Method for measuring particle size index

Abstract

To provide a particle size index measuring method capable of accurately measuring a particle size index to be a standard indicating a particle size distribution.SOLUTION: Preparing, for a plurality of brands, a reference sample that is a sample composed of particles having different particle size distributions depending on brands, and calculating a particle diameter and an area of a surface layer particle based on a distance image of the reference sample; A step ST1 of creating a calibration curve indicating a relationship between a sum of areas of surface particles in a specific particle size division calculated for each of a plurality of reference samples and a particle size indicator acquired in advance for each of the plurality of reference samples using a sieve, the particle size indicator being a mass ratio of particles having a specific particle size or less to all particles constituting the reference sample; And a step ST3 of calculating the particle size index of the object to be measured on the basis of the calibration curve and the total sum of the areas of the surface layer particles in the specific particle size section.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a method for measuring a particle size index, which serves as an indicator of the particle size distribution of a measurement object composed of particles such as coal particles, and in particular to a particle size index measurement method that can accurately measure the particle size index of a measurement object, even if the measurement object contains particles with small particle sizes that cannot be detected in distance images obtained by a 3D camera or the like, and particles that generate pseudo-particles due to aggregation due to moisture. [Background technology]

[0002] In blast furnace operation, coke, a raw material for the blast furnace, is required to have a certain strength to ensure good ventilation within the furnace. Therefore, in the production of coke in a coke oven, the bulk density of the coal charged into the coke oven is increased to improve the contact between adjacent coal particles so that strong adhesion occurs between the coal particles when the coal softens and melts. Coal is crushed before being charged into the coke oven. To increase the bulk density of the coal at the time of charging, it is desirable to measure the particle size distribution of the coal before crushing and adjust the particle size by crushing to approach the ideal particle size distribution.

[0003] However, the current method of measuring coal particle size distribution is, for example, performed using a sieve once per shift, and the infrequent measurement makes it difficult to implement timely improvement actions, such as adjusting grinding conditions, in response to changes in coal brand or moisture content.For this reason, it is desirable to adopt a method that can continuously measure particle size without contact and evaluate particle size distribution using range images obtained by a 3D camera or similar device.

[0004] As a method capable of continuously measuring particle diameters in a non-contact manner, for example, the method described in Non-Patent Document 1 has been proposed. The method described in Non-Patent Document 1 is a measurement method using a light-section type 3D camera in which a laser light source that emits linear laser light and an area scan camera are integrated. In the method described in Non-Patent Document 1, a moving distance detection device such as a rotary encoder in contact with the belt conveyor is used. A 3D camera measures the position of the upper edge of the cross section of particles deposited on the belt conveyor every time the belt conveyor moves a certain distance. This generates a distance image (sometimes called a 3D image or depth image) in which the pixel value of each pixel indicates the distance from a reference position (e.g., the distance from the 3D camera). Near the boundaries of stacked particles, the irradiated laser light is interrupted and dark, and the unevenness of the particles increases. Therefore, in the distance image, the pixel values ​​of pixel regions corresponding to the boundaries of particles tend to differ from the pixel values ​​of other pixel regions. The method described in Non-Patent Document 1 utilizes this characteristic to determine particle boundaries, identify each particle, and calculate the particle size of each particle. The dimensions of stacked particles with portions hidden by other particles are smaller than their actual dimensions. For this reason, in the method described in Non-Patent Document 1, height information of each particle (height from the bottom of the conveyor belt) that can be calculated using a 3D camera is used to preferentially extract surface particles (hereinafter referred to as "surface particles" as appropriate), and the diameter of the minor axis of each surface particle in the range image when it is considered as an ellipse is used as the particle size.

[0005] As a method capable of continuously measuring particle diameters in a non-contact manner, the methods described in Non-Patent Document 2 and Patent Document 1 have also been proposed, which use a 3D camera in the same way as the method described in Non-Patent Document 1. Non-Patent Document 2 and Patent Document 1 describe in detail an edge detection method for identifying each particle and an image processing method for recognizing surface particles, and Patent Document 1 in particular also describes a method for speeding up measurement.

[0006] In the method described in Non-Patent Document 2, image processing is performed on a distance image acquired of deposited particles to extract surface particles in the surface layer with little overlap, the particle size of each surface particle is sorted into particle size categories determined by the mesh size of a sieve, and the number distribution of surface particles (surface number distribution), which is the relationship between the particle size categories and the number of surface particles in each particle size category, is calculated. The method described in Non-Patent Document 2 then estimates the number distribution of all deposited particles, including not only surface particles but also hidden particles, using a surface probability model, which represents the degree of surface visibility and appearance in the surface layer according to particle size, i.e., a model that estimates the number distribution of all deposited particles (total number distribution) from the number distribution of surface particles. Furthermore, the method described in Non-Patent Document 2 estimates the distribution of the mass fraction of all deposited particles (total mass distribution) using the volume ratio (or mass ratio) for each particle size category.

