Method for estimating the amount of voids around molded coal

Abstract

To provide a method for estimating voids around molded coal that can easily estimate the voids around molded coal even when any volume of molded coal is combined with any amount of powdered coal. [Solution] A method for estimating the amount of voids around molded coal when blended coal containing molded coal and powdered coal is filled into a container, wherein the cumulative filling loss per unit mass of molded coal is determined for a blended coal sample, and based on the relationship between the particle size ratio of powdered coal and molded coal sample, the moisture content of the powdered coal sample, and the cumulative filling loss per unit mass of molded coal, a relationship equation (I) is determined between the particle size ratio of powdered coal and molded coal, the moisture content of powdered coal, and the cumulative filling loss per unit mass of molded coal, and for blended coal which is a combination of powdered coal and molded coal intended for use in coke production, the particle size ratio of the powdered coal and molded coal intended for use, and the moisture content of the powdered coal intended for use are substituted into relationship equation (I) to calculate an estimated value of the cumulative filling loss per unit mass of molded coal.

Description

[Technical Field] 【0001】 This invention relates to a method for estimating the amount of voids surrounding molded charcoal. [Background technology] 【0002】 Conventionally, in the production of coke used in blast furnace operations, various methods have been considered to maintain good coke strength while increasing the proportion of inferior coal, such as non-coking coal, in the blended coal composed of molded coal and pulverized coal, in order to address the depletion of resources of high-quality, strongly coking coal. In order to obtain the desired coke strength using blended coal containing inferior coal, a coal pretreatment process may be useful. For example, known coal drying processes include the Coal Moisture Coal (CMC) method and the Dry-cleaned and Agglomerated Pre-compaction System (DAPS), while processes for blending molded materials such as molded coal include the Dry-cleaned and Agglomerated Precompaction System (DAPS), molded coal blending methods, and other methods for adjusting the particle size of crushed coal. These are combined as appropriate. 【0003】 According to the molded coal blending method, the overall bulk density of the charged coal can be improved by blending high-density molded coal, potentially improving coke strength. Furthermore, by utilizing the high density of the molded coal, it is possible to concentrate lower-grade coal within the molded coal without reducing coke strength. The molded coal blending method can also be combined with other coal pretreatment processes, such as drying processes and crushing particle size adjustment methods. 【0004】 In the molten coal blending method, voids are created around the molten coal due to the blending process, and these voids may persist even after carbonization. Since these residual voids around the molten coal cause a decrease in coke strength, it is important to produce molten coal with sufficient expandability to fill these voids during carbonization. To understand the expandability of the molten coal needed to fill these voids and to determine the coal blending ratios for the blended coal, it is crucial to accurately estimate the amount of voids around the molten coal in the blended coal charged into the coke oven. 【0005】 Patent Document 1 describes a method for producing coke in which molded coal and blended coal made from crushed coal are charged into a coke oven and carbonized, characterized in that a test device is used to fill a container with molded coal and blended coal made from crushed coal by gravity, a cross-sectional image of the inside of the container is taken by X-ray CT, the maximum width W of the void formed around the molded coal is quantified from the obtained cross-sectional image, and further, the maximum expansion volume of the molded coal during carbonization is measured using the test device to determine the amount of expansion of the molded coal as the change in the equivalent circle diameter Δr (mm) before and after expansion, and if the determined change Δr is less than 40% of the maximum width W (mm), the coal composition constituting the molded coal is changed to determine a coal composition in which the change Δr is 40% or more of the maximum width W, and molded coal produced based on this composition is used. 【0006】 Patent Document 2 describes a method for analyzing the amount of voids around molded charcoal when a blend of charcoal containing molded charcoal and powdered charcoal is filled into a container. The method involves filling a test container with molded charcoal and powdered charcoal by gravity using a test apparatus, capturing a cross-sectional image of the inside of the test container using X-ray CT, and determining the amount of voids around the molded charcoal by 3D analysis of the obtained cross-sectional image. In the 3D analysis, high-density areas where the density exceeds a predetermined value and low-density areas where the density is below the predetermined value are defined, the high-density areas are filtered using predetermined shape parameters to define the molded charcoal area, and optionally, areas within the high-density areas other than the molded charcoal area and whose volume exceeds a predetermined value are defined as agglomerated charcoal areas. The agglomerated charcoal areas are excluded from the analysis by being treated as having no pixel data, and an expansion process is performed (n+1) times, expanding the periphery of the molded charcoal area in a similar shape to the molded charcoal area by 1 unit volume each time. The above procedure is followed, where n is a natural number, and (n+1) is the number at which the average density of the region whose volume increased in the (n+1)th expansion treatment is approximately the same as the average density of the region whose volume increased in the nth expansion treatment. The average density of the region whose volume increased in the nth and subsequent expansion treatments is used as a threshold. For each expansion treatment, the region whose density exceeds the threshold is defined as the pulverized coal portion, and the region whose density is below the threshold is defined as the void portion. For each expansion treatment, the difference between the average density of the pulverized coal portion and the average density of the void portion is calculated. The amount of filling reduction is calculated by multiplying the volume of the region whose volume increased in each expansion treatment by the difference. The amount of filling reduction is accumulated over the total number of expansion treatments to calculate the cumulative amount of filling reduction, and the cumulative amount of filling reduction is used as an indicator of the amount of voids around the molded coal. [Prior art documents] [Patent Documents] 【0007】 [Patent Document 1] Japanese Patent Publication No. 2014-224242 [Patent Document 2] Japanese Patent Publication No. 2024-059325 [Overview of the project] [Problems that the invention aims to solve] 【0008】 The method described in Patent Document 1 evaluates the voids around molded coal in a blend of molded coal and powdered coal as width using two-dimensional image analysis. Therefore, there is still room for improvement in terms of the accuracy of the evaluation when the voids around the molded coal are too small, or when powdered coal containing agglomerate is used. On the other hand, the method described in Patent Document 2 evaluates the voids around molded coal by accumulating the packing loss, which is a value that takes density and volume into account, using 3D image analysis. In the method described in Patent Document 2, if the volume of molded coal is constant, the accumulated packing loss can be calculated regardless of the particle size composition of the powdered coal. However, if the volume of molded coal is changed, it was necessary to re-evaluate the accumulated packing loss. 【0009】 One aspect of the present invention aims to solve the above problems and provide a method for estimating voids around molded charcoal that can easily estimate the voids around molded charcoal even when any volume of molded charcoal and any powdered charcoal are combined. [Means for solving the problem] 【0010】 The gist of this invention is as follows: [1] A method for estimating the amount of voids around molded coal when a blend of coal containing molded coal and powdered coal is filled into a container, The estimation method described above is Multiple blended coal samples are prepared by combining multiple types of powdered coal samples, each selected to have different particle size configurations and different moisture content levels, with multiple types of molded coal samples, each selected to have different volume levels. For each blended coal sample, the cumulative packing loss obtained according to the following analysis method (A) is divided by the mass of molded coal to obtain the cumulative packing loss per unit mass of molded coal. Based on the relationship between the particle size ratio of the pulverized coal sample and the molded coal sample, the moisture content of the pulverized coal sample, and the cumulative packing loss per unit mass of molded coal, an equation (I) relating the particle size ratio of the pulverized coal and molded coal, the moisture content of the pulverized coal, and the cumulative packing loss per unit mass of molded coal is obtained. For blended coal, which