Self-flowing type intelligent photoelectric sorting system

Abstract

The invention provides a self-flowing type intelligent photoelectric sorting system. The self-flowing type intelligent photoelectric sorting system comprises a feeding device, a self-flowing device, a recognition device, a material separation execution device, a material distribution device and an industrial personal computer which are sequentially arranged from upstream to downstream. A self-flowing chute is arranged in the self-flowing device, and a preset angle is formed between a bottom plate of the self-flowing chute and the horizontal plane, so that the accelerated speed for moving along the self-flowing chute is provided by utilizing self gravity component force of materials; the recognition device is connected with the industrial personal computer and used for collecting attribute data of materials flowing to a recognition area in a self-flowing mode and transmitting the attribute data to the industrial personal computer to be processed so as to determine to-be-executed materials and non-target materials. The material separation execution device is used for receiving an instruction of the industrial personal computer to change the motion trail of the to-be-executed material so as to execute a physical separation action on the to-be-executed material; the material distributing device is used for storing the to-be-executed materials and the non-target materials respectively. Mechanical power systems such as a motor and a belt conveyor do not need to be arranged in the self-flowing device, the structure is simple, and energy consumption is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of material sorting, and more particularly, to a self-flow type intelligent optoelectronic sorting system. Background Art

[0002] In the field of material sorting, it involves the precise differentiation of different materials in order to reduce the cost of subsequent material processing processes. Most of the existing equipment for processing materials is that a vibrating feeder evenly conveys the materials to a belt, and then the belt conveyor drives the belt to move so that the materials pass through the recognition area for recognition, and then the materials are sent to the sorting system by the material distribution part. Among them, the belt transportation structure is complex, requiring a belt conveyor, rollers, a deviation correction system, etc., with relatively high design difficulty, large installation and commissioning difficulty, and high cost.

[0003] Therefore, eliminating the complex belt conveyor material transportation system, saving costs, reducing the design and installation and commissioning difficulties, and reducing the application power of the entire sorting system are urgent problems to be solved at present. Summary of the Invention

[0004] In view of this, the present invention proposes a self-flow type intelligent optoelectronic sorting system, aiming to solve the above technical problems existing in the prior art.

[0005] The present invention proposes a self-flow type intelligent optoelectronic sorting system, comprising: a feeding device, a self-flow device, an identification device, a material separation execution device, a material distribution device, and an industrial control computer; wherein,

[0006] The feeding device is arranged upstream of the self-flow device and is used to supply materials to the self-flow device;

[0007] A self-flow chute is arranged in the self-flow device, and the bottom plate of the self-flow chute forms a preset angle with the horizontal plane to utilize the gravity component of the materials themselves to provide an acceleration for moving along the self-flow chute, so that the materials make an accelerated movement in the self-flow chute to flow through the recognition area of the identification device and the material separation execution device;

[0008] The identification device is arranged at the lower end of the self-flow device and is connected to the industrial control computer, and is used to collect the material attribute data of the materials flowing to the recognition area and transmit the attribute data to the industrial control computer for processing to determine the materials to be executed and non-target materials;

[0009] The material separation execution device is arranged at the end of the self-flow device, and the material separation execution device is connected to the industrial control computer, and is used to receive the instruction of the industrial control computer to change the movement trajectory of the materials to be executed, so as to perform a physical separation action on them;

[0010] The material separation device is arranged downstream of the execution device and is used for separately storing the to-be-executed material and the non-target material separated by the material separation execution device.

[0011] Further, in the above-mentioned gravity-flow type intelligent optoelectronic sorting system, the gravity-flow device includes: a gravity-flow chute and a plurality of first support columns; wherein,

[0012] The bottom plate and side walls of the gravity-flow chute form a closed cavity to provide a gravity-flow channel for the material, and both the bottom plate and the side walls are made of smooth materials;

[0013] The first support columns are arranged at the bottom of the gravity-flow chute to support the gravity-flow chute.

[0014] Further, in the above-mentioned gravity-flow type intelligent optoelectronic sorting system, a water injection port is provided at a position near the upper end of the side wall of the gravity-flow chute, and is used for selectively injecting water into the gravity-flow chute according to the properties of the material, so that the material can smoothly slide in the gravity-flow chute.

[0015] Further, in the above-mentioned gravity-flow type intelligent optoelectronic sorting system, the feeding device includes: a feeding chute and a plurality of second support columns; wherein,

[0016] A vibration motor is arranged on one side of the feeding chute and is used for uniformly dispersing the material in the feeding chute;

[0017] Each of the second support columns is arranged at the bottom of the feeding chute to support the feeding chute.

