, used to judge whether there is a data value in the characteristic data value whose curvature exceeds a preset curvature threshold; a large block abnormality determination submodule, used to determine the abnormality detection result as a large block abnormality if there is a data value in the characteristic data value whose curvature exceeds a preset curvature threshold.

[0016] Optionally, in an embodiment of the present application, the abnormal result determination module includes: a center distance judgment submodule, used to determine whether the distance between the line center point of the material cross-section curve and the surface center point of the cross-section of the conveyor belt exceeds a preset distance threshold; a deviation abnormality determination submodule, used to determine the abnormal detection result as a deviation abnormality if the distance between the line center point of the material cross-section curve and the surface center point of the cross-section of the conveyor belt exceeds a preset distance threshold.

[0017] Optionally, in an embodiment of the present application, the abnormal result determination module includes: a point cloud data processing sub-module, used to obtain the point cloud data of the transport belt in a no-load state, and identify the belt scanning curve from the scanning surface of the point cloud data in the no-load state; a material blockage abnormality determination sub-module, used to determine whether the abnormal detection result is a material blockage abnormality based on the material cross-sectional curve and the belt scanning curve.

[0018] Optionally, in an embodiment of the present application, the material blockage anomaly includes: accumulated material blockage; a material blockage anomaly determination submodule, including: an actual height determination unit, used to subtract the lowest point value on the belt scanning curve from the highest point value on the material cross-section curve to obtain the actual height of the cross section; an accumulated material blockage determination unit, used to determine the abnormal detection result as accumulated material blockage if the actual height of the cross section exceeds a preset height threshold.

[0019] Optionally, in an embodiment of the present application, the material blockage anomaly includes: complete material blockage; a material blockage anomaly determination submodule includes: a material flow calculation unit, used to calculate the upstream material flow of the transport belt and the downstream material flow of the transport belt according to the material cross-sectional curve and the belt scanning curve; a flow difference judgment unit, used to judge whether the difference between the upstream material flow and the downstream material flow is greater than a preset difference threshold; a complete material blockage determination unit, used to determine the abnormal detection result as a complete material blockage if the difference between the upstream material flow and the downstream material flow is greater than a preset difference threshold.

[0020] Optionally, in an embodiment of the present application, the abnormal result determination module includes: an abnormal result acquisition sub-module, which is used to use a neural network model to predict the material cross-section curve and the belt scanning curve to obtain an abnormal prediction result of the belt conveyor, wherein the neural network model is obtained by training the neural network using the material cross-section curve and the belt scanning curve as training data and the abnormal detection results as training labels.

[0021] Optionally, in an embodiment of the present application, the abnormal result determination module includes: a model incremental training sub-module, which is used to perform incremental training on the neural network model using the material cross-section curve and the belt scanning curve as incremental training data and the belt conveyor operation abnormality detection result as incremental training label if the abnormal prediction result of the belt conveyor is different from the belt conveyor operation abnormality detection result, and the belt conveyor operation abnormality prediction result is verified to be an erroneous result.

[0022] An embodiment of the present application also provides a belt conveyor operation abnormality detection device, including: a point cloud data recognition module, used to obtain point cloud data of the conveyor belt of the belt conveyor in a fully loaded state, and identify the material cross-section curve from the cross-section of the point cloud data in the fully loaded state; a scanning curve recognition module, used to obtain point cloud data of the conveyor belt in an unloaded state, and identify the belt scanning curve from the scanning surface of the point cloud data in the unloaded state; a detection result determination module, used to determine the belt conveyor operation abnormality detection result according to the material cross-section curve and the belt scanning curve through a rule algorithm; an abnormal result determination module, used to predict the material cross-section curve and the belt scanning curve using a neural network model to obtain the belt conveyor abnormality prediction result; a final result determination module, used to determine the belt conveyor abnormality prediction result as the final abnormality result if the belt conveyor abnormality prediction result is the same as the belt conveyor operation abnormality detection result.

[0023] Optionally, in an embodiment of the present application, the belt conveyor operation abnormality detection device also includes: a neural network training module, which is used to perform incremental training on the neural network model using the material cross-section curve and the belt scanning curve as incremental training data and the belt conveyor operation abnormality detection result as incremental training label if the abnormality prediction result of the belt conveyor is different from the abnormality detection result of the belt conveyor operation. After the abnormality prediction result of the belt conveyor is verified to be an erroneous result,

[0024] An embodiment of the present application further provides an electronic device, including: a processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the machine-readable instructions execute the method described above when executed by the processor.

[0025] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method described above is executed.