is a combination of pulverized coal and molded coal intended for use in coke production, the particle size ratio of the pulverized coal and molded coal intended for use, and the moisture content of the pulverized coal intended for use are substituted into the above relational formula (I) to calculate an estimated cumulative packing loss per unit mass of molded coal, and this estimated cumulative packing loss per unit mass of molded coal is used as an indicator of the void amount around the molded coal. It is a method, The aforementioned analysis method (A) is: Using a test apparatus, molded coal and powdered coal are filled into a test container by gravity. Cross-sectional images of the inside of the test container were acquired using X-ray CT. The amount of voids around the molded coal was determined by 3D analysis of the obtained cross-sectional images. In the aforementioned 3D analysis, A high-density section where the density exceeds a predetermined value and a low-density section where the density is less than or equal to the predetermined value are defined, The high-density portion is filtered using a predetermined shape parameter to define the molded char portion. Optionally, within the high-density portion, any region other than the molded charcoal portion and whose volume exceeds a predetermined value is defined as the agglomerated charcoal portion, and this agglomerated charcoal portion is excluded from the analysis by being treated as having no pixel data. An expansion process is performed (n+1) or more times, starting from the periphery of the molded charcoal portion and expanding by 1 unit volume each time in a shape similar to the molded charcoal portion, where n is a natural number. The (n+1) mentioned above is the number at which the average density of the region whose volume increased in the (n+1)th expansion process is approximately the same as the average density of the region whose volume increased in the nth expansion process. The average density of the region whose volume increased in the nth and subsequent expansion processes is used as the threshold. In each expansion treatment, the region whose volume increases is defined as follows: the region where the density exceeds the threshold is defined as the pulverized coal portion, and the region where the density is below the threshold is defined as the void portion. For each expansion treatment, the difference between the average density of the pulverized coal portion and the average density of the void portion is calculated for the region whose volume has increased. The amount of filling reduction is calculated by multiplying the volume of the region that has increased in volume during each expansion process by the difference value. Calculating the integrated filling reduction amount by integrating the filling reduction amount with the total number of full expansion processes is a method for estimating the void volume around formed coal [2] A method for estimating the void volume around formed coal that occurs when a blended coal containing formed coal and pulverized coal is filled into a container, The estimation method is preparing a plurality of types of blended coal samples, each of which is a combination of each of a plurality of types of pulverized coal samples selected so as to include levels with different particle size compositions and levels with different moisture contents, and each of a plurality of types of formed coal samples selected so as to include levels with different volumes, For each blended coal sample, dividing the integrated filling reduction amount obtained according to the following analysis method (A) by the volume of the formed coal to obtain the integrated filling reduction amount per unit volume of the formed coal, and based on the relationship between the particle size ratio of the pulverized coal sample and the formed coal sample, the moisture content of the pulverized coal sample, and the integrated filling reduction amount per unit volume of the formed coal, obtaining a relational expression (I) between the particle size ratio of the pulverized coal and the formed coal, the moisture content of the pulverized coal, and the integrated filling reduction amount per unit volume of the formed coal, For the blended coal, which is a combination of the pulverized coal and the formed coal planned to be used in coke production, substituting the value of the particle size ratio of the planned pulverized coal and the planned formed coal and the value of the moisture content of the planned pulverized coal into the relational expression (I) to calculate an estimated value of the integrated filling reduction amount per unit volume of the formed coal, and using the estimated value of the integrated filling reduction amount per unit volume of the formed coal as an index of the void volume around the formed coal is a method, and the analysis method (A) is filling the formed coal and the pulverized coal into a test container by natural fall using a test apparatus, imaging a cross-sectional image inside the test container by X-ray CT, and obtaining the void volume around the formed coal by 3D analysis of the obtained cross-sectional image, in the 3D analysis, defining a high-density part with a density exceeding a predetermined value and a low-density part with a density not exceeding the predetermined value, respectively, Optionally, within the high-density portion, any region other than the molded charcoal portion and whose volume exceeds a predetermined value is defined as the agglomerated charcoal portion, and this agglomerated charcoal portion is excluded from the analysis by being treated as having no pixel data. An expansion process is performed (n+1) or more times, starting from the periphery of the molded charcoal portion and expanding by 1 unit volume each time in a shape similar to the molded charcoal portion, where n is a natural number. The (n+1) mentioned above is the number at which the average density of the region whose volume increased in the (n+1)th expansion process is approximately the same as the average density of the region whose volume increased in the nth expansion process. The average density of the region whose volume increased in the nth and subsequent expansion processes is used as the threshold. In each expansion treatment, the region whose volume increases is defined as follows: the region where the density exceeds the threshold is defined as the pulverized coal portion, and the region where the density is below the threshold is defined as the void portion. For each expansion treatment, the difference between the average density of the pulverized coal portion and the average density of the void portion is calculated for the region whose volume has increased. The amount of filling reduction is calculated by multiplying the volume of the region that has increased in volume during each expansion process by the difference value. The cumulative amount of filling loss is calculated by accumulating the aforementioned filling loss over the total number of expansion treatments. The method is A method for estimating the amount of voids around molded coal. [Effects of the Invention] 【0011】 According to one aspect of the present invention, a method for estimating voids around molded charcoal is provided, which allows for the easy estimation of voids around molded charcoal even when any volume of molded charcoal and any powdered charcoal are combined. [Brief explanation of the drawing] 【0012】 [Figure 1] This is a schematic diagram illustrating the image processing flow. [Figure 2] This is a schematic diagram illustrating the amount of filling loss. [Figure 3]This figure shows the relationship between the estimated cumulative packing loss per unit mass of molded coal obtained from a regression equation (Example 1) and the cumulative packing loss per unit mass of molded coal obtained from image analysis (Reference Example 1). [Figure 4] This figure shows the relationship between the estimated cumulative packing loss per unit volume of molded coal obtained from the regression equation (Example 2) and the cumulative packing loss per unit volume of molded coal obtained from image analysis (Reference Example 2). [Modes for carrying out the invention] 【0013】 The following describes exemplary embodiments of the present invention (which may also be referred to as these embodiments in this disclosure), but the present invention is not limited to these embodiments. 