Automatic Classification and Recognition System and Method for Regenerated High-Quality Metals Based on LIBS Technology

Abstract

The present invention discloses a system and method for automatic classification and recognition of recycled high-quality metals based on LIBS technology. In this system, a material transmission module conveys the material to be tested along a classification and recognition production line; a pre-treatment device for the material to be tested pre-treats the material to be tested; an information perception system for the material to be tested collects the basic information of the material to be tested; a control system controls a LIBS multi-dimensional high-throughput detection system based on the basic information of the material to be tested; the LIBS multi-dimensional high-throughput detection system performs detection and analysis on the material to be tested based on LIBS technology to obtain the classification and recognition result of the material to be tested; and a post-treatment device for the material classifies and collects different types of materials to be tested. The present invention can perform full-automatic online classification and recognition on recycled high-quality metals of different categories and different forms, greatly improving the classification and recognition efficiency and accuracy of recycled high-quality metals, and enabling the effective recycling and utilization of recycled high-quality metals.

Description

Technical Field

[0001] The present invention relates to the technical field of recycled metal classification and recycling, and particularly to an automatic classification and identification system and method for recycled high-quality metals based on LIBS technology. Background Art

[0002] The recycling and utilization of renewable resources can not only turn waste into treasure, reduce resource waste, but also reduce environmental pollution; especially the recycling and utilization of high-quality and rare metals in recycled metals, such as metal materials like stainless steel, titanium alloy, and superalloy. Due to the scarcity of the metal elements they contain, coupled with complex smelting processes and low product yields, if the recycled high-quality metals can be recycled and utilized, it will greatly reduce resource waste. Therefore, the recycling and utilization of recycled high-quality metals become more important; the main sources of recycled high-quality metals are scraps, defective metal parts, and metal parts that have reached the service life during the metal processing process. Due to the uncontrollable sources, there is a phenomenon of mixed metal grades. If the grades of recycled high-quality metals can be classified before entering the smelting process, it can not only prevent the degradation and utilization of high-value alloys, but also provide important guidance for subsequent smelting processes and reduce energy consumption and pollution during the smelting process.

[0003] Laser-induced breakdown spectroscopy (LIBS) detection technology uses pulsed lasers to perform non-contact high-energy excitation ablation on the material surface. The ablation products generate a large amount of plasma, and spectral signal acquisition and analysis are performed on the plasma signal to finally obtain the composition and content of the object to be measured. When using LIBS technology for testing, there is no need to directly contact the object to be measured, and online dynamic analysis can be realized, which is suitable for online, high-throughput, large-scale, and rapid classification detection of recycled high-quality metal materials. Therefore, how to use LIBS technology to solve the problem of recycling and classification of recycled metals has become a topic worthy of in-depth discussion by those skilled in the art. Summary of the Invention

[0004] The purpose of the present invention is to provide an automatic classification and identification system and method for recycled high-quality metals based on LIBS technology, which can perform full-automatic online classification and identification on recycled high-quality metals of different categories and different forms, greatly improving the classification and identification efficiency and accuracy of recycled high-quality metals, and enabling the effective recycling and utilization of recycled high-quality metals.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] An automatic classification and identification system for recycled high-quality metals based on LIBS technology, comprising: a material transmission module, a pre-treatment device for the material to be measured, an information perception system for the material to be measured, a LIBS multi-dimensional high-throughput detection system, a control system, and a post-treatment device for the material.

[0007] The material transfer module is used to convey the material to be tested along the classification and identification production line; the material to be tested is recycled high-quality metal material;

[0008] The pretreatment device for the material to be tested is used to pretreat the material to be tested so that the form and surface state of the material to be tested have relative consistency;

[0009] The information perception system for the material to be tested is used to collect the basic information of the material to be tested and transmit it to the control system; the basic information includes the presence, size, shape, height, and position information of the material to be tested;

[0010] The control system is used to control the LIBS multi-dimensional high-throughput detection system based on the basic information of the material to be tested;

[0011] The LIBS multi-dimensional high-throughput detection system is used to detect and analyze the material to be tested based on LIBS technology to obtain the classification and identification results of the material to be tested;

[0012] The post-treatment device for the material is used to receive the execution instruction of the control system and perform classification and collection actions on different types of materials to be tested according to the execution instruction; the control system generates the execution instruction according to the classification and identification results of the material to be tested.

