AU2020427333A1 - Intelligent photoelectric sorting machine and product separation method thereof - Google Patents
Intelligent photoelectric sorting machine and product separation method thereof Download PDFInfo
- Publication number
- AU2020427333A1 AU2020427333A1 AU2020427333A AU2020427333A AU2020427333A1 AU 2020427333 A1 AU2020427333 A1 AU 2020427333A1 AU 2020427333 A AU2020427333 A AU 2020427333A AU 2020427333 A AU2020427333 A AU 2020427333A AU 2020427333 A1 AU2020427333 A1 AU 2020427333A1
- Authority
- AU
- Australia
- Prior art keywords
- separated
- ray
- products
- label
- probabilities
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000926 separation method Methods 0.000 title claims abstract description 38
- 230000007246 mechanism Effects 0.000 claims abstract description 72
- 230000005540 biological transmission Effects 0.000 claims abstract description 9
- 239000000047 product Substances 0.000 claims description 194
- 238000001228 spectrum Methods 0.000 claims description 28
- 238000001514 detection method Methods 0.000 claims description 16
- 230000007480 spreading Effects 0.000 claims description 13
- 239000000126 substance Substances 0.000 claims description 10
- 238000002189 fluorescence spectrum Methods 0.000 claims description 9
- 238000000034 method Methods 0.000 claims description 7
- 238000004846 x-ray emission Methods 0.000 claims description 7
- 238000004140 cleaning Methods 0.000 claims description 6
- 239000000428 dust Substances 0.000 claims description 6
- 239000013589 supplement Substances 0.000 claims description 3
- 230000002776 aggregation Effects 0.000 abstract 6
- 238000004220 aggregation Methods 0.000 abstract 6
- 239000003245 coal Substances 0.000 description 10
- 238000004876 x-ray fluorescence Methods 0.000 description 6
- 239000000463 material Substances 0.000 description 5
- 230000002411 adverse Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000001420 photoelectron spectroscopy Methods 0.000 description 2
- 238000004833 X-ray photoelectron spectroscopy Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 239000004566 building material Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 230000005693 optoelectronics Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
Landscapes
- Analysing Materials By The Use Of Radiation (AREA)
- Sorting Of Articles (AREA)
- Separation, Sorting, Adjustment, Or Bending Of Sheets To Be Conveyed (AREA)
Abstract
Disclosed are an intelligent photoelectric sorting machine and a product separation method thereof. The intelligent photoelectric sorting machine comprises a belt transmission mechanism (4), a discrimination apparatus, and an aggregation mechanism (9). The belt transmission mechanism (4) is configured to convey multiple products to be sorted; the discrimination apparatus is configured to discriminate said multiple products and define same one by one with different label numbers; the aggregation mechanism (9) comprises multiple aggregation channels. Each aggregation channel is electrically connected to the discrimination apparatus, and is configured to identify the label number corresponding to the aggregation channel and control a product to be sorted corresponding to the label number to enter the aggregation channel.
Description
This application relates to the field of dry separation technology, for example, to an intelligent photoelectric separator and a product separation method.
Intelligent dry separation technology has been widely used in many fields such as coal, minerals, food, garbage, building materials, etc. The separating method includes two processes: identification and separation. These processes are performed by spreading materials on a moving distributing device, identifying each material by an identification mechanism in a manner of image, color, X-ray transmission, X-ray-excited fluorescence and the like; and after the identifying is completed, picking the materials, or changing their falling trajectories by being hit by an elastic component or ejected by an instantaneous high-pressure air during falling of the materials from a tail end of the distributing device to realize separation of the materials.
In the related art, using intelligent dry separation, generally on only products of two categories can be separated. For products of multiple categories, due to that the identification mechanism cannot effectively define the products after they are identified, which adversely affects subsequent separation of the products of multiple categories.
In addition, in an X-ray separator, detection is performed based on the detection principle of transmitting both high-energy X-rays and low-energy X-rays. Due to that X-rays with continuous energy spectrum are used, the high-energy X-rays and the low-energy X-rays cannot be distinguished well, and the adverse effects caused by overlapping between the high-energy X rays and the low-energy X-rays is serious. Besides, the thicknesses of to-be-detected substances are not uniform. Although, based on the two-dimensional information of high-energy and low energy, part of the adverse effects of the non-uniform thickness on the to-be-detected substances are eliminated by algorithms, the detection error caused by the non-uniform thicknesses of the substances is still very large, especially the substances with similar atomic numbers cannot be effectively identified.
An intelligent photoelectric separator and a product separation method are provided according to the present application.