【0014】 One aspect of the present invention provides a method for estimating the amount of voids around molded coal when a blend of molded coal and powdered coal is filled into a container. In this estimation method, Multiple blended coal samples are prepared by combining multiple types of powdered coal samples, each selected to have different particle size configurations and different moisture content levels, with multiple types of molded coal samples, each selected to have different volume levels. For each blended coal sample, the cumulative packing loss obtained according to the following analysis method (A) is divided by the mass of molded coal to obtain the cumulative packing loss per unit mass of molded coal. Based on the relationship between the particle size ratio of the pulverized coal sample and the molded coal sample, the moisture content of the pulverized coal sample, and the cumulative packing loss per unit mass of molded coal, an equation (I) relating the particle size ratio of the pulverized coal and molded coal, the moisture content of the pulverized coal, and the cumulative packing loss per unit mass of molded coal is obtained. For blended coal, which is a combination of pulverized coal and molded coal intended for use in coke production, the particle size ratio of the pulverized coal and molded coal intended for use, and the moisture content of the pulverized coal intended for use are substituted into relational equation (I) to calculate an estimated cumulative packing loss per unit mass of molded coal, and this estimated cumulative packing loss per unit mass of molded coal is used as an indicator of the amount of voids around the molded coal. One aspect of the present invention also provides a method for estimating the amount of voids around molded coal as described above, wherein the cumulative amount of packing loss per unit volume of molded coal is used instead of the cumulative amount of packing loss per unit mass of molded coal. (Analysis method (A)) Using a test apparatus, molded coal and powdered coal are filled into a test container by gravity. Cross-sectional images of the inside of the test container were acquired using X-ray CT. The amount of voids around the molded coal was determined by 3D analysis of the obtained cross-sectional images. In the aforementioned 3D analysis, A high-density section where the density exceeds a predetermined value and a low-density section where the density is less than or equal to the predetermined value are defined, The high-density portion is filtered using a predetermined shape parameter to define the molded char portion. Optionally, within the high-density portion, any region other than the molded charcoal portion and whose volume exceeds a predetermined value is defined as the agglomerated charcoal portion, and this agglomerated charcoal portion is excluded from the analysis by being treated as having no pixel data. An expansion process is performed (n+1) or more times, starting from the periphery of the molded charcoal portion and expanding by 1 unit volume each time in a shape similar to the molded charcoal portion, where n is a natural number. The (n+1) mentioned above is the number at which the average density of the region whose volume increased in the (n+1)th expansion process is approximately the same as the average density of the region whose volume increased in the nth expansion process. The average density of the region whose volume increased in the nth and subsequent expansion processes is used as the threshold. In each expansion treatment, the region whose volume increases is defined as follows: the region where the density exceeds the threshold is defined as the pulverized coal portion, and the region where the density is below the threshold is defined as the void portion. For each expansion treatment, the difference between the average density of the pulverized coal portion and the average density of the void portion is calculated for the region whose volume has increased. The amount of filling reduction is calculated by multiplying the volume of the region that has increased in volume during each expansion process by the difference value. The cumulative amount of filling loss is calculated by accumulating the aforementioned filling loss over the total number of expansion treatments. 【0015】 In this disclosure, "pulverized coal" refers to crushed coal, and includes coal that has been further size-adjusted after crushing, and agglomerated coal if it is present. In this disclosure, "agglomerated coal" refers to coal with an equivalent spherical radius of less than 6 mm obtained by adding a binding filler to pulverized coal (pulverized coal below a 0.3 mm sieve in one embodiment) and pressurizing it. In this disclosure, "molded coal" refers to coal with an equivalent spherical radius of 6 mm or more obtained by adding a binding filler to pulverized coal (pulverized coal below a 0.3 mm sieve in one embodiment) and pressurizing it. 【0016】 By using the cumulative packing loss obtained by the analysis method (A) of this disclosure, it is possible to estimate the stacking packing loss of various types of pulverized coal with different properties without actual image analysis, based on a predetermined relationship between the properties of the pulverized coal, such as particle size composition and moisture content, and the stacking packing loss. Factors that can influence the cumulative packing loss include the particle size composition of the pulverized coal, the moisture content of the pulverized coal, and other factors. However, based on the inventors' prior investigations, it is considered that the particle size composition and moisture content of the pulverized coal have a particularly large influence on the amount of voids around the molded coal. Among the particle size composition, the mass ratio of coarse particles contained in the pulverized coal is thought to have a particularly strong influence on the amount of voids around the molded coal. This is thought to be because if the difference in particle size between the pulverized coal and the molded coal is small, the interparticle gaps are less likely to be filled with particles. Furthermore, it is thought that the moisture content of the pulverized coal affects the amount of voids around the molded coal by affecting the fluidity of the pulverized coal. 【0017】 As described above, the cumulative filling loss is thought to be affected by the degree of difference in particle size between pulverized coal and molded coal. Therefore, when the volume of molded coal is changed, it is necessary to re-examine the relationship between the particle size composition and moisture content of the pulverized coal and the cumulative filling loss, and for this purpose, image analysis must be performed again. 【0018】 The inventors have investigated various methods to accurately calculate the cumulative filling loss without having to perform image analysis again even when the volume of molded coal is changed, and have conceived of calculating an estimated value of the cumulative filling loss per unit mass of molded coal or per unit volume of molded coal using the particle size ratio of pulverized coal and molded coal. According to the method of this embodiment, the cumulative filling loss when using any volume of molded coal and any pulverized coal can be calculated from the particle size ratio of pulverized coal and molded coal and the moisture content of the pulverized coal, and the amount of voids around the molded coal can be grasped more easily. 【0019】 <<Estimation of void space around molded coal using cumulative packing loss per unit mass of molded coal as an indicator>> The following provides a more detailed example of the case where the cumulative fill-down per unit mass of molded coal is used. 【0020】 <Selection of charcoal powder samples> The pulverized coal samples are selected so as to include levels with different particle size configurations and levels with different moisture content. The pulverized coal samples may also include levels with the same particle size configuration or moisture content. In one embodiment, the pulverized coal samples consist of groups of levels where each of the particle size configuration and moisture content has two or more, preferably three or more, different values. From the viewpoint of estimation accuracy, a larger number of different values ​​is advantageous, but from the viewpoint of work efficiency, the number of different values ​​may be 10 or less, or 8 or less, in one embodiment. 【0021】 The particle size composition value used for estimation is, in one embodiment, the value of the mass percentage of coarse particles, and preferably, one value selected from the group consisting of sieve mass percentage and particle size. The size of the molded coal is usually assumed to be 3cc to 130cc. The sieve size for the sieve mass percentage, which is an indicator of the mass percentage of coarse particles in the pulverized coal, more specifically, the mesh opening size according to JIS Z 8801-1, is typically 2mm or larger, preferably 2.8mm or larger, or 5.6mm or larger. From the viewpoint of more accurately estimating the influence of coarse particles in the pulverized coal on the amount of voids around the molded coal, the sieve size may, in one embodiment, be 8mm or smaller, or 6.7mm or smaller. As the particle size of the pulverized coal sample, the mass-average diameter, which is an indicator in which the presence of coarse particles contributes significantly to the value, is used. 【0022】 <Selection of molded charcoal samples> The molded charcoal samples are selected so that they include levels with different volumes. The volume of the molded charcoal may be an actual measured value obtained by the small-quantity apparent density measurement method in accordance with JIS K2151:2004. The molded charcoal samples may also include levels with the same volume. The molded charcoal samples consist of a group of levels with two or more, preferably three or more, different volume values ​​in one embodiment. From the viewpoint of estimation accuracy, a larger number of different values ​​is advantageous, but from the viewpoint of work efficiency, the number of different values ​​may be 10 or less, or 8 or less, in one embodiment. As for the particle size of the molded charcoal samples, for convenience, as the length of each axis differs depending on the shape, making comparison difficult, the spherical equivalent particle size is used. 【0023】 <Analysis of voids around molded coal using analysis method (A)> The analysis of the amount of voids around the molded coal may be performed using "Analysis Method (A)" of this disclosure. An example of this analysis method will be described below. In analysis method (A), a cross-sectional image of the blended coal obtained using X-ray CT (Computed Tomography) is analyzed in 3D. When attempting to evaluate the voids around molded coal by 3D analysis of cross-sectional images, the evaluation results vary depending on which areas are considered voids, as these voids are regions with low packing density but do not have a completely zero density. In this embodiment, the voids around molded coal are evaluated by 3D analysis of cross-sectional images, and in doing so, an index called packing loss is adopted, which takes volume into account by multiplying density by volume in the areas considered to be voids. Evaluation using packing loss allows for highly accurate quantitative evaluation of the amount of voids without constraints on the size and shape of the voids around molded coal. 