Intelligent coal mine sorting treatment equipment based on machine learning

Abstract

The invention relates to the technical field of coal mine sorting, and discloses intelligent coal mine sorting treatment equipment based on machine learning, the intelligent coal mine sorting treatment equipment comprises a sorting equipment main body, the top of the sorting equipment main body is provided with a feeding cylinder, the side surface of the feeding cylinder is provided with a driving motor, and the output end of the driving motor is fixedly connected with a rotating rod. According to the intelligent coal mine sorting treatment equipment based on machine learning, through cooperative use of a feeding cylinder and a driving motor, the output end of the driving motor can drive a rotating rod to rotate, the rotating rotating rod can drive a smashing roller to smash coal, and through cooperative use of a supporting frame and an air cylinder, the coal is smashed; according to the coal pulverizing device, the air cylinder can drive the transverse pipe to ascend and descend, meanwhile, the transverse pipe can be limited through the installed fixing column and the installed connecting block, displacement of the transverse pipe during ascending and descending is avoided, a water storage tank and a first water pump are used in cooperation, so that water is conveniently supplied to the interior of the transverse pipe through a corrugated pipe, and dust falling can be conducted during coal pulverizing.

Description

Technical Field

[0001] The present invention relates to the technical field of coal mine sorting, and in particular to intelligent coal mine sorting and processing equipment based on machine learning. Background Art

[0002] In the field of coal sorting, the existing technology mainly uses traditional methods such as mechanical screening, vibration classification and simple conveying devices to separate impurities and gangue from raw coal. This type of equipment uses the combination of grading screens and vibration motors to achieve preliminary separation of materials of different particle sizes.

[0003] However, this traditional device has the following significant defects in practical application:

[0004] During the coal sorting process, a large amount of dust is generated due to material crushing, vibration screening and other links. Traditional equipment lacks effective dust control measures. These dusts float in the air, not only polluting the working environment, but also threatening the health of operators. At the same time, floating dust may accumulate inside the equipment, increasing the difficulty of equipment maintenance.

[0005] The separation of coal and impurities by traditional mechanical screening equipment depends on the difference in material particle size. When the particle size of the impurities is close to that of the coal or the shape is irregular, the sorting efficiency drops significantly. In addition, traditional equipment is difficult to handle impurities attached to the surface of coal blocks, resulting in low purity of the sorted coal, which affects market competitiveness.

[0006] Existing equipment usually adopts fixed parameter operation mode, such as crushing roller speed, screen vibration intensity, etc. Such fixed parameters are difficult to adapt to the changes of different coal types and working conditions. For example, high-hardness coal blocks require higher crushing efficiency, while low-hardness coal blocks require lower energy consumption. Equipment lacking dynamic adjustment capabilities not only reduces operating efficiency, but also increases energy consumption and equipment wear.

[0007] Traditional spray dust suppression systems usually use fixed spray intensity or coverage, and cannot be dynamically adjusted according to changes in dust concentration. This results in wasting water resources when the dust concentration is low, and failing to effectively suppress dust in a timely manner when the dust concentration is high. Excessive waste of water resources and insufficient spray dust suppression effect are both problems that need to be solved urgently.

[0008] The existing coal mine sorting equipment relies on manual intervention during operation, including the adjustment of operating parameters, fault monitoring and evaluation of sorting results. This low level of automation not only increases labor costs, but may also lead to unstable equipment operation due to differences in operator experience, affecting sorting efficiency and safety.

[0009] The subsystems in the sorting equipment (such as the crushing system, screening system and dust reduction system) lack an effective collaborative working mechanism. For example, the dust generated during the crushing process cannot be linked with the dust reduction system, and the changes in screening efficiency cannot be fed back to the crushing system in time for adjustment. This isolated operation mode makes it difficult to optimize the overall performance of the system.

