A method and system for sorting plastic particles

Abstract

This invention relates to a method and system for sorting plastic granules, relating to the field of plastic granule sorting. The method includes: acquiring spectral information; determining the location of impurities from the spectral information; acquiring an image of the impurities based on their locations; determining the impurity outline from the image and determining the impurity material from the spectral information based on the impurity location; determining a discharge location based on the impurity material and determining the impurity volume based on the impurity outline; determining a discharge distance by comparing the discharge location and the impurity location, and determining the impurity weight by combining the impurity volume and material; determining a sorting airflow force by combining the discharge distance and the impurity weight, and determining a sorting angle based on the impurity location; and generating and sending an impurity sorting command in response to the sorting airflow force and the sorting angle. This application improves the accuracy of plastic granule sorting and enables precise control of the nozzle according to the different shapes of the granules.

Description

Technical Field

[0001] This invention relates to the field of plastic sorting, and in particular to a method and system for sorting plastic particles. Background Technology

[0002] Plastic particle sorting technology is mainly used to separate impurities (such as plastics of different materials, metal scraps, dust, etc.) from target plastic particles in mixed plastic particles, thereby improving the purity of plastic particles.

[0003] Existing plastic particle sorting technologies generally employ a spectral recognition-based sorting method. This method involves collecting the spectral information of plastic particles, comparing it with a preset spectral library to determine the particle material, and then filtering out non-target particles. These non-target particles are then blown away from the discharge port through nozzles, thereby achieving the separation of particles of different materials.

[0004] Spectral identification has difficulty identifying the appearance parameters of particles, and the shapes of particles are relatively complex, which leads to a mismatch between the air force used by the nozzle and the shape of the particles, resulting in low sorting accuracy and large errors. Summary of the Invention

[0005] To improve the accuracy of plastic particle sorting and to precisely control the nozzles according to the different shapes of the particles, this invention provides a method and system for sorting plastic particles.

[0006] In a first aspect, the present invention provides a method for sorting plastic particles, employing the following technical solution:

[0007] A method for sorting plastic granules, comprising:

[0008] Step 100: Collect spectral information;

[0009] Step 101: Determine the location of the impurity from the spectral information;

[0010] Step 102: Acquire an image of the impurity based on its location;

[0011] Step 103: Determine the impurity outline from the impurity image, and determine the impurity material from the spectral information based on the impurity location;

[0012] Step 104: Determine the discharge location based on the material of the impurity, and determine the volume of the impurity based on its outline;

[0013] Step 105: Determine the discharge distance by comparing the discharge location and the impurity location, and determine the impurity weight by combining the impurity volume and impurity material;

[0014] Step 106: Determine the sorting air force based on the discharge distance and the weight of the impurities, and determine the sorting angle based on the location of the impurities;

[0015] Step 107: In response to the sorting wind force and sorting angle, generate and send an impurity sorting command.

[0016] By adopting the above technical solution, spectral information is collected and the location of impurities is determined. Based on the location, images of impurities are collected, the outline and material of impurities are identified, the discharge distance, the weight of impurities, and the sorting wind force and angle are calculated. Finally, sorting instructions are generated and sent, thereby achieving accurate identification and sorting of impurities in plastic particles, reducing the problems of inaccurate spectral identification and positioning, coarse control, and easy missorting and omission, and improving the sorting accuracy of plastic particles.

[0017] Optional, also includes:

[0018] Step 108: Determine the sorting direction based on the discharge distance;

[0019] Step 109: Determine the wind action surface from the impurity profile based on the sorting direction;

[0020] Step 110: Determine the angle of action from the wind action surface;

[0021] Step 111: Determine the correction coefficient based on the stated angle of action;

[0022] Step 112: Update the sorting wind force in response to the correction coefficient.

[0023] By adopting the above technical solution, the sorting direction is determined according to the discharge distance. Based on the sorting direction, the wind force action surface is determined from the impurity profile and the action angle is extracted. The correction coefficient is calculated through the action angle, and finally the sorting wind force is updated based on the correction coefficient. This optimizes the control accuracy of the sorting wind force, reduces the problems of poor wind force adaptability and easy blowing deviation, and further improves the accuracy of plastic particle sorting.

[0024] Optional, also includes:

[0025] Step 113: Determine the angle deviation based on the stated angle of action;

[0026] Step 114: When the angle deviation is greater than the preset fluctuation threshold, determine the candidate action surface by combining the action angle and the preset selection range;

[0027] Step 115: Determine the normal direction based on the candidate action surface;

[0028] Step 116: Determine the mean value of the normal based on the normal direction;

[0029] Step 117: Calculate the difference between the mean normal value and the sorting direction, and define it as the action deviation;

[0030] Step 118: Determine the target action surface according to the aforementioned action deviation;

[0031] Step 119: Determine the position of the mean value from the target action surface based on the mean value of the normal;

[0032] Step 120: Update the sorting angle in response to the mean position.

[0033] By adopting the above technical solution, the angle deviation is determined based on the action angle. When the angle deviation is greater than the preset fluctuation threshold, the candidate action surface is determined and the normal direction is extracted. The action deviation between the mean normal value and the sorting direction is calculated. Finally, the sorting angle is updated based on the action deviation, thereby reducing the sorting offset problem caused by the fluctuation of the action angle, refining the sorting angle control, and improving the stability of plastic particle sorting.

[0034] Optional, also includes:

[0035] Step 121: Determine the effective area based on the candidate effective surface, and determine the impurity centroid based on the impurity profile;

[0036] Step 122: Determine the effective distance by comparing the position of the impurity's centroid and mean value;

[0037] Step 123: Determine the candidate evaluation values ​​by combining the action distance, action area, and action deviation;

[0038] Step 124: Update the target action surface in response to the candidate evaluation value.