【0024】 In this disclosure, the expansion process refers to a process in which one voxel (unit volume) (central voxel) and 26 adjacent voxels (peripheral voxels) are considered, the maximum value of the peripheral voxels is calculated, and if it is greater than the value of the central voxel, the value of the central voxel is replaced with the maximum value. In the cross-sectional image, the molded coal is denser than its surroundings (i.e., has a larger X-ray CT value), so by performing this expansion process on the periphery of the molded coal, the molded coal expands in a similar shape at a rate of 1 voxel per process. 【0025】 Figure 1 is a schematic diagram illustrating the image processing flow. In the method of this embodiment, the voids around the molded coal are analyzed by performing 3D analysis on the X-ray CT cross-sectional image acquired in step S11 in steps S12 to S18. Below, an example of the procedure for analyzing the amount of voids around molded coal according to this embodiment will be described with reference to Figure 1. 【0026】 (Step S11) In this step, molded coal and pulverized coal are filled into a test container by gravity using a test apparatus, and a cross-sectional image of the inside of the test container is captured by X-ray CT. The coal constituting each of the molded coal and pulverized coal may be one type or a combination of two or more types. The pulverized coal may include agglomerated coal in one embodiment, or it may not include agglomerated coal in another embodiment. The test apparatus and X-ray CT apparatus may be commercially available devices, and the measurement conditions of the X-ray CT may be set as desired. The voxel size is preferably small and is not particularly limited from the viewpoint of obtaining good analytical accuracy. In one embodiment of the present invention, the size was set to 0.488 mm × 0.488 mm × 0.488 mm. 【0027】 (Step S12) In steps S12 to S18, the software attached to the X-ray CT scanner may be used for 3D analysis. In step S12, the analysis area (hereinafter also referred to as ROI) within the cross-sectional image (hereinafter also referred to as the original image) acquired in step S11 is binarized by density, and high-density areas where the density exceeds a predetermined value and low-density areas where the density is below a predetermined value are defined. The above default value may be set appropriately so that molded coal and agglomerated coal, if present, are classified as high-density areas, and pulverized coal other than agglomerated coal are classified as low-density areas, for example, 1.0 g / cm³. 3 That is acceptable. 【0028】 (Step S13) In this step, the high-density portion described above is filtered using predetermined shape parameters to define the molded char portion. The molded char portion may be defined using the following procedure. a) For the high-density section defined in step S12, small particles (in one embodiment, small particles of 100 voxels or less) are removed to reduce noise. b) Separate the particles in the high-density portion after the processing in a) above so that each particle can be distinguished. Separation may be performed by labeling. c) From the high-density portion after the processing in b) above, the region corresponding to molded coal is extracted by filtering using predetermined shape parameters. The predetermined shape parameters are set so as to accurately define the molded coal portion, and in one embodiment, they may be one or more of the following: Anisotropy, Flatness, Elongation, Volume, etc. A combination of Anisotropy, Flatness, and Elongation is preferred because it allows for easy implementation of the desired filtering regardless of the size of the molded coal. For example, filtering may be performed using Anisotropy < 0.9, Flatness < 0.4, and Elongation > 0.4. d) In the region extracted in c) above, detect and remove any constrictions. Generally, molded charcoal often has constrictions in the burr area. Constrictions may be detected using the watershed method. e) The area after processing d) above is subjected to one shrink (i.e., one voxel shrink), small particle removal using the same procedure as a) above, one expansion (i.e., one voxel expansion), and smoothing in that order, and the remaining area is defined as the molded charcoal portion. 【0029】 (Step S14) This step may only be performed if the pulverized coal contains agglomerated coal. In this step, the agglomerated coal region is defined as the area of ​​the high-density region defined in step S12 that is not the molded coal region defined in step S13 and whose volume exceeds a predetermined value. The agglomerated coal region is treated as having no pixel data. When pulverized coal contains agglomerated coal, the agglomerated coal has an equivalent sphere radius of less than 6 mm, but its density is as high as that of molded coal. If such agglomerated coal is included in the expansion treatment area in addition to molded coal, there is a risk that the voids will not be accurately evaluated. Therefore, if the blended coal contains agglomerated coal, the agglomerated coal is excluded from the analysis. 【0030】 Specifically, the molded char portion defined in step S13 is subtracted from the high-density portion defined in step S12, and the remaining area is labeled. From this labeled area, volume filtering is performed (in one embodiment, Volume > 15 mm). 3 The aggregated charcoal portion is extracted by selecting only the region that satisfies the volume parameter. 【0031】 Furthermore, if the blended coal contains agglomerate, voids are formed not only around the molded coal but also around the agglomerate. When agglomerate is present around the molded coal, it is conceivable that the amount of packing loss around the agglomerate should be included in the packing loss around the molded coal. Therefore, it is conceivable to determine the packing loss around the agglomerate and subtract it from the packing loss around the molded coal. However, according to the inventors' studies, the proportion of the packing loss around the agglomerate included in the packing loss around the molded coal is negligible, and the influence of the presence of voids around the agglomerate on coke strength is negligible. Therefore, in the evaluation of packing loss in this embodiment, the packing loss around the agglomerate does not need to be considered. Specifically, in the 3D analysis of this embodiment, the area corresponding to the agglomerate is excluded from the analysis of this embodiment by being treated as an area without pixel data. In this disclosure, the cumulative packing loss refers to the packing loss around the molded coal. 【0032】 (Step S15) In this step, within the ROI of the original image, the molded charcoal portion defined in step S13 above is subjected to an expansion process of 1 unit volume (i.e., 1 voxel) at a time, starting from the periphery of the molded charcoal portion and expanding in a similar shape to the molded charcoal portion, at least (n+1) times, where n is a natural number. (n+1) is the number at which the average density of the region whose volume increased in the (n+1)th expansion process is approximately the same as the average density of the region whose volume increased in the nth expansion process. Here, "approximately the same" means, in one embodiment, that the average density of the region whose volume increased in the (n+1)th expansion process is within ±0.3% of the average density of the region whose volume increased in the nth expansion process. Here, ±0.3% corresponds to measurement variability and is not limited to this value. 【0033】 After each expansion process is completed, the volume of the region increased by that expansion process (i.e., the volume increase due to that expansion process) and the average density are calculated. The volume corresponds to the number of voxels in the region whose volume increased. The average density is calculated by taking the number average of the CT values ​​of each voxel in the region whose volume increased by the number of voxels in that region. The expansion process is carried out at least until the average density of the region increased by one expansion process becomes approximately the same even if the number of expansion processes is increased. The total number of expansion processes may be (n+1), or it may be more than (n+1), for example, 10 or more, or 20 or more. In one embodiment, the total number of expansion processes may be 30 to 50, for example, 30. 【0034】 (Step S16) In this step, the average density of the region whose volume increased in the nth expansion treatment is used as a threshold. For each expansion treatment, the region whose density exceeds the threshold is defined as the pulverized coal portion, and the region whose density is below the threshold is defined as the void portion. 