A rubber particle detection screening system and method

Abstract

The application provides a rubber particle detection and screening system and method. The rubber particle detection and screening system comprises a vibrating conveying device, a visual detection device arranged above the vibrating conveying device, the visual detection device comprising at least three groups of industrial cameras arranged above a conveying groove, the industrial cameras being used to collect images of rubber particles conveyed on the conveying groove, a negative pressure screening device arranged above the vibrating conveying device and adjacent to a discharge end of the conveying groove, the negative pressure screening device comprising a vacuum generator, a negative pressure pipeline and an electric valve, the negative pressure pipeline being provided with a suction port and a discharge port, the suction port being adjacent to the discharge end of the conveying groove, the electric valve being arranged on the negative pressure pipeline and adjacent to the suction port, the vacuum generator being connected to the negative pressure pipeline and adjacent to the discharge port, and a control device electrically connected to the industrial cameras, the vacuum generator and the electric valve. The application realizes automatic and accurate detection and rapid screening of rubber particles, and improves the detection and screening efficiency.

Description

Technical Field

[0001] This invention relates to a screening system and method, specifically a rubber particle detection and screening system and method, belonging to the field of rubber production technology. Background Technology

[0002] During rubber production, rubber particles with substandard colors are inevitably produced; these particles are called "colored rubber." Although the probability of colored rubber occurring is small, its presence is difficult to avoid, thus seriously affecting the quality of finished rubber products. Therefore, it is necessary to detect and screen for defective colored rubber during the rubber production process.

[0003] In existing technologies, the detection and screening of colored rubber is usually carried out by a vision inspection system after the rubber block pressing step. The specific process is as follows: after the rubber granules are dried in a fluidized bed, they are transported into a briquetting machine by a vibrating conveyor belt and pressed into rubber blocks of a certain size. Then, the rubber blocks are sent to the industrial camera of the vision inspection system by a conveyor. The industrial camera takes pictures of the six sides of the rubber block. The control device executes a preset algorithm to detect the pictures taken by the industrial camera. If any color abnormality is found in any part, the rubber block is guided to the waste rubber bin by the conveyor for further processing.

[0004] While the aforementioned visual inspection system can detect colored glue and screen out defective rubber blocks containing colored glue, it has the following drawbacks: First, by acquiring photos of the six surfaces of the rubber block using an industrial camera, it can only identify whether colored glue appears on the surface of the rubber block, and cannot determine whether colored glue appears inside the rubber block; Second, the rubber block is relatively large compared to the rubber granules, and the pressed rubber block generally weighs 25 kg, resulting in significant waste of discarded rubber blocks containing colored glue. Summary of the Invention

[0005] Based on the above background, the purpose of this invention is to provide a rubber particle detection and screening system that performs color rubber detection and screening on rubber particles before pressing rubber blocks, thereby achieving automated, accurate detection and rapid screening of rubber particles and improving detection and screening efficiency.

[0006] Another objective of this invention is to provide a method for detecting and screening rubber particles, thereby improving the accuracy and efficiency of screening rubber particles.

[0007] To achieve the above-mentioned objectives, the present invention provides the following technical solution:

[0008] A rubber particle detection and screening system, comprising:

[0009] A vibrating conveying device includes a load-bearing base and a conveying trough, a vibration damper, and a vibrating motor mounted on the load-bearing base. The conveying trough is located on the top of the load-bearing base and is connected to the load-bearing base through the vibration damper. One end of the conveying trough is the feeding end and the other end is the discharging end. The vibrating motor is connected to the outer wall of the conveying trough.

[0010] A visual inspection device is disposed above the vibrating conveyor and adjacent to the feed end of the conveying trough. The visual inspection device includes at least three sets of industrial cameras disposed above the conveying trough. The industrial cameras are used to capture images of the rubber particles conveyed on the conveying trough.

[0011] A negative pressure screening device is located above the vibrating conveyor and near the discharge end of the conveying trough. The negative pressure screening device includes a vacuum generator, a negative pressure pipeline, and an electric valve. The negative pressure pipeline has a suction port and a discharge port. The suction port is near the discharge end of the conveying trough. The electric valve is located on the negative pressure pipeline and near the suction port. The vacuum generator is connected to the negative pressure pipeline and is near the discharge port.