[0039] By adopting the above technical solution, the effective area is determined based on the candidate effective surface, the centroid of the impurity is identified according to the impurity outline, the effective distance is determined by comparing the centroid with the mean position, the candidate evaluation value is calculated by combining the effective distance, area and deviation, and finally the target effective surface of the wind force is selected based on the candidate evaluation value, thereby improving the rationality of the target effective surface selection and further improving the accuracy and stability of plastic particle sorting.

[0040] Optionally, it also includes an overlapping sorting method, the overlapping sorting method comprising:

[0041] Step 200: Determine the material deviation based on the spectral information;

[0042] Step 201: When the material deviation is greater than the preset overlap threshold, determine the suspected location based on the material deviation;

[0043] Step 202: Acquire a suspected image based on the suspected location;

[0044] Step 203: Identify overlapping features from the suspected image;

[0045] Step 204: Determine the upper region and the exposed region in response to the overlapping feature;

[0046] Step 205: Update the spectral information based on the upper region and the exposed region, and determine the upper material and the exposed material;

[0047] Step 206: Update the impurity position based on the upper material and the exposed material.

[0048] By adopting the above technical solution, when the spectral information of the particles deviates significantly from the spectral data of each material, it is determined that there is particle overlap. At this time, images are collected based on the suspected location and the overlapping features are identified, thereby dividing the upper layer and the exposed area, updating the spectral information and determining the upper and lower layer materials, thus achieving accurate identification of overlapping particles and improving the accuracy of plastic particle sorting.

[0049] Optionally, the overlapping sorting method further includes:

[0050] Step 207: Determine the upper layer position and the exposed position based on the upper layer material and the exposed material;

[0051] Step 208: When the upper layer position and the exposed position are inconsistent, update the impurity profile according to the upper layer region and determine the stable profile according to the exposed region;

[0052] Step 209: Determine the stable area from the stable profile, and determine the profile spacing by combining the stable profile and the impurity profile;

[0053] Step 210: Determine the stability evaluation value by combining the contour spacing and the stable area;

[0054] Step 211: Determine the stabilizing point of action based on the stability evaluation value, and determine the stabilizing angle based on the stabilizing point of action;

[0055] Step 212: Determine the stable wind force according to the stated stable angle;

[0056] Step 213: In response to the stabilizing point of action and the generation of stabilizing wind, a lower-level stabilization command is sent.

[0057] By adopting the above technical solution, when the discharge ports corresponding to the upper and lower layers of particles are inconsistent, the stable contour exposed by the lower layer particles is extracted, thereby calculating the stable area and contour spacing, and combining the two to select the stable action point. Then, when the nozzle blows the upper layer particles toward the corresponding discharge port, other nozzles blow the lower layer particles to reduce the movement of the lower layer particles with the upper layer particles, thereby achieving precise separation of impurities between the upper and lower layers and improving the accuracy of sorting overlapping plastic particles.

[0058] Optionally, the overlapping sorting method further includes:

[0059] Step 214: When the upper layer position and the exposed position are inconsistent, determine the upper layer distance and the exposed distance based on the upper layer position and the exposed position;

[0060] Step 215: Calculate the sum of the upper layer distance and the exposed distance, and define it as the total travel distance;

[0061] Step 216: Determine the sorting speed based on the total stroke;

[0062] Step 217: Determine the acceleration coefficient according to the sorting speed;

[0063] Step 218: Update the sorting wind force in response to the acceleration coefficient.

[0064] By adopting the above technical solution, when the discharge ports corresponding to the upper and lower layers of particles are inconsistent, the upper layer particles need to be blown away first and then the lower layer particles need to be blown away, which compresses the blowing process of the lower layer particles. At this time, the total stroke required for the upper and lower layers of particles is calculated, and the stroke ratio is allocated according to the distance and an appropriate sorting speed is selected, thereby improving the efficiency and accuracy of overlapping sorting.

[0065] Optionally, a collaborative sorting method may also be included, wherein the collaborative sorting method includes:

[0066] Step 300: Determine the sorting sequence based on the location of the impurities;

[0067] Step 301: Determine the sorting interval based on the sorting sequence, and retrieve the spacing wind force based on the sorting interval;

[0068] Step 302: Determine the radius of influence based on the wind force at the specified spacing;

[0069] Step 303: When the radius of influence is greater than the sorting spacing, determine the spacing ratio based on the radius of influence and the sorting spacing;

[0070] Step 304: Determine the influencing wind force by combining the spacing ratio and spacing wind force, and retrieve the spacing position based on the sorting spacing;

[0071] Step 305: Determine the offset direction based on the spacing position;

[0072] Step 306: Determine the offset distance based on the influencing wind force;

[0073] Step 307: Update the spacing position by combining the offset direction and offset distance.

[0074] By adopting the above technical solution, when the nozzle blows away non-target particles, the wind force can easily affect the surrounding particles. The sorting spacing corresponding to the nearest particle around the non-target particle is extracted, and the offset direction and offset distance of the spacing wind force generated by the nozzle are combined to correct the spacing position of the surrounding particles, thereby reducing the problem of mutual interference in the sorting of multiple impurities, realizing efficient collaborative sorting of multiple impurities, and improving the stability of large-scale sorting.

[0075] Optionally, the collaborative sorting method further includes:

[0076] Step 308: When the offset distance is greater than the preset offset threshold, determine the falling time difference based on the distance position;

[0077] Step 309: Determine the independent moments by combining the falling time difference and the spacing ratio;

[0078] Step 310: Update the sorting sequence based on the independent time points, and determine the delay ratio according to the independent time points;

[0079] Step 311: Update the spacing wind force in response to the delay ratio.