[0007] In Patent Document 2, the inventors applied the method described in Non-Patent Document 2 to a mixed particle size deposit in which particles (coke particles) belonging to multiple particle size classes are mixed and deposited at a predetermined mass ratio, and conducted a confirmation test to determine whether the overall mass distribution can be accurately estimated. As a result of this confirmation test, it was found that even if the sample was a single-grained sample consisting only of particles with the same grain size, in the case of irregular particles such as coke or sintered ore, the grain size distribution of the single-grained sample would be broader than the grain size distribution because the grain size of the surface layer particles varies depending on the particle orientation. For this reason, it was found that simply sorting the grain sizes of surface layer particles measured using distance images obtained by a 3D camera or the like into grain size divisions determined by sieves, counting the number of surface layer particles for each grain size division, and applying the method described in Non-Patent Document 2 would result in a blurred grain size distribution (total mass distribution) that did not accurately match the results measured using sieves.

[0008] Therefore, in the method described in Patent Document 2, single-grain-size samples are prepared for each of the multiple grain-size ranges to which the particles constituting the mixed-grain-size deposit belong. The particle size of the surface layer of each single-grain-size sample is measured to calculate a first particle size distribution, which indicates the relationship between the particle size and the number (or area or volume) of particles in the surface layer of the single-grain-size sample. Furthermore, the particle size of the surface layer of a mixed-grain-size deposit containing particles belonging to the multiple grain-size ranges is measured to calculate a second particle size distribution, which indicates the relationship between the particle size and the number (or area or volume) of particles in the surface layer of the mixed-grain-size deposit. The calculated second particle size distribution is then approximated by a linear sum of the calculated first particle size distributions, and each coefficient of this linear sum is considered to represent the number ratio of different particle size ranges in the surface layer of the mixed-grain-size deposit, assuming that the mixed-grain-size deposit is composed of a combination of single-grain-size samples of multiple grain-size ranges. The coefficients are then used to calculate the overall mass distribution using a procedure similar to that described in Non-Patent Document 2, enabling highly accurate calculations that closely match the results of measurements using sieves.

[0009] However, when the measurement target is something like coal, which contains particles with small particle sizes (for example, less than 6 mm) that cannot be measured using distance images obtained by a 3D camera or the like, the method described in Patent Document 2 cannot correctly evaluate the particle size distribution (total mass distribution). FIG. 1 shows an example of the results of investigating the particle size distribution of coal samples (Coal Nos. 1 to 5) each composed of five different types of coal particles. The horizontal axis in FIG. 1 represents particle size, and the vertical axis represents the cumulative mass fraction of the overall mass distribution. For example, when the horizontal axis in FIG. 1 is 10 mm, the value on the vertical axis represents the mass fraction of particles with a particle size of 10 mm or less, and when the horizontal axis is 20 mm, the value on the vertical axis represents the mass fraction of particles with a particle size of 20 mm or less. Table 1 below shows the particle size index, which is the mass fraction of particles with a specific particle size or less, for Coal Nos. 1 to 5 shown in FIG. 1. The particle size index is an index that serves as a guide for particle size distribution, and the specific particle size is, for example, 3 mm. Therefore, the particle size index shown in Table 1 is the mass fraction of particles with a particle size of 3 mm or less (hereinafter, appropriately referred to as the "-3 mm fraction"). [Table 1] As can be seen from Figure 1, the proportion of larger particle sizes increases as the number increases, from Coal No. 1 to Coal No. 5. Reflecting this, Table 1 shows that, from Coal No. 1 to Coal No. 5, the particle size index (-3 mm proportion) decreases and the proportion of larger particle sizes increases as the number increases.

[0010] As shown in Figure 1 and Table 1, although coal contains large particles with particle sizes of 6 mm or more, which correspond to particle sizes of 6 mm or more that can be measured using distance images obtained by a 3D camera or the like, the proportion of particles with particle sizes of 3 mm or less is high. Therefore, the method described in Patent Document 2, which targets only particle size categories that correspond to particle sizes that can be measured using distance images, cannot accurately evaluate the particle size distribution of coal. Furthermore, because coal contains particles that generate pseudo-particles due to aggregation caused by moisture, it is affected by changes in moisture content, which causes the actual particle size distribution to be unable to be accurately evaluated.

[0011] In adjusting the coal crushing conditions, it is not necessary to measure the particle size distribution for all particle size divisions (calculate the mass proportion for each particle size division); it is sufficient to calculate a particle size index, such as the -3 mm proportion, which serves as a guideline for expressing the particle size distribution. [Prior art documents] [Patent documents]

[0012] [Patent Document 1] Japanese Patent Application Publication No. 2019-174155 [Patent Document 2] Japanese Patent Publication No. 2022-172620 [Non-patent literature]

[0013] [Non-Patent Document 1] MJ Thurley, "Automated, On-line, Calibration-Free, Particle Size Measurement using 3D Profile Data", Measurement and Analysis of Blast Fragmentation: Workshop Hosted by FRAGBLAST 10 - The 10th International Symposium on Rock Fragmentation by Blasting, 2013, pp.23-32 [Non-patent document 2] MJ Thurley, "Three Dimensional Data Analysis for the Separation and Sizing of Rock Piles in Mining", Ph.D. Thesis, Monash University, December 2002, chapter 4, pp.27-60 Summary of the Invention [Problem to be solved by the invention]

