Intelligent coal flow management system based on visual AI autonomous decision

Abstract

The invention discloses an intelligent coal flow management system based on visual AI autonomous decision, and relates to the technical field of coal flow management, and the system comprises the following modules: a visual monitoring module which carries out the picture shooting of a coal flow on a conveyor belt to obtain a plurality of continuous reference pictures, carries out the visual analysis of the plurality of reference pictures to obtain a visual coal flow coefficient, and sends the visual coal flow coefficient to a database; establishing a visual database, and incorporating the visual coal flow coefficient and the reference picture into the visual database; the visual detection module is used for shooting pictures of coal flow, and visual analysis is carried out on a plurality of reference pictures to obtain a visual coal flow coefficient, so that the quantity of coal on the conveyor belt can be predicted through the visual pictures, and the rotating speed of a conveyor belt driving motor can be adjusted subsequently according to the predicted quantity of coal; and the conveying speed of the conveying belt can be adjusted according to the actual coal conveying quantity.

Description

Technical Field

[0001] The present invention relates to the technical field of coal flow management, and in particular to an intelligent coal flow management system based on visual AI autonomous decision-making. Background Art

[0002] With the rapid development of my country's industry, underground coal energy mining has become an important research direction. Coal flow refers to the flow of coal after it is crushed and transported through conveyor belts. The smoothness of coal flow transportation is directly related to the production efficiency and supply stability of coal. Efficient coal flow management can ensure that coal is delivered to the demand side in a timely and accurate manner, providing strong guarantees for industrial production and energy supply. At the same time, with the continuous advancement of technology, modern coal flow monitoring and control systems are also constantly developing to improve the safety and reliability of coal flow.

[0003] There are the following problems in the process of coal transportation through conveyor belts:

[0004] 1. The conveyor belt for transporting coal flow is driven by a motor. However, the amount of coal piled on the conveyor belt is not constant. With the change of the progress of coal mining and crushing in the mine, the amount of coal on the conveyor belt is sometimes more, sometimes less, and sometimes even no coal is transported. A large amount of coal on the conveyor belt means that the efficiency of coal mining and crushing is high at this time. Therefore, it is necessary to speed up the transportation and transmission of coal to prevent the blockage of coal transportation after mining. A small amount of coal on the conveyor belt means that the efficiency of coal mining and crushing is low at this time. If the conveyor belt still maintains a high-speed transportation state, it will not only waste electricity but also waste the service life of the conveyor belt and the rotating roller due to friction.

[0005] 2. During the transportation of coal flow, the coal moves with the conveyor belt and is shaken, and the accumulated coal will roll down. Therefore, some coal will approach the edge of the conveyor belt or even roll off the conveyor belt. The rolled coal is not only not transported effectively, but also has the risk of injuring workers or mining equipment. In order to facilitate the intelligent management of the transportation speed of the coal flow conveyor belt and to prevent the coal from rolling off the conveyor belt, we propose an intelligent coal flow management system based on visual AI autonomous decision-making. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In view of the deficiencies in the prior art, the present invention provides an intelligent coal flow management system based on visual AI autonomous decision-making to solve the above-mentioned problems in the prior art.

[0008] (II) Technical solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent coal flow management system based on visual AI autonomous decision-making, comprising the following modules:

[0010] The visual monitoring module takes pictures of the coal flow on the conveyor belt to obtain multiple continuous reference pictures, obtains the visual coal flow coefficient by visual analysis of the multiple reference pictures, establishes a visual database, and incorporates the visual coal flow coefficient and the reference pictures into the visual database;

[0011] The data correction module obtains the running speed of the conveyor belt, obtains the instantaneous power of the driving motor, obtains the conveyor belt load value through the running speed of the conveyor belt and the instantaneous power of the driving motor, sets a preset threshold value of the conveyor belt load value, obtains the load difference coefficient by subtracting the conveyor belt load value from the preset threshold value of the conveyor belt load value, and obtains the coal flow transport coefficient through the load difference coefficient and the visual coal flow coefficient;

[0012] The speed control module sets a preset threshold value of the coal flow transport coefficient and adjusts the speed of the driving motor by comparing the coal flow transport coefficient with the preset threshold value of the coal flow transport coefficient;

[0013] The offset alarm module obtains the coal edge distance in the reference image, sets the edge safety distance preset threshold, and determines whether it is necessary to turn on the alarm to remind the management personnel by comparing the coal edge distance with the edge safety distance preset threshold.