An intelligent photoelectric separator, includes a belt conveying mechanism, an identification device and a collection mechanism. The belt conveying mechanism is configured to convey to be-separated products of multiple categories; the identification device is configured to identify the to-be-separated products of multiple categories and define the to-be-separated products of multiple categories with different labels one by one; and, the collection mechanism includes multiple collection channels, the number of the multiple collection channels is equal to the number of categories of the products of multiple categories, each of the collection channels is electrically connected to the identification device, and each of the collection channels is configured to identify a label corresponding to a respective collection channel, and control to-be separated products corresponding to the label corresponding to the respective collecting channel to enter the respective collection channel.
A product separation method for an intelligent photoelectric separator, includes the following steps:
defining, in a descending order of a proportion of products of each category respectively in products of N categories, a first label to an Nth label for the Products of N categories sequentially, where N is greater than 1;
!0 setting, starting from a nearest end of the belt conveying mechanism, each of the N collection channels to respectively correspond to the first label to the Nth label insequence;
controlling a to-be-separated product to pass through a spreading device and to enter an identification device;
controlling an X-ray line detection sensor to obtain an equivalent energy attenuation rate of X ray with continuous energy spectrum of the to-be-separated product, and obtaining first probabilities that the to-be-separated product corresponds to different labels based on the equivalent energy attenuation rate of X-ray with continuous energy spectrum;
controlling an X-ray-excited fluorescence receiver to obtain fluorescence spectrum information of the to-be-separated product, and obtaining second probabilities that the to-be-separated product corresponds to different labels based on the fluorescence spectrum information; controlling an image identification system to obtain image information of the to-be-separated product, and obtaining third probabilities that the to-be-separated product corresponds to different labels based on the image information; performing a weighted sum on the first probabilities, the second probabilities and the third probabilities of the to-be-separated product corresponding to a same label, to obtain probabilities of the to-be-separated product corresponding to each of the labels; determining a label corresponding to the to-be-separated product based on a set probability predefined threshold; transmitting the determined label to the N collection channels, and comparing the determined label with labels corresponding to the N collection channels, to determine a collection channel corresponding to the to-be-separated product; and controlling the collection channel corresponding to the to-be-separated product to transmit a signal to an electromagnetic valve based on a label corresponding to the collecting channel; and controlling an air nozzle by the electromagnetic valve to eject the to-be-separated product to enter the collection channel.
FIG. 1 is a schematic structural diagram of an intelligent photoelectric separator according to the present application; and
FIG. 2 is a flowchart of a separation method for an intelligent photoelectric separator according to the present application.
Reference list
1. Cleaning mechanism; 2. X-Ray system; 21. X-Ray emission mechanism; 22. X-ray line detection sensor; 23 X-ray-excited fluorescence receiver; 3 Image identification system; 4 Belt conveying mechanism;
5 Feeding mechanism; 6 Spreading device; 7 Air tank; 8 Air nozzle; and 9 Collection mechanism.
The technical solutions of the present application are described hereinafter with reference to the drawings and embodiments. The embodiments described herein are only intended to explain the present application rather than limiting the present application. For the convenience of description, only a part rather than the whole related to the present application is shown in the drawings.
In the description of this application, unless otherwise specified and defined, the terms "installation", "connecting" and "connection" should be understood in a broad sense, for example, it may be a fixed connection or a detachable connection; it may be a mechanical connection or an electrical connection; it may be a direct connection or an indirect connection through an intermediate medium, or an internal communication between two components. The meanings of the above terms in the present application can be understood according to the situations.
As shown in FIG. 1, an intelligent photoelectric separator is provided according to the present application, which includes a belt conveying mechanism 4, an identification device and a collection mechanism 9. The belt conveying mechanism 4 is configured to convey to-be-separated products of multiple categories; the identification device is configured to identify to-be-separated products of multiple categories and define the to-be-separated products of multiple categories with different labels one by one. The collection mechanism 9 includes multiple collection channels, and the number of the multiple collection channels is the same as the number of categories of the to-be-separated products of multiple categories, and each of the collection channels is electrically connected to the identification device, and is configured to identify a label corresponding to a respective collection channel, and to control to-be-separated products corresponding to the label corresponding to the respective collecting channel to enter the respective collection channel.
In an embodiment, the number of the multiple collection channels is greater than or equal to the number of predefined product categories.
In an embodiment, the separator further includes a spreading device 6, and the spreading device 6 is arranged below the belt conveying mechanism 4, and the spreading device 6 drives a conveyor belt of the belt conveying mechanism 4 to vibrate, to spread the to-be-separated products on the conveyor belt to prevent heaping up of the to-be-separated products which may result in inaccurate subsequent identification and separation. In this embodiment, the spreading device 6 is a vibrator, which can make the conveyor belt vibrate. In other embodiments, it may also be other forms of vibration mechanisms, as long as it can make the conveyor belt vibrate to spread the to-be-separated products on the conveyor belt.
