Method for predicting amount of coke degradation

Abstract

To simply, quickly, and non-destructively achieve prediction of an amount of coke degradation.SOLUTION: A method for predicting an amount of coke degradation when an impact is applied to a coke lump after reaction, comprising: a first step of estimating a critical apparent density of a reacted coke lump based on an X-ray CT image; a second step of obtaining an X-ray CT image of the coke lump after the reaction and performing a three-dimensional image analysis to determine a first low apparent density region held in the high apparent density region and a first high apparent density region held in a low apparent density region; and a third step of predicting an amount of coke degradation by subtracting a coke mass in the first low apparent density region from a sum of a coke mass in the low apparent density region and a coke mass in the first high apparent density region.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method for predicting the amount of coke degradation. [Background technology]

[0002] In blast furnace operation, coke releases heat through a combustion reaction with hot air blown in from the tuyere. It acts as a fuel to generate iron oxide, and as a reducing agent to reduce iron oxide directly with the carbon (C) of the coke itself, or indirectly through CO generated by gasification of the coke. is responsible for:

[0003] In blast furnace operation, a decrease in furnace permeability due to coke powdering is a problem. To maintain stable blast furnace operation, various methods are used to evaluate the properties of the coke charged into the furnace.

[0004] It is known that the porosity of coke increases as it reacts with CO2 and deteriorates, and that there is a close relationship between the high porosity region and the amount of attrition (see, for example, Non-Patent Document 1). Methods for evaluating the porosity of coke after reaction with CO2 include microscopic observation of resin-embedded polished samples, and measuring the true density and apparent density using mercury intrusion porosimetry, X-ray CT, and JIS K2151 to calculate the porosity.

[0005] Patent Document 1 discloses a method for predicting the amount of coke disintegration that occurs due to reaction degradation of coke that has reacted with CO2 in a heated environment, the method comprising: a first step of determining the porosity distribution within the coke lump after reaction with CO2 using an X-ray CT device; a second step of calculating a critical porosity, which is the limit porosity at which the coke lump can withstand disintegration after the reaction; and a third step of predicting the amount of disintegration of the coke lump based on the porosity distribution within the coke lump determined in the first step and the critical porosity of the coke lump calculated in the second step. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2005-089617 [Non-patent literature]

[0007] [Non-Patent Document 1] Iron and Steel: 87(5), 259-266, 2001-05-01 [Non-patent document 2] S.Nomura et al.,ISIJ International,47,831(2007) Summary of the Invention [Problem to be solved by the invention]

[0008] An object of the present invention is to realize a simple, rapid, and non-destructive prediction of the amount of coke breakage. [Means for solving the problem]

[0009] In order to solve the above problems, the method for predicting the amount of coke disintegration according to the present invention is (1) a method for predicting the amount of coke disintegration when an impact is applied to a lump of coke that has reacted with CO2 in a heated environment, the method comprising: a first step of estimating a critical apparent density, which is the limit density at which the lump of coke that has reacted with CO2 in a heated environment can withstand disintegration, based on an X-ray CT image; and defining a region where the apparent density is lower than the critical apparent density as a low apparent density region and a region where the apparent density is equal to or higher than the critical apparent density as a high apparent density region. The method is characterized by comprising: a second step of reacting a lump of coke to be used in a blast furnace with CO2 in a heated environment, acquiring X-ray CT images and performing three-dimensional image analysis to determine a first low apparent density region held in the high apparent density region and a first high apparent density region held in the low apparent density region; and a third step of predicting the amount of coke breakage by subtracting the coke mass in the first low apparent density region from the sum of the coke mass in the low apparent density region and the coke mass in the first high apparent density region.

[0010] (2) The method for predicting the amount of coke breakage described in (1) above, characterized in that the first step comprises: a mass acquisition step of preparing a plurality of coke lumps reacted with CO2 in a heated environment, and performing image analysis of X-ray CT images of each coke lump to determine the coke mass in a region where the apparent density is equal to or less than a predetermined value; a correlation acquisition step of determining the correlation between the amount of coke breakage actually measured when an impact is applied to the coke lump and the coke mass determined in the mass acquisition step; and a critical apparent density determination step of performing the mass acquisition step and the correlation acquisition step multiple times while changing the predetermined value and determining the predetermined value corresponding to the correlation with the highest correlation as the critical apparent density.