[0014] The present invention aims to provide a particle size index measurement method that can accurately measure the particle size index, which serves as a guide to particle size distribution, even for objects that include particles such as coal particles that have small particle sizes that cannot be measured using distance images obtained by a 3D camera, etc., and that generate pseudo-particles due to aggregation caused by moisture. [Means for solving the problem]

[0015] To solve the above problems, the present inventors conducted extensive research. As a result, they found that there is a high correlation between the sum of the areas of surface particles of a measurement object calculated based on a distance image and a particle size index, which is the mass fraction of particles having a specific particle size or less (for example, a particle size of 3 mm or less). Furthermore, they found that when considering the sum of the areas of surface particles in a specific particle size range (for example, a range with a particle size of more than 25 mm) among surface particles, the correlation between this and the particle size index can be maintained with little effect from changes in moisture content, even for measurement objects containing particles that generate pseudoparticles due to aggregation with moisture. The present invention was completed based on the findings of the inventors described above.

[0016] That is, in order to solve the above problem, the present invention provides a method for preparing a plurality of brands of reference samples, which are samples made up of particles with different particle size distributions depending on the brand, taking images of the surface layers of the reference samples while changing the deposition state of the reference samples, obtaining distance images showing the distance from a reference position to the particles in the surface layer, calculating the particle sizes and areas of the particles in the surface layer based on the distance images, and calculating a particle size index, which is the mass ratio of particles of a specific particle size or less out of all the particles constituting the reference samples, obtained in advance for each of the plurality of reference samples using a sieve, for each of the plurality of reference samples. an area calculation step of imaging the surface layer of a measurement object made up of particles whose particle size distribution is unknown while changing the deposition state of the measurement object, obtaining a distance image showing the distance from a reference position to the particles in the surface layer, calculating the particle size and area of ​​the particles in the surface layer based on the distance image, and calculating the sum of the areas of the particles in the surface layer in the specific particle size range; and a particle size index calculation step of calculating the particle size index of the measurement object based on the calibration curve and the sum of the areas of the particles in the surface layer in the specific particle size range for the measurement object.

[0017] In the present invention, a "distance image" refers to an image in which the pixel value of each pixel indicates the distance from a reference position (e.g., the distance from the distance image acquisition means). The "distance image" acquired in the calibration curve creation step is an image indicating the distance from the reference position to the surface particles of the reference sample, and the "distance image" acquired in the area calculation step is an image indicating the distance from the reference position to the surface particles to be measured. The distance image acquisition means for acquiring the distance image is not particularly limited as long as it can acquire the distance to the target surface layer, but examples include a light-section type 3D camera that combines a laser light source that emits linear laser light with an area scan camera. The reference position can be set to any position; for example, the position of the distance image acquisition means can be used as the reference position. Once a "distance image" can be acquired, it is possible to calculate the particle size and area by performing calculations based on the "distance image" using a calculation device connected to the distance image acquisition means, for example, by using the method described in Non-Patent Document 2.

[0018] According to the present invention, in the calibration curve creation step, a plurality of brands of standard samples composed of particles with different particle size distributions (and therefore different particle size indices, which serve as a guide for particle size distribution), are used, and a calibration curve is created showing the relationship between the sum of the areas of surface particles in a specific particle size range, calculated for each of the plurality of standard samples based on these distance images, and the particle size index (the mass proportion of particles of a specific particle size or less among all the particles constituting the standard sample) previously obtained for each of the plurality of standard samples using a sieve. According to the aforementioned findings of the inventors, there is a high correlation between the two, and the effect of changes in moisture content is small, making it possible to create a highly accurate calibration curve. Therefore, in the area calculation process, if the sum of the areas of surface particles in a specific particle size range (the same specific particle size range as the reference sample) is calculated based on a distance image of the measurement object composed of particles whose particle size distribution is unknown (and therefore whose particle size index is also unknown), then in the particle size index calculation process, the particle size index of the measurement object can be measured with high accuracy based on the calibration curve and the sum of the areas of surface particles in the specific particle size range for this measurement object.

[0019] When the object to be measured for particle size index by the present invention is piled up in a tray, the area of ​​the object to be measured in the distance image (the area occupied by the object to be measured in a plane perpendicular to the direction in which the distance image is acquired) can be considered to be constant, but when the object to be measured is transported in a piled up state on a conveyor belt, the area of ​​the object to be measured in the distance image is considered to fluctuate. If the area of ​​the object to be measured fluctuates, the total area of ​​the surface particles of a specific particle size range of the object to be measured will also fluctuate, which may result in a decrease in the accuracy of the particle size index measurement. To prevent this decrease in measurement accuracy, the sum of the areas of surface particles in a specific particle size range, which is one of two correlated variables represented by the calibration curve created in the calibration curve creation step, can be normalized by the area of ​​the reference sample. Then, in the particle size index calculation step, the particle size index of the object to be measured can be calculated based on the calibration curve and the sum of the areas of surface particles in a specific particle size range for the object to be measured, normalized by the area of ​​the object to be measured. This allows the particle size index of the object to be measured with high accuracy even if the area of ​​the object to be measured fluctuates.

