Control method and conveying device for air conveying

Abstract

The embodiment of the application discloses a kind of control method and conveying equipment of winnowing, belong to the technical field of plastic winnowing, solve the problem of low overall efficiency of winnowing machine when existing plastic winnowing.The image of material to be selected on the conveying belt is obtained, and the distribution information of the plastic to be selected is determined based on the pixel point information corresponding to the image of the material to be selected.Based on the distribution information of the plastic to be selected, the corresponding winnowing area is determined, and the air outlet to be adjusted on the winnowing machine is determined based on the plastic proportion corresponding to each winnowing area.The reference air outlet associated with the air outlet to be adjusted is determined, and the parameter control function is constructed based on the data of the plastic to be selected corresponding to the reference air outlet.The wind speed and air outlet angle corresponding to the reference air outlet are output based on the parameter control function, to control the air outlet to be adjusted.The winnowing machine is adjusted, and the plastic composition of the material after winnowing is detected, to control the winnowing machine based on the detection result.

Description

Technical Field

[0001] This application relates to the field of plastic air classification technology, and in particular to a control method and conveying equipment for air classification conveying. Background Technology

[0002] In the field of plastic recycling and processing, air classifiers, as highly efficient material separation equipment, are widely used in the classification and screening of plastics. The traditional working principle of an air classifier relies primarily on the force of airflow. By adjusting the air velocity and outlet angle, plastic materials of different weights, shapes, and materials are separated. In this process, a conveyor belt first evenly spreads and transports the plastic material to be selected to the air classification area. Subsequently, the air classifier releases airflow of specific intensity and direction through preset outlets, blowing the material and causing plastics with different properties to be guided to different collection areas due to the different forces applied.

[0003] However, existing air classifiers face a significant problem in actual operation: due to the often uneven distribution of the plastic materials to be classified and the large differences in material properties such as density and shape, the workload of each air outlet during the air classification process is extremely uneven. Specifically, some air outlets, although handling less material, still release a large airflow, while other air outlets are overloaded due to handling too much or too heavy material, resulting in increased equipment energy consumption and reduced overall efficiency of the air classifier. Summary of the Invention

[0004] This application provides a control method and conveying equipment for air separation conveying, which solves the following technical problems: In the prior art, some air outlets, although they handle less material, still release a large airflow, while some air outlets are overloaded due to handling too much or too heavy material, resulting in increased equipment energy consumption and reduced overall efficiency of the air separator.

[0005] The embodiments of this application adopt the following technical solutions:

[0006] This application provides a control method for air-separated conveying. The method includes: acquiring an image of materials to be selected on a conveyor belt; determining the distribution information of plastics to be selected based on the pixel information corresponding to the material image; determining the air-separation area to which the plastics belong based on the distribution information; determining the air outlet to be adjusted on the air separator based on the plastic percentage corresponding to each air-separation area; determining a reference air outlet that is related to the air outlet to be adjusted; constructing a parameter control function based on the plastic data to be selected corresponding to the reference air outlet; outputting the wind speed and outlet angle corresponding to the reference air outlet based on the parameter control function; controlling the air outlet to be adjusted based on the wind speed and outlet angle; performing air separation on the plastics to be selected based on the adjusted air separator; and detecting the plastic composition of the separated material; and performing secondary control of the air separator based on the detection results.

[0007] This application utilizes image processing technology to acquire information about the materials to be selected in real time, avoiding errors and delays caused by manual intervention and improving the accuracy and efficiency of data acquisition. By analyzing the plastic content in different areas, areas requiring key adjustments are identified, allowing for targeted adjustment of the air vents and optimization of the air separation effect. Through correlation analysis, other air vents affecting the effect of the vent to be adjusted are identified, and data from these vents is used to construct a control function, achieving a more precise control strategy. The optimal wind speed and outlet angle are calculated using the parameter control function, enabling precise control of the vent to be adjusted, improving air separation efficiency and quality, balancing the air separation workload of each vent, and enhancing the overall performance of the air separator. Secondary adjustments based on the detection results can further optimize the performance of the air separator, improve sorting accuracy and efficiency, and form a closed-loop optimization process.

[0008] In one implementation of this application, an image of the materials to be selected on a conveyor belt is acquired. Based on the pixel information corresponding to the image of the materials to be selected, the distribution information of the plastics to be selected is determined. Specifically, this includes: determining each pixel in the output tensor corresponding to the image of the materials to be selected; determining the material category corresponding to each pixel based on the material category probability corresponding to each pixel; dividing adjacent pixels with the same material category to obtain a set of pixels corresponding to different materials to be selected; determining a set of reference pixels belonging to the plastic category in the set of pixels to obtain the image region of the plastics to be selected based on the set of reference pixels; marking the image region of the plastics to be selected in the image of the materials to be selected, and inputting the marked image into a recurrent neural network to determine the conveying order and conveying time correlation of the plastics to be selected on the conveyor belt through the recurrent neural network; and obtaining the distribution information of the plastics to be selected based on the conveying order, the conveying time correlation, and the area of ​​the image region of the plastics to be selected.

