Coal and gangue separation system and method based on multi-stage screening and slope separation

Abstract

The invention provides a coal and gangue separation system and method based on multistage screening and slope separation, and belongs to the field of underground coal and gangue intelligent separation. The system comprises a coal gangue particle size multistage screening subsystem, a slope type coal gangue sorting subsystem and a high-precision coal gangue identifying and sorting subsystem. Firstly, the coal gangue particle size multi-stage screening subsystem conducts multi-stage screening according to the particle size of coal and gangue, and the coal and the gangue are conveyed to the slope type coal and gangue sorting subsystem after screening; secondly, the slope type coal and gangue separation subsystem realizes preliminary separation of coal and gangue by controlling key parameters of a slope type belt conveyor according to the physical property difference of coal and gangue with different particle sizes on the slope type belt conveyor; and finally, the high-precision coal and gangue identifying and sorting subsystem adopts a Relief-MRMR-SVM model to accurately identify the coal and the gangue, and a mechanical arm is controlled to sort, so that deep optimized sorting is realized. According to the system and the method, accurate separation of the coal and the gangue is realized through the three closely matched subsystems and the method.

Description

Technical Field

[0001] The present invention relates to the technical field of underground coal gangue intelligent sorting, and in particular to a coal gangue sorting system and method based on multi-stage screening and inclined separation. Background Art

[0002] Traditional coal gangue sorting methods have the problem of low intelligence, which makes it difficult to meet the green and intelligent transformation of the modern coal industry. At present, some coal gangue separation technologies have been proposed and applied, such as physical screening, gravity sorting, flotation, etc. However, these technologies often have problems such as unstable separation effect and complex equipment in practical applications, and are easily affected by environmental factors in practical applications, resulting in limited sorting effects and unable to achieve comprehensive and refined processing of coal and gangue.

[0003] Some early sorting equipment can only perform simple rough sorting, and cannot deeply identify and separate coal and gangue, resulting in a certain amount of gangue still mixed in the sorted coal products, so that the gangue cannot be fully separated and effectively utilized. Therefore, a new type of coal-gangue sorting system is urgently needed to solve the problems existing in the existing technology, realize the intelligent and precise separation of coal and gangue, improve the efficiency and quality of coal sorting, and reduce the labor cost in the sorting process, which is of great significance to the rational utilization of coal resources and environmental protection. Summary of the invention

[0004] The purpose of the present invention is to provide a coal gangue sorting system and method based on multi-stage screening and inclined separation, which realizes intelligent and precise separation of coal and gangue through the cooperation of multi-stage screening, inclined separation technology, high-precision coal gangue identification model and mechanical arm, thereby improving the efficiency and quality of sorting.

[0005] To achieve the above object, the present invention provides a coal gangue separation system based on multi-stage screening and inclined separation, comprising:

[0006] The gangue particle size multi-stage screening subsystem comprises a main belt conveyor, a steel plate frame and n screening units; wherein the n screening units are arranged on the steel plate frame in a stepped manner, the ends of the n screening units are connected to the inclined-surface gangue sorting subsystem, any two adjacent screening units are connected by a ramp, each screening unit is equipped with a secondary belt conveyor, the first n-1 secondary belt conveyors are provided with a roller screen, and the roller spacing decreases from the first secondary belt conveyor to the n-1 secondary belt conveyors, and the first screening unit is connected to the main belt conveyor;

[0007] Inclined-plane coal-gangue separation subsystem, including n downward belt conveyors, n inclined-plane belt conveyors and 1 programmable logic controller; the end of the downward belt conveyor is connected to the head of the inclined-plane belt conveyor, each inclined-plane belt conveyor is driven by a servo motor and equipped with a force sensor, and the head of each downward belt conveyor is connected to the end of the auxiliary belt conveyor;

[0008] High-precision coal-gangue identification and separation subsystem, including n gangue conveyor belts, n coal conveyor belts, 1 server and 1 DeltaV distributed control system; among them, n gangue conveyor belts and n coal conveyor belts are arranged in sequence along the conveying direction of n inclined-plane belt conveyors, and image acquisition devices and robotic arms are arranged on one side of each gangue conveyor belt and coal conveyor belt.