[0011] (3) The method for predicting the amount of powdered coke described in (1) above, characterized in that the first step comprises a differential image acquisition step of acquiring X-ray CT images of a coke lump that has been reacted with CO2 in a heated environment and a coke lump that has been impacted, aligning these X-ray CT images and analyzing the images to obtain a two-dimensional or three-dimensional differential image corresponding to the powdered region, a relationship information acquisition step of analyzing the differential image acquired in the differential image acquisition step to obtain a relationship between the apparent density and the amount of powdered coke, and a critical apparent density determination step of determining the critical apparent density based on the information acquired in the relationship information acquisition step.

[0012] (4) The method for predicting the amount of coke degradation described in (1) above, characterized in that the first step includes: a first mass acquisition step of acquiring and analyzing an X-ray CT image of the lump coke that has been reacted with CO2 in a heated environment to determine a coke mass that falls within a predetermined apparent density range; a second mass acquisition step of acquiring and analyzing an X-ray CT image of the lump coke that has been subjected to an impact after the reaction to determine a coke mass that falls within the predetermined apparent density range; a degradation rate calculation step of calculating a degradation rate based on the coke masses acquired in the first mass acquisition step and the second mass acquisition step; a relationship information acquisition step of performing the first mass acquisition step, the second mass acquisition step, and the degradation rate calculation step multiple times while changing the apparent density range to determine a relationship between the apparent density and the degradation rate of the lump coke; and a critical apparent density determination step of determining a critical apparent density based on the information acquired in the relationship information acquisition step.

[0013] (5) A method for predicting the amount of coke powder described in any one of (1) to (4) above, characterized in that in the three-dimensional image analysis of the second step, the first low apparent density region is obtained by performing a closing process multiple times. [Effects of the Invention]

[0014] According to the present invention, the amount of coke breakage can be predicted simply, quickly, and non-destructively. [Brief explanation of the drawings]

[0015] [Figure 1] FIG. 1 is a distribution diagram in which coke lumps are divided into regions based on density regions, etc. [Figure 2] This is an X-ray CT image of coke 2-2 before reaction. [Figure 3] This is an X-ray CT image of coke 2-2 after reaction. [Figure 4] This is an X-ray CT image of coke 3-1 before reaction. [Figure 5] This is an X-ray CT image of coke 3-1 after reaction. [Figure 6]These are X-ray CT images of coke 2-2 and coke 3-1 after the reaction, with each apparent density. [Figure 7] This is a graph showing the relationship between the amount of fines generated and the mass of coke contained in the apparent density region (0.0 to 0.60 g / cm3). [Figure 8] This is a graph showing the relationship between the amount of fines generated and the mass of coke contained in the apparent density region (0.0 to 0.70 g / cm3). [Figure 9] This is a graph showing the relationship between the amount of fines generated and the mass of coke contained in the apparent density region (0.0 to 0.80 g / cm3). [Figure 10] FIG. 10 is an explanatory diagram of a method for determining the number of times of closing processing. [Figure 11] FIG. 10 is a graph showing the relationship between predicted values ​​and actually measured powder amounts. [Figure 12] These are X-ray CT images of coke 2-2 (before post-reaction type I test, after post-reaction type I test, difference image). [Figure 13] FIG. 2 is a diagram showing the relationship between apparent density and amount of powder (coke 2-2). [Figure 14] FIG. 2 is a diagram showing the relationship between apparent density and powdering rate (coke 2-2). DETAILED DESCRIPTION OF THE INVENTION

[0016] A method for predicting the amount of coke disintegration, which is an embodiment of the present invention, is a method for predicting the amount of coke disintegration when an impact is applied to a coke lump that has reacted with CO2 in a heated environment (hereinafter also referred to as a coke lump after reaction), and includes the first to third steps described below.

[0017] (Regarding the first step) The critical apparent density, which is the limit density at which the reacted coke lump can withstand pulverization, is estimated based on X-ray CT images. The reacted coke lump can be obtained by a reaction test in which the coke lump is reacted with CO2. The reaction conditions can be the same as the CRI conditions specified in ISO 18894 (reaction temperature: 1100°C, CO2 gas flow rate: 5 NL / min, reaction time: 2 hours), but can also be set arbitrarily. The preferred condition range is 900-1500°C, where the reaction between CO2 and steam and coke is active. Alternatively, coke lump samples taken from inside a blast furnace can be used instead of a reaction test. The size of the coke lump can be set arbitrarily.

[0018] When the critical apparent density is exceeded, powdering becomes difficult to occur, and the critical apparent density can be determined by Method A, Method B, Method C, etc., which will be explained below.

