WO2023197510A1 - 一种铝土矿识别方法及装置 - Google Patents

一种铝土矿识别方法及装置 Download PDF

Info

Publication number
WO2023197510A1
WO2023197510A1 PCT/CN2022/116015 CN2022116015W WO2023197510A1 WO 2023197510 A1 WO2023197510 A1 WO 2023197510A1 CN 2022116015 W CN2022116015 W CN 2022116015W WO 2023197510 A1 WO2023197510 A1 WO 2023197510A1
Authority
WO
WIPO (PCT)
Prior art keywords
bauxite
tailings
concentrate
water
identification
Prior art date
Application number
PCT/CN2022/116015
Other languages
English (en)
French (fr)
Inventor
李太友
葛小冬
罗洋
秦野
Original Assignee
天津美腾科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 天津美腾科技股份有限公司 filed Critical 天津美腾科技股份有限公司
Publication of WO2023197510A1 publication Critical patent/WO2023197510A1/zh

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
    • B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
    • B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
    • B07C5/34—Sorting according to other particular properties
    • B07C5/342—Sorting according to other particular properties according to optical properties, e.g. colour
    • B07C5/3422—Sorting according to other particular properties according to optical properties, e.g. colour using video scanning devices, e.g. TV-cameras
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
    • B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
    • B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
    • B07C5/02—Measures preceding sorting, e.g. arranging articles in a stream orientating
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
    • B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
    • B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
    • B07C5/36—Sorting apparatus characterised by the means used for distribution
    • B07C5/38—Collecting or arranging articles in groups
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
    • Y02W30/00—Technologies for solid waste management
    • Y02W30/50—Reuse, recycling or recovery technologies
    • Y02W30/52—Mechanical processing of waste for the recovery of materials, e.g. crushing, shredding, separation or disassembly