In an embodiment, above an end of the belt conveying mechanism 4 where the product falls, a cleaning mechanism 1 is provided, and the cleaning mechanism 1 includes a dust hood and a vacuum cleaner connected to the dust hood, the vacuum cleaner can absorb, through the dust hood, the dust brought in the process of product separation, and prevent the generated dust from polluting the air.
In an embodiment, the separator further includes a feeding mechanism 5, the feeding mechanism 5 is arranged above the belt conveying mechanism 4, and the to-be-separated products that are delivered slip onto the belt conveying mechanism 4 through the feeding mechanism 5, the feeding mechanism 5 makes the to-be-separated products slip smoothly onto the conveyor belt, so as to prevent the to-be-separated products from sticking/overlapping on the conveyor belt which may result in the inability to effectively identify and separate the products subsequently. The feeding mechanism 5 in this embodiment is a vibrating feeder, and in other embodiments, it can also be a chute or a sliding plate, etc., which is not limited here.
In an embodiment, the identification device includes an X-ray system 2 and/or an image identification system 3. In this embodiment, the identification device includes an X-ray system 2 and an image identification system 3. The X-ray system 2 can perform ray identification on the to-be-separated product based on the photoelectric properties of the to-be-separated product, and the image identification system 3 can perform image identification on the to-be-separated product based on the image property of the to-be-separated product. In an embodiment, the X-ray system 2 uses pixels of the captured image of the product to be equivalent to the atomic number of the product. When X-rays transmit an object with a low tissue density, the amount of the X-rays absorbed is less, and the remaining amount of the X-rays is more, so the resulting picture is brighter, while when X-rays transmit an object with a high tissue density, the amount of the X rays absorbed is more, and the remaining amount of the X-rays is less, so the image formed is darker, and eventually an image with black-white contrast and light-shade difference will be formed on the screen. Different atomic numbers within a unit volume represent different products.
When the X-ray system 2 and the image identification system 3 are used together, the category of the product can be more accurately confirmed, where the X-ray system 2 includes an X-ray emission mechanism 21, an X-ray-excited fluorescence receiver 23 and an X-ray line detection sensor 22. The X-ray emission mechanism 21 emits X-rays with continuous energy spectrum which are received by the X-ray line detection sensor 22 after being transmitted through the to be-separated product, and are received by the X-ray-excited fluorescence receiver 23 after being reflected by the to-be-separated product. The image identification system 3 obtains image information of the to-be-separated product, and the image information of the to-be-separated product, together with the above X-ray photoelectron spectroscopy and X-ray fluorescence spectroscopy, is used to obtain category information of the to-be-separated product in accordance with a weighted voting mechanism, to determine the category of the product, and to define products of multiple categories with different labels one by one, and the labels are transmitted to the collection channels. In an embodiment, the X-ray source emits X-rays with continuous energy spectrum, and the X-rays with continuous energy spectrum are transmitted through the to-be separated product. Due to different densities of the products, the X-rays received by the X-ray line detection sensor 22 are also different, and the equivalent energy attenuation rate of X-ray with continuous energy spectrum of the to-be-separated product can be calculated and obtained through the received X-rays, to determine the probability that which label the product corresponds to; the X-ray source emits X-rays with continuous energy spectrum, and the X-rays with continuous energy spectrum, after being reflected by the to-be-separated product, excites a fluorescence spectrum which is received by the X-ray-excited fluorescence receiver 23, and the probability that which label the product corresponds to is determined based on the received fluorescence spectrum.
In other embodiments, for scenarios where the accuracy requirement imposed on the identification is not high, the identification device may include only the X-ray system 2, and the X-ray system 2 performs the ray identification on the to-be-separated products based on the optoelectronic properties of the to-be-separated products, and defines products of multiple categories with different labels one by one, and transmits the labels to the collection channels.
In an embodiment, the belt conveying mechanism 4 includes a driving part and a driving roller and a driven roller. The driving roller and a driven roller are spaced apart and arranged side by side. The conveyor belt is wound on the driving roller and the driven roller, and the driving roller and the driven roller are connected through the conveyor belt. An output shaft of the driving part is in transmission connection with the driving roller, the output shaft of the driving part can be in transmission connection with the driving roller. A belt, a coupling or a gear assembly can be used to enable the driving part to be in transmission connection, to drive the driving roller to rotate, so as to drive the conveyor belt to convey the to-be-separated products. In this embodiment, the driving part is a motor, such as a speed-adjustable servo motor, stepper motor or reducing motor. The driving part can control the rotation speed of the conveyor belt, so as to control the conveying speed of the products, and avoid interference between two adjacent products when they are falling due to a too fast speed.