[0009] In one implementation of this application, after determining the distribution information of the plastics to be selected based on the pixel information corresponding to the image of the material to be selected, the method further includes: determining the shape of the plastics to be selected based on the distribution information of the plastics to be selected; dividing the plastics to be selected with the same shape into the same set, so as to determine the plastic shape distribution information based on the number of plastics to be selected in each set; comparing the plastic shape distribution information with the plastic shape distribution information in a preset shape distribution table, so as to obtain the air classifier control coefficient corresponding to the reference plastic shape distribution information with the highest similarity; wherein, the preset shape distribution table includes multiple reference plastic shape distribution information, and also includes the air classifier control coefficients corresponding to the multiple reference plastic shape distribution information respectively.

[0010] In one implementation of this application, the distribution information of the plastics to be selected is used to determine the air classification area to which they belong. Based on the distribution ratio of each air classification area, the air outlet to be adjusted on the air classifier is determined. Specifically, this includes: determining the placement area of ​​each plastic on the conveyor belt based on the distribution information of the plastics to be selected; dividing the conveyor belt into multiple air classification areas based on the number of air outlets corresponding to the air classifier; obtaining the plastic ratio corresponding to each of the multiple air classification areas based on the placement area of ​​each plastic; comparing the plastic ratio corresponding to each air classification area with a preset ratio threshold; and if the plastic ratio is greater than the preset ratio threshold, the air classifier outlet corresponding to the air classification area is taken as the air outlet to be adjusted.

[0011] In one implementation of this application, a reference air outlet associated with the air outlet to be adjusted is determined. Based on the candidate plastic data corresponding to the reference air outlet, a parameter control function is constructed. Specifically, this includes: determining the reference air outlet adjacent to the air outlet to be adjusted and obtaining the plastic percentage corresponding to the reference air outlet; determining the adjustable plastic percentage based on the difference between the plastic percentage and a preset percentage threshold; obtaining the placement area of ​​the candidate plastic corresponding to the reference air outlet; determining the air separation distance between the placement area and the air separation area of ​​the air outlet to be adjusted; determining the candidate plastic in a stacked state based on the placement area of ​​the candidate plastic corresponding to the reference air outlet; obtaining plastic stacking distribution data based on the quantity of candidate plastic in a stacked state and the area of ​​the stacked area; and constructing a parameter control function based on the adjustable plastic percentage, air separation distance, and plastic stacking distribution data.

[0012] In one implementation of this application, a parameter control function is constructed based on adjustable plastic percentage, distance, and plastic accumulation distribution data. Specifically, the constructed wind speed control function is as follows:

[0013]

[0014] The constructed air outlet angle control function is as follows:

[0015]

[0016] Based on the wind speed control function and the outlet angle control function, a parameter control function is obtained. The reference air outlet is then controlled based on the output value of this parameter control function. Here, c is the wind speed of the reference air outlet; θ is the wind direction angle of the reference air outlet; p is the adjustable plastic percentage; d is the wind separation distance; s is the plastic accumulation distribution data; v0 is the initial wind speed of the reference air outlet; θ0 is the initial wind direction angle of the reference air outlet; k1 is the first weight; k2 is the second weight; k3 is the third weight; k4 is the fourth weight; k5 is the fifth weight; k6 is the sixth weight; k7 is the seventh weight; k8 is the eighth weight; s d s represents the average bulk density of the candidate plastics in their stacked state;a The area of ​​the accumulation region; This is the first noise term; This is the second noise term.

[0017] In one implementation of this application, after constructing the parameter control function, the method further includes: determining the coefficient type of the air outlet selector control coefficient; wherein the coefficient type includes speed type and angle type; adjusting the wind speed control function based on the control coefficient corresponding to the speed type; and adjusting the air outlet angle control function based on the control coefficient corresponding to the angle type.

[0018] In one implementation of this application, the plastic composition of the material after air classification is detected, and the air classifier is then adjusted based on the detection results. Specifically, this includes: sampling the impurity collection area after air classification to obtain the plastic content based on the sampling results; if the plastic content does not meet a preset content condition, inputting the plastic sample image from the impurity collection area into a preset plastic classification model to output the plastic category corresponding to the impurity collection area based on the preset plastic classification model; determining the quantity of plastics corresponding to different plastic categories in the impurity area to determine the plastic category to be processed based on the quantity; and determining the air outlet control data corresponding to the plastic category to be processed in a preset plastic information database to perform secondary adjustment of the air classifier based on the air outlet control data.

