Apparatus and method for evaluating particle size of blast furnace fuel raw materials

Abstract

An object of the present invention is to provide an apparatus and method for evaluating particle size of raw fuel for blast furnace, which detects individual particles of raw fuel in a video image taken for evaluating the particle size of raw fuel and classifies the types of raw fuel to evaluate the particle size by classifying the raw fuel by type. [Solution] The particle size evaluation device for blast furnace fuel raw materials of the present invention includes an image acquisition unit that acquires continuous images of the fuel raw materials being transported to the blast furnace, and an evaluation unit that evaluates the particle size of the fuel raw materials by determining the boundaries of the fuel raw materials due to the brightness difference of the fuel raw materials image included in the images and the type of the fuel raw materials according to the surface color and pattern of each of the fuel raw materials. The particle size evaluation method for blast furnace fuel raw materials includes a step in which the image acquisition unit acquires continuous images of the fuel raw materials being transported to the blast furnace, a step in which the evaluation unit uses a preset deep learning algorithm to determine the type of the fuel raw materials according to the boundaries of the fuel raw materials due to the brightness difference of the fuel raw materials image included in the images and the surface color and pattern of each of the fuel raw materials, a step in which the evaluation unit determines a representative size value of each particle of the fuel raw materials using size information according to the number of pixels occupied by each individual particle of the fuel raw materials in the fuel raw materials image, and a step in which the evaluation unit calculates an average particle size of the entire fuel raw materials according to the determined representative size value of the individual particles of the fuel raw materials, and calculates the particle size for each type of fuel raw materials according to the type of fuel raw materials classified by the deep learning algorithm.

Description

[Technical field]

[0001] The present invention relates to an apparatus and method for evaluating the particle size of blast furnace fuel materials. [Background technology]

[0002] The conventional method for measuring the particle size of fuel materials involves taking a sample of the target fuel material, passing the sample through a mesh sieve with holes of a certain size, and calculating the particle size based on the weight ratio of the samples that pass through each section based on the size of the mesh, using the concept of a weighted average. This method does not take into account the fact that the size distribution may differ even within the section, and since the particle size information is obtained from an extremely small, localized sample, it does not tend to be consistent throughout, and therefore has low representativeness.

[0003] To complement this, a method was developed to measure the average particle size by taking a video image of the fuel material being fed from a conveyor belt, using video techniques to find the boundaries of each individual fuel material particle, and evaluating their individual sizes. Although this is a limited measurement method in which measurements are centered on the upper observed surface, it is more representative as the entire feed section is evenly photographed, and it can show variability in real time.

[0004] On the other hand, although it has the advantage of continuous measurement, since the entire fuel supply is photographed, if the fuel is supplied in a mixed state, it is often difficult to determine the cause of the particle size variation. In the case of a blast furnace, the composition of the fuel supply is divided into three patterns, such as cokes, large iron ore, and small iron ore, depending on the combination of fuel supplies supplied. When coke is supplied, only coke is supplied in a limited manner and the fuel supplies are not mixed. However, when iron ore, i.e., sintered ore, is supplied, raw materials other than sintered ore or coke are also supplied, and when the particle size evaluation of the fuel supplies using the previous image was performed, the individual particle size changes of the mixed fuel supplies appear as an overall variation, so there was a lack of information. For example, when iron ore is supplied as large particles, sintered ore, pellets, and medium-sized coke are supplied together. When three or four or more types of raw materials are supplied in this way, if the particle size of one of them changes, it appears as a variation in the overall average particle size, but it was not possible to determine which raw material changed and caused the overall average particle size to change. During operations, if there is a fluctuation in particle size, the changed information must be confirmed and, if necessary, a request must be made to correct the particle size in the previous process or the charging conditions of the fuel raw materials must be adjusted to match the changed particle size. However, unless it is determined which raw material is causing the fluctuation, it is difficult to proceed with subsequent measures. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Korean Patent Publication No. 10-2021-0125529 Summary of the Invention [Problem to be solved by the invention]