[0020] Therefore, in the present invention, when the measurement object is transported in a state of being piled up on a belt conveyor, preferably, in the calibration curve creation step, the sum of the areas of the particles in the surface layer of the specific particle size range is normalized by the area of ​​the reference sample in the distance image, and the relationship between the normalized sum of the areas of the particles in the surface layer of the specific particle size range calculated for each of the plurality of reference samples and the particle size index previously acquired for each of the plurality of reference samples is created as the calibration curve, and in the area calculation step, images of the surface layer of the measurement object and the belt conveyor are taken while the measurement object is transported on the belt conveyor, and the particles in the surface layer and the conveyor are imaged from a reference position. a first distance image that is a distance image showing the distance to the conveyor belt; a difference image between the first distance image and a second distance image that is a distance image showing the distance from a reference position to the conveyor belt, which has been acquired in advance by capturing an image of only the belt conveyor; a sum of the areas of the particles in the surface layer of the specific particle size range is calculated based on the difference image; and the area of ​​the object to be measured in the difference image; and in the particle size index calculation step, the particle size index of the object to be measured is calculated based on the calibration curve and the sum of the areas of the particles in the surface layer of the specific particle size range for the object to be measured, normalized by the area of ​​the object to be measured in the difference image.

[0021] According to the above-mentioned preferred method, in the calibration curve creation step, the sum of the areas of particles in the surface layer of a specific particle size range is normalized by the area of ​​a reference sample in the distance image, and a calibration curve is created showing the relationship between the normalized sum of the areas of particles in the surface layer of a specific particle size range calculated for each of the multiple reference samples and the particle size index previously obtained for each of the multiple reference samples. Then, in the area calculation step, the surface layer of the measurement object and the belt conveyor are imaged to obtain a first distance image, which is a distance image showing the distance from a reference position to the surface particles and the conveyor belt, and a difference image is calculated between the first distance image and a second distance image, which is a distance image showing the distance from the reference position to the conveyor belt and was previously obtained by image-taking only the belt conveyor. In the area calculation step, the area of ​​the measurement object can be calculated based on this difference image. It is also possible to calculate the sum of the areas of surface particles of a specific particle size range for the measurement object based on the difference image. Therefore, in the particle size index calculation process, the particle size index of the object to be measured can be measured accurately without a decrease in measurement accuracy, based on the calibration curve and the sum of the area of ​​surface particles in a specific particle size range for the object to be measured, normalized by the area of ​​the object to be measured in the differential image, even when the object to be measured is transported in a piled state on a belt conveyor.

[0022] The present invention is suitably used when the particles are coal particles. [Effects of the Invention]

[0023] According to the present invention, it is possible to accurately measure the particle size index, which serves as a guide for particle size distribution, even for measurement objects that include particles such as coal particles that have small particle sizes that cannot be measured using distance images obtained by a 3D camera or the like, and that include particles that generate pseudo-particles due to aggregation due to moisture. [Brief explanation of the drawings]

[0024] [Figure 1] FIG. 1 shows an example of the results of investigating the particle size distribution of coal samples (coal Nos. 1 to 5) each composed of five different types of coal particles of different brands. [Figure 2] 1 is a flow chart showing steps of a particle size index measurement method according to one embodiment of the present invention. [Figure 3] FIG. 3 is an explanatory diagram for schematically explaining step ST1 shown in FIG. 2. [Figure 4]4 is a diagram showing an example of a distance image acquired by the 3D camera 2 shown in FIG. 3 and surface layer particles detected from the distance image. FIG. [Figure 5] The deposition state of each of the reference samples 1 of coals No. 1 to No. 5 was changed and images were taken repeatedly 10 times. This shows the relationship between the sum of the areas of surface particles (surface particles of all particle size categories) calculated from the 10 distance images and the particle size index (-3 mm ratio) of coals No. 1 to No. 5 shown in Table 1. [Figure 6] FIG. 1 shows an example of distance images acquired by a 3D camera 2 and surface layer particles detected from the distance images when reference samples 1 with different moisture contents are prepared for coal No. 1, which has the smallest proportion of large-particle-size particles. [Figure 7] Images were taken 10 times with different deposition conditions for each of the reference samples 1 of coal No. 1 with moisture contents of 8% and 13%. The relationship between the total area of ​​surface particles (surface particles of all particle size categories) calculated from the 10 range images and the particle size index (-3 mm ratio) of coal No. 1 shown in Table 1 is shown in addition to Fig. 5. [Figure 8] This figure shows the relationship between the moisture content and the sum of the surface particle areas calculated from 10 distance images taken 10 times while changing the deposition state of each of the reference samples 1 of coals No. 3 to No. 5. [Figure 9] This figure summarizes the relationship between the sum of the surface particle areas obtained by changing the moisture content of each of the reference samples 1 of coals No. 1 to No. 5 in the range of 6 to 13%, which is close to 10%, and the particle size index (-3 mm proportion) of coals No. 1 to No. 5 shown in Table 1. [Figure 10] FIG. 3 is an explanatory diagram for schematically explaining step ST3 shown in FIG. 2. [Figure 11] FIG. 3 is an explanatory diagram for schematically explaining a modified example of step ST2 shown in FIG. 2. DETAILED DESCRIPTION OF THE INVENTION

[0025] Hereinafter, with reference to the accompanying drawings as appropriate, an embodiment of the present invention will be described, taking as an example a case where the particles constituting the reference sample and the measurement object are coal particles. Fig. 2 is a flow chart showing the steps of a particle size index measuring method according to one embodiment of the present invention. As shown in Fig. 2, the particle size index measuring method according to this embodiment includes steps ST1 to ST3. Each of steps ST1 to ST3 will be described in order below.