[0009] The present invention also provides a coal-gangue separation method based on multi-stage screening and inclined-plane separation, adopting the above-mentioned coal-gangue separation system based on multi-stage screening and inclined-plane separation, including the following steps:

[0010] Step S1, during the trial operation stage, adjust and determine the relevant equipment and their parameters of each subsystem;

[0011] Step S11, adjust the roller spacing of the roller screens on the first n-1 screening units in the coal-gangue particle size multi-stage screening subsystem, and the roller spacing decreases in sequence from the first auxiliary belt conveyor to the n-1th auxiliary belt conveyor;

[0012] Step S12, determine the inclination angle θ of the inclined-plane belt conveyor in the inclined-plane coal-gangue separation subsystem: 0 < θ < min{arctanμ1, arctanμ2}, where μ1 represents the friction coefficient of coal on the inclined-plane belt conveyor, and μ2 represents the friction coefficient of gangue on the inclined-plane belt conveyor;

[0013] Step S13, determine the speed V1 of the inclined-plane belt conveyor in the inclined-plane coal-gangue separation subsystem: V1 > V2, V1 > V3, where V2 represents the speed at which coal leaves the inclined-plane belt conveyor, and V3 represents the speed at which gangue leaves the inclined-plane belt conveyor;

[0014] Step S14, determine the length S of the inclined-plane belt conveyor in the inclined-plane coal-gangue separation subsystem: where a1 represents the acceleration of coal moving on the inclined-plane belt conveyor, a2 represents the acceleration of gangue moving on the inclined-plane belt conveyor, and V0 represents the speed of coal and gangue when they reach the head of the inclined-plane belt conveyor;

[0015] Step S15, determine the positions of the gangue conveyor belt and the coal conveyor belt in the high-precision coal-gangue identification and separation subsystem: Wherein, L1 represents the distance between the gangue conveyor belt and the inclined belt conveyor, L2 represents the distance between the coal conveyor belt and the inclined belt conveyor, and g represents the acceleration of gravity;

[0016] Step S2: The system is officially put into operation, and the coal and gangue particle size multi-stage screening subsystem is used to complete the screening of coal and gangue with different particle sizes;

[0017] Step S21, the main belt conveyor conveys the coal and gangue to the first screening unit of the coal gangue particle size multi-stage screening subsystem;

[0018] Step S22, when the coal and gangue pass through the roller screen on the first screening unit, the coal and gangue larger than the roller spacing pass through the roller screen, and then are conveyed into the inclined coal-gangue sorting subsystem via the auxiliary belt conveyor, and the coal and gangue smaller than the roller spacing pass through the roller spacing into the ramp and finally leak into the next screening unit, and then are screened for the next particle size until the remaining coal and gangue fall into the nth screening unit;

[0019] Step S3, the inclined coal gangue sorting subsystem performs preliminary sorting on coal and gangue of the same particle size range;

[0020] Step S31, the coal and gangue of various particle sizes screened out in step S22 are transferred to the corresponding inclined belt conveyor through the descending belt conveyor;

[0021] Step S32, the coal and gangue are conveyed to the head end of the corresponding inclined belt conveyor at a speed of V0, and begin to accelerate along the inclined direction;

[0022] Step S33, under the difference of physical properties of coal and gangue of different particle sizes, through the speed, length and inclination angle set by each inclined belt conveyor, the coal and gangue leave the inclined belt conveyor and perform oblique throwing motion with different initial velocities, and fall into the coal conveying belt conveyor at a distance of L2 from the inclined belt conveyor and the gangue conveyor at a distance of L1 respectively;

[0023] Step S4, the high-precision coal gangue identification and sorting subsystem performs deep optimization sorting on the initially sorted coal and gangue;

[0024] Step S41, the image acquisition device acquires images of coal and gangue on the gangue conveyor belt conveyor and the coal conveyor belt conveyor;

[0025] Step S42, transmitting the image to the server for preprocessing, including graying, smoothing and sharpening the image to obtain a preprocessed image;

[0026] Step S43, using the image processing module of the MATLAB library in the server, based on the preprocessed image, using the grayscale co-occurrence matrix, extracting the contrast and angular second-order moment feature parameters, and using the grayscale information, extracting the grayscale mean and grayscale variance feature parameters; fusing the output texture and grayscale features to form contrast, angular second-order moment, grayscale mean, and grayscale variance features as feature vector data sets;

[0027] Step S44: taking the feature vector data set as input and the actual coal and gangue real labels as output, inputting them into the Relief-MRMR-SVM model, training the model, obtaining the weight parameters of the Relief-MRMR-SVM model, and embedding the weight parameters of the Relief-MRMR-SVM model into the server;

[0028] Step S45, the coal and gangue images newly collected by the image acquisition device are transmitted to the server for processing and feature extraction, and then the extracted features are classified using the trained Relief-MRMR-SVM model;

[0029] Step S46: The server receives the identification results of coal and gangue, and transmits the classification information to the DeltaV distributed control system. The DeltaV distributed control system controls the robot arm to perform deep optimization sorting of coal and gangue according to the classification information and instructions.

[0030] Therefore, the present invention adopts the above-mentioned coal gangue sorting system and method based on multi-stage screening and inclined separation, and the beneficial technical effects are as follows:

[0031] (1) The present invention uses a multi-stage coal gangue particle size screening subsystem to finely classify coal and gangue according to the particle size range, combines the preliminary sorting of the inclined coal gangue sorting subsystem with the deep optimization sorting of the high-precision coal gangue identification and sorting subsystem, and realizes the intelligent and precise separation of coal and gangue.