[0019] In one implementation of this application, the air classifier is subjected to secondary control based on air outlet control data. Specifically, this includes: determining the air velocity difference and angle difference based on the current air outlet data and the air outlet control data corresponding to the type of plastic to be processed; wherein the air outlet control data includes different air outlet velocities and different air outlet angles; adjusting each air outlet of the air classifier based on the air velocity difference and angle difference; after a preset time interval, re-inspecting the impurity collection area; if the plastic content of the sampled plastic does not meet the preset content condition, selecting a new air outlet velocity and a new air outlet angle from the air outlet control data to re-control the air classifier; wherein the new air outlet velocity is greater than the previously selected air outlet velocity, and the new air outlet angle is greater than the previously selected air outlet angle.

[0020] This application provides an air-separation conveying device, which consists of an air-separation channel, a support frame, and an air-separator housing. The device further includes: a conveyor belt evenly divided into multiple air-separation zones along its width, used to convey materials to be separated; an image monitoring device mounted on the air-separation box, used to acquire images of the materials to be separated on the conveyor belt; an air-separation box located on the side of the image monitoring device facing the discharge direction of the conveyor belt, with the top of the air-separation box connected to the air-separation channel and mounting holes at the bottom corresponding to each air-separation zone; a retractable air outlet, including a retractable air duct mounted in the mounting hole, the upper end of the air duct connected to the inside of the air-separation box; and a driving component capable of moving the air duct up and down.

[0021] The above-mentioned technical solutions adopted in this application embodiment can achieve the following beneficial effects: This application embodiment acquires the information of the material to be selected in real time through image processing technology, avoiding errors and delays caused by manual intervention, and improving the accuracy and efficiency of data acquisition. By analyzing the plastic ratio in different areas, areas that need to be adjusted are identified, thereby allowing for targeted adjustment of the air outlets and optimization of the air separation effect. Through correlation analysis, other air outlets affecting the effect of the air outlet to be adjusted are identified, and the data of these air outlets are used to construct a control function to achieve a more precise control strategy. Through the parameter control function, the optimal wind speed and air outlet angle are calculated, achieving precise control of the air outlet to be adjusted, improving air separation efficiency and quality, balancing the air separation workload of each air outlet, and improving the overall performance of the air separator. Secondary adjustment based on the detection results can further optimize the performance of the air separator, improve the accuracy and efficiency of sorting, and form a closed-loop optimization process. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application 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 some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0023] Figure 1 A flowchart illustrating a control method for air separation and conveying provided in an embodiment of this application;

[0024] Figure 2 This is a front view of an air-separating conveyor provided in an embodiment of the present application;

[0025] Figure 3 This is a schematic diagram of the rear of an air-separating conveyor provided in an embodiment of this application;

[0026] Figure 4This is a side view of an air separation conveyor provided in an embodiment of this application.

[0027] Figure label:

[0028] 101 Air separation channel, 102 Image monitoring device, 103 Air separation box, 104 Drive unit, 105 Telescopic air outlet, 106 Mounting hole, 107 Bracket, 108 Air outlet, 109 Conveyor belt, 110 Air separator housing. Detailed Implementation

[0029] This application provides a control method and conveying equipment for air separation and conveying.

[0030] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0031] Figure 1 A flowchart of a control method for air separation and conveying provided in an embodiment of this application is shown below. Figure 1 As shown, the control method for air separation conveying includes the following steps:

[0032] Step 101: Obtain the image of the material to be selected on the conveyor belt, and determine the distribution information of the plastic to be selected based on the pixel information corresponding to the image of the material to be selected.

[0033] In one implementation of this application, each pixel in the output tensor corresponding to the image of the candidate material is determined. Based on the material category probability corresponding to each pixel, the material category corresponding to each pixel is determined. Adjacent pixels with the same material category are divided to obtain pixel sets corresponding to different candidate materials. Within the pixel sets, a reference pixel set belonging to the plastic category is determined to obtain the candidate plastic image region based on the reference pixel set. The candidate plastic image region is marked in the candidate material image, and the marked image is input into a recurrent neural network to determine the conveying order and conveying time correlation of the candidate plastic on the conveyor belt. Based on the conveying order, conveying time correlation, and the area of ​​the candidate plastic image region, the distribution information of the candidate plastic is obtained.

[0034] Specifically, firstly, an image of the materials to be selected on the conveyor belt is acquired, and an output tensor is generated. This output tensor is a digital representation of the image, containing information about each pixel in the image. For each pixel in the output tensor, the system calculates the probability that the pixel belongs to different material categories based on a pre-trained model, such as a convolutional neural network. Based on these probabilities, each pixel is assigned a most probable material category. The training process of the convolutional neural network involves using preset image pixel samples as input and the pixel categories corresponding to these input samples as output to train the preset model. If the training results meet the requirements, the convolutional neural network is obtained.