[0026] The embodiment of the present application also provides a computer program product, including: a computer program or a computer instruction, and the computer program or the computer instruction executes the method described above when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0028] Figure 1 A schematic diagram of a process flow of a belt conveyor operation abnormality detection method provided by an embodiment of the present application is shown;

[0029] Figure 2 A schematic cross-sectional view of a transport belt provided by an embodiment of the present application in a normal state is shown;

[0030] Figure 3 A schematic diagram showing a large abnormality of a conveyor belt provided by an embodiment of the present application when the force is unevenly applied;

[0031] Figure 4 A schematic diagram showing the actual height of a cross section provided by an embodiment of the present application is shown;

[0032] Figure 5 A flow chart of a belt conveyor operation abnormality detection method using a neural network model provided in an embodiment of the present application is shown;

[0033] Figure 6 A schematic diagram of the process flow of a belt conveyor operation abnormality detection device provided by an embodiment of the present application is shown;

[0034] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0035] To make the purpose, technical scheme and advantages of the embodiment of the present application clearer, the technical scheme in the embodiment of the present application will be clearly and completely described below in conjunction with the drawings in the embodiment of the present application. It should be understood that the drawings in the embodiment of the present application only serve the purpose of explanation and description, and are not used to limit the protection scope of the embodiment of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in the embodiment of the present application shows the operation implemented according to some embodiments of the embodiment of the present application. It should be understood that the operation of the flowchart can be implemented out of order, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart under the guidance of the content of the embodiment of the present application, or remove one or more operations from the flowchart.

[0036] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all embodiments. The components of the embodiments of the present application described and shown in the drawings generally here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the embodiments of the present application claimed, but merely represents the selected embodiments of the embodiments of the present application.

[0037] It is understandable that the "first" and "second" in the embodiments of the present application are used to distinguish similar objects. Those skilled in the art will understand that the words "first" and "second" do not limit the quantity and execution order, and the words "first" and "second" do not necessarily limit the difference. In the description of the embodiments of the present application, the term "and / or" is merely a description of the association relationship of associated objects, indicating that there may be three relationships, such as A and / or B, which can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the related objects before and after are in an "or" relationship. The term "multiple" refers to more than two (including two), and similarly, "multiple groups" refers to more than two groups (including two groups).

[0038] It should be noted that the belt conveyor operation abnormality detection method provided in the embodiment of the present application can be executed by an electronic device, where the electronic device refers to a device terminal or server with the function of executing a computer program, such as a smart phone, a personal computer, a tablet computer, a personal digital assistant or a mobile Internet device. A server refers to a device that provides computing services through a network, such as an x86 server and a non-x86 server, and a non-x86 server includes a mainframe, a minicomputer and a UNIX server.

[0039] The following is an example of an application scenario in which the belt conveyor operation abnormality detection method is applicable: After the belt conveyor in industrial fields such as mines, quarries, coal, and sand and gravel has been running continuously for a period of time, the conveyor belt of the belt conveyor is prone to abnormalities, such as large block abnormalities, deviation abnormalities, and material blocking abnormalities. Among them, large block abnormalities are non-standard, oversized, or abnormally shaped objects encountered in the process of conveying materials on the conveyor belt. These objects may have a negative impact on the normal operation of the belt conveyor (such as rollover). Deviation abnormalities are when the conveyor belt deviates from the predetermined center line during operation, causing one or both sides of the belt to contact and rub against the frame or rollers. Material blocking abnormalities are when the conveyor belt is conveying materials. Due to the accumulation, jamming, or improper distribution of materials, a certain part of the belt conveyor or the entire system cannot smoothly transport materials. If these belt conveyor abnormalities are not identified and handled in a timely manner, the belt conveyor is prone to safety accidents in the process of transporting materials.

[0040] See also Figure 1The flowchart of the belt conveyor operation abnormality detection method provided by the embodiment of the present application is shown; the main idea of ​​the belt conveyor operation abnormality detection method is to identify the material cross-section curve from the cross-section of the point cloud data of the transport belt, and determine the belt conveyor operation abnormality detection result based on the cross-section curve. Effective use of point cloud data can achieve accurate material contour and volume measurement, which makes the accuracy of abnormality detection much higher than traditional manual inspection and image detection-based methods, and effectively improves the accuracy of detecting abnormalities of the conveyor belt of the belt conveyor. The implementation method of the above-mentioned belt conveyor operation abnormality detection method may include:

[0041] Step S110: obtaining point cloud data of the conveyor belt of the belt conveyor in a fully loaded state, the point cloud data including: cross sections of the conveyor belt at different times.

[0042] It can be understood that the above-mentioned point cloud data includes the cross-section of the conveyor belt at different times, that is, the point cloud data is composed of cross-sectional data at different times. It can be understood that the vertical axis (Y axis) of the point cloud data is composed of cross-sectional data frame by frame.

[0043] See also Figure 2 A schematic diagram of a cross section of a conveyor belt in a normal state provided by an embodiment of the present application is shown; an implementation of the above step S110 is, for example: scanning the fully loaded conveyor belt with a depth camera above the conveyor belt to obtain point cloud data of the fully loaded conveyor belt, and then the depth camera can directly send the point cloud data to the electronic device, and the electronic device can receive the above point cloud data of the fully loaded conveyor belt. The camera head of the above depth camera can be set from top to bottom, so that the depth camera shoots downward the materials transported on the conveyor belt. This camera setting method makes the depth camera less likely to accumulate dust and is easy to maintain.

[0044] Step S120: Identify the material cross-section curve from the cross-section.

