JP7807657B2 - Method for estimating the fragmentation rate of iron ore pellets - Google Patents

Description

The present invention relates to a method for estimating the rate of powdering of iron ore pellets when they are powdered due to impact during transportation.

When iron ore pellets are transported, they may be subjected to physical impacts, which can cause some of the iron ore pellets to break down into powder. If the iron ore pellets are charged into a blast furnace or direct reduction shaft furnace while still containing the powdered portions (hereinafter referred to as "pellet chips"), the pellet chips will impair the air permeability inside the furnace. For this reason, the iron ore pellets are typically sieved to remove the pellet chips before being charged into a blast furnace, etc. The removed pellet chips are typically transported to a sintering plant, where they are agglomerated and then charged into the blast furnace.

Regarding the pulverization of iron oxide raw materials (pellets and sintered ore) charged into blast furnaces, etc., Patent Document 1 discloses a technique for evaluating the reduction pulverization of pellets, and Patent Document 2 discloses a technique for determining the fracture toughness value of sintered ore.

In Patent Document 1, the porosity and average pore diameter of steelmaking pellets are measured, the reaction zone width is estimated from these measurements, the total crack area is estimated from this reaction zone width, and the reduction disintegration index of the steelmaking pellets is estimated from this total crack area. Here, the reaction zone width is the reciprocal of the slope representing the change in magnetite concentration in the radial direction of the pellet due to reduction. Furthermore, the total crack area is the total crack area per unit volume within the pellet caused by reduction.

In Patent Document 2, the length of the crack formed when a Vickers indenter is pressed into a sample (sintered ore) is calculated, and the fracture toughness value of the sample is calculated based on the crack length, the length of the indentation formed when the Vickers indenter is pressed into the sample, and the Vickers hardness of the sample. Here, the crack length is calculated from the AE energy calculated from the AE wave when the Vickers indenter is pressed into the sample, the sample thickness, and the porosity of the sample.

JP 2016-60920 A JP 2018-44210 A

The object of the present invention is to estimate the pulverization of iron ore pellets caused by impacts during transportation. Patent Document 1 focuses on pulverization when steelmaking pellets are reduced (reduction pulverization), and does not focus on pulverization of iron ore pellets caused by impacts during transportation, as in the present invention. Furthermore, Patent Document 2 estimates the fracture toughness value of sintered ore, and does not estimate pulverization of iron ore pellets caused by impacts during transportation, as in the present invention.

The present invention provides a method for estimating the pulverization rate, which is the ratio of the mass of pulverized material to the mass of the iron ore pellets before pulverization, for iron ore pellets that are pulverized due to impact during transportation. First, voids contained in the iron ore pellets are extracted based on three-dimensional images obtained by X-ray CT scanning of the iron ore pellets before pulverization. Next, of the extracted voids, voids with a sphericity below a threshold are identified as cracks, and the crack volume rate, which is the ratio of the total volume of cracks to the apparent volume of the iron ore pellets, is calculated. Next, the pulverization rate is calculated from the calculated crack volume rate based on the correlation between the crack volume rate and the pulverization rate. Here, the correlation between the crack volume rate and the pulverization rate is expressed as a linear function and is calculated in advance.

The threshold value can be set to 0.5 [-]. The powdered material can be particles with a particle size of 6.3 mm or less. The correlation between the crack volume fraction and the powdered rate can be calculated for each iron ore pellet transport path. The sphericity can be calculated from the volume and surface area of the voids in the three-dimensional image.

According to the present invention, by focusing on the correlation between the crack volume fraction and the pulverization rate, it is possible to estimate the pulverization rate from the crack volume fraction of iron ore pellets before pulverization.

1 is a flowchart illustrating a method for estimating the fineness rate of iron ore pellets. FIG. 1 shows an example X-ray image of an iron ore pellet in one cross section. FIG. 1 is a diagram showing an example of an X-ray image of an iron ore pellet in a cross section, in which voids are extracted. FIG. 1 is a diagram showing an example of an image in which voids (cracks) are extracted from an X-ray image of an iron ore pellet in one cross section. FIG. 1 is a diagram showing an example of an image in which voids (pores) are extracted from an X-ray image of an iron ore pellet in one cross section.

This embodiment estimates the pulverization rate when iron ore pellets are pulverized by impacts that occur during transportation. The pulverization referred to here does not include pulverization of iron ore pellets due to reduction in a furnace. The transport route for iron ore pellets is the route taken from immediately after production until they are charged into a blast furnace, etc., and it is during loading and unloading of iron ore pellets (for example, loading and unloading from a ship) that the iron ore pellets are most susceptible to impacts, making them more likely to pulverize. The estimation of the pulverization rate assumes that the transport route for iron ore pellets is the same, and assumes that the impacts acting on iron ore pellets along the same transport route are the same.

Iron ore pellets are fired products made by agglomerating fine iron ore of a specified particle size (for example, 100 μm or less) into agglomerates of about 10 mm in size, and are used as a raw material in blast furnace processes and direct reduction processes. The manufacturing process for iron ore pellets mainly involves a granulation process and a firing process. In the granulation process, a granulator is used to granulate fine iron ore with adjusted particle size and moisture content to produce green pellets. In the firing process, the green pellets are heated and fired to produce iron ore pellets.

The pulverization of iron ore pellets due to impact can be broadly divided into pulverization due to wear on the surface of the iron ore pellets (hereinafter referred to as "abrasion pulverization") and pulverization due to cracks inside the iron ore pellets (hereinafter referred to as "crack pulverization"). Cracks in iron ore pellets occur when the internal pressure of the raw pellets increases due to the rapid evaporation of water during the firing process and the thermal decomposition of the water of crystallization in the iron ore.