[0035] Furthermore, adjacent pixels belonging to the same material category are grouped into sets, so that each set represents a region of a specific material in the image. Among the resulting sets, the set of pixels belonging to the plastic category is identified; this set of plastic pixels is the reference set. The region corresponding to the reference set is marked on the original image, which is the candidate plastic image region. This allows for a visual indication of which parts contain plastic material.

[0036] Furthermore, the labeled images are input into a recurrent neural network. In this embodiment, the recurrent neural network is used to analyze the correlation between the conveying sequence and conveying time of plastic materials on the conveyor belt. Specifically, the recurrent neural network analyzes the plastic materials in the image and determines their conveying sequence and conveying time correlation based on their position, shape, and possible movement patterns on the conveyor belt. Simultaneously, this embodiment also calculates the area of ​​each candidate plastic image region to determine the quantity and scale of the material. Finally, by combining the conveying sequence, conveying time correlation, and area information, distribution information of the candidate plastics is generated. The distribution information in this embodiment is a detailed report, containing information such as the specific position, quantity, and estimated arrival time of the plastic materials on the conveyor belt.

[0037] In one implementation of this application, the shape of the candidate plastic is determined based on the distribution information of the candidate plastics. Candidate plastics with the same shape are grouped into the same set, and the plastic shape distribution information is determined based on the quantity of candidate plastics in each set. The plastic shape distribution information is compared with the plastic shape distribution information in a preset shape distribution table to obtain the air classifier control coefficient corresponding to the reference plastic shape distribution information with the highest similarity. The preset shape distribution table includes multiple reference plastic shape distribution information and the air classifier control coefficients corresponding to each of the multiple reference plastic shape distribution information.

[0038] Specifically, based on the obtained distribution information of the candidate plastics, including location, quantity, and area, the shape characteristics of each candidate plastic are further analyzed. This application embodiment uses image processing techniques, such as edge detection, to obtain a precise shape description of each candidate plastic. Candidate plastics with the same shape are grouped into the same set; that is, if two or more candidate plastics have similar or matching shape characteristics, they will be classified into one category. For each shape set, the number of candidate plastics within it is calculated, thereby obtaining the plastic shape distribution information.

[0039] Furthermore, this embodiment of the application includes a preset shape distribution table, which contains multiple reference plastic shape distribution information and their corresponding air classifier control coefficients. The currently obtained plastic shape distribution information is compared with the reference information in the preset table to find the match with the highest similarity. After determining the reference plastic shape distribution information with the highest similarity, the air classifier control coefficient corresponding to that information is selected. This coefficient will be used to adjust the air classifier settings to optimize the separation effect on plastics of specific shapes.

[0040] Step 102: Based on the distribution information of the plastics to be selected, determine the corresponding air classification area, and based on the plastic proportion of each air classification area, determine the air outlet to be adjusted on the air classifier.

[0041] In one implementation of this application, the placement area of ​​each candidate plastic on the conveyor belt is determined based on the distribution information of the candidate plastics. The conveyor belt is divided into multiple air-classifying zones based on the number of air vents corresponding to the air classifier. The plastic percentage corresponding to each candidate plastic is obtained from its respective placement area. The plastic percentage of each air-classifying zone is compared with a preset percentage threshold. If the plastic percentage is greater than the preset percentage threshold, the air vent corresponding to that air-classifying zone is designated as the air vent to be adjusted.

[0042] Specifically, based on the obtained distribution information of the plastics to be selected, the specific placement area of ​​each plastic on the conveyor belt can be determined. The conveyor belt is divided into multiple air-classifying zones according to the number of air vents corresponding to the air classifier. Each air-classifying zone corresponds to one or more air vents, which are responsible for performing air-classification operations on the plastics within that zone.

[0043] Furthermore, for each air-separation zone, the proportion of plastic to be separated within that zone is calculated. In this embodiment, this is achieved by statistically analyzing the quantity or area of ​​plastic within that zone, and then dividing it by the area or quantity of the entire air-separation zone to obtain the plastic proportion. The plastic proportion of each air-separation zone is then compared with a preset proportion threshold. This threshold is preset based on factors such as the performance of the air separator, the physical properties of the plastic, and the expected separation effect. If the plastic proportion of a certain air-separation zone exceeds this threshold, it means that there is too much plastic in that zone, and additional airflow or a more refined air-separation strategy may be needed to ensure the separation effect.

[0044] Furthermore, if the proportion of plastic exceeds a preset threshold, the air separator outlet corresponding to that air separation area is designated as the air outlet to be adjusted. That is, the wind speed, direction, or other parameters of these outlets need to be adjusted to optimize the separation effect of plastic within that area.

[0045] Step 103: Identify the reference air outlets that are related to the air outlet to be adjusted, and construct the parameter control function based on the candidate plastic data corresponding to the reference air outlets.