[0018] Further, in the above-mentioned gravity-flow type intelligent optoelectronic sorting system, the identification device includes: a light source system and a sensor unit; wherein,

[0019] The light source system is arranged facing the material in the gravity-flow device and is used for emitting electromagnetic wave signals to the material;

[0020] The sensor unit is connected to the light source system and is used for receiving the electromagnetic wave signals after being affected by the material;

[0021] The industrial control computer is connected to the sensor unit and is used for receiving the electromagnetic wave signals after being affected by the material and analyzing the properties of the material based on this to determine the to-be-executed material and the non-target material.

[0022] Further, in the above-mentioned gravity-flow type intelligent optoelectronic sorting system, a neural network training module is built in the industrial control computer, and it configures a Tanh neural network model, which is used for analyzing and processing according to the data characteristics of different material attribute collections, and setting different output intervals according to different material properties to achieve precise distinction between the to-be-executed material and the non-target material.

[0023] Further, in the above-mentioned self-flowing intelligent optoelectronic sorting system, the Tanh neural network model uses as the activation function, where x represents the material attribute data; tanh(x) represents the value for material classification judgment, and its value range is (-1, 1).

[0024] Further, in the above-mentioned self-flowing intelligent optoelectronic sorting system, the material separation execution device includes: an execution air pipe, a cylinder, a push rod, a kicking plate, and a withdrawal air pipe; where

[0025] The execution air pipe is connected to the interface on the cylinder and is used to fill high-pressure gas into the cylinder.

[0026] A piston is connected inside the cylinder, and one end of the piston outside the cylinder is connected to the first end of the push rod, and is used to push the push rod to move under the action of high-pressure gas.

[0027] The second end of the push rod is connected to the kicking plate and is used to drive the kicking plate to push and strike the material and change its running trajectory.

[0028] The withdrawal air pipe is connected to another interface of the cylinder and is used to fill high-pressure gas into the cylinder so that the piston pulls the push rod and then pulls the kicking plate back to the initial position to wait for the next execution.

[0029] Further, in the above-mentioned self-flowing intelligent optoelectronic sorting system, the material separation execution device includes: a first box body, a solenoid valve group, and a baffle; where

[0030] The solenoid valve group is arranged on the top of the first box body and is used to release the high-pressure gas stored inside the first box body to blow the material to be executed so that it falls into the first bin of the material distribution device.

[0031] The baffle is arranged on the first box body at a position on one side of the solenoid valve group and is used to prevent the material from hitting the first box body.

[0032] Further, in the above-mentioned self-flowing intelligent optoelectronic sorting system, the material distribution device includes: a second box body with an open top; where

[0033] A partition is arranged inside the second box body to divide the space inside the box body into a first bin and a second bin, so as to store the material to be executed and the non-target material respectively.

[0034] In the present invention, a gravity flow device is arranged downstream of the feeding device. The bottom plate of the gravity flow chute forms a preset angle with the horizontal plane, so as to utilize the component of the gravity of the material itself to provide an acceleration for the movement along the gravity flow chute, enabling the material to perform an accelerating movement in the gravity flow chute to flow through the downstream identification device and the material separation execution device. There is no need to set mechanical power systems such as motors and belt conveyors in the gravity flow device, with a simple structure and reduced energy consumption. Further, the identification device is connected to the industrial control computer, and is used to collect the material attribute data flowing by gravity to the identification area, and transmit the attribute data to the industrial control computer for processing to accurately identify the materials to be executed and non-target materials, greatly improving the accuracy of material sorting. Description of the Drawings

[0035] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0036] Figure 1 It is a schematic diagram of the overall structure of the gravity flow type intelligent optoelectronic sorting system provided by an embodiment of the present invention;

[0037] Figure 2 It is a side view of the gravity flow type intelligent optoelectronic sorting system provided by an embodiment of the present invention;

[0038] Figure 3 It is a top view of the gravity flow type intelligent optoelectronic sorting system provided by an embodiment of the present invention;

[0039] Figure 4 It is a schematic side view of the transmission X-ray identification system installed above the gravity flow chute in the implementation of the present invention;

[0040] Figure 5 It is a schematic side view of the ultraviolet light identification system (UV), visible light identification system (VLC) and infrared light identification system (IR) installed above the gravity flow chute in the embodiment of the present invention;

[0041] Figure 6 It is a schematic side view of the ultraviolet light identification system (UV), visible light identification system (VLC) and infrared light identification system (IR) installed below the gravity flow chute in the embodiment of the present invention;

[0042] Figure 7 It is a schematic diagram of the structure of the high-speed jet valve of the execution system in the embodiment of the present invention;

[0043] Figure 8 It is a schematic diagram of the structure of the kick plate of the execution system in the embodiment of the present invention;

[0044] Figure 9This is a schematic structural diagram of the material distribution system in the embodiment of the present invention. Detailed implementation manners

[0045] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art. It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the drawings and in conjunction with the embodiments.