[0010] In summary, although the existing coal mine sorting and processing equipment can meet the basic sorting needs, it has obvious deficiencies in dust control, sorting accuracy, dynamic adaptability, resource utilization efficiency and automation level. These defects not only affect the operating efficiency and environmental performance of the equipment, but also impose certain restrictions on the economic benefits and industry competitiveness of coal mining enterprises. Therefore, it is urgent to develop an intelligent coal mine sorting and processing equipment based on machine learning that can effectively overcome the above defects, so as to achieve the goals of high efficiency, energy saving and environmental protection in the sorting process. Summary of the invention

[0011] In view of the shortcomings of the prior art, the present invention provides an intelligent coal mine sorting and processing equipment based on machine learning, which has the advantages of strong practicality, good stability, easy dust reduction and easy separation of coal and impurities. It solves the problem that the traditional intelligent coal mine sorting and processing equipment based on machine learning is not convenient for dust reduction when crushing coal, which makes the floating dust easily affect the environment.

[0012] The present invention provides the following technical solution: an intelligent coal mine sorting and processing equipment based on machine learning, comprising a sorting equipment main body, a feeding drum is provided on the top of the sorting equipment main body, a driving motor is provided on the side of the feeding drum, the output end of the driving motor is fixedly connected to a rotating rod, the surface of the rotating rod is fixedly connected to a crushing roller, a baffle plate is provided inside the feeding drum, a support frame is provided on the top of the feeding drum, a cylinder is provided on the top of the support frame, the output end of the cylinder is fixedly connected to a cross pipe, a slide groove is provided on the side of the support frame, a fixed column is provided inside the slide groove, the surface of the fixed column is slidably connected to the circular groove provided on the surface of the connecting block, the side of the connecting block close to the inside of the support frame is fixedly connected to one end of the cross pipe, a water tank is provided on the top of the support frame, a first water pump is provided on the top of the water tank, a bellows is provided on the other end of the first water pump, and the bottom of the bellows passes through the top of the support frame and is connected to the surface of the cross pipe.

[0013] As a preferred technical solution of the present invention, a screen is provided inside the main body of the sorting device, a vibration motor is provided at the bottom of the screen, and a baffle is provided inside the main body of the sorting device.

[0014] As a preferred technical solution of the present invention, a filter frame is provided inside the main body of the sorting equipment, a filter plate is provided inside the filter frame, a filter partition installed inside the filter frame passes through a rectangular groove opened on the top of the filter plate, a pumping pipe is installed on the front of the filter frame, and the pumping pipe passes through the surface of the filter frame and is connected to the inside of the filter partition, a second water pump is connected to the front of the pumping pipe, and a drain pipe is provided at the other end of the second water pump.

[0015] As a preferred technical solution of the present invention, a connecting groove is provided on the front of the sorting equipment body, and the height of the connecting groove is the same as the height of the filter frame. A slide plate is provided inside the sorting equipment body, and the bottom of the slide plate is close to the top of the filter frame.

[0016] As a preferred technical solution of the present invention, the slide plate is an oblique plate, and the interior of the connecting groove is slidably connected with a filter frame.

[0017] As a preferred technical solution of the present invention, an atomizing head is provided on the surface of the transverse tube, and the atomizing head is close to the surface of the crushing roller.

[0018] As a preferred technical solution of the present invention: the rotation speed of the crushing roller of the equipment is dynamically adjusted by integrating a supervised learning algorithm, and the following model is used for prediction

[0019] R t =σ(W·X t +b)

[0020] Among them, R t Indicates the real-time speed of the crushing roller; the eigenvector X t =[ΔP t ,S t ,T t ,C t ], including the feed flow rate change rate ΔP t , Material hardness S t , Equipment temperature T t And the dust concentration C t ; W and b are weight matrix and bias vector respectively; σ is activation function. The weight W and bias b are optimized by back propagation algorithm. The model uses real-time data collected by equipment sensors to dynamically adjust the working state of the crushing roller to meet the processing requirements of different materials.

