A particle size detection device and method for simulating the motion state of metallurgical particles

Abstract

The present invention discloses a particle size detection device and method for simulating the motion state of metallurgical particles, and belongs to the technical field of material particle detection in the sintering process. The present invention includes an image acquisition system, four conveyor belts connected in sequence, a laser rangefinder, and an image detection and material particle size calculation system. An image acquisition system is provided on the outside of each conveyor belt for collecting pictures of the material in an inclined surface, a plane, and a falling state, and preliminarily obtaining the particle size and particle size distribution by using an image processing method. In combination with the detection value of the laser rangefinder, a two-dimensional plane-space volume conversion model is constructed to obtain the final particle size and distribution. The present invention can reproduce the material motion state of the production site under laboratory conditions, and accurately detect the particle size and distribution of the material, thereby reducing the number of experiments on the production site, improving the safety of the experiment and the reliability of the detection results, and ensuring the efficient deployment of the detection system.

Description

Technical Field

[0001] The present invention belongs to the technical field of material particle detection in a sintering process, and more specifically, relates to a device and a method for simulating the real motion state of metallurgical particles and realizing particle size detection of particles. Background Art

[0002] In the field of iron and steel metallurgy, the sintering process is an indispensable link in the long process. The sintering process produces a large number of granular materials, with various types, shapes and particle sizes. In the preparation of sintering raw materials, granular materials include iron ore, solvents, coke powder, return ore, etc.; in the material preparation process, they are divided into mixed materials and green balls; after sintering, the granular materials include finished sintered ore and pellets. Granular materials exist throughout the entire sintering process, and their particle size can directly or indirectly reflect the normal operation of the sintering process and the quality of related products, and even affect the smooth operation of subsequent blast furnace production.

[0003] Therefore, particle size detection and analysis of granular materials during the sintering process is crucial for improving production and enhancing product quality. Research on this topic has long been a hot topic for researchers. However, due to the complex production environment, characterized by high temperatures, high pressures, confined spaces, and high levels of dust and dirt, the safety of researchers cannot be fully guaranteed. Currently, researchers studying particle size detection typically collect particles from the production site and bring them to the laboratory for imaging and sampling. While this approach ensures research safety, it lacks the potential for interference from the production environment. For example, laboratories cannot capture images of high-speed moving materials, simulate material motion in various scenarios such as inclined surfaces, flat surfaces, and cross-sections, or simulate the light and shadow effects of the production site. Consequently, many particle size detection research results are limited to simple laboratory environments and can frequently encounter problems when deployed in production sites. Consequently, research progress has been relatively slow, and many studies have failed to be implemented.

[0004] A search revealed relevant patent documents related to particle size detection of sintered materials. For example, Chinese patent application number 202110275265.2, published on May 10, 2017, is titled "An Online Particle Size Detection Device for Sintered Fuel Coal." The device in this application includes a feeding device; a negative pressure housing with a material inlet, a material outlet, and an air outlet; the feeding device is used to convey material into the material inlet; a rotatable feed roller disposed within the housing and below the material inlet for changing the direction and controlling the feed rate; an air inlet device disposed below the feed roller for blowing air onto the material passing through the feed roller; a camera for recording the distribution of material particles within the housing; and a processing module for capturing images captured by the camera and calculating the material particle size and particle size distribution. The negative pressure housing, the feed roller, and the air inlet keep the material suspended. The camera and processing module then process the suspended material to obtain the desired particle size analysis. This application uses air to blow up the material, causing it to be suspended, and then uses a camera to collect data on the suspended material. This can, to a certain extent, eliminate the problems of material accumulation and overlap. This can simulate the motion state of the particles as much as possible and detect the particle size of the particles in motion. However, photographing the material in a suspended state results in a single state, making it difficult to capture images of the material in more states and difficult to adapt to actual production conditions. At the same time, the material particles are in a suspended state and the particle speed is relatively fast, which places high requirements on the hardware equipment of the camera and the image acquisition cost. Summary of the Invention

[0005] 1. Problem to be solved

[0006] In response to the above-mentioned difficulties in the existing technology for detecting the particle size of particles in motion during the sintering process, the present invention provides a particle size detection device and method that simulates the motion state of metallurgical particles. By optimizing the structure of the detection device, a detection environment that is more closely related to the actual production environment can be provided, thereby improving the detection effect and accuracy, which is conducive to accelerating the R&D process, shortening the on-site product deployment cycle, and providing a safe detection environment to ensure the safety of scientific researchers.

[0007] 2. Technical solution

[0008] In order to solve the above problems, the technical solutions adopted by the present invention are as follows:

[0009] The particle size detection device of the present invention, which simulates the motion state of metallurgical particles, includes an image acquisition system, a conveyor belt, a laser rangefinder, and an image detection and material particle size calculation system, wherein:

[0010] The image acquisition system includes a camera and a fill light. The camera is electrically connected to the image detection and material particle size calculation system and is used to transmit the collected particle images to the image detection and material particle size calculation system for processing, and calculate the particle size and particle size distribution of the pellet particles in conjunction with the detection results of the laser rangefinder;

[0011] The conveyor belt comprises a first conveyor belt, a second conveyor belt, a third conveyor belt and a fourth conveyor belt which are sequentially arranged around and connected;

[0012] A camera and a fill light are provided on one side of the first conveyor belt for collecting pictures of particles in a state of rolling down an inclined surface; a camera and a fill light are provided above the second conveyor belt for collecting pictures of particles in a state of flat accumulation; a camera and a fill light are provided on one side of the third conveyor belt for collecting pictures of particles in a state of falling cross section; a laser rangefinder and a fill light are provided on one side of the fourth conveyor belt for collecting particle size information on the side of the particles.