[0014] Preferably, in the visual monitoring module, a visual coal flow coefficient is obtained by visually analyzing a plurality of reference images, specifically:

[0015] Step 1: Classify all reference images according to the top-view angle and the side-view angle, then grayscale all reference images, obtain the brightness information of each reference image after grayscale processing, set the brightness preset threshold range, capture the brightness area within the brightness preset threshold range in each top-view reference image and mark it as the coal top-view area, and capture the brightness area within the brightness preset threshold range in each side-view reference image and mark it as the coal side-view area;

[0016] Step 2: Perform repeated area comparison and deletion on all coal top-view areas to obtain the final coal top-view area, obtain the area of ​​each final coal top-view area respectively, perform repeated area comparison and deletion on all coal side-view areas to obtain the final coal side-view area respectively, obtain the height of each final coal side-view area respectively, arrange the heights of all final coal side-view areas in ascending order and select the maximum value to mark as the final coal side-view area height;

[0017] Step 3: Set the preset threshold of area ratio, and obtain the visual coal flow coefficient through the preset threshold of area ratio, the area of ​​all final coal overlooking areas and the height of the final coal side viewing area.

[0018] Preferably, the calculation method of the visual coal flow coefficient is as follows:

[0019]

[0020] Among them, Sx represents the visual coal flow coefficient, Mj represents the preset threshold of the area ratio, Fs represents the area of ​​the final coal overlooking area, Es represents the height of the final coal side viewing area, and n represents the number of areas of the final coal overlooking area, n = 1, 2, 3,...

[0021] Preferably, the running speed of the conveyor belt is obtained in the data correction module, specifically:

[0022] Step 1: Obtain two consecutive reference images taken in the visual monitoring module, obtain the shooting interval, and identify and capture multiple common reference objects in the two reference images by brightness, contrast, area and shape;

[0023] Step 2: Overlap the two reference images, measure the distance between each group of common reference objects to obtain multiple distance values, arrange the multiple distance values ​​in ascending order and delete the minimum and maximum values, sum and average the remaining multiple distance values ​​to obtain the reference distance, set the distance ratio preset threshold, and obtain the running speed of the conveyor belt through the reference distance, the distance ratio preset threshold and the shooting time interval.

[0024] Preferably, the calculation method of the running speed of the conveyor belt is specifically as follows:

[0025]

[0026] Wherein, Sd represents the running speed of the conveyor belt, Cj represents the reference distance, Jy represents the preset threshold of the distance ratio, and Sj represents the shooting time interval.

[0027] Preferably, in the data correction module, the conveyor belt load value is obtained by the running speed of the conveyor belt and the instantaneous power of the driving motor, specifically:

[0028] Step 1: obtaining the running speeds of multiple conveyor belts at different moments, obtaining the instantaneous powers of multiple corresponding driving motors, and obtaining the traction forces of multiple conveyor belts by dividing the instantaneous powers of multiple driving motors by the running speeds of multiple corresponding conveyor belts at different moments;

[0029] Step 2: Arrange the traction forces of multiple conveyor belts in ascending order and take the median to obtain the reference traction force, obtain the friction coefficient between the conveyor belt and the coal flow, and obtain the conveyor belt load value by dividing the reference traction force by the friction coefficient between the conveyor belt and the coal flow.