[0026] [Process ST1] Step ST1 corresponds to the calibration curve creation step of the present invention. Figure 3 is an explanatory diagram that schematically explains step ST1. 3, in step ST1, reference samples 1, which are samples composed of particles C with different particle size distributions depending on the brand, are prepared for multiple brands. Specifically, for example, from a pail P storing particles C of each brand with a mass of 15 kg, particles C with a mass of 3 kg, equivalent to one tray T, are dug out from the top and deposited in the tray T, respectively, to prepare reference samples 1.

[0027] Next, in step ST1, the surface layer of the reference sample 1 is repeatedly imaged while changing the deposition state of the reference sample 1, and a distance image showing the distance from the reference position to the surface layer particles is obtained. In this embodiment, the distance image is obtained using a 3D camera 2 that is placed above the tray T and uses a light-section method with a linear laser beam L, as shown in FIG. 3 . Specifically, the distance image can be obtained by moving the tray T relative to the 3D camera 2 in a direction (a direction perpendicular to the paper surface of FIG. 3 ) perpendicular to the direction in which the linear laser beam L, which serves as the light-section line, extends (the left-right direction in FIG. 3 ). Furthermore, the deposition state of the reference sample 1 can be changed, for example, by returning the particles C in the tray T to another container (not shown) after one distance image has been obtained, and then returning the particles C from the other container to the tray T and depositing them again.

[0028] Next, in step ST1, the calculation device 3 connected to the 3D camera 2 calculates the particle size and area of ​​the surface layer particles of the reference sample 1 based on the distance image acquired by the 3D camera 2. Then, in step ST1, a calibration curve is created showing the relationship between the sum of the areas of surface particles in a specific particle size range (in this embodiment, the range with particle sizes larger than 25 mm) calculated for each of the multiple reference samples 1 and the particle size index, which is the mass proportion of particles with a specific particle size or less among all the particles C constituting the reference sample 1, obtained in advance for each of the multiple reference samples 1 using a sieve (in this embodiment, the -3mm proportion, which is the mass proportion of particles with a particle size of 3 mm or less). The reason for creating the calibration curve in the above manner will be explained below.

[0029] FIG. 4 shows an example of a distance image acquired by a 3D camera 2 and surface layer particles detected from the distance image. The upper part of FIG. 4 is a distance image, and the lower part is a color display of the surface layer particles detected in the distance image in the upper part (although FIG. 4 attached to this specification shows a monochrome display, in reality it is a color display). The example shown in FIG. 4 is a distance image obtained for each of the reference samples 1 of coals No. 1 to No. 5 having the particle size distribution shown in FIG. 1 above. The moisture content of coals No. 1 to No. 5, from which the distance images shown in FIG. 4 were acquired, is 10%. The range image shown in Figure 4 was obtained using a Sick Ranger-E50 (area scan camera pixel count: 1536 x 512, measurement speed: 35,000 cross sections / second) as 3D camera 2, and by adjusting the height from tray T to 3D camera 2 so that the entire tray T (width 300 x length 400 x height 60 mm) on which coal particles had accumulated was within the field of view. Tray T was also mounted on a belt-driven traveling cart, and a rotary encoder installed on the traveling cart was used to detect the travel distance. One range image was obtained when the traveling cart had traveled the length of tray T. As mentioned above, from coal No. 1 to coal No. 5, the proportion of larger particles increases as the number increases. However, as shown in the bottom panel of Figure 4, the greater the proportion of larger particles, the greater the number of detected surface particles and the larger the total area of ​​the surface particles.

[0030] Figure 5 shows the relationship between the sum of the surface particle areas (surface particle areas of all particle size classes) calculated from 10 distance images taken by repeatedly capturing images of each of the above-mentioned reference samples 1 for coals No. 1 to No. 5, with the deposition conditions changed, and the particle size index (-3 mm proportion) for coals No. 1 to No. 5 shown in Table 1. Note that in Figure 5, coal No. 1 has two plot points because two reference samples 1 for coal No. 1 were prepared, and the sum of the surface particle areas was calculated by repeatedly capturing images of each of them, with the deposition conditions changed. The same is true for coals No. 2 to No. 5. As shown in Figure 5, there is a high correlation between the total area of ​​surface particles and the particle size index.