[0046] Refer to Figures 1 to 3 , the self-flow type intelligent optoelectronic sorting system of the embodiment of the present invention includes: a feeding device 300, a self-flow device 400, an identification device 500, a material separation execution device 600, a material distribution device 700, and an industrial control computer; wherein, the feeding device 300 is arranged upstream of the self-flow device 400 and is used to provide materials to the self-flow device 400; a self-flow chute is arranged in the self-flow device 400, and the bottom plate of the self-flow chute forms a preset angle with the horizontal plane to utilize the gravity component of the materials to provide an acceleration for moving along the self-flow chute, so that the materials make an accelerated movement in the self-flow chute to flow through the identification area of the identification device 500 and the material separation execution device 600; the identification device 500 is arranged at the lower end of the self-flow device 400 and is connected to the industrial control computer, and is used to collect the material attribute data of the materials flowing to the identification area and transmit the attribute data to the industrial control computer for processing to determine the materials to be executed and the non-target materials; the material separation execution device 600 is arranged at the end of the self-flow device 400, and the material separation execution device 600 is connected to the industrial control computer, and is used to receive the instruction of the industrial control computer to change the movement trajectory of the materials to be executed, so as to perform a physical separation action on it; the material distribution device 700 is arranged downstream of the execution device and is used to store the materials to be executed and the non-target materials separated by the material separation execution device 600 respectively.

[0047] Specifically, the self-flow device 400 includes: a self-flow chute and a plurality of first support columns; wherein, the bottom plate and the side wall of the self-flow chute form a closed cavity to provide a self-flow channel for the materials, and both the bottom plate and the side wall are made of smooth materials; each of the first support columns is arranged at the bottom of the self-flow chute to support the self-flow chute.

[0048] In practice, the top of the gravity chute can be closed or partially open. The surface of the gravity chute should be smooth and flat. In practice, materials with a low coefficient of friction and wear resistance can be selected for the gravity chute to reduce the friction between the material and the gravity chute. The identification device 500 can be arranged on the top cover of the gravity chute. The upper end of the gravity chute and the lower end of the feeding device 300 can be connected by welding. There can be two first support columns, and the two first support columns can be arranged on both sides of the bottom of the upper end of the gravity chute.

[0049] The bottom plate of the gravity chute forms a preset angle α with the horizontal plane. In practice, this angle can be adjusted up and down according to the material properties and separation requirements. This angle should enable the component of the material's gravity to overcome the friction force and provide a constant acceleration, so that the material moves with a uniform acceleration in the gravity chute.

[0050] Preferably, the bottom plate of the gravity chute forms a preset angle α with the horizontal plane, and the preset angle α is between 20° and 35°, so that the material moves with a uniform acceleration in the gravity chute to flow through the identification area of the identification device 500 and the material separation execution device 600.

[0051] Preferably, a water injection port 401 is provided at a position near the upper end of the side wall of the gravity chute for selectively injecting water into the gravity chute according to the material properties, so that the material can slide smoothly in the gravity chute. Of course, a flushing mechanism can also be arranged in the gravity chute to further improve the fluidity of the material. Appropriately adjusting the water flow size can change the movement state of the material in the water flow, making it easier to distinguish materials with different characteristics, thereby improving the separation accuracy. For example, the flushing mechanism can be a spray pipe with multiple water outlet holes. The spray pipe is arranged parallel to the chute direction of the gravity chute, which can not only provide uniform water flow coverage, but also accurately adjust the water flow rate and velocity according to actual needs, so that the material maintains an ideal flow state under the action of the water flow, and thus significantly improves the separation accuracy.

[0052] In this embodiment, the material mainly refers to minerals, and the feeding device 300 is mainly used for the preliminary screening of minerals, that is, pre-discarding and screening out the waste in the minerals.

[0053] The feeding device 300 includes: a feeding trough and several second support columns; wherein, a vibration motor 301 is arranged on one side of the feeding trough for uniformly dispersing the materials in the feeding trough; each of the second support columns is arranged at the bottom of the feeding trough for supporting the feeding trough.