[0080] By adopting the above technical solution, when the surrounding particles are greatly affected, the time difference of impurity falling is calculated, and the independent moment when the surrounding particles are not affected is determined by combining the time difference. Thus, the spacing wind force is corrected according to the independent moment, which further reduces the problem of mutual interference in the sorting of multiple impurities and improves the stability and accuracy of large-scale plastic particle sorting.

[0081] Secondly, this application provides a sorting system for plastic granules, which adopts the following technical solution:

[0082] A plastic pellet sorting system, comprising:

[0083] The acquisition module is used to acquire spectral information and impurity images;

[0084] A memory for storing the program for any of the above-mentioned plastic particle sorting methods;

[0085] The processor is the unit of memory that allows programs to be loaded and executed by the processor.

[0086] By adopting the above technical solution, spectral information is collected and the location of impurities is determined. Based on the location, images of impurities are collected, the outline and material of impurities are identified, the discharge distance, the weight of impurities, and the sorting wind force and angle are calculated. Finally, sorting instructions are generated and sent, thereby achieving accurate identification and sorting of impurities in plastic particles, reducing the problems of inaccurate spectral identification and positioning, coarse control, and easy missorting and omission, and improving the sorting accuracy of plastic particles.

[0087] In summary, this application includes at least one of the following beneficial technical effects:

[0088] 1. Collect spectral information and determine the location of impurities. Based on the location, collect images of impurities, identify the impurity outline and material, calculate the discharge distance, impurity weight, and sorting wind force and angle, and finally generate and send sorting instructions. This enables accurate identification and sorting of impurities in plastic granules, reduces problems such as inaccurate spectral identification and positioning, coarse control, and easy missorting and omission, and improves the sorting accuracy of plastic granules.

[0089] 2. Determine the sorting direction based on the discharge distance, determine the wind force action surface from the impurity profile based on the sorting direction and extract the action angle, calculate the correction coefficient through the action angle, and finally update the sorting wind force based on the correction coefficient, thereby optimizing the control accuracy of the sorting wind force, reducing the problems of poor wind force adaptability and easy blowing deviation, and further improving the accuracy of plastic particle sorting;

[0090] 3. Determine the angle deviation based on the action angle. When the angle deviation is greater than the preset fluctuation threshold, determine the candidate action surface and extract the normal direction. Calculate the action deviation between the mean normal value and the sorting direction. Finally, update the sorting angle based on the action deviation to reduce the sorting offset problem caused by the fluctuation of the action angle, refine the sorting angle control, and improve the stability of plastic particle sorting. Attached Figure Description

[0091] Figure 1 This is a flowchart of a method for sorting plastic granules;

[0092] Figure 2 This is a flowchart of the overlapping sorting method;

[0093] Figure 3 This is a flowchart of the collaborative sorting method. Detailed Implementation

[0094] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0095] This application discloses a method and system for sorting plastic particles.

[0096] Reference Figure 1 A method for sorting plastic granules, comprising:

[0097] Step 100: Collect spectral information.

[0098] Spectral information refers to the characteristic data obtained by non-contact scanning of plastic particles to be sorted using hyperspectral or near-infrared sensors. It is used to distinguish plastics of different materials and impurities. Spectral information includes the absorption peaks, reflectance, characteristic wavelengths, etc. of the particles. The acquisition process is completed continuously in real time by the detection system, and the acquisition frequency is set by the staff according to the conveying speed.

[0099] Step 101: Determine the location of the impurity from the spectral information.

[0100] Impurity location refers to the coordinate information of non-target particles, representing the spatial position of non-target particles in the image. The system identifies non-target material areas by comparing standard spectral library with real-time spectral information, and marks the center coordinates of the area as the impurity location. The method for determining the impurity location is common knowledge known to those in the art.

[0101] Step 102: Acquire an image of the impurity based on its location.

[0102] An impurity image is a high-definition image obtained by focusing on a local area using an industrial camera based on the determined location of the impurity. The system drives the industrial camera to perform regional imaging based on the location of the impurity. The image contains appearance information such as the outline, size, and stacking status of the impurity. The impurity image is selected by the operator based on the actual situation.

[0103] Step 103: Determine the impurity outline from the impurity image, and determine the impurity material from the spectral information based on the impurity location.

[0104] Impurity contours refer to the external boundaries of non-target particles. Impurity contours can be extracted from impurity images using image recognition technology and edge detection algorithms. The methods for identifying impurity contours are common knowledge in the field.

[0105] Impurity material refers to the material category determined by matching the spectral characteristics of the corresponding position of the impurity, such as PET, PP, PE, PVC, etc. The method for identifying impurity material is common knowledge to those in the field.

[0106] Step 104: Determine the discharge location based on the material of the impurity, and determine the volume of the impurity based on its outline.

[0107] The discharge position refers to the rejection channel or drop point preset by the system based on the material of the impurity. Different materials correspond to different discharge positions. The discharge position corresponding to the material of the impurity can be found in the discharge correspondence table. The discharge correspondence table is a data table that records different impurity materials and their corresponding discharge positions.

[0108] Impurity volume refers to the volume of non-target particles. The space occupied by the impurity can be calculated based on the impurity profile as the impurity volume. The method for determining the impurity volume is common knowledge to those skilled in the art.

[0109] Step 105: Determine the discharge distance by comparing the discharge location and the impurity location, and determine the impurity weight by combining the impurity volume and impurity material.

[0110] The discharge distance refers to the path distance from the current position of the impurity to the corresponding discharge port, that is, the distance between the discharge position and the position of the impurity. The method for determining the discharge distance is selected by the staff according to the actual situation.

[0111] Impurity weight refers to the weight of non-target particles. It can be estimated by multiplying the volume of the impurity by the density of the corresponding material. The impurity density corresponding to the impurity material can be found in the density correspondence table, which is a data table that records different impurity materials and their corresponding impurity densities.