[0045] An implementation example of the above-mentioned step S120 is as follows: in the cross-sections of the material area of ​​the conveyor belt at different times, these cross-sections are perpendicular to the conveying direction of the conveyor belt, and a fitting algorithm is used to perform curve fitting on the cross-section curves in the cross-sections to obtain identified cross-section curves. The fitting algorithm here can adopt least squares method, Bezier curve fitting, spline curve fitting and other algorithms.

[0046] Step S130: determining the abnormal operation detection result of the belt conveyor according to the material cross-section curve.

[0047] It can be understood that since the abnormal detection results include large block abnormalities, deviation abnormalities and blockage abnormalities, etc., and the blockage abnormalities are divided into accumulation blockages and complete blockages, etc., there are many ways to implement the detection results of the belt conveyor's operation abnormalities. The following will describe in detail the implementation methods for determining the detection results of the belt conveyor's operation abnormalities.

[0048] In the implementation process of the above scheme, by obtaining the point cloud data of the conveyor belt of the belt conveyor in a fully loaded state, and identifying the material cross-section curve from the cross-section of the point cloud data at different times, and then determining the belt conveyor operation abnormality detection result based on the cross-section curve, the point cloud data can be used to achieve accurate material contour and volume measurement, which makes the accuracy of anomaly detection much higher than traditional manual inspections and image-based detection methods, effectively improving the accuracy of detecting abnormalities in the conveyor belt of the belt conveyor.

[0049] See also Figure 3 The schematic diagram of the large-block anomaly of the conveyor belt provided by the embodiment of the present application when the force is unevenly applied is shown; it can be seen from the schematic diagram of the large-block anomaly that the curvature of at least one point in the material cross-section curve is very high, and this characteristic can be calculated through the above point cloud data and its historical data. Specifically, as an optional implementation of the above step S130, the above implementation of determining the operation anomaly detection result of the belt conveyor according to the material cross-section curve may include:

[0050] Step S131: acquiring historical data of the transport belt within a preset time period, and calculating the point cloud average value based on the historical data within the preset time period.

[0051] The historical data of the above-mentioned conveyor belt within the preset time period may be the historical data within the past 3 minutes, or may be the historical data of a fixed number of frames, for example, the point cloud data of the past 300 frames is used as the historical data within the above-mentioned preset time period.

[0052] Step S132: Slice the point cloud average value along the transverse direction of the conveyor belt to obtain the average line of the conveyor belt.

[0053] Step S133: Subtract the point value on the average line of the conveyor belt from the point value on the material cross-section curve to obtain the characteristic data value.

[0054] Step S134: Determine whether there is a data value in the characteristic data value whose curvature exceeds a preset curvature threshold.

[0055] After subtracting the point value on the average line of the conveyor belt from the point value on the material cross-section curve, it can be seen that the curvature of the curve point value shows a sudden increase trend. In order to determine whether it is a large block anomaly based on this trend, a curvature threshold can be set to determine whether it is a large block anomaly. Specifically, an executable program can be used to determine whether there is a data value in the characteristic data value whose curvature exceeds the preset curvature threshold.

[0056] Step S135: If there is a data value whose curvature exceeds a preset curvature threshold value in the feature data value, the anomaly detection result is determined as a large block anomaly.

[0057] An implementation example of the above steps S134 to S135 is: using an executable program to determine whether there are data values ​​in the feature data values ​​whose curvature exceeds a preset curvature threshold. If there are data values ​​in the feature data values ​​whose curvature exceeds the preset curvature threshold, the electronic device can determine the above abnormality detection result as a large block abnormality; wherein the above preset curvature threshold can be set according to a specific scenario or according to historical experience.

[0058] As an optional implementation of the above step S130, the above implementation of determining the abnormal operation detection result of the belt conveyor according to the material cross-section curve may include:

[0059] Step S136: Determine whether the distance between the center point of the line of the material cross-section curve and the center point of the surface of the cross-section of the conveyor belt exceeds a preset distance threshold.

[0060] Assuming that the cross section of the conveyor belt is represented by the xy coordinate axis, the maximum value of the x-axis is set to xMax, and the maximum value of the y-axis is set to yMax, then the coordinates of the center point of the cross section above can be expressed as

[0061] Step S137: If the distance between the center point of the line of the material cross-section curve and the center point of the surface of the cross-section of the conveyor belt exceeds a preset distance threshold, the abnormality detection result is determined as a deviation abnormality.

[0062] The implementation method of the above steps S136 to S137 is, for example: using an executable program to determine whether the distance between the line center point of the material cross-section curve and the surface center point of the cross-section of the conveyor belt exceeds a preset distance threshold; the line center point of the above material cross-section curve can be a center point or a center of gravity in a mathematical sense. If the distance between the line center point of the material cross-section curve and the surface center point of the cross-section of the conveyor belt exceeds the preset distance threshold, the abnormality detection result is determined as a deviation abnormality. The above preset distance threshold can also be set according to specific scenarios and historical experience values. Therefore, the preset distance threshold is manually controllable. For example, the preset distance threshold can be set to 10 cm or 20 cm, etc.