[0021] As a preferred technical solution of the present invention: the lifting control of the cylinder is based on a reinforcement learning algorithm, and dynamic lifting control is achieved by co-optimizing with the grinding roller speed model. The objective function of the reinforcement learning is defined as:

[0022]

[0023] Among them, Q(s,a) is the value of performing action a in state s, r(s,a) is the current reward value, γ is the discount factor, s′ and a′ are the next state and action respectively; the input parameter of s comes from the rotation speed R of the crushing roller t and the eigenvector X t The reinforcement learning model optimizes the lifting path in real time to ensure accurate positioning of the cylinder during the material sorting process and improve equipment operation efficiency.

[0024] As a preferred technical solution of the present invention: the separation system processes the image data on the screen through a convolutional neural network (CNN) to achieve separation of coal and impurities, and its loss function is:

[0025]

[0026] Among them, y i is the actual category (coal or impurities), is the predicted class probability, λ||W|| 2 is a regularization term used to reduce overfitting. The sorting system links the image analysis results with the lifting control of the cylinder and the speed adjustment of the crushing roller, and updates the control parameters of other modules through the output features to further improve the sorting accuracy and system coordination.

[0027] As a preferred technical solution of the present invention: the spray dust suppression system realizes water volume control based on a multi-objective optimization algorithm, and its optimization objective function is:

[0028]

[0029] Among them, Dust (C i ) indicates the dust concentration after spraying, Water (W i ) represents water consumption, α and β are trade-off coefficients. The dust concentration C predicted by combining convolutional neural network t The spray water volume is adjusted dynamically based on real-time monitoring data to ensure dust reduction effect while reducing water consumption. The spray system and the sorting system work together to improve the environmental performance and economic benefits of the equipment.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] This technical solution achieves precise regulation of the crushing, screening and sorting processes by introducing a dynamic control mechanism based on machine learning. The speed of the crushing roller is dynamically adjusted through real-time data analysis to ensure that coal blocks of different hardness and particle sizes can be crushed in the best way; the sorting system uses convolutional neural networks to perform image analysis on coal and impurities to accurately separate different types of materials, and the sorting effect can be guaranteed even if the particle size of impurities and coal is close. This intelligent control and sorting method significantly improves the purity and efficiency of coal sorting, and solves the problem of insufficient sorting accuracy of traditional equipment.

[0032] Through reinforcement learning and multi-objective optimization algorithms, this equipment achieves coordinated optimization of the crushing roller, cylinder lifting and spray dust suppression system. During the spray dust suppression process, the system can dynamically adjust the spray intensity and coverage according to the real-time monitored dust concentration, effectively reducing the waste of water resources; the dynamic control mechanism of the crushing roller speed also achieves a balance between energy consumption and efficiency. Overall, the equipment minimizes resource consumption and operating costs while ensuring sorting performance.

[0033] This equipment uses machine learning algorithms to build a fully closed-loop control system for data collection, analysis, decision-making and execution. The subsystems achieve a high degree of collaboration through data sharing, avoiding the problem of isolated operation of each module in traditional equipment. In addition, the improvement of automation reduces the dependence on manual intervention and improves the stability and reliability of equipment operation. The operator only needs to set basic parameters, and the equipment can adjust itself according to the real-time working conditions to ensure that the equipment can operate efficiently in various complex environments.

[0034] To sum up, this technical solution has significant advantages in sorting accuracy, resource conservation and intelligence, meeting the urgent needs of the coal mining industry for efficient, environmentally friendly and intelligent sorting equipment, and bringing higher economic benefits and environmental value to the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 It is a schematic diagram of the front structure of the present invention;

[0037] Figure 2 It is a schematic diagram of the lifting structure of the present invention;

[0038] Figure 3 It is a schematic diagram of the cross-sectional structure of the present invention;

[0039] Figure 4 It is a schematic diagram of the explosion structure of the present invention.