[0031] FIG. 6 shows examples of distance images acquired by a 3D camera 2 and surface particles detected from the distance images when reference samples 1 with different moisture contents were prepared for Coal No. 1, which has the smallest proportion of large particles. The upper part of FIG. 6 shows the distance image, and the lower part shows a color display of the surface particles detected in the upper distance image (although the display appears monochrome in FIG. 6 attached to this specification, the display is actually color). The examples shown in FIG. 6 are distance images acquired for Reference Samples 1 of Coal No. 1 with moisture contents of 8%, 10%, and 13%, respectively. The conditions for acquiring the distance images were the same as those for acquiring the distance images shown in FIG. 4. The distance image acquired for Reference Sample 1 of Coal No. 1 with a moisture content of 10% shown in FIG. 6 is the same as that shown in FIG. 4. The moisture content of the coal (the moisture content of Reference Sample 1) was adjusted by adding water to achieve the desired moisture content, and the moisture content value was calculated from the change in mass of Reference Sample 1 until it reached an absolute dry state and the amount of water added to Reference Sample 1 in the absolute dry state. All moisture content adjustments and calculations described herein were performed in this manner. As shown in the bottom panel of Figure 6, the higher the moisture content, the greater the number of detected surface particles. As shown in the top panel of Figure 6, the surface (the powdery particle portion with small particle size) of the distance images obtained for Reference Sample 1 with a moisture content of 8% and 10% is smooth with few irregularities, whereas the surface of the distance image obtained for Reference Sample 1 with a moisture content of 13% has rough irregularities, which are thought to be due to pseudo-particles generated by aggregation due to moisture. It is thought that some of these pseudo-particles are detected as surface particles, which is why the number of detected surface particles is so high.

[0032] Fig. 7 is a diagram supplementing Fig. 5, showing the relationship between the sum of the surface particle areas (surface particle areas of all particle size categories) calculated from 10 range images taken by repeatedly capturing images 10 times for each of the Reference Samples 1 of Coal No. 1 with moisture contents of 8% and 13%, while changing the deposition state, and the particle size index (-3 mm proportion) for Coal No. 1 shown in Table 1. Note that Fig. 7 has four plot points for Coal No. 1 with moisture contents of 8% and 13%, because four Reference Samples 1 of Coal No. 1 with moisture contents of 8% and 13% were prepared, and four images were taken 10 times for each of them while changing the deposition state, to calculate the sum of the surface particle areas. As shown in Figure 7, there is not a significant difference in the total area of ​​surface particles between Reference Sample 1 for Coal No. 1, which has a moisture content of 8%, and Reference Sample 1 for Coal No. 1, which has a moisture content of 10%, but the total area of ​​surface particles for Reference Sample 1 for Coal No. 1, which has a moisture content of 13%, is larger and is therefore larger than the total area of ​​surface particles for Reference Sample 1 for Coal No. 2, which has a larger proportion of larger particles than Reference Sample 1 for Coal No. 1 (and therefore a smaller proportion of -3 mm particles, which is the particle size index). This shows that the correlation between the total area of ​​surface particles and the particle size index is affected by changes in moisture content.

[0033] While Figures 6 and 7 show the effect of changes in moisture content on Reference Sample 1 of Coal No. 1, which contains a small proportion of large-sized particles, the effect of changes in moisture content was also confirmed for Reference Samples 1 of Coal Nos. 3 to 5, which contain a large proportion of large-sized particles. Figure 8 shows the relationship between the moisture content and the sum of the surface particle areas calculated from 10 distance images taken 10 times while changing the deposition state of each of the reference samples 1 of coal Nos. 3 to 5, which have different moisture contents. Figure 8(a) shows the relationship between the moisture content and the sum of the surface particle areas of all particle size categories. Figure 8(b) shows the relationship between the moisture content and the sum of the surface particle areas of particle sizes of 25 mm or less. Figure 8(c) shows the relationship between the moisture content and the sum of the surface particle areas of particle sizes of more than 25 mm. As shown in Figures 8(a) and 8(b), the sum of the surface particle areas of all particle sizes or those of 25 mm or smaller shows a complex trend: from the dry state to a moisture content of 6% to 8%, the sum of the surface particle areas decreases, and as moisture content increases, the sum of the surface particle areas increases. In contrast, as shown in Figure 8(c), the sum of the surface particle areas of those of larger than 25 mm shows a trend toward a decrease from the dry state to a moisture content of 6%, but as moisture content increases, the change in the sum of the surface particle areas becomes smaller and more constant as the moisture content increases, approaching 10%. The decrease in the sum of the surface particle areas is thought to be due to the adhesion of powdery particles caused by moisture, obscuring the particle outlines. The increase in the sum of the surface particle areas is thought to be due to the formation of pseudoparticles caused by moisture-induced aggregation, but this only occurs in Figures 8(a) and 8(b). Therefore, in order to reduce the effect of changes in moisture content on the correlation shown in Figure 5 (to reduce the effect of the generation of pseudo-particles), it is considered desirable to relate the particle size index on the horizontal axis to the sum of the area of ​​surface particles in the particle size category larger than 25 mm, as shown in Figure 8(c), instead of the sum of the area of ​​surface particles in all particle size categories as shown on the vertical axis of Figure 5.

[0034] Figure 9 shows the relationship between the sum of the surface particle areas of the reference samples 1 of coals No. 1 to No. 5 obtained by varying the moisture content in the range of 6 to 13%, around 10%, and the particle size index (-3 mm fraction) of coals No. 1 to No. 5 shown in Table 1. Figure 9(a) shows the relationship between the sum of the surface particle areas of all particle size categories and the particle size index. Figure 9(b) shows the relationship between the sum of the surface particle areas of the particle size category larger than 25 mm and the particle size index. As shown in Figure 9(a), when considering the sum of the surface particle areas of all particle size categories, the vertical axis values ​​vary widely for the same particle size index value on the horizontal axis. However, as shown in Figure 9(b), when considering the sum of the surface particle areas of particle sizes larger than 25 mm, the vertical axis values ​​vary narrowly for the same particle size index value on the horizontal axis, indicating a high correlation between the two. In other words, the effect of changes in moisture content on the correlation between the two can be reduced.