[0013] Further, the pretreatment device for the material to be tested includes a shearing device and a impurity removal and cleaning device. The shearing device is used to shear and crush the material to be tested into materials with consistent forms. The forms of the material to be tested include: chip-shaped materials with a shape size of 3-15 mm; sheet-block materials with a shape size of more than 15 mm;

[0014] The impurity removal and cleaning device includes at least one of ultrasonic, high-temperature, and laser ablation devices, and is used to remove impurities and clean the surface of the material to be tested so that the surface state of the material to be tested is consistent.

[0015] Further, the information perception system for the material to be tested includes an adjustable brightness illumination system, a variable focal length optical system, and an adaptive height sensing system. The adjustable brightness illumination system is used to illuminate the surface of the material to be tested so as to distinguish the material to be tested from the conveyor belt of the material transfer module. The variable focal length optical system is used to dynamically adjust the lens focal length to adapt to the size, position change, and detection requirements of the material to be tested, and ensure the imaging quality and detail capture ability; the adaptive height sensing system is used to sense the height fluctuation of the material to be tested in real time and dynamically adjust the sensor pose.

[0016] Further, the LIBS multi-dimensional high-throughput detection system includes multiple groups of single-channel optical systems. The single-channel optical system includes a laser, an excitation optical path, a return optical path, and a spectrometer;

[0017] The LIBS multi-dimensional high-throughput detection system uses laser-induced laser spectroscopy technology and is equipped with a fully automatic analysis program. The fully automatic analysis program automatically calculates the start time and end time of excitation and the working state of the laser in the LIBS multi-dimensional high-throughput detection system according to the material identification status and the area of the horizontal projection.

[0018] The LIBS multi-dimensional high-throughput detection system excites the material to be tested through multiple single-channel optical systems, obtains the atomic emission spectrum information of the material, compares the obtained atomic emission spectrum information of the material with the material model database, and classifies and identifies the material to be tested in combination with the classification algorithm.

[0019] Furthermore, the construction method of the material model database is as follows: The atomic emission spectrum information of multiple regenerated high-quality metals of different grades is used as the training set, the SVM parameters are iteratively optimized by GA, a classification model is trained, and the material model database is constructed based on the classification model.

[0020] Furthermore, the control system controls the start and stop times and the scanning mode of the excitation of the LIBS multi-dimensional high-throughput detection system; the control system displays the classification and identification results on the visualization panel.

[0021] Furthermore, the material post-processing device includes multiple different material channels for collecting different types of materials to be tested.

[0022] The present invention also provides a method for automatically classifying and identifying regenerated high-quality metals based on LIBS technology, which is applied to the above-mentioned system for automatically classifying and identifying regenerated high-quality metals based on LIBS technology, and includes the following steps:

[0023] S1. Excitation area division:

[0024] During the detection process, according to the position of the material to be tested on the material transmission module, the detection area of the material to be tested is divided into a horizontal area and a vertical area. The horizontal area is the transmission direction of the material to be tested, and the detection area is divided into a front single-point scanning excitation area and a rear multi-point scanning excitation area; the vertical area division is perpendicular to the transmission direction of the material to be tested, and the detection area is divided into areas 1, 2, 3,..., N.

[0025] S2. Detection mode confirmation and material spectrum information acquisition:

[0026] According to the detection requirements, the corresponding excitation mode is matched, and the material to be tested is excited to obtain the atomic emission spectrum information of the material to be tested; the excitation modes include single-point excitation mode, single-point and key material accumulation mode, line scan sampling mode, and line scan sampling and area multi-point accumulation mode.

[0027] S3. Data processing of the spectrum information of the material to be tested:

[0028] It is used to detect and analyze the material to be tested based on LIBS technology, obtain the detection information of the material to be tested, and compare it with the model database to obtain the classification and identification results of the material to be tested;

[0029] S4. Output of classification and identification results of materials to be tested:

[0030] The control system displays the classification and identification results on a visual panel; the material post-processing device classifies and collects different types of materials to be tested.