[0030] Preferably, the calculation method of the coal flow transport coefficient is as follows:

[0031]

[0032] Among them, My represents the coal flow transportation coefficient, Sx represents the visual coal flow coefficient, and Fh represents the load difference coefficient. and are weights,

[0033] Preferably, the speed control module is specifically:

[0034] Step 1: Obtain the coal flow transport coefficient, set a preset threshold value of the coal flow transport coefficient, and obtain a transport coefficient difference by subtracting the coal flow transport coefficient from the preset threshold value of the coal flow transport coefficient;

[0035] Step 2: Set the comparison period to 10 seconds, set the drive motor speed adjustment range, and compare the transport coefficient difference with 0 every 10 seconds. If the transport coefficient difference is greater than 0, the conveyor belt drive motor is controlled to increase the speed based on the drive motor speed adjustment range. If the transport coefficient difference is less than 0, the conveyor belt drive motor is controlled to reduce the speed based on the drive motor speed adjustment range.

[0036] Preferably, the deviation alarm module is specifically:

[0037] Step 1: Obtain a reference image in the visual monitoring module, obtain the coal overlooking area in the reference image after grayscale processing, obtain the brightness of the conveyor belt edge, capture the conveyor belt edge position in the reference image after grayscale processing through the conveyor belt edge brightness, and measure the distance between the coal overlooking area and the conveyor belt edge position to obtain multiple margin values;

[0038] Step 2: setting a preset threshold value for the margin ratio, obtaining multiple margin reference values ​​by multiplying the preset threshold value for the margin ratio by multiple margin values, arranging the multiple margin reference values ​​in ascending order, and obtaining a minimum margin value and a majority margin value of the margin reference values;

[0039] Step 3: Set the preset threshold of edge safety distance, and compare the preset threshold of edge safety distance with the minimum margin value and the majority margin value respectively. If both the minimum margin value and the majority margin value are greater than the preset threshold of edge safety distance, the coal position is marked as safe. If both the minimum margin value and the majority margin value are less than the preset threshold of edge safety distance, the coal position is marked as dangerous and an alarm is issued to alert the management personnel.

[0040] (III) Beneficial effects

[0041] The present invention provides an intelligent coal flow management system based on visual AI autonomous decision-making, which has the following beneficial effects:

[0042] (1) In this scheme, the coal flow is photographed by the visual detection module, and the visual coal flow coefficient is obtained by visual analysis of multiple reference images, so that the amount of coal on the conveyor belt can be predicted by the visual image, and then the speed of the conveyor belt drive motor can be adjusted according to the predicted amount of coal, and the conveyor belt transportation speed can be adjusted according to the actual amount of coal transported. This is beneficial to speed up the transportation efficiency when the amount of coal is large, and it is also convenient to reduce the ineffective workload of the conveyor belt and the drive motor when the amount of coal is low, so as to avoid wasting electricity while wasting the service life of the conveyor belt and the roller, and then it is convenient to intelligently manage the transportation speed of the coal flow conveyor belt in coal flow management.

[0043] (2) In this scheme, the data correction module is used to combine the running speed of the conveyor belt and the instantaneous power of the driving motor to obtain the load value of the conveyor belt, and then the load difference coefficient is obtained by subtracting the conveyor belt load value from the preset threshold value of the conveyor belt load value. Finally, the coal flow transportation coefficient is obtained by combining the load difference coefficient with the visual coal flow coefficient, so as to facilitate a more accurate prediction of the coal quantity by combining the visual monitoring information, the conveyor belt speed and the driving motor power, which is beneficial to improve the accuracy of the coal quantity prediction results, and then facilitate adjusting the speed of the driving motor according to the prediction results, and then facilitate adjusting the transportation speed according to the actual coal transportation quantity.