[0112] Step 106: Determine the sorting air force based on the discharge distance and the weight of the impurities, and determine the sorting angle based on the location of the impurities.

[0113] Sorting air force refers to the air pressure required to be output by the nozzle. The air force increases with the increase of discharge distance and impurity weight. The sorting air force corresponding to the discharge distance and impurity weight can be matched from the air force correspondence table. The air force correspondence table is a data table that records different discharge distances and impurity weights and their corresponding sorting air forces.

[0114] The sorting angle refers to the pointing angle of the airflow nozzle, that is, the angle at which it points to the location of the impurities. The sorting angle is directly obtained from coordinate calculation.

[0115] Step 107: In response to the sorting wind force and sorting angle, generate and send an impurity sorting command.

[0116] Impurity sorting instructions refer to the control signals sent by the controller to the pneumatic actuator, which include parameters such as sorting wind force and sorting angle. The impurity sorting instructions are generated and sent synchronously by the system in real time. The actuator completes the impurity removal action according to the instructions, realizing precise sorting of single particles.

[0117] The system collects spectral information and determines the location of impurities. Based on this location, it acquires images of the impurities, identifies their outlines and materials, calculates the discharge distance, impurity weight, and sorting wind force and angle, and finally generates and sends sorting instructions. This enables accurate identification and sorting of impurities in plastic granules, reducing problems such as inaccurate spectral identification and positioning, coarse control, and easy missorting and omission, thereby improving the sorting accuracy of plastic granules.

[0118] A method for sorting plastic granules further includes:

[0119] Step 108: Determine the sorting direction based on the discharge distance.

[0120] The sorting direction refers to the direction in which non-target particles head towards the discharge position. It represents the direction of movement of non-target particles that need to be blown away. The method for determining the sorting direction is selected by the staff based on the actual situation.

[0121] Step 109: Determine the wind action surface from the impurity profile based on the sorting direction.

[0122] The wind-driven surface refers to the surface area that drives non-target particles to move along the sorting direction. The highest point can be extracted from the impurity profile, and the back side of the sorting direction located at the highest point is taken as the wind-driven surface.

[0123] Step 110: Determine the angle of action from the wind action surface.

[0124] The action angle refers to the angle between the normal to the wind action surface and the sorting direction. It is used to characterize the degree of contact between the airflow and the surface of impurities. The smaller the action angle, the higher the wind power utilization efficiency. The action angle is obtained directly through contour geometry calculation.

[0125] Step 111: Determine the correction coefficient based on the angle of action.

[0126] The correction factor is an adjustment factor used to compensate for wind power loss; the larger the angle of application, the larger the correction factor. The correction factor corresponding to the angle of application can be found in the correction correspondence table. Then, the product of the original sorting wind force and the correction factor is calculated as the new sorting wind force. The correction correspondence table is a data table that records different angles of application and their corresponding correction factors.

[0127] Step 112: Update the sorting wind force in response to the correction coefficient.

[0128] The sorting direction is determined based on the discharge distance. The wind action surface is determined from the impurity profile based on the sorting direction and the action angle is extracted. The correction coefficient is calculated based on the action angle. Finally, the sorting wind force is updated based on the correction coefficient, thereby optimizing the control accuracy of the sorting wind force, reducing the problems of poor wind force adaptability and easy blowing deviation, and further improving the accuracy of plastic particle sorting.

[0129] A method for sorting plastic granules further includes:

[0130] Step 113: Determine the angle deviation based on the stated angle of action.

[0131] Angle deviation refers to the difference between the maximum and minimum angles of action on the wind-affected surface, reflecting the undulation of the wind-affected surface.

[0132] Step 114: When the angle deviation is greater than the preset fluctuation threshold, the candidate action surface is determined by combining the action angle and the preset selection range.

[0133] The fluctuation threshold is the critical value of the angle deviation used to judge whether the wind-affected surface is too undulating. The fluctuation threshold is selected by the staff according to the actual situation. An angle deviation greater than the fluctuation threshold means that the wind-affected surface is too undulating. At this time, the effect of the nozzle blowing to different positions on non-target particles is different.

[0134] The selection range refers to the angle deviation value used to select the candidate action surface. It can be selected by the staff according to the actual situation. The candidate action surface refers to multiple candidate force-bearing surfaces that have been re-screened on the impurity profile. That is, the continuous area where the angle difference is located within the angle deviation. The selection method of the candidate action surface is selected by the staff according to the actual situation.

[0135] Step 115: Determine the normal direction based on the candidate action surface.

[0136] The normal direction refers to the direction perpendicular to the surface to be acted upon, and is used to determine the optimal incident direction of the airflow. Each point on the surface to be acted upon corresponds to a normal, and the normal direction is the angle value of each normal.

[0137] Step 116: Determine the mean value of the normal based on the normal direction.

[0138] The mean normal value refers to the comprehensive direction obtained by weighted averaging of all normal directions of each candidate surface. The calculation method for the mean normal value is selected by the staff based on the actual situation.

[0139] Step 117: Calculate the difference between the mean normal value and the sorting direction, and define it as the action deviation.

[0140] Action deviation refers to the angular difference between the mean normal and the sorting direction, which characterizes the degree of deviation of the overall force direction. The smaller the action deviation, the higher the sorting stability.

[0141] Step 118: Determine the target action surface according to the action deviation.

[0142] The target action surface refers to the action surface with the highest wind stability, that is, the candidate action surface with the smallest action deviation.

[0143] Step 119: Determine the mean position from the target action surface based on the mean normal value.