Definitions

  • This application relates to the technical field of bauxite sorting, for example, to a bauxite identification method and device.
  • Bauxite also known as bauxite, is a general name for ores composed of gibbsite, boehmite or diaspore as the main minerals. Bauxite is widely used in my country's industrial field, and the demand for bauxite in my country is huge every year. Bauxite often coexists with iron oxides and hydroxides, anatase, and clay minerals such as kaolinite and chlorite, and sometimes contains calcium, magnesium, sulfur and other minerals. Bauxite ore can be divided into three categories: high-alkali bauxite, high-titanium bauxite and high-iron bauxite according to the impurities contained in it.
  • the process of sorting bauxite concentrate from bauxite ore is actually a process of removing gangue minerals and harmful impurities, and separating high-aluminum minerals and low-aluminum minerals to obtain a concentrate with a high aluminum-to-silicon ratio.
  • the main mineral processing process of bauxite will adopt different mineral processing processes according to different types of ores.
  • the beneficiation process of gibbsite-kaolinite bauxite usually uses mud and sand separation first, coarse-grade grinding, magnetic separation to remove iron, and mud grinding and flotation.
  • This application provides a bauxite identification method that can effectively improve the bauxite sorting accuracy.
  • This application provides a bauxite identification method, including:
  • the bauxite is marked as concentrate or tailings based on the parameter information.
  • Figure 1 is a flow chart of the bauxite identification method provided by this application.
  • Figure 2 is a schematic structural diagram of a bauxite identification device provided by this application.
  • Figure 3 is a schematic structural diagram of another bauxite identification device provided by this application.
  • first position and second position are two different positions, and the first feature “on”, “above” and “above” the second feature include the first feature on the second feature. Directly above and diagonally above, or simply means that the level of the first feature is higher than that of the second feature. “Below”, “under” and “under” the first feature is the second feature includes the first feature being directly below and diagonally below the second feature, or simply means that the first feature is less horizontally than the second feature.
  • connection should be understood in a broad sense.
  • it can be a fixed connection, a detachable connection, or an integral connection; it can be A mechanical connection can also be an electrical connection; it can be a direct connection or an indirect connection through an intermediary, or it can be an internal connection between two components.
  • a mechanical connection can also be an electrical connection; it can be a direct connection or an indirect connection through an intermediary, or it can be an internal connection between two components.
  • a photoelectric sorter using an X-ray source conducts transmission scanning of the selected ore through X-rays to obtain the atomic number data of the minerals contained inside the ore, establish an identification model, identify the ore and miscellaneous rocks, and then drive the actuator Carry out ore sorting.
  • the sorting goal is to increase the aluminum-silicon ratio. Since the atomic numbers of aluminum and silicon are close, after X-ray transmission scanning, the K values of the two elements are very close and difficult to distinguish. Therefore, bauxite cannot be sorted by X-ray scanning photoelectric sorting machine.
  • Photoelectric sorting machines that use image technology use color or other identifiable features on the image to sort the types of ores using technical means such as deep learning.
  • image recognition technology works by detecting salient areas, i.e. the parts of an image or object that contain the most information. It does this by isolating and locating the most informative parts or features in a selected image, while ignoring what might be less likely Other characteristics of interest.
  • This process uses an image recognition algorithm, also known as an image classifier, that takes an image as input and outputs what the image contains.
  • This application provides a bauxite identification method that can improve the bauxite sorting accuracy.
  • the above-mentioned bauxite identification method includes the following steps:
  • the bauxite in order to ensure that the surface of the bauxite is completely sprayed wet, during the spraying process of bauxite, the bauxite can be vibrated to tumble at the same time, thereby ensuring that the surface of the bauxite is wetted with water to avoid Since no part of the bauxite surface is wetted by water, the identification effect is affected, which is beneficial to ensuring the accuracy of bauxite identification.
  • one of the following methods can be used to detect bauxite to obtain the required parameter information:
  • the bauxite concentrate and tailings can be identified by obtaining a grayscale image of the bauxite surface. The steps are:
  • the wetted bauxite is irradiated with near-infrared light of two different bands;
  • one of the two segments of near-infrared light can be selected to be a band near the water absorption peak, such as 1900nm to 1940nm, for example, it can be 1900nm, 1910nm, 1920nm, 1930nm or 1940nm, and the other segment can be selected to be far away from the water absorption peak.
  • the wavelength band such as 780nm to 820nm, can be, for example, 780nm, 790nm, 800nm, 810nm or 820nm.
  • One of the two segments of near-infrared light selects a band near the water absorption peak, while the other selects a band far away from the water absorption peak.
  • the grayscale difference between the pixels of the photos taken in different bands is large. It facilitates the identification of concentrates and tailings and also helps improve the identification accuracy.
  • the concentrate has strong water absorption.
  • the surface of the concentrate When photographed by a near-infrared camera, the surface of the concentrate will appear reflective and bright in the absence of water. Since the tailings absorb almost no water, the water on the surface of the tailings will absorb light in the near-infrared band. When photographed by a near-infrared camera, the surface of the tailings appears black, and the precision can be identified through the pixel grayscale value of the photo. Mines and tailings.
  • the second preset time is to allow the concentrate to absorb water. The time it takes to completely absorb water avoids affecting the identification accuracy of bauxite due to the short water absorption time.
  • the second preset time can be set according to the type of bauxite and the water absorption of the bauxite concentrate, and can be 1s to 60s.
  • it can be 1s, 10s, 20s, 30s, 40s, or 50s. or 60s, just set it according to the actual situation.
  • the camera takes pictures of the bauxite after being illuminated in different wavebands, compares the two photos taken of the same bauxite, and calculates the pixel grayscale difference between the two photos. Since there is basically no water on the surface of the concentrate, the pixel grayscale difference is small, while the tailings have water on the surface, and the pixel grayscale difference is generally large.
  • the preset value is marked as tailings, and the pixel grayscale difference less than or equal to the first preset value is marked as concentrate, thereby realizing the identification of concentrate and tailings.
  • This method is suitable for most bauxite ores. Compared with bauxite identification methods in related technologies, the identification accuracy is high and is conducive to improving the aluminum-silicon ratio of bauxite entering subsequent processes.
  • the first preset value of the pixel grayscale difference can be set according to the type of bauxite. Generally, it can be set to 80 to 85. For example, it can be 80, 81, 82, 83, 84 or 85. , you can set it according to actual needs.
  • the bauxite concentrate and tailings can be identified by obtaining a spectral image of the bauxite surface. The steps are:
  • the hyperspectral camera can identify the full range of light, so a halogen lamp is used to illuminate the wetted bauxite.
  • Hyperspectral imaging technology is based on many narrow-band image data technologies. It combines imaging technology with spectral technology to detect the two-dimensional geometric space and one-dimensional spectral information of the target, and obtain continuous, narrow-band images with high spectral resolution. data.
  • the one-dimensional information on the bauxite surface passes through the lens and slit of the hyperspectral camera, light of different wavelengths spreads according to different degrees of bending and dispersion.
  • Each point on this one-dimensional image is then diffracted and split by a grating to form a
  • the spectral band is illuminated on the detector, and the position and intensity of each pixel on the detector characterizes the spectrum and intensity.
  • a point corresponds to a spectrum section, and a line corresponds to a spectrum.
  • the second preset time is to allow the concentrate to absorb water. The time it takes to completely absorb water avoids affecting the identification accuracy of bauxite due to the short water absorption time.
  • the second preset time can be set according to the type of bauxite and the water absorption of the bauxite concentrate, and can be 1s to 60s.
  • it can be 1s, 10s, 20s, 30s, 40s, or 50s. or 60s, just set it according to the actual situation.
  • tailings Since the tailings absorb almost no water, there is a complete water film on the surface of the tailings. When identified by a hyperspectral camera, its spectral characteristics are consistent with those of water, while the concentrate has no moisture on the surface due to its strong water absorption. It is significantly different from the spectral characteristics of tailings.
  • the spectral image of bauxite can be compared with the spectral image of water, and a second preset value can be set to mark the bauxite whose similarity to the spectral image of water is greater than or equal to the second preset value.
  • a second preset value can be set to mark the bauxite whose similarity to the spectral image of water is greater than or equal to the second preset value.
  • bauxite whose similarity to the spectral image of water is smaller than the second preset value is marked as concentrate.
  • computer programming can be used to assist identification, and the information of the spectral image of water can be trained through the neural network method. Then, using the established neural network model, only the spectral image information of bauxite can be input to obtain the concentrate or The identification results of tailings have high identification efficiency and high accuracy.
  • the second preset value of similarity can be set according to the type of bauxite, and can generally be set to 80% to 95%.
  • it can be 80%, 81%, 82%, or 83%.
  • the bauxite concentrate and tailings can be identified by obtaining the spectral curve of the bauxite surface. The steps are:
  • the excited state is also different, that is, the wavelength of the reflected light is different.
  • the reflected light information of the object is captured and analyzed through an online spectrometer, so that the content of the object can be detected. What kind of element.
  • the second preset time is to allow the concentrate to absorb water. The time it takes to completely absorb water avoids affecting the identification accuracy of bauxite due to the short water absorption time.
  • the second preset time can be set according to the type of bauxite and the water absorption of the bauxite concentrate, and can be 1s to 60s.
  • it can be 1s, 10s, 20s, 30s, 40s, or 50s. or 60s, just set it according to the actual situation.
  • tailings Since the tailings absorb almost no water, there is a complete water film on the surface of the tailings. When identified by an online spectrometer, its spectral characteristics are consistent with those of water, while the concentrate has strong water absorption, so there is no moisture on the surface of the concentrate. , which is significantly different from the spectral characteristics of tailings.