[0031] Furthermore, in the S2, the single-point excitation mode implements single-point excitation for flake-like materials to obtain single-point material detection information; the single-point and key material accumulation mode, based on the single-point excitation mode, performs multi-point cumulative excitation on the materials to be tested with large information differences, so as to achieve accurate classification and identification of the materials to be tested; the line scan sampling mode, for chip-like materials, implements sampling scanning to obtain large-scale regional spectral information; the line scan sampling and regional multi-point accumulation mode, based on the line scan sampling mode, performs small-area multi-point cumulative excitation on areas with large information differences within a large area, so as to achieve accurate classification and identification of chip-like materials.

[0032] Compared with the traditional metal sorting technology in existing technologies, the automatic classification and identification system and method for recycled high-quality metals based on LIBS technology provided by the present invention integrates multiple system modules such as material transmission module, pre-treatment device for materials to be tested, information perception system for materials to be tested, LIBS multi-dimensional high-throughput detection system, control system, material post-processing device, etc., to realize the complete pipeline operation of material transmission, pre-treatment, basic information collection, LIBS technical analysis and classification collection, improve the efficiency of automatic classification and identification of high-quality metal materials, and realize the collection of high-quality metal materials of different categories, facilitating the recycling and reuse of high-quality metals.

[0033] The present invention uses LIBS technology to perform online fully automatic classification and identification of recycled high-quality metals. It does not require complicated manual operations and automatically matches the corresponding excitation mode according to the identification and classification requirements, greatly improving the efficiency and accuracy of classification and identification. It can perform fully automatic online classification and identification of recycled high-quality metals of different categories and forms (chips, flakes), and is suitable for online, high-throughput, large-scale, and rapid classification and detection of recycled high-quality metal materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0035] Figure 1 It is a schematic diagram of the overall structure of the automatic classification and recognition system for recycled high-quality metals based on LIBS technology of the present invention;

[0036] Figure 2 It is a schematic diagram of the conveying direction of the material transmission module of the present invention;

[0037] Figure 3 It is a working flow chart of the automatic classification and recognition system for recycled high-quality metals based on LIBS technology;

[0038] Figure 4 It is a schematic diagram of the optical system of the LIBS multi-dimensional high-throughput detection system of the present invention;

[0039] Figure 5 It is a schematic diagram of the detection method for the automatic classification and recognition of recycled high-quality metals based on LIBS technology of the present invention. (a) represents the division of the excitation area of chip-shaped materials, where A is the pre-positioned single-point scanning excitation area and B is the post-positioned multi-point scanning excitation area; (b) represents the excitation mode of chip-shaped materials. In the figure, it is the line-scan sampling and area multi-point accumulation mode, and the curve represents the excitation scanning path; (c) represents the division of the excitation area of sheet and block materials, where C is the pre-positioned single-point scanning excitation area and D is the post-positioned multi-point scanning excitation area; (d) represents the excitation mode of sheet and block materials. In the figure, it is the single-point and key material accumulation mode, and the dots represent the excitation spots;

[0040] Explanation of reference numerals: 1. Material transmission module; 2. Pretreatment device for the material to be tested; 3. Information perception system for the material to be tested; 4. LIBS multi-dimensional high-throughput detection system; 5. Control system; 6. Post-treatment device for materials; 7. Sheet and block materials; 8. Excitation spots. Detailed implementation manners

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0042] The object of the present invention is to provide a regenerated high-quality metal automatic classification and recognition system and method based on LIBS technology, which integrates six modules including a material transmission module, a pre-treatment device for the material to be tested, an information perception system for the material to be tested, a LIBS multi-dimensional high-throughput detection system, a control system, and a post-treatment device for the material, etc. It can realize the fully automatic on-line classification and recognition of the regenerated metal grades, greatly improve the classification and recognition efficiency and accuracy of the regenerated metal grades, and finally quickly classify and recognize the regenerated high-quality metal grades.