[0019] (Method A) Multiple coke lumps (reacted coke lumps) with different coal properties and blending conditions are prepared, and multiple X-ray CT images are acquired for each coke lump and analyzed to determine the coke mass contained in the area where the apparent density is below a predetermined value (corresponding to the mass acquisition process). The X-ray CT images can be acquired, for example, by a medical X-ray CT. The imaging conditions may be set appropriately so that the distribution of the apparent density of the coke lump can be grasped. Specific examples of the imaging conditions will be described in the examples below. The coke mass can be determined, for example, by dividing the apparent density region where the apparent density is equal to or less than a predetermined value into multiple regions according to the apparent density value, multiplying the average apparent density and volume of each divided apparent density region, and integrating the results. The coke mass may be determined from a two-dimensional X-ray CT image or from a three-dimensional image constructed from the X-ray CT image. This process is performed for each coke lump. The apparent density value can be calculated from the CT value.

[0020] After determining the mass of coke falling within the region where the apparent density is equal to or less than a predetermined value, the correlation between the mass of coke and the measured amount of disintegration when the reacted lump coke is impacted is determined (corresponding to the correlation obtaining step). The measured amount of disintegration may be determined by placing a sample in an I-type rotating drum used for CSR measurement, rotating it 600 times, removing it, and measuring the mass of the sample that falls below the 9.5 mm sieve relative to the mass of the sample placed in the drum. However, the impact and sieve size can be set arbitrarily. Furthermore, instead of the I-type strength rotation test, other impact tests such as a drum test may also be used. The correlation may be a correlation equation defined by a linear function, which can be obtained by fitting the plotted data to the linear function.

[0021] The mass acquisition step and correlation acquisition step described above are performed multiple times while changing the predetermined value. The specified value cannot be uniquely defined because it differs depending on the type of impact test, but is generally between 0.50 and 0.90 g / cm 3 In order to improve accuracy, it is preferable that the number of predetermined values ​​to be changed is three or more. This gives a linear function for each given value, R 2 The critical apparent density can be determined as a predetermined value corresponding to the correlation equation with a high value (corresponding to the critical apparent density determination step). In a general I-type strength rotation test, the critical apparent density is set to 0.70 g / cm. 3 It is recommended to set it to a certain level.

[0022] (Method B) X-ray CT images of the reacted coke lump and the impacted coke lump are acquired, and these X-ray CT images are aligned and analyzed to obtain two-dimensional or three-dimensional difference images (corresponding to the difference image acquisition step). Unlike Method A, this method is simpler because it only requires the preparation of one type of arbitrary coke lump (the same applies to Method C, which will be described later). The contents of Method A are used to describe the reacted coke lump and the impact method (the same applies to Method C, which will be described later).

[0023] To align the coke mass, it is necessary to obtain identical cross-sectional images of the reacted and impacted coke masses. For example, common 3D image processing software such as Avizo Inspect (Thermo Fisher Scientific) can be used for alignment. For example, the centers of gravity of the reacted and impacted coke masses are aligned, and the positions are finely adjusted so that the sum of the absolute values ​​of the differences in CT values ​​between the superimposed CT images of the reacted and impacted coke masses is small. After alignment, a difference image can be obtained by subtracting the CT value of the impacted coke mass image from the CT value of the reacted coke mass image. If the image is an X-ray CT image, a difference image is obtained for each X-ray CT image. If the image is a 3D image constructed from X-ray CT images, a single difference image is obtained. The difference image visualizes the area pulverized by the impact.

[0024] After obtaining the difference image, the difference image is divided into multiple regions according to the apparent density value, and the average apparent density and volume of each divided apparent density region are multiplied to determine the coke mass of each pulverized apparent density region. This obtains the relationship between apparent density and pulverized coke mass (corresponding to the relational information acquisition step). From the obtained relational information (shown in FIG. 13 below), the apparent density corresponding to the peak value of the coke mass can be determined, and this apparent density can be determined as the critical apparent density (corresponding to the critical apparent density determination step). The peak value of the coke mass in each pulverized apparent density region is the density range with the highest mass ratio among the pulverized coke. Density ranges higher than the peak value density range indicate a low rate of pulverization due to impact. Therefore, it is inferred that the peak value of the coke mass in each apparent density region corresponds to the critical apparent density, which has a good correlation with the amount of pulverization. Therefore, it is desirable to carry out a related information acquisition process so that the peak value can be grasped. Although it is expected that the apparent density corresponding to the peak value will change depending on the impact applied and the reaction rate with CO2, the range of apparent density in which the peak value can be grasped using the I-type rotational strength device under the conditions specified in ISO 18894 is 0.5 to 0.9 g / cm. 3 Therefore, 0.5 to 0.9 g / cm 3 The apparent density corresponding to the peak value may be searched for in the range of .