[0006] An object of the present invention is to provide an apparatus and method for evaluating particle size of raw fuel for blast furnace, which detects individual particles of raw fuel in a video image taken for particle size evaluation of raw fuel and classifies the types of raw fuel to evaluate particle size by classifying the raw fuel by type. [Means for solving the problem]

[0007] In order to solve the above-mentioned problems, the particle size evaluation device for blast furnace raw fuel of the present invention is characterized by including an image acquisition unit that acquires continuous images of raw fuel transported to a blast furnace, and an evaluation unit that evaluates the particle size of the raw fuel by determining the boundaries of the raw fuel based on the brightness difference of the raw fuel image contained in the images and the type of the raw fuel according to the surface color and pattern of each raw fuel.

[0008] Furthermore, the particle size evaluation method for blast furnace raw fuel of the present invention is characterized by including the steps of: an image acquisition unit acquiring continuous images of raw fuel transported to a blast furnace; an evaluation unit using a preset deep learning algorithm to determine each boundary of the raw fuel due to the brightness difference of the raw fuel image included in the image and the type of the raw fuel according to the surface color and pattern of each raw fuel; the evaluation unit determining a representative size value of each individual particle of the raw fuel using size information according to the number of pixels occupied by each individual particle of the raw fuel in the raw fuel image; and the evaluation unit calculating an average particle size of the entire raw fuel according to the determined representative size value of the individual particles of the raw fuel, and calculating a particle size for each type of raw fuel according to the type of raw fuel classified by the deep learning algorithm. Effect of the Invention

[0009] According to the present invention, when multiple types of fuel raw materials are mixed when the fuel raw materials are supplied, the raw materials can be classified by type and the particle size can be evaluated, and the variation in particle size can be confirmed by raw material type, and measures can be taken or responses can be made to related operational matters. [Brief description of the drawings]

[0010] [Figure 1] 1 is a schematic configuration diagram of a particle size evaluation device for blast furnace fuel materials according to an embodiment of the present invention. [Diagram 2] 1 is a schematic diagram of an image acquisition unit of a particle size evaluation device for blast furnace fuel materials according to an embodiment of the present invention. FIG. [Diagram 3] 1 is a schematic flow chart of a method for evaluating particle size of blast furnace fuel material according to an embodiment of the present invention. [Figure 4] FIG. 2 is a diagram showing an example of a mixture of sintered ore and pellets among images acquired by the particle size evaluation device for blast furnace fuel materials according to an embodiment of the present invention. [Diagram 5] FIG. 2 is a diagram showing an example of a mixture of sintered ore and granulated ore, among images acquired by the particle size evaluation device for blast furnace fuel materials according to an embodiment of the present invention. [Figure 6] FIG. 2 is a diagram showing an example of a mixture of sintered ore and small coke among images acquired by the particle size evaluation device for blast furnace fuel materials according to an embodiment of the present invention. [Figure 7] FIG. 2 is a diagram showing an image acquired by the particle size evaluation device for blast furnace raw fuel according to an embodiment of the present invention and actual measurement data for learning artificial intelligence. [Figure 8] 1 is a diagram showing an image acquired by the particle size evaluation device for blast furnace fuel raw materials according to an embodiment of the present invention and classification and detection by type by an artificial intelligence that learned the image. [Figure 9] 1 is a diagram showing an image acquired by the particle size evaluation device for blast furnace fuel raw materials according to an embodiment of the present invention and classification and detection by type by an artificial intelligence that learned the image. [Figure 10] FIG. 1 is a diagram showing how artificial intelligence classifies and determines sizes by type from an image acquired by a particle size evaluation device for blast furnace fuel raw materials according to an embodiment of the present invention. [Figure 11] FIG. 1 is a diagram showing how artificial intelligence classifies and determines sizes by type from an image acquired by a particle size evaluation device for blast furnace fuel raw materials according to an embodiment of the present invention. [Figure 12] FIG. 1 illustrates an exemplary computing environment in which an apparatus for evaluating particle size of blast furnace fuel raw materials according to an embodiment of the present invention may be implemented. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments will be described in detail with reference to the accompanying drawings so that those skilled in the art to which the present invention pertains can easily carry out the present invention.