[0063] As an optional implementation of the above step S130, the above implementation of determining the abnormal operation detection result of the belt conveyor according to the material cross-section curve may include:

[0064] Step S138: Acquire the point cloud data of the transport belt in the no-load state, and identify the belt scanning curve from the scanning surface of the point cloud data in the no-load state.

[0065] Step S139: Determine whether the abnormality detection result is a material blocking abnormality based on the material cross-section curve and the belt scanning curve.

[0066] See also Figure 4 A schematic diagram of the actual height of the cross section provided by the embodiment of the present application is shown; under normal circumstances, the material cross-section curve and the belt scanning curve in the cross section have a small amplitude. If there is no amplitude or a small amplitude within a certain range, it means that there is no material blocking abnormality. This amplitude can be specifically measured by the actual height of the cross section. As an optional implementation of the above-mentioned step S139, the above-mentioned material blocking abnormality may include: accumulated material blocking; the above-mentioned implementation method of determining whether the abnormal detection result is a material blocking abnormality based on the material cross-section curve and the belt scanning curve may include:

[0067] Step S139a: Subtract the lowest point value on the belt scanning curve from the highest point value on the material cross-section curve to obtain the actual height of the cross section.

[0068] In order to judge the accumulation and blockage of materials, the concept of actual height is introduced here, that is, the highest point value on the material cross-section curve is subtracted from the lowest point value on the belt scanning curve to obtain the actual height of the cross section, which can be understood as the distance between the bottom of the belt and the highest position of the material pile.

[0069] Step S139b: If the actual height of the cross section exceeds the preset height threshold, the abnormal detection result is determined as accumulated blockage.

[0070] An example of an implementation of the above step S139b is: Under normal circumstances, the actual height of the cross section will be within a certain range. If the actual height of the cross section exceeds the preset height threshold of the range, the abnormal detection result is determined as an accumulation of material blockage. The actual height of the cross section is obtained by subtracting the lowest point value on the belt scanning curve from the highest point value on the material cross-section curve through the executable program. If the actual height of the cross section exceeds the preset height threshold, the electronic device can determine the abnormal detection result as an accumulation of material blockage. By continuously checking the material cross-section curve and the belt scanning curve obtained by the point cloud data, it can be found that the material position curvature increases suddenly when there is a large block abnormality, while the material position curvature increases gradually when there is a blockage abnormality.

[0071] As an optional implementation of the above step S139, the above material blocking abnormality may include: complete material blocking; the above implementation of determining whether the abnormality detection result is a material blocking abnormality based on the material cross-section curve and the belt scanning curve may include:

[0072] Step S139c: Calculate the upstream material flow rate of the conveyor belt and the downstream material flow rate of the conveyor belt according to the material cross-section curve and the belt scanning curve.

[0073] In the specific practice, depth cameras can be installed upstream and downstream of the conveyor belt respectively. The point cloud data obtained by the upstream depth camera is used to calculate the upstream material flow of the conveyor belt, and the point cloud data obtained by the downstream depth camera is used to calculate the downstream material flow of the conveyor belt.

[0074] Step S139d: Determine whether the difference between the upstream material flow rate and the downstream material flow rate is greater than a preset difference threshold.

[0075] Step S139e: If the difference between the upstream material flow rate and the downstream material flow rate is greater than the preset difference threshold, the abnormal detection result is determined as complete material blockage.

[0076] The implementation of the above step S139d and step S139e is, for example: the electronic device uses an executable program to determine whether the difference between the upstream material flow and the downstream material flow is greater than a preset difference threshold, wherein the preset difference threshold here can be set according to a specific scenario or a historical experience value. If the difference between the upstream material flow and the downstream material flow is greater than the preset difference threshold, the electronic device can directly determine the abnormal detection result as complete material blockage.

[0077] Optionally, the center point of the conveyor belt can be used as the boundary, and the material flow on both sides of the center point (including the left flow and the right flow) can be counted respectively, and then the imbalance degree can be calculated according to the material flow, and the left or right material blockage can be determined according to the flow imbalance degree formula. Among them, the above flow imbalance degree formula is for example:

[0078]

[0079] Among them, leftblock represents the above-mentioned left-side flow, and rightblock represents the above-mentioned right-side flow. If the left-side flow is greater than the right-side flow, the flow imbalance is the difference between the left-side flow and the left-right average flow divided by the left-right average flow; if the left-side flow is less than the right-side flow, the flow imbalance is the difference between the right-side flow and the left-right average flow divided by the left-right average flow.

[0080] As an optional implementation of the above method, after the belt scanning curve is identified from the scanning surface of the point cloud data in the no-load state, a neural network model can also be used to predict abnormalities. This implementation may include:

[0081] Step S139f: Use the neural network model to predict the material cross-section curve and the belt scanning curve to obtain the abnormal prediction result of the belt conveyor, wherein the neural network model is obtained by training the neural network with the material cross-section curve and the belt scanning curve as training data and the abnormal detection results as training labels.