  • the spectral data of each bauxite is calculated through the algorithm software of the online spectrometer, and the spectral curve of bauxite is compared with the spectral curve of water, and a third preset value is set to compare with the spectral curve of water.
  • the bauxite whose spectral curve similarity is greater than or equal to the third preset value is marked as tailings, and the bauxite whose similarity to the spectral curve of water is less than the third preset value is marked as concentrate.
  • the third preset value of similarity can be set according to the type of bauxite, and can generally be set to 80% to 95%.
  • it can be 80%, 81%, 82%, or 83%.
  • an online spectrometer in the 780nm to 2100nm band can be selected, and the integration time of the online spectrometer can be set to 100ms.
  • the bauxite concentrate and tailings can be identified by obtaining the moisture value of the bauxite surface.
  • the steps are:
  • Water molecules are not static. When they encounter specific energy bands, they vibrate, and the bonds between the two hydrogen atoms and the oxygen atoms stretch, contract, or distort in other forms. In different parts of the entire spectrum, some absorption bands are very strong and some are very weak. In the near-infrared part of the spectrum, the absorption of water molecules is particularly strong.
  • the hydrogen-oxygen bonds in the water will absorb the near-infrared rays of a specific wavelength (the specific wavelength is 1940nm). At a specific wavelength, the reflected near-infrared energy is absorbed by the material. The amount of near-infrared energy absorbed by water molecules is inversely proportional.
  • the infrared moisture detector can take advantage of this feature and calculate the moisture content of the bauxite surface under test based on the energy loss.
  • the second preset time is the time for the concentrate to completely absorb water, so as to avoid affecting the identification accuracy of bauxite due to the short water absorption time.
  • the second preset time can be set according to the type of bauxite and the water absorption of the bauxite concentrate, and can be 1s to 60s.
  • it can be 1s, 10s, 20s, 30s, 40s, or 50s. or 60s, just set it according to the actual situation.
  • the tailings absorb almost no water, there is a complete water film on the surface of the tailings, while the concentrate has strong water absorption, so there is almost no moisture on the surface of the concentrate.
  • the energy lost When identified by an infrared moisture detector, the energy lost will be significant. The difference is that the water content is significantly different.
  • the moisture on the bauxite surface is detected by an infrared moisture detector, and a moisture content threshold is set.
  • Bauxite with a moisture content greater than the moisture content threshold is marked as tailings
  • bauxite with a moisture content less than or equal to the moisture content threshold is marked as tailings.
  • Bauxite is labeled as concentrate.
  • the moisture content threshold can be set according to the type of bauxite, generally it can be set to 80% to 95%, for example, it can be 80%, 81%, 82%, 83%, 84%, 85% %, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94% or 95%, you can set it according to actual needs.
  • the bauxite concentrate and tailings can be identified by obtaining infrared imaging spectra of the bauxite surface. The steps are:
  • the time for bauxite to absorb water can be set as the first preset time to ensure that the bauxite water absorption time reaches the first preset time. After setting, dry the moisture on the bauxite surface.
  • the bauxite surface needs to be blown dry to remove the surface moisture.
  • the moisture on the surface of the bauxite can be blown dry using an air knife.
  • microwave heating can be used to heat bauxite. Since the tailings absorb little water, there is almost no water inside the tailings, while the concentrate has strong water absorption, so there is more moisture inside the concentrate. It is known that the specific heat capacity of water is 4.2KJ/kg °C, while the specific heat capacity of bauxite ore is 0.75-1.2KJ/kg °C. There is a big difference between the two. Due to the difference in whether there is water inside the concentrate and the tailings, microwave heating of bauxite will lead to different temperature rises in each. The temperature rise of tailings without moisture inside is greater than that of concentrates with moisture inside.
  • the bauxite when microwave heating of bauxite, the bauxite can be placed in the cabin, and a suppressor to prevent microwave leakage can be set in the cabin. On the one hand, it can prevent microwave leakage and ensure microwave heating efficiency; On the other hand, harmful microwave pollution to the environment can be avoided.
  • the infrared imager can draw a surface temperature map of the object based on the infrared rays emitted by the monitored object. Therefore, the infrared imager can form an infrared imaging spectrum corresponding to different temperatures and different colors, and then analyze the infrared imaging spectrum through the algorithm software in the server, and mark the bauxite with a temperature difference greater than the fourth preset value as tail. Bauxite ore whose temperature difference is less than or equal to the fourth preset value is marked as concentrate.
  • the fourth preset value of the temperature difference can be set according to the type of bauxite. Generally, it can be set to 80°C to 85°C. For example, it can be 80°C, 81°C, 82°C, or 83°C. , 84°C or 85°C, you can set it according to actual needs.
  • the drop point of the concentrate or tailings can be changed by blowing, and the concentrate can be sorted to the first Preset position, sort the tailings to the second preset position to achieve separation of concentrate and tailings.
  • the tailing operation is generally carried out to transport the concentrate to the next process.
  • the concentrate or tailings can also be picked out by grabbing to achieve separation of concentrates and tailings.
  • the bauxite identification method provided in this application utilizes the characteristics of strong water absorption of concentrates and almost no water absorption of tailings to wet the surface of bauxite, and then utilizes the different moisture content on the surface of bauxite concentrate and tailings after water absorption.
  • the bauxite surface is detected to obtain the parameter information of the bauxite surface, and the bauxite concentrate and tailings are identified based on the obtained parameter information.
  • This method is generally applicable to bauxite and is beneficial to the pre-disposal of bauxite. This stage improves the sorting accuracy, thereby increasing the aluminum-silicon ratio that enters subsequent processes.
  • This application also provides a bauxite identification device, which can identify bauxite concentrate and bauxite tailings using the above-mentioned bauxite identification method, with high identification accuracy.
  • the above-mentioned bauxite identification device includes a wetting device 100, a detection device 300 and an identification device 400, wherein the wetting device 100 is configured to spray the surface of the bauxite 200 water so that the surface of bauxite 200 is completely sprayed wet.
  • the detection device 300 is configured to detect the surface of the bauxite 200 and can be a near-infrared camera, a hyperspectral camera, an online spectrometer, an infrared moisture detector or an infrared imager, etc., which can be selected according to actual needs.
  • the identification device 400 is data connected to the detection device 300 and is configured to obtain parameter information on the surface of the bauxite 200 and to identify the concentrate and tailings based on the above parameter information.
  • the above device can realize the identification of bauxite 200 concentrate and tailings. Since bauxite 200 generally has the characteristics of strong water absorption in concentrates and almost no water absorption in tailings, it is comparable to bauxite identification devices in related technologies. Compared with other bauxite identification devices, the above-mentioned bauxite identification device has a wider application range and is conducive to improving the identification accuracy and sorting accuracy of bauxite 200.
  • the above-mentioned bauxite identification device may include a first conveying device 500.
  • the first conveying device 500 can convey the bauxite 200, thereby realizing automatic identification of the bauxite 200.
  • the wetting device 100 may be disposed above the first conveying device 500 and configured to spray wet the surface of the bauxite 200 .
  • the humidifying device 100 may also be disposed below the first conveying device 500 .
  • the humidifying device 100 can also be disposed above and below the first conveying device 500 at the same time, and can be disposed according to actual needs.
  • the distance between the detection device 300 and the starting end of the first conveying device 500 is defined as a preset distance.
  • the preset distance is the distance from the starting end of the bauxite 200 to the detection device 300 through the second preset time to ensure that the bauxite ore 200 has sufficient water absorption time to avoid the insufficient water absorption time of bauxite 200 causing more water on the surface of the concentrate to affect the identification effect.
  • the first conveying device 500 may also include a vibrating feeder 510.
  • the bauxite 200 is placed in the vibrating feeder 510, tumbled under the vibration of the vibrating feeder 510, and the spray device is used to Spraying the bauxite 200 in the vibrating feeder 510 is conducive to ensuring that the surface of the bauxite 200 is completely sprayed, thereby preventing the detection results from being affected by not being sprayed on the surface of the bauxite 200, and ensuring the identification accuracy. .
  • the above-mentioned bauxite identification device may also include a second conveying device 600.
  • the starting end of the second conveying device 600 is connected to the tail end of the first conveying device 500, capable of transporting the first The bauxite 200 on the conveying device 500 is transported to the identification device 400 for identification.
  • the second conveying device 600 may be a slide.
  • the second conveying device 600 can also be a conveyor belt, which can be set according to actual needs.
  • the above-mentioned bauxite identification device can also include a sorting device 700, which is configured to pre-dispose the bauxite 200.
  • the sorting device 700 can sort the concentrate to the first preset position 710, and the tailings can be separated. The ore is sorted to the second preset position 720 to realize the sorting of concentrate and tailings.
  • the sorting device 700 may be a nozzle.
  • the dispensing device can also be a robot, which can be selected according to actual needs.
  • the above-mentioned bauxite identification device can also include a feeding device 800.
  • the feeding device 800 is provided with a feeding port, and can control the opening and closing of the feeding port through a hydraulic gate, and then Achieve control over the feed amount.
  • the bauxite 200 concentrate and tailings can be identified, and the concentrates and tailings can be sorted out, which is beneficial to dumping the tailings and transporting the concentrate to the downstream.
  • the above-mentioned bauxite identification device is suitable for identifying bauxite 200, which is beneficial to improving the sorting accuracy of bauxite 200, thereby increasing the aluminum-silicon ratio entering the next process.
  • this application uses spraying wet bauxite to detect parameter information on the bauxite surface to obtain parameters related to moisture on the bauxite surface. Then the concentrate and tailings are identified. Compared with the bauxite identification method in the related technology, it can be applied to the identification of bauxite concentrate and tailings, and has a wider scope of application, and can then be used in bauxite pre-throwing waste. This stage greatly improves the sorting accuracy of bauxite and effectively improves the aluminum-silicon ratio of bauxite entering subsequent processes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)
  • Sorting Of Articles (AREA)