[0044] (3) In this solution, the reference image in the visual inspection module is further analyzed through the offset alarm module to estimate the distance between the coal and the edge of the conveyor belt. When the distance is too small, the alarm is turned on to remind the management personnel to intervene and clean up the coal at the edge, which is helpful to prevent the coal from falling from the conveyor belt during transportation and injuring workers or mining equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a structural schematic diagram of an intelligent coal flow management system based on visual AI autonomous decision-making in the present invention;

[0046] Figure 2 This is a flow chart of an intelligent coal flow management system based on visual AI autonomous decision-making according to the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] See also Figure 1-Figure 2The present invention provides an intelligent coal flow management system based on visual AI autonomous decision-making, including the following modules:

[0049] The visual monitoring module takes pictures of the coal flow on the conveyor belt to obtain multiple continuous reference pictures, obtains the visual coal flow coefficient by visual analysis of the multiple reference pictures, establishes a visual database, and incorporates the visual coal flow coefficient and the reference pictures into the visual database;

[0050] The data correction module obtains the running speed of the conveyor belt, obtains the instantaneous power of the driving motor, obtains the conveyor belt load value through the running speed of the conveyor belt and the instantaneous power of the driving motor, sets a preset threshold value of the conveyor belt load value, obtains the load difference coefficient by subtracting the conveyor belt load value from the preset threshold value of the conveyor belt load value, and obtains the coal flow transport coefficient through the load difference coefficient and the visual coal flow coefficient;

[0051] The speed control module sets a preset threshold value of the coal flow transport coefficient and adjusts the speed of the driving motor by comparing the coal flow transport coefficient with the preset threshold value of the coal flow transport coefficient;

[0052] The offset alarm module obtains the coal edge distance in the reference image, sets the preset threshold of the edge safety distance, and determines whether it is necessary to open the alarm to remind the management personnel by comparing the coal edge distance with the preset threshold of the edge safety distance;

[0053] The calculation method of coal flow transport coefficient is as follows:

[0054]

[0055] Among them, My represents the coal flow transportation coefficient, Sx represents the visual coal flow coefficient, and Fh represents the load difference coefficient. and are weights,

[0056] In this embodiment, in this scheme, the coal flow is photographed by a visual detection module, and a visual coal flow coefficient is obtained by visually analyzing multiple reference images, so that the amount of coal on the conveyor belt is predicted by the visual image, and then the speed of the conveyor belt drive motor is adjusted according to the predicted amount of coal, and then the conveyor belt transportation speed is adjusted according to the actual amount of coal transported, which is beneficial to speed up the transportation efficiency when the amount of coal is large, and is also convenient to reduce the ineffective workload of the conveyor belt and the drive motor when the amount of coal is low, so as to avoid wasting electricity while wasting the service life of the conveyor belt and the rotating roller, and then it is convenient to intelligently manage the transportation speed of the coal flow conveyor belt in coal flow management, and by establishing a visual database, it is convenient to calibrate and correct the predicted amount of coal through historical reference images and historical visual coal flow coefficients in the subsequent prediction process, so as to facilitate more accurate prediction of the coal quantity in the subsequent prediction;

[0057] In this scheme, the data correction module is used to combine the running speed of the conveyor belt and the instantaneous power of the driving motor to obtain the load value of the conveyor belt, and then the load difference coefficient is obtained by subtracting the conveyor belt load value from the preset threshold value of the conveyor belt load value, and finally the coal flow transportation coefficient is obtained by combining the load difference coefficient with the visual coal flow coefficient, so as to facilitate a more accurate prediction of the coal quantity by combining the visual monitoring information, the conveyor belt speed and the driving motor power, which is beneficial to improve the accuracy of the coal quantity prediction results, and then it is convenient to adjust the speed of the driving motor according to the prediction results, and then it is convenient to adjust the transportation speed according to the actual coal transportation quantity, which is further beneficial to speed up the transportation efficiency when the amount of coal is large, and further facilitate to reduce the ineffective workload of the conveyor belt and the driving motor when the amount of coal is low, which is further beneficial to avoid wasting electricity while wasting the service life of the conveyor belt and the rotating roller, and further facilitate intelligent management of the transportation speed of the coal flow conveyor belt in coal flow management;