[0013] Furthermore, the first conveyor belt and the fourth conveyor belt are respectively configured as a horizontal section, an inclined section and a horizontal section connected in sequence, and the second conveyor belt and the third conveyor belt are both configured as horizontal sections; the two horizontal sections of the first conveyor belt are respectively perpendicular to the movement directions of the horizontal sections of the second conveyor belt and the fourth conveyor belt, the movement directions of the second conveyor belt and the third conveyor belt are perpendicular, and the movement directions of the horizontal sections of the third conveyor belt and the fourth conveyor belt are perpendicular.

[0014] Furthermore, one of the horizontal sections of the first conveyor belt is located 5 to 10 cm above the second conveyor belt, and the tail end of the horizontal section extends 5 to 15 cm into the side of the second conveyor belt. This design ensures that the pellets transported by the first conveyor belt fall into the middle area of ​​the second conveyor belt and are evenly distributed on the second conveyor belt from the middle area, avoiding material accumulation that affects the camera set above the second conveyor belt from collecting pictures.

[0015] Similarly, one of the horizontal sections of the fourth conveyor belt is located 5 to 10 cm above the first conveyor belt, and the tail end of the horizontal section extends 5 to 15 cm into the side of the first conveyor belt. This arrangement ensures that the pellets transported by the fourth conveyor belt fall into the middle area of ​​the first conveyor belt and are evenly distributed on the first conveyor belt from the middle area, providing a more evenly distributed material state for camera shooting.

[0016] The horizontal tail of the second conveyor belt is flush with the side of the third conveyor belt, and the tail of the third conveyor belt is flush with the side of one of the horizontal sections of the fourth conveyor belt; the inclination angle of the inclined sections of the first conveyor belt and the fourth conveyor belt is 45°.

[0017] Furthermore, the conveyor belts move at a speed of 0.02 to 0.05 m / s. The cameras and fill lights installed on the sides of the first and third conveyor belts, as well as the laser rangefinder and fill light installed on the side of the fourth conveyor belt, are each positioned 8 to 12 cm from the corresponding conveyor belt edge. This effectively prevents laser rangefinder positioning loss and data interference caused by fill lights being too close or too far apart. Furthermore, the camera and fill light installed above the second conveyor belt are positioned 5 to 30 cm above the second conveyor belt. This arrangement allows the camera to clearly capture material features without overexposure due to being too low, or blurry or dark due to being too high.

[0018] Furthermore, the fill light and the camera or laser rangefinder are arranged side by side horizontally and in the same direction, and the distance between the fill light and the camera or laser rangefinder is 8 to 12 cm, thereby ensuring that the camera will not cause the collected pictures to be overexposed or too dark due to the fill light being set too close or too far.

[0019] Furthermore, it also includes a control system, which is electrically connected to the image acquisition system, the conveyor belt, the laser rangefinder and the image detection and material particle size calculation system.

[0020] The present invention provides a method for detecting particle size in a simulated particle metallurgical motion state, using the above-mentioned detection device for detection, and specifically includes the following steps:

[0021] Step 1: Start the conveyor belt to circulate and transport the material. When the material passes through the first conveyor belt, the second conveyor belt, and the third conveyor belt, the corresponding image acquisition system collects images of all pellet particles, and collects images of the pellet particles in the inclined rolling state, the plane stacking state, and the cross-section falling state respectively. The images are transmitted to the image detection and material particle size calculation system, and then the particle sizes of all pellet particles in the images in the three states are calculated respectively, and the particle size distribution data of the pellet particles are obtained. The three sets of cameras can obtain three sets of particle size distribution data of the pellet particles. Finally, the three sets of particle size distribution data are integrated to obtain the initial particle size distribution data of the pellet particles;

[0022] Step 2: When the material passes through the fourth conveyor belt, the laser rangefinder measures the pellet particles from the side and collects the point distance data on the side of the pellet particles;

[0023] Step 3: Based on the point distance data on the side of the pellet particles collected in step 2, a two-dimensional plane-space volume conversion model is used to process the data to calculate the particle size and particle size distribution of the pellet particles to be tested, and the initial particle size distribution data of the pellet particles obtained in step 2 is corrected to obtain the final particle size distribution of the pellet particles;

[0024] Step 4: According to the final pellet particle size obtained in step 3, adjust the material feeding amount, conveyor belt inclination and movement speed, and fill light irradiation angle and brightness;

[0025] Step 5: Repeat steps 1 to 4 to obtain the pellet particle size and particle size distribution data at the next round of testing time.

[0026] Furthermore, in step 1, the image processing method includes:

[0027] Step S1, using an image filtering noise reduction algorithm to perform noise reduction on the collected photos;

[0028] Step S2: using an image enhancement algorithm to enhance the feature information of the target object;

[0029] Step S3: Use the intelligent detection and segmentation model trained with the material sample image to identify and segment the pellet particle contour information in three different states, obtain the contour coordinate information of each pellet particle, and preliminarily calculate the particle size of all pellet particles revealed in the image.