[0046] In one implementation of this application, a reference air outlet adjacent to the air outlet to be adjusted is determined, and the plastic percentage corresponding to the reference air outlet is obtained. The adjustable plastic percentage is then determined based on the difference between the plastic percentage and a preset percentage threshold. Additionally, the placement area of ​​the candidate plastics corresponding to the reference air outlet is obtained, and the air separation distance between the placement area and the air separation area of ​​the air outlet to be adjusted is determined. Furthermore, based on the placement area of ​​the candidate plastics corresponding to the reference air outlet, the candidate plastics in a stacked state are determined, and plastic stacking distribution data is obtained based on the quantity of the stacked candidate plastics and the area of ​​the stacked area. A parameter control function is constructed based on the adjustable plastic percentage, air separation distance, and plastic stacking distribution data.

[0047] Specifically, after identifying the air outlet to be adjusted, other adjacent air outlets are identified; these are called reference air outlets. Assuming the workload of the reference air outlets does not exceed their limits, for each reference air outlet, its corresponding plastic percentage is obtained—that is, the ratio of the quantity or area of ​​plastic within the air separation area of ​​that air outlet to the total area of ​​that area. The difference between the plastic percentage of the reference air outlet and a preset percentage threshold is calculated. This difference reflects the degree of excess or deficiency of plastic within the air separation area of ​​that air outlet. Based on this difference, the adjustable plastic percentage is determined; that is, without affecting the separation effect of the current area, the amount of plastic from other areas that the air outlet can accommodate is determined.

[0048] Furthermore, the placement areas of the candidate plastics corresponding to the reference air vent are obtained, and the distances between these areas and the air-separation areas of the vent to be adjusted are calculated. This distance, called the air-separation distance, affects the attenuation and diffusion of air force during transmission. For the placement areas corresponding to the reference air vent, candidate plastics in a piled-up state are identified, i.e., plastics piled together due to excessive quantity or irregular shape. The quantity and area of ​​the piled-up plastics are calculated to obtain plastic pile distribution data. This data reflects the distribution of plastics on the conveyor belt. Based on the adjustable plastic ratio, air-separation distance, and plastic pile distribution data, a parameter control function is constructed.

[0049] In one implementation of this application, the constructed wind speed control function is as follows:

[0050]

[0051] The constructed air outlet angle control function is as follows:

[0052]

[0053] Based on the wind speed control function and the air outlet angle control function, a parameter control function is obtained. The reference air outlet is then controlled based on the output value of this parameter control function.

[0054] Where c is the wind speed at the reference air outlet; θ is the wind direction angle at the reference air outlet; p is the adjustable plastic percentage; d is the wind separation distance; s is the plastic accumulation distribution data; v0 is the initial wind speed at the reference air outlet; θ0 is the initial wind direction angle at the reference air outlet; k1 is the first weight; k2 is the second weight; k3 is the third weight; k4 is the fourth weight; k5 is the fifth weight; k6 is the sixth weight; k7 is the seventh weight; k8 is the eighth weight; s d s represents the average bulk density of the candidate plastics in their stacked state; a The area of ​​the accumulation region; This is the first noise term; This is the second noise term.

[0055] In addition to adjustable plastic percentage and air separation distance, this application's embodiments introduce the bulk density and packing area of ​​the plastics to be selected in their stacked state, thus taking into account the material's state more comprehensively. Bulk density s d The larger the value, the denser the packing, requiring higher wind speeds to break up the deposits and ensure effective wind separation; therefore, its coefficient k3 is positive. The area of ​​the packing region s aThe larger the value, the more extensive the airflow adjustment is required; therefore, its coefficient k4 is positive. The impact on large-area accumulation areas must be considered when adjusting wind speed and direction. In the wind speed control function, the distance d is taken as the reciprocal. This is because the closer the reference outlet, the greater its influence on the outlet to be adjusted. Therefore, when adjusting wind speed, the influence of a closer reference outlet is more significant.

[0056] Step 104: Output the wind speed and air outlet angle corresponding to the reference air outlet based on the parameter control function, and control the air outlet to be adjusted based on the wind speed and air outlet angle.

[0057] In one implementation of this application, the coefficient type of the outlet air separator control coefficient is determined; wherein, the coefficient type includes speed type and angle type. Based on the control coefficient corresponding to the speed type, the wind speed control function is adjusted. Based on the control coefficient corresponding to the angle type, the outlet air angle control function is adjusted.

[0058] Specifically, in this application embodiment, the speed type control coefficient refers to a coefficient used to adjust the rotational speed of the air separator. This coefficient allows adjustment of the wind speed control function to achieve precise control of the output wind speed of the air separator; for example, this coefficient can be multiplied by the wind speed control function. Similarly, in this application embodiment, the angle type control coefficient refers to a coefficient used to adjust the angle of the air separator. This coefficient allows adjustment of the angle control function to achieve precise control of the output angle of the air separator; for example, this coefficient can be multiplied by the angle control function.