[0012] The control device is electrically connected to an industrial camera, a vacuum generator, and an electric valve. The control device is used to receive images captured by the industrial camera, identify the color adhesive based on the images, and control the opening and closing of the electric valve and the vacuum generator based on the identification results.

[0013] Preferably, each group of industrial cameras has at least two cameras, and the image acquisition direction of at least two of the industrial cameras is perpendicular to the surface of the conveyor trough.

[0014] Preferably, there are multiple negative pressure pipes and multiple electric valves. Each negative pressure pipe is equipped with an electric valve and is connected to a vacuum generator. The multiple negative pressure pipes are arranged linearly and equally spaced along the width of the conveying trough, and the suction port of each negative pressure pipe is perpendicular to the surface of the conveying trough.

[0015] A method for detecting and screening rubber particles using any of the rubber particle detection and screening systems described above, the method comprising the following steps:

[0016] Rubber granules are placed at the feed end of the conveyor trough. The vibration frequency and conveying speed v are set, and the vibration motor is turned on to make the rubber granules vibrate and move forward along the conveyor trough.

[0017] An industrial camera acquires images of rubber particles at a set sampling frequency. The control device executes an image recognition algorithm to identify colored rubber based on the acquired images. When the rubber is identified as colored rubber, the control device records the first coordinates (x1, y1) of the colored rubber on the conveying surface of the conveying trough and the detection time t1.

[0018] The control device executes a trajectory prediction algorithm. Based on the first coordinate of the color glue and the maximum vibration deviation u, it predicts the abscissa x2 corresponding to the longitudinal coordinate y2 of the conveying surface of the conveying trough where the center line of the suction port of the negative pressure pipeline is located, and obtains the predicted second coordinate range of the color glue on the conveying surface of the conveying trough (x2-u, y2)~(x2+u, y2).

[0019] The control device divides the conveying surface area of ​​the conveying channel corresponding to the negative pressure pipe into n partitions according to the number of negative pressure pipes n. The width of each partition is d / n, and the boundary coordinates of each partition are obtained, where d is the width of the conveying channel.

[0020] The control device compares the predicted second coordinate range with the boundary coordinates of each partition. When the predicted second coordinate range is completely within a single partition, the electric valve corresponding to the negative pressure pipeline of that partition is opened at the material suction time t2 to suction material. When the predicted second coordinate range is in multiple adjacent partitions, all the electric valves corresponding to the negative pressure pipelines of multiple adjacent partitions are opened at the material suction time t2 to suction material.

[0021] After the control device maintains the electric valve open for a set period of time, it closes the electric valve. The colored rubber and several adjacent rubber particles are sucked into the negative pressure pipe and discharged through the discharge port. The control device then turns on the vacuum generator and maintains it for a set period of time, keeping the negative pressure pipe in a vacuum state in preparation for the next material suction operation.

[0022] Preferably, the trajectory prediction algorithm is one of the following: Kalman recursive filtering algorithm, ARIMA time series analysis algorithm, Markov chain algorithm, random forest algorithm, and SVM support vector machine regression algorithm.

[0023] Preferably, the trajectory prediction algorithm is the Kalman recursive filtering algorithm. In the model construction of the trajectory prediction algorithm, the state equation adopts the following mathematical expression.

[0024] x k =ax k-1 +bu k-1 +w k-1 ;

[0025] In the formula, x k Let x represent the position and velocity vector of the pigment at time k. k-1 Let a represent the position and velocity vector of the pigment at time k-1, a represent the 2×2 state transition matrix, b represent the 2×1 control input matrix, and u represent the position and velocity vector of the pigment at time k-1. k-1 w represents the control vector containing the vibration frequency. k-1 Indicates process noise;

[0026] The measurement equation uses the following mathematical expression:

[0027] zk =cx k +v k ;

[0028] In the formula, z k Let c represent the position of the pigment measured at time k, and let v represent the 1×2 measurement matrix. k This indicates measurement noise.

[0029] Preferably, the method for calculating the material suction time t2 includes the following steps:

[0030] Second-order state estimation is performed using the Kalman recursive filtering algorithm to iteratively predict the position and velocity of the colorant.