[0054] The feed chute can be a rectangular chute with an open top. The two side walls of the rectangular chute are inclined inward near the discharge port, forming a constricted structure. In this embodiment, the second support columns can be 4 in number, and each pair of the second support columns is symmetrically arranged on both sides of the feed chute. The vibration motor 301 can be connected to the feed chute through a base. Preferably, a spring is arranged between the bottom of the base 302 and the second support columns to buffer the vibration of the vibration motor 301.

[0055] The identification device 500 is used to collect the attribute data of the materials flowing through the gravity chute and transmit the collected data to the industrial control computer for processing to identify different materials.

[0056] In this embodiment, the identification device 500 includes: a light source system 501 and a sensor unit 502; wherein, the light source system 501 is arranged facing the materials in the gravity device 400 to emit electromagnetic wave signals to the materials; the sensor unit 502 is connected to the light source system 501 to receive the electromagnetic wave signals after being affected by the materials; the industrial control computer is connected to the sensor unit 502 to receive the electromagnetic wave signals after being affected by the materials and analyze the properties of the materials based on this to determine the materials to be executed and non-target materials. In this embodiment, the materials to be executed are the target materials.

[0057] Continue to refer to Figure 2 , the specific position of the identification device 500 is at the lower end of the gravity device 400, close to the material separation execution device 600. Among them, the light source system 501 can be at the upper part of the lower end of the gravity device 400. At the same time, according to different material properties, the light source system 501 can also be selected to be arranged at the lower part of the lower end of the gravity device 400, such as Figure 2 the position of the sensor system shown. The same as the light source system 501, according to different forms of electromagnetic waves excited by the materials, the sensor unit 502 can also be installed at the position of the light source system 501 and the position of the sensor unit 502. That is, the installation positions of the light source system 501 and the sensor unit 502 can be selected to be installed at the upper or lower part of the lower end of the gravity device 400 according to different material properties, and can be on the same side or on different sides.

[0058] Refer to Figure 9 , in this embodiment, the material distribution device 700 includes: a second box body 704 with an open top; wherein, a partition 702 is arranged inside the second box body 704 to divide the space inside the box body into a first bin 703 and a second bin 701, so as to store the materials to be executed and non-target materials respectively.

[0059] Specifically, the second box body 704 can be an open rectangular container, with a partition 702 arranged in the middle, dividing the interior of the second box body 704 into two bins. The first bin 703 stores the materials to be processed, and the second bin 701 stores the non-target materials; or the first bin 703 stores the non-target materials and the second bin 701 stores the materials to be processed, which is not limited in this embodiment.

[0060] The specific process of the self-flowing intelligent optoelectronic sorting system of the embodiment of the present invention for sorting materials can be divided into a familiar material sorting process and an unfamiliar material sorting process according to the presence or absence of material information in the database.

[0061] The light source system 501 is a transmission X-ray, a fluorescent X-ray, an ultraviolet light, a visible light or an infrared light. The ultraviolet light recognition system (UV), the visible light recognition system (VLC) and the infrared light recognition system (IR) mainly use the visible fluorescence excitation technology or the spectral absorption technology to distinguish the types of materials.

[0062] Correspondingly, the sensor system is an X-ray detector, a photomultiplier tube (BMT), a high-speed industrial camera (CCD) or a spectrometer (Spectroscope).

[0063] Combined Figure 2 、 Figure 4 and Figure 5 , according to different material properties and recognition principles, the light source system 501 and the sensor unit 502 can be arranged on the same side or different sides of the self-flow system. When on the same side, they can be set together on the top cover side of the self-flow chute or on the bottom side of the self-flow chute. When the light source system 501 and the sensor unit 502 are arranged on the same side, the sensor unit 502 is connected to the side wall of the self-flow chute, and a light source bracket 5031 is arranged on one side of the sensor unit 502, and the light source system 501 is arranged on the light source bracket.

[0064] Refer to Figure 4 , in a specific implementation manner of this embodiment, the light source system 501 selects a transmission X-ray light source system 5011, and the sensor unit 502 selects a dual-energy transmission X-ray detector. The sensor corresponding to the transmission X-ray recognition system (XRT) is a dual-energy transmission X-ray sensor 5012. The transmission X-ray recognition system (XRT) is erected in the recognition area of the self-flow device 400, the transmission X-ray light source system 5011 is erected above the cavity of the self-flow chute, the X-ray light source vertically emits fan-shaped X-rays to the self-flow chute, and the dual-energy transmission X-ray sensor 5012 is erected below the self-flow chute to receive the X-rays emitted by the X-ray light source. When the material passes through the X-ray scanning area, the X-ray signal received by the dual-energy transmission X-ray sensor changes, and the industrial control computer analyzes the composition of the material according to the changed X-ray signal.