[0043] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] Embodiment 1

[0045] As Figure 1 shown, the regenerated high-quality metal automatic classification and recognition system based on LIBS technology provided by the present invention includes: a material transmission module 1, a pre-treatment device 2 for the material to be tested, an information perception system 3 for the material to be tested, a LIBS multi-dimensional high-throughput detection system 4, a control system 5, and a post-treatment device 6 for the material; the specific introduction is as follows:

[0046] As Figure 2 shown, the material transmission module 1 is used to convey the material to be tested along the classification and recognition pipeline, and convey the material to be tested to the corresponding positions of different systems and modules, so as to perform corresponding operations on the material to be tested; the material to be tested is a regenerated high-quality metal material; the regenerated high-quality metal includes metal and alloy materials with high-quality characteristics, and its sources include but are not limited to processing leftovers, defective metal parts, metal parts that have reached the service life, etc.; for example, the material transmission module 1 includes a conveyor belt and a motor drivingly connected to the conveyor belt, and the pre-treatment device 2 for the material to be tested, the information perception system 3 for the material to be tested, the LIBS multi-dimensional high-throughput detection system 4, the control system 5, and the post-treatment device 6 for the material are arranged in sequence along the conveyor belt, and the motor is electrically connected to the controller of the control system 5.

[0047] The pretreatment device 2 for the material to be measured is used to pretreat the material to be measured, so that the form and surface state of the material to be measured have relative consistency; for example, the pretreatment device 2 for the material to be measured includes a shearing device and an impurity removal and cleaning device, and the impurity removal and cleaning device removes impurities and cleans the surface of the material by means of ultrasonic, high temperature, laser ablation, etc. For example, the shearing device can adopt a pair-roll crusher, and the two rolls rotate relatively and squeeze and crush to shear and crush the material to be measured into materials with consistent forms; in the impurity removal and cleaning device, ultrasonic: an industrial multi-tank ultrasonic cleaning device is adopted, and the multi-tank device realizes the full process automation of degreasing-rinsing-drying; high temperature: a high-temperature water-based solvent is used to clean the oil stains on the surface of the material to be measured; laser ablation (high-temperature plasma-driven cleaning): the surface of the material is excited by high-frequency pulses, and the high energy ablates the surface of the material and decomposes surface pollutants (such as metal oxides and organic matter residues); the shearing device and the impurity removal and cleaning device can be selected from existing devices with the above functions according to actual needs, and no specific limitations are made here.

[0048] The form of the material to be measured includes: chip-shaped materials with a shape size of 3-15 mm; sheet-shaped and block-shaped materials with a shape size of more than 15 mm; the impurity removal and cleaning device includes at least one of ultrasonic, high temperature and laser ablation devices, and is used to remove impurities and clean the surface of the material to be measured, so that the surface state of the material to be measured is consistent.

[0049] The information perception system 3 for the material to be measured is used to collect the basic information of the material to be measured and transmit it to the control system 5; the basic information includes the presence, size, shape, height and position information of the material to be measured; for example, the information perception system 3 for the material to be measured includes an adjustable brightness illumination system, a variable focal length optical system and an adaptive height sensing system. The adjustable brightness illumination system adjusts the brightness of the LED light strip through a built-in adjustment circuit. The adjustable brightness illumination system is used to illuminate the surface of the material to be measured, so as to distinguish the material to be measured from the conveyor belt of the material transmission module and capture the material information more clearly and accurately; the variable focal length optical system is used to dynamically adjust the lens focal length. The variable focal length optical system includes a variable focal length optical lens, and the focal length of the zoom lens is realized according to the clarity of the picture, adapting to the size, position change and detection requirements of the material to be measured, and ensuring the imaging quality and detail capture ability; the adaptive height sensing system uses a 3D line laser measuring instrument to sense the height fluctuation of the material to be measured in real time and dynamically adjust the sensor pose.

[0050] The information perception system 3 for the material to be measured transmits the basic information of the material to be measured perceived through a data communication line to the control system 5, and the control system 5 triggers the full-automatic analysis program of the LIBS multi-dimensional high-throughput detection system 4 to collect high-throughput spectral receipts for the material to be measured.

[0051] The control system 5 is used to control the LIBS multi-dimensional high-throughput detection system 4 based on the basic information of the material to be measured; for example, the control system includes a software operating system, an industrial control computer, a display, a control circuit, etc. The control system 5 realizes the coordinated work between system modules through data transmission and displays the classification and identification information on the visualization panel; the control system 5 is also used to control the start-stop time and scanning mode of the excitation of the LIBS multi-dimensional high-throughput detection system 4.