[0031] Record the perpendicular distance between the predicted position of the colored glue and the straight line on the vertical coordinate of the conveying surface of the conveying trough where the center line of the suction port of the negative pressure pipeline is located;

[0032] Accumulate the predicted speed of the colored glue each time and calculate the cumulative distance traveled by the colored glue;

[0033] When the cumulative travel distance of the color glue first exceeds the aforementioned vertical distance, the predicted time at this moment is recorded as the material suction time t2. Preferably, the maximum vibration deviation u is calculated using the following mathematical expression.

[0034] u = 3σ(x);

[0035] In the formula, σ(x) represents the standard deviation of the x-coordinate of the conveying surface of the conveying tank due to the vibration of the color glue. σ(x) is calculated using the following mathematical expression.

[0036]

[0037] In the formula, x i x represents the value of the collected pigment at the i-th position. p This represents the average value of the color glue at different times when it was collected, where m is the number of data collection points.

[0038] Compared with the prior art, the present invention has the following advantages:

[0039] This invention discloses a rubber particle detection and screening system. A vibrating conveyor transports rubber particles through a visual inspection device. As the rubber particles continuously change posture during vibration, at least three industrial cameras can comprehensively capture images of each side of the particles. When a control device identifies colored rubber based on the captured images, it controls the opening and closing of an electric valve and a vacuum generator. This causes a negative pressure screening device to draw the colored rubber and several adjacent rubber particles into a negative pressure pipe, thus removing the colored rubber from the vibrating conveyor. The good quality rubber particles continue to vibrate and advance into a subsequent briquetting machine to be pressed into rubber blocks. This rubber particle detection and screening system achieves automated, accurate detection and rapid screening of rubber particles, improving detection and screening efficiency.

[0040] The present invention provides a rubber particle detection and screening method that can accurately predict the movement trajectory of the colored rubber after the control device identifies the colored rubber based on the image acquired by the vision detection device. Based on the predicted future position of the colored rubber, the electric valve of the corresponding zone is opened to allow the negative pressure pipeline to suck up the material, thereby reducing the number of good rubber particles that are rejected along with the colored rubber, further improving screening efficiency and reducing the loss of good products. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0042] Figure 1 This is a top view schematic diagram of a rubber particle detection and screening system according to the present invention;

[0043] Figure 2 This is a partial front view schematic diagram of a rubber particle detection and screening system according to the present invention;

[0044] Figure 3 This is a schematic diagram of the predicted second coordinate range being completely within a single partition in this invention;

[0045] Figure 4 This is a schematic diagram of the predicted second coordinate range in this invention when it is located in multiple adjacent partitions.

[0046] In the diagram: 1. Vibrating conveyor; 2. Visual inspection device; 3. Negative pressure screening device; 4. Control device; 5. Fluidized bed; 6. Measuring scale; 7. Briquetting machine; 101. Load-bearing base; 102. Conveying trough; 103. Vibration damper; 104. Vibrating motor; 1021. Feeding end; 1022. Discharge end; 201. Industrial camera; 301. Vacuum generator; 302. Negative pressure pipeline; 303. Electric valve; 3021. Suction port; 3022. Discharge port. Detailed Implementation

[0047] The technical solution of the present invention will be further described in detail below through specific embodiments and in conjunction with the accompanying drawings. It should be understood that the implementation of the present invention is not limited to the following embodiments, and any modifications and / or alterations made to the present invention will fall within the protection scope of the present invention.

[0048] In this invention, unless otherwise specified, all parts and percentages are by weight, and the equipment and raw materials used are commercially available or commonly used in the art. Unless otherwise specified, the methods in the following embodiments are conventional methods in the art. Unless otherwise specified, the components or equipment in the following embodiments are general standard parts or components known to those skilled in the art, and their structures and principles can be learned by those skilled in the art through technical manuals or conventional experimental methods.

[0049] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings. In this detailed description, numerous specific details are set forth to facilitate explanation and provide a thorough understanding of the embodiments of the present invention. However, one or more embodiments may be practiced by those skilled in the art without these specific details.