[0065] Due to the different X-ray absorption capabilities of materials, the intensity of X-rays received by the dual-energy transmission X-ray sensor changes. According to the Lambert-Beer law I = I 0 e -μx Calculate the intensity of low-energy X-rays I L and the intensity of high-energy X-rays I H , where I is the intensity of X-rays after passing through the material, I 0 is the initial intensity of the incident X-rays, μ is the linear absorption coefficient of the material for X-rays, which is related to factors such as the type of material and the energy of the X-rays, x is the distance that the X-rays propagate in the material, that is, the thickness of the material; μx represents the attenuation degree generated when the X-rays pass through the material. The relevant constant related to the material composition is obtained by the ratio method In the formula, the R value is only related to the nature of the material itself, I L0 and I H0 are the incident intensities of low-energy X-rays and high-energy X-rays respectively, I L and I H are the transmitted intensities of low-energy X-rays and high-energy X-rays after passing through the material respectively, μ L and μ H are the absorption coefficients of the material for low-energy X-rays and high-energy X-rays respectively. When the material constant is R1, the material is marked as material a; when the material constant is R2, the material is marked as material b; when the material constant is R3, the material is marked as material c, and so on. Materials with different constants are marked with different material names, and at the same time, the material names that need to be executed are set in the control software for the execution system to execute conveniently.

[0066] Refer to Figure 5 , in another specific implementation manner of this embodiment, the light source system 501 selects the fluorescent X-ray light source system 5012, and the sensor unit 502 selects the fluorescent X-ray detector 5022. The sensor corresponding to the fluorescent X-ray light source system 5012 (XRF) is the fluorescent X-ray detector 5022, and the fluorescent X-ray light source is erected on the upper side of the self-flow chute cavity by the light source support 5031 on one side of the fluorescent X-ray detector 5022. The fluorescent X-ray light source system 5012 uses characteristic X-ray technology to identify the type of material and distinguishes materials by the different characteristic X-rays excited by different materials.

[0067] Refer to Figure 6, in another specific implementation manner of this embodiment, the identification device 500 can also select an ultraviolet light identification system (UV), a visible light identification system (VLC), or an infrared light identification system (IR), mainly using visible fluorescence excitation technology or spectral absorption technology to distinguish the types of materials. The specific operation is as follows: Select a suitable excitation light source and sensor according to the material properties. For example, spodumene will show a red fluorescence reaction under ultraviolet light irradiation at 365 nm. Use a high-speed industrial camera as the receiving sensor, and analyze whether the captured material shows a red part in the picture taken by the camera to determine whether this material is spodumene. Similarly, the visible light identification system (VLC) and the infrared light identification system (IR) can be selected according to the corresponding material properties. The light sources (5022) of the three systems and the corresponding sensors (5012) are supported by brackets (5031) on the same side of the self-flow system cavity, and the specific position can be either above or below the self-flow trough.

[0068] The industrial control computer is built-in with a data analysis and processing module, a neural network module, and a control module. The data analysis and processing module is connected to the sensor unit 502 in the identification device 500, receives the data collected by the sensor unit 502, and processes the data. The neural network model uses these data as input data for training and then outputs a prediction result. The control module controls the actions of the material separation execution device 600 according to the prediction result.

[0069] The neural network training module in this embodiment configures a Tanh neural network model, which is used to analyze and process the data characteristics of different material attribute data collected, and sets different output intervals according to different material properties to achieve precise distinction between the materials to be executed and non-target materials.

[0070] Further, the Tanh neural network model uses as the activation function; x represents the material attribute data, such as the R value, gray value, brightness, RGB color component value, or photon number in dual-energy XRT, etc.; tanh(x) represents the value for material classification judgment, and its value range is (-1, 1). The output value of the Tanh neural network model is mapped in the interval (-1, 1) and is symmetric about the center of 0. When the identification system collects material information, the material attribute data is used as x and input into the Tanh neural network model, and the output value (ranging from -1 to 1) is used as the basis for material classification judgment. For example, if the material information collected in the transmission X-ray light source system is the R value, the R value is used as x and input into the activation function, and the output real number is used as the basis for material classification judgment.