Abstract

一种铝土矿识别方法及装置，该铝土矿识别方法包括：将铝土矿表面润湿；对润湿后的铝土矿的表面进行检测，获得铝土矿表面的参数信息；根据参数信息将铝土矿标记为精矿或尾矿。

Description

一种铝土矿识别方法及装置
本申请要求在2022年04月11日提交中国专利局、申请号为202210371175.8的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。
技术领域
本申请涉及铝土矿分选技术领域，例如涉及一种铝土矿识别方法及装置。
背景技术
铝土矿又称铝矾土，是以三水铝石、一水软铝石或一水硬铝石为主要矿物所组成的矿石的统称。铝土矿在我国工业领域有着广泛的用途，每年我国的铝土矿需求量十分庞大。铝土矿经常与铁的氧化物和氢氧化物、锐钛矿及高岭石、绿泥石等粘土矿物共生，有时还含钙、镁、硫等矿物。铝土矿石按其所含杂质可分为高碱铝土矿、高钛铝土矿、高铁铝土矿三类。
从铝土矿矿石中分选出铝土矿精矿的过程其实就是一个除去脉石矿物和有害杂质，分离高铝矿物和低铝矿物，以获得高铝硅比的精矿的过程。相关技术中，铝土矿的主要选矿流程会根据矿石的不同类型，采用不同的选矿工艺流程。如三水铝石-高岭石类铝土矿的选矿流程，常采用先进行泥、砂分选，粗级别磨矿后用磁选除铁，矿泥磨矿后浮选。但是这样的分选方法经济效益非常低，因此往往会采取预选抛废的选矿工艺来大幅降低入磨量，有效的降低选矿和铝土矿冶炼后续成本，有利于低铝硅比铝土矿矿石的开发利用，扩大资源利用率。
发明内容
本申请提供一种铝土矿识别方法，能够有效提高铝土矿的分选精度。
本申请提供一种铝土矿识别方法，包括：
将铝土矿表面润湿；
对润湿后的所述铝土矿的表面进行检测，获得所述铝土矿表面的参数信息；
根据所述参数信息将所述铝土矿标记为精矿或尾矿。
附图说明
图1为本申请提供的铝土矿识别方法的流程图；
图2为本申请提供的一种铝土矿识别装置的结构示意图；
图3为本申请提供的另一种铝土矿识别装置的结构示意图。
图中：
100、湿润装置；200、铝土矿；300、检测装置；400、识别装置；500、第一输送装置；510、振动给料器；600、第二输送装置；700、分选装置；710、第一预设位置；720、第二预设位置；800、入料装置。
具体实施方式
下面将结合附图对本申请进行描述，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
在本申请的描述中，需要说明的是，术语“中心”、“上”、“下”、“左”、“右”、“竖直”、“水平”、“内”、“外”等指示的方位或位置关系为基于附图所示的方位或位置关系，是为了便于描述本申请和简化描述，而不是指示或暗示所指的装置或元件具有特定的方位、以特定的方位构造和操作。此外，术语“第一”、“第二”、仅用于描述，而不能理解为指示或暗示相对重要性。其中，术语“第一位置”和“第二位置”为两个不同的位置，而且，第一特征在第二特征“之上”、“上方”和“上面”包括第一特征在第二特征正上方和斜上方，或仅仅表示第一特征水平高度高于第二特征。第一特征在第二特征“之下”、“下方”和“下面”包括第一特征在第二特征正下方和斜下方，或仅仅表示第一特征水平高度小于第二特征。
在本申请的描述中，需要说明的是，术语“安装”、“相连”、“连接”应做广义理解，例如，可以是固定连接，也可以是可拆卸连接，或一体地连接；可以是机械连接，也可以是电连接；可以是直接相连，也可以通过中间媒介间接相连，可以是两个元件内部的连通。对于本领域的普通技术人员而言，可以根据实际情况理解上述术语在本申请中的实际含义。
下面详细描述本申请的实施例，所述实施例的示例在附图中示出，其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。下面通过参考附图描述的实施例是示例性的，用于解释本申请，。
相关技术中，一般有如下铝土矿分选的方法：
(1)使用X射源的光电分选机通过X射线对所选矿石进行透射扫描，获得矿石内部所含矿物质的原子序数数据，建立识别模型，识别出矿石与杂石，进而驱动执行机构进行矿石分选。但是对于铝土矿而言，其分选目标是提高铝硅比，由于铝元素和硅元素原子序数接近，X射线透射扫描后，这两种元素得 出的K值非常接近，很难区分。因此无法通过X光扫描的光电分选机进行铝土矿的分选。
(2)使用图像技术的光电分选机通过颜色或者图像上其他可识别特征，采用深度学习等技术手段，对矿石的种类进行分选。但是图像识别技术的工作原理是检测显著区域，即包含图像或物体最多信息的部分，它通过隔离所选图像中信息量最大的部分或特征并对其定位来实现这一点，同时忽略可能不太感兴趣的其他特征。该过程使用图像识别算法，也称为图像分类器，以图像作为输入并输出图像包含的内容。由于图像识别的关键是各个不同分选目标之间具备统一的信息差异性，部分铝土矿的精矿和尾矿确实具备这样的特点，可以通过图像识别技术进行分选，但仍有一大部分铝土矿矿石复杂，来料多样，精矿和尾矿之间并不具备明显的差异性，这也就导致图像识别训练成本极高，精度很低，并不能适用于这类目标的分选。
本申请提供一种铝土矿识别方法，能够提高铝土矿的分选精度。
在一实施例中，如图1所示，上述铝土矿识别方法包括以下步骤：
S1、将铝土矿表面喷湿。
可选地，为了保证铝土矿表面被完全喷湿，在对铝土矿进行喷湿工序时，可以同时振动铝土矿使其翻滚，进而保证铝土矿的表面都被水淋湿，避免由于铝土矿表面一处未被水喷湿而影响识别效果，有利于保证铝土矿识别的精准性。
S2、对润湿后的铝土矿的表面进行检测，获得铝土矿表面的参数信息。
在一实施例中，可以采用以下方法中的一种对铝土矿进行检测，以获得所需的参数信息：
S21、获取铝土矿表面的灰度图像。
S22、获取铝土矿表面的光谱图像。
S23、获取铝土矿表面的光谱曲线。
S24、获取铝土矿表面的水分值。
S25、获取铝土矿表面的红外成像图谱。
可选地，在一个实施例中，可以通过获取铝土矿表面的灰度图像的方法识别铝土矿精矿和尾矿，步骤为：
S211、对润湿后的铝土矿先后采用两种不同波段的近红外光进行照射；
由于近红外波段的光可以被水吸收，且铝土矿具有精矿吸水性强，尾矿几乎不吸水的特点，因此，可以利用这一特点对精矿和尾矿进行识别。
可选地，两段近红外光中一段可以选择水吸收峰值附近的波段，如1900nm～1940nm，示例性地，可以是1900nm、1910nm、1920nm、1930nm或1940nm，另一段可以选择远离水吸收峰值附近的波段，如780nm～820nm，示例性地，可以是780nm、790nm、800nm、810nm或820nm。两段近红外光一段选择水吸收峰值附近的波段，而另一段选择远离水吸收峰值附近的波段，对于表面有水的尾矿来说，不同波段下拍摄出的照片像素灰度差较大，便于精矿与尾矿的识别，也有利于提高识别精度。
S212、通过近红外相机对铝土矿被近红外光照射的位置进行拍照，以获得两张不同波段的近红外光照射下的铝土矿的照片。
精矿吸水性强，在近红外相机的拍照下，精矿表面在没有水的情况下会呈现反射的光亮状态。由于尾矿几乎不吸水，因此尾矿表面的水会将近红外波段的光吸收，在近红外相机的拍照下，尾矿表面呈发黑的状态，进而能够通过照片的像素灰度值识别出精矿和尾矿。