[0012] FIG. 1 is a schematic diagram of a particle size evaluation device for blast furnace fuel raw materials according to one embodiment of the present invention, FIG. 2 is a schematic diagram of an image acquisition unit of the particle size evaluation device for blast furnace fuel raw materials according to one embodiment of the present invention, and FIG. 3 is a schematic flowchart of a particle size evaluation method for blast furnace fuel raw materials according to one embodiment of the present invention.

[0013] As shown in FIG. 1, an apparatus 100 for evaluating particle size of blast furnace fuel according to an embodiment of the present invention may include an image acquisition unit 110 and an evaluation unit 120, and may further include a lighting controller .

[0014] The image acquisition unit 110 can acquire images of the fuel materials being transported to the blast furnace via a belt conveyor.

[0015] As shown in FIG. 2 along with FIG. 1, the image capture unit 110 may include a lighting unit 111, a camera 112, a purge hood 113, and a frame 114.

[0016] There is a belt conveyor at the bottom end of the image capture unit 110, and the lighting of the image capture unit 110 can irradiate the fuel material supplied onto the belt conveyor with light as wide as the width of the belt conveyor.

[0017] The lights 111a and 111b of the lighting unit 111 can be installed at the front and rear of the belt conveyor in the traveling direction, and when two lights are irradiated, the lights cross and gather at the middle part, and the upper camera 112 can capture an image in that area. In order to block the inflow of dust from the fuel material transported to the camera 112, a purge hood 113 can be installed directly below the camera 112, and a positive pressure can be constantly formed in the internal space of the purge hood 113, so that the dust at the lower part does not rise and rise toward the camera. A frame 114 is configured at the lower part of the purge hood 113, and the frame 114 may be as wide as the range of the angle of view of the camera 112 in the width direction. When the upper camera 112 is a line scan camera, the frame 114 opens only a very narrow gap space, and can limit the amount of surrounding dust rising.

[0018] The image acquisition unit 110 may be applied with a fuel material conveying mechanism such as a belt conveyor to acquire images of the continuously supplied fuel material, and in order to capture images synchronized with the conveying speed of the belt conveyor, the rotation speed is sensed using an encoder connected to the pulley shaft of the belt conveyor. The image signal captured by the camera 112 is transmitted to the evaluation unit 120.

[0019] The evaluation unit 120 collects image data frame by frame using a frame grabber 121, and detects each individual particle of the fuel material image using an artificial intelligence algorithm in the measurement unit 122 while classifying and detecting the type of the fuel material particle. The measurement unit 122 including such an artificial intelligence algorithm is composed of a graphic processing unit (GPU) and can process images at high speed, and can process images of the fuel material continuously supplied from a belt conveyor or the like in a short time, enabling continuous measurement. The evaluation unit 120 can control the lighting controller 130 to turn on the lighting during shooting according to the supply state of the fuel material, and can turn off the lighting during a period when the supply of the fuel material is interrupted and no image is shot.

[0020] As shown in FIG. 1 and FIG. 3, first, the image acquired by the image acquisition unit 110 is transmitted to the evaluation unit 120 (S10). The frame grabber 121 of the evaluation unit 120 acquires image data in units of frames. The image may be a black and white gray image, but using a color image may be advantageous in distinguishing the type of fuel material. The image may be acquired in units of frames using an area camera, but in a facility where the material is continuously supplied, such as a belt conveyor, it is possible to use a line scan camera to capture images in a line in the width direction and connect these images in succession to form an image in units of frames. In particular, when a lot of dust is present, if a line scan camera is used in this way, the opening for capturing the image of the frame 114 can be configured as a rectangular slit shape that is narrow in the length direction and is sufficient for the capturing angle in the width direction, which is advantageous in capturing the desired image while minimizing the inflow of dust.