[0030] Furthermore, the particle size distribution ratio of the pellets is determined based on the particle size range, specifically following the following rules:

[0031] First, the particle size proportions of the pellet particles in three different states, namely, the particle size of less than 0.5 cm, 0.5-1 cm (including pellets with a diameter of 0.5 cm, excluding pellets with a diameter of 1 cm), 1 cm-1.5 cm (including pellets with a diameter of 1 cm, excluding pellets with a diameter of 1.5 cm), 1.5-2 cm (including pellets with a diameter of 1.5 cm, excluding pellets with a diameter of 2 cm), 2 cm-2.5 cm (including pellets with a diameter of 2 cm, excluding pellets with a diameter of 2.5 cm), 2.5-3 cm (including pellets with a diameter of 2.5 cm, excluding pellets with a diameter of 3 cm), and pellets with a diameter of 3 cm and above were counted respectively.

[0032] Then, the particle size distribution percentages calculated under the three states are compared. When the difference between any two percentages in the corresponding particle size distribution ranges under the three states is within 5%, the average value of the three groups of percentages is taken as the initial particle size distribution data of the pellet particles; otherwise, the particle size distribution percentage under the plane stacking state is taken as the initial particle size distribution data of the pellet particles.

[0033] For example, if the three collected states have particle sizes falling within the 0.5-1cm range of 2% (slope rolling state), 3% (flat accumulation state), and 8% (cross-section falling state), then the proportion of particles falling within the 0.5-1cm range is directly taken as 3%. For another example, if the three collected states have particle sizes falling within the 0.5-1cm range of 3% (slope rolling state), 4% (flat accumulation state), and 5% (cross-section falling state), then the average of the three groups is taken, and the proportion of particles falling within the 0.5-1cm range is taken as 4%. The proportions of the remaining particle size ranges are calculated according to the above rules.

[0034] In step 3, the method for determining the particle size and proportion of pellets based on the data collected by the laser rangefinder (104) includes the following steps:

[0035] Step 1: The collected pellet particle point distances, i.e., pellet particle edge point data and pellet particle center point data, are fitted into an arc using the least squares method;

[0036] Step 2: Use Hough circle transform to fit the arc to obtain the semicircle of the pellet particles in three-dimensional space;

[0037] Step 3: Using the principle of symmetry, fit the shape of the entire pellet in three-dimensional space and calculate the particle size of the pellet;

[0038] Step 4: Count the particle size distribution proportions of all pellets;

[0039] Step 5: Compare the initial pellet size distribution data with the pellet size distribution data obtained in step 4:

[0040] When the difference between the two proportions in the corresponding particle size range between the initial particle size distribution data and the pellet size distribution data obtained in step 4 is within 5%, the average of the two proportions is used as the final pellet particle size distribution data; otherwise, the pellet particle size distribution data obtained in step 4 is used as the final pellet particle size distribution data.

[0041] For example, if the proportion of particles with a diameter within the range of 0.5 to 1 cm in the initial particle size distribution data is 2%, and the proportion of particles with a diameter within the range of 0.5 to 1 cm in the pellet size distribution data obtained in step 4 is 4%, then the proportion of particles with a diameter within the range of 0.5 to 1 cm in the final pellet size distribution data is 3%. For another example, if the proportion of particles with a diameter within the range of 0.5 to 1 cm in the initial particle size distribution data is 2%, and the proportion of particles with a diameter within the range of 0.5 to 1 cm in the pellet size distribution data obtained in step 4 is 8%, then the proportion of particles with a diameter within the range of 0.5 to 1 cm in the final pellet size distribution data is 8%.

[0042] By adopting the above method, the accuracy of pellet particle measurement can be improved, thereby significantly improving the detection accuracy of the device.

[0043] Furthermore, in step 4,

[0044] When the particle size of the pellets detected is less than 0.5cm or greater than 3cm, the system will prompt you to adjust the amount of pellets to be added. The pellets with a particle size less than 0.5cm or greater than 3cm will be removed from the currently added pellets, and other equipment parameters will not be adjusted.

[0045] When the particle size of the pellets being tested is between 0.5cm and 3cm, inclusive, the amount of feed does not need to be adjusted. The camera image acquisition efficiency is then evaluated. When the image acquisition efficiency is greater than or equal to 90%, this set of data is deemed appropriate, and there is no need to adjust the amount of feed and other equipment parameters. When the image acquisition efficiency is less than 90%, this set of data is deemed worthless. The conveyor belt inclination angle and movement speed are reduced by 10% to 15% based on the original inclination angle and movement speed, and the fill light brightness is increased by 5% to 10% based on the original brightness. The above measures are used to improve the accuracy of the test results.

[0046] 3. Beneficial effects

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] The present invention simulates the motion of metallurgical particles. By employing multiple conveyor belts and optimizing the layout of each belt, combined with an image acquisition system and a laser rangefinder, the device can simultaneously detect particle size on three surfaces: particles rolling down an inclined surface, particles moving on a plane, and particles falling from a cross-section. A more accurate particle size is calculated using a "two-dimensional plane-to-space volume" conversion model. Because the present invention utilizes multi-angle photography for sampling, it reduces sample contingency and prevents detection system downtime due to individual equipment failures, thereby ensuring smooth production.