[0032] (2) The system and method proposed in the present invention not only improve the efficiency and quality of coal sorting, but also effectively reduce the energy consumption and labor costs in the sorting process, which is of great significance to the rational utilization of coal resources and environmental protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a structural schematic diagram of a first embodiment of a coal gangue separation system based on multi-stage screening and inclined separation of the present invention;

[0034] Figure 2 It is a schematic diagram of the structure of a multi-stage roller screen according to a first embodiment of a coal gangue separation system based on multi-stage screening and inclined separation of the present invention;

[0035] Figure 3It is a schematic diagram of the information transmission process of a first embodiment of a coal gangue sorting system based on multi-stage screening and inclined separation of the present invention;

[0036] Figure 4 It is a schematic flow chart of a second embodiment of a coal gangue separation method based on multi-stage screening and inclined separation of the present invention;

[0037] Figure 5 This is a force analysis diagram of coal or gangue on an inclined belt conveyor according to Example 2 of the present invention.

[0038] Reference numerals

[0039] 1. Main belt conveyor; 2. Gangue; 3. Coal; 4. Auxiliary belt conveyor; 5. Roller screen; 6. Inclined road; 7. Steel plate frame; 8. Descending belt conveyor; 9. Inclined belt conveyor; 10. Servo motor; 11. Force sensor; 12. Gangue belt conveyor; 13. Coal belt conveyor; 14. Robotic arm; 15. Image acquisition equipment; 16. Programmable logic controller; 17. Server; 18. DeltaV distributed control system. DETAILED DESCRIPTION

[0040] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0041] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0042] Embodiment 1

[0043] like Figure 1 FIG. 1 is a schematic diagram of a multi-stage screening and inclined separation type coal gangue separation system according to the present invention, comprising:

[0044] The gangue particle size multi-stage sieve subsystem is used to classify the coal 3 and gangue 2 according to the particle size.

[0045] The gangue particle size multi-stage screening subsystem includes n screening units (n is set to 4 in this embodiment) arranged in a stepped manner from top to bottom on a steel plate frame 7.

[0046] The four screening units are respectively a primary screening unit, a secondary screening unit, a tertiary screening unit and a quaternary screening unit, wherein the primary screening unit, the secondary screening unit and the tertiary screening unit each include a secondary belt conveyor 4 and a roller screen 5, the roller screen 5 being arranged on the secondary belt conveyor 4, the quaternary screening unit includes a secondary belt conveyor 4, any two adjacent secondary belt conveyors 4 are connected by a ramp 6, and the ends of the four screening units are connected to the inclined coal gangue sorting subsystem. The primary screening unit is also connected to the main belt conveyor 1.

[0047] In this embodiment, one roller screen 5 is composed of three rollers.

[0048] like Figure 2 As shown, the main belt conveyor 1 transports gangue 2 and coal 3 to a coal gangue particle size multi-stage screening subsystem built by a certain mine, and coal 3 and gangue 2 fall into the primary screening unit. According to the actual sorting requirements of the mine, the roller spacing of the roller screen 5 on the primary screening unit is adjusted to 60 mm, and the coal 3 and gangue 2 with a particle size greater than 60 mm will continue to be transported along the auxiliary belt conveyor 4 to the inclined coal gangue sorting subsystem for the next step of sorting, and the coal 3 and gangue 2 with a particle size less than 60 mm will leak through the roller spacing to the secondary screening unit.

[0049] Similarly, the screening process on the secondary screening unit and the screening on the tertiary screening unit is similar to the screening process on the above-mentioned primary screening unit. The roller spacing of the roller screen 5 on the secondary screening unit is adjusted to 40mm, and the coal 3 and gangue 2 with a particle size greater than 40mm will continue to be transported to the inclined coal gangue sorting subsystem along the auxiliary belt conveyor 4, and the coal 3 and gangue 2 with a particle size less than 40mm will fall into the tertiary screening unit. The roller spacing of the roller screen 5 on the tertiary screening unit is adjusted to 20mm, and the coal 3 and gangue 2 with a particle size greater than 20mm will continue to be transported to the inclined coal gangue sorting subsystem along the auxiliary belt conveyor 4, and the coal 3 and gangue 2 with a particle size less than 20mm will fall into the quaternary screening unit.

[0050] After multi-stage screening, coal 3 and gangue 2 are divided into four ranges according to particle size: greater than 60mm, 40-60mm, 20-40mm, and 0-20mm. Each type of coal 3 and gangue 2 enters its corresponding inclined coal and gangue sorting subsystem for subsequent sorting and processing.

[0051] The inclined coal gangue separation subsystem is used for preliminarily separating the coal 3 and the gangue 2.