[0059] Step 105: Based on the adjusted air classifier, perform air classification on the plastic to be selected, and perform plastic composition detection on the air-classified material, so as to perform secondary adjustment of the air classifier based on the detection results.

[0060] In one implementation of this application, a random inspection is performed on the impurity collection area after air classification to obtain the plastic content based on the inspection results. If the plastic content does not meet the preset content conditions, the plastic sample image in the impurity collection area is input into a preset plastic classification model to output the plastic category corresponding to the impurity collection area based on the preset plastic classification model. The quantity of plastics corresponding to different plastic categories in the impurity area is determined to identify the plastic category to be processed based on the quantity. The air outlet control data corresponding to the plastic category to be processed is determined from a preset plastic information database to perform secondary control of the air classifier based on the air outlet control data.

[0061] Specifically, in the impurity collection area after air classification, multiple sampling points are randomly selected, and a certain amount of impurity samples are collected from each sampling point to test the plastic content in the impurity samples. Based on the test results, the average plastic content in the impurity samples is calculated, and the average content is compared with the preset plastic content conditions.

[0062] Furthermore, when the plastic content does not meet the preset conditions, images of plastic samples from the impurity collection area are input into a pre-set plastic classification model. This model, based on image recognition technology, can identify different plastic categories within the impurities and count the identified categories to determine the quantity of each category. The training process for this plastic classification model involves using images of different plastic categories as input and the corresponding plastic category as output to train a pre-set neural network model, thus obtaining the pre-set plastic classification model.

[0063] Furthermore, based on the quantity of each type of plastic and the preset processing priority, the categories of plastics to be processed are determined. The corresponding airflow control data for each category is then retrieved from a pre-set plastic information database. This airflow control data includes parameters such as the air separator's rotation speed and damper opening. Based on the retrieved airflow control data, the air separator is then subjected to secondary control. The purpose of this control is to optimize the air separation process and improve the separation efficiency and purity of the plastic categories to be processed.

[0064] In one implementation of this application, the air velocity difference and angle difference are determined based on the current air outlet data and the air outlet control data corresponding to the type of plastic to be processed; wherein, the air outlet control data includes different air outlet velocities and different air outlet angles. Based on the air velocity difference and angle difference, the air outlets of the air classifier are adjusted. After a preset time interval, the impurity collection area is re-inspected. If the plastic content in the sampled area does not meet the preset content condition, a new air outlet velocity and a new air outlet angle are selected from the air outlet control data to perform a secondary adjustment of the air classifier. The new air outlet velocity is greater than the previously selected air outlet velocity, and the new air outlet angle is greater than the previously selected air outlet angle.

[0065] Specifically, the current air velocity and outlet angle of each air outlet of the air separator are recorded. From a pre-set plastic information database, the corresponding air outlet control data for the plastic category to be processed is retrieved. This data includes different air outlet velocities and outlet angles. For each air outlet, the difference between the current air velocity and the desired air velocity is calculated, as well as the difference between the current outlet angle and the desired outlet angle. Based on the calculated velocity and angle differences, the air outlets of the air separator are adjusted.

[0066] Furthermore, during the adjustment process, the operating status of the air classifier and the changes in plastic content in the impurity collection area are continuously monitored. After adjusting the air outlets, the impurity collection area is re-inspected at preset intervals (e.g., 1 hour, 2 hours, etc.). Based on the inspection results, it is determined whether the plastic content in the impurity collection area meets the preset conditions. If the plastic content still does not meet the preset conditions, a new air outlet speed and a new air outlet angle are selected from the air outlet control data. The new air outlet speed should be greater than the previously selected air outlet speed, and the new air outlet angle should also be greater than the previously selected air outlet angle. Based on the new air outlet speed and air outlet angle, the air classifier is adjusted a second time. Through this process, the air outlet settings of the air classifier can be gradually optimized, improving the separation efficiency and purity of plastics.

[0067] The air separation and conveying control method in this application embodiment can also be used to separate different materials based on actual application needs, such as garbage.

[0068] Figure 2 This is a front view of an air-separating conveyor provided in an embodiment of this application. Figure 3 This is a rear view of an air-separation conveying device provided in an embodiment of this application. Figure 4 This is a side view of an air-separating conveyor provided in an embodiment of this application. Figure 2 , Figure 3 as well as Figure 4 As shown, the air-separating conveyor consists of an air-separating channel 101, a support 107, and an air-separating machine housing 110. The air-separating conveyor also includes: a conveyor belt 109, which is evenly divided into multiple air-separating zones along its width and is used to convey materials to be separated; an image monitoring device 102, mounted on the air-separating box 103, used to acquire images of the materials to be separated on the conveyor belt 109; the air-separating box 103, located on the side of the image monitoring device 102 facing the discharge direction of the conveyor belt 109, with the air-separating channel 101 connected to its top and mounting holes 106 at its bottom corresponding to each air-separating zone; and a retractable air vent 105, including a retractable air duct mounted in the mounting holes 106, the upper end of which is connected to the inside of the air-separating box 103; and a drive component 104 capable of moving the air duct up and down.