【0035】 (Step S17) Next, for the region whose volume increased in each expansion treatment, the difference between the average density of the pulverized coal portion and the average density of the void portion is calculated. The average density of the pulverized coal portion is the number average of the CT values ​​of each voxel contained in the pulverized coal portion, multiplied by the total number of voxels in the pulverized coal portion. The average density of the void portion is the number average of the CT values ​​of each voxel contained in the void portion, multiplied by the total number of voxels in the void portion. Next, the amount of filling reduction is calculated by multiplying the volume of the region whose volume increased in each expansion treatment by the above difference value. 【0036】 (Step S18) In this step, the cumulative filling loss is calculated by accumulating the filling loss amount calculated in step S17 over the total number of expansion treatments. Figure 2 is a schematic diagram illustrating the filling loss amount. The filling loss amount calculated as described above in this step changes due to the presence of voids up to a certain number of expansion treatments (n times in one embodiment), but beyond that number (more than (n+1) times in one embodiment), it becomes almost constant because there are no longer any voids in the region increased by the expansion treatment. For example, in Figure 2, the filling loss amount in the region increased by each expansion treatment changes up to the 8th expansion treatment, but from the 9th expansion treatment onward, the filling loss amount becomes almost constant from the value after the 8th expansion treatment. The cumulative value of the filling loss amount over the number of expansion treatments until the filling loss amount becomes almost constant reflects the total amount of voids present around the molded coal. From the above viewpoint, in one embodiment, the cumulative filling loss amount obtained by summing the filling loss amounts over the total number of expansion treatments of (n+1) times or more is useful as an indicator of the amount of voids around the molded coal. 【0037】 <Calculation of cumulative filling loss per unit mass of molded coal> For each blended coal sample, the cumulative packing loss obtained according to the analysis method described above is divided by the mass of the molded coal used in the analysis (in one embodiment, the dry mass) to obtain the cumulative packing loss per unit mass of molded coal. 【0038】 <Derivation of relation (I)> Next, based on the relationship between the particle size ratio of the pulverized coal sample and the molded coal sample, the moisture content of the pulverized coal sample, and the cumulative packing loss per unit mass of molded coal, we derive equation (I) relating the particle size ratio of the pulverized coal and the molded coal sample, the moisture content of the pulverized coal, and the cumulative packing loss per unit mass of molded coal. In one embodiment, equation (I) is expressed as shown in equation (I-1) below. In one embodiment, the particle size ratio of the pulverized coal sample and the molded coal sample is the ratio of [spherical equivalent particle size of the molded coal sample] / [mass-average diameter of the pulverized coal sample]. Cumulative packing loss per unit mass of molded coal = a × [ratio of particle size between pulverized coal and molded coal] + b × [moisture content of pulverized coal] + c (I-1) (In the formula, a and b are coefficients derived from the relationship between the particle size ratio of the pulverized coal sample and the molded coal sample, the moisture content of the pulverized coal sample, and the cumulative packing loss per unit mass of molded coal of the blended coal sample; and c is a constant term derived from the relationship between the particle size ratio of the pulverized coal sample and the molded coal sample, the moisture content of the pulverized coal sample, and the cumulative packing loss per unit mass of molded coal of the blended coal sample.) The method for deriving relation (I) is not limited to this, but may include multivariate analysis such as simple linear regression or multiple linear regression. The analytical procedures for each of these examples will be described below. 【0039】 (Simple linear regression analysis) In one embodiment of the simple regression analysis, the particle size ratio of pulverized coal and molded coal, and the moisture content of pulverized coal are used as variables, and the relationship between each of these and the cumulative filling loss per unit mass of molded coal may be examined using linear regression or polynomial regression, typically linear regression. Specifically, the values ​​of the particle size ratio of pulverized coal and molded coal and the moisture content of pulverized coal are plotted on the x-axis, and the value of the cumulative filling loss per unit mass of molded coal is plotted on the y-axis, and from this plot, a simple regression equation may be obtained, for example, by linear regression using the least squares method. 【0040】 Next, the coefficient a of the above relational equation (I-1) is defined as the value of the regression coefficient (in one embodiment, the slope of the linear regression equation) obtained from a simple regression equation showing the relationship between the particle size ratio of pulverized coal and molded coal and the cumulative filling loss per unit mass of molded coal. The coefficient b of the relational equation (I-1) is defined as the value of the regression coefficient (in one embodiment, the slope of the linear regression equation) obtained from a simple regression equation showing the relationship between the moisture content of pulverized coal and the cumulative filling loss per unit mass of molded coal. The coefficient c may be, for example, 0 or non-zero. In this way, relational equation (I-1) can be obtained as relational equation (I) between the particle size ratio of pulverized coal and molded coal, the moisture content of pulverized coal, and the cumulative filling loss per unit mass of molded coal. 【0041】 (Multiple regression analysis) In one embodiment of the multiple regression analysis, both the particle size ratio of pulverized coal and molded coal and the moisture content of pulverized coal are used as variables, and the relationship between these variables and the cumulative filling loss per unit mass of molded coal may be examined using linear regression or polynomial regression, typically linear regression. Specifically, a multiple regression analysis may be performed with the particle size ratio of pulverized coal and molded coal and the moisture content of pulverized coal as explanatory variables, and the cumulative filling loss per unit mass of molded coal as the dependent variable. According to the inventors' studies, there is no strong correlation between the particle size ratio of pulverized coal and molded coal and the moisture content of pulverized coal, so using these as explanatory variables may be advantageous in performing a significant multiple regression analysis. The regression is not limited to this, but may be performed, for example, by the least squares method. 【0042】 Next, the regression coefficients obtained from a multiple regression equation showing the relationship between the particle size ratio of pulverized coal and molded coal, the moisture content of pulverized coal, and the cumulative filling loss per unit mass of molded coal are set as coefficients a and b of the above relational equation (I-1), and the value of the intercept of the said multiple regression equation is set as coefficient c. In this way, the above relational equation (I-1) can be obtained as relational equation (I) representing the relationship between the particle size ratio of pulverized coal and molded coal, the moisture content of pulverized coal, and the cumulative filling loss per unit mass of molded coal. In multiple regression analysis, relational equation (I-1) with an intercept c (i.e., constant term) of 0 may be obtained by standardization, but standardization may also be omitted. 【0043】 There are no particular limitations on the method for confirming whether the multiple regression analysis was performed effectively, and the test may be performed using standard methods. The significance level may be selected as desired, for example, 5% or 1%. If the regression results do not meet the significance level, the operation of repeating the multiple regression analysis after changing the parameter selected, for example, the particle size ratio of pulverized coal to molded coal, may be repeated until the significance level is met. If the multiple regression analysis with a certain particle size ratio does not meet the significance level, the particle size ratio may be changed and the multiple regression analysis may be repeated. 【0044】 In the above, we have described the case in which the particle size ratio of pulverized coal to molten coal and the moisture content of pulverized coal are used as variables in estimating the amount of voids around molten coal. However, if there are other factors that significantly affect the amount of reduction in stacking capacity, it is possible to use those other factors instead of, or in addition to, the particle size ratio and / or moisture content mentioned above. For example, the above example shows a case where there are two explanatory variables in the multiple regression analysis, but it is also possible to use three or more explanatory variables. 【0045】 <Calculation of estimated cumulative filling loss per unit mass of molded coal> By using the relational equation (I) derived using the procedure exemplified above, an estimated cumulative packing loss per unit mass of molded coal in the blended coal can be calculated. In one embodiment, for blended coal which is a combination of pulverized coal and molded coal to be used in coke production, the particle size ratio of the pulverized coal and molded coal to be used and the moisture content of the pulverized coal to be used are substituted into relational equation (I) above to calculate an estimated cumulative packing loss per unit mass of molded coal. This estimated cumulative packing loss per unit mass of molded coal can be used as an indicator of the amount of voids around the molded coal. 