在一实施例中，由于铝土矿吸水需要一定的时间，因此，在对铝土矿拍照之前，应该保证铝土矿的吸水时间达到第二预设时间，第二预设时间是使精矿能够完全将水吸收的时间，避免由于吸水时间短而影响铝土矿的识别精度。
可选地，第二预设时间可以根据铝土矿的品种和铝土矿精矿的吸水性设置，可以为1s～60s，示例性地，可以为1s、10s、20s、30s、40s、50s或60s，根据实际情况设置即可。
S213、获取两张照片的像素灰度差，将像素灰度差大于第一预设值的铝土矿标记为尾矿，将像素灰度差小于或等于第一预设值的铝土矿标记为精矿。
由于水对不同波段的近红外光的吸收情况不同，因此，为了确保铝土矿识别的精准性，选取两个不同波段的近红外光先后对喷湿的铝土矿进行照射，并使用近红外相机对不同波段照射后的铝土矿进行拍照，对同一个铝土矿拍摄的两张照片进行对比，计算两张照片的像素灰度差。精矿由于表面基本没有水，因此像素灰度差较小，而尾矿表面有水，像素灰度差值一般较大，所以，设置一个第一预设值，将像素灰度差大于第一预设值的标记为尾矿，将像素灰度差小于或等于第一预设值的标记为精矿，实现精矿和尾矿的识别。这种方法适用于绝大多数铝土矿，与相关技术中的铝土矿识别方法相比，识别精度高，有利于提高进入后续工序的铝土矿的铝硅比。
在一实施例中，像素灰度差的第一预设值可以根据铝土矿的品种设置，一般可以设置为80～85，示例性地，可以是80、81、82、83、84或85，根据实际需要设置即可。
可选地，在另一个实施例中，可以通过获取铝土矿表面的光谱图像的方法识别铝土矿精矿和尾矿，步骤为：
S221、对润湿后的铝土矿采用卤素灯进行照射。
由于铝土矿表面的光能量很低，难以精确识别出精矿和尾矿，因此，需要增加光源对润湿后的铝土矿进行照射，以保证精矿和尾矿的识别精确性，而高光谱相机能够对全波段的光进行识别，因此选用卤素灯对润湿后的铝土矿进行照射。
S222、通过高光谱相机对铝土矿被卤素灯照射的位置进行拍照。
高光谱成像技术是基于非常多窄波段的影像数据技术，它将成像技术与光谱技术相结合，探测目标的二维几何空间及一维光谱信息，获取高光谱分辨率的连续、窄波段的图像数据。
铝土矿表面的一维信息通过高光谱相机的镜头和狭缝后，不同波长的光按照不同程度的弯散传播，这一维图像上的每个点，再通过光栅进行衍射分光，形成一个谱带，照射到探测器上，探测器上的每个像素位置和强度表征光谱和强度。一个点对应一个谱段，一条线就对应一个谱面。
在一实施例中，由于铝土矿吸水需要一定的时间，因此，在对铝土矿拍照之前，应该保证铝土矿的吸水时间达到第二预设时间，第二预设时间是使精矿能够完全将水吸收的时间，避免由于吸水时间短而影响铝土矿的识别精度。
可选地，第二预设时间可以根据铝土矿的品种和铝土矿精矿的吸水性设置，可以为1s～60s，示例性地，可以为1s、10s、20s、30s、40s、50s或60s，根据实际情况设置即可。
S223、获取铝土矿的光谱图像，将上述光谱图像与水的光谱图像进行比较，将铝土矿的光谱图像与水的光谱图像的相似度大于或等于第二预设值的铝土矿标记为尾矿，将铝土矿的光谱图像与水的光谱图像的相似度小于第二预设值的铝土矿标记为精矿。
由于尾矿几乎不吸水，因此尾矿表面有完整的水膜，在通过高光谱相机进行识别时，其光谱特性和水的光谱特性保持一致，而精矿由于吸水性强而使得表面没有水分，和尾矿的光谱特征有显著的区别。
因此，可以将铝土矿的光谱图像与水的光谱图像进行比对，并设置一个第二预设值，将与水的光谱图像的相似度大于或等于第二预设值的铝土矿标记为尾矿，将与水的光谱图像的相似度小于第二预设值的铝土矿标记为精矿。
可选地，可以通过计算机编程辅助识别，通过神经网络的方法对水的光谱 图像的信息进行训练，然后利用建立的神经网络模型，只需输入铝土矿的光谱图像信息即可获得精矿或尾矿的识别结果，识别效率高，且准确性高。
在一实施例中，相似度的第二预设值可以根据铝土矿的品种设置，一般可以设置为80％～95％，示例性地，可以是80％、81％、82％、83％、84％、85％、86％、87％、88％、89％、90％、91％、92％、93％、94％或95％，根据实际需要设置即可。
可选地，在又一个实施例中，可以通过获取铝土矿表面的光谱曲线的方法识别铝土矿精矿和尾矿，步骤为：
S231、通过在线光谱仪对铝土矿表面进行拍照；
由于每一种元素的基态是不同的，所以激发态也是不一样的，也就是反射的光的波长不同，通过在线光谱仪对物块反射光信息进行抓取、分析，从而可以测知物品中含有何种元素。
在一实施例中，由于铝土矿吸水需要一定的时间，因此，在对铝土矿拍照之前，应该保证铝土矿的吸水时间达到第二预设时间，第二预设时间是使精矿能够完全将水吸收的时间，避免由于吸水时间短而影响铝土矿的识别精度。
可选地，第二预设时间可以根据铝土矿的品种和铝土矿精矿的吸水性设置，可以为1s～60s，示例性地，可以为1s、10s、20s、30s、40s、50s或60s，根据实际情况设置即可。
S232、获取铝土矿表面单点的光谱曲线，将铝土矿的光谱曲线与水的光谱曲线进行比较，将铝土矿的光谱曲线与水的光谱曲线的相似度大于或等于第三预设值的铝土矿标记为尾矿，将铝土矿的光谱曲线与水的光谱曲线的相似度小于第三预设值的铝土矿标记为精矿。
由于尾矿几乎不吸水，因此尾矿表面有完整的水膜，在通过在线光谱仪进行识别时，其光谱特性和水的光谱特性保持一致，而精矿由于吸水性强，因此精矿表面没有水分，和尾矿的光谱特征有显著的区别。
因此，通过在线光谱仪的算法软件对每个铝土矿的光谱数据进行计算，并将铝土矿的光谱曲线与水的光谱曲线进行比对，并设置一个第三预设值，将与水的光谱曲线的相似度大于或等于第三预设值的铝土矿标记为尾矿，将与水的光谱曲线的相似度小于第三预设值的铝土矿标记为精矿。
在一实施例中，相似度的第三预设值可以根据铝土矿的品种设置，一般可以设置为80％～95％，示例性地，可以是80％、81％、82％、83％、84％、85％、86％、87％、88％、89％、90％、91％、92％、93％、94％或95％，根据实际需要设置即可。
可选地，根据水的吸收峰特点，可以选用780nm到2100nm波段的在线光谱仪，并将在线光谱仪的积分时间设置为100ms。
可选地，在再一个实施例中，可以通过获取铝土矿表面的水分值的方法识别铝土矿精矿和尾矿，步骤为：
S241、通过红外水分检测仪对铝土矿表面的水分进行检测；
水分子不是静止的，当遇到特定的能量带时，它们会振动，两个氢原子与氧原子的键会伸展、收缩、或以其它形态扭曲。在整个光谱的不同部位，有一些吸收波段十分强烈，有一些十分微弱。在光谱的近红外部位，对于水分子的吸收特别强烈，水中的氢-氧键会吸收特定波长的近红外线(特定波长为1940nm)，在特定波长下，所反射回去的近红外线能量和物料中水分子吸收的近红外线能量成反比。红外水分检测仪可以利用这一特点，根据能量的损失量计算出被测铝土矿表面的含水率。
在一实施例中，由于铝土矿吸水需要一定的时间，因此，在通过红外水分检测仪对铝土矿表面的水分进行检测之前，应该保证铝土矿的吸水时间达到第二预设时间，第二预设时间是使精矿能够完全将水吸收的时间，避免由于吸水时间短而影响铝土矿的识别精度。
可选地，第二预设时间可以根据铝土矿的品种和铝土矿精矿的吸水性设置，可以为1s～60s，示例性地，可以为1s、10s、20s、30s、40s、50s或60s，根据实际情况设置即可。
S242、获取铝土矿表面的含水率，将含水率大于含水率阈值的铝土矿标记为尾矿，将含水率小于或等于含水率阈值的铝土矿标记为精矿。