[0058] In this scheme, the predicted coal flow transport coefficient is compared through the speed control module, and the speed of the driving motor is regulated according to the comparison result, which is beneficial to speed up the transportation efficiency when the amount of coal is large, and is also convenient to reduce the ineffective workload of the conveyor belt and the driving motor when the amount of coal is low, thereby avoiding wasting electricity while wasting the service life of the conveyor belt and the rotating roller, and then facilitating the intelligent management of the transportation speed of the coal flow conveyor belt in the coal flow management;

[0059] In this solution, the reference image in the visual inspection module is further analyzed by the offset alarm module to estimate the distance between the coal and the edge of the conveyor belt. When the distance is too small, the alarm is turned on to remind the management personnel to intervene and clean up the coal at the edge, which is helpful to prevent the coal from falling from the conveyor belt during transportation and injuring workers or mining equipment.

[0060] It is worth mentioning that the weight values ​​in this scheme can be obtained through the hierarchical analysis method, and the values ​​of the preset thresholds in this scheme can be obtained through the weight analysis method, which will not be elaborated here.

[0061] In the visual monitoring module, the visual coal flow coefficient is obtained by visually analyzing multiple reference images, specifically:

[0062] Step 1: Classify all reference images according to the top-view angle and the side-view angle, then grayscale all reference images, obtain the brightness information of each reference image after grayscale processing, set the brightness preset threshold range, capture the brightness area within the brightness preset threshold range in each top-view reference image and mark it as the coal top-view area, and capture the brightness area within the brightness preset threshold range in each side-view reference image and mark it as the coal side-view area;

[0063] Step 2: Perform repeated area comparison and deletion on all coal top-view areas to obtain the final coal top-view area, obtain the area of ​​each final coal top-view area respectively, perform repeated area comparison and deletion on all coal side-view areas to obtain the final coal side-view area respectively, obtain the height of each final coal side-view area respectively, arrange the heights of all final coal side-view areas in ascending order and select the maximum value to mark as the final coal side-view area height;

[0064] Step 3: setting a preset threshold value for the area ratio, and obtaining a visual coal flow coefficient through the preset threshold value for the area ratio, the area of ​​all final coal overlooking areas, and the height of the final coal side-viewing area;

[0065] The calculation method of the visual coal flow coefficient is as follows:

[0066]

[0067] Among them, Sx represents the visual coal flow coefficient, Mj represents the preset threshold of the area ratio, Fs represents the area of ​​the final coal overlooking area, Es represents the height of the final coal side viewing area, and n represents the number of areas of the final coal overlooking area, n = 1, 2, 3,...

[0068] In this embodiment, the captured image is processed in grayscale, the coal in the image is identified by brightness, the coal area is identified by a bird's-eye view, and the coal height is identified by a side view to calculate the volume of the coal. This makes it easier to predict the amount of coal on the conveyor belt through visual monitoring, and then to adjust the speed of the drive motor according to the prediction results, thereby adjusting the coal transportation efficiency according to the actual amount of coal transported.

[0069] The running speed of the conveyor belt is obtained in the data correction module, specifically:

[0070] Step 1: Obtain two consecutive reference images taken in the visual monitoring module, obtain the shooting interval, and identify and capture multiple common reference objects in the two reference images by brightness, contrast, area and shape;

[0071] Step 2: Overlap the two reference images, measure the distance between each group of common reference objects to obtain multiple distance values, arrange the multiple distance values ​​in ascending order and delete the minimum and maximum values, sum and average the remaining multiple distance values ​​to obtain the reference distance, set the distance ratio preset threshold, and obtain the running speed of the conveyor belt through the reference distance, the distance ratio preset threshold and the shooting time interval;

[0072] The calculation method of the conveyor belt running speed is as follows:

[0073]

[0074] Wherein, Sd represents the running speed of the conveyor belt, Cj represents the reference distance, Jy represents the preset threshold of the distance ratio, and Sj represents the shooting time interval.