[0021] Next, the measurement unit 122 of the evaluation unit 120 can detect individual particles of the fuel material and classify the type (S20). The boundary between the fuel material appears as a continuous shadow formed in the rectangular area of ​​the lights 111a, 111b, and a boundary due to the difference in brightness appears. Since the surface color and surface pattern of each individual particle are different, the type of fuel material can be classified by utilizing these characteristics. In this way, classification of types based on the boundaries of individual particles and the surface characteristics of particles can be easily realized by using an artificial intelligence algorithm.

[0022] Thereafter, the measurement unit 122 of the evaluation unit 120 can calculate the size from the detection area of ​​the individual particles of the detected fuel material (S30). For fuel material that is a certain distance away from the camera 112, there is a size standard indicated by an interval of one pixel. The measurement unit 122 can obtain the area from the number of pixels occupied by the area of ​​the individual particles of the fuel material in the fuel material image, and can obtain information on dimensions such as diameter, horizontal / vertical length, etc. Using such size information, the measurement unit 122 can determine a representative size value of the individual particles of the fuel material.

[0023] Next, the measurement unit 122 of the evaluation unit 120 can calculate the average particle size using the obtained size values ​​of the individual particles (S40). The average particle size of all fuel materials supplied during the designated period can be calculated, and since the types of fuel materials are classified, the fuel materials can be classified by type and the particle size of each type of fuel material can be calculated.

[0024] As described above, by calculating the particle size for each type of fuel material, it is possible to confirm which of the fuel material materials supplied simultaneously has a varying particle size.

[0025] Fig. 4 is a diagram showing an example of a mixture of sintered ore and pellets, among images acquired by a particle size evaluation device for blast furnace fuel raw materials according to an embodiment of the present invention. As shown in Fig. 4, the sintered ore has a rough gray-white surface and some parts have holes like basalt (identification code a). On the other hand, the pellets are brown, spherical, and bead-like in shape, although they have an uneven diameter (identification code b). These can be easily distinguished by a human eye, and the artificial intelligence algorithm of the measurement unit 122 can distinguish between sintered ore and pellets through a learning process, just as a human would judge.

[0026] FIG. 5 is a diagram showing an example of a mixture of sintered ore and pelletized ore, among images acquired by a particle size evaluation device for blast furnace raw fuel according to an embodiment of the present invention. As shown in FIG. 5, the raw fuel contains sintered ore (identification code a) and is mixed with brown, gravel-like pelletized ore (identification code c). Although pelletized ore is not uniform in shape, it has a relatively smooth surface compared to sintered ore and has the same brown color as pellets. In other words, the artificial intelligence algorithm can classify the type of raw fuel through a learning process according to the color and surface pattern of the raw fuel.

[0027] FIG. 6 is a diagram showing an example of a mixture of sintered ore (identification code a) and small coke (identification code d) among images acquired by the particle size evaluation device for blast furnace fuel materials according to one embodiment of the present invention.

[0028] FIG. 7 is a diagram showing an image acquired by the particle size evaluation device for blast furnace fuel raw materials according to an embodiment of the present invention and actual measurement data for learning artificial intelligence. FIG. 7 shows an example in which sintered ore and granulated ore are mixed in the fuel raw materials, in which the sintered ore is represented by an area with an identification code "a" and the granulated ore is represented by an area with an identification code "c". FIG. 7 also shows an image in which sintered ore and small-sized coke are mixed in the fuel raw materials, in which, as shown, there is grayish white sintered ore and the darker gray particles correspond to coke. In the conventional boundary detection based on the difference in brightness, it was not possible to classify and detect these particles by type, but by using the artificial intelligence algorithm of the measurement unit 122, not only the area of ​​each individual particle can be detected, but also the type can be detected at once. The artificial intelligence algorithm can be a deep learning algorithm.

[0029] 8 and 9 are respectively an image acquired by the particle size evaluation device for blast furnace raw fuel according to an embodiment of the present invention and an image detected by the AI ​​that learned the image and classified by type. As shown in Fig. 8 and Fig. 9, it can be seen that the AI ​​algorithm learned by the measurement unit 122 receives the original image image in the upper part and outputs the image in the lower part in which the raw fuel is classified by type.