由于尾矿几乎不吸水，因此，尾矿表面有完整的水膜，而精矿吸水性强，因此精矿表面几乎没有水分，在通过红外水分检测仪进行识别时，其损失的能量会有显著的区别，也就是含水量有显著区别。
因此，通过红外水分检测仪对铝土矿表面的水分进行检测，并设置一个含水率阈值，将含水率大于含水率阈值的铝土矿标记为尾矿，将含水率小于或等于含水率阈值的铝土矿标记为精矿。
在一实施例中，含水率阈值可以根据铝土矿的品种设置，一般可以设置为80％～95％，示例性地，可以是80％、81％、82％、83％、84％、85％、86％、87％、88％、89％、90％、91％、92％、93％、94％或95％，根据实际需要设置即可。
可选地，在其它实施例中，可以通过获取铝土矿表面的红外成像图谱的方法识别铝土矿精矿和尾矿，步骤为：
S251、在铝土矿表面吸收水的时间达到第一预设时间时，吹干铝土矿表面的水分。
由于铝土矿吸水需要一定的时间，因此，为了保证铝土矿的内部吸水完全，可以将铝土矿吸收完水分的时间设置为第一预设时间，保证铝土矿吸水时间达到第一预设之间后再将铝土矿表面的水分吹干。
为了避免铝土矿表面没被吸收的水分对铝土矿的含水率进行影响，需要将铝土矿表面吹干，去除表面的水分。可选地，可以通过风刀对铝土矿表面的水分进行风力吹干。
S252、加热铝土矿，对吹干后的铝土矿进行加热；
在一实施例中，可以采用微波加热的方法对铝土矿进行加热，由于尾矿几乎不吸水，因此尾矿内部几乎没有水，而精矿吸水性强，因此，精矿内部水分较多。已知水的比热容为4.2KJ/kg ℃，而铝土矿石的比热容在0.75-1.2KJ/kg ℃，二者具有较大的差别。由于精矿和尾矿内部是否有水的不同，对铝土矿进行微波加热会导致各自温度上升的幅度不同。内部没有水分的尾矿，温度上升的幅度要大于内部有水分的精矿。
可选地，在对铝土矿进行微波加热时，可以将铝土矿设置在舱体内，并在舱体内设置防止微波泄露的抑制器，一方面，能够防止微波泄露，保证微波的加热效率；另一方面，能够避免有害的微波污染环境。
S253、通过红外成像仪获取加热后的铝土矿的红外成像图谱，将温度差大于第四预设值的铝土矿标记为尾矿，将温度差小于或等于第四预设值的所述铝土矿标记为精矿。
在红外成像仪的监测下，精矿表面由于没有水，水都渗入内部，进行微波加热后，表面温度会上升较少，尾矿则因为内部没有水，温度上升的更多。
红外成像仪根据监测物体发射的红外线，便可绘制此物体的表面温度图谱。由此，红外成像仪可以将不同的温度对应不同的颜色形成红外成像图谱，然后通过服务器里面的算法软件对红外成像图谱进行分析，将温度差大于第四预设值的铝土矿标记为尾矿，对温度差小于或等于第四预设值的铝土矿标记为精矿。
在一实施例中，温度差的第四预设值可以根据铝土矿的品种设置，一般可以设置为80℃～85℃，示例性地，可以是80℃、81℃、82℃、83℃、84℃或85℃，根据实际需要设置即可。
S3、根据参数信息将铝土矿标记为精矿或尾矿。
考虑到铝土矿在进行预抛废时，需要将尾矿剔除，因此，在一个实施例中， 可以通过喷吹的方式改变精矿或尾矿的落点，将精矿分选到第一预设位置，将尾矿分选到第二预设位置，实现精矿和尾矿的分离。对于尾矿，一般进行抛尾作业，将精矿输送给下一道工序。在其他实施例中，还可以通过抓取的方式将精矿或尾矿挑选出来，实现精矿和尾矿的分选。
本申请提供的铝土矿识别方法，利用精矿吸水性强，尾矿几乎不吸水的特点，将铝土矿表面润湿，然后利用吸水后铝土矿精矿和尾矿表面水分的不同对铝土矿表面进行检测，得到铝土矿表面的参数信息，根据所得参数信息实现铝土矿精矿和尾矿的识别，该方法普遍适用于铝土矿，有利于在铝土矿预抛废阶段提高分选精度，进而提高进入后续工序的铝硅比。
本申请还提供一种铝土矿识别装置，采用上述的铝土矿识别方法，能够识别出铝土矿精矿和铝土矿尾矿，识别精度较高。
在一实施例中，如图2和图3所示，上述铝土矿识别装置包括湿润装置100、检测装置300以及识别装置400，其中，湿润装置100被设置为给铝土矿200表面喷淋水，使铝土矿200表面被完全喷湿。检测装置300被设置为对铝土矿200的表面进行检测，可以为近红外相机、高光谱相机、在线光谱仪、红外水分检测仪或红外成像仪等，根据实际需要选择即可。识别装置400与检测装置300数据连接，被设置为获得铝土矿200表面的参数信息，并能够根据上述参数信息识别出精矿和尾矿。
通过上述装置能够实现铝土矿200精矿和尾矿的识别，由于铝土矿200普遍具有精矿吸水性强，尾矿几乎不吸水的特点，因此与相关技术中的铝土矿识别装置相比，上述铝土矿识别装置的适用范围更广，有利于提高铝土矿200的识别精度以及分选精度。
在一实施例中，上述铝土矿识别装置可以包括第一输送装置500，第一输送装置500能够传送铝土矿200，进而实现铝土矿200的自动识别。在一个实施例中，湿润装置100可以设置在第一输送装置500的上方，被设置为将铝土矿200表面喷湿。在另一个实施例中，湿润装置100也可以设置在第一输送装置500的下方。在其他实施例中，湿润装置100还可以同时设置在第一输送装置500的上方和下方，根据实际需要设置即可。
将检测装置300与第一输送装置500的始端的距离定义为预设距离，预设距离为铝土矿200从始端经第二预设时间传送到检测装置300处的距离，以保证铝土矿200具有足够的吸水时间，避免由于铝土矿200吸水时间不够导致精矿表面的水较多而影响识别效果。
可选地，第一输送装置500还可以包括振动给料器510，将铝土矿200放置 在振动给料器510中，在振动给料器510的振动下进行翻滚，并使喷淋装置对振动给料器510内的铝土矿200进行喷淋，有利于保证铝土矿200表面被完全喷湿，进而避免由于铝土矿200表面一处未被喷湿而影响检测结果，保证识别精度。
可选地，如图2和图3所示，上述铝土矿识别装置还可以包括第二输送装置600，第二输送装置600的始端与第一输送装置500的尾端相连，能够将第一输送装置500上的铝土矿200输送至识别装置400进行识别。在一个实施例中，第二输送装置600可以为滑板。在其他实施例中，第二输送装置600也可以为传送带，根据实际需要设置即可。
可选地，上述铝土矿识别装置还可以包括分选装置700，被设置为铝土矿200的预抛废，分选装置700能够将精矿分选到第一预设位置710，将尾矿分选到第二预设位置720，实现精矿和尾矿的分选。在一个实施例中，分选装置700可以为喷嘴。在其他实施例中，分装装置也可以为机械手，根据实际需要选择即可。
在一实施例中，如图3所示，上述铝土矿识别装置还可以包括入料装置800，入料装置800设有入料口，可以通过液压闸板控制入料口的开闭，进而实现对入料量的控制。
通过采用上述铝土矿识别装置，可以实现铝土矿200精矿和尾矿的识别，并将精矿和尾矿分选出来，有利于对尾矿进行抛尾，并将精矿输送给下一道工序。上述铝土矿识别装置适用于铝土矿200的识别，有利于提高铝土矿200的分选精度，进而提高进入下一道工序的铝硅比。
本申请根据铝土矿精矿吸水性强，尾矿几乎不吸水性的特点，采用将铝土矿喷湿后通过检测铝土矿表面的参数信息，获得铝土矿表面与水分相关的参数，进而识别出精矿和尾矿，与相关技术中的铝土矿识别方法相比，能够适用于铝土矿精矿和尾矿的识别，适用范围更广，进而能够在铝土矿预抛废阶段大幅提高铝土矿的分选精度，有效提高进入后续工序的铝土矿的铝硅比。