[0025] (Method C) An X-ray CT image of the reacted coke lump is acquired and analyzed to determine the coke mass within a predetermined apparent density range (corresponding to the first mass acquisition process), and an X-ray CT image of the impacted coke lump is acquired and analyzed to determine the coke mass within a predetermined apparent density range (corresponding to the second mass acquisition process). The coke mass can be calculated by multiplying the average apparent density and volume for each apparent density range based on two-dimensional X-ray CT images or three-dimensional images constructed from X-ray CT images. The apparent density value can be calculated from the CT values.

[0026] The degradation rate is calculated based on the coke masses acquired in the first mass acquisition step and the second mass acquisition step (corresponding to a degradation rate calculation step). The general formula for the powdering rate is as follows: Pulverization rate = (100 - (mass of coke lump after impact in a given apparent density range / mass of coke lump after reaction in a given density range) x 100)

[0027] The above-described first mass acquisition step, second mass acquisition step, and powdering rate calculation step are performed multiple times while changing the apparent density range.

[0028] Through the above steps, a one-to-one relationship between the apparent density range and the powdering rate can be obtained for each change in the apparent density range, and thus, relationship information between the apparent density and the powdering rate of the lump coke (as exemplified in FIG. 14 described later) can be acquired (corresponding to the relationship information acquisition step). Based on the acquired relationship information, for example, the apparent density corresponding to a powdering rate of 50% can be determined as the critical apparent density. An apparent density region with a powdering rate of less than 50% means that the rate of powdering due to impact is low, and an apparent density region with a powdering rate of more than 50% means that the rate of powdering due to impact is high. Therefore, it is inferred that the apparent density region with a powdering rate of 50% corresponds to the critical apparent density, which has a good correlation with the amount of powdering. Therefore, it is desirable to determine the "predetermined apparent density range" in the first mass acquisition process and the second mass acquisition process so that the apparent density at a 50% powdering rate can be determined. Although it is expected that the apparent density corresponding to the peak value will change depending on the impact applied and the reaction rate with CO2, the apparent density range in which the peak value can be determined using the I-type rotational strength device under the conditions specified in ISO 18894 is 0.5 to 0.9 g / cm. 3 Examples include 0.6 to 0.8 g / cm 3 , which is narrower than the above range. 3 It is also possible to search for an apparent density with a powdering rate of 50% within this range.

[0029] (2nd process) In the reacted coke lump, the region where the apparent density is lower than the critical apparent density determined in step 1 is defined as the low apparent density region, and the region where the apparent density is equal to or higher than the critical apparent density is defined as the high apparent density region. The low apparent density region is a region with high porosity, and the high apparent density region is a region with low porosity.

[0030] After reacting coke blocks intended for use in a blast furnace with CO2 in a heated environment, X-ray CT images are acquired and three-dimensional image analysis is performed to identify low apparent density regions (non-pulverized) and high apparent density regions (pulverized). X-ray CT images can be acquired using, for example, a medical X-ray CT, but this is similar to the first step and will not be described here. For three-dimensional image analysis, software capable of closing processing (e.g., Avizo Inspect by Thermo Fisher Scientific) can be used. Closing processing is a common image processing method for shaping, in which an expansion process using image processing is performed X times, followed by a contraction process X times in the same way. X represents multiple values.

[0031] The low apparent density region (non-powdered) is a region of the low apparent density region that is maintained within the high apparent density region, and corresponds to the "first low apparent density region." The high apparent density region (powdered) is a region of the high apparent density region that is maintained within the low apparent density region, and corresponds to the "first high apparent density region." The meaning of "hold" will be explained later.

[0032] The reason for dividing the lump coke into regions in this way will be explained. Generally, low apparent density regions have low density (high porosity) and are therefore prone to pulverization due to insufficient strength, while high apparent density regions have high density (low porosity) and are therefore less prone to pulverization. The inventors of the present invention considered that the low apparent density regions (non-pulverized) and the high apparent density regions (pulverized) do not follow this general rule. The inventors of the present invention considered the reason for this as follows.