[0049] The particle size detection method of the present invention that simulates the metallurgical movement state of particles builds a material image acquisition environment that fits the actual situation of the sintering production site by simulating the metallurgical movement process of the particles, thereby realizing the identification and detection of the particle size of the metallurgical granular material, significantly improving the detection accuracy, and eliminating the need for scientific researchers to go to the site in person to take samples during detection, solving the problems of high labor intensity of manual particle size screening and delayed detection information feedback, while also ensuring the personal safety of scientific researchers. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 Schematic diagram of the structure of the particle size detection device for simulating the motion state of metallurgical particles in the present invention;

[0051] In the picture:

[0052] 101. First camera; 102. Second camera; 103. Third camera; 104. Laser rangefinder.

[0053] 201, the first fill light; 202, the second fill light; 203, the third fill light; 204, the fourth fill light.

[0054] 301, first conveyor belt; 302, second conveyor belt; 303, third conveyor belt; 304, fourth conveyor belt. DETAILED DESCRIPTION

[0055] It should be noted that it should be understood that the terms "center", "longitudinal", "lateral", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, they cannot be understood as limiting the protection content of the present invention.

[0056] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0057] The present invention provides a particle size detection device for simulating the motion state of metallurgical particles, such as Figure 1 As shown, it includes an image acquisition system, a conveyor belt, a laser rangefinder 104) and an image detection and material particle size calculation system, wherein:

[0058] The image acquisition system includes a camera and a fill light. The camera model is MV-EM510C. The present invention is equipped with a first camera 101, a second camera 102, and a third camera 103. All cameras are electrically connected to the image detection and material particle size calculation system. The captured relevant images can be transmitted to the image detection and material particle size calculation system in real time for processing. In combination with the detection results of the laser rangefinder 104, the particle size is calculated, significantly improving the detection accuracy. The fill light adopts the Hikvision LED strobe fill light model: CXBG-1-PS-DS-TL2000A-L1. The present invention is equipped with a first fill light 201, a second fill light 202, a third fill light 203, and a fourth fill light 204.

[0059] Specifically, the conveyor belt described in the present invention includes a first conveyor belt 301, a second conveyor belt 302, a third conveyor belt 303 and a fourth conveyor belt 304 that are connected in sequence. The four conveyor belts are powered by rollers, and the brackets connected to the rollers provide support for the entire conveyor belt. For the sake of detection accuracy and safety, the speeds of the four conveyor belts are the same, and the speeds are all set to 0.02~0.05m / s.

[0060] The first conveyor belt 301 and the fourth conveyor belt 304 are respectively configured as a horizontal section, an inclined section and a horizontal section connected in sequence, and the second conveyor belt 302 and the third conveyor belt 303 are both configured as horizontal sections; the two horizontal sections of the first conveyor belt 301 are respectively perpendicular to the movement directions of the horizontal sections of the second conveyor belt 302 and the fourth conveyor belt 304, the movement directions of the second conveyor belt 302 and the third conveyor belt 303 are perpendicular, and the movement directions of the third conveyor belt 303 and the horizontal sections of the fourth conveyor belt are perpendicular.

[0061] by Figure 1 The following is an example of the orientation of the conveyor belts 301. This orientation is not restrictive and is provided for convenience in describing the placement of the conveyor belts and the camera and fill light locations. A horizontal section of the first conveyor belt 301 is positioned 5 to 10 cm above the second conveyor belt 302, with the tail end of the horizontal section extending 15 to 15 cm from the side of the second conveyor belt 302. A horizontal section of the fourth conveyor belt 304 is positioned 5 to 10 cm above the first conveyor belt 301, with the tail end of the horizontal section extending 5 to 15 cm from the side of the first conveyor belt 301. The tail end of the second conveyor belt 302 is flush with the side of the third conveyor belt 303, and the tail end of the third conveyor belt 303 is flush with the side of one of the horizontal sections of the fourth conveyor belt 304. The inclined sections of the first and fourth conveyor belts 301 and 304 have an inclination angle of 45°.

[0062] Among them, a first camera 101 and a first fill light 201 are arranged side by side in the same direction 8 to 12 cm to the right side of the movement direction of the first conveyor belt 301 (that is, the outside of the first conveyor belt 301) for collecting pictures of particles rolling down the inclined surface. A second camera 102 and a second fill light 202 are arranged side by side in the same direction 5 to 30 cm above the right side of the moving direction of the second conveyor belt 302, which are used to collect pictures of the particles in a flat stacking state; a third camera 103 and a third fill light 203 are arranged side by side in the same direction 10 cm to the right side of the moving direction of the third conveyor belt 303, which are used to collect pictures of the particles in a falling cross-section state; a laser rangefinder 104 and a fourth fill light 204 are arranged side by side in the same direction 8 to 12 cm to the right side of the moving direction of the fourth conveyor belt 304, which are used to collect particle side size information of the particles. The above-mentioned positions of the conveyor belts, the conveyor belts, the fill lights and the cameras (or laser rangefinders 104) can maximize the recognition clarity and ensure the safety of the equipment. The distance between the fill light and the camera or laser rangefinder 104 is 10 cm. This designed distance can make the materials be identified more clearly.

[0063] At the same time, the first conveyor belt 301, the third conveyor belt 303, and the fourth conveyor belt 304 are designed to be horizontal with the fill light and the camera (or laser rangefinder 104), so that small items are not missed. The second conveyor belt 302 is arranged perpendicular to the fill light and the camera, ensuring comprehensive recognition from a bird's-eye view.