[0082] The above neural network model can use a convolutional neural network (CNN) as the backbone network of the image classification model.

[0083] As an optional implementation of the above method, after obtaining the abnormality prediction result of the belt conveyor, the subsequent processing method can also be determined according to the abnormality prediction result and the abnormality detection result of the belt conveyor. The implementation may include:

[0084] Step S139g: If the abnormality prediction result of the belt conveyor is the same as the abnormality detection result of the belt conveyor operation, the abnormality detection result of the belt conveyor operation is used as the final abnormality result.

[0085] It is understandable that if the abnormal prediction result of the belt conveyor is the same as the abnormal operation detection result of the belt conveyor, then the abnormal operation detection result of the belt conveyor is credible. Since the abnormal prediction result of the belt conveyor is the same as the abnormal operation detection result of the belt conveyor, the abnormal prediction result of the belt conveyor or the abnormal operation detection result of the belt conveyor can be directly used as the final abnormal result. If the abnormal prediction result of the belt conveyor is different from the abnormal operation detection result of the belt conveyor, then the abnormal prediction result of the belt conveyor can be manually verified to be the correct result (indicating that the result of the above rule algorithm is wrong, and the algorithm engineer can supplement or modify the rule algorithm at this time). If the abnormal prediction result of the belt conveyor is manually verified to be an incorrect result, the following incremental training can be performed.

[0086] Step S139h: If the abnormal prediction result of the belt conveyor is different from the abnormal operation detection result of the belt conveyor, and the abnormal prediction result of the belt conveyor is verified to be an incorrect result, the material cross-section curve and the belt scanning curve are used as incremental training data, and the abnormal operation detection result of the belt conveyor is used as the incremental training label to perform incremental training on the neural network model.

[0087] Specifically, when the abnormal prediction result of the belt conveyor is different from the abnormal operation detection result of the belt conveyor, and the abnormal prediction result of the belt conveyor is verified to be an erroneous result, the material cross-section curve and the belt scanning curve can be added to the training data set as training data, and the abnormal operation detection result of the belt conveyor can be added to the training data set as a training label. At this time, the material cross-section curve and the belt scanning curve in the training data set can be directly used as training data, and the abnormal operation detection result of the belt conveyor can be used as a training label to fully train the neural network model. Of course, the newly added material cross-section curve and belt scanning curve in the training data set can also be used as incremental training data, and the newly added abnormal detection result in the training data set can be used as an incremental training label to perform incremental training on the neural network model.

[0088] In the process of implementing the above solution, incremental training is used to enable the model to adapt to changes in the production environment in real time, such as changes in material properties, equipment aging, or changes in operating conditions, to ensure that the prediction results always conform to the actual situation. Compared with retraining the entire model from scratch, incremental training only needs to update part of the model's weights, which greatly reduces computing resources and time costs. The model can continuously learn the latest data, gradually correct errors, and improve the accuracy and reliability of the prediction.

[0089] See also Figure 5 The flowchart of the belt conveyor operation abnormality detection method using a neural network model provided in the embodiment of the present application is shown; the embodiment of the present application also provides a belt conveyor operation abnormality detection method, including:

[0090] Step S210: acquiring point cloud data of the conveyor belt of the belt conveyor in a fully loaded state, and identifying a material cross-section curve from the cross-section of the point cloud data in the fully loaded state.

[0091] It can be understood that the implementation of the above step S210 is similar to the implementation of the above steps S110 to S120, and thus will not be described in detail.

[0092] Step S220: acquiring point cloud data of the transport belt in an unloaded state, and identifying a belt scanning curve from a scanning surface of the point cloud data in the unloaded state.

[0093] Step S230: Determine the abnormal operation detection result of the belt conveyor according to the material cross-section curve and the belt scanning curve through a rule algorithm.

[0094] It is understandable that the implementation of the above steps S220 to S230 is similar to the implementation of the above steps S138 to S139, so they are not repeated here. The above rule algorithm is also called an expert rule algorithm, which can be an algorithm that uses mathematical methods to implement calculation rules, such as the rules and / or algorithms used in the above steps S110 to S130 to determine the large block abnormality, deviation abnormality and blocking abnormality of the belt conveyor.

[0095] Step S240: Use the neural network model to predict the material cross-section curve and the belt scanning curve to obtain an abnormal prediction result of the belt conveyor.

[0096] Step S250: If the abnormality prediction result of the belt conveyor is the same as the abnormality detection result of the belt conveyor operation, the abnormality prediction result of the belt conveyor is determined as the final abnormality result.

[0097] It can be understood that the implementation of the above steps S240 to S250 is similar to the implementation of the above steps S139f to S139g, and therefore will not be repeated.

[0098] As an optional implementation of the above method, after obtaining the abnormal prediction result of the belt conveyor, the training neural network model can also be determined according to the abnormal prediction result and the abnormal detection result of the belt conveyor. The implementation may include:

[0099] Step S260: If the abnormality prediction result of the belt conveyor is different from the abnormality detection result of the belt conveyor, after the abnormality prediction result of the belt conveyor is verified to be an incorrect result, the material cross-section curve and the belt scanning curve are used as incremental training data, and the abnormality detection result of the belt conveyor is used as the incremental training label to perform incremental training on the neural network model.