【0046】 <<Estimation of void space around molded coal using cumulative packing loss per unit volume of molded coal as an indicator>> The above describes the case where the cumulative packing loss per unit mass of molded coal is used as an indicator of the amount of voids around molded coal. However, instead of the cumulative packing loss per unit mass of molded coal, the cumulative packing loss per unit volume of molded coal may also be evaluated. The cumulative packing loss per unit volume of molded coal is the value obtained by dividing the cumulative packing loss by the volume of molded coal obtained by the method of this disclosure. In this case, the same procedure as described above in "Estimation of the amount of voids around molded coal using the cumulative packing loss per unit mass of molded coal as an indicator" may be adopted, except that the "cumulative packing loss per unit mass of molded coal" mentioned above in "Estimation of the amount of voids around molded coal using the cumulative packing loss per unit mass of molded coal as an indicator" is replaced with "cumulative packing loss per unit volume of molded coal". Even by this method, the amount of voids around molded coal when any volume of molded coal and any amount of pulverized coal is used can be estimated simply and accurately. [Examples] 【0047】 The following describes exemplary embodiments of the present invention with reference to examples, but the present invention is not limited to these embodiments. 【0048】 <Cumulative reduction in filling volume per unit mass of molded coal> [Reference example 1] <Image analysis of void volume around molded coal> (Coal used) For the analysis, pulverized coal under conditions 1 to 7 shown in Table 1 and molded coal under conditions 1 to 3 shown in Table 2 were used. In Table 1, "pulverized coal" refers to the coal itself obtained by pulverizing pulverized coal, "coarse-grained coal" refers to coal from which fine particles with a particle size of less than 0.3 mm have been removed, "smooth-grained coal" refers to coal with a particle size of 0.3 mm to 3 mm, and "aggregate coal" refers to coal with an equivalent spherical radius of less than 6 mm obtained by adding a tar-based binding agent as a binding agent to the above pulverized coal and then press-molding it. In Table 1, the moisture content is a value measured using the moisture content determination method in accordance with JIS M8812:2006, and the mass-average diameter is a value obtained by weighting the particle size and mass ratio based on the particle size distribution measured using a sieve with a mesh size in accordance with JIS Z8801-1. Furthermore, in Table 2, the cup volume is the value indicated in the molding roll drawing, the spherical equivalent particle size refers to the particle size of spheres with the same actual volume, and the actual volume is the value obtained using the small-volume apparent density measurement method. 【0049】 The voids around the molded coal were analyzed by the amount of filling reduction using the following procedure. 【0050】 (Step S11) Using a test apparatus, molded coal and pulverized coal were filled into a test container by gravity, and cross-sectional images of the inside of the test container were captured using X-ray CT. 【0051】 As the test apparatus, a Mini ASTM apparatus (drop height 1m) was used, which is a 1 / 2 scale version of the ASTM improved bulk density measuring apparatus (Qingtang et al., Coke Circular, 30(11), 13-5(1981)), which is an improved version of the bulk density test in accordance with ASTM D 291-86, with the following conditions improved. 【0052】 The Mini ASTM test container (150 × 150 × 150 mm) was filled with coal by dropping 1.5 kg of pulverized coal of each level, followed by molded coal of each level, and finally the remaining 2.75 kg of pulverized coal of each level. The sample container filled with coal was imaged using an X-ray CT diagnostic device (TSX-201 (Aquilion LB) manufactured by Toshiba Medical Systems Corporation). The X-ray CT imaging conditions were as follows. Under the following imaging conditions, a resolution of 0.488 mm per pixel was obtained. Scan Mode: Helical Tube voltage: 120kV Tube current: 400mA FOV (Field of View): 440mm Image slice thickness: 0.5mm 【0053】 The obtained cross-sectional images were processed using the image analysis software Avizo. All analyses were performed in 3D. The region of interest (ROI) was defined as the central 120mm rectangular area (i.e., the central part (120×120×120mm) within the container dimensions (150×150×150mm)) to exclude wall effects. 【0054】 (Step S12) Based on the X-ray CT value of each pixel within the ROI, the density (BD) is expressed by the following relationship: BD(t / m 3 ) = 0.001 × (X-ray CT value) + 1 The calculation was performed according to the following formula. Density is 1.0 g / cm³. 3 Areas exceeding 1.0 g / cm³ are high-density areas. 3 The following regions were defined as low-density areas. 【0055】 (Step S13) a) Small particles of 100 voxels or less were removed from the high-density region defined in step S12. b) The high-density portion after the processing in a) above was subjected to a labeling process so that the particles could be separated and distinguished one by one. c) From the high-density region after the processing described in b) above, only those particles satisfying the shape parameters of Anisotropy < 0.9, Flatness < 0.4, and Elongation > 0.4 were selected as heteromorphic particles. d) The irregularly shaped particles extracted in c) above were subjected to the watershed method to detect and remove constrictions. e) The irregularly shaped particles after the process in d) above were processed in the following order: 1 voxel shrinkage, removal of small particles using the same procedure as in a), 1 voxel expansion, and smoothing. The remaining area was defined as the molded carbon portion. All expansion and shrinkage processes were performed in a spherical shape (ball dilation / ball erosion). 【0056】 (Step S14) This step was performed only at levels containing agglomerated charcoal. The molded charcoal portion defined in step S13 was subtracted from the high-density portion defined in step S12, and the remaining area was labeled. From this labeled area, Volume > 15 mm 3 Only the region satisfying the volume parameter was selected as the agglomerate carbon region. This region was excluded from the analysis by masking it. 【0057】 (Step S15) Within the ROI of the original image, the molded char portion defined in step S13 above was subjected to an expansion process 30 times, in which one voxel was expanded each time from the periphery of the molded char portion in a shape similar to the molded char portion. 【0058】 (Step S16) After each expansion treatment was completed, the volume of the area increased by that treatment (i.e., the volume increase due to one expansion treatment) and the average density were calculated. After more than 21 expansion treatments, the average density remained almost unchanged, so the average density after 21 expansion treatments was defined as the average density of the pulverized coal portion. This average density was used as a threshold, and the area exceeding this threshold was defined as the pulverized coal portion, while the area below this threshold was defined as the void portion. 【0059】 (Step S17) The average density of the coal pulverized portion was calculated by taking the number average of the densities of each voxel corresponding to the coal pulverized portion, and the average density of the void portion was calculated by taking the number average of the densities of each voxel corresponding to the void portion, and the difference between the average density of the coal pulverized portion and the average density of the void portion was calculated. For each expansion treatment cycle, the amount of filling reduction was calculated according to the following formula. Filling reduction amount = [Volume of the region that increased in volume during each expansion treatment] × [Difference between the average density of the pulverized coal portion and the average density of the void portion] 【0060】 (Step S18) The cumulative amount of filling loss was calculated by accumulating the above-mentioned filling loss over the number of expansion treatments. The results are shown in Table 3. 【0061】 <Calculation of cumulative filling loss per unit mass of molded coal> For each level, the cumulative fill-down amount obtained above was divided by the dry mass of the molded coal to calculate the cumulative fill-down amount per unit mass of molded coal (measured value). The results are shown in Table 4. 【0062】 [Example 1] We investigated a method to estimate the amount of voids around molten coal using factors that affect the voids around the molten coal, without image analysis. From the cumulative filling reduction results shown in Table 3, it can be seen that the size of the voids around the molten coal is influenced by the particle size composition of the pulverized coal (more specifically, the proportion of coarse particles), the moisture content of the pulverized coal, and the particle size of the molten coal. From the perspective of filling the interparticle voids, it is thought that the ratio of the particle size of the pulverized coal to the particle size of the molten coal influences the amount of voids around the molten coal. Therefore, in this example, we decided to use the particle size ratio of pulverized coal to molten coal, and the moisture content of the pulverized coal, as explanatory variables. 