[0040] In the figure: 1. main body of the sorting equipment; 2. feed barrel; 3. drive motor; 4. rotating rod; 5. crushing roller; 6. baffle plate; 7. support frame; 8. cylinder; 9. cross pipe; 10. slide; 11. fixed column; 12. connecting block; 13. water storage tank; 14. first water pump; 15. bellows; 16. screen; 17. baffle; 18. filter frame; 19. filter plate; 20. filter partition; 21. suction pipe; 22. second water pump; 23. drain pipe; 24. connecting groove; 25. slide plate. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0042] Example 1: Please refer to Figure 1-4, an intelligent coal mine sorting and processing equipment based on machine learning, including a sorting equipment body 1, a feeding drum 2 is provided on the top of the sorting equipment body 1, a driving motor 3 is provided on the side of the feeding drum 2, a rotating rod 4 is fixedly connected to the output end of the driving motor 3, a crushing roller 5 is fixedly connected to the surface of the rotating rod 4, a baffle plate 6 is provided inside the feeding drum 2, a supporting frame 7 is provided on the top of the supporting frame 7, a cylinder 8 is provided on the top of the supporting frame 7, a transverse pipe 9 is fixedly connected to the output end of the cylinder 8, a chute 10 is provided on the side of the supporting frame 7, a fixed column 11 is provided inside the chute 10, the surface of the fixed column 11 is slidably connected to the circular groove provided on the surface of the connecting block 12, and the side of the connecting block 12 close to the inside of the supporting frame 7 is fixedly connected to one end of the transverse pipe 9, a water storage tank 13 is provided on the top of the supporting frame 7, a first water pump 14 is provided on the top of the water storage tank 13, a bellows 15 is provided on the other end of the first water pump 14, and the bottom of the bellows 15 passes through the top of the supporting frame 7 and is connected to the surface of the transverse pipe 9. A feed barrel 2 is welded on the top of the equipment body 1, which is convenient for fixing the two driving motors 3 on the front side of the feed barrel 2 by bolts. The output ends of the two driving motors 3 rotate in opposite directions, so that the output ends of the driving motors 3 can drive the crushing rollers 5 welded on the surface of the rotating rod 4 to rotate, and the rotating crushing rollers 5 can crush the coal. The baffle plate 6 installed inside the feed barrel 2 can prevent the coal from entering the device from between the crushing rollers 5 and the inside of the feed barrel 2. A support frame 7 is welded on the top of the feed barrel 2, so that the top of the support frame 7 can fix the cylinder 8 by bolts, and the output end of the cylinder 8 can drive the cross pipe 9 to rise and fall. At the same time, the installed fixing column 11 facilitates the movement of the connecting block 12, so that the connecting block 12 can limit the cross pipe 9. The water tank 13 installed on the top of the support frame 7 is convenient for using the first water pump 14 to pump water, and the pumped water can enter the inside of the cross pipe 9 through the bellows 15, so that the cross pipe 9 can be used to spray water on the crushed coal for dust reduction.

[0043] See also Figure 3 A screen 16 is provided inside the main body 1 of the sorting equipment, a vibration motor is provided at the bottom of the screen 16, and a baffle 17 is provided inside the main body 1 of the sorting equipment. The screen 16 installed inside the main body 1 of the sorting equipment can screen the coal through the vibration motor at the bottom, and the baffle 17 can intercept the screened coal.

[0044] See also Figure 4A filter frame 18 is provided inside the main body 1 of the sorting equipment, and a filter plate 19 is provided inside the filter frame 18. A filter partition 20 installed inside the filter frame 18 passes through a rectangular groove opened on the top of the filter plate 19. A pumping pipe 21 is installed on the front of the filter frame 18, and the pumping pipe 21 passes through the surface of the filter frame 18 and is connected to the inside of the filter partition 20. A second water pump 22 is connected to the front of the pumping pipe 21, and a drain pipe 23 is provided at the other end of the second water pump 22. The filter frame 18 installed inside the main body 1 of the sorting equipment can utilize the hydrophobicity of the surface of the coal particles to float the coal particles through the adhesion of bubbles, thereby realizing separation from impurities. The installed filter plate 19 can filter large impurities so that small impurities can be deposited inside the filter frame 18. At the same time, the installed filter partition 20 is convenient for pumping water after the coal is taken out, and then the water inside the filter partition 20 is extracted by the second water pump 22, so that it is convenient to use the drain pipe 23 for drainage.