[0100] It can be understood that the implementation of the above step S260 is similar to the implementation of the above step S139h, so it is not repeated here.

[0101] See also Figure 6 The flowchart of the belt conveyor operation abnormality detection device provided by the embodiment of the present application is shown; the embodiment of the present application provides a belt conveyor operation abnormality detection device 300, including:

[0102] The point cloud data acquisition module 310 is used to acquire point cloud data of the conveyor belt of the belt conveyor in a fully loaded state, and the point cloud data includes: cross sections of the conveyor belt at different times.

[0103] The material curve identification module 320 is used to identify the material cross-section curve from the cross-section.

[0104] The abnormal result determination module 330 is used to determine the abnormal operation detection result of the belt conveyor according to the material cross-section curve.

[0105] As an optional implementation of the above device, the abnormal result determination module includes:

[0106] The historical data processing submodule is used to obtain the historical data of the conveyor belt within a preset time period and calculate the point cloud average value based on the historical data within the preset time period.

[0107] The cloud average value segmentation module is used to segment the point cloud average value along the cross-sectional direction of the conveyor belt to obtain the average line of the conveyor belt.

[0108] The characteristic data acquisition submodule is used to obtain the characteristic data value by subtracting the point value on the average line of the transport belt from the point value on the material cross-section curve.

[0109] The feature data judgment submodule is used to judge whether there is a data value whose curvature exceeds a preset curvature threshold in the feature data value.

[0110] The large block anomaly determination submodule is used to determine the anomaly detection result as a large block anomaly if there is a data value whose curvature exceeds a preset curvature threshold in the feature data value.

[0111] As an optional implementation of the above device, the abnormal result determination module includes:

[0112] The center distance judgment submodule is used to judge whether the distance between the center point of the line of the material cross-section curve and the center point of the surface of the cross-section of the conveyor belt exceeds a preset distance threshold.

[0113] The deviation abnormality determination submodule is used to determine the abnormality detection result as a deviation abnormality if the distance between the line center point of the material cross-section curve and the surface center point of the cross-section of the conveyor belt exceeds a preset distance threshold.

[0114] As an optional implementation of the above device, the abnormal result determination module includes:

[0115] The point cloud data processing submodule is used to obtain the point cloud data of the transport belt in the no-load state, and identify the belt scanning curve from the scanning surface of the point cloud data in the no-load state.

[0116] The material blocking abnormality determination submodule is used to determine whether the abnormality detection result is a material blocking abnormality based on the material cross-section curve and the belt scanning curve.

[0117] As an optional implementation of the above device, the material blocking anomaly includes: material accumulation; the material blocking anomaly determination submodule includes:

[0118] The actual height determination unit is used to subtract the lowest point value on the belt scanning curve from the highest point value on the material cross-section curve to obtain the actual height of the cross section.

[0119] The accumulated blockage determination unit is used to determine the abnormal detection result as accumulated blockage if the actual height of the cross section exceeds a preset height threshold.

[0120] As an optional implementation of the above device, the material blocking abnormality includes: complete material blocking; the material blocking abnormality determination submodule includes:

[0121] The material flow calculation unit is used to calculate the upstream material flow of the conveyor belt and the downstream material flow of the conveyor belt according to the material cross-section curve and the belt scanning curve.

[0122] The flow difference judgment unit is used to judge whether the difference between the upstream material flow and the downstream material flow is greater than a preset difference threshold.

[0123] The complete material blockage determination unit is used to determine the abnormal detection result as a complete material blockage if the difference between the upstream material flow rate and the downstream material flow rate is greater than a preset difference threshold.

[0124] As an optional implementation of the above device, the abnormal result determination module includes:

[0125] The abnormal result acquisition submodule is used to use the neural network model to predict the material cross-section curve and the belt scanning curve to obtain the abnormal prediction result of the belt conveyor. The neural network model is obtained by training the neural network with the material cross-section curve and the belt scanning curve as training data and the abnormal detection results as training labels.

[0126] As an optional implementation of the above device, the abnormal result determination module includes:

[0127] The model incremental training submodule is used to perform incremental training on the neural network model using the material cross-section curve and the belt scanning curve as incremental training data and the belt conveyor operation abnormality detection result as incremental training label if the belt conveyor anomaly prediction result is different from the belt conveyor operation abnormality detection result, and the belt conveyor anomaly prediction result is verified to be an incorrect result.

[0128] The embodiment of the present application also provides a belt conveyor operation abnormality detection device, comprising:

[0129] The point cloud data recognition module is used to obtain the point cloud data of the conveyor belt of the belt conveyor in a fully loaded state, and to recognize the material cross-section curve from the cross-section of the point cloud data in the fully loaded state.

[0130] The scanning curve recognition module is used to obtain the point cloud data of the transport belt in the no-load state, and recognize the belt scanning curve from the scanning surface of the point cloud data in the no-load state.

