Fault Prediction Method for Belt Conveyors Based on Intelligent Control

Abstract

This invention relates to the field of conveyor management technology, specifically disclosing a fault prediction method for belt conveyors based on intelligent control. The method includes setting monitoring periods and acquiring impact values, offset values, and category coefficients based on monitoring data within those periods. Then, an offset coefficient is obtained based on the impact value, offset value, and category coefficient. Within multiple consecutive and adjacent monitoring periods, the offset coefficient is compared with a control interval to obtain a judgment result indicating whether the offset coefficient falls within the control interval in different monitoring periods. An influence coefficient is obtained based on multiple judgment results, and then the deviation fault is predicted based on the influence coefficient. This invention enables effective monitoring and prediction of ultra-long belt conveyors while saving costs, and can predict the probability of impending deviation faults.

Description

Technical Field

[0001] This invention relates to the field of conveyor management technology, and more specifically, to a fault prediction method for belt conveyors based on intelligent control. Background Technology

[0002] Belt conveyors are versatile and can be used for transferring products or raw materials in various situations. Currently, the development trend of belt conveyors is towards intelligent control, which utilizes intelligent control technologies such as condition monitoring, intelligent video monitoring, and intelligent speed regulation, along with an integrated management and control platform, to achieve centralized control and unattended operation of belt conveyors. This reduces the failure rate of belt conveyors and achieves the goal of intelligent management and control.

[0003] In applications such as transporting minerals or delivering goods, where belt conveyors need to be used centrally from multiple locations, they are typically used in combination with multiple sub-conveyors and a main conveyor. In such cases, the main conveyor is usually set to be extremely long, making comprehensive data monitoring difficult. Furthermore, the failure of the main conveyor can result in significant and unacceptable losses. Therefore, extending the service life of the main conveyor and reducing its failure rate in such applications is of utmost importance.

[0004] There are various types of faults in the main conveyor, which can usually be detected through condition monitoring or intelligent video monitoring. Among them, conveyor belt misalignment is a special type of fault. Generally speaking, misalignment can cause great damage to the belt conveyor. However, conveyor belt misalignment can be corrected in time by a belt correction device. The belt correction device adjusts the running direction of the belt by increasing friction, which leads to increased wear on the belt edge. In other words, frequent use of the belt correction device will deteriorate the service condition of the belt conveyor. Therefore, how to reduce the use of the belt correction device is a problem that needs to be solved. In view of this, this invention proposes a belt conveyor fault prediction method based on intelligent control, which predicts the misalignment fault of the main conveyor and adjusts the sub-conveyors based on intelligent control to reduce the occurrence of the main conveyor misalignment problem. Summary of the Invention

[0005] The purpose of this invention is to provide a fault prediction method for belt conveyors based on intelligent control, and to solve the following technical problems:

[0006] How to predict the misalignment fault of the main conveyor and adjust the sub-conveyors based on intelligent control to reduce the occurrence of the main conveyor misalignment problem.

[0007] The objective of this invention can be achieved through the following technical solutions:

[0008] The fault prediction method for belt conveyors based on intelligent control includes the following steps:

[0009] At the intersection of the sub-conveyor and the main conveyor, video monitoring and quality monitoring are carried out on the transported materials on the main conveyor and the sub-conveyor to obtain monitoring data;

[0010] Set a monitoring period and obtain the impact value, offset value and category coefficient based on the monitoring data within the monitoring period. Then obtain the offset coefficient based on the impact value, offset value and category coefficient.

[0011] In multiple consecutive and adjacent monitoring periods, the offset coefficient is compared with the control interval to obtain the judgment result of whether the offset coefficient falls within the control interval in different monitoring periods;

[0012] The influence coefficient is obtained based on multiple judgment results, and then the deviation fault is predicted based on the influence coefficient.

[0013] As a further technical solution of the present invention: the step of obtaining the impact value includes:

[0014] The quality of transported goods entering the main conveyor from the sub-conveyor during the monitoring period is obtained through quality monitoring;

[0015] By acquiring image data of the sub-conveyors during the monitoring period through video monitoring, the height difference between the sub-conveyors and the main conveyor, as well as the type of transported goods and the corresponding category coefficient, are determined.

[0016] The first correction factor is set based on the category coefficient. The impact value is the product of the first correction value after de-normalization, the square of the weight and height difference of the transported goods entering the main conveyor from the sub-conveyor during the monitoring period, and the first correction factor.

[0017] As a further technical solution of the present invention: the process of obtaining the offset value includes:

[0018] During the monitoring period, video frames of the main conveyor belt before and after an impact action are monitored.