Claims (15)

  1. 一种铝土矿识别方法，包括：
    将铝土矿表面润湿；
    对润湿后的所述铝土矿的表面进行检测，获得所述铝土矿表面的参数信息；
    根据所述参数信息将所述铝土矿标记为精矿或尾矿。
  2. 根据权利要求1所述的铝土矿识别方法，其中，所述对润湿后的所述铝土矿的表面进行检测，获得所述铝土矿表面的参数信息，包括：
    获取所述铝土矿表面的灰度图像
    或，获取所述铝土矿表面的光谱图像；
    或，获取所述铝土矿表面的光谱曲线；
    或，获取所述铝土矿表面的水分值；
    或，获取所述铝土矿表面的红外成像图谱。
  3. 根据权利要求2所述的铝土矿识别方法，其中，所述获取所述铝土矿表面的灰度图像，包括：
    对润湿后的铝土矿先后采用两种不同波段的近红外光进行照射；
    通过近红外相机对铝土矿被所述近红外光照射的位置进行拍照，以获得两张不同波段的所述近红外光照射下的铝土矿的照片；
    获取两张所述照片的像素灰度差，将所述像素灰度差大于第一预设值的铝土矿标记为尾矿，将所述像素灰度差小于或等于所述第一预设值的铝土矿标记为精矿。
  4. 根据权利要求2所述的铝土矿识别方法，其中，所述获取所述铝土矿表面的光谱图像，包括：
    对润湿后的铝土矿采用卤素灯进行照射；
    通过高光谱相机对铝土矿被所述卤素灯照射的位置进行拍照；
    获取所述铝土矿的光谱图像，将所述光谱图像与水的光谱图像进行比较，将所述铝土矿的光谱图像与所述水的光谱图像的相似度大于或等于第二预设值的铝土矿标记为尾矿，将所述铝土矿的光谱图像与所述水的光谱图像的相似度小于所述第二预设值的铝土矿标记为精矿。
  5. 根据权利要求2所述的铝土矿识别方法，其中，所述获取所述铝土矿表面的光谱曲线，包括：
    通过在线光谱仪对润湿后的铝土矿表面进行拍照；
    获取所述铝土矿表面单点的光谱曲线，将所述铝土矿的光谱曲线与水的光谱曲线进行比较，将所述铝土矿的光谱曲线与所述水的光谱曲线的相似度大于或等于第三预设值的铝土矿标记为尾矿，将所述铝土矿的光谱曲线与所述水的光谱曲线的相似度小于所述第三预设值的铝土矿标记为精矿。
  6. 根据权利要求2所述的铝土矿识别方法，其中，所述获取所述铝土矿表面的水分值，包括：
    通过红外水分检测仪对所述铝土矿表面的水分进行检测；
    获取所述铝土矿表面的含水率，将含水率大于含水率阈值的铝土矿标记为尾矿，将含水率小于或等于所述含水率阈值的铝土矿标记为精矿。
  7. 根据权利要求2所述的铝土矿识别方法，其中，所述获取所述铝土矿表面的红外成像图谱，包括：
    在所述铝土矿表面吸收水的时间达到第一预设时间时，吹干所述铝土矿表面的水分；
    对吹干后的所述铝土矿进行加热；
    通过红外成像仪获取加热后的所述铝土矿的红外成像图谱，将温度差大于第四预设值的所述铝土矿标记为尾矿，将温度差小于或等于所述第四预设值的所述铝土矿标记为精矿。
  8. 根据权利要求3-5任一项所述的铝土矿识别方法，还包括：
    在所述铝土矿表面吸收水的时间达到第二预设时间时，对所述铝土矿拍照。
  9. 根据权利要求8所述的铝土矿识别方法，其中，所述第二预设时间为1s-60s。
  10. 根据权利要求1所述的铝土矿识别方法，其中，在对所述铝土矿进行润湿时，振动所述铝土矿使所述铝土矿翻滚。
  11. 根据权利要求1-7任一项所述的铝土矿识别方法，还包括：
    对所述精矿和所述尾矿进行分选，采用喷吹所述精矿或所述尾矿的方式改变所述精矿或所述尾矿的落点，以将所述精矿分选至第一预设位置，以及将所述尾矿分选至第二预设位置。
  12. 一种铝土矿识别装置，采用权利要求1-11任一项所述的铝土矿识别方法，包括：
    湿润装置，被设置为将铝土矿表面均匀覆盖预设量的水，以将铝土矿表面润湿；
    检测装置，被设置为对湿润后的所述铝土矿的表面进行检测；
    识别装置，所述识别装置与所述检测装置数据连接，被设置为获取所述铝土矿表面的参数信息并根据所述参数信息将所述铝土矿标记为精矿或尾矿。
  13. 根据权利要求12所述的铝土矿识别装置，还包括第一输送装置，所述湿润装置设置在所述第一输送装置的上方和下方中的至少之一，对所述第一输送装置内的所述铝土矿进行喷淋。
  14. 根据权利要求13所述的铝土矿识别装置，还包括第二输送装置，所述第二输送装置被设置为将所述第一输送装置输出的铝土矿输送至识别装置。
  15. 根据权利要求14所述的铝土矿识别装置，还包括分选装置，所述分选装置设置在所述第二输送装置的尾端，被设置为将所述精矿分选至第一预设位置，以及将所述尾矿分选至第二预设位置。
PCT/CN2022/116015 2022-04-11 2022-08-31 一种铝土矿识别方法及装置 WO2023197510A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210371175.8 2022-04-11
CN202210371175.8A CN114433509B (zh) 2022-04-11 2022-04-11 一种铝土矿识别方法及装置