[0071] Specifically, according to different material properties, aluminum alloy, copper alloy, stainless steel, non-ferrous metal ores and their tailings, coal ores, gangue, etc. are selected for signal acquisition, analysis and training. When sorting different materials, different output intervals of the Tanh neural network model can be selected according to the material properties for distinction. For example, when sorting materials, it is set that the output interval is (-1, 0) for material 1 and the output interval is (0, 1) for material 2, or it is set that the output interval is (-1, -0.5) for material 1 and the output interval is (-0.5, 1) for material 2. Taking non-ferrous metal ores as an example, the output interval of (-1, -0.5) is the concentrate rich in metals, and the output interval of (-0.5, 1) is the tailings with less metal content. The demarcation of the interval can be flexibly changed according to the material properties. It can be seen that the material moves to the recognition area of the recognition device 500 through the gravity flow device 400 and is scanned by electromagnetic waves. When different materials pass through, the electromagnetic wave signals received by the sensor change differently. Through software design, neural network technology and machine learning technology are used to analyze the changes in electromagnetic waves to achieve the purpose of identifying different materials.

[0072] Obviously, it can be concluded from the above that in the gravity-flow type intelligent optoelectronic sorting system provided in this embodiment, by arranging a gravity flow device 400 downstream of the feeding device 300, the bottom plate of the gravity flow chute forms a preset angle with the horizontal plane to utilize the gravity component of the material itself to provide an acceleration for moving along the gravity flow chute, so that the material moves in a uniformly accelerated motion in the gravity flow chute to flow through the downstream recognition device 500 and the material separation execution device 600. There is no need to set mechanical power systems such as motors and belt conveyors in the gravity flow device 400, and the structure is simple and the energy consumption is reduced; further, the recognition device 500 is connected to the industrial control computer, and is used to collect the material attribute data flowing by gravity to the recognition area and transmit the attribute data to the industrial control computer for processing to accurately identify the materials to be executed and non-target materials, greatly improving the accuracy of material sorting.

[0073] Refer to Figure 7, in a specific implementation manner of this embodiment, the material separation execution device 600 includes: an execution air pipe 605, a cylinder 6010, a push rod 607, a kicking plate 608, and a retracting air pipe 606; wherein, the execution air pipe 605 is connected to the interface on the cylinder and is used to fill high-pressure gas into the interior of the cylinder 6010; a piston is connected inside the cylinder 6010, and one end of the piston outside the cylinder 6010 is connected to the first end of the push rod 607 and is used to push the push rod 607 to move under the action of high-pressure gas; the second end of the push rod 607 is connected to the kicking plate 608 and is used to drive the kicking plate 608 to strike the material and change its running trajectory; the retracting air pipe 606 is connected to another interface of the cylinder 6010 and is used to fill high-pressure gas into the cylinder to enable the piston to pull the push rod 607 and then pull the kicking plate 608 back to the initial position to wait for the next execution.

[0074] Specifically, the execution air pipe 605, the cylinder 6010, the push rod 607, the kicking plate 608, and the retracting air pipe 606 are all arranged on the base 609. When the material to be executed passes through the area of the kicking plate 608, high-pressure gas is filled into the piston through the execution air pipe 605, so that the push rod 607 is pushed out, driving the kicking plate 608 to strike the material to be executed, changing the running trajectory of the material, and enabling the material to enter the execution bin. After the kicking plate 608 strikes the material, high-pressure gas is filled through the retracting air pipe 606, so that the push rod 607 drives the kicking plate 608 back to the initial position to wait for the next execution. The industrial control computer controls the start of the execution air pipe 605 when detecting that the target material reaches the specified position, and controls the start of the retracting air pipe 606 after the kicking plate 608 completes the task.

[0075] During specific implementation, the kicking plate 608 covers the lower part of the gravity flow device 400, so that the materials at various positions of the gravity flow device 400 can all be executed. The identification device 500 transmits the position information of the material to be executed to the kicking plate 608. When the material to be executed passes through the area of the kicking plate 608, high-pressure gas corresponding to the position is filled into the execution piston through the execution air pipe 605, so that the execution push rod 607 is pushed out, driving the kicking plate 608 to strike the material to be executed, changing the running trajectory of the material, and enabling the material to enter the first bin 703. After the kicking plate 608 strikes the material, high-pressure gas is filled through the retracting air pipe 606, so that the execution push rod 607 drives the kicking plate 608 back to the initial position to wait for the next execution. The materials that do not need to be executed identified by the identification device 500 enter the second bin 701. Different execution methods can be selected according to different material properties, such as the kicking plate 608 ( Figure 7 ), high-speed jet valve group ( Figure 8 ), etc. The execution methods are flexible and changeable, and can more accurately sort out the materials.

[0076] It should be noted that in practice, multiple sets of material separation execution devices 600 can be set up to separate the material to be executed and the non-target materials. The multiple sets of material separation execution devices 600 are connected to the industrial control computer to perform corresponding execution or non-execution operations on the material to be executed and the non-target materials respectively under the instructions of the industrial control computer. The kick plates 608 in the multiple sets of material separation execution devices 600 can be arranged in parallel to form a kick plate 608 assembly, and the width of the kick plate 608 assembly should adapt to the width of the gravity chute channel and the particle size of the materials.