[0033] The low apparent density region (non-pulverization) is less likely to pulverize because it is held by the high apparent density region with high strength, which reduces the impact of the drop. Furthermore, the high apparent density region (pulverization) is more likely to separate from the coke lump and turn into powder because the surrounding low apparent density region is more likely to pulverize due to the drop impact. Therefore, by regarding the low apparent density region (non-pulverization) as a region that does not pulverize and the high apparent density region (pulverization) as a region that pulverizes, we considered that the accuracy of estimating the amount of pulverization would improve, and performed the above-mentioned region division in three dimensions. In other words, the technical idea of ​​the present invention is to make a correction to exclude the coke contained in the low apparent density region (non-breakdown) from the breakdown amount and to include the coke contained in the high apparent density region (breakdown) in the breakdown amount.

[0034] The method for determining each density region will be described with reference to Figure 1. Figure 1 is a distribution map in which an X-ray CT image of a certain coke lump is divided into a low apparent density region (powdered), a low apparent density region (non-powdered), a high apparent density region (powdered), and a high apparent density region (non-powdered). These regions are defined by a three-dimensional image, but are shown in two dimensions for convenience of explanation. Unlike the first process, the second process cannot be performed using a two-dimensional image.

[0035] First, a three-dimensional image is constructed from an X-ray CT image of the reacted coke lump, and the image is analyzed to separate the low apparent density region, where the apparent density is lower than the critical apparent density, from the high apparent density region, where the apparent density is equal to or higher than the critical apparent density. This processing can be performed based on the CT value. Next, high apparent density regions whose size (longer diameter of the high apparent density region) is larger than a predetermined value are closed to extract low apparent density regions (non-powdered). These extracted low apparent density regions (non-powdered) are "low apparent density regions retained in high apparent density regions." The predetermined value may be the same as the sieve opening (for example, 9.5 mm) when measuring the strength, but is not limited to this.

[0036] The number of closing processes cannot be uniquely defined because it varies depending on factors such as the size of the voxel. However, the change in the total mass of the high apparent density region (i.e., the high apparent density region (non-powdered) in Figure 1) that is larger than the aforementioned predetermined value and the low apparent density region (non-powdered) extracted by the closing process can be examined, and the number of times when this change decreases (hereinafter also referred to as the reference number) or a number greater than the reference number can be determined to be the appropriate number of closing processes. However, since an increase in the number of closing processes increases the load on image processing, it is desirable to stop the closing process after the reference number is exceeded before the load on image processing becomes excessively large.

[0037] Referring to FIG. 1, the "low apparent density region (non-powdered) held in the high apparent density region (non-powdered)" includes a low apparent density region located in a closed space within the high apparent density region (non-powdered) (in other words, three-dimensionally surrounded by the high apparent density region) and a low apparent density region fitted into a recess formed on the outer surface of the high apparent density region (non-powdered). The recess may or may not have its end closed by the high apparent density region (non-powdered) in the direction normal to the page. The low apparent density region (non-powdered) held in the high apparent density region (non-powdered) can be grasped by appropriately setting the number of closing processes, so the explanation of "holding" will be limited to the above.

[0038] The high apparent density region (powdered) is a high apparent density region that is located in a closed space within the low apparent density region and has a density smaller than the predetermined value described above, and corresponds to a "high apparent density region held in a low apparent density region." The high apparent density region (powdered) can also be determined by image analysis.

[0039] By constructing a three-dimensional image from cross-sectional images obtained by medical X-ray CT and performing the above-mentioned closing process, the three-dimensional position, volume, etc. of each density region can be determined.

[0040] (3rd step) The predicted amount of coke breakage is calculated by subtracting the mass of coke contained in the low apparent density region (non-breakage) from the sum of the mass of coke contained in the low apparent density region and the mass of coke contained in the high apparent density region (breakage). This can be expressed by the following formula: Predicted coke powder amount = Mass of coke contained in the low apparent density region + Mass of coke contained in the high apparent density region (degraded) - Mass of coke contained in the low apparent density region (non-degraded) The low apparent density region is divided into multiple regions according to the apparent density value, and the average apparent density and volume of each divided apparent density region are multiplied and integrated to determine the coke mass contained in the low apparent density region. The high apparent density region (breakdown) is divided into multiple regions according to the apparent density value, and the average apparent density and volume of each divided apparent density region are multiplied and integrated to determine the coke mass contained in the high apparent density region (breakdown). The extracted low apparent density region (non-breakdown) is divided into multiple regions according to the apparent density value, and the average apparent density and volume of each divided apparent density region are multiplied and integrated to determine the coke mass contained in the low apparent density region (non-breakdown). These processes for determining the coke mass can also be performed using the above-mentioned three-dimensional image analysis software.

[0041] According to the prediction method of this embodiment, the amount of coke disintegration after reaction can be predicted simply, quickly, non-destructively, and using a small amount of sample. In addition, the amount of disintegration can be predicted using a small amount of sample taken from the lower part of the blast furnace (for example, near the tuyere).