[0019] Within a video frame, contour recognition is used to obtain the contour image of the goods that have fallen onto the main conveyor belt;

[0020] Obtain the centroid of the profile of the main conveyor belt before and after an impact action is completed, and obtain the first and second distances from the two centroids to the centerline of the main conveyor belt.

[0021] The offset value is calculated based on the first distance and the second distance.

[0022] As a further technical solution of the present invention: the process of calculating and obtaining the offset value includes:

[0023] Through the formula:

[0024]

[0025] Among them, De j is the offset value for the j-th monitoring period, and n is the number of impact actions that occurred during the monitoring period. It is the first distance from the centroid of the main conveyor belt's profile after the i-th impact action is completed, which is the centerline of the main conveyor belt. σ1 is the second distance from the centroid of the main conveyor belt profile after the i-th impact action is completed, the centerline of the main conveyor belt, σ1 is the first de-normalization coefficient, δ is the preset correction distance conversion function, and i is a positive integer greater than 0 and less than n.

[0026] As a further technical solution of the present invention: the process of obtaining the offset coefficient includes:

[0027] Through the formula:

[0028]

[0029] Where Ex is the offset coefficient for the j-th monitoring period, α1 is the preset first weight coefficient, α2 is the preset second weight coefficient, ty is the category coefficient, m is the mass of transported goods entering the main conveyor from the sub-conveyor during the monitoring period, h is the height difference, σ2 is the second de-normalization coefficient, and De0 is the preset standard value of the offset value.

[0030] As a further technical solution of the present invention: the process of obtaining the judgment result includes:

[0031] The offset coefficient is compared with a pre-defined pair of critical intervals (-k1, -k2) and (k3, k4);

[0032] If the offset coefficient falls within any critical interval, the output judgment result is that there is a point offset within the monitoring period.

[0033] The above technical solution provides a method for obtaining the judgment result. The judgment of the landing point offset of the present invention is based on the offset coefficient. The offset coefficient uses the offset amount, impact amount and transport type as reference variables. The larger the offset amount or impact amount, the greater the impact of the transport on the conveyor belt during the impact process, and it can help to judge the impact direction on the conveyor belt during the impact process.

[0034] As a further technical solution of the present invention: the process of obtaining the influence coefficient based on the judgment result includes:

[0035] Through the formula:

[0036]

[0037] Where Inf is the influence coefficient, m1 is the number of monitoring periods judged to have landing point offset, m2 is the number of monitoring periods judged not to have landing point offset, and j is a positive integer greater than 0 and less than m1+m2.

[0038] As a further technical solution of the present invention: the process of predicting deviation faults based on the influence coefficient includes:

[0039] Based on the changes in the influence coefficient in historical data, a linear trend model is selected to predict the next influence coefficient and obtain the predicted value.

[0040] The predicted value is compared with the safe range. If the predicted value falls within the safe range, it is predicted that there is no possibility of conveyor belt misalignment.

[0041] If the predicted value is not within the safe range, it indicates that a conveyor belt misalignment fault is likely to occur.

[0042] The above technical solution provides a process for predicting conveyor belt misalignment faults. In the process of misalignment prediction, the influence coefficient obtained in the time series is used as the basis for prediction, so as to achieve the purpose of early prediction and facilitate the implementation of subsequent maintenance or adjustment plans.

[0043] As a further technical solution of the present invention: the end of the sub-conveyor is configured to be rotatable between [a°, b°], and the rotation angle is set;

[0044] If the sign of the influence coefficient obtained after rotation is different from that obtained before rotation, then stop rotating until the sum of the influence coefficient before rotation and the influence coefficient after multiple rotations has changed from positive to negative, and then return to the initial state.

[0045] If the sign of the influence coefficient obtained after rotation is the same as that obtained before rotation, then continue to rotate in the same direction with the same rotation angle.

[0046] The above technical solution provides a method for controlling and adjusting an intelligent conveyor belt. The end of the sub-conveyor in this invention is configured as a rotatable structure. Rotation changes the landing point of the transported material on the main conveyor, thereby adjusting the influence coefficient. Conveyor belt misalignment is caused by excessive unidirectional impact. By adjusting the end of the sub-conveyor, the effects before and after adjustment can be offset, or the cumulative speed of the effects before and after adjustment can be reduced. Both of these methods can reduce the number of times the correction device needs to be activated, thus extending the service life of the main conveyor.

[0047] The beneficial effects of this invention are:

[0048] (1) The monitoring and prediction method based on offset coefficient provided by the present invention has low difficulty in obtaining monitoring data. It can effectively monitor and predict ultra-long belt conveyors while saving costs, and can predict the probability of impending deviation failure.