[0052] The LIBS multi-dimensional high-throughput detection system 4 can adapt to different heights and angles of the material and realizes multi-dimensional high-throughput detection from dimensions such as height, angle, the superposition of height and angle, and multiple excitation modes; it is used to detect and analyze the material to be measured based on LIBS technology to obtain the classification and identification results of the material to be measured; for example, the LIBS multi-dimensional high-throughput detection system 4 includes multiple groups of single-channel optical systems. The single-channel optical system includes a laser, an excitation optical path, a return optical path, and a spectrometer. After receiving the signal from the control system 5, the laser emits light in a set excitation mode. The excitation optical path shapes, reflects, and focuses the laser emitted by the laser on the surface of the material to excite the material; after the material is excited by high-energy laser, it emits atomic emission spectra, which are collected by the return optical path and sent to the spectrometer. The spectrometer will complete the photoelectric signal conversion and complete the acquisition of the material spectral information.

[0053] The LIBS multi-dimensional high-throughput detection system 4 adopts laser-induced laser spectroscopy technology and is equipped with a full-automatic analysis program. The full-automatic analysis program automatically calculates the start time and end time of excitation and the working state of the laser in the LIBS multi-dimensional high-throughput detection system according to the material identification status and the horizontal projection area.

[0054] The LIBS multi-dimensional high-throughput detection system 4 excites the material to be measured through multiple groups of single-channel optical systems, obtains the atomic emission spectral information of the material, compares the obtained atomic emission spectral information of the material with the material model database, and classifies and identifies the material to be measured in combination with the classification algorithm.

[0055] The construction method of the material model database is as follows: The atomic emission spectral information of multiple materials of recycled high-quality metals with different grades is used as the training set. The parameters of the SVM (Support Vector Machine) are iteratively optimized by the GA (Genetic Algorithm), and a classification model is trained. The material model database is constructed based on the classification model. The material model database contains the grade and its corresponding element types and contents. The element composition of the material to be measured can be determined through the atomic emission spectral information of the material. By comparing with the element types and contents in the material model database, the grade of the material to be measured can be confirmed.

[0056] For example, the LIBS multi-dimensional high-throughput detection system 4 consists of three sets of single-path optical systems, each including a laser, an excitation optical path, a return optical path, and a spectrometer. These three sets of single-path optical systems form a comprehensive optical system that performs excitation scanning on nine titanium alloy grades one by one in a sampling scanning mode, collecting spectral data for nine different titanium alloy grades: TA3, TA7, TA15, TB2, TB13, TC1, TC4, TC6, and TC11. The classification model is constructed by optimizing the Support Vector Machine (SVM) using the Genetic Algorithm (GA). The SVM parameters are iteratively optimized using GA to train the classification model.

[0057] The material post-processing device 6 is composed of multiple material channels, each of which is provided with a rocker mechanism. The rocker swings by being pushed by a cylinder, thereby finally achieving the separation of the materials. The controller of the material post-processing device 6 receives the execution instruction of the control system 5 and performs classification and collection actions on different types of materials to be tested according to the execution instruction; the control system 5 generates the execution instruction according to the classification and identification results of the materials to be tested; for example, the material post-processing device (6) includes multiple different material channels for collecting different types of materials to be tested. Multiple material collection buckets can also be provided under different material channels for collecting different materials to be tested. For example, for the classification of recycled titanium alloy materials, it includes: 9 different grades of recycled high-quality titanium alloy materials, namely TA3, TA7, TA15, TB2, TB13, TC1, TC4, TC6, and TC11, are provided with a rocker mechanism to assign different materials to different material channels, thereby achieving the final material identification and classification.