[0131] The detection result determination module is used to determine the abnormal operation detection result of the belt conveyor according to the material cross-section curve and the belt scanning curve through a rule algorithm.

[0132] The abnormal result determination module is used to use the neural network model to predict the material cross-section curve and the belt scanning curve to obtain the abnormal prediction result of the belt conveyor.

[0133] The final result determination module is used to determine the abnormality prediction result of the belt conveyor as the final abnormality result if the abnormality prediction result of the belt conveyor is the same as the abnormality detection result of the operation of the belt conveyor.

[0134] As an optional implementation of the above device, the belt conveyor operation abnormality detection device further includes:

[0135] The neural network training module is used to perform incremental training on the neural network model using the material cross-section curve and the belt scanning curve as incremental training data and the belt conveyor operation abnormality detection result as incremental training label if the belt conveyor anomaly prediction result is different from the belt conveyor operation abnormality detection result. After the belt conveyor anomaly prediction result is verified to be an erroneous result, the neural network model is incrementally trained using the material cross-section curve and the belt scanning curve as incremental training data and the belt conveyor operation abnormality detection result as incremental training label.

[0136] It should be understood that the device corresponds to the above-mentioned belt conveyor operation abnormality detection method embodiment, and can execute the various steps involved in the above-mentioned method embodiment. The specific functions of the device can be referred to in the above description, and the detailed description is appropriately omitted here. The device includes at least one software function module that can be stored in a memory in the form of software or firmware or solidified in the operating system (OS) of the device.

[0137] See also Figure 7 An electronic device 400 provided in an embodiment of the present application includes: a processor 410 and a memory 420, wherein the memory 420 stores machine-readable instructions executable by the processor 410, and when the machine-readable instructions are executed by the processor 410, the above method is executed.

[0138] The embodiment of the present application also provides a computer-readable storage medium 430, on which a computer program is stored, and the computer program is executed by the processor 410 to execute the above method. The computer-readable storage medium 430 can be implemented by 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 memory, flash memory, disk or optical disk.

[0139] The embodiment of the present application also provides a computer program product, including: a computer program or a computer instruction, and the computer program or the computer instruction executes the method described above when executed by a processor.

[0140] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0141] In several embodiments provided by the embodiments of the present application, it should be understood that the disclosed device and method can also be implemented in other ways. The device embodiments described above are only schematic, for example, the flowcharts and block diagrams in the accompanying drawings show the possible implementation architecture, functions and operations of the devices, methods and computer program products according to the multiple embodiments of the embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a part of a module, a program segment or a code, and a part of a module, a program segment or a code includes 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 box can also be different from the order of occurrence marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which is mainly determined by the functions involved.

[0142] In addition, each functional module of each embodiment in the embodiment of the present application can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part. In addition, in the description of this specification, the description of reference terms "one embodiment", "some embodiments", "example", "specific example", "some examples", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of the present application. In this specification, the schematic representation of the above terms does not necessarily target the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples without contradiction.

[0143] The above description is only an optional implementation manner of the embodiments of the present application, but the protection scope of the embodiments of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the embodiments of the present application, which should be covered within the protection scope of the embodiments of the present application.

Claims

1. A belt conveyor operation abnormality detection method, characterized in that: include: Acquire point cloud data of a conveyor belt of a belt conveyor in a fully loaded state, the point cloud data comprising: cross sections of the conveyor belt at different times; identifying a material cross-section curve from the cross-section; Determining the abnormal operation detection result of the belt conveyor according to the material cross-section curve; Among them, the method of determining the operation abnormality detection result of the belt conveyor according to the material cross-section curve includes: obtaining historical data of the conveying belt within a preset time length, and calculating the point cloud average value according to the historical data within the preset time length; dividing the point cloud average value along the cross-section direction of the conveying belt to obtain the average line of the conveying belt; subtracting the point value on the average line of the conveying belt from the point value on the material cross-section curve to obtain a characteristic data value; judging whether there is a data value in the characteristic data value whose curvature exceeds a preset curvature threshold; if so, determining the abnormality detection result as a large block abnormality; obtaining the point cloud data of the conveying belt in a no-load state, and identifying the belt scanning curve from the scanning surface of the point cloud data in the no-load state; determining whether the abnormality detection result is a material blocking abnormality according to the material cross-section curve and the belt scanning curve.

2. The method according to claim 1, characterized in that The step of determining the abnormal operation detection result of the belt conveyor according to the material cross-section curve includes: Determine whether the distance between the center point of the line of the material cross-section curve and the center point of the surface of the cross-section of the conveyor belt exceeds a preset distance threshold; If so, the abnormal detection result is determined as a deviation abnormality.

3. The method according to claim 1, characterized in that The material blocking anomaly includes: material accumulation blocking; and determining whether the abnormality detection result is a material blocking anomaly based on the material cross-section curve and the belt scanning curve includes: Subtract the lowest point value on the belt scanning curve from the highest point value on the material cross-section curve to obtain the actual height of the cross-section; If the actual height of the cross section exceeds a preset height threshold, the abnormal detection result is determined as the accumulated blockage.