[0064] Furthermore, the detection device of the present invention is further optimized to include a control system. The control system utilizes a conventional automated electronic control system. The control system of the present invention is electrically connected to the image acquisition system, the conveyor belt, the laser rangefinder 104, and the image detection and material particle size calculation system, respectively, to achieve automated detection. Compared to existing manual sampling and screening methods for particle size detection, this effectively solves the problems of high labor intensity, poor sampling representativeness, and the inability to achieve 24-hour monitoring. Compared to several other current particle size detection methods, such as sedimentation, electric induction, electromagnetic wave scattering, and laser diffraction, this method is more suitable for detecting material particles in the sintering process.

[0065] The present invention also provides a method for detecting particle size based on the above detection device, comprising the following steps:

[0066] Step 1: Start the conveyor belt to circulate and transport the material. When the material passes through the first conveyor belt 301, the second conveyor belt 302, and the third conveyor belt 303, the corresponding image acquisition system collects images of all pellet particles, and collects images of the pellet particles in the inclined rolling state, the plane stacking state, and the cross-section falling state respectively. The images are transmitted to the image detection and material particle size calculation system, and the particle sizes of all pellet particles in the images in the three states are calculated respectively, and the particle size distribution data of the pellet particles are obtained, thereby obtaining the initial particle size distribution data of the pellet particles;

[0067] Specifically, the image processing method includes:

[0068] Step S1: Using an image filtering and noise reduction algorithm to perform noise reduction on the collected photos to reduce photoelectric interference during the collection process;

[0069] Step S2: using an image enhancement algorithm to enhance the feature information of the target object and improve the contrast between the foreground and background;

[0070] Step S3: Use the intelligent detection and segmentation model trained with the material sample image to identify and segment the pellet particle contour information in three different states, obtain the contour coordinate information of each pellet particle, and preliminarily calculate the particle size of all pellet particles revealed in the image.

[0071] At the same time, in step 1, the particle size distribution ratio of the pellets is determined according to the particle size range of the pellets, and the specific rules are as follows:

[0072] First, the particle size proportions of the pellet particles in three different states, namely, the particle size of less than 0.5 cm, 0.5-1 cm (including pellets with a diameter of 0.5 cm, excluding pellets with a diameter of 1 cm), 1 cm-1.5 cm (including pellets with a diameter of 1 cm, excluding pellets with a diameter of 1.5 cm), 1.5-2 cm (including pellets with a diameter of 1.5 cm, excluding pellets with a diameter of 2 cm), 2 cm-2.5 cm (including pellets with a diameter of 2 cm, excluding pellets with a diameter of 2.5 cm), 2.5-3 cm (including pellets with a diameter of 2.5 cm, excluding pellets with a diameter of 3 cm), and pellets with a diameter of 3 cm and above were counted respectively.

[0073] Then, the particle size distribution percentages calculated under the three states are compared. When the difference between any two percentages in the corresponding particle size distribution ranges under the three states is within 5%, the average value of the three groups of percentages is taken as the initial particle size distribution data of the pellet particles; otherwise, the particle size distribution percentage under the plane stacking state is taken as the initial particle size distribution data of the pellet particles.

[0074] Step 2: When the material passes through the fourth conveyor belt 304, the laser rangefinder 104 measures the pellets from the side and collects point distance data on the side of the pellets;

[0075] Step 3: Based on the point distance data collected in step 2, a two-dimensional plane-space volume conversion model is used to process the data to calculate the particle size and particle size distribution of the pellet particles to be tested, and the initial particle size distribution data of the pellet particles obtained in step 2 is corrected to obtain the final particle size distribution of the pellet particles;

[0076] Specifically, in step 3, the method for determining the particle size and proportion of pellets based on the data collected by the laser rangefinder (104) includes the following steps:

[0077] Step 1: The collected pellet particle point distances, i.e., pellet particle edge point data and pellet particle center point data, are fitted into an arc using the least squares method;

[0078] Step 2: Use Hough circle transform to fit the arc to obtain the semicircle of the pellet particles in three-dimensional space;

[0079] Step 3: Using the principle of symmetry, fit the shape of the entire pellet in three-dimensional space and calculate the particle size of the pellet;

[0080] Step 4: Count the particle size distribution proportions of all pellets;

[0081] Step 5: Compare the initial pellet size distribution data with the pellet size distribution data obtained in step 4:

[0082] When the difference between the two proportions in the corresponding particle size range between the initial particle size distribution data and the pellet size distribution data obtained in step 4 is within 5%, the average of the two proportions is used as the final pellet particle size distribution data; otherwise, the pellet particle size distribution data obtained in step 4 is used as the final pellet particle size distribution data.

[0083] Step 4: According to the final pellet particle size obtained in step 3, adjust the material feeding amount, conveyor belt inclination and movement speed, and fill light brightness;

[0084] When the particle size of the pellets detected is less than 0.5cm or greater than 3cm, the system will prompt you to adjust the amount of pellets to be added. The pellets with a particle size less than 0.5cm or greater than 3cm will be removed from the currently added pellets, and other equipment parameters will not be adjusted.

[0085] When the particle size of the pellets being tested is between 0.5cm and 3cm, inclusive, the amount of feed does not need to be adjusted. The camera image acquisition efficiency is evaluated. When the image acquisition efficiency is greater than or equal to 90%, this set of data is judged to be appropriate, and there is no need to adjust the amount of feed and other equipment parameters. When the image acquisition efficiency is less than 90%, this set of data is judged to be worthless. The inclination angle and movement speed of the conveyor belt are reduced by 10% to 15% based on the original inclination angle and movement speed, and the brightness of the fill light is increased by 5 to 10% based on the original brightness. The above measures are used to improve the accuracy of the test results.