[0035] Based on the results described above, in step ST1, a calibration curve is created that shows the relationship between the particle size index (-3 mm ratio in this embodiment) and the total area of ​​surface particles in a specific particle size range (a range with particle sizes larger than 25 mm in this embodiment) calculated for each of the multiple reference samples 1. One possible method for creating the calibration curve is to apply an approximation method such as the least squares method to the plotted points shown in Fig. 9(b) and approximate the result with a function such as a polynomial. In this embodiment, the particle size range larger than 25 mm is defined as the specific particle size range, but the present invention is not limited to this. For example, it is also possible to create a calibration curve by sequentially applying tentative particle size ranges as the specific particle size range, and to determine the particle size range that minimizes the error between the calibration curve and the plotted points as shown in Figure 9(b) as the specific particle size range.

[0036] [Process ST2] Step ST2 corresponds to the area calculation step of the present invention. In step ST2, the surface layer of the measurement object, which is made up of particles with an unknown particle size distribution, is repeatedly imaged while changing the deposition state of the measurement object, and a distance image showing the distance from the reference position to the surface layer particles is obtained. As the measurement object, for example, particles C deposited in a tray T can be used, as in the above-mentioned reference sample 1, and distance images can also be obtained using a 3D camera 2, as in the above-mentioned reference sample 1. In step ST2, the particle sizes and areas of the surface particles of the measurement object are calculated based on the distance image, and the sum of the areas of the surface particles in a specific particle size range (in this embodiment, the range with particle sizes larger than 25 mm) is calculated. The sum of the areas of the surface particles in a specific particle size range for the measurement object can be calculated by the calculation device 3, as in the case of the reference sample 1 described above.

[0037] [Process ST3] Step ST3 corresponds to the particle size index calculation step of the present invention. Fig. 10 is an explanatory diagram that schematically illustrates step ST3. In Fig. 10, when the particle size index is x and the sum of the areas of surface particles in a specific particle size range is y, the calibration curve created in step ST1 is expressed by the function y = f(x). Then, the sum of the areas of surface particles in a specific particle size range for the measurement object calculated in step ST2 is assumed to be y1. In this case, as shown in Fig. 10, in step ST3, the particle size index x1 of the object to be measured is calculated based on the calibration curve y = f(x) and the sum y1 of the areas of surface particles in a specific particle size range for the object to be measured. When y = f(x) is a polynomial, a numerical solution such as Newton's method can be used to calculate x1 based on y1.

[0038] As explained above, in the particle size index measurement method according to this embodiment, in step ST1 (calibration curve creation step), multiple brands of reference samples composed of particles with different particle size distributions (and therefore different particle size indexes, which serve as a guide for particle size distribution), are used, and a calibration curve is created showing the relationship between the sum of the areas of surface particles in a specific particle size range, calculated for each of the multiple reference samples based on these distance images, and the particle size index (the mass proportion of particles of a specific particle size or smaller among all the particles constituting the reference sample) previously obtained for each of the multiple reference samples using a sieve. As mentioned above, there is a high correlation between the two, and the effect of changes in moisture content is small, making it possible to create a highly accurate calibration curve. Therefore, in step ST2 (area calculation step), if the sum of the areas of surface particles in a specific particle size range (the same specific particle size range as the reference sample) is calculated based on a distance image of the measurement object composed of particles whose particle size distribution is unknown (and therefore whose particle size index is also unknown), then in step ST3 (particle size index calculation step), the particle size index of the measurement object can be measured accurately based on the calibration curve and the sum of the areas of surface particles in the specific particle size range for this measurement object.

[0039] In this embodiment, the case where the measurement object whose particle size index is measured is piled up in tray T has been described as an example, but the present invention is not limited to this and can also be applied to a case where the measurement object is transported in a piled up state on a belt conveyor. However, when the measurement object is piled up in tray T, the area of ​​the measurement object in the distance image (the area occupied by the measurement object in a plane (horizontal plane in the example shown in FIG. 3) perpendicular to the direction in which the distance image is acquired (up and down direction in the example shown in FIG. 3)) can be considered to be constant. However, when the measurement object is transported in a piled up state on a belt conveyor, the area of ​​the measurement object in the distance image is considered to fluctuate. If the area of ​​the measurement object fluctuates, the sum of the areas of surface particles in a specific particle size range of the measurement object will also fluctuate, which may reduce the measurement accuracy of the particle size index. To prevent this decrease in measurement accuracy, the sum of the areas of surface particles in a specific particle size range, among the two correlated variables represented by the calibration curve created in step ST1, can be normalized by the area of ​​reference sample 1. Then, in step ST3, the particle size index of the object to be measured can be calculated based on the calibration curve and the sum of the areas of surface particles in a specific particle size range for the object to be measured, normalized by the area of ​​the object to be measured. This makes it possible to measure the particle size index of the object to be measured with high accuracy even if the area of ​​the object to be measured varies.