[0052] The inclined coal gangue sorting subsystem includes four descending belt conveyors 8, four inclined belt conveyors 9 and one programmable logic controller 16 (PLC).

[0053] Each inclined belt conveyor 9 is driven by a servo motor 10 and is equipped with a force sensor 11. The head end of the descending belt conveyor 8 is connected to the end of each level of screening unit.

[0054] The inclined belt conveyor 9 can better separate the coal 3 and the gangue 2 with the same particle size range. The coal 3 and the gangue 2 with the same particle size range are similar in shape, so the difference in physical properties between the two on the inclined belt conveyor 9 can be better used to separate the two, reducing the influence of other irrelevant factors such as volume. At the same time, it is convenient to meet the specific coal mine's requirements for the particle size range of coal 3 and gangue 2, which is conducive to the later treatment of coal 3 and gangue 2.

[0055] The high-precision coal gangue identification and sorting subsystem is used to carry out fine sorting of the initially separated coal 3 and gangue 2.

[0056] The high-precision gangue identification and sorting subsystem includes 4 gangue belt conveyors 12, 4 coal belt conveyors 13, 1 server 17 and 1 DeltaV distributed control system 18, as well as 8 robotic arms 14 and 8 image acquisition devices 15 arranged on both sides of the gangue belt conveyors 12 and the coal belt conveyors 13.

[0057] The gangue conveyor belt 12 and the coal conveyor belt 13 are sequentially arranged in front of the inclined belt conveyor 9 along the conveying direction.

[0058] In this embodiment, the mine uses a pseudo dual-energy X-ray transmission image acquisition device, a portable dual-energy X-ray flat panel detector, to respectively acquire clear images of coal 3 and gangue 2. During the image acquisition process, the device not only exerts its high-precision dual-energy X-ray technology, but also ensures the accuracy and efficiency of image acquisition through intelligent operation. At the same time, in order to further improve the automation level of operation, the mine is also equipped with a mechanical arm 14, which works in conjunction with the image acquisition device 15 to complete the entire process from sample placement to image acquisition.

[0059] like Figure 3 As shown, the programmable logic controller 16 is used to receive and process the data of the inclined belt conveyor 9 collected by the force sensor 11, and then transmit the processing result to the servo motor 10. The servo motor 10 drives the inclined belt conveyor 9 and adjusts the inclination angle and speed of the inclined belt conveyor 9.

[0060] The server 17 is used to collect and process the image information of the image acquisition device 15, run the recognition algorithm to accurately identify the coal 3 and the gangue 2, and then transmit the processing results to the DeltaV distributed control system 18. The DeltaV distributed control system 18 is used to control the parallel SCARA type robot arm 14 to complete the deep optimization sorting of the coal 3 and the gangue 2 according to the recognition results of the server 17.

[0061] Embodiment 2

[0062] like Figure 4As shown, based on the sorting system of embodiment 1, this embodiment provides a coal gangue sorting method based on multi-stage screening and inclined separation, comprising the following steps:

[0063] Step S1: During the trial operation phase, the relevant equipment and parameters of each subsystem are adjusted and determined.

[0064] Step S11, adjust the roller spacing of the roller screen 5 on the first-stage screening unit, the second-stage screening unit and the third-stage screening unit in the coal gangue particle size multi-stage screening subsystem to 60mm, 40mm and 20mm respectively, and divide the screened coal 3 and gangue 2 into four levels according to the particle size, namely, extra large particle size, large particle size, medium particle size and small particle size.

[0065] Step S12: To ensure that the coal 3 and the gangue 2 can accelerate upward along the inclined surface on the inclined belt conveyor 9 and achieve the maximum difference in horizontal displacement after the coal 3 and the gangue 2 are thrown out, the inclination angle θ of the inclined belt conveyor 9 needs to be determined.

[0066] like Figure 5 As shown, the force analysis of coal 3 and gangue 2 is carried out, and the following results are obtained:

[0067] G = mg;

[0068] f=μ×F N ;

[0069] F N =G×cosθ;

[0070] fG×sinθ=ma;

[0071] Among them, G represents the gravity exerted on the coal 3 or gangue 2 on the inclined belt conveyor 9, m represents the mass of the coal 3 or gangue 2, g represents the acceleration of gravity, f represents the friction force of the coal 3 or gangue 2 on the inclined belt conveyor 9, a represents the acceleration of the movement of the coal 3 or gangue 2 on the inclined belt conveyor 9, and μ represents the friction coefficient of the coal 3 or gangue 2 on the inclined belt conveyor 9.