[0050] like Figure 1 and Figure 2 As shown, an embodiment of the present invention discloses a rubber particle detection and screening system, including a vibrating conveyor 1, a visual inspection device 2, a negative pressure screening device 3, and a control device 4. This rubber particle detection and screening system can be used in conjunction with a rubber particle pressing production line. The vibrating conveyor 1 is located between a fluidized bed 5 and a briquetting machine 7. Rubber particles conveyed from the fluidized bed 5 are transported by the vibrating conveyor 1 to a weighing scale 6 and then fed into the briquetting machine 7 to be pressed into rubber blocks.

[0051] The vibrating conveyor 1 includes a load-bearing base 101 and a conveying trough 102, a vibration damper 103, and a vibrating motor 104 disposed on the load-bearing base 101. The conveying trough 102 is located on the top of the load-bearing base 101 and is connected to the load-bearing base 101 through the vibration damper 103. One end of the conveying trough 102 is the feed end 1021 and the other end is the discharge end 1022. The vibrating motor 104 is connected to the outer wall of the conveying trough 102. The vibration damper 103 has a damping spring inside, which is used to reduce the vibration transmitted from the conveying trough 102 to the load-bearing base 101.

[0052] The vision inspection device 2 is located above the vibrating conveyor 1 and adjacent to the feed end 1021 of the conveying trough 102. The vision inspection device 2 includes three sets of industrial cameras 201 positioned above the conveying trough 102. The industrial cameras 201 are used to capture images of the rubber granules conveyed on the conveying trough 102 and are equipped with necessary light sources. The number of sets of industrial cameras 201 can be increased according to actual application needs. Each set of industrial cameras 201 consists of two cameras, with the image acquisition direction of both cameras perpendicular to the surface of the conveying trough 102. The number of sets of industrial cameras 201 can also be increased according to actual application needs. The image acquisition process involves the rubber granules moving along the conveying trough 102 under the vibration of the vibrating motor 104. During continuous vibration, the posture of the rubber granules changes continuously, thus achieving automatic flipping. The three sets of industrial cameras 201 capture images at different times, ensuring that approximately every side of all rubber granules is captured.

[0053] The negative pressure screening device 3 is located above the vibrating conveyor 1 and adjacent to the discharge end 1022 of the conveying trough 102. The negative pressure screening device 3 includes a vacuum generator 301, a negative pressure pipe 302, and an electric valve 303. The negative pressure pipe 302 has a suction port 3021 and a discharge port 3022. The suction port 3021 is adjacent to the discharge end 1022 of the conveying trough 102. The electric valve 303 is located on the negative pressure pipe 302 and adjacent to the suction port 3021. The vacuum generator 301 is connected to the negative pressure pipe 302 and adjacent to the discharge port 3022. There are three negative pressure pipes 302 and three electric valves 303. Each negative pressure pipe 302 is equipped with one electric valve 303, and each negative pressure pipe 302 is connected to the vacuum generator 301. Three negative pressure pipes 302 are arranged linearly and at equal intervals along the width of the conveying trough 102, and the suction port 3021 of each negative pressure pipe 302 is perpendicular to the surface of the conveying trough 102. Of course, the number of negative pressure pipes 302 and electric valves 303 can be increased or decreased appropriately according to actual application needs.

[0054] The control device 4 is electrically connected to the industrial camera 201, the vacuum generator 301, and the electric valve 303, respectively. The control device 4 receives images captured by the industrial camera 201, identifies color adhesive based on the images, and controls the opening and closing of the electric valve 303 and the vacuum generator 301 based on the identification results. When identifying color adhesive, if any image captured by the three sets of industrial cameras 201 contains a color that does not meet the requirements, it is determined that the image contains color adhesive.

[0055] This rubber granule detection and screening system uses a vibrating conveyor 1 to transport rubber granules through a vision inspection device 2. As the rubber granules continuously change posture during their vibration, at least three industrial cameras 201 can comprehensively capture images of each side of the granules. When the control device 4 identifies the colored rubber based on the captured images, it controls the opening and closing of the electric valve 303 and the vacuum generator 301, causing the negative pressure screening device 3 to draw the colored rubber and several adjacent rubber granules into the negative pressure pipe 302. This removes the colored rubber from the vibrating conveyor 1, while the good quality rubber granules continue to vibrate and enter the subsequent briquetting machine 7 to be pressed into rubber blocks. This rubber granule detection and screening system achieves automated, accurate detection and rapid screening of rubber granules, improving detection and screening efficiency.