[0075] In this embodiment, since the position of the shooting camera is fixed, the reference object is determined by comparing continuous pictures, which makes it convenient to calculate the moving speed of the conveyor belt according to the shooting time interval and the moving distance of the reference object, and further facilitates the subsequent accurate prediction of the amount of coal transported on the conveyor belt.

[0076] In the data correction module, the conveyor belt load value is obtained by the running speed of the conveyor belt and the instantaneous power of the drive motor, specifically:

[0077] Step 1: obtaining the running speeds of multiple conveyor belts at different moments, obtaining the instantaneous powers of multiple corresponding driving motors, and obtaining the traction forces of multiple conveyor belts by dividing the instantaneous powers of multiple driving motors by the running speeds of multiple corresponding conveyor belts at different moments;

[0078] Step 2: Arrange the traction forces of multiple conveyor belts in ascending order and take the median to obtain the reference traction force, obtain the friction coefficient between the conveyor belt and the coal flow, and obtain the conveyor belt load value by dividing the reference traction force by the friction coefficient between the conveyor belt and the coal flow.

[0079] In this embodiment, the running speed of the conveyor belt and the instantaneous power of the drive motor are combined to obtain the load value of the conveyor belt, so as to facilitate the prediction of the actual amount of coal transported on the conveyor belt, so as to facilitate the subsequent adjustment of the rotation speed of the conveyor belt drive motor according to the predicted amount of coal, and then to facilitate the adjustment of the transport speed of the conveyor belt according to the actual amount of coal transported, which is beneficial to speed up the transportation efficiency when the amount of coal is large, and is also convenient for reducing the ineffective workload of the conveyor belt and the drive motor when the amount of coal is low, so as to help avoid wasting electricity while wasting the service life of the conveyor belt and the rotating roller, and then facilitate the intelligent management of the transport speed of the coal flow conveyor belt in coal flow management.

[0080] Speed ​​control module, specifically:

[0081] Step 1: Obtain the coal flow transport coefficient, set a preset threshold value of the coal flow transport coefficient, and obtain a transport coefficient difference by subtracting the coal flow transport coefficient from the preset threshold value of the coal flow transport coefficient;

[0082] Step 2: Set the comparison period to 10 seconds, set the drive motor speed adjustment range, and compare the transport coefficient difference with 0 every 10 seconds. If the transport coefficient difference is greater than 0, the conveyor belt drive motor is controlled to increase the speed based on the drive motor speed adjustment range. If the transport coefficient difference is less than 0, the conveyor belt drive motor is controlled to reduce the speed based on the drive motor speed adjustment range.

[0083] In this embodiment, by setting the comparison period and the speed adjustment range, it is convenient to detect the coal in a periodic cycle and adjust the speed of the drive motor, thereby facilitating the intelligent adjustment of the coal transportation speed according to the actual coal transportation volume.

[0084] The deviation alarm module is as follows:

[0085] Step 1: Obtain a reference image in the visual monitoring module, obtain the coal overlooking area in the reference image after grayscale processing, obtain the brightness of the conveyor belt edge, capture the conveyor belt edge position in the reference image after grayscale processing through the conveyor belt edge brightness, and measure the distance between the coal overlooking area and the conveyor belt edge position to obtain multiple margin values;

[0086] Step 2: setting a preset threshold value for the margin ratio, obtaining multiple margin reference values ​​by multiplying the preset threshold value for the margin ratio by multiple margin values, arranging the multiple margin reference values ​​in ascending order, and obtaining a minimum margin value and a majority margin value of the margin reference values;

[0087] Step 3: Set the preset threshold of edge safety distance, and compare the preset threshold of edge safety distance with the minimum margin value and the majority margin value respectively. If both the minimum margin value and the majority margin value are greater than the preset threshold of edge safety distance, the coal position is marked as safe. If both the minimum margin value and the majority margin value are less than the preset threshold of edge safety distance, the coal position is marked as dangerous and an alarm is issued to alert the management personnel.