Publications (1)

Publication Number Publication Date
WO2023197510A1 true WO2023197510A1 (zh) 2023-10-19

Family

ID=81359116

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/116015 WO2023197510A1 (zh) 2022-04-11 2022-08-31 一种铝土矿识别方法及装置

Country Status (2)

Country Link
CN (1) CN114433509B (zh)
WO (1) WO2023197510A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114433509B (zh) * 2022-04-11 2022-08-16 天津美腾科技股份有限公司 一种铝土矿识别方法及装置
CN115672780B (zh) * 2022-11-01 2024-05-03 山东黄金矿业科技有限公司选冶实验室分公司 一种入磨前矿石品位预富集方法及预富集系统
CN116871177B (zh) * 2023-09-05 2023-11-17 国擎(山东)信息科技有限公司 一种基于多光谱技术的高岭土原矿分选方法及系统

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5396260A (en) * 1992-12-22 1995-03-07 The Center For Innovative Technology Video instrumentation for the analysis of mineral content in ores and coal
CN1830574A (zh) * 2006-04-12 2006-09-13 中南大学 一种铝土矿铝硅矿物分离方法
CN102256712A (zh) * 2008-12-19 2011-11-23 Omya发展股份公司 通过x射线拣选从含碳酸钙的岩石中分离矿物杂质的方法
CN204866817U (zh) * 2015-05-22 2015-12-16 深圳好朋友信息科技有限公司 钨矿矿石分选机内置成像背景保护装置
CN111921696A (zh) * 2020-07-02 2020-11-13 中国铝业股份有限公司 一种铝土矿中的多种有价矿物综合回收方法
CN111957596A (zh) * 2020-07-15 2020-11-20 郑鸿 一种基于人工智能图像识别的选煤系统
CN112221657A (zh) * 2020-09-03 2021-01-15 湖北杉树垭矿业有限公司 磷矿光电选矿分选工艺
CN114433509A (zh) * 2022-04-11 2022-05-06 天津美腾科技股份有限公司 一种铝土矿识别方法及装置

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100595575C (zh) * 2007-11-19 2010-03-24 南京国晟科技有限公司 散射式矿石成分实时在线检测装置
CN102128900A (zh) * 2010-12-15 2011-07-20 中国铝业股份有限公司 一种检测铝土矿成份的方法
CN103769375A (zh) * 2013-04-23 2014-05-07 湖南久泰冶金科技有限公司 一种用于矿石物料分选装置
CN203556583U (zh) * 2013-11-08 2014-04-23 黄山恒源石英材料有限公司 一种矿石分选设备
CN106203336A (zh) * 2016-07-11 2016-12-07 陕西科技大学 一种基于灰度图像标记的矿井巷道灯识别方法
CN110211107A (zh) * 2019-05-28 2019-09-06 太原理工大学 一种基于双波段红外图像的矿用胶带损伤检测方法
CN110302978B (zh) * 2019-07-22 2024-05-17 湖南金石分选智能科技有限公司 一种模块化固体物料分选机
CN112371559A (zh) * 2020-08-31 2021-02-19 江苏旷博智能技术有限公司 矸石识别方法和矸石自动分离系统

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5396260A (en) * 1992-12-22 1995-03-07 The Center For Innovative Technology Video instrumentation for the analysis of mineral content in ores and coal
CN1830574A (zh) * 2006-04-12 2006-09-13 中南大学 一种铝土矿铝硅矿物分离方法
CN102256712A (zh) * 2008-12-19 2011-11-23 Omya发展股份公司 通过x射线拣选从含碳酸钙的岩石中分离矿物杂质的方法
CN204866817U (zh) * 2015-05-22 2015-12-16 深圳好朋友信息科技有限公司 钨矿矿石分选机内置成像背景保护装置
CN111921696A (zh) * 2020-07-02 2020-11-13 中国铝业股份有限公司 一种铝土矿中的多种有价矿物综合回收方法
CN111957596A (zh) * 2020-07-15 2020-11-20 郑鸿 一种基于人工智能图像识别的选煤系统
CN112221657A (zh) * 2020-09-03 2021-01-15 湖北杉树垭矿业有限公司 磷矿光电选矿分选工艺
CN114433509A (zh) * 2022-04-11 2022-05-06 天津美腾科技股份有限公司 一种铝土矿识别方法及装置

Also Published As

Publication number Publication date
CN114433509A (zh) 2022-05-06
CN114433509B (zh) 2022-08-16

Similar Documents

Publication Publication Date Title
WO2023197510A1 (zh) 一种铝土矿识别方法及装置
US7449655B2 (en) Apparatus for, and method of, classifying objects in a waste stream
Hu et al. Multispectral imaging: A new solution for identification of coal and gangue
US8875901B2 (en) Sorting mined material on the basis of two or more properties of the material
JP6152845B2 (ja) 光学式粒状物選別機
EP3801934B1 (en) Process and system for in-line inspection of product stream for detection of foreign objects
CN114600169A (zh) 用于堆料分选的神经网络
CN104198497B (zh) 一种基于视觉显著图和支持向量机的表面缺陷检测方法
Wang et al. License plate detection using gradient information and cascade detectors
CN105479014A (zh) 自动打标检测装置
Diaz-Romero et al. Deep learning computer vision for the separation of Cast-and Wrought-Aluminum scrap
CN104634792A (zh) Pss良率在线监测系统及其检测方法
CN110665848A (zh) 基于ccd相机与高光谱相机检测的建筑固废分拣系统
CN113552028B (zh) 混合料视觉分析装置、混合料粒度组成检测方法及系统
CN101726492A (zh) 高光谱食品在线实时检测装置及其工作方法
CN205464821U (zh) 自动打标检测装置
Tong Technical amenability study of laboratory-scale sensor-based ore sorting on a Mississippi Valley type lead-zinc ore
Moirogiorgou et al. Intelligent robotic system for urban waste recycling
CN107944394A (zh) 一种对传输带上物料进行检测的视频分析方法及系统
CN110773455A (zh) 一种基于红外视觉的垃圾智能分拣装置及方法
Picon et al. Automation of waste recycling using hyperspectral image analysis
US20220355342A1 (en) Sorting of contaminants
Bobulski et al. The triple histogram method for waste classification
WO2023080385A1 (ko) 테라헤르츠 기술과 초분광 기술을 적용한 다기능 복합 이물질 검사 장치
US20230044783A1 (en) Metal separation in a scrap yard

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22937148

Country of ref document: EP

Kind code of ref document: A1