[0144] The mean position refers to the center point of force corresponding to the mean normal on the target surface, that is, the position where the normal direction is consistent with the mean normal. The sorting angle is adjusted so that the nozzle is oriented towards the mean position.

[0145] Step 120: Update the sorting angle in response to the mean position.

[0146] The angle deviation is determined based on the action angle. When the angle deviation is greater than the preset fluctuation threshold, the candidate action surface is determined and the normal direction is extracted. The action deviation between the mean normal value and the sorting direction is calculated. Finally, the sorting angle is updated based on the action deviation, thereby reducing the sorting offset problem caused by the fluctuation of the action angle, refining the sorting angle control, and improving the stability of plastic particle sorting.

[0147] A method for sorting plastic granules further includes:

[0148] Step 121: Determine the effective area based on the candidate effective surface, and determine the impurity centroid based on the impurity profile.

[0149] The effective area refers to the projected area of ​​the selected effective surface on the nozzle orientation. The larger the area, the higher the wind power utilization rate. The center of gravity of the impurity refers to the position of the mass center of the impurity obtained through contour geometry calculation. It is used to judge whether the force is balanced. The methods for determining the effective area and the center of gravity of the impurity are selected by the staff according to the actual situation.

[0150] Step 122: Determine the effective distance by comparing the position of the impurity's centroid and mean position.

[0151] The effective distance refers to the spatial distance between the point of action of the wind force and the center of gravity of the impurity. The smaller the distance, the less likely the impurity is to overturn and the more stable the sorting trajectory is.

[0152] Step 123: Determine the candidate evaluation value by combining the action distance, action area, and action deviation.

[0153] The candidate evaluation value refers to the optimal score obtained by comprehensively considering the action distance, action area, and action deviation. The smaller the action distance, the larger the action area, and the smaller the action deviation, the larger the candidate evaluation value, which means that the action surface has a better sorting effect. The candidate action surface with the largest candidate evaluation value is selected as the target action surface. The candidate evaluation values ​​corresponding to the action distance, action area, and action deviation can be found in the evaluation correspondence table. The evaluation correspondence table is a data table that records different action distances, action areas, and action deviations and their corresponding candidate evaluation values.

[0154] Step 124: Update the target action surface in response to the candidate evaluation value.

[0155] The effective area is determined based on the candidate effective surface. The centroid of the impurity is identified based on the impurity profile. The effective distance is determined by comparing the centroid with the mean position. The candidate evaluation value is calculated by combining the effective distance, area and deviation. Finally, the target effective surface for wind action is selected based on the candidate evaluation value, thereby improving the rationality of the target effective surface selection and further improving the accuracy and stability of plastic particle sorting.

[0156] Reference Figure 2 Overlapping sorting methods include:

[0157] Step 200: Determine the material deviation based on the spectral information.

[0158] Material deviation refers to the degree of difference between the spectral characteristics at the same spatial location and the standard material library. The calculation method of material deviation is common knowledge to those in the field.

[0159] Step 201: When the material deviation is greater than the preset overlap threshold, determine the suspected location based on the material deviation.

[0160] The overlap threshold is the critical value for determining the material deviation when different materials overlap. The overlap threshold is selected by the staff according to the actual situation. The suspected location is the suspicious area that is determined to be the overlap of different materials, that is, the location of impurities with material deviation.

[0161] Step 202: Collect suspected images based on the suspected locations.

[0162] Suspected images refer to high-resolution images acquired by magnifying overlapping suspected locations. Suspected images can be acquired using industrial cameras, and the acquisition method for suspected images is selected by the staff based on the actual situation.

[0163] Step 203: Identify overlapping features from the suspected image.

[0164] Overlapping features refer to visual features that indicate particle stacking, such as double-layer contours, light and dark boundaries, and edge misalignment. Overlapping features can be extracted using image recognition technology, and the methods for recognizing overlapping features are common knowledge in the field.

[0165] Step 204: Determine the upper region and the exposed region in response to the overlapping feature.

[0166] The upper region refers to the image area occupied by the uppermost particle in the overlapping particles, and the exposed region refers to the exposed part of the lower particle that is not obscured. The two regions together constitute the complete structure of the overlapping particles. The upper region and the exposed region can be determined by image recognition technology. The method for determining the upper region and the exposed region is common knowledge to those in the art.

[0167] Step 205: Update the spectral information based on the upper region and the exposed region, and determine the upper material and the exposed material.

[0168] The upper material refers to the material result obtained by re-spectrally matching the upper region, and the exposed material refers to the material result obtained by re-spectrally matching the exposed region, thus realizing the layered material identification of overlapping particles.

[0169] Step 206: Update the impurity position based on the upper material and the exposed material.

[0170] When the spectral information of the particles deviates significantly from the spectral data of each material, it is determined that there is particle overlap. At this time, images are acquired based on the suspected location and the overlapping features are identified, thereby dividing the upper layer and the exposed area, updating the spectral information, and determining the upper and lower layer materials, thus achieving accurate identification of overlapping particles and improving the accuracy of plastic particle sorting.

[0171] Overlapping sorting methods also include:

[0172] Step 207: Determine the upper layer position and the exposed position based on the upper layer material and the exposed material.

[0173] The upper position refers to the discharge position corresponding to the upper material, and the exposed position refers to the discharge position corresponding to the exposed material. The upper position and exposed position corresponding to the upper material and exposed material can be found from the discharge correspondence table.

[0174] Step 208: When the upper layer position and the exposed position are inconsistent, update the impurity profile according to the upper layer region and determine the stable profile according to the exposed region.

[0175] The inconsistency between the upper layer position and the exposed position means that the particles in the upper layer and the particles in the lower layer need to be blown to different directions. In this case, the non-target particles in the upper layer are first blown to the upper layer position according to the above steps 104 to 107. The stable profile refers to the boundary of the exposed part of the non-target particles in the lower layer. The stable profile can be determined by image recognition technology. The method for determining the stable profile is common knowledge known to those in the art.