[0088] In this embodiment, the reference image in the visual inspection module is further analyzed to estimate the distance between the coal and the edge of the conveyor belt. When the distance is too small, an alarm is activated to remind the management staff to intervene and clean up the coal at the edge, thereby preventing the coal from falling from the conveyor belt and injuring workers or mining equipment during transportation.

[0089] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0090] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0091] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. An intelligent coal flow management system based on visual AI autonomous decision-making, characterized in that: Includes the following modules: The visual monitoring module takes pictures of the coal flow on the conveyor belt to obtain multiple continuous reference pictures, obtains the visual coal flow coefficient by visual analysis of the multiple reference pictures, establishes a visual database, and incorporates the visual coal flow coefficient and the reference pictures into the visual database; The data correction module obtains the running speed of the conveyor belt, obtains the instantaneous power of the driving motor, obtains the conveyor belt load value through the running speed of the conveyor belt and the instantaneous power of the driving motor, sets a preset threshold value of the conveyor belt load value, obtains the load difference coefficient by subtracting the conveyor belt load value from the preset threshold value of the conveyor belt load value, and obtains the coal flow transport coefficient through the load difference coefficient and the visual coal flow coefficient; The speed control module sets a preset threshold value of the coal flow transport coefficient and adjusts the speed of the driving motor by comparing the coal flow transport coefficient with the preset threshold value of the coal flow transport coefficient; The offset alarm module obtains the coal edge distance in the reference image, sets the edge safety distance preset threshold, and determines whether it is necessary to turn on the alarm to remind the management personnel by comparing the coal edge distance with the edge safety distance preset threshold.

2. According to claim 1, an intelligent coal flow management system based on visual AI autonomous decision-making is characterized by: In the visual monitoring module, the visual coal flow coefficient is obtained by visually analyzing multiple reference images, specifically: Step 1: Classify all reference images according to the top-view angle and the side-view angle, then grayscale all reference images, obtain the brightness information of each reference image after grayscale processing, set the brightness preset threshold range, capture the brightness area within the brightness preset threshold range in each top-view reference image and mark it as the coal top-view area, and capture the brightness area within the brightness preset threshold range in each side-view reference image and mark it as the coal side-view area; Step 2: Perform repeated area comparison and deletion on all coal top-view areas to obtain the final coal top-view area, obtain the area of ​​each final coal top-view area respectively, perform repeated area comparison and deletion on all coal side-view areas to obtain the final coal side-view area respectively, obtain the height of each final coal side-view area respectively, arrange the heights of all final coal side-view areas in ascending order and select the maximum value to mark as the final coal side-view area height; Step 3: Set the preset threshold of area ratio, and obtain the visual coal flow coefficient through the preset threshold of area ratio, the area of ​​all final coal overlooking areas and the height of the final coal side viewing area.

3. According to claim 2, an intelligent coal flow management system based on visual AI autonomous decision-making is characterized by: The calculation method of the visual coal flow coefficient is as follows: Among them, Sx represents the visual coal flow coefficient, Mj represents the preset threshold of the area ratio, Fs represents the area of ​​the final coal overlooking area, Es represents the height of the final coal side viewing area, and n represents the number of areas of the final coal overlooking area, n = 1, 2, 3,...