[0030] 10 and 11 are diagrams in which artificial intelligence classifies and determines the size of individual particles from images acquired by a particle size evaluation device for blast furnace raw fuel according to an embodiment of the present invention. Referring to Fig. 10 and Fig. 11, the results show the size of each individual particle determined using the output image of an artificial intelligence algorithm that classifies and detects the raw fuel, and each particle is represented by a rectangular frame.

[0031] FIG. 10 shows an example where sintered ore and pelletized ore are mixed, with the outer size of the sintered ore shown as a rectangle with the identification code "a" and the outer size of the pelletized ore shown as a rectangle with the identification code "c." FIG. 11 shows an example where sintered ore and pellets are mixed, with the outer size of the sintered ore shown as a rectangle with the identification code "a" and the outer size of the pellets shown as a rectangle with the identification code "b."

[0032] The fuel materials supplied to the blast furnace are mixed in batches of coke, sintered ore, pellets, and pellets according to the composition, and are supplied at the same time. However, because the size distribution of these materials is different, the overall average particle size can change even if the composition ratio changes. Pellets are small, about 10 mm, but the coke supplied together with sintered ore is 30-40 mm or larger. Therefore, as the composition ratio of pellets increases, the overall average particle size decreases, and as the composition ratio of coke increases, the overall average particle size increases. However, if the particle size is calculated by dividing the fuel materials, the particle size is calculated for each fuel material, such as the particle size of sintered ore by collecting only sintered ore, or the particle size of pellets by separating only pellets, and the particle size change for each fuel material can be obtained. Therefore, the optical illusion caused by the particle size change due to the change in the composition ratio of each fuel material can be reduced.

[0033] 12 is a diagram showing an exemplary computing environment in which a particle size evaluation apparatus for blast furnace fuel raw materials according to an embodiment of the present invention can be implemented, and shows an example of a system 1000 including a computing device 1100 configured to implement one or more of the above-described embodiments. For example, the computing device 1100 can include, but is not limited to, a personal computer, a server computer, a handheld or laptop device, a mobile device (such as a mobile phone, PDA, media player, etc.), a multiprocessor system, a consumer electronics device, a minicomputer, a mainframe computer, a distributed computing environment including any of the above-mentioned systems or devices, etc.

[0034] The computing device 1100 may include at least one processing unit 1110 and a memory 1120. Here, the processing unit 1110 may include, for example, a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor, an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), etc., and may have multiple cores. The memory 1120 may be a volatile memory (e.g., RAM, etc.), a non-volatile memory (e.g., ROM, flash memory, etc.), or a combination thereof.

[0035] Computing device 1100 may also include additional storage 1130. Storage 1130 may include, but is not limited to, magnetic storage, optical storage, and the like. Storage 1130 may store computer readable instructions for implementing one or more embodiments disclosed herein, as well as other computer readable instructions for implementing an operating system, application programs, and the like. The computer readable instructions stored in storage 1130 may be loaded into memory 1120 for execution by processing unit 1110.

[0036] Furthermore, the computing device 1100 may include an input device 1140 and an output device 1150. Here, the input device 1140 may include, for example, a keyboard, a mouse, a pen, a voice input device, a touch input device, an infrared camera, a video input device, or any other input device. Also, the output device 1150 may include, for example, one or more displays, speakers, a printer, or any other output device. Note that the computing device 1100 may also use an input device or an output device provided in another computing device as the input device 1140 or the output device 1150.

[0037] Computing device 1100 may also include communication connections 1160 that allow it to communicate with other devices (e.g., computing device 1300) over network 1200. Here, communication connections 1160 may include a modem, a network interface card (NIC), an integrated network interface, a radio frequency transmitter / receiver, an infrared port, a USB connection, or other interface for connecting computing device 1100 to other computing devices. Additionally, communication connections 1160 may include wired or wireless connections.

[0038] The components of the computing device 1100 described above may be connected by various interconnects such as buses (e.g., Peripheral Component Interconnect (PCI), USB, firmware (IEEE 1394), optical bus structures, etc.) or may be interconnected by a network.