[0077] Refer to Figure 8 , in another specific implementation manner of this embodiment, the material separation execution device 600 includes: a first box body 603, a solenoid valve group 601, and a baffle 602; wherein, the solenoid valve group 601 is arranged on the top of the first box body 603 and is used to release the high-pressure gas stored inside the first box body 603 to blow the material to be executed so that it falls into the first bin 703 of the material distribution device 700; the baffle 602 is arranged at a position on the first box body 603 on one side of the solenoid valve group 601 and is used to prevent the material from hitting the first box body 603.

[0078] Specifically, a fixed bracket 604 can be arranged at the bottom of the first box body 603. The solenoid valve group 601 can be composed of densely arranged high-speed jet solenoid valves. The solenoid valve group 601 is connected to the industrial control computer so that the corresponding solenoid valve performs corresponding blowing operations on the material to be executed after being opened under the instructions of the industrial control computer. When the material to be executed passes above the high-speed jet valve group, the position information of the material recognized by the recognition system is transmitted to the execution system, and the high-speed solenoid valve at the corresponding position is opened to release the high-pressure gas stored inside, blowing the material to be executed and changing the movement trajectory of the material so that the material falls into the execution bin. When the material that does not need to be executed passes by, the solenoid valve is not opened, the movement trajectory of the material does not change, and the material finally moves to the non-execution bin.

[0079] It should be noted that the width of the solenoid valve group should adapt to the width of the gravity chute channel and the particle size of the materials.

[0080] During specific implementation, when the solenoid valve in the solenoid valve group is opened, the high-pressure gas stored inside is released. The high-pressure gas blows the material that has moved above the jet valve, changing the movement trajectory of the material so that the material moves to the first bin 703. The material baffle prevents the material from hitting the high-speed jet valve box body and causing damage to the solenoid valve. The specific execution process of the jet valve group is as follows: The identification device 500 transmits the position information of the material to be executed identified to the material separation execution device 600. When the material passes above the jet valve group, the high-speed jet valve at the corresponding position is opened, and the solenoid valve releases the high-pressure gas stored inside. The high-pressure gas blows the material to be executed, changing the movement trajectory of the material so that the material falls into the first bin 703. For the material that is identified by the identification device 500 as not to be executed, the solenoid valve is not opened, the movement trajectory of the material remains unchanged, and the material finally moves to the second bin 701.

[0081] For familiar materials, the sorting system of the present invention can perform sorting according to the following process: (1) Set the material discrimination curve: Based on the existing information in the database, a specific discrimination curve is set for the material. When the system receives the material data, the data processing and analysis module of the industrial control computer will first query the database to find if there is matching known material information, which may include the discrimination curve set for the material before. Once the relevant material information is obtained from the database, the data processing and analysis module will set or adjust the current material discrimination criteria based on this information; (2) Turn on the gravity-fed intelligent optoelectronic sorting system, feed the material into the sorting system for sorting and discrimination: After starting the system, the sensor unit of the identification device collects the change in the electromagnetic wave signal generated when the material passes through and transmits it to the data processing and analysis module in real time; The data processing and analysis module generates corresponding control instructions based on the classification result and sends them to the control and coordination module to guide the execution system to perform the separation operation.

[0082] For unfamiliar materials, the sorting can be carried out according to the following process:

[0083] (1) Manually select and distinguish materials: Manually select and distinguish the unknown material type, and distinguish the unknown material according to the artificially defined discrimination criteria; (2) Turn on the neural network learning module of the gravity-fed intelligent optoelectronic sorting system to learn: Put in the artificially defined materials to make the neural network model start to learn features from the new material samples (send the unfamiliar material into the system, and each sensor of the identification device collects relevant data. The neural network model automatically learns the data features of these new materials by adjusting weights and optimizing parameters); (3) Record the material information into the database to become familiar materials: After the neural network learning is completed, the newly obtained material feature information is saved to the database and becomes the reference basis for material sorting; (4) Set the material discrimination curve: Just like familiar materials, set an appropriate discrimination curve for the new material; (5) Turn on the gravity-fed optoelectronic sorting system and, based on the updated database, achieve effective sorting of all materials including unfamiliar materials.