[0058] Example 2

[0059] like Figures 3 - 5 As shown, the present invention also provides a method for automatically classifying and identifying recycled high-quality metals based on LIBS technology, which is applied to the above-mentioned system for automatically classifying and identifying recycled high-quality metals based on LIBS technology, comprising the following steps:

[0060] S1. Excitation area division:

[0061] During the detection process, according to the position of the material to be tested on the material transmission module, the detection area of the material to be tested is divided into a horizontal area and a vertical area. The horizontal area is the transmission direction of the material to be tested, and the detection area is divided into a front single-point scanning excitation area and a rear multi-point scanning excitation area. The vertical area division is perpendicular to the transmission direction of the material to be tested, and the detection area is divided into 1, 2, 3, ..., N areas;

[0062] S2. Confirmation of detection mode and collection of material spectrum information:

[0063] According to the detection requirements, the corresponding excitation mode is matched, the material to be tested is excited, and the material atomic emission spectrum information of the material to be tested is obtained; the excitation modes include single-point excitation mode, single-point and key material accumulation mode, line scan sampling mode, line scan sampling and regional multi-point accumulation mode; among them, the single-point excitation mode is for flake-like materials, and single-point material detection information is obtained; the single-point and key material accumulation mode, based on the single-point excitation mode, performs multi-point cumulative excitation on the material to be tested with large information differences, so as to achieve accurate classification and identification of the material to be tested; the line scan sampling mode, for chip-like materials, implements sampling scanning to obtain large-scale regional spectral information; the line scan sampling and regional multi-point accumulation mode, based on the line scan sampling mode, performs small-area multi-point cumulative excitation on areas with large information differences within a large area, so as to achieve accurate classification and identification of chip-like materials;

[0064] S3. Data processing of spectrum information of materials to be tested:

[0065] It is used to detect and analyze the material to be tested based on LIBS technology, obtain the detection information of the material to be tested, and compare it with the model database to obtain the classification and identification results of the material to be tested;

[0066] S4. Output of classification and identification results of materials to be tested:

[0067] The control system displays the classification and identification results on a visual panel; the material post-processing device classifies and collects different types of materials to be tested.

[0068] Example 2-1

[0069] Taking recycled titanium alloy materials as an example, the present invention provides a method for automatically classifying and identifying recycled high-quality metals based on LIBS technology, comprising the following steps:

[0070] S1. Excitation area division:

[0071] For recycled titanium alloy chips, such as Figure 5 As shown in (a), since the scraps can cover the entire conveyor belt, the entire surface of the conveyor belt can be defined as scraps, and the material detection area is divided into horizontal and vertical areas. The horizontal area is the material conveying direction. In this embodiment, the horizontal excitation area is divided into the front single-point scanning excitation area A and the rear multi-point scanning excitation area B. The vertical area division is perpendicular to the material conveying direction, and the detection area is divided into 1, 2, and 3 areas.

[0072] S2. Detection mode confirmation and material spectrum information collection:

[0073] For recycled titanium alloy scrap, such as Figure 5 In (b), the detection mode is line-scan sampling and area multi-point accumulation mode. Among them, in the pre-positioned single-point scanning excitation area, linear sampling scans are performed on 3 areas of the material, and the average spectral data of the 3 areas is used as the classification and identification information of the overall material; in the post-positioned multi-point scanning excitation area, the multi-point accumulation mode can be used, but no excitation is performed in this example; since the scrap is covered on the conveyor belt and the scrap moves along the conveyor belt, the frequency of the laser is high. Scanning the entire scrap forms a curved scanning path. Figure 5 The curve shown in (b) represents the path of the laser scanning the scrap, and the density of the curve represents different excitation modes;

[0074] S3. Processing of spectral information data of the material to be measured:

[0075] Process the spectral information of the obtained recycled titanium alloy material. Use the average spectral value of 3 areas as the information of the material to be measured, compare it with the model database, and give the classification and identification information;

[0076] S4. Output of the spectral information result of the material to be measured:

[0077] According to the data comparison result of S3, give the classification and identification information of the material. For example, classify and identify 9 different grades of recycled high-quality titanium alloy materials, namely TA3, TA7, TA15, TB2, TB13, TC1, TC4, TC6, and TC11, as shown in Table 1. The classification and identification accuracy rates are 96.70%, 96.40%, 95.90%, 96.81%, 96.86%, 96.79%, 95.86%, 95.72%, and 96.13% respectively.