【0063】 Furthermore, the cumulative packing loss, which represents the amount of voids around the molded coal, increases with increasing particle size of the molded coal. From the perspective of particle packing efficiency, the larger the ratio of the particle size of molded coal to the particle size of pulverized coal, the better the particle packing efficiency. As the particle size of molded coal increases, the volume around the molded coal expands, which is thought to increase the cumulative packing loss. Therefore, in this example, we decided to reorganize the amount of voids around the molded coal into the cumulative packing loss per unit mass of molded coal, and used this as the dependent variable. 【0064】 It should be noted that factors influencing the cumulative filling loss are not limited to the particle size ratio of pulverized coal and molded coal and the moisture content of pulverized coal. However, in this example, estimation was performed focusing on the particle size ratio of pulverized coal and molded coal and the moisture content of pulverized coal, which are the main factors that have a large influence on the cumulative filling loss. For convenience, in this example, sieves with mesh openings of 2.8 mm and 5.6 mm in accordance with JIS Z 8801-1 are referred to as 3 mm sieves and 6 mm sieves, respectively. 【0065】 In this example, multiple regression analysis was performed using linear regression by least squares, with the particle size ratio of pulverized coal and molded coal and the moisture content of pulverized coal as explanatory variables, and the cumulative packing loss per unit mass of molded coal as the dependent variable. For the regression analysis, the values ​​for each level shown in Table 3 were used for the particle size ratio of pulverized coal and molded coal and the moisture content of pulverized coal, and the value from Reference Example 1 shown in Table 4 was used for the cumulative packing loss per unit mass of molded coal. Using the regression coefficient, which is the slope of the obtained regression equation, the relationship equation (I) between the particle size ratio of pulverized coal and molded coal, the moisture content of pulverized coal, and the cumulative packing loss per unit mass of molded coal was derived. At this time, a significance level of 5%, which is common in multiple regression analysis, was adopted. Note that abs-g means the absolute value of g. Cumulative packing loss per unit mass of molded coal (×10 -2 abs-g) = [Ratio of spherical equivalent particle size of molded coal / mass-average diameter of powdered coal] × (-0.0318) + [Moisture content of powdered coal (mass%)] × 0.1012 + 1.4622 (I) 【0066】 The cumulative filling loss per unit mass of molded coal (estimated value) was calculated by substituting the values ​​for each level shown in Table 3 into the relationship equation (I) [ratio of spherical equivalent particle size of molded coal to mass-average diameter of pulverized coal] and [moisture content of pulverized coal (mass%)]. The results are shown in Table 4. 【0067】 Figure 3 is a plot of the values ​​shown in Table 4, illustrating the relationship between the measured value (Reference Example 1) and the estimated value (Example 1) of the cumulative filling loss per unit mass of molded coal. The upper tail probability (significance F) of the ratio of variances between the two groups, the estimated cumulative filling loss per unit mass of molded coal calculated according to the relational equation (I) obtained in Example 1 and the measured value of the cumulative filling loss per unit mass of molded coal obtained in Reference Example 1, was 0.0047, which is less than 0.05, so it was determined that the regression analysis was meaningful. In addition, the P-values, which are the probabilities that the regression coefficients for the intercept, the particle size ratio of pulverized coal to molded coal, and the moisture content of pulverized coal are at extreme values, were 0.0029, 0.0040, and 0.0229, respectively, and all were less than 0.05, so it was determined that each regression coefficient is significant. 【0068】 Furthermore, it was confirmed in advance that there was no strong correlation between the particle size ratio of pulverized coal and molded coal and the moisture content of the pulverized coal, which were used as explanatory variables, by the following method. In the figure showing the relationship between the particle size ratio of pulverized coal and molded coal and the moisture content of the pulverized coal, simple linear regression was performed using the least squares method from the plotted values, and the coefficient of determination (R-squared value) of the regression equation was obtained, and it was confirmed that the tolerance (=1-R-squared value) was greater than 0.1. 【0069】 <Cumulative reduction in filling volume per unit volume of molded coal> [Reference example 2] The procedure was the same as in Reference Example 1, except that the procedure described in the section "Calculation of cumulative filling loss per unit mass of molded coal" was replaced with the following. 【0070】 <Calculation of cumulative filling loss per unit volume of molded coal> For each level, the cumulative filling loss obtained above was divided by the actual volume of molded coal shown in Table 2 to calculate the cumulative filling loss per unit volume of molded coal (measured value). The results are shown in Table 5. 【0071】 [Example 2] A method for estimating the amount of voids around molded coal using factors that affect the voids around molded coal, without image analysis, was investigated. Similar to Example 1, the particle size ratio of pulverized coal to molded coal and the moisture content of pulverized coal were used as explanatory variables. In this example, the measured values ​​from Reference Example 2 shown in Table 5 were used, and the cumulative packing loss per unit volume of molded coal was used as the dependent variable. 【0072】 Multiple regression analysis was performed using linear regression by least squares, with the particle size ratio of pulverized coal and molded coal and the moisture content of pulverized coal as explanatory variables, and the cumulative filling loss per unit volume of molded coal as the dependent variable. The particle size ratio of pulverized coal and molded coal and the moisture content of pulverized coal were set to the values ​​shown in Table 3, and the cumulative filling loss per unit volume of molded coal was set to the value calculated in Reference Example 2. Using the regression coefficient, which is the slope of the obtained regression equation, the relationship equation (II) between the particle size ratio of pulverized coal and molded coal, the moisture content of pulverized coal, and the cumulative filling loss per unit volume of molded coal was derived. At this time, a significance level of 5%, which is common in multiple regression analysis, was adopted. Cumulative packing loss per unit volume of molded coal (×10 -2 abs-g) = [Ratio of spherical equivalent particle size of molded coal / mass-average diameter of powdered coal] × (-0.0365) + [Moisture content of powdered coal (mass%)] × 0.1134 + 1.7027 (II) 【0073】 The cumulative filling loss per unit volume of molded coal (estimated value) was calculated by substituting the values ​​for each level shown in Table 3 into the relationship equation (II) [ratio of spherical equivalent particle size of molded coal / mass-average diameter of pulverized coal] and [moisture content of pulverized coal (mass%)]. The results are shown in Table 5. 【0074】 Figure 4 is a plot of the values ​​shown in Table 5, illustrating the relationship between the measured value (Reference Example 2) and the estimated value (Example 2) of the cumulative filling loss per unit volume of molded coal. The upper tail probability (significance F) of the ratio of variances between the two groups, the estimated cumulative filling loss per unit volume of molded coal calculated according to relational equation (II) obtained in Example 2 and the measured value of the cumulative filling loss per unit volume of molded coal obtained in Reference Example 2, was 0.0053, which is less than 0.05, so it was determined that the regression analysis was meaningful. In addition, the p-values, which are the probabilities that the regression coefficients for the intercept, the particle size ratio between pulverized coal and molded coal, and the moisture content of pulverized coal are extreme values, were 0.0029, 0.0043, and 0.0265, respectively, and all were less than 0.05, so it was determined that each regression coefficient is significant. 【0075】 As mentioned above in Example 1, it was previously confirmed that there is no strong correlation between the particle size ratio of pulverized coal and molded coal, which were used as explanatory variables, and the moisture content of the pulverized coal. 【0076】 [Examples 3 and 4, Reference Examples 3 and 4] <Validity evaluation of cumulative packing loss per unit mass or unit volume of molded coal, calculated from the particle size ratio of pulverized coal and molded coal and the moisture content of pulverized coal> As shown in Table 6, at level 10, when pulverized charcoal with a moisture content of 10% by mass was mixed with pillow-type molded charcoal (spherical equivalent particle size 33 mm, mass 21 g) with a cup volume of 15 cc and an actual volume of 18 cc, the cumulative packing loss, cumulative packing loss per unit mass of molded charcoal (Reference Example 3), and cumulative packing loss per unit volume of molded charcoal (Reference Example 4) were determined by image analysis. The image analysis followed the same procedure as in Reference Examples 1 and 2. 