4. According to claim 1, the intelligent coal flow management system based on visual AI autonomous decision-making is characterized by: The running speed of the conveyor belt is obtained in the data correction module, specifically: Step 1: Obtain two consecutive reference images taken in the visual monitoring module, obtain the shooting interval, and identify and capture multiple common reference objects in the two reference images by brightness, contrast, area and shape; Step 2: Overlap the two reference images, measure the distance between each group of common reference objects to obtain multiple distance values, arrange the multiple distance values ​​in ascending order and delete the minimum and maximum values, sum and average the remaining multiple distance values ​​to obtain the reference distance, set the distance ratio preset threshold, and obtain the running speed of the conveyor belt through the reference distance, the distance ratio preset threshold and the shooting time interval.

5. According to claim 4, an intelligent coal flow management system based on visual AI autonomous decision-making is characterized by: The calculation method of the conveyor belt running speed is as follows: Wherein, Sd represents the running speed of the conveyor belt, Cj represents the reference distance, Jy represents the preset threshold of the distance ratio, and Sj represents the shooting time interval.

6. The intelligent coal flow management system based on visual AI autonomous decision-making according to claim 1 is characterized by: In the data correction module, the conveyor belt load value is obtained by the running speed of the conveyor belt and the instantaneous power of the drive motor, specifically: Step 1: obtaining the running speeds of multiple conveyor belts at different moments, obtaining the instantaneous powers of multiple corresponding driving motors, and obtaining the traction forces of multiple conveyor belts by dividing the instantaneous powers of multiple driving motors by the running speeds of multiple corresponding conveyor belts at different moments; Step 2: Arrange the traction forces of multiple conveyor belts in ascending order and take the median to obtain the reference traction force, obtain the friction coefficient between the conveyor belt and the coal flow, and obtain the conveyor belt load value by dividing the reference traction force by the friction coefficient between the conveyor belt and the coal flow.

7. The intelligent coal flow management system based on visual AI autonomous decision-making according to claim 1 is characterized by: The calculation method of coal flow transport coefficient is as follows: Among them, My represents the coal flow transportation coefficient, Sx represents the visual coal flow coefficient, and Fh represents the load difference coefficient. and are weights, 8. The intelligent coal flow management system based on visual AI autonomous decision-making according to claim 1 is characterized by: Speed ​​control module, specifically: Step 1: Obtain the coal flow transport coefficient, set a preset threshold value of the coal flow transport coefficient, and obtain a transport coefficient difference by subtracting the coal flow transport coefficient from the preset threshold value of the coal flow transport coefficient; Step 2: Set the comparison period to 10 seconds, set the drive motor speed adjustment range, and compare the transport coefficient difference with 0 every 10 seconds. If the transport coefficient difference is greater than 0, the conveyor belt drive motor is controlled to increase the speed based on the drive motor speed adjustment range. If the transport coefficient difference is less than 0, the conveyor belt drive motor is controlled to reduce the speed based on the drive motor speed adjustment range.

9. The intelligent coal flow management system based on visual AI autonomous decision-making according to claim 2 is characterized by: The deviation alarm module is as follows: Step 1: Obtain a reference image in the visual monitoring module, obtain the coal overlooking area in the reference image after grayscale processing, obtain the brightness of the conveyor belt edge, capture the conveyor belt edge position in the reference image after grayscale processing through the conveyor belt edge brightness, and measure the distance between the coal overlooking area and the conveyor belt edge position to obtain multiple margin values; Step 2: setting a preset threshold value for the margin ratio, obtaining multiple margin reference values ​​by multiplying the preset threshold value for the margin ratio by multiple margin values, arranging the multiple margin reference values ​​in ascending order, and obtaining a minimum margin value and a majority margin value of the margin reference values; Step 3: Set the preset threshold of edge safety distance, and compare the preset threshold of edge safety distance with the minimum margin value and the majority margin value respectively. If both the minimum margin value and the majority margin value are greater than the preset threshold of edge safety distance, the coal position is marked as safe. If both the minimum margin value and the majority margin value are less than the preset threshold of edge safety distance, the coal position is marked as dangerous and an alarm is issued to alert the management personnel.

Images