[0086] Step 5: loop step 1 to step 4 to obtain the particle size and particle size distribution data of the pellet particles at the next round of detection time.

[0087] Furthermore, it is worth noting that the present invention utilizes a laser rangefinder 104 with a camera function to measure the distance between the laser rangefinder 104 and the pellet surface from the side where the pellet falls. The maximum value of the measured values ​​is taken as the center point data, and the minimum value as the edge point data. A least-squares method is used to fit the resulting circular arc curve, which is then fitted using the Hough circle algorithm to obtain spatial data of the material in three-dimensional space. This allows for precise processing and construction of specific pellet materials, and is used to verify and correct the accuracy of the pellet size data collected and calculated using the first, second, and third cameras 101, 102, and 103. Because the images captured by these three cameras are planar data, they only reflect the size of the pellet projected at a certain angle and cannot accurately represent the spatial size of the object. Therefore, the particle size is supplemented by the side information detected by the laser rangefinder 104, achieving a two-dimensional to three-dimensional conversion. The conversion process from a two-dimensional plane to a three-dimensional space is mainly completed by a camera in conjunction with a laser rangefinder 104. The cooperation method is based on the particle size of the pellets detected by the front image, and then the laser rangefinder 104 is used to perform multiple and intensive measurements on the side of the pellets to collect the particle size information of the pellets on the side, which can provide users with more accurate material information in three-dimensional space.

[0088] The present invention will be further described below with reference to specific embodiments.

[0089] Example 1

[0090] The particle size detection method for simulating the particle metallurgical motion state of this embodiment includes the following steps:

[0091] (1) equipped with the detection device of the present invention;

[0092] 1. Layout of conveyor belt;

[0093] The diameter of the rollers of each conveyor belt is 10 cm, and the speed of the conveyor belt is 0.05 m / s.

[0094] The width of the first, second, and third conveyor belts 301, 302, and 303 is 40 cm, while the width of the fourth conveyor belt 304 is 30 cm. The first conveyor belt 301 is 90 cm long when viewed from above, with an inclined section at a 45-degree angle. One horizontal section of the first conveyor belt 301 is located 5 cm above the second conveyor belt 302 and 10 cm beyond the sides of the second conveyor belt 302. The second conveyor belt 302 is 155 cm long when viewed from above, located 5 cm above the third conveyor belt 303 and 10 cm beyond the sides of the third conveyor belt 303. The third conveyor belt 303 is 95 cm long when viewed from above, located 5 cm above the horizontal section of the fourth conveyor belt 304 and 10 cm beyond the sides of the horizontal section. The fourth conveyor belt 304 is 175 cm long when viewed from above, with an inclined section at a 45-degree angle. It is located 5 cm above another horizontal section of the first conveyor belt 301 and 10 cm beyond the sides of another horizontal section of the first conveyor belt 301.

[0095] 2. Layout of the image acquisition system and laser rangefinder 104;

[0096] A first camera 101 and a first fill light 201 are arranged side by side and in the same direction 10 cm outside the first conveyor belt 301 , and the distance between the first camera 101 and the first fill light 201 is 10 cm.

[0097] A second camera 102 and a second fill light 202 are arranged side by side and in the same direction 20 cm above the outer side of the second conveyor belt 302 , and the distance between the second camera 102 and the second fill light 202 is 10 cm.

[0098] A third camera 103 and a third fill light 203 are arranged side by side and in the same direction 10 cm outside the third conveyor belt 303 , and the distance between the third camera 103 and the third fill light 203 is 10 cm.

[0099] A laser rangefinder 104 and a fourth fill light 204 are arranged side by side and in the same direction 10 cm outside the fourth conveyor belt 304 , and the distance between the laser rangefinder 104 and the fourth fill light 204 is 10 cm.

[0100] 3. Detection system connection

[0101] The camera and the laser rangefinder 104 are electrically connected to the image detection and material particle size calculation system, and the control system is electrically connected to the image acquisition system, the conveyor belt roller power equipment, the laser rangefinder 104 and the image detection and material particle size calculation system respectively.

[0102] (2) Fabric and testing

[0103] 100 pellets of mineral material with a diameter of 2 cm to be tested are evenly laid on the horizontal section of the first conveyor belt 301, and the control system and each conveyor belt, image acquisition system, laser rangefinder 104, and image detection material particle size calculation system are turned on and circulated. During this process, pictures of the material particles in the inclined rolling state, flat stacking state, and cross-section falling state are collected, as well as the detection data of the laser rangefinder 104, namely the distance between the edge point and the center point of the pellets.