[0056] The embodiments of the present invention also disclose a method for detecting and screening rubber particles using the above-described rubber particle detection and screening system, the method comprising the following steps:

[0057] Rubber granules are placed at the feed end 1021 of the conveying trough 102. The vibration frequency and conveying speed v are set, and the vibration motor 104 is turned on so that the rubber granules vibrate and move forward along the conveying trough 102.

[0058] The industrial camera 201 acquires images of rubber particles at a set sampling frequency. The control device 4 executes an image recognition algorithm to identify colored rubber based on the acquired images. When the rubber is identified as colored rubber, the control device 4 records the first coordinates (x1, y1) of the colored rubber on the conveying surface of the conveying trough 102 and the detection time t1.

[0059] The control device 4 executes a trajectory prediction algorithm. Based on the first coordinate of the color glue and the maximum vibration deviation u, it predicts the abscissa x2 corresponding to the longitudinal coordinate y2 of the color glue on the conveying surface of the conveying trough 102 where the center line of the suction port 3021 of the negative pressure pipe 302 is located. The predicted second coordinate range of the color glue on the conveying surface of the conveying trough 102 is obtained (x2-u, y2)~(x2+u, y2).

[0060] The control device 4 divides the conveying surface area of ​​the conveying channel 102 corresponding to the negative pressure pipe 302 into n partitions according to the number n of negative pressure pipes 302. The width of each partition is d / n, and the boundary coordinates of each partition are obtained, where d is the width of the conveying channel 102.

[0061] Control device 4 compares the predicted second coordinate range with the boundary coordinates of each partition, such as... Figure 3 As shown, when the predicted second coordinate range is entirely within a single zone, the electric valve 303 corresponding to the negative pressure pipe 302 of that zone is opened at the material suction time t2 to suction material; as Figure 4 As shown, when the predicted second coordinate range is in multiple adjacent zones, all electric valves 303 corresponding to the negative pressure pipelines 302 of the multiple adjacent zones will be opened at the material suction time t2 to suction material.

[0062] After the control device 4 maintains the electric valve 303 open for a set period of time, it closes the electric valve 303. The colored glue and multiple adjacent rubber particles are sucked into the negative pressure pipe 302 and discharged through the discharge port 3022. The control device 4 turns on the vacuum generator 301 and maintains it for a set period of time, so that the negative pressure pipe 302 maintains a vacuum state in preparation for the next material suction operation.

[0063] In the above steps, the trajectory prediction algorithm is one of the following: Kalman recursive filtering algorithm, ARIMA time series analysis algorithm, Markov chain algorithm, random forest algorithm, or SVM support vector machine regression algorithm. The position coordinates of the rubber particles at multiple different times during their vibration and movement are pre-collected to form a dataset. Based on this dataset, one of the above algorithms is used to construct a trajectory prediction algorithm suitable for predicting the colored rubber trajectory.

[0064] The following section provides a detailed explanation of the Kalman recursive filtering algorithm using trajectory prediction as an example. The Kalman recursive filtering algorithm utilizes state equations to estimate the future state of the process, enabling it to predict the dynamic trajectory of colored rubber. Considering the requirements of rubber particle detection scenarios and the need for rapid prediction of the colored rubber's vibration trajectory, the Kalman recursive filtering algorithm is well-suited because it is faster than other algorithms and saves computational resources. Furthermore, due to the given constraint of the maximum vibration deviation u, even though the prediction accuracy of the Kalman recursive filtering algorithm is slightly lower than other algorithms, the maximum deviation range can tolerate a certain prediction error. The state equation of the Kalman recursive filtering algorithm uses the following mathematical expression...

[0065] x k =ax k-1 +bu k-1 +w k-1 ;

[0066] In the formula, x k Let x represent the position and velocity vector of the pigment at time k. k-1 Let a represent the position and velocity vector of the pigment at time k-1, a represent the 2×2 state transition matrix, b represent the 2×1 control input matrix, and u represent the position and velocity vector of the pigment at time k-1. k-1w represents the control vector containing the vibration frequency. k-1 Indicates process noise;

[0067] The measurement equation uses the following mathematical expression:

[0068] z k =cx k +v k ;

[0069] In the formula, z k Let c represent the position of the pigment measured at time k, and let v represent the 1×2 measurement matrix. k This indicates measurement noise.