[0078]

[0079] Example 2-2

[0080] Taking recycled high-quality metal stainless steel sheet-like materials as an example, the difference from Example 2-1 is that for recycled stainless steel sheet-like materials, such as Figure 5 In (c) and (d), in this example, the pre-positioned single-point scanning excitation area C and the post-positioned multi-point scanning excitation area D are divided. The number of dots on one material represents different excitation modes. The detection mode of the pre-positioned single-point scanning excitation area C is the single-point excitation mode, that is, single-point tracking excitation is performed on 3 areas of the material. The detection mode of the post-positioned multi-point scanning excitation area D is the single-point and key material accumulation mode, and no excitation is performed in this example; Figure 5 The dots shown in (d) represent the laser excitation spots, and the number of spots represents different excitation modes.

[0081] Taking the average value of the spectral data of three regions as the classification and identification information of the overall material, the grades of stainless steel materials tested are 201, 301, 304, 309, 310, 316, 321, 410, 430, and the test accuracies are 98.93%, 97.56%, 96.31%, 97.42%, 98.67%, 98.51%, 98.07%, 94.72%, 94.72% respectively.

[0082]

[0083] Compared with traditional metal sorting, the present invention uses LIBS technology to classify and identify recycled metals by elemental composition, and sets different classification and identification methods for different materials to be tested and classification and identification requirements. It integrates six modules such as a material transmission module, a pre-treatment device for materials to be tested, an information perception system for materials to be tested, a LIBS multi-dimensional high-throughput detection system, a control system, and a post-treatment device for materials, and can realize the fully automated online classification and identification of the grades of recycled metals, greatly improving the classification and identification efficiency and accuracy of the grades of recycled metals, and finally quickly classifying and identifying the grades of high-quality recycled metals.

[0084] Matters not covered in the present invention are well-known technologies.

[0085] Specific examples are used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, there will be changes in the specific implementation manner and application scope according to the idea of the present invention. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A high-quality regenerated metal automatic classification and recognition system based on LIBS technology, characterized in that, Including: A material transmission module (1), a pretreatment device for the material to be tested (2), an information sensing system for the material to be tested (3), a LIBS multi-dimensional high-throughput detection system (4), a control system (5), and a post-treatment device for the material (6); The material transmission module (1) is used to convey the material to be tested along a classification and identification production line; the material to be tested is a recycled high-quality metal material; The pretreatment device for the material to be tested (2) is used to pretreat the material to be tested so that the form and surface state of the material to be tested have relative consistency; The information sensing system for the material to be tested (3) is used to collect the basic information of the material to be tested and transmit it to the control system (5); the basic information includes the presence or absence, size, shape, height, and position information of the material to be tested; The control system (5) is used to control the LIBS multi-dimensional high-throughput detection system (4) based on the basic information of the material to be tested, including confirmation of the detection mode and collection of the material spectral information: According to the detection requirements, match the corresponding excitation mode, excite the material to be tested, and obtain the atomic emission spectral information of the material to be tested; the excitation modes include a single-point excitation mode, a single-point and key material accumulation mode, a line-scan sampling mode, and a line-scan sampling and area multi-point accumulation mode; The LIBS multi-dimensional high-throughput detection system (4) is used to detect and analyze the material to be tested based on LIBS technology to obtain the classification and identification result of the material to be tested; The post-treatment device for the material (6) is used to receive the execution instruction of the control system (5) and perform classification and collection actions on different types of materials to be tested according to the execution instruction; the control system (5) generates the execution instruction according to the classification and identification result of the material to be tested.

2. The high-quality metal automatic classification and recognition system based on LIBS technology according to claim 1, characterized in that, The pretreatment device for the material to be tested (2) includes a shearing device and a impurity-removing and cleaning device. The shearing device is used to shear and crush the material to be tested into materials with consistent forms. The forms of the material to be tested include: chip-shaped materials with a shape size of 3-15 mm; sheet-shaped and block-shaped materials with a shape size of more than 15 mm; The impurity-removing and cleaning device includes at least one of ultrasonic, high-temperature, and laser ablation devices, and is used to remove impurities and clean the surface of the material to be tested so that the surface state of the material to be tested is consistent.