[0069] Specifically, the air-separating channel 101 on the air-separating conveyor is used to introduce and distribute airflow. The airflow is guided through the air-separating channel 101 into the air-separating box 103 to achieve air-separation of the material on the conveyor belt 109. The bracket 107 is the frame supporting the entire air-separating conveyor structure, ensuring that all components are stably fixed in their proper positions. The air-separator casing 110 is the outer shell surrounding and protecting the internal components of the air-separating conveyor, used to reduce noise and prevent dust leakage. The conveyor belt 109 is the component in the air-separating conveyor used to transport the material to be sorted. It is inclined and positioned below the air-separating conveyor, and is evenly divided into multiple air-separating zones along its width to allow for independent sorting of the material in each zone. An image monitoring device 102, such as a camera, is installed on the air-separating box to acquire images of the material to be sorted on the conveyor belt 109. These images are used for material identification and classification. The image monitoring device 102 is typically connected to a computer control system to automate the aforementioned methods for control and data analysis. The air separator 103 is located on the side of the image monitoring device 102 facing the discharge direction of the conveyor belt 109. The top of the air separator 103 is connected to the air separation channel 101, and the bottom is provided with mounting holes 106 corresponding to each air separation area. These mounting holes 106 are used to install retractable air outlets 105. The retractable air outlet 105 includes an air duct that is retractably installed in the mounting hole 106 and a drive component 104. The upper end of the air duct is connected to the inside of the air separator 103 to provide airflow to the air separation area. The drive component 104 (such as a cylinder, motor, etc.) can drive the air duct to move up and down, thereby adjusting the opening size and position of the air outlet, so as to adjust the airflow intensity according to the properties of the material and the screening requirements.

[0070] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0071] The above descriptions are merely embodiments of this application and are not intended to limit the scope of this application. For those skilled in the art, various modifications and variations can be made to the embodiments of this application. These modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions in the embodiments of this application.

Claims

1. A control method for air-separated conveying, characterized in that, The method includes: The process involves acquiring images of materials to be selected on a conveyor belt, determining the distribution information of plastics to be selected based on the pixel information corresponding to the images, specifically including: determining each pixel in the output tensor corresponding to the images of materials to be selected; determining the material category corresponding to each pixel based on the material category probability corresponding to each pixel; dividing adjacent pixels with the same material category to obtain pixel sets corresponding to different materials to be selected; determining a reference pixel set belonging to the plastic category in the pixel set to obtain the image region of plastics to be selected based on the reference pixel set; marking the image region of plastics to be selected in the images of materials to be selected, and inputting the marked image into a recurrent neural network to determine the conveying order and conveying time correlation of the plastics to be selected on the conveyor belt through the recurrent neural network; and obtaining the distribution information of plastics to be selected based on the conveying order, conveying time correlation, and the area of ​​the image region of plastics to be selected. After determining the distribution information of the plastics to be selected based on the pixel information corresponding to the image of the material to be selected, the method further includes: Based on the distribution information of the candidate plastics, the shape of the candidate plastics is determined; Candidate plastics with the same shape are grouped into the same set to determine the plastic shape distribution information based on the number of candidate plastics in each set; The plastic shape distribution information is compared with the plastic shape distribution information in a preset shape distribution table to obtain the air classifier control coefficient corresponding to the reference plastic shape distribution information with the highest similarity; wherein, the preset shape distribution table includes multiple reference plastic shape distribution information, and also includes multiple air classifier control coefficients corresponding to each of the reference plastic shape distribution information; Based on the distribution information of the plastics to be selected, the corresponding air classification areas are determined. Then, based on the plastic percentage corresponding to each air classification area, the air outlets on the air classifier to be adjusted are determined. Specifically, this includes: determining the placement area of ​​each plastic on the conveyor belt based on the distribution information of the plastics to be selected; dividing the conveyor belt into multiple air classification areas based on the number of air outlets corresponding to the air classifier; obtaining the plastic percentage corresponding to each of the multiple air classification areas based on the placement area corresponding to each plastic; comparing the plastic percentage corresponding to each air classification area with a preset percentage threshold; and if the plastic percentage is greater than the preset percentage threshold, using the air classifier outlet corresponding to that air classification area as the air outlet to be adjusted. The step involves identifying a reference air outlet that is correlated with the air outlet to be adjusted, and constructing a parameter control function based on the candidate plastic data corresponding to the reference air outlet. Specifically, this includes: Identify a reference air outlet adjacent to the air outlet to be adjusted, and obtain the plastic percentage corresponding to the reference air outlet. Based on the difference between the plastic percentage and the preset percentage threshold, determine the adjustable plastic percentage. In addition, the placement area of ​​the plastic to be selected corresponding to the reference air outlet is obtained, so as to determine the air selection distance between the two based on the placement area and the air selection area of ​​the air outlet to be adjusted; Furthermore, based on the placement area of ​​the candidate plastics corresponding to the reference air vent, the candidate plastics in the stacked state are determined, and based on the number of candidate plastics in the stacked state and the area of ​​the stacked area, plastic stacking distribution data is obtained. Based on the adjustable plastic percentage, the air separation distance, and the plastic stacking distribution data, a parameter control function is constructed. The parameter control function, constructed based on the adjustable plastic percentage, the air separation distance, and the plastic packing distribution data, specifically includes: The constructed wind speed control function is as follows: ； The constructed air outlet angle control function is as follows: ； Based on the wind speed control function and the air outlet angle control function, the parameter control function is obtained, and the reference air outlet is controlled based on the output value of the parameter control function. in, c For reference wind speed at the vent; For reference wind direction angle; p The plastic content can be adjusted. d For wind selection distance; s This is data on the distribution of plastic stockpiles; v 0 represents the initial wind speed at the reference air outlet; The initial wind direction angle is used as a reference point; k 1 is the first weight; k 2 is the second weight; k 3 is the third weight; k 4 is the fourth weight; k 5 is the fifth weight; k 6 is the sixth weight; k 7 is the seventh weight; k 8 represents the eighth weight; The average bulk density of the candidate plastics in their stacked state; The area of ​​the accumulation region; This is the first noise term; This is the second noise term; A reference air outlet that is related to the air outlet to be adjusted is identified, and a parameter control function is constructed based on the candidate plastic data corresponding to the reference air outlet. Based on the parameter control function, the wind speed and air outlet angle corresponding to the reference air outlet are output, so as to control the air outlet to be adjusted based on the wind speed and air outlet angle. Based on the adjusted air classifier, the plastic to be selected is air-classified, and the plastic composition of the air-classified material is detected, so as to perform secondary adjustment of the air classifier based on the detection results.