[0070] The calculation method for the material suction time t2 in the above steps includes the following steps:

[0071] Second-order state estimation is performed using the Kalman recursive filtering algorithm to iteratively predict the position and velocity of the colorant.

[0072] Record the perpendicular distance between the predicted position of the color glue and the straight line on the vertical coordinate of the conveying surface of the conveying trough 102 where the center line of the suction port 3021 of the negative pressure pipe 302 is located.

[0073] Accumulate the predicted speed of the colored glue each time and calculate the cumulative distance traveled by the colored glue;

[0074] When the cumulative travel distance of the colored adhesive first exceeds the aforementioned vertical distance, the predicted time at this moment is recorded as the material suction time t2. In the above steps, the maximum vibration deviation u is calculated using the following mathematical expression.

[0075] u = 3σ(x);

[0076] In the formula, σ(x) represents the standard deviation of the x-coordinate of the conveying surface of the color glue vibration in the conveying tank 102. σ(x) is calculated using the following mathematical expression.

[0077]

[0078] In the formula, x i x represents the value of the collected pigment at the i-th position. p This represents the average value of the color glue at different times when it was collected, where m is the number of data collection points.

[0079] This method can accurately predict the movement trajectory of the colored rubber after the control device 4 identifies the colored rubber based on the image collected by the vision detection device 2. Based on the predicted future position of the colored rubber, the electric valve 303 of the corresponding zone is opened to allow the negative pressure pipeline 302 to suck up the material, thereby reducing the good rubber particles that are rejected along with the colored rubber, further improving the screening efficiency and reducing the loss of good products.

[0080] This article uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. It should be noted that those skilled in the art can make several improvements and modifications to the present invention without departing from the principles of the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims

1. A method for detecting and screening rubber particles, wherein the method employs a rubber particle detection and screening system, characterized in that: The rubber particle detection and screening system includes: Vibrating conveyor (1), the vibrating conveyor (1) includes a load-bearing base (101) and a conveying trough (102), a damper (103) and a vibrating motor (104) disposed on the load-bearing base (101). The conveying trough (102) is located on the top of the load-bearing base (101). The conveying trough (102) is connected to the load-bearing base (101) through the damper (103). One end of the conveying trough (102) is the feeding end (1021) and the other end is the discharging end (1022). The vibrating motor (104) is connected to the outer wall of the conveying trough (102). A visual inspection device (2) is located above the vibrating conveying device (1) and adjacent to the feed end (1021) of the conveying trough (102). The visual inspection device (2) includes at least three sets of industrial cameras (201) located above the conveying trough (102). The industrial cameras (201) are used to collect images of rubber particles conveyed on the conveying trough (102). A negative pressure screening device (3) is located above the vibrating conveying device (1) and near the discharge end (1022) of the conveying trough (102). The negative pressure screening device (3) includes a vacuum generator (301), a negative pressure pipe (302) and an electric valve (303). The negative pressure pipe (302) is provided with a suction port (3021) and a discharge port (3022). The suction port (3021) is near the discharge end (1022) of the conveying trough (102). The electric valve (303) is located on the negative pressure pipe (302) and near the suction port (3021). The vacuum generator (301) is connected to the negative pressure pipe (302) and is near the discharge port (3022). The control device (4) is electrically connected to the industrial camera (201), the vacuum generator (301) and the electric valve (303) respectively. The control device (4) is used to receive the image captured by the industrial camera (201), identify the color glue according to the image, and control the opening and closing of the electric valve (303) and the opening and closing of the vacuum generator (301) according to the identification result. The number of each group of industrial cameras (201) is at least two, and the image acquisition direction of at least two of the industrial cameras (201) is perpendicular to the surface of the conveying groove (102); The number of negative pressure pipes (302) and the number of electric valves (303) are both multiple. Each negative pressure pipe (302) is equipped with an electric valve (303). Each negative pressure pipe (302) is connected to a vacuum generator (301). The multiple negative pressure pipes (302) are arranged linearly and equally spaced along the width direction of the conveying groove (102), and the suction port (3021) of each negative pressure pipe (302) is perpendicular to the surface of the conveying groove (102). The method includes the following steps: Rubber granules are placed at the feed end (1021) of the conveying trough (102), the vibration frequency and conveying speed v are set, and the vibration motor (104) is turned on so that the rubber granules vibrate and move forward along the conveying trough (102). An industrial camera (201) acquires images of rubber particles at a set sampling frequency. The control device (4) executes an image recognition algorithm to identify colored glue based on the acquired image. When the color glue is identified, the control device (4) records the first coordinate (x1, y1) of the color glue on the conveying surface of the conveying groove (102) and the detection time t1. The control device (4) executes a trajectory prediction algorithm. Based on the first coordinate of the color glue and the maximum vibration deviation u, it predicts the abscissa x2 corresponding to the longitudinal coordinate y2 of the conveying surface of the conveying groove (102) where the center line of the suction port (3021) of the negative pressure pipe (302) is located. It obtains the predicted second coordinate range (x2-u, y2)~(x2+u, y2) of the color glue on the conveying surface of the conveying groove (102). The control device (4) divides the conveying surface area of ​​the conveying groove (102) corresponding to the negative pressure pipe (302) into n partitions according to the number n of the negative pressure pipe (302), the width of each partition is d / n, and obtains the boundary coordinates of each partition, where d is the width of the conveying groove (102); The control device (4) compares the predicted second coordinate range with the boundary coordinates of each partition. When the predicted second coordinate range is completely within a single partition, the electric valve (303) corresponding to the negative pressure pipe (302) of that partition is opened at the material suction time t2 to suction the material. When the predicted second coordinate range is in multiple adjacent partitions, all the electric valves (303) corresponding to the negative pressure pipes (302) of multiple adjacent partitions are opened at the material suction time t2 to suction the material. After the control device (4) maintains the electric valve (303) open for a set period, it closes the electric valve (303), and the colored glue and multiple adjacent rubber particles are sucked into the negative pressure pipe (302) and discharged through the discharge port (3022). The control device (4) turns on the vacuum generator (301) and maintains it for a set period, so that the negative pressure pipe (302) maintains a vacuum state in preparation for the next material suction operation. The maximum vibration deviation u is calculated using the following mathematical expression: u = 3σ(x); where σ(x) represents the standard deviation of the x-coordinate of the conveying surface of the color glue vibration in the conveying groove (102), and σ(x) is calculated using the following mathematical expression. , In the formula, x i x represents the value of the collected pigment at the i-th position. p This represents the average value of the color glue at different times when it was collected, where m is the number of data collection points.