[0072] In order to enable the coal 3 or gangue 2 to accelerate upward along the inclined surface on the inclined belt conveyor 9, the following relationship needs to be satisfied: the upward force of the coal 3 or gangue 2 along the inclined belt conveyor 9 must be greater than the downward force along the inclined belt conveyor 9, that is, the requirement:

[0073] f>G×sinθ;

[0074] From this we can get:

[0075] μ>tanθ;

[0076] The friction coefficient value of the coal 3 on the inclined belt conveyor 9 is obtained as 0.5, and the friction coefficient value of the gangue 2 on the inclined belt conveyor 9 is 0.4.

[0077] Calculate the inclination angle θ of the inclined belt conveyor 9:

[0078] 0 < θ < min{arctanμ1, arctanμ2};

[0079] Among them, μ1 represents the friction coefficient of the coal 3 on the inclined belt conveyor 9, with a value of 0.5; μ2 represents the friction coefficient of the gangue 2 on the inclined belt conveyor 9, with a value of 0.4.

[0080] Four inclined belt conveyors 9 are respectively arranged 3.00 m below and 4.00 m horizontally to the right of each auxiliary belt conveyor 4, and the coal 3 and gangue 2 are transferred to the four inclined belt conveyors 9 through a descending belt conveyor 8 with a length of 5.00 m.

[0081] Through trial operation, relevant parameters are adjusted. The force sensor 11 will monitor the data of the inclined belt conveyor 9 in real time and transmit the data to the programmable logic controller 16. The programmable logic controller 16 calculates according to the above principle, and then obtains the accurate best inclination angle of the inclined belt conveyor 9. If the inclination angle θ of the four inclined belt conveyors 9 < arctan(0.4) = 21.8°, then the inclination angle θ of the four inclined belt conveyors 9 = 21°.

[0082] Step S13: Determine the running speed V1 of the inclined belt conveyor 9:

[0083] To achieve the speed difference between the coal 3 and the gangue 2 on the inclined belt conveyor 9 and enhance the separation effect, determine the running speed V1 of the inclined belt conveyor 9. If the coal 3 and the gangue 2 reach the same speed as the inclined belt conveyor 9 before reaching the end of the inclined belt conveyor 9, the initial velocities of their oblique projectile motions will be the same, and then the coal 3 and the gangue 2 cannot be effectively separated. Therefore, the formula is as follows:

[0084] V1 > V2, V1 > V3;

[0085] Among them, V2 represents the speed required for the coal 3 to leave the inclined belt conveyor 9, and V3 represents the speed required for the gangue 2 to leave the inclined belt conveyor 9.

[0086] From the formula:

[0087] a1 = μ1gcosθ - gsinθ;

[0088] a2 = μ2gcosθ - gsinθ;

[0089] Wherein, a1 represents the acceleration of the coal 3 moving on the inclined belt conveyor 9, and a2 represents the acceleration of the gangue 2 moving on the inclined belt conveyor 9.

[0090] The servo motor 10 receives the speed requirement and the tilt angle requirement, and adjusts the speed of the inclined belt conveyor 9 .

[0091] It can be calculated that the accelerations of coal 3 and gangue 2 on the inclined belt conveyor 9 are 1.08 m / s 2 and 0.16m / s 2 Therefore, it can be obtained that the initial velocities of coal 3 and gangue 2 leaving the inclined belt conveyor 9 are V2=4.76m / s and V3=2.01m / s respectively, which must be less than the operating speed V1 of the inclined belt conveyor 9. Therefore, it can be ensured that coal 3 and gangue 2 can always accelerate on the inclined belt conveyor 9 until they fall off. Therefore, the operating speed V1 of the four inclined belt conveyors 9 installed in the mine is set to 5m / s.

[0092] The optimal inclination angle and the optimal speed are transmitted to the programmable logic controller 16, and the programmable logic controller 16 adjusts the inclination angle of the inclined belt conveyor 9 to fit the optimal inclination angle by controlling the servo motor 10, and adjusts the speed of the inclined belt conveyor 9 to fit the optimal speed.

[0093] Step S14, determine the length S of the inclined belt conveyor 9:

[0094] In order to prevent the coal 3 and gangue 2 from reaching the same speed as the inclined belt conveyor 9 before reaching the end of the inclined belt conveyor 9 and failing to separate the two, the length S of the inclined belt conveyor 9 needs to be determined:

[0095]

[0096] Wherein, V0 represents the speed of the coal and gangue when they reach the head end of the inclined belt conveyor 9.

[0097] Through the above calculation, it can be determined that the length of the inclined belt conveyor 9 is 10m.

[0098] Step S15, determining the positions of the gangue conveyor belt 12 and the coal conveyor belt 13:

[0099]

[0100] Wherein, L1 represents the distance between the gangue conveying belt conveyor 12 and the inclined belt conveyor 9, and L2 represents the distance between the coal conveying belt conveyor 13 and the inclined belt conveyor 9.