[0049] (2) The offset coefficient of this invention uses the offset amount, impact amount and transport type as reference variables. The larger the offset amount or impact amount, the greater the impact of the transport on the conveyor belt during the impact process, and it can help determine the impact direction on the conveyor belt during the impact process.

[0050] (3) The present invention can change the landing point of the transported material on the main conveyor by rotating the conveyor, thereby adjusting the influence coefficient. The conveyor belt runs off-center because of excessive unidirectional impact. By adjusting the end of the conveyor, the influence before and after the adjustment can be offset or the cumulative speed of the influence before and after the adjustment can be reduced. Both can achieve the purpose of reducing the number of times the correction device is started, thereby improving the service life of the main conveyor. Attached Figure Description

[0051] The invention will now be further described with reference to the accompanying drawings.

[0052] Figure 1 This is a flowchart of the overall steps of the present invention;

[0053] Figure 2 This is a flowchart of the impact value acquisition steps of the present invention;

[0054] Figure 3 This is a flowchart of the offset value acquisition steps of the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Please see Figure 1 As shown, in one embodiment, a fault prediction method for belt conveyors based on intelligent control is provided, including:

[0057] S100. At the intersection of the sub-conveyor and the main conveyor, video monitoring and quality monitoring are carried out on the transported materials on the main conveyor and the sub-conveyor to obtain monitoring data;

[0058] S200. Set the monitoring period and obtain the impact value, offset value and category coefficient based on the monitoring data within the monitoring period. Then, obtain the offset coefficient based on the impact value, offset value and category coefficient. The category coefficient is obtained by obtaining the image of the transported goods through the image data of the video monitoring and distinguishing them. Different types of goods are set with different values, and this value is taken as the category coefficient.

[0059] S300. In multiple consecutive and adjacent monitoring periods, the offset coefficient is compared with the control interval to obtain the judgment result of whether the offset coefficient falls into the control interval in different monitoring periods.

[0060] S400: Obtain the influence coefficient based on multiple judgment results, and then predict the deviation fault based on the influence coefficient.

[0061] This embodiment provides a method for predicting belt misalignment faults. Specifically, the monitoring and prediction method based on the offset coefficient provided by this invention has low difficulty in acquiring monitoring data. For ultra-long belt conveyors, it can effectively monitor and predict the occurrence of belt misalignment faults while saving costs.

[0062] S210, The steps for obtaining the impact value include:

[0063] S211. The quality of the transported goods entering the main conveyor from the sub-conveyor during the monitoring period is obtained through quality monitoring. Different monitoring methods can be used for different transported goods. For example, express delivery can be pre-weighed and the weight of the transported goods can be obtained by label identification. Preferably, the conveyor belt of the sub-conveyor is divided into several parts, and a weighing sensor is installed in each part, so as to sum the weighing values ​​of each section to obtain the required mass.

[0064] S212. Obtain image data of the sub-conveyor during the monitoring period through video monitoring, and determine the height difference between the sub-conveyor and the main conveyor, as well as the type of transported goods and the corresponding category coefficient;

[0065] S213. Set a first correction coefficient based on the category coefficient. The impact value is the product of the first correction value after de-normalization, the square of the weight and height difference of the transported goods entering the main conveyor from the sub-conveyor during the monitoring period, and the first correction coefficient.

[0066] S220, The process of obtaining the offset value includes:

[0067] S221. During the monitoring period, video frames of the main conveyor belt before and after an impact action are monitored by video monitoring. An impact action refers to the process in which the corresponding position of the main conveyor and the sub-conveyor undergoes a quality change and then stabilizes again.

[0068] S222. Obtain the outline image of the goods falling on the main conveyor belt through contour recognition within the video frame;

[0069] S223. Obtain the centroid of the outline of the main conveyor belt before and after an impact action is completed, and obtain the first distance and the second distance from the two centroids to the center line of the main conveyor belt. The center line of the conveyor belt is parallel to the direction of conveyor belt movement. The first distance and the second distance have positive and negative values. The positive and negative values ​​are distinguished as follows: if the centroid is located to the right of the direction vector of conveyor belt movement, a positive value is assigned; if it is located to the left, a negative value is assigned.

[0070] S224. Calculate and obtain the offset value based on the first distance and the second distance.