3. The high-quality regenerated metal automatic classification and recognition system based on LIBS technology according to claim 1, characterized in that, The information sensing system for the material to be tested (3) includes an adjustable brightness illumination system, a variable focal length optical system, and an adaptive height sensing system. The adjustable brightness illumination system is used to illuminate the surface of the material to be tested to distinguish the material to be tested from the conveyor belt of the material transmission module (1). The variable focal length optical system is used to dynamically adjust the lens focal length to adapt to the size, position change, and detection requirements of the material to be tested, and ensure the imaging quality and detail capture ability; the adaptive height sensing system is used to continuously sense the height fluctuation of the material to be tested and dynamically adjust the sensor pose.

4. The high-quality metal automatic classification and recognition system based on LIBS technology according to claim 1, characterized in that The LIBS multi-dimensional high-throughput detection system (4) includes multiple groups of single-channel optical systems. The single-channel optical system includes a laser, an excitation optical path, a return optical path, and a spectrometer; The LIBS multi-dimensional high-throughput detection system (4) adopts laser-induced laser spectroscopy technology and is equipped with a fully automatic analysis program. The fully automatic analysis program automatically calculates the start time and end time of excitation and the working state of the laser in the LIBS multi-dimensional high-throughput detection system according to the material identification state and the area of the horizontal projection; The LIBS multi-dimensional high-throughput detection system (4) excites the material to be tested through multiple sets of single-path optical systems to obtain the atomic emission spectrum information of the material, compares the obtained atomic emission spectrum information of the material with the material model database, and classifies and identifies the material to be tested in combination with a classification algorithm.

5. The high-quality regenerated metal automatic classification and recognition system based on LIBS technology according to claim 4, wherein The material model database is constructed by using atomic emission spectrum information of multiple recycled high-quality metals of different brands as a training set, optimizing SVM parameters through GA iteration, training to obtain a classification model, and constructing a material model database based on the classification model.

6. The automatic classification and recognition system for recycled high-quality metals based on LIBS technology according to claim 1, wherein The control system (5) controls the start and stop time and scanning mode of the LIBS multi-dimensional high-throughput detection system (4); the control system (5) displays the classification and recognition results on a visual panel.

7. The automatic classification and recognition system for recycled high-quality metals based on LIBS technology according to claim 1, characterized in that, The material post-processing device (6) comprises a plurality of different material channels for collecting different types of materials to be tested.

8. A method for automatically classifying and identifying high-quality recycled metals based on LIBS technology, which is applied to the system for automatically classifying and identifying high-quality recycled metals based on LIBS technology according to any one of claims 1-7, and is characterized in that, The following steps are involved: S1. Excitation area division: During the detection process, the detection area of the material to be tested is divided into a horizontal area and a vertical area according to the position of the material to be tested on the material transmission module. The horizontal area is the transmission direction of the material to be tested, and the detection area is divided into a front single-point scanning excitation area and a rear multi-point scanning excitation area; the vertical area division is perpendicular to the transmission direction of the material to be tested, and the detection area is divided into 1, 2, 3, ..., N areas; S2. Confirmation of detection mode and collection of material spectrum information: According to the detection requirements, the corresponding excitation mode is matched, the material to be tested is excited, and the material atomic emission spectrum information of the material to be tested is obtained; the excitation modes include single-point excitation mode, single-point and key material accumulation mode, line scan sampling mode, line scan sampling and regional multi-point accumulation mode; S3. Data processing of spectrum information of materials to be tested: It is used to detect and analyze the material to be tested based on LIBS technology, obtain the detection information of the material to be tested, and compare it with the model database to obtain the classification and identification results of the material to be tested; S4. Output of classification and identification results of materials to be tested: The control system displays the classification and identification results on a visual panel; the material post-processing device classifies and collects different types of materials to be tested.

9. The method for automatically classifying and identifying high-quality recycled metals based on LIBS technology according to claim 8, characterized in that In the S2, the single-point excitation mode is used to implement single-point excitation for flake-like materials to obtain single-point material detection information; the single-point and key material accumulation mode, based on the single-point excitation mode, performs multi-point accumulation excitation on the materials to be tested with large information differences, so as to achieve accurate classification and identification of the materials to be tested; the line scan sampling mode is used to implement sampling scanning for chip-like materials to obtain spectral information of a large range of areas; the line scan sampling and regional multi-point accumulation mode, based on the line scan sampling mode, performs small-area multi-point accumulation excitation on areas with large information differences within a large range of areas, so as to achieve accurate classification and identification of chip-like materials.

Images