[0176] Step 209: Determine the stable area from the stable profile, and determine the profile spacing by combining the stable profile and the impurity profile.

[0177] The stable area refers to the projected area of ​​different stable profiles in the direction of the nozzle. The larger the stable area, the greater the wind power utilization rate. The profile spacing refers to the distance between the farthest point of the impurity profile in the upper layer on different stable profiles. The larger the profile spacing, the less likely it is to affect the non-target particles in the upper layer. The method for determining the stable area and profile spacing is selected by the staff according to the actual situation.

[0178] Step 210: Determine the stability evaluation value by combining the contour spacing and the stable area.

[0179] The stability rating value is a comprehensive score used to evaluate the stability of lower-level particles under wind action. The larger the profile spacing and stable area, the higher the stability rating value. The stability rating value corresponding to the profile spacing and stable area can be found in the stability correspondence table, which is a data table that records different profile spacings and stable areas and their corresponding stability rating values.

[0180] Step 211: Determine the stabilizing point of action based on the stability evaluation value, and determine the stabilizing angle based on the stabilizing point of action.

[0181] The stabilizing point of action refers to the optimal stress point of the lower particles, which is the exposed area corresponding to the maximum stability evaluation value. The geometric center of the exposed area is then taken as the stabilizing point of action. The stabilizing angle is the angle of the nozzle toward the stabilizing point of action. The methods for determining the stabilizing point of action and the stabilizing angle are selected by the staff according to the actual situation.

[0182] Step 212: Determine the stable wind force according to the stated stable angle.

[0183] Stable wind force refers to the wind force value used when blowing air through nozzles to reduce the movement of lower non-target particles with upper non-target particles. The greater the deviation of the stabilization angle from vertical, the greater the stabilization wind force is used. The stabilization wind force corresponding to the stabilization angle can be found in the stabilization wind force table, which is a data table that records different stabilization angles and their corresponding stabilization wind forces.

[0184] Step 213: In response to the stabilizing point of action and the generation of stabilizing wind, a lower-level stabilization command is sent.

[0185] The lower-level stabilization command refers to the command to control other nozzles to blow air onto the exposed parts of non-target particles in the lower layer according to the stabilizing point of action and stabilizing wind force. The method for generating the lower-level stabilization command is common knowledge to those skilled in the art.

[0186] When the discharge ports corresponding to the upper and lower layers of particles are inconsistent, the stable contour exposed by the lower layer particles is extracted to calculate the stable area and contour spacing. The stable action point is selected by combining the two. Then, when the nozzle blows the upper layer particles toward the corresponding discharge port, other nozzles blow the lower layer particles to reduce the movement of the lower layer particles with the upper layer particles, thereby achieving precise separation of impurities between the upper and lower layers and improving the accuracy of sorting overlapping plastic particles.

[0187] Overlapping sorting methods also include:

[0188] Step 214: When the upper layer position and the exposed position are inconsistent, determine the upper layer distance and the exposed distance based on the upper layer position and the exposed position.

[0189] The upper layer distance refers to the path length of non-target particles in the upper layer to the discharge port, and the exposed distance refers to the path length of non-target particles in the lower layer to the discharge port. The method for determining the upper layer distance and the exposed distance is selected by the staff according to the actual situation.

[0190] Step 215: Calculate the sum of the upper layer distance and the exposed distance, and define it as the total travel distance.

[0191] The total path length refers to the total path length required for two layers of particles to complete the sorting process, and is used to assess the overall sorting time and speed requirements.

[0192] Step 216: Determine the sorting speed based on the total stroke.

[0193] Sorting speed refers to the speed at which non-target particles move under the action of airflow. The longer the stroke, the higher the required sorting speed, in order to ensure that the particles are removed within a limited area. The sorting speed corresponding to the total stroke can be found in the speed correspondence table, which is a data table that records different total strokes and their corresponding sorting speeds.

[0194] Step 217: Determine the acceleration coefficient according to the sorting speed.

[0195] The acceleration coefficient refers to the gain coefficient that increases the sorting wind force according to the sorting speed. The higher the sorting speed, the larger the acceleration coefficient. The acceleration coefficient corresponding to the sorting speed can be found in the acceleration correspondence table. Then, the product of the original sorting wind force and the acceleration coefficient is calculated as the new sorting wind force. The acceleration correspondence table is a data table that records different sorting speeds and their corresponding acceleration coefficients.

[0196] Step 218: Update the sorting wind force in response to the acceleration coefficient.

[0197] When the discharge ports corresponding to the upper and lower layers of particles are not the same, the upper layer particles need to be blown away first and then the lower layer particles need to be blown away. This results in the lower layer particles being compressed during the blowing process. At this time, the total stroke required for the upper and lower layers of particles is calculated, and the stroke ratio is allocated according to the distance and an appropriate sorting speed is selected, thereby improving the efficiency and accuracy of overlapping sorting.

[0198] Reference Figure 3 Collaborative sorting methods include:

[0199] Step 300: Determine the sorting sequence based on the location of the impurities.

[0200] The sorting sequence refers to the order in which multiple non-target particles are sorted sequentially. The sorting sequence is arranged in ascending order based on the distance between the impurity position of the non-target particle and the nozzle position. The method for determining the sorting sequence is selected by the staff based on the actual situation.

[0201] Step 301: Determine the sorting interval based on the sorting sequence, and retrieve the spacing wind force based on the sorting interval.

[0202] The sorting spacing refers to the closest distance between each non-target particle to be sorted and its nearest neighbor. It can be obtained by extracting the particle position after the non-target particle to be sorted from the sorting sequence, calculating the distance between the impurity position and the particle position as the particle spacing, and selecting the smallest particle spacing as the sorting spacing.