[0042] (Example) The present invention will be specifically described below with reference to examples. Cokes with different CRI and CSR were produced in a test coke oven by changing the coal blending conditions and coal packing density. Table 1 shows the properties of the coal used. Table 2 shows the coal blending conditions and coke quality (coke strength). The particle size of the coal was 1 mm or less (100% by mass). [Table 1] [Table 2]

[0043] Coke was reacted under the same reaction conditions as those for CRI specified in ISO 18894 (reaction temperature: 1100°C, CO2 gas flow rate: 5 NL / min, reaction time: 2 hours), and the CRI was calculated based on the mass before and after the reaction. Normally, CRI measurements use approximately 200 g (approximately 40-45 grains) of 19-21 mm granular coke, but in this example, three grains of 19-21 mm granular coke were used for each level. Thereafter, an impact of 600 revolutions was applied using an I-type rotational strength device (φ130 mm, length 700 mm) and the amount of powder of 9.5 mm or less generated was measured.

[0044] Table 3 shows the CRI and CSR measured for a single coke grain. [Table 3]

[0045] Cross-sectional images of each granular coke before and after the reaction were taken using a medical X-ray CT (TSX-201A, manufactured by Canon Medical Co., Ltd.) under the following imaging conditions: tube voltage: 120 kV, tube current: 350 mA, imaging area: φ38 mm, number of pixels in the cross-sectional image: 512 × 512 pixels, slice width: 0.5 mm.

[0046] Furthermore, three-dimensional images (voxel images: 0.073 × 0.073 × 0.073) were constructed from cross-sectional images obtained by medical X-ray CT using the three-dimensional image processing software Avizo Inspect (Thermo Fisher Scientific). 3 ) CT value is -1000, water (density 0.998g / cm 3 ) was set to 0, and the apparent density of each voxel was calculated from the CT value.

[0047] Figures 2 and 3 are X-ray CT images (cross-sectional images with a slice pitch of 1.5 mm) of coke 2-2, a single coke with a low CRI, measured before and after the reaction. Figure 2 shows the image before the reaction, and Figure 3 shows the image after the reaction. (a1) to (a16) in Figure 2 correspond to (b1) to (b16) in Figure 3, respectively. In the X-ray CT images, the brightness increases as the apparent density increases. Comparing Figures 2 and 3, it can be seen that the brightness of the image after the reaction decreases overall, and the apparent density of the coke decreases as the solution-loss reaction progresses. Furthermore, the brightness of the coke after the reaction in Figure 3 is lower toward the surface, indicating that the solution-loss reaction progresses and the apparent density decreases. On the other hand, the change in brightness and apparent density inside the coke before and after the reaction is small.

[0048] Figures 4 and 5 are X-ray CT images (cross-sectional images with a slice pitch of 1.5 mm) of coke 3-1, which has a high CRI measured on a single coke grain, before and after the reaction. Figure 4 shows the image before the reaction, and Figure 5 shows the image after the reaction. (a1) to (a29) in Figure 4 correspond to (b1) to (b29) in Figure 5, respectively. It can be seen that the image brightness of Coke 3-1 is significantly lower after the reaction than that of Coke 2-2. This is because the CRI of Coke 3-1 is higher than that of Coke 2-2, and the coke mass after the reaction is smaller. It can also be seen that the solution loss reaction progressed to the inside of the sample, resulting in a decrease in the apparent density of Coke 3-1.

[0049] Figure 6 shows X-ray CT images for each apparent density (8 divisions) obtained by analyzing the X-ray CT images of coke 2-2 after reaction (Fig. 3(b7)) and coke 3-1 after reaction (Fig. 5(b19)). It can be seen that both coke 2-2 and coke 3-1 have low apparent densities on the sample surface and high apparent densities inside the sample. In addition, the apparent density of coke 2-2 after reaction was 0.7 g / cm. 3 Although the above regions are clearly present, the apparent density of coke 3-1 after the reaction is 0.7 g / cm 3 It can be seen that such areas hardly exist.

[0050] The CSR of coke 2-2 (on a 9.5 mm sieve after 600 rotations in the I-type rotation strength test) was 59.9 points, while the CSR of coke 3-1 was 0.0 points. This indicates that the CSR tends to be low when there are many regions with low apparent density in the coke after reaction. Therefore, the relationship between the amount of fines of 9.5 mm or less generated in the I-type rotation strength test and the coke mass in each apparent density region was investigated. The coke mass was calculated using the method described in the embodiment.