[0045] See also Figure 3 A connecting groove 24 is provided on the front of the sorting device body 1, and the height of the connecting groove 24 is the same as the height of the filter frame 18. A slide plate 25 is provided inside the sorting device body 1, and the bottom of the slide plate 25 is close to the top of the filter frame 18. The connecting groove 24 provided on the front of the sorting device body 1 makes it convenient to take out the filter frame 18, so as to facilitate the processing of impurities inside the filter frame 18.

[0046] Working principle: during operation, when the coal is crushed, the output end of the driving motor 3 installed on the front of the feed barrel 2 can drive the rotating rod 4 to rotate, and the rotating rotating rod 4 can drive the crushing roller 5 to crush the coal. When the coal is crushed, a large amount of dust will be generated, and then the support frame 7 installed on the top of the feed barrel 2 is convenient for the cylinder 8 to drive the cross pipe 9 to rise and fall, and the installed fixed column 11 is convenient for the connection block 12 to move, so that the connection block 12 can limit the cross pipe 9, and the water storage tank 13 installed on the top of the support frame 7 is convenient for the first water pump 14 to pump water, and the pumped water can enter the interior of the cross pipe 9 through the bellows 15, so as to use the cross pipe 9 to crush the coal. The crushed coal is sprayed with water to reduce dust, and the crushed coal can be screened by the screen 16. The screened coal can fall into two different filter frames 18 through the baffle 17 and the slide plate 25. The hydrophobicity of the surface of the coal particles is utilized, and the coal particles are floated through the adhesion of bubbles, thereby achieving separation from impurities, so that the floating coal can be collected, and the installed filter plate 19 can filter the large ones so that the small impurities can be deposited inside the filter frame 18. At the same time, the installed filter baffle 20 is convenient for pumping water after the coal is taken out, and then the water inside the filter baffle 20 is pumped out by the second water pump 22, and the pumped water can be discharged through the drain pipe 23.

[0047] Example 2: During the operation of the intelligent coal mine sorting and processing equipment, the sensor installed inside the equipment collects the feed flow rate change rate (ΔP t ), material hardness (S t ), equipment temperature (T t ) and dust concentration (C t ) and other key parameters. The sensor data is input into the supervised learning model, which calculates the rotation speed of the crushing roller (5) using the following formula:

[0048] R t =σ(W·X t )+b)

[0049] Among them, X t is the vector of the above features, W and b are the model weights and biases obtained through training. After training with historical data, the model can predict the optimal speed R t , thereby dynamically adjusting the working state of the crushing roller to adapt to the characteristics of different materials.

[0050] Through this embodiment, the equipment can increase the rotation speed of harder coal blocks to enhance the crushing efficiency, and at the same time reduce the rotation speed of materials containing more impurities to reduce energy consumption.

[0051] Example 3: The cylinder (8) in the device controls the lifting and lowering of the horizontal pipe (9) to adjust the separation efficiency of the coal blocks after crushing. This process is controlled by a reinforcement learning algorithm. In the initial stage, the cylinder obtains the current state s through the sensor, combined with the real-time speed R of the crushing roller t And the feed feature X, the reinforcement learning algorithm predicts the optimal action a. The specific value function is as follows:

[0052]

[0053] The reward value r(s,a) is calculated by the balance between sorting accuracy and energy consumption. The cylinder dynamically adjusts the height of the cross tube by executing the action (lifting and lowering operation) recommended by the reinforcement learning algorithm, so that the screening accuracy of the sorting equipment is optimized.