[0101] The distance d between the coal conveying belt conveyor 13 and the gangue conveying belt conveyor 12 is:

[0102] d = L1-L2;

[0103] After deduction, it can be concluded that:

[0104] d = 4sinθcos 2 θ(μ1-μ2)S;

[0105] The distance d between the landing points of coal 3 and gangue 2 is affected by the inclination angle θ of the inclined belt conveyor 9, the length S, and the friction coefficients of coal 3 and gangue 2 respectively. Under the condition that θ, μ1 and μ2 remain unchanged, increasing S can increase the distance between the landing points of coal 3 and gangue 2 and achieve better sorting effect.

[0106] The inclination angle θ=21° of the inclined belt conveyor 9 that meets the actual sorting needs of the mine has been obtained, and it is known that μ1=0.5, μ2=0.4, S=10m, so the calculation can obtain the distance L1=0.27m between the gangue belt conveyor 12 and the inclined belt conveyor 9, and the distance L2=1.52m between the coal belt conveyor 13 and the inclined belt conveyor 9. Therefore, the distance d=1.25m between the coal belt conveyor 13 and the gangue belt conveyor 12 can achieve the effect of separating the two.

[0107] Step S2, the system is officially put into operation, and the screening of coal 3 and gangue 2 with different particle sizes is completed by using the gangue particle size multi-stage screening subsystem.

[0108] Step S21, the main belt conveyor 1 conveys the coal 3 and the gangue 2 to the first screening unit of the coal gangue particle size multi-stage screening subsystem.

[0109] Step S22, coal 3 and gangue 2 fall from the main belt conveyor 1 to the primary screening unit. When passing through the roller screen 5 on the primary screening unit, coal 3 and gangue 2 larger than 60 mm will pass through the roller screen 5 and enter the descending belt conveyor 8, and then be sent to the inclined belt conveyor 9. Coal 3 and gangue 2 smaller than 60 mm will fall into the secondary screening unit, and then pass through the roller screen 5 on the secondary screening unit for a second screening. Coal 3 and gangue 2 smaller than 40 mm fall into the tertiary screening unit and are screened again. Finally, coal 3 and gangue 2 are divided into four parts: extra large particle size larger than 60 mm, large particle size larger than 40 mm and less than 60 mm, medium particle size larger than 20 mm and less than 40 mm, and small particle size smaller than 20 mm, completing the multi-stage screening of coal 3 and gangue 2, and then being transported to the corresponding four inclined belt conveyors 9.

[0110] Step S3, the inclined coal gangue sorting subsystem performs preliminary sorting on the coal 3 and gangue 2 of the same particle size range;

[0111] Step S31, the coal 3 and gangue 2 of various particle sizes screened in step S22 are transferred to the corresponding inclined belt conveyor 9 through the descending belt conveyor 8;

[0112] Step S32, coal 3 and gangue 2 are conveyed to the bottom of the corresponding inclined belt conveyor 9 at a speed of V0, and begin to accelerate along the inclined direction;

[0113] Step S33, under the difference in physical properties of coal 3 and gangue 2 of different particle sizes, through the set speed, length and inclination angle of each inclined belt conveyor 9, the coal 3 and gangue 2 leave the inclined belt conveyor 9 and make oblique throwing motion with different initial velocities, and fall into the coal transport belt conveyor 13 1.52m away from the inclined belt conveyor 9 and the gangue transport belt conveyor 12 0.27m away respectively.

[0114] Step S4, the high-precision coal gangue identification and sorting subsystem performs deep optimization sorting on the preliminarily sorted coal 3 and gangue 2;

[0115] Step S41, the image acquisition device 15 acquires images of the coal 3 and the gangue 2 on the gangue conveyor belt 12 and the coal conveyor belt 13, a total of 2000 images;

[0116] Step S42, transmitting the image to the server 17 for preprocessing, including graying, smoothing and sharpening the image to obtain a preprocessed image;

[0117] Step S43, using the image processing module of the MATLAB library in the server 17, based on the preprocessed image, using the grayscale co-occurrence matrix, extracting the contrast and angular second-order moment feature parameters, and using the grayscale information, extracting the grayscale mean and grayscale variance feature parameters; fusing the output texture and grayscale features to form contrast, angular second-order moment, grayscale mean, and grayscale variance features as feature vector data sets;

[0118] The gray level co-occurrence matrix is ​​used to extract the contrast and angular second-order moment feature parameters, which are calculated as follows:

[0119]

[0120] Where N represents contrast, E represents the angular second-order moment, P(i,j,d,q) represents the gray-level co-occurrence matrix, i and j represent the gray values ​​of two pixels, d and q represent the distance and orientation of two pixels, respectively.