[0203] Spacing wind force refers to the sorting wind force used to separate non-target particles. The spacing wind force can be retrieved from the system, and the method for retrieving the spacing wind force is selected by the staff according to the actual situation.

[0204] Step 302: Determine the radius of influence based on the wind force at the specified spacing.

[0205] The radius of influence refers to the radius of the maximum range of influence of the spacing wind force, that is, the farthest distance that can be affected when sorting non-target particles. The larger the spacing wind force, the larger the radius of influence. The radius of influence corresponding to the spacing wind force can be found in the influence correspondence table, which is a data table that records different spacing wind forces and their corresponding radii of influence.

[0206] Step 303: When the radius of influence is greater than the sorting spacing, determine the spacing ratio based on the radius of influence and the sorting spacing.

[0207] An influence radius greater than the sorting spacing means that when sorting non-target particles, the wind force can easily affect adjacent particles, causing the position of adjacent particles to shift. The spacing ratio is a value that shows the influence on adjacent particles. The difference between the influence radius and the sorting spacing can be calculated as the spacing difference value, and the quotient of the spacing difference and the influence radius can be calculated as the spacing ratio.

[0208] Step 304: Combine the spacing ratio and spacing wind force to determine the influencing wind force, and retrieve the spacing position based on the sorting spacing.

[0209] The influence of wind force refers to the actual wind force experienced by adjacent particles. The greater the spacing ratio and the spacing wind force, the greater the influence of wind force. The product of the spacing ratio and the spacing wind force can be calculated as the influence of wind force.

[0210] Spacing position refers to the coordinate information of the nearest adjacent particle to the non-target particle to be sorted. It can be retrieved directly from the system. The method of retrieving the spacing position is selected by the staff according to the actual situation.

[0211] Step 305: Determine the offset direction based on the distance position.

[0212] The offset direction refers to the lateral deviation trend that adjacent particles may exhibit under airflow interference. It is determined by the direction of the wind force and the distance position. The wind force direction corresponding to the distance wind force and the air outlet position of the nozzle can be retrieved, and the direction of the wind force affecting adjacent particles can be determined by combining the distance position as the offset direction.

[0213] Step 306: Determine the offset distance based on the influencing wind force.

[0214] Offset distance refers to the length by which impurities deviate from their ideal trajectory under the influence of interfering wind forces. The greater the influencing wind force, the greater the offset distance. The offset distance corresponding to the influencing wind force can be found in the distance correspondence table, which is a data table that records different influencing wind forces and their corresponding offset distances.

[0215] Step 307: Update the spacing position by combining the offset direction and offset distance.

[0216] When the nozzle blows away non-target particles, the wind force can easily affect the surrounding particles. The sorting spacing corresponding to the nearest particle around the non-target particle is extracted, and the offset direction and offset distance of the spacing wind force generated by the nozzle are combined to correct the spacing position of the surrounding particles, thereby reducing the problem of mutual interference in the sorting of multiple impurities, realizing efficient collaborative sorting of multiple impurities, and improving the stability of large-scale sorting.

[0217] Collaborative sorting methods also include:

[0218] Step 308: When the offset distance is greater than the preset offset threshold, determine the falling time difference based on the distance position.

[0219] The offset threshold is a critical value used to judge particle offset. The offset threshold is selected by the operator based on the actual situation. An offset distance greater than the offset threshold indicates that the particle offset amplitude is too large, which can easily cause the particle to interfere with adjacent particles, resulting in more particles having inaccurate impurity positions. The fall time difference refers to the time difference between the non-target particles to be sorted and adjacent particles entering the air separation stage. The air separation stage is the process of blowing non-target particles towards the corresponding discharge port through nozzles. The system can calculate it based on the distance position and the conveying speed of the particles to the air separation stage. The calculation method of the fall time difference is common knowledge in the field.

[0220] Step 309: Determine the independent moment by combining the falling time difference and the spacing ratio.

[0221] An independent moment refers to the moment when the nozzle blows non-target particles toward the corresponding discharge port without affecting adjacent particles, i.e., the moment when the spacing ratio is equal to 0. The independent duration corresponding to the falling time difference, spacing ratio, and influence radius can be found in the duration correspondence table. The larger the falling time difference and the smaller the spacing ratio and influence radius, the smaller the independent duration. The independent moment is then calculated as the sum of the independent duration and the actual moment, where the actual moment refers to the time scale at this time, which can be retrieved from the system.

[0222] Step 310: Update the sorting time sequence based on the independent time and determine the delay ratio according to the independent time.

[0223] The delay ratio refers to the coefficient that adjusts the wind force according to the independent time and blows non-target particles into the corresponding discharge port in a timely manner. The longer the independent time, the higher the delay ratio. The delay ratio corresponding to the independent time can be found from the delay correspondence table. Then, the product of the original spacing wind force and the delay ratio is calculated as the new spacing wind force. The delay correspondence table is a data table that records different independent times and their corresponding delay ratios.

[0224] Step 311: Update the spacing wind force in response to the delay ratio.

[0225] When the surrounding particles are significantly affected, the time difference of impurity fall is calculated, and the independent moment when the surrounding particles are unaffected is determined by combining the time difference. The spacing wind force is then adjusted according to the independent moment to further reduce the problem of mutual interference between multiple impurities and improve the stability and accuracy of large-scale plastic particle sorting.

[0226] Based on the same inventive concept, embodiments of the present invention provide a plastic particle sorting system, comprising:

[0227] The acquisition module is used to acquire spectral information and impurity images;

[0228] A memory for storing the program for any of the above-mentioned plastic particle sorting methods;

[0229] The processor is the unit of memory that allows programs to be loaded and executed by the processor.