[0104] (3) Image processing to obtain particle size and distribution;

[0105] The image processing method of the present invention is used to process the image, and the particle size and particle size distribution of the pellets to be measured in the two-dimensional plane are obtained as follows:

[0106] 1) When rolling down the inclined plane:

[0107] Particle size: 1 particle with a diameter of 1.4 cm, 5 particles with a diameter of 1.8 cm, 93 particles with a diameter of 2 cm, and 1 particle with a diameter of 2.2 cm;

[0108] Particle size distribution: Particles with a diameter of 1cm to 1.5cm account for 1%, particles with a diameter of 1.5 to 2cm account for 98%, and particles with a diameter of 2cm to 2.5cm account for 1%;

[0109] 2) In the plane stacking state:

[0110] Particle size: 2 with a particle size of 1.4 cm, 6 with a particle size of 1.8 cm, 91 with a particle size of 2 cm, and 1 with a particle size of 2.1 cm;

[0111] Particle size distribution: Particles with a diameter of 1cm to 1.5cm account for 2%, particles with a diameter of 1.5 to 2cm account for 97%, and particles with a diameter of 2cm to 2.5cm account for 1%;

[0112] 3) When the cross section falls:

[0113] Particle size: 1 particle with a diameter of 1.3 cm, 4 particles with a diameter of 1.7 cm, 94 particles with a diameter of 2 cm, and 1 particle with a diameter of 2.1 cm

[0114] Particle size distribution: Particles with a diameter of 1cm to 1.5cm account for 1%, particles with a diameter of 1.5 to 2cm account for 98%, and particles with a diameter of 2cm to 2.5cm account for 1%;

[0115] The calculated initial particle size distribution data of the pellet particles are as follows:

[0116] Particle size below 0.5cm accounts for 0%, 0.5-1cm accounts for 0%, 1cm-1.5cm accounts for 1.33%, particle size between 1.5-2cm accounts for 97.77%, 2cm-2.5cm accounts for 1%, 2.5-3cm accounts for 0%, and above 3cm accounts for 0%.

[0117] Combined with the detection data of the laser rangefinder 104, the two-dimensional plane-space volume conversion fitting model is used for processing, and the particle size distribution is obtained as follows: 1 particle with a particle size of 1.45 cm (i.e., 1%), 98 particles with a particle size of 2 cm (i.e., 98%), and 1 particle with a particle size of 2.1 cm (i.e., 1%). The corresponding pellet particle size distribution data is:

[0118] Particle size below 0.5cm accounts for 0%, 0.5-1cm accounts for 0%, 1cm-1.5cm accounts for 1%, particle size between 1.5-2cm accounts for 98%, 2cm-2.5cm accounts for 1%, 2.5-3cm accounts for 0%, and 3cm and above accounts for 0%.

[0119] The final pellet particle size distribution data is as follows:

[0120] Particle sizes below 0.5 cm account for 0%, 0.5-1 cm account for 0%, 1 cm-1.5 cm account for 1.2%, particle sizes between 1.5 and 2 cm account for 97.8%, particles larger than 2 cm account for 1%, 2.5-3 cm account for 0%, and 3 cm and above account for 0%.

[0121] The above results are consistent with the results of 100 pellets with a known particle size of 2 cm used in experimental testing, and the detection method of the present invention has high accuracy.

[0122] The particle sizes of the pellets tested are all within the range of 1cm to 3cm. There is no need to adjust the feed rate and other equipment parameters. Continue with the currently set feed rate and equipment parameters for the next round of testing.

[0123] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A particle size detection device that simulates the motion state of metallurgical particles, characterized by: It includes an image acquisition system, a conveyor belt, a laser rangefinder (104) and an image detection and material particle size calculation system, wherein: The image acquisition system includes a camera and a fill light, and the camera is electrically connected to the image detection and material particle size calculation system, and is used to transmit the collected particle image to the image detection and material particle size calculation system for processing, and in conjunction with the detection results of the laser rangefinder (104), calculate the particle size and particle size distribution of the pellet particles; The conveyor belt comprises a first conveyor belt (301), a second conveyor belt (302), a third conveyor belt (303) and a fourth conveyor belt (304) which are sequentially arranged around and connected to each other; A camera and a fill light are provided on one side of the first conveyor belt (301) for collecting images of particles in a state of rolling down an inclined surface; a camera and a fill light are provided above the second conveyor belt (302) for collecting images of particles in a state of plane accumulation; a camera and a fill light are provided on one side of the third conveyor belt (303) for collecting images of particles in a state of falling cross-section; a laser rangefinder (104) and a fill light are provided on one side of the fourth conveyor belt (304) for collecting particle size information on the side of the particles; The first conveyor belt (301) and the fourth conveyor belt (304) are respectively configured as a horizontal section, an inclined section and a horizontal section connected in sequence, and the second conveyor belt (302) and the third conveyor belt (303) are both configured as horizontal sections; the two horizontal sections of the first conveyor belt (301) are respectively perpendicular to the movement directions of the horizontal sections of the second conveyor belt (302) and the fourth conveyor belt (304), the movement directions of the second conveyor belt (302) and the third conveyor belt (303) are perpendicular, and the movement directions of the third conveyor belt (303) and the horizontal section of the fourth conveyor belt are perpendicular.

2. The particle size detection device according to claim 1, characterized in that: One of the horizontal sections of the first conveyor belt (301) is located 5 to 10 cm above the second conveyor belt (302), and the tail end of the horizontal section extends into the side of the second conveyor belt (302) by 5 to 15 cm. One of the horizontal sections of the fourth conveyor belt (304) is located 5 to 10 cm above the first conveyor belt (301), and the tail end of the horizontal section extends into the side of the first conveyor belt (301) by 5 to 15 cm. The horizontal tail of the second conveyor belt (302) is flush with the side of the third conveyor belt (303), and the tail of the third conveyor belt (303) is flush with the side of one of the horizontal sections of the fourth conveyor belt (304). The inclination angle of the inclined sections of the first conveyor belt (301) and the fourth conveyor belt (304) is 45°.