In an embodiment, a separation assembly includes an air nozzle 8, an air tank 7 and an air compressor. The air nozzle 8 is arranged below and/or obliquely above the belt conveying mechanism 4, the position for arrangement and the quantity can be set based on practical requirements of the site. The air tank 7 is connected to the air nozzle 8, and can provide high pressure air for the air nozzle 8. The air nozzle 8 is located at an end where the product falls. The air nozzle 8 is provided with an electromagnetic valve, and the electromagnetic valve is electrically connected to the collection channels. When a product with a label the same as the label corresponding to a collection channel is delivered, the collection channel controls the electromagnetic valve to control the air nozzle 8 to eject, to enable the product to enter the !0 corresponding collection channel. The air compressor is connected to the air tank 7, and the air compressor can supplement high-pressure air for the air tank 7. A pressure sensor is arranged in the air tank 7, and whether to supplement the high-pressure air to the air tank 7 is determined based on a detection value of the pressure sensor. In this embodiment, two rows of air nozzles 8 are provided, air nozzles in one row are large nozzles, and air nozzles in the other row are small nozzles. The air ejected by the large nozzles is strong, and can eject a product for a long distance. The air ejected by the small nozzles is not so strong, and only can eject the product for a shorter distance. A product that is not ejected moves forward for a shortest distance under the action of an inertial force applied by the belt conveying mechanism 4.
In an embodiment, since a product depositary has a fixed location, and must store products of specific categories. Proportions of products of multiple categories in the to-be-separated products are different in different batches of products, and in order to save energy and reduce the error of separation, generally the products with a largest proportion fall into a collection channel close to the belt conveying mechanism 4, and the products with a smallest proportion are ejected into a collection channel farthest from the belt conveying mechanism 4, and the products with a middle proportion are ejected into a middle collection channel. The products in each of the collection channels are conveyed to a corresponding product depositary by the conveying mechanism. In this embodiment, the products of the first label enter the collection channel close to the belt conveying mechanism 4; the products of the second label are driven by the small nozzles to enter the middle collection channel, and the products of the third label are driven by the large nozzles to enter the collection channel farthest from the belt conveying mechanism 4. In order to ensure that the products of the first label are the one with the largest proportion in all the products, it is necessary to define the products with the largest proportion to have thefirst label before separation.
In the intelligent photoelectric separator according to the present application, when the belt conveying mechanism 4 thereof conveys the to-be-separated products to a lower side of the identification device, the identification device identifies the to-be-separated products, and defines the to-be-separated products of multiple categories one by one with different labels. Each of the collection channels is electrically connected to the identification device, and can identify the label corresponding to the respective collection channel, and control the air nozzle 8 of the separation assembly to eject the products with the label corresponding to the respective collection channel to the respective collection channel. Different labels are used to define the products, which facilitates the identification of products of multiple categories and enables the products to enter the corresponding collection channels for separation.
As shown in FIG. 2, a product separation method for an intelligent photoelectric separator is further provided according to this embodiment, which is applicable to the above-described intelligent photoelectric separator, and includes the following steps: S10 to SI10.
In S10, to-be-separated products are defined to correspond to a first label, a second label to an Nth label in sequence.
In S20, starting from a nearest end of the belt conveying mechanism, the collection channels are marked sequentially in a descending order of the numbers of the labels.
In S30, the to-be-separated products enter the identification device after passing the spreading device 6.
In S40, the X-ray line detection sensor 22 obtains an equivalent energy attenuation rate of X-ray with continuous energy spectrum of each of the to-be-separated products, and determines probabilities that each of the to-be-separated products corresponds to the labels defined above.
In S50, the image identification system 3 obtains image information of each of the to-be-separated products, and determines probabilities that each of the to-be-separated products corresponds to the labels defined above.
In S60, the X-ray-excited fluorescence receiver 23 obtains fluorescence spectrum information of each of the to-be-separated products, and determines probabilities each of the to-be-separated products corresponds to the labels defined above.
In S70, the probabilities of the same label obtained in the above steps S40 to S60 are weighted and summed to obtain the probabilities of each of the to-be-separated products corresponding to each of the labels.
In S80, it is determined which label each of the to-be-separated products corresponds to, based on predefined probability thresholds corresponding to the to-be-separated products.
In S90, a signal for the determined label is sent to the collection channels, and is compared with the labels marked for the collection channels.
In S100, a collection channel transmits a signal to a separation mechanism based on the label corresponding to the collection channel.