[0039] As used herein, terms such as "evaluator," "frame grabber," "measurement unit," "lighting controller," and the like generally refer to computer-related entities that are hardware, a combination of hardware and software, software, or software in execution. For example, a component may be, but is not limited to, a process running on a processor, a processor, an object, an executable, a thread of execution, a program, and / or a computer. For example, an application running on a controller and the controller may all be components. One or more components may reside within a process and / or thread of execution, and a component may be localized on one computer or distributed among two or more computers.

[0040] As described above, according to the present invention, when multiple types of fuel raw materials are mixed when the fuel raw materials are supplied, the particle size of each raw material can be evaluated by classifying them by type, and the variation in particle size by raw material type can be confirmed, and measures or responses can be taken for related operational matters.

[0041] The present invention described above is not limited by the above-mentioned embodiments and the attached drawings, but by the claims below, and it will be easily understood by those having ordinary knowledge in the technical field to which the present invention pertains that the configuration of the present invention can be changed and modified in various ways without departing from the technical concept of the present invention. [Explanation of symbols]

[0042] 100 Particle Size Evaluation Device 110 Video acquisition unit 111 Lighting Department 111a, 111b Lighting 112 Camera 113 Purge Hood 114 Frames 120 Evaluation Department 121 Frame Grabber 122 Measuring section 130 Lighting Controller 1000 Systems 1100 Computing Devices 1110 Processing Unit 1120 Memory 1130 Storage 1140 Input Devices 1150 Output Device 1160 Communication Connection 1200 Network 1300 Computing Devices

Claims

1. an image acquisition unit for acquiring continuous images of fuel materials being transported to a blast furnace; and an evaluation unit that distinguishes the type of fuel material and evaluates the particle size of the fuel material according to the boundaries of the fuel material based on the brightness difference of the fuel material image included in the image and the surface color and pattern of the fuel material.

2. The evaluation unit is a frame grabber that collects images from the image acquisition unit in units of frames in synchronization with a conveying speed of a belt conveyor that conveys the fuel material to the blast furnace; 2. The blast furnace fuel raw material particle size evaluation device according to claim 1, further comprising a measurement unit for determining the type of fuel raw material and measuring the particle size of the fuel raw material according to the boundaries of the fuel raw material based on the brightness difference of the fuel raw material image in the image collected by the frame grabber and the surface color and pattern of the fuel raw material using a preset artificial intelligence algorithm.

3. 3. The particle size evaluation device for blast furnace raw fuel materials according to claim 2, wherein the measurement unit determines a representative size value of an individual particle of the raw fuel material by using size information corresponding to a number of pixels occupied by the individual particle of the raw fuel material in the raw fuel material image.

4. 4. The particle size evaluation device for blast furnace raw fuel according to claim 3, wherein the measurement unit calculates an average particle size of the entire raw fuel according to the determined representative size value of each particle of the raw fuel, and calculates particle sizes for each type of raw fuel according to the type of raw fuel classified by the artificial intelligence algorithm.

5. The image acquisition unit is A camera for taking pictures above the fuel material; A lighting unit for providing illumination to the fuel material at the front and rear of the belt conveyor in the traveling direction of the fuel material; A purge hood disposed under the camera to block the inflow of dust from the fuel material; 2. The apparatus for evaluating particle size of fuel materials for a blast furnace according to claim 1, further comprising: a frame disposed under the purge hood and blocking an inflow of dust of the fuel materials together with the purge hood.

6. An image acquisition unit acquires continuous images of fuel materials being transported to a blast furnace; The evaluation unit uses a preset artificial intelligence algorithm to determine the boundaries of the fuel materials according to the contrast of the fuel material images included in the image and the types of the fuel materials according to the surface colors and patterns of the fuel materials; the evaluation unit determining a representative size value of each individual particle of the fuel raw material using size information corresponding to the number of pixels occupied by each individual particle of the fuel raw material in the fuel raw material image; the evaluation unit calculates an average particle size of the entire fuel raw materials according to the determined representative size value of the individual particles of the fuel raw materials, and calculates particle sizes for each type of fuel raw materials according to the types of fuel raw materials classified by the artificial intelligence algorithm.

Images