【0077】 Furthermore, the particle size ratio of pulverized coal to molded coal and the moisture content of pulverized coal at level 10 were substituted into relational equation (I) obtained in Example 1 and relational equation (II) obtained in Example 2, respectively, to obtain the estimated cumulative filling loss per unit mass of molded coal (Example 3) and the estimated cumulative filling loss per unit volume of molded coal (Example 4). 【0078】 The integrated filling reduction amount per unit mass of the formed carbon was 1.79×10 -2 g in Reference Example 3 (measured value based on image analysis) and 1.73×10 -2 g in Example 3 (estimated value based on the regression equation). The integrated filling reduction amount per unit volume of the formed carbon was 2.09×10 -2 g in Reference Example 4 (measured value based on image analysis) and 1.98×10 -2 g in Example 4 (estimated value based on the regression equation). From the above results, for Level 10, it was confirmed that both the integrated filling reduction amount per unit mass of the formed carbon and the integrated filling reduction amount per unit volume of the formed carbon were generally in agreement between the measured value by image analysis and the estimated value by regression analysis. 【0079】 [Table 1] 【0080】 [Table 2] 【0081】 [Table 3] 【0082】 [Table 4] 【0083】 [Table 5] 【0084】 [Table 6]

Claims

[Claim 1] A method for estimating the amount of voids around molded coal when a blend of coal containing molded coal and powdered coal is filled into a container, The estimation method described above is Multiple blended coal samples are prepared by combining multiple types of powdered coal samples, each selected to have different particle size configurations and different moisture content levels, with multiple types of molded coal samples, each selected to have different volume levels. For each blended coal sample, the cumulative packing loss obtained according to the following analysis method (A) is divided by the mass of molded coal to obtain the cumulative packing loss per unit mass of molded coal. Based on the relationship between the particle size ratio of the pulverized coal sample and the molded coal sample, the moisture content of the pulverized coal sample, and the cumulative packing loss per unit mass of molded coal, a relationship formula (I) is obtained between the particle size ratio of the pulverized coal and molded coal, the moisture content of the pulverized coal, and the cumulative packing loss per unit mass of molded coal. For blended coal, which is a combination of pulverized coal and molded coal intended for use in coke production, the particle size ratio of the pulverized coal and molded coal intended for use, and the moisture content of the pulverized coal intended for use are substituted into the above relational formula (I) to calculate an estimated cumulative packing loss per unit mass of molded coal, and this estimated cumulative packing loss per unit mass of molded coal is used as an indicator of the void amount around the molded coal. It is a method, The aforementioned analysis method (A) is: Using a test apparatus, molded coal and powdered coal are filled into a test container by gravity. Cross-sectional images of the inside of the test container were acquired using X-ray CT. The amount of voids around the molded coal was determined by 3D analysis of the obtained cross-sectional images. In the aforementioned 3D analysis, A high-density section where the density exceeds a predetermined value and a low-density section where the density is less than or equal to the predetermined value are defined, The high-density portion is filtered using a predetermined shape parameter to define the molded char portion. Optionally, within the high-density portion, any region other than the molded charcoal portion and whose volume exceeds a predetermined value is defined as the agglomerated charcoal portion, and this agglomerated charcoal portion is excluded from the analysis by being treated as having no pixel data. An expansion process is performed (n+1) or more times, starting from the periphery of the molded charcoal portion and expanding by one unit volume each time in a shape similar to the molded charcoal portion, where n is a natural number. The (n+1) is the number at which the average density of the region whose volume increased in the (n+1)th expansion process is approximately the same as the average density of the region whose volume increased in the nth expansion process. The average density of the region whose volume increased in the nth and subsequent expansion processes is used as the threshold. In each expansion treatment, the region whose volume increases is defined as follows: the region where the density exceeds the threshold is defined as the pulverized coal portion, and the region where the density is below the threshold is defined as the void portion. For each expansion treatment, the difference between the average density of the pulverized coal portion and the average density of the void portion is calculated for the region whose volume has increased. The amount of filling reduction is calculated by multiplying the volume of the region that has increased in volume during each expansion process by the difference value. The cumulative amount of filling loss is calculated by accumulating the aforementioned filling loss over the total number of expansion treatments. The method is A method for estimating the amount of voids around molded coal. [Claim 2] A method for estimating the amount of voids around molded coal when a blend of coal containing molded coal and powdered coal is filled into a container, The estimation method described above is Multiple blended coal samples are prepared by combining multiple types of powdered coal samples, each selected to have different particle size configurations and different moisture content levels, with multiple types of molded coal samples, each selected to have different volume levels. For each blended coal sample, the cumulative packing loss obtained according to the following analysis method (A) is divided by the volume of molded coal to obtain the cumulative packing loss per unit volume of molded coal. Based on the relationship between the particle size ratio of the pulverized coal sample and the molded coal sample, the moisture content of the pulverized coal sample, and the cumulative packing loss per unit volume of molded coal, a relationship formula (I) is obtained between the particle size ratio of the pulverized coal and molded coal, the moisture content of the pulverized coal, and the cumulative packing loss per unit volume of molded coal. For blended coal, which is a combination of pulverized coal and molded coal intended for use in coke production, the particle size ratio of the pulverized coal and molded coal intended for use, and the moisture content of the pulverized coal intended for use are substituted into the above relational formula (I) to calculate an estimated cumulative packing loss per unit volume of molded coal, and this estimated cumulative packing loss per unit volume of molded coal is used as an indicator of the void amount around the molded coal. It is a method, The aforementioned analysis method (A) is: Using a test apparatus, molded coal and powdered coal are filled into a test container by gravity. Cross-sectional images of the inside of the test container were acquired using X-ray CT. The amount of voids around the molded coal was determined by 3D analysis of the obtained cross-sectional images. In the aforementioned 3D analysis, A high-density section where the density exceeds a predetermined value and a low-density section where the density is less than or equal to the predetermined value are defined, The high-density portion is filtered using a predetermined shape parameter to define the molded char portion. Optionally, within the high-density portion, any region other than the molded charcoal portion and whose volume exceeds a predetermined value is defined as the agglomerated charcoal portion, and this agglomerated charcoal portion is excluded from the analysis by being treated as having no pixel data. An expansion process is performed (n+1) or more times, starting from the periphery of the molded charcoal portion and expanding by one unit volume each time in a shape similar to the molded charcoal portion, where n is a natural number. The (n+1) is the number at which the average density of the region whose volume increased in the (n+1)th expansion process is approximately the same as the average density of the region whose volume increased in the nth expansion process. The average density of the region whose volume increased in the nth and subsequent expansion processes is used as the threshold. In each expansion treatment, the region whose volume increases is defined as follows: the region where the density exceeds the threshold is defined as the pulverized coal portion, and the region where the density is below the threshold is defined as the void portion. For each expansion treatment, the difference between the average density of the pulverized coal portion and the average density of the void portion is calculated for the region whose volume has increased. The amount of filling reduction is calculated by multiplying the volume of the region that has increased in volume during each expansion process by the difference value. The cumulative amount of filling loss is calculated by accumulating the aforementioned filling loss over the total number of expansion treatments. The method is A method for estimating the amount of voids around molded coal.

Images