[0071] The process of calculating and obtaining the offset value includes:

[0072] Through the formula:

[0073]

[0074] Among them, De j is the offset value for the j-th monitoring period, and n is the number of impact actions that occurred during the monitoring period. It is the first distance from the centroid of the main conveyor belt's profile after the i-th impact action is completed, which is the centerline of the main conveyor belt. σ1 is the second distance from the centroid of the main conveyor belt's profile to the centerline of the main conveyor belt after the i-th impact action. σ1 is the first de-normalization coefficient, and δ is a preset correction distance conversion function. The correction distance conversion function is a preset lookup table function used to correct the landing point of transported items that may stack. When there is no stacking problem, ... Both δ and 0 are taken as 0, and i is a positive integer greater than 0 and less than n.

[0075] The process of obtaining the offset coefficients includes:

[0076] Through the formula:

[0077]

[0078] Where Ex is the offset coefficient for the j-th monitoring period, α1 is the preset first weight coefficient, α2 is the preset second weight coefficient, ty is the category coefficient, m is the mass of transported goods entering the main conveyor from the sub-conveyor during the monitoring period, h is the height difference, σ2 is the second de-normalization coefficient, and De0 is the preset standard value of the offset value.

[0079] The process of obtaining the judgment result includes:

[0080] S310. Compare the offset coefficient with a preset pair of critical intervals (-k1, -k2) and (k3, k4);

[0081] S320. If the offset coefficient falls within any critical interval, the output judgment result is that there is a point offset within the monitoring period.

[0082] This embodiment provides a method and steps for obtaining the judgment result. The judgment of the landing point offset of the present invention is based on the offset coefficient. The offset coefficient takes the offset amount, impact amount and transport type as reference variables. The larger the offset amount or impact amount, the greater the impact of the transport on the conveyor belt during the impact process, and it can help to determine the impact direction on the conveyor belt during the impact process.

[0083] The process of obtaining the influence coefficient based on the judgment results includes:

[0084] Through the formula:

[0085]

[0086] Where Inf is the influence coefficient, m1 is the number of monitoring periods judged to have landing point offset, m2 is the number of monitoring periods judged not to have landing point offset, and j is a positive integer greater than 0 and less than m1+m2.

[0087] The process of predicting deviation faults based on the influence coefficient includes:

[0088] S410. Based on the changes in the influence coefficient in historical data, a linear trend model is selected to predict the next influence coefficient and obtain the predicted value. The influence coefficient tends to change linearly over a long period of time, so a linear trend model such as the Hole linear trend model is used for data prediction. The linear trend model is an existing technology, so it will not be elaborated on.

[0089] S420. Compare the predicted value with the safe interval [Inf1, Inf2]. If the predicted value falls within the safe interval, it is predicted that there is no possibility of conveyor belt misalignment.

[0090] S430. If the predicted value is not within the safe range, it is predicted that a conveyor belt misalignment fault may occur.

[0091] This embodiment provides a process for predicting conveyor belt misalignment faults. In the process of misalignment prediction, it is necessary to use the influence coefficient obtained in time series as the basis for prediction, so as to achieve the purpose of early prediction and facilitate the implementation of subsequent maintenance or adjustment plans.

[0092] The end of the sub-conveyor is configured to rotate between [a°, b°], and the rotation angle is set to control the rotation of the end of the sub-conveyor to change the landing point of the transported items on the sub-conveyor on the main conveyor.

[0093] If the sign of the influence coefficient obtained after rotation is different from that obtained before rotation, then stop rotating until the sum of the influence coefficient before rotation and the influence coefficient after multiple rotations has changed from positive to negative, and then return to the initial state.

[0094] If the sign of the influence coefficient obtained after rotation is the same as that obtained before rotation, then continue to rotate in the same direction with the same rotation angle.

[0095] It should be noted that the combination of influence coefficient and rotation adjustment can also be used to offset environmental effects. For example, the influence of different wind forces and wind directions on the conveyor belt can be described by data, such as by vectors. Then, a comparison relationship between the data description and the influence coefficient can be established. After establishing the comparison relationship, different influence coefficients are output under different wind conditions. The influence coefficient is added to the influence coefficient obtained in step S400 as a new influence coefficient and applied to the adjustment process at the end of the sub-conveyor.

[0096] This embodiment provides a method for controlling the adjustment of an intelligent conveyor belt. The end of the sub-conveyor of this invention is configured as a rotatable structure. By rotating, the landing point of the transported items on the sub-conveyor on the main conveyor can be changed, thereby adjusting the influence coefficient. Conveyor belt deviation is caused by excessive unidirectional impact. By adjusting the end of the sub-conveyor, the effects before and after adjustment can be offset or the cumulative speed of the effects before and after adjustment can be reduced. Both can achieve the purpose of reducing the number of times the correction device is activated, thereby improving the service life of the main conveyor.