4. The method according to claim 1, characterized in that: The material blocking anomaly includes: complete material blocking; and determining whether the anomaly detection result is a material blocking anomaly based on the material cross-section curve and the belt scanning curve includes: Calculating the upstream material flow of the conveyor belt and the downstream material flow of the conveyor belt according to the material cross-section curve and the belt scanning curve; Determine whether the difference between the upstream material flow rate and the downstream material flow rate is greater than a preset difference threshold; If so, the abnormal detection result is determined as the complete blockage.

5. The method according to claim 1, characterized in that After the belt scanning curve is identified from the scanning surface of the point cloud data in the no-load state, the method further includes: A neural network model is used to predict the material cross-section curve and the belt scanning curve to obtain an abnormal prediction result of the belt conveyor, wherein the neural network model is obtained by training the neural network using the material cross-section curve and the belt scanning curve as training data and the abnormal detection results as training labels.

6. The method according to claim 5, characterized in that After obtaining the abnormal prediction result of the belt conveyor, the method further includes: If the abnormality prediction result of the belt conveyor is different from the abnormality detection result of the belt conveyor, and the abnormality prediction result of the belt conveyor is verified to be an erroneous result, the material cross-section curve and the belt scanning curve are used as incremental training data, and the abnormality detection result of the belt conveyor is used as an incremental training label to perform incremental training on the neural network model.

7. A belt conveyor operation abnormality detection method, characterized in that: include: Acquire point cloud data of a conveyor belt of a belt conveyor in a fully loaded state, and identify a material cross-section curve from a cross section of the point cloud data in the fully loaded state; Acquiring point cloud data of the transport belt in an unloaded state, and identifying a belt scanning curve from a scanning surface of the point cloud data in the unloaded state; Determining the abnormal operation detection result of the belt conveyor according to the material cross-section curve and the belt scanning curve through a rule algorithm; Using a neural network model to predict the material cross-section curve and the belt scanning curve, to obtain an abnormal prediction result of the belt conveyor; If the abnormality prediction result of the belt conveyor is the same as the abnormality detection result of the operation of the belt conveyor, the abnormality prediction result of the belt conveyor is determined as the final abnormality result; Among them, the method of determining the operation abnormality detection result of the belt conveyor according to the material cross-section curve and the belt scanning curve includes: obtaining historical data of the conveying belt within a preset time length, and calculating the point cloud average value according to the historical data within the preset time length; dividing the point cloud average value along the cross-section direction of the conveying belt to obtain the average line of the conveying belt; subtracting the point value on the average line of the conveying belt from the point value on the material cross-section curve to obtain a characteristic data value; judging whether there is a data value in the characteristic data value whose curvature exceeds a preset curvature threshold; if so, determining the abnormality detection result as a large block abnormality; obtaining the point cloud data of the conveying belt in a no-load state, and identifying the belt scanning curve from the scanning surface of the point cloud data in the no-load state; determining whether the abnormality detection result is a material blocking abnormality according to the material cross-section curve and the belt scanning curve.

8. The method according to claim 7, characterized in that Also includes: If the abnormality prediction result of the belt conveyor is different from the abnormality detection result of the belt conveyor operation, after the abnormality prediction result of the belt conveyor is verified to be an erroneous result, the material cross-section curve and the belt scanning curve are used as incremental training data, and the abnormality detection result of the belt conveyor operation is used as an incremental training label to perform incremental training on the neural network model.

9. A belt conveyor operation abnormality detection device, characterized in that: include: A point cloud data acquisition module is used to acquire point cloud data of a conveyor belt of a belt conveyor in a fully loaded state, wherein the point cloud data includes: cross sections of the conveyor belt at different times; A material curve identification module, used for identifying a material cross-section curve from the cross-section; An abnormal result determination module, used to determine the abnormal operation detection result of the belt conveyor according to the material cross-section curve; Among them, the method of determining the operation abnormality detection result of the belt conveyor according to the material cross-section curve includes: obtaining historical data of the conveying belt within a preset time length, and calculating the point cloud average value according to the historical data within the preset time length; dividing the point cloud average value along the cross-section direction of the conveying belt to obtain the average line of the conveying belt; subtracting the point value on the average line of the conveying belt from the point value on the material cross-section curve to obtain a characteristic data value; judging whether there is a data value in the characteristic data value whose curvature exceeds a preset curvature threshold; if so, determining the abnormality detection result as a large block abnormality; obtaining the point cloud data of the conveying belt in a no-load state, and identifying the belt scanning curve from the scanning surface of the point cloud data in the no-load state; determining whether the abnormality detection result is a material blocking abnormality according to the material cross-section curve and the belt scanning curve.

10. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the machine-readable instructions are executed by the processor to perform any method according to claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is executed.

12. A computer program product, characterized in that include: A computer program or a computer instruction, wherein when the computer program or the computer instruction is executed by a processor, the method according to any one of claims 1 to 8 is executed.

Images