2. The method for detecting and screening rubber particles according to claim 1, characterized in that: The trajectory prediction algorithm is one of the following: Kalman recursive filtering algorithm, ARIMA time series analysis algorithm, Markov chain algorithm, random forest algorithm, and SVM support vector machine regression algorithm.

3. The method for detecting and screening rubber particles according to claim 2, characterized in that: The trajectory prediction algorithm is a Kalman recursive filtering algorithm. In the model construction of the trajectory prediction algorithm, the state equation adopts the following mathematical expression: x k =ax k-1 + bu k-1 + w k-1 In the formula, x k Let x represent the position and velocity vector of the pigment at time k. k-1 Let a represent the position and velocity vector of the pigment at time k-1, a represent the 2×2 state transition matrix, b represent the 2×1 control input matrix, and u represent the position and velocity vector of the pigment at time k-1. k-1 w represents the control vector containing the vibration frequency. k-1 Indicates process noise; The measurement equation uses the following mathematical expression, z k = cx k + v k In the formula, z k Let c represent the position of the pigment measured at time k, and let v represent the 1×2 measurement matrix. k This indicates measurement noise.

4. The method for detecting and screening rubber particles according to claim 3, characterized in that: The method for calculating the material suction time t2 includes the following steps: Second-order state estimation is performed using the Kalman recursive filtering algorithm to iteratively predict the position and velocity of the colorant. Record the vertical distance between the predicted position of the color glue and the straight line of the conveying surface of the conveying trough (102) where the center line of the suction port (3021) of the negative pressure pipe (302) is located; Accumulate the predicted speed of the colored glue each time and calculate the cumulative distance traveled by the colored glue; When the cumulative travel distance of the color glue exceeds the above-mentioned vertical distance for the first time, the predicted time at this moment is recorded as the material suction time t2.

Images