[0121] Based on the grayscale information, the grayscale mean and grayscale variance are extracted, which are calculated as follows:

[0122]

[0123] Among them, m represents the grayscale mean, L represents the number of grayscale levels, and z i The discrete random variable representing the gray level of the image, P(z i ) represents the estimated value of the probability distribution; then calculate the center distance s n and grayscale variance σ 2 :

[0124] s n (z)=(z i -m) n P(z i );

[0125] σ 2 =s2(z);

[0126] Where n represents the order of the moment and s2(z) represents the center distance of order 2.

[0127] Tables 1 to 4 show the characteristic parameters of five images of coal 3 and gangue 2 processed by the image processing module of the MATLAB library.

[0128] Table 1 Gray characteristic parameters of coal

[0129]

[0130] Table 2 Grayscale characteristic parameters of gangue

[0131]

[0132] Table 3 Coal texture characteristic parameters

[0133]

[0134]

[0135] Table 4 Texture characteristic parameters of gangue

[0136]

[0137] Table 5 Coal and gangue related feature data set construction table (partial)

[0138]

[0139] Step S44, taking the feature vector data set (Table 5) as input, taking the actual coal 3 and gangue 2 real labels as output, inputting them into the Relief-MRMR-SVM model, training the model, obtaining the weight parameters of the Relief-MRMR-SVM model, and embedding the weight parameters of the Relief-MRMR-SVM model into the server;

[0140] The specific process of Relief-MRMR-SVM model training is as follows:

[0141] In order to build a recognition model for coal 3 and gangue 2, the grayscale features and texture features were first fused to form a data set containing 2000 sets of data. The data set includes four input features: grayscale mean, grayscale variance, contrast, and angular second-order moment, as well as an output label, namely the actual classification of coal 3 or gangue 2. In this data set, 80% of the data was used as a training set, and the remaining 20% ​​was used as a validation set. Next, the feature vector composed of contrast, angular second-order moment, grayscale mean, and grayscale variance was used as input features and input into the Relief-MRMR-SVM model for the recognition of coal 3 and gangue 2. In the training stage, the Relief-MRMR-SVM model was trained using 80% of the data in the training set to learn the mapping relationship from features to labels. After the training is completed, the model is validated using the remaining 20% ​​validation set data, and the performance of the model is evaluated by comparing the recognition results of the model with the true labels of the validation set. When the correct samples of the validation set are highly consistent with the recognition results of the model, it is considered that the model has learned effective weight parameters and can be used for accurate identification of coal 3 and gangue 2.

[0142] Step S45, the coal 3 and gangue 2 images newly collected by the image acquisition device 15 are transmitted to the server 17 for processing and feature extraction, and then the extracted features are classified using the trained Relief-MRMR-SVM model;

[0143] Step S46, the server 17 receives the identification results of the coal 3 and the gangue 2, and transmits the classification information to the DeltaV distributed control system 18. The DeltaV distributed control system 18 controls the robot arm 14 to perform fine sorting of the coal 3 and the gangue 2 according to the classification information and instructions.

[0144] The test results show that the gangue picking rate on the coal conveying belt conveyor 13 and the coal picking rate on the gangue conveyor 12 are both over 90% on average, and the mispicking rate is less than 5%, which meets the requirements for high-precision coal and gangue sorting.

[0145] It is worth noting that the contents not elaborated in detail in the present invention are all prior art and are well known to those skilled in the art.

[0146] Therefore, the present invention adopts the above-mentioned coal gangue sorting system and method based on multi-stage screening and inclined separation, and realizes the intelligent and precise separation of coal and gangue through the cooperation of multi-stage screening, inclined separation technology, high-precision coal gangue identification model and robotic arm, thereby improving the efficiency and quality of sorting.

[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A coal gangue separation system based on multi-stage screening and inclined separation, characterized in that: include: The gangue particle size multi-stage screening subsystem comprises a main belt conveyor, a steel plate frame and n screening units; wherein the n screening units are arranged on the steel plate frame in a stepped manner, the ends of the n screening units are connected to the inclined-surface gangue sorting subsystem, any two adjacent screening units are connected by a ramp, each screening unit is equipped with a secondary belt conveyor, the first n-1 secondary belt conveyors are provided with a roller screen, and the roller spacing decreases from the first secondary belt conveyor to the n-1 secondary belt conveyors, and the first screening unit is connected to the main belt conveyor; Inclined coal gangue sorting subsystem, including n descending belt conveyors, n inclined belt conveyors and a programmable logic controller; the end of the descending belt conveyor is connected to the head end of the inclined belt conveyor, each inclined belt conveyor is driven by a servo motor and equipped with a force sensor, and the head end of each descending belt conveyor is connected to the end of the auxiliary belt conveyor; The high-precision gangue identification and sorting subsystem includes n gangue belt conveyors, n coal belt conveyors, 1 server and 1 DeltaV distributed control system; wherein, the n gangue belt conveyors and n coal belt conveyors are arranged in sequence along the conveying direction of the n inclined belt conveyors, and image acquisition equipment and a robotic arm are arranged on one side of each gangue belt conveyor and coal belt conveyor.