2. The control method for air separation and conveying according to claim 1, characterized in that, After constructing the parameter tuning function, the method further includes: The coefficient type of the control coefficient of the air separator is determined; wherein, the coefficient type includes speed type and angle type; The wind speed control function is adjusted based on the control coefficient corresponding to the speed type. The air outlet angle control function is adjusted based on the control coefficient corresponding to the angle type.

3. The control method for air separation and conveying according to claim 1, characterized in that, The step of performing plastic component testing on the air-classified material, and then making secondary adjustments to the air classifier based on the test results, specifically includes: Random sampling is conducted on the impurity collection area after air separation, and the plastic content is obtained based on the sampling results; If the plastic content does not meet the preset content conditions, the plastic sample image in the impurity collection area is input into the preset plastic classification model, so as to output the plastic category corresponding to the impurity collection area based on the preset plastic classification model; The quantity of plastics corresponding to different plastic categories in the impurity collection area is determined, so as to determine the plastic category to be processed based on the quantity of plastics; The air vent control data corresponding to the category of plastic to be processed is determined from the pre-set plastic information database, so as to perform secondary control on the air separator based on the air vent control data.

4. The control method for air separation and conveying according to claim 3, characterized in that, The secondary control of the air separator based on the air outlet control data specifically includes: Based on the current air outlet data and the air outlet control data corresponding to the type of plastic to be processed, the air velocity difference and angle difference are determined; wherein, the air outlet control data includes different air outlet velocities and different air outlet angles; Based on the wind speed difference and the angle difference, the air outlets of the wind separator are adjusted. After a preset time interval, the impurity collection area is re-inspected. If the plastic content of the sampled area does not meet the preset content condition, a new air outlet speed and a new air outlet angle are selected from the air outlet control data to re-adjust the air separator. Wherein, the new air outlet velocity is greater than the previously selected air outlet velocity, and the new air outlet angle is greater than the previously selected air outlet angle.

5. A control method for air-separated conveying according to any one of claims 1-4, characterized in that, The equipment consists of a conveyor belt, an image monitoring device, an air separator, a retractable air outlet, and a drive unit. A conveyor belt, which is evenly divided into multiple air-separation zones along its width, is used to convey materials to be sorted. An image monitoring device is installed on the air separator box and is used to acquire images of the materials to be selected on the conveyor belt. An air separator is located on the side of the image monitoring device facing the discharge direction of the conveyor belt. The top of the air separator is connected to an air separation channel, and the bottom is provided with mounting holes corresponding to each of the air separation areas. A retractable air outlet includes a retractable air duct installed in the mounting hole, the upper end of the air duct being connected to the inside of the air separator box; The driving component can move the air duct up and down.

Images