[0054] During operation, the system continuously optimizes the lifting and lowering control strategy by updating the state value s′ and action value Q(s′,a′) in real time, thereby maintaining high sorting efficiency in a dynamic environment.

[0055] Embodiment 4: The cylinder 8 in the device controls the lifting and lowering of the horizontal pipe 9 to adjust the separation efficiency of the coal blocks after crushing. This process is controlled by a reinforcement learning algorithm.

[0056] In the initial stage, the cylinder obtains the current state s through the sensor, combined with the real-time speed R of the crushing roller t And the feed characteristics X t, the optimal action a is predicted by the reinforcement learning algorithm.

[0057] The specific value function is as follows:

[0058]

[0059] The reward value r(s,a) is calculated by the balance between sorting accuracy and energy consumption. The cylinder dynamically adjusts the height of the cross tube by executing the action (lifting and lowering operation) recommended by the reinforcement learning algorithm, so that the screening accuracy of the sorting equipment is optimized.

[0060] During operation, the system continuously optimizes the lifting and lowering control strategy by updating the state value s′ and action value Q(s′,a′) in real time, thereby maintaining high sorting efficiency in a dynamic environment.

[0061] Embodiment 5: After being crushed and preliminarily screened, the coal blocks pass through the screen 16 and enter the sorting system.

[0062] The sorting system uses a convolutional neural network (CNN) to analyze the image on the screen to determine the location and proportion of coal and impurities.

[0063] The image data is processed by the model, and its loss function is:

[0064]

[0065] Among them, y i is the actual label (coal or impurity), The model predicts the result. Through the trained model, the sorting equipment can quickly separate the coal from the impurities. The separated coal falls into the filter frame 18 through the horizontal pipe, while the impurities are transported to the waste collection box.

[0066] This embodiment significantly improves the accuracy of separation of coal and impurities and reduces the workload of manual screening.

[0067] Example 4: The large amount of dust generated during the operation of the equipment is treated by the spray dust suppression system. The spray system combines a multi-objective optimization algorithm to dynamically adjust the water volume. The optimization objectives are:

[0068]

[0069] The sensor monitors the dust concentration C in the environment in real time i and spray water volume W i ,The spray intensity is adjusted through optimization algorithm.

[0070] The image data analyzed by the convolutional neural network further provides dust distribution information, and the system dynamically optimizes the spray coverage based on the analysis results. Through this embodiment, the spray system can reduce dust concentration while minimizing water consumption, achieving the purpose of energy saving and environmental protection.

[0071] Embodiment 6: The various systems in the above embodiments work together by sharing data. For example: The real-time adjustment result of the rotation speed of the crushing roller R t And the dust concentration prediction value C t The image analysis results of the sorting module are used to optimize the dust reduction of the spray system and adjust the water volume control strategy through dynamic prediction.

[0072] The entire set of equipment forms a closed-loop intelligent sorting system through collaborative control technology based on machine learning, which significantly improves the coal mine sorting efficiency and equipment intelligence level.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent coal mine sorting and processing equipment based on machine learning, comprising a sorting equipment body (1), characterized in that: A feed barrel (2) is provided on the top of the main body (1) of the sorting equipment, a drive motor (3) is provided on the side of the feed barrel (2), a rotating rod (4) is fixedly connected to the output end of the drive motor (3), a crushing roller (5) is fixedly connected to the surface of the rotating rod (4), a baffle plate (6) is provided inside the feed barrel (2), a support frame (7) is provided on the top of the support frame (7), a cylinder (8) is provided on the top of the support frame (7), a transverse pipe (9) is fixedly connected to the output end of the cylinder (8), and a slide groove (10) is provided on the side of the support frame (7). A fixing column (11) is provided inside the slide groove (10), and the surface of the fixing column (11) is slidably connected to a circular groove provided on the surface of a connecting block (12). The connecting block (12) is fixedly connected to one end of a transverse pipe (9) on one side close to the inside of the support frame (7). A water storage tank (13) is provided on the top of the support frame (7), and a first water pump (14) is provided on the top of the water storage tank (13). A bellows (15) is provided at the other end of the first water pump (14), and the bottom of the bellows (15) passes through the top of the support frame (7) and is connected to the surface of the transverse pipe (9).