In S110, the separation mechanism controls the electromagnetic valve to control the air nozzle 8 to eject each of the to-be-separated products into a corresponding channel respectively.
In an embodiment, the product separation method for the intelligent photoelectric separator includes: defining, in a descending order of predefined proportions of products of each category of N categories, a first label to an Nth label for the products of N categories sequentially, where N is greater than 1; starting from the nearest end of the belt conveying mechanism 4, setting N collection channels corresponding to the first label to the Nth label insequence; controlling the to be-separated products to pass the spreading device 6 and then enter the identification device; !5 controlling the X-ray line detection sensor 22 to obtain an equivalent energy attenuation rate of X-ray with continuous energy spectrum of the to-be-separated product, and obtaining first probabilities that the to-be-separated product corresponds to different labels based on the equivalent energy attenuation rate of X-ray with continuous energy spectrum; controlling the X ray-excited fluorescence receiver 23 to obtain fluorescence spectrum information of the to-be separated products, and obtaining second probabilities that the to-be-separated product corresponds to different labels based on the fluorescence spectrum information; controlling the image identification system 3 to obtain image information of the to-be-separated products, and obtaining third probabilities that the to-be-separated product corresponds to different labels based on the image information; performing weighted sum on the obtained first probabilities, second probabilities and third probabilities of the to-be-separated products corresponding to the same label, to obtain probabilities of the to-be-separated products corresponding to each of the labels; determining a label corresponding to the to-be-separated product based on a set probability predefined threshold; transmitting a determined label to the N collection channels, and comparing the determined label with the labels corresponding to the N collection channels, to determine a collection channel corresponding to the to-be-separated products; and controlling the collection channel to transmit a signal to the electromagnetic valve based on the label corresponding to the collecting channel; and controlling the air nozzle 8 by the electromagnetic valve to eject the to be-separated products to enter the collection channel.
Optionally, the equivalent energy attenuation rate of X-ray with continuous energy spectrum of the to-be-separated products is denoted as a parameter , and the parameter [is calculated as follows:
[t=(Q1l/[n)x[n/(1+2+...+n)]+[t2/[t(n-1)x[(n !0 1)/(1+2+...+n)]+...+(tn/2)/(ptn/2+1)x[(n/2+1)/(1+2+...+n)];
where, 1 is an energy attenuation rate of rays at an energy level of 1Kev after the rays transmit a substance; 2 is an energy attenuation rate of rays at an energy level of 2Kev after the rays transmit the substance; n is an energy attenuation rate of rays at an energy level of nKev after the rays transmit the substance; n is the maximum energy level of the X-ray system, and n is an even number.
In the related art, high-energy spectrum and low-energy spectrum are used for transmission. According to the present application, a detection algorithm of using full energy spectrum for transmission is adopted, and calculations of energy attenuation rate of rays are performed on rays generated after the X-rays at multiple energy levels are transmitted. The boundaries of the '0 multiple energy levels are clear, and the data dimensions are rich, which addresses the issue that the boundary of high and low energies in the continuum X-ray identification are not clear.
Through multi-dimensional data, the adverse effects of the non-uniform thickness of the to-be detected products on the detection result are effectively eliminated.
In an embodiment, because the products with a large proportion in the to-be-separated products are not always the same every time. For example, in the coal mine separation, when the proportion of coal is greater, the coal is selected while the gangue is discharged, and when the proportion of gangue is greater, the gangue is selected while the coal is discharged. For color separation in the field of grain separation or for solid separation in the field of garbage separation, the categories of to-be-separated products change, and the products that account for a larger proportion in each batch of products change, so it is necessary to define the products with a largest quantity to correspond to the first label, in order to reduce the number of ejection times while separation.
For products having thresholds which are close to each other, wrong determination often occurs, especially in the field of coal separation, the X-ray equivalent energy attenuation rates of gangue containing coal and coal-containing gangue may be close to or even the same as coal or gangue due to different coal or gangue containing rate, and in this case, wrong determination will occur in the separation. In order to avoid such a situation, in the present application, the image identification system and the X-ray fluorescence are used for auxiliary identification, and a voting mechanism is provided. Since X-ray with continuous energy spectrum photoelectron spectroscopy is used to identify the atomic number of the product, it has a higher accuracy than those of the other two identification methods, and therefore has a higher voted weight. for !0 example, the weight of X-ray with continuous energy spectrum photoelectron spectroscopy is set to 0.7, the weight of the image identification system is set to 0.2, and the weight ratio of the X ray fluorescence is set to 0.1. The setting of the weights is not limited to the above-mentioned values, and the weights are set and adjusted based on properties of different to-be-separated products in different fields.