[0051] Figures 7 to 9 show the relationship between the amount of fines of 9.5 mm or less (horizontal axis) generated by performing an I-type rotational strength test on the coke after reaction and the coke mass in each apparent density range (vertical axis). The apparent density range in Figure 7 is 0.0 to 0.60 g / cm. 3 The apparent density range in Figure 8 is 0.0 to 0.70 g / cm 3 The apparent density range in Figure 9 is 0.0 to 0.80 g / cm 3 These 0.60g / cm 3 , 0.70g / cm 3 and 0.80 g / cm 3 corresponds to the "predetermined value" described in Method A of the first step. The data plotted in each figure were fitted to a linear function, and it was found that the correlation in Figure 8 was high. In other words, the apparent density ranged from 0.0 to 0.70 g / cm 3 It was confirmed that there is a high correlation between the coke mass contained in the region and the amount of fines generated. 3 was set as the critical apparent density.

[0052] For each level of post-reaction coke, the low and high apparent density regions were determined, and then the low apparent density region (non-powdered) and high apparent density region (powdered) were determined. The closing treatment was performed on the high apparent density region of post-reaction coke 1-2, which was larger than 9.5 mm (corresponding to the sieve opening used for strength measurement). The relationship between the number of closing treatments and the total mass was examined, and the relationship shown in Figure 10 was obtained. The total mass refers to the total mass of the coke contained in the high apparent density region (non-powdered) and the low apparent density region (non-powdered). Based on the results shown in Figure 10, the appropriate number of closing treatments was determined to be 40. The change in the total mass between the high apparent density region (non-powdered) in Figure 1 and the low apparent density region (non-powdered) extracted by the closing treatment was examined, and 40 was determined to be the appropriate number of closing treatments, as this was the number of times when this change decreased.

[0053] Therefore, the closing process was performed 40 times on the high apparent density region larger than 9.5 mm of each coke after the reaction to extract the low apparent density region (non-powdered).In addition, the high apparent density region (powdered) smaller than 9.5 mm that was retained in the low apparent density region (powdered) was identified by image processing. These images were processed using the above-mentioned three-dimensional image processing software Avizo Inspect (manufactured by Thermo Fisher Scientific), and the mass of each density region was determined.

[0054] The predicted amount of disintegration was calculated by subtracting the mass of coke in the low apparent density region (non-disintegrated) from the sum of the masses of coke in the low apparent density region and the high apparent density region (disintegrated), and the results are shown in Table 4. The amount of fines of 9.5 mm or less measured using the above-mentioned I-type rotational strength device is also shown in Table 4 as the actual amount of disintegration. [Table 4]

[0055] The predicted values ​​and the measured powder amounts were plotted in Figure 11 and fitted to a linear function. The obtained linear function has a slope closer to 1 than the linear function in Figure 8, so it was found that the estimation accuracy can be improved by understanding the spatial position of the density region.

[0056] According to the prediction method of this embodiment, the amount of powder can be predicted non-destructively and with a small amount of sample based on an X-ray CT image of the coke lump after reaction, making the prediction method simpler and faster than conventional I-type strength tests, etc.

[0057] An experiment to determine the critical apparent density was also conducted according to Method B in Step 1. Figure 12 shows X-ray CT images of coke 2-2, where (a1) to (a17) are X-ray CT images before the post-reaction I-type test, (b1) to (b17) are X-ray CT images after the post-reaction I-type test, and (c1) to (c17) are difference images. Coke 2-2 before and after the I-type rotational strength test was semi-automatically aligned using the Register Images module (Normalized Mutual Information) (the centers of gravity of the coke before and after the reaction were automatically aligned, then manually adjusted, and finally, automatic fine adjustment was performed using Normalized Mutual Information to reduce the sum of the absolute values ​​of the differences in CT values). A difference image was constructed by subtracting the CT value of the image after the I-type rotation strength test from the CT value of the image before the I-type rotation strength test. This visualized the area that was pulverized in the I-type rotation strength test (600 rpm) for coke 2-2, which had deteriorated due to a solution loss reaction. 3 ) has a CT value of -1000 and water (density 0.998g / cm 3 ) was set to 0, and the apparent density of each voxel was calculated from the CT value. Furthermore, the coke mass in each pulverized apparent density region was calculated based on the method described in the embodiment, and the relationship information in Fig. 13 was obtained. Referring to the same figure, the apparent density of each pulverized coke mass was calculated based on the CT value of 0.7 g / cm3, which corresponds to the peak value of the pulverized coke mass. 3 Since the peak value is consistent with the critical apparent density obtained by Method A, it is possible to estimate the critical apparent density based on the peak value.