[0084] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims

1. A self-flowing intelligent photoelectric sorting system, characterized in that: include: Feeding device, gravity device, identification device, material separation execution device, material distribution device and industrial computer; Among them, The feeding device is arranged upstream of the gravity flow device and is used to supply materials to the gravity flow device; The gravity flow device is provided with a gravity chute, the bottom plate of the gravity chute is at a preset angle with the horizontal plane, so as to utilize the gravity component of the material itself to provide acceleration along the gravity chute, so that the material moves in the gravity chute with acceleration to flow through the identification area of ​​the identification device and the material separation execution device; The identification device is arranged at the lower end of the gravity flow device and is connected to the industrial computer, and is used to collect the attribute data of the material flowing to the identification area by gravity flow, and transmit the attribute data to the industrial computer for processing to determine the materials to be executed and the non-target materials; The material separation execution device is arranged at the end of the gravity flow device, and the material separation execution device is connected to the industrial computer, and is used to receive instructions from the industrial computer to change the movement trajectory of the material to be executed, so as to perform a physical separation action on it; The material separation device is arranged downstream of the execution device, and is used to separately store the to-be-executed material and the non-target material separated by the material separation execution device.

2. The self-flowing intelligent photoelectric sorting system according to claim 1 is characterized in that: The gravity flow device comprises: a gravity flow chute and a plurality of first support columns; wherein, The bottom plate and side walls of the gravity chute form a closed cavity to provide a gravity channel for the material, and the bottom plate and side walls are made of smooth material; Each of the first support columns is arranged at the bottom of the gravity chute to support the gravity chute.

3. The self-flowing intelligent photoelectric sorting system according to claim 1 is characterized in that: A water injection port is provided near the upper end of the side wall of the gravity chute for selectively injecting water into the gravity chute according to the properties of the material, so that the material can slide smoothly in the gravity chute.

4. The self-flowing intelligent photoelectric sorting system according to claim 1 is characterized in that: The feeding device comprises: a feeding trough and a plurality of second supporting columns; wherein, A vibration motor is provided on one side of the feed trough for evenly dispersing the material in the feed trough; Each of the second support columns is arranged at the bottom of the feed trough to support the feed trough.

5. The self-flowing intelligent photoelectric sorting system according to claim 1 is characterized in that: The identification device comprises: a light source system and a sensor unit; wherein, The light source system is arranged toward the material in the gravity flow device to emit electromagnetic wave signals to the material; The sensor unit is connected to the light source system to receive the electromagnetic wave signal after being acted upon by the material; The industrial computer is connected to the sensor unit to receive the electromagnetic wave signal after being acted upon by the material and analyze the properties of the material accordingly to determine the material to be executed and the non-target material.

6. The self-flowing intelligent photoelectric sorting system according to claim 1 is characterized in that: The industrial computer has a built-in neural network training module, which is configured with a Tanh neural network model for analyzing and processing the collected data characteristics of different material attributes, and setting different output intervals according to the properties of different materials to achieve accurate distinction between the materials to be executed and non-target materials.

7. The self-flowing intelligent photoelectric sorting system according to claim 6 is characterized in that: The Tanh neural network model uses As an activation function, x represents the material attribute data; tanh(x) represents the numerical value of the material classification judgment, and its value range is (-1, 1).

8. The self-flowing intelligent photoelectric sorting system according to claim 1 is characterized in that: The material separation execution device comprises: an execution air pipe, a cylinder, a push rod, a kick plate and a withdrawal air pipe; wherein, The actuating gas pipe is connected to the interface on the cylinder and is used to fill the cylinder with high-pressure gas; The cylinder is internally connected with a piston, and one end of the piston located outside the cylinder is connected to the first end of the push rod, so as to push the push rod to move under the action of high-pressure gas; The second end of the push rod is connected to the kick plate, and is used to drive the kick plate to push and hit the material and change its running trajectory; The withdrawal air pipe is connected to another interface of the cylinder and is used to fill the cylinder with high-pressure gas so that the piston pulls the push rod and then pulls the kick plate back to the initial position to wait for the next execution.

9. The self-flowing intelligent photoelectric sorting system according to claim 1 is characterized in that: The material separation execution device comprises: a first box body, a solenoid valve group and a baffle; wherein, The solenoid valve group is arranged on the top of the first box body, and is used to release the high-pressure gas stored in the first box body to spray the material to be processed so that it falls into the first material bin of the material distribution device; The baffle is arranged on the first box body at a position on one side of the solenoid valve group, and is used to prevent materials from damaging the first box body.

10. The self-flowing intelligent photoelectric sorting system according to claim 1, characterized in that: The material distribution device comprises: a second box body with an open top; wherein A partition is provided inside the second box body to separate the space inside the box body into a first silo and a second silo, so as to store the materials to be executed and the non-target materials respectively.

Images