In an embodiment, comparison is made for the X-ray with continuous energy spectrum, the image identification system, and the X-ray fluorescence respectively and separately based on the pre-set thresholds to define label probabilities, and then a vote is made for the label probabilities acquired by the X-ray with continuous energy spectrum, the image identification system, and the X-ray fluorescence based on their weights. For example, it is determined based on the X-ray with continuous energy spectrum that the probability of the product corresponding to the first label is 80%, and the probability of the product corresponding to the second label is 20%, it is determined based on the image identification system that the probability of the product corresponding to the first label is 70%, and the probability of the product corresponding to the second label is 30%, and it is determined based on the X-ray fluorescence that the probability of the product corresponding to the first label is 40%, and the probability of the product corresponding to the second label is 60%, the weight of the X-ray with continuous energy spectrum is set to 0.7, the weight of the image identification system is set to 0.2, the weight of the X-ray fluorescence is set to 0.1, and by performing the weighted voting, the first label gets votes of 0.8x0.7+0.2x0.7+0.1x0.4=0.74, and the second label gets votes of 0.7x0.2+0.2x0.3+0.1x0.6=0.26. The predefined threshold is defined that when the first label gets votes of more than 0.6, it is determined that the products correspond to the first label, thus it is determined based on the above votes that the product corresponds to the first label. If the predefined threshold is defined as that when the first label gets votes exceeding 0.75, it is determined that the products correspond to the first label, for the above-mentioned votes, it is determined that the products correspond to the second label. The predefined threshold is defined based on product properties, and the above thresholds are set based on product properties, and are not limited to the above settings. According to the present application, the three identification methods are used to identify the products separately, and the category of the product isfinally determined by calculation and comparison in accordance with a weighted voting mechanism.
In the present application, X-ray identification, image identification system, and X-ray fluorescence spectroscopy are adopted to assist the identification, and the weighted voting mechanism is adopted. After performing weighted sum, the label with the largest proportion is determined for the products, which significantly improves the separation accuracy, and reduces the mis-separation rate.
Claims (11)
1. An intelligent photoelectric separator, comprising a belt conveying mechanism (4), an identification device and a collection mechanism (9); wherein
the belt conveying mechanism (4) is configured to convey to-be-separated products of a plurality of categories;
the identification device is configured to identify the to-be-separated products of the plurality of categories and define the to-be-separated products of the plurality of categories with different labels one by one; and
the collection mechanism (9) comprises a plurality of collection channels, each of the plurality of collection channels is electrically connected to the identification device, and each of the plurality of collection channels is configured to identify a label corresponding to a respective collection channel, and control to-be-separated products corresponding to the label to enter the respective collection channel.
2. The intelligent photoelectric separator of claim 1, wherein a number of the plurality of collection channels is greater than or equal to a number of predefined product categories.
3. The intelligent photoelectric separator of claim 1, comprising a spreading device (6), wherein the spreading device (6) is arranged below the belt conveying mechanism (4), the spreading device (6) is configured to drive a conveyor belt of the belt conveying mechanism (4) to vibrate, to allow the to-be-separated products of the plurality of categories to spread on the conveyor belt.
4. The intelligent photoelectric separator of claim 1, comprising a cleaning mechanism (1), wherein the cleaning mechanism (1) is arranged above an end where the to-be-separated products of the plurality of categories fall, and the cleaning mechanism (1) is configured to remove dust generated by the to-be-separated products of the plurality of categories.
5. The intelligent photoelectric separator of claim 1, comprising a feeding mechanism (5), !5 wherein the feeding mechanism (5) is arranged above the belt conveying mechanism (4), and the feeding mechanism (5) is configured to deliver the to-be-separated products of the plurality of categories to the belt conveying mechanism (4).
6. The intelligent photoelectric separator of claim 1, wherein the identification device comprises at least one of the following: an X-ray system (2), an image identification system (3); wherein the X-ray system (2) is configured to perform ray identification on the to-be-separated products of the plurality of categories; and the image identification system (3) is configured to perform image identification on the to be-separated products of the plurality of categories.