3. The particle size detection device according to claim 1, characterized in that: The movement speed of the conveyor belt is 0.02~0.05m / s; the camera and the fill light arranged on one side of the first conveyor belt (301) and the third conveyor belt (303), and the laser rangefinder (104) and the fill light arranged on one side of the fourth conveyor belt (304) are respectively arranged at a distance of 8~12cm from the corresponding conveyor belt side; the camera and the fill light arranged above the second conveyor belt (302) are 15~30cm above the second conveyor belt (302).

4. The particle size detection device according to any one of claims 1 to 2, characterized in that: The fill light and the camera or laser rangefinder (104) are arranged side by side horizontally and in the same direction, and the distance between the fill light and the camera or laser rangefinder (104) is 8 to 12 cm.

5. The particle size detection device according to any one of claims 1 to 2, characterized in that: It also includes a control system, which is electrically connected to the image acquisition system, the conveyor belt, the laser rangefinder (104), and the image detection and material particle size calculation system.

6. A method for detecting particle size in a simulated particle metallurgical motion state, characterized by: The detection device according to any one of claims 1 to 5 is used for detection, which specifically comprises the following steps: Step 1: Start the conveyor belt to circulate and transport the material. When the material passes through the first conveyor belt (301), the second conveyor belt (302) and the third conveyor belt (303), the corresponding image acquisition system acquires images of all pellet particles, and respectively acquires images of the pellet particles in the inclined rolling state, the plane stacking state and the cross-section falling state, and transmits the images to the image detection and material particle size calculation system, and then calculates the particle sizes of all pellet particles in the images in the three states, and obtains the pellet particle size distribution data, thereby obtaining the initial particle size distribution data of the pellet particles; Step 2: When the material passes through the fourth conveyor belt (304), the laser rangefinder (104) measures the pellet particles from the side, and collects point distance data on the side of the pellet particles; Step 3: Based on the point distance data collected in step 2, a two-dimensional plane-space volume conversion model is used to process the data to calculate the particle size and particle size distribution of the pellet particles to be tested, and the initial particle size distribution data of the pellet particles obtained in step 2 is corrected to obtain the final particle size distribution of the pellet particles; Step 4: According to the final pellet particle size obtained in step 3, adjust the material feeding amount, conveyor belt inclination and movement speed, and fill light brightness; Step 5: loop step 1 to step 4 to obtain the particle size and particle size distribution data of the pellet particles at the next round of detection time.

7. The particle size detection method according to claim 6, characterized in that: In step 1, the image processing method includes: Step S1, using an image filtering noise reduction algorithm to perform noise reduction on the collected photos; Step S2: using an image enhancement algorithm to enhance the feature information of the target object; Step S3: Use the intelligent detection and segmentation model trained with the material sample image to identify and segment the pellet particle contour information in three different states, obtain the contour coordinate information of each pellet particle, and preliminarily calculate the particle size of all pellet particles revealed in the image.

8. The particle size detection method according to claim 6, characterized in that: In step 1, the particle size distribution ratio of the pellets is determined based on the particle size range, specifically following the following rules: First, the particle size proportions of the pellets in three different states, namely, the proportions of pellets with diameters below 0.5 cm, 0.5-1 cm, 1 cm-1.5 cm, 1.5-2 cm, 2 cm-2.5 cm, 2.5-3 cm, and above 3 cm, were counted respectively. Then, compare the particle size distribution percentages calculated under the three states. When the difference between any two percentages in the corresponding particle size distribution ranges under the three states is within 5%, take the average value of the three groups of percentages as the initial particle size distribution data of the pellet particles; otherwise, take the particle size distribution percentage under the plane stacking state as the initial particle size distribution data of the pellet particles; In step 3, the method for determining the particle size and proportion of pellets based on the data collected by the laser rangefinder (104) includes the following steps: Step 1: The collected pellet particle point distances, i.e., pellet particle edge point data and pellet particle center point data, are fitted into an arc using the least squares method; Step 2: Use Hough circle transform to fit the arc to obtain the semicircle of the pellet particles in three-dimensional space; Step 3: Using the principle of symmetry, fit the shape of the entire pellet in three-dimensional space and calculate the particle size of the pellet; Step 4: Count the particle size distribution proportions of all pellets; Step 5: Compare the initial pellet size distribution data with the pellet size distribution data obtained in step 4: When the difference between the two proportions in the corresponding particle size range between the initial particle size distribution data and the pellet size distribution data obtained in step 4 is within 5%, the average of the two proportions is used as the final pellet particle size distribution data; otherwise, the pellet particle size distribution data obtained in step 4 is used as the final pellet particle size distribution data.

9. The particle size detection method according to claim 6, characterized in that: In step four, When the particle size of the pellets detected is less than 0.5cm or greater than 3cm, the system will prompt you to adjust the amount of pellets to be added. The pellets with a particle size less than 0.5cm or greater than 3cm will be removed from the currently added pellets, and other equipment parameters will not be adjusted. When the particle size of the pellets detected is between 0.5cm and 3cm, inclusive, the feed amount does not need to be adjusted. The camera image acquisition efficiency is evaluated. When the image acquisition efficiency is greater than or equal to 90%, this set of data is judged to be appropriate and there is no need to adjust the feed amount and other equipment parameters. When the image acquisition efficiency is less than 90%, this set of data is judged to be worthless. The conveyor belt inclination angle and movement speed are reduced, and the brightness of the fill light is increased.

Images