[0058] An experiment to determine the critical apparent density was also conducted according to Method C in the first step. X-ray CT images of coke 2-2 used in the example of Method B were analyzed, and the degree of degradation was determined based on the method described in the embodiment, thereby obtaining the relevant information shown in Figure 14. Referring to the figure, the value of 0.7 g / cm3, which corresponds to a degree of degradation of 50%, was used. 3 Since this agrees with the critical apparent density obtained by Method A, it was found that the critical apparent density can be estimated based on a powdering rate of 50%.

[0059] (Variation) Based on the finding that the reaction rate due to solution loss in a blast furnace does not change significantly even if the reactivity of the coke changes, many studies have been conducted on methods for measuring the strength index of coke at a constant reaction rate (see, for example, Non-Patent Document 2). The method of the present invention can be applied to the strength index at a constant reaction rate and to coke blended with biomass by changing the sample to be photographed by X-ray CT.

[0060] (Reference example) The present invention can also be applied to a coke reaction under an H2O gas atmosphere.

Claims

1. CO in a heated environment ２ A method for predicting the amount of coke pulverization when an impact is applied to a lump of coke that has reacted with CO in a heated environment ２ a first step of estimating a critical apparent density, which is a limit density at which the coke lump reacted with the coke can withstand pulverization, based on an X-ray CT image; When the region where the apparent density is lower than the critical apparent density is defined as the low apparent density region, and the region where the apparent density is equal to or higher than the critical apparent density is defined as the high apparent density region, the coke lump to be used in the blast furnace is heated in a CO ２ a second step of obtaining an X-ray CT image and performing three-dimensional image analysis to determine a first low apparent density region held in the high apparent density region and a first high apparent density region held in the low apparent density region; a third step of predicting the amount of coke degradation by subtracting the coke mass in the first low apparent density region from the sum of the coke mass in the low apparent density region and the coke mass in the first high apparent density region; A method for predicting the amount of coke degradation, comprising:

2. The first step is a step of heating the gas containing CO ２ a mass obtaining step of preparing a plurality of lump cokes reacted with the above and analyzing an X-ray CT image of each lump coke to obtain a mass of the coke in a region where the apparent density is equal to or less than a predetermined value; a correlation obtaining step of obtaining a correlation between an actually measured amount of pulverization when an impact is applied to the lump coke and the coke mass obtained in the mass obtaining step; a critical apparent density determining step of performing the mass obtaining step and the correlation obtaining step a plurality of times while changing the predetermined value, and determining the predetermined value corresponding to the correlation with the highest correlation as the critical apparent density; 2. The method for predicting the amount of coke breakage according to claim 1, further comprising:

3. The first step is a step of heating the gas containing CO ２ a difference image acquisition step of acquiring X-ray CT images of the coke lump that has been reacted with the coke lump and the coke lump that has been impacted, and aligning and analyzing these X-ray CT images to obtain a two-dimensional or three-dimensional difference image corresponding to the powdered region; a relational information acquiring step of determining the relationship between the apparent density and the powdering amount by analyzing the difference image acquired in the difference image acquiring step; a critical apparent density determining step of determining a critical apparent density based on the information acquired in the related information acquiring step; 2. The method for predicting the amount of coke breakage according to claim 1, further comprising:

4. The first step is a step of heating the gas containing CO ２ a first mass acquisition step of acquiring an X-ray CT image of the lump coke reacted with the coke and analyzing the image to determine a coke mass falling within a predetermined apparent density range; a second mass acquisition step of acquiring an X-ray CT image of the coke lump obtained by applying an impact to the coke lump after the reaction and analyzing the image to determine a coke mass falling within a predetermined apparent density range; a powdering rate calculation step of calculating a powdering rate based on the coke masses acquired in the first mass acquisition step and the second mass acquisition step; a relationship information acquiring step of determining a relationship between the apparent density and the powdering rate of the lump coke by performing the first mass acquiring step, the second mass acquiring step, and the powdering rate calculating step a plurality of times while changing the apparent density range; a critical apparent density determining step of determining a critical apparent density based on the information acquired in the related information acquiring step; 2. The method for predicting the amount of coke breakage according to claim 1, further comprising:

5. In the three-dimensional image analysis of the second step, the first low apparent density region is obtained by performing a closing process multiple times.

5. The method for predicting the amount of coke breakage according to claim 1, wherein the method comprises:

Images