7. The intelligent photoelectric separator of claim 6, wherein the X-ray system (2) comprises an X-ray emission mechanism (21), an X-ray-excited fluorescence receiver (23) and an X-ray line detection sensor (22); wherein
the X-ray emission mechanism (21) is configured to emit X-rays with continuous energy spectrum;
the X-ray line detection sensor (22) is configured to receive X-rays generated after the X rays with continuous energy spectrum are transmitted through a to-be-separated product of each category, and to determine first probabilities that the to-be-separated product of each category corresponds to different labels based on the received X-rays; and
the X-ray-excited fluorescence receiver (23) is configured to receive X-ray-excited fluorescence generated after the X-rays with continuous energy spectrum are transmitted through the to-be-separated product of each category, and to determine second probabilities that the to be-separated product of each category corresponds to the different labels based on the received X-ray-excited fluorescence;
wherein the image identification system (3) is configured to acquire image information of the to-be-separated product of each category, and to determine third probabilities that the to-be separated product of each category corresponds to the different labels based on the image information; and
wherein the identification device is configured to obtain a label corresponding to the to-be separated product of each category and transmit the label corresponding to the to-be-separated product of each category to the plurality of collection channels, wherein the label corresponding to the to-be-separated product of each category is obtained in accordance with a weighting mechanism based on at least one of the following:
the first probabilities and the second probabilities;
the third probabilities; or,
the first probabilities, the second probabilities and the third probabilities.
8. The intelligent photoelectric separator of claim 3, wherein the belt conveying mechanism (4) comprises a driving part, a driving roller and a driven roller, and the conveying belt is annular, and is sleeved on each of the driving roller and the driven roller, the driving roller is connected to the driven roller through the conveying belt, and the driving part is in transmission connection with the driving roller.
9. The intelligent photoelectric separator of claim 1, comprising a separation assembly, wherein the separation assembly is configured to eject to-be-separated products of each category into a respective collecting channel;
the separation assembly comprises an air nozzle (8), an air tank (7) and an air compressor; and
wherein the air nozzle (8) is arranged at an end of the belt conveying mechanism (4) where the to-be-separated products of the plurality of categories fall, and the air nozzle (8) is provided with an electromagnetic valve, and the electromagnetic valve is electrically connected to the plurality of collection channels;
the air tank (7) is configured to provide high-pressure air for the air nozzle (8); and
the air compressor is configured to supplement high-pressure air for the air tank (7).
10. A product separation method for an intelligent photoelectric separator, applicable to the intelligent photoelectric separator of any one of claims 1 to 9, comprising:
defining, in a descending order of a respective predefined proportion of products of each category among products of N categories, a first label to an Nth label for the products of N categories sequentially, wherein N is greater than 1;
setting, starting from a nearest end of the belt conveying mechanism (4), each of the N collection channels to respectively correspond to the first label to the Nth label insequence;
controlling a to-be-separated product of the to-be-separated products of N categories to pass through a spreading device (6) and to enter an identification device;
controlling an X-ray line detection sensor (22) to obtain an equivalent energy attenuation rate of X-ray with continuous energy spectrum of the to-be-separated product, and obtaining first probabilities that the to-be-separated product correspond to different labels based on the equivalent energy attenuation rate of X-ray with continuous energy spectrum; controlling an X-ray-excited fluorescence receiver (23) to obtain fluorescence spectrum information of the to-be-separated product, and obtaining second probabilities that the to-be separated product corresponds to different labels based on the fluorescence spectrum information; controlling an image identification system (3) to obtain image information of the to-be separated product, and obtaining third probabilities that the to-be-separated product corresponds to different labels based on the image information; performing a weighted sum on the first probabilities, the second probabilities and the third probabilities of the to-be-separated product corresponding to a same label, to obtain probabilities of the to-be-separated product corresponding to each of the labels; determining a label corresponding to the to-be-separated product based on a set probability predefined threshold; transmitting the determined label to the N collection channels, and comparing the determined label with labels corresponding to the N collection channels, to determine a collection channel corresponding to the to-be-separated product; and controlling the collection channel corresponding to the to-be-separated product to transmit a signal to an electromagnetic valve based on a label corresponding to the collecting channel; and controlling an air nozzle (8) by the electromagnetic valve to eject the to-be-separated product to enter the collection channel.