[0040] That is, when the measurement object is transported in a piled state on a belt conveyor, in step ST1, the sum of the areas of surface particles in a specific particle size range is normalized by the area of ​​the reference sample 1 in the distance image, and a calibration curve is created showing the relationship between the normalized sum of the areas of surface particles in a specific particle size range calculated for each of the multiple reference samples 1 and the particle size index previously obtained for each of the multiple reference samples 1.

[0041] Moreover, for step ST2, for example, the following method may be carried out. FIG. 11 is an explanatory diagram schematically illustrating a modified example of step ST2. The upper part of FIG. 11 shows a distance image, and the lower part shows a cross-sectional view. As shown in FIG. 11(a), when the measurement object 4 is transported in a piled state on a belt conveyor BC, in step ST2, images of the surface of the measurement object 4 and the belt conveyor BC are captured while the measurement object 4 is being transported on the belt conveyor BC, and a first distance image is obtained, which is a distance image showing the distance from a reference position to the surface particles of the measurement object 4 and the conveyor belt BC. On the other hand, as shown in Figure 11(b), an image of only the belt conveyor BC is captured when no measurement objects 4 are piled up, and a second distance image, which is a distance image showing the distance from the reference position to the conveyor belt BC, is obtained in advance. Then, in step ST2, a difference image between the first distance image and the second distance image is calculated, and the sum of the areas of surface particles in a specific particle size range for the measurement object 4 is calculated based on the difference image. Also, the area S of the measurement object 4 in the difference image (the area of ​​the hatched region in FIG. 11(c)) is calculated based on the difference image. To calculate the area S of the measurement object 4, for example, the difference image may be binarized using a predetermined small threshold value, and the area of ​​the pixel region having a pixel value equal to or greater than this threshold value may be calculated.

[0042] Finally, in step ST3, the particle size index of the object to be measured 4 is calculated based on the calibration curve created as described above and the sum of the areas of surface particles in a specific particle size range for the object to be measured 4 normalized by the area S of the object to be measured 4 in the difference image.

[0043] According to the above method, even if the measurement object 4 is transported in a piled state on the belt conveyor BC (i.e., even if the area S of the measurement object 4 varies), a calibration curve is created by normalizing the sum of the area of ​​surface particles of a specific particle size range for the reference sample 1 by the area of ​​the reference sample 1 in the distance image, and this calibration curve is applied with the sum of the area of ​​surface particles of a specific particle size range for the measurement object 4 normalized by the area S of the measurement object 4 in the difference image, so that the particle size index of the measurement object 4 can be measured accurately without a decrease in measurement accuracy. [Explanation of symbols]

[0044] 1. Reference sample 2. 3D camera (means for acquiring distance images) 3...Arithmetic unit 4. Measurement target

Claims

1. a calibration curve creation step of preparing reference samples for a plurality of brands, which are samples composed of particles having different particle size distributions depending on the brand, imaging the surface layers of the reference samples while changing the deposition state of the reference samples, obtaining distance images showing the distances from a reference position to the particles in the surface layers, calculating the particle sizes and areas of the particles in the surface layers based on the distance images, and creating a calibration curve showing the relationship between the sum of the areas of particles in a specific particle size range in the surface layers calculated for each of the plurality of reference samples and a particle size index which is the mass proportion of particles of a specific particle size or less out of all the particles constituting the reference samples, which has been obtained in advance for each of the plurality of reference samples using a sieve; an area calculation step of imaging a surface layer of a measurement object composed of particles with an unknown particle size distribution while changing the deposition state of the measurement object, obtaining a distance image showing the distance from a reference position to the particles of the surface layer, calculating the particle sizes and areas of the particles of the surface layer based on the distance image, and calculating the sum of the areas of the particles of the surface layer in the specific particle size range; a particle size index calculation step of calculating the particle size index of the object to be measured based on the calibration curve and the sum of the areas of the particles in the surface layer of the specific particle size range of the object to be measured; A particle size index measuring method having the following.

2. The measurement object is transported in a piled state on a belt conveyor, In the calibration curve creation step, the sum of the areas of the particles in the surface layer in the specific particle size range is normalized by the area of ​​the reference sample in the distance image, and the relationship between the normalized sum of the areas of the particles in the surface layer in the specific particle size range calculated for each of the plurality of reference samples and the particle size index previously obtained for each of the plurality of reference samples is created as the calibration curve; In the area calculation step, while the measurement object is being transported on the belt conveyor, images of the surface layer of the measurement object and the belt conveyor are captured to obtain a first distance image which is a distance image showing the distance from a reference position to the particles in the surface layer and the conveyor belt, a difference image is calculated between the first distance image and a second distance image which is a distance image showing the distance from the reference position to the conveyor belt and which has been previously captured by capturing an image of only the belt conveyor, and a sum of the areas of the particles in the surface layer in the specific particle size range is calculated based on the difference image, and the area of ​​the measurement object in the difference image is calculated, 2. The particle size index measuring method according to claim 1, wherein in the particle size index calculation step, the particle size index of the object to be measured is calculated based on the calibration curve and the sum of the areas of the particles in the surface layer of the specific particle size range of the object to be measured normalized by the area of ​​the object to be measured in the difference image.

3. The particle size index measuring method according to claim 1 or 2, wherein the particles are coal particles.

Images