[0230] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0231] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for sorting plastic granules, characterized in that, include: Step 100: Collect spectral information; Step 101: Determine the location of the impurity from the spectral information; Step 102: Acquire an image of the impurity based on its location; Step 103: Determine the impurity outline from the impurity image, and determine the impurity material from the spectral information based on the impurity location; Step 104: Determine the discharge location based on the material of the impurity, and determine the volume of the impurity based on its outline; Step 105: Determine the discharge distance by comparing the discharge location and the impurity location, and determine the impurity weight by combining the impurity volume and impurity material; Step 106: Determine the sorting air force based on the discharge distance and the weight of the impurities, and determine the sorting angle based on the location of the impurities. The sorting angle refers to the pointing angle of the airflow nozzle. Step 107: In response to the sorting wind force and sorting angle, generate and send an impurity sorting command; Also includes: Step 108: Determine the sorting direction based on the discharge distance; Step 109: Determine the wind action surface from the impurity profile based on the sorting direction; Step 110: Determine the angle of action from the wind action surface; Step 111: Determine the correction coefficient based on the stated angle of action; Step 112: Update the sorting wind force in response to the correction coefficient; Also includes: Step 113: Determine the angle deviation based on the stated angle of action; Step 114: When the angle deviation is greater than the preset fluctuation threshold, determine the candidate action surface by combining the action angle and the preset selection range; Step 115: Determine the normal direction based on the candidate action surface; Step 116: Determine the mean value of the normal based on the normal direction; Step 117: Calculate the difference between the mean normal value and the sorting direction, and define it as the action deviation; Step 118: Determine the target action surface according to the aforementioned action deviation; Step 119: Determine the position of the mean value from the target action surface based on the mean value of the normal; Step 120: Update the sorting angle in response to the mean position.

2. The method for sorting plastic granules according to claim 1, characterized in that, Also includes: Step 121: Determine the effective area based on the candidate effective surface, and determine the impurity centroid based on the impurity profile; Step 122: Determine the effective distance by comparing the position of the impurity's centroid and mean value; Step 123: Determine the candidate evaluation values ​​by combining the action distance, action area, and action deviation; Step 124: Update the target action surface in response to the candidate evaluation value.

3. The method for sorting plastic granules according to claim 2, characterized in that, It also includes an overlapping sorting method, which includes: Step 200: Determine the material deviation based on the spectral information; Step 201: When the material deviation is greater than the preset overlap threshold, determine the suspected location based on the material deviation; Step 202: Acquire a suspected image based on the suspected location; Step 203: Identify overlapping features from the suspected image; Step 204: Determine the upper region and the exposed region in response to the overlapping feature; Step 205: Update the spectral information based on the upper region and the exposed region, and determine the upper material and the exposed material; Step 206: Update the impurity position based on the upper material and the exposed material.

4. The method for sorting plastic granules according to claim 3, characterized in that, The overlapping sorting method further includes: Step 207: Determine the upper layer position and the exposed position based on the upper layer material and the exposed material; Step 208: When the upper layer position and the exposed position are inconsistent, update the impurity profile according to the upper layer region and determine the stable profile according to the exposed region; Step 209: Determine the stable area from the stable profile, and determine the profile spacing by combining the stable profile and the impurity profile; Step 210: Determine the stability evaluation value by combining the contour spacing and the stable area; Step 211: Determine the stabilizing point of action based on the stability evaluation value, and determine the stabilizing angle based on the stabilizing point of action; Step 212: Determine the stable wind force according to the stated stable angle; Step 213: In response to the stabilizing point of action and the generation of stabilizing wind, a lower-level stabilization command is sent.

5. The method for sorting plastic granules according to claim 4, characterized in that, The overlapping sorting method further includes: Step 214: When the upper layer position and the exposed position are inconsistent, determine the upper layer distance and the exposed distance based on the upper layer position and the exposed position; Step 215: Calculate the sum of the upper layer distance and the exposed distance, and define it as the total travel distance; Step 216: Determine the sorting speed based on the total stroke; Step 217: Determine the acceleration coefficient according to the sorting speed; Step 218: Update the sorting wind force in response to the acceleration coefficient.

6. The method for sorting plastic granules according to claim 1, characterized in that, It also includes a collaborative sorting method, which includes: Step 300: Determine the sorting sequence based on the location of the impurities; Step 301: Determine the sorting interval based on the sorting sequence, and retrieve the spacing wind force based on the sorting interval; Step 302: Determine the radius of influence based on the wind force at the specified spacing; Step 303: When the radius of influence is greater than the sorting spacing, determine the spacing ratio based on the radius of influence and the sorting spacing; Step 304: Determine the influencing wind force by combining the spacing ratio and spacing wind force, and retrieve the spacing position based on the sorting spacing; Step 305: Determine the offset direction based on the spacing position; Step 306: Determine the offset distance based on the influencing wind force; Step 307: Update the spacing position by combining the offset direction and offset distance.

7. The method for sorting plastic granules according to claim 6, characterized in that, The collaborative sorting method further includes: Step 308: When the offset distance is greater than the preset offset threshold, determine the falling time difference based on the distance position; Step 309: Determine the independent moments by combining the falling time difference and the spacing ratio; Step 310: Update the sorting sequence based on the independent time points, and determine the delay ratio according to the independent time points; Step 311: Update the spacing wind force in response to the delay ratio.

8. A system for sorting plastic granules, characterized in that, include: The acquisition module is used to acquire spectral information and impurity images; A memory for storing a program for a method of sorting plastic particles as described in any one of claims 1 to 7; The processor is the unit of memory that allows programs to be loaded and executed by the processor.

Images