11. The method of claim 10, wherein the equivalent energy attenuation rate of X-ray with continuous energy spectrum is denoted as , and is calculated as follows:
a=(j1/ltn)x[n/(1+2+...+n)]+ 2/(n-1)x[(n 1)/(1+2+...+n)]+...+(an/2)/(jan/2+1)x[(n/2+1)/(1+2+...+n)];
wherein, 1is an energy attenuation rate of rays at an energy level of 1Kev after the rays transmit a substance; 2 is an energy attenuation rate of rays at an energy level of 2Kev after the rays transmit a substance; nis an energy attenuation rate of rays at an energy level of nKev after the rays transmit a substance; n is a maximum energy level of an X-ray system, and n is an even number.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2020/074327 WO2021155506A1 (en) | 2020-02-05 | 2020-02-05 | Intelligent photoelectric sorting machine and product separation method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| AU2020427333A1 true AU2020427333A1 (en) | 2022-09-29 |
| AU2020427333B2 AU2020427333B2 (en) | 2024-05-16 |
Family
ID=77199649
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| AU2020427333A Active AU2020427333B2 (en) | 2020-02-05 | 2020-02-05 | Intelligent photoelectric sorting machine and product separation method thereof |
Country Status (3)
| Country | Link |
|---|---|
| CN (1) | CN115190825A (en) |
| AU (1) | AU2020427333B2 (en) |
| WO (1) | WO2021155506A1 (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112403937A (en) * | 2020-12-04 | 2021-02-26 | 湖州霍里思特智能科技有限公司 | Mineral product sorting machine |
| CN112495832A (en) * | 2020-12-04 | 2021-03-16 | 湖州霍里思特智能科技有限公司 | Mineral product sorting machine and mineral product sorting method |
| CN113926725A (en) * | 2021-10-26 | 2022-01-14 | 东北大学秦皇岛分校 | Coal and gangue rapid sorting device and method based on density estimation |
| CN114405861A (en) * | 2021-12-31 | 2022-04-29 | 安徽中科光电色选机械有限公司 | Shellfish material sorting device |
| CN114602821A (en) * | 2022-03-28 | 2022-06-10 | 同方威视技术股份有限公司 | Ore dressing device |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202411010U (en) * | 2012-01-06 | 2012-09-05 | 肇庆维胜环保资源科技有限公司 | Photoelectric material separator |
| CN104923486A (en) * | 2015-07-06 | 2015-09-23 | 合肥美亚光电技术股份有限公司 | Crawler type X-ray detection device and material sorting device |
| CN106040617A (en) * | 2016-05-29 | 2016-10-26 | 内蒙古科技大学 | Radioactive ore sorting machine |
| CN106733721A (en) * | 2017-02-16 | 2017-05-31 | 天津美腾科技有限公司 | Three product intelligent dry-dressing machines |
| WO2018200866A1 (en) * | 2017-04-26 | 2018-11-01 | UHV Technologies, Inc. | Material sorting using a vision system |
| CN207013301U (en) * | 2017-07-28 | 2018-02-16 | 合肥安杰特光电科技有限公司 | A kind of intelligent opto-electrical separator based on orientating and conveying mechanisms |
-
2020
- 2020-02-05 CN CN202080097700.0A patent/CN115190825A/en active Pending
- 2020-02-05 WO PCT/CN2020/074327 patent/WO2021155506A1/en active Application Filing
- 2020-02-05 AU AU2020427333A patent/AU2020427333B2/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021155506A1 (en) | 2021-08-12 |
| CN115190825A (en) | 2022-10-14 |
| AU2020427333B2 (en) | 2024-05-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2020427333B2 (en) | Intelligent photoelectric sorting machine and product separation method thereof | |
| CN104085667B (en) | The automatic adjustment module of charging and method thereof, device, bulk cargo foreign matter testing agency | |
| US4194634A (en) | Method and apparatus for sorting radioactive material | |
| CA1162884A (en) | Size accommodating, nipped rollers or belt, vertical stream, particle sensor and separator | |
| KR100910963B1 (en) | Apparatus for counting coin using camera and control method thereof | |
| US4122951A (en) | Machine for the automatic detection of blemishes in olives and other fruits | |
| CN114798488B (en) | Material sorting system and sorting method | |
| CN105598026A (en) | Automatic and efficient dry sorting system | |
| CN111617987A (en) | Intelligent automatic sorting system for coal | |
| JP2017080732A (en) | Ore screening method and device | |
| CN110641949A (en) | Equipment and system for monitoring goods jam on conveyor belt | |
| AU2021446611A1 (en) | Material intelligent sorting apparatus and method | |
| CN115945403A (en) | Intelligent ore sorting machine | |
| US20230264230A1 (en) | Efficient material recovery facility | |
| CN212576904U (en) | Intelligent automatic sorting system for coal | |
| CN114101095B (en) | Control method of sorting equipment and sorting equipment | |
| CN203922011U (en) | Charging self-checking device and bulk cargo foreign matter testing agency | |
| CN218743310U (en) | Multi-vision integrated intelligent sorting equipment | |
| JPS5837664A (en) | Sorter | |
| CN111689173B (en) | Feeding control method, device and system for material pile | |
| CN215142214U (en) | Missing detection device for bill | |
| CN114602821A (en) | Ore dressing device | |
| CN216728309U (en) | Sorting machine with multiplied processing capacity | |
| CN112820022B (en) | Anti-stacking device for goods taking bin of vending machine | |
| DE102010014105A1 (en) | Method and device for measuring objects during transport |