2. The intelligent coal mine sorting and processing equipment based on machine learning according to claim 1 is characterized in that: A screen (16) is provided inside the main body (1) of the sorting device, a vibration motor is provided at the bottom of the screen (16), and a baffle (17) is provided inside the main body (1) of the sorting device.

3. The intelligent coal mine sorting and processing equipment based on machine learning according to claim 1 is characterized in that: A filter frame (18) is provided inside the main body (1) of the sorting device, a filter plate (19) is provided inside the filter frame (18), a filter baffle (20) installed inside the filter frame (18) passes through a rectangular groove opened on the top of the filter plate (19), a water pump (21) is installed on the front of the filter frame (18), and the water pump (21) passes through the surface of the filter frame (18) and is connected to the inside of the filter baffle (20), a second water pump (22) is connected to the front of the water pump (21), and a drainage pipe (23) is provided at the other end of the second water pump (22).

4. The intelligent coal mine sorting and processing equipment based on machine learning according to claim 1 is characterized in that: A connecting groove (24) is provided on the front of the main body (1) of the sorting device, and the height of the connecting groove (24) is the same as the height of the filter frame (18). A sliding plate (25) is provided inside the main body (1) of the sorting device, and the bottom of the sliding plate (25) is close to the top of the filter frame (18).

5. The intelligent coal mine sorting and processing equipment based on machine learning according to claim 4 is characterized in that: The slide plate (25) is an oblique plate, and the interior of the connecting groove (24) is slidably connected to a filter frame (18).

6. The intelligent coal mine sorting and processing equipment based on machine learning according to claim 1 is characterized by: An atomizing head is provided on the surface of the transverse tube (9), and the atomizing head is close to the surface of the crushing roller (5).

7. The intelligent coal mine sorting and processing equipment based on machine learning according to claim 1 is characterized by: The rotation speed of the crushing roller (5) of the device is dynamically adjusted by integrating a supervised learning algorithm, and the following model is used for prediction: R t =σ(W·X t +b) Among them, R t Indicates the real-time speed of the crushing roller; the characteristic vector X t =[ΔP t ,S t ,T t ,C t ], including the feed flow rate change rate ΔP t , Material hardness S t , Equipment temperature T t And the dust concentration C t ; W and b are the weight matrix and bias vector respectively; σ is the activation function that optimizes the weight W and bias b through the back propagation algorithm.

8. The intelligent coal mine sorting and processing equipment based on machine learning according to claim 7 is characterized by: The lifting control of the cylinder (8) is based on a reinforcement learning algorithm, and dynamic lifting control is achieved by co-optimizing with the grinding roller speed model. The objective function of the reinforcement learning is defined as: Among them, Q(s,a) is the value of performing action a in state s, r(s,a) is the current reward value, γ is the discount factor, s′ and a′ are the next state and action respectively; the input parameter of s comes from the speed R of the crushing roller t and the eigenvector X t .

9. The intelligent coal mine sorting and processing equipment based on machine learning according to claim 8 is characterized by: The separation system processes the image data on the screen (16) through a convolutional neural network to separate the coal from the impurities, and its loss function is: Among them, y i For the actual category, is the predicted class probability, λ||W|| 2 is a regularization term used to reduce overfitting.

10. The intelligent coal mine sorting and processing equipment based on machine learning according to claim 9, characterized in that: The spray dust suppression system realizes water volume control based on a multi-objective optimization algorithm, and its optimization objective function is: Among them, Dust (C i ) indicates the dust concentration after spraying, Water (W i ) represents water consumption, and α and β are trade-off coefficients.

Images