[0097] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A fault prediction method for belt conveyors based on intelligent control, characterized in that, Includes the following steps: At the intersection of the sub-conveyor and the main conveyor, video monitoring and quality monitoring are carried out on the transported materials on the main conveyor and the sub-conveyor to obtain monitoring data; Set a monitoring period and obtain the impact value, offset value and category coefficient based on the monitoring data within the monitoring period. Then obtain the offset coefficient based on the impact value, offset value and category coefficient. The steps to obtain the impact value include: The quality of transported goods entering the main conveyor from the sub-conveyor during the monitoring period is obtained through quality monitoring; By acquiring image data of the sub-conveyors during the monitoring period through video monitoring, the height difference between the sub-conveyors and the main conveyor, as well as the type of transported goods and the corresponding category coefficient, are determined. The first correction factor is set based on the category coefficient. The impact value is the product of the first correction value after de-normalization, the square of the weight and height difference of the transported goods entering the main conveyor from the sub-conveyor during the monitoring period, and the first correction factor. The process of obtaining the offset coefficients includes: Through the formula: Among them, De j is the offset value of the j-th monitoring period, Ex is the offset coefficient of the j-th monitoring period, α1 is the preset first weight coefficient, α2 is the preset second weight coefficient, ty is the category coefficient, m is the mass of transported goods from the sub-conveyor to the main conveyor during the monitoring period, h is the height difference, σ2 is the second de-normalization coefficient, and De0 is the preset standard value of the offset value. In multiple consecutive and adjacent monitoring periods, the offset coefficient is compared with the control interval to obtain the judgment result of whether the offset coefficient falls within the control interval in different monitoring periods; The influence coefficient is obtained based on multiple judgment results, and then the deviation fault is predicted based on the influence coefficient. The process of obtaining the influence coefficient based on the judgment results includes: Through the formula: Where Inf is the influence coefficient, m1 is the number of monitoring periods judged to have landing point offset, m2 is the number of monitoring periods judged not to have landing point offset, and j is a positive integer greater than 0 and less than m1+m2.

2. The fault prediction method for belt conveyors based on intelligent control according to claim 1, characterized in that, The process of obtaining the offset value includes: During the monitoring period, video frames of the main conveyor belt before and after an impact action are monitored. Within a video frame, contour recognition is used to obtain the contour image of the goods that have fallen onto the main conveyor belt; Obtain the centroid of the profile of the main conveyor belt before and after an impact action is completed, and obtain the first and second distances from the two centroids to the centerline of the main conveyor belt. The offset value is calculated based on the first distance and the second distance.

3. The fault prediction method for belt conveyors based on intelligent control according to claim 2, characterized in that, The process of calculating and obtaining the offset value includes: Through the formula: Among them, De j is the offset value for the j-th monitoring period, and n is the number of impact actions that occurred during the monitoring period. It is the first distance from the centroid of the main conveyor belt's profile after the i-th impact action is completed, which is the centerline of the main conveyor belt. σ1 is the second distance from the centroid of the main conveyor belt profile after the i-th impact action is completed, the centerline of the main conveyor belt, σ1 is the first de-normalization coefficient, δ is the preset correction distance conversion function, and i is a positive integer greater than 0 and less than n.

4. The fault prediction method for belt conveyors based on intelligent control according to claim 1, characterized in that, The process of obtaining multiple judgment results includes: The offset coefficient is compared with a pre-defined pair of critical intervals (-k1, -k2) and (k3, k4); If the offset coefficient falls within any critical interval, the output judgment result is that there is a point offset within the monitoring period.

5. The fault prediction method for belt conveyors based on intelligent control according to claim 1, characterized in that, The process of predicting deviation faults based on the influence coefficient includes: Based on the changes in the influence coefficient in historical data, a linear trend model is selected to predict the next influence coefficient and obtain the predicted value. The predicted value is compared with the safe range. If the predicted value falls within the safe range, it is predicted that there is no possibility of conveyor belt misalignment. If the predicted value is not within the safe range, it indicates that a conveyor belt misalignment fault is likely to occur.

6. The fault prediction method for belt conveyors based on intelligent control according to claim 1, characterized in that, include: The end of the sub-conveyor is configured to rotate between [a°, b°], and the rotation angle is set. If the sign of the influence coefficient obtained after rotation is different from that obtained before rotation, then stop rotating until the sum of the influence coefficient before rotation and the influence coefficient after multiple rotations has changed from positive to negative, and then return to the initial state. If the sign of the influence coefficient obtained after rotation is the same as that obtained before rotation, then continue to rotate in the same direction with the same rotation angle.

Images