2. A coal gangue separation method based on multi-stage screening and inclined separation, characterized in that: The coal gangue separation system based on multi-stage screening and inclined separation as claimed in claim 1 comprises the following steps: Step S1: During the trial operation phase, adjust and determine the relevant equipment and parameters of each subsystem; Step S11, adjusting the roller spacing of the roller screens on the first n-1 screening units in the multi-stage screening subsystem for the particle size of coal gangue, and reducing the roller spacing from the first auxiliary belt conveyor to the n-1 auxiliary belt conveyor; Step S12, determining the inclination angle θ of the inclined belt conveyor in the inclined coal gangue sorting subsystem: 0<θ<min{arctanμ1, arctanμ2}, wherein μ1 represents the friction coefficient of coal on the inclined belt conveyor, and μ2 represents the friction coefficient of gangue on the inclined belt conveyor; Step S13, determining the speed V1 of the inclined belt conveyor in the inclined coal gangue sorting subsystem: V1>V2, V1>V3, wherein V2 represents the speed at which coal leaves the inclined belt conveyor, and V3 represents the speed at which gangue leaves the inclined belt conveyor; Step S14, determining the length S of the inclined belt conveyor in the inclined coal gangue sorting subsystem: Wherein, a1 represents the acceleration of coal moving on the inclined belt conveyor, a2 represents the acceleration of gangue moving on the inclined belt conveyor, and V0 represents the speed of coal and gangue when they reach the head end of the inclined belt conveyor; Step S15, determining the positions of the gangue conveyor belt conveyor and the coal conveyor belt conveyor in the high-precision gangue identification and sorting subsystem: Wherein, L1 represents the distance between the gangue conveyor belt and the inclined belt conveyor, L2 represents the distance between the coal conveyor belt and the inclined belt conveyor, and g represents the acceleration of gravity; Step S2, the system is officially put into operation, and the coal and gangue particle size multi-stage screening subsystem is used to complete the screening of coal and gangue with different particle sizes; Step S21, the main belt conveyor conveys the coal and gangue to the first screening unit of the coal gangue particle size multi-stage screening subsystem; Step S22, when the coal and gangue pass through the roller screen on the first screening unit, the coal and gangue larger than the roller spacing pass through the roller screen, and then are conveyed into the inclined coal-gangue sorting subsystem via the auxiliary belt conveyor, and the coal and gangue smaller than the roller spacing pass through the roller spacing into the ramp and finally leak into the next screening unit, and then are screened for the next particle size until the remaining coal and gangue fall into the nth screening unit; Step S3, the inclined coal gangue sorting subsystem performs preliminary sorting on coal and gangue of the same particle size range; Step S31, the coal and gangue of various particle sizes screened out in step S22 are transferred to the corresponding inclined belt conveyor through the descending belt conveyor; Step S32, the coal and gangue are conveyed to the head end of the corresponding inclined belt conveyor at a speed of V0, and begin to accelerate along the inclined direction; Step S33, under the difference of physical properties of coal and gangue of different particle sizes, through the speed, length and inclination angle set by each inclined belt conveyor, the coal and gangue leave the inclined belt conveyor and perform oblique throwing motion with different initial velocities, and fall into the coal conveying belt conveyor at a distance of L2 from the inclined belt conveyor and the gangue conveyor at a distance of L1 respectively; Step S4, the high-precision coal gangue identification and sorting subsystem performs deep optimization sorting on the initially sorted coal and gangue; Step S41, the image acquisition device acquires images of coal and gangue on the gangue conveyor belt conveyor and the coal conveyor belt conveyor; Step S42, transmitting the image to the server for preprocessing, including graying, smoothing and sharpening the image to obtain a preprocessed image; Step S43, using the image processing module of the MATLAB library in the server, based on the preprocessed image, using the grayscale co-occurrence matrix, extracting the contrast and angular second-order moment feature parameters, and using the grayscale information, extracting the grayscale mean and grayscale variance feature parameters; fusing the output texture and grayscale features to form contrast, angular second-order moment, grayscale mean, and grayscale variance features as feature vector data sets; Step S44: taking the feature vector data set as input and the actual coal and gangue real labels as output, inputting them into the Relief-MRMR-SVM model, training the model, obtaining the weight parameters of the Relief-MRMR-SVM model, and embedding the weight parameters of the Relief-MRMR-SVM model into the server; Step S45, the coal and gangue images newly collected by the image acquisition device are transmitted to the server for processing and feature extraction, and then the extracted features are classified using the trained Relief-MRMR-SVM model; Step S46: The server receives the identification results of coal and gangue, and transmits the classification information to the DeltaV distributed control system. The DeltaV distributed control system controls the robot arm to perform deep optimization sorting of coal and gangue according to the classification information and instructions.

Images