Abrasion and fragmentation are thought to occur most frequently when the matrix strength of the iron ore pellets is reduced due to the firing temperature being lower than the target temperature or the firing time being shorter than the target time (so-called insufficient firing) during the firing process. However, in normal firing processes, the firing temperature and firing time are properly managed, so abrasion and fragmentation are thought to be less likely to occur. For this reason, cracking and fragmentation are thought to be the predominant form of fragmentation of iron ore pellets due to impact.

(Method for estimating the powdering rate of iron ore pellets)
A method for estimating the iron ore pellet fragmentation rate Rp will be described with reference to the flowchart shown in Fig. 1. As will be described later, the iron ore pellet fragmentation rate Rp is estimated based on the crack volume rate Rvc of the iron ore pellet. First, the powdering rate Rp and the crack volume rate Rvc will be described below.

The pulverization rate Rp is the ratio (wt%) of the total mass M2 of pulverized material generated by the pulverization of iron ore pellets due to impact to the mass M1 of the iron ore pellets before pulverization, and is expressed by the following formula (1):

The iron ore pellets before pulverization can be iron ore pellets immediately after they have been produced by the above-described production process, but before they have been transported, which would cause pulverization. As mentioned above, given that iron ore pellets are most susceptible to impact and pulverization during loading and unloading (e.g., loading and unloading a ship), the iron ore pellets before pulverization can be iron ore pellets before loading and unloading (e.g., before being loaded onto a ship). Meanwhile, pulverized material can be defined as particles with a predetermined particle size or less, which can be, for example, 6.3 mm.

The crack volume ratio Rvc is the ratio of the total volume V2 of all cracks present in the iron ore pellets (before pulverization) to the apparent volume V1 of the iron ore pellets (before pulverization), and is expressed by the following formula (2):

The apparent volume V1 is determined from the surface shape of a three-dimensional image obtained by performing an X-ray CT scan on the iron ore pellets (before pulverization). The iron ore pellets contain voids, and the volume of these voids is also included in the apparent volume V1. The total volume V2 of cracks is determined by identifying cracks in the three-dimensional image obtained by performing an X-ray CT scan on the iron ore pellets (before pulverization). In other words, all cracks contained in the iron ore pellets (before pulverization) are identified, and the total volume V2 is the sum of the volumes of these cracks. The method for identifying cracks will be described later.

In step S101, an X-ray CT scan is performed on the iron ore pellets (before pulverization). The X-ray CT scan can obtain multiple X-ray images of the iron ore pellets in cross sections perpendicular to the scanning direction. Figure 2 shows an example X-ray image of an iron ore pellet in one cross section. A three-dimensional image of the iron ore pellets can be obtained by superimposing the multiple X-ray images described above.

In step S102, voids contained in the iron ore pellets are extracted based on the three-dimensional images obtained by the X-ray CT scan in step S101. Specifically, by performing binarization processing on the X-ray images (cross-sectional images) of the iron ore pellets, voids are distinguished from non-voids (hereinafter referred to as "non-voids") in each of the X-ray images. The non-voids are the parts that correspond to the matrix of the iron ore pellets. Here, the matrix refers to the combined part of the solid and microscopic pores (voids smaller than the spatial resolution) that cannot be detected by the X-ray CT scan.

In the binarization process, a threshold value is set for brightness values between 0 and 255, and parts with brightness values above the threshold are distinguished from parts with brightness values below the threshold. Here, the brightness value of void parts is smaller than the brightness value of non-void parts, so parts with brightness values below the threshold are void parts, and parts with brightness values above the threshold are non-void parts. This makes it possible to extract the void parts (three-dimensional shapes) contained in the iron ore pellets. In addition to the binarization process, image processing (for example, top-hat transformation processing) required to extract void parts can also be performed.

When extracting void areas, one continuous space surrounded by non-void areas constitutes one void area, and iron ore pellets typically have multiple void areas. Figure 3 shows an example of an image in which void areas have been extracted from an X-ray image of an iron ore pellet in a cross section (corresponding to Figure 2).

In step S103, the sphericity Ψm of each void portion is measured. The sphericity Ψm is calculated using the following formula (3). The sphericity Ψm is a value of 1.0 [-] or less, and a sphericity Ψm of 1.0 indicates a perfect sphere.

In the above formula (3), Ψm is the sphericity [-], Vv is the volume of one void portion [mm ³ ], and Av is the surface area of one void portion [mm ² ]. The volume Vv and surface area Av can be determined from a three-dimensional image obtained by X-ray CT scanning. In other words, the volume Vv and surface area Av can be determined by identifying each void portion contained in the iron ore pellet.

In step S104, cracks are identified from the void portions extracted in step S102 based on the sphericity Ψm measured in step S103. Typically, void portions contained in iron ore pellets contain pores and cracks, but to calculate the crack volume fraction Rvc described above, only the cracks contained in the void portions are identified. Considering the respective formation processes of pores and cracks, pores have shapes that are close to spherical, while cracks have shapes that cannot be described as spherical. Therefore, pores and cracks can be distinguished by their shapes (i.e., sphericity Ψm).

Specifically, a threshold value Ψth for sphericity Ψm is set to distinguish between pores and cracks, and voids where the sphericity Ψm is equal to or less than the threshold value Ψth are identified as cracks. Voids where the sphericity Ψm is greater than the threshold value Ψth are considered to be pores. Figure 4 shows an example of an X-ray image (corresponding to Figure 2) of an iron ore pellet at a cross section in which voids (cracks) have been extracted. Figure 5 shows an example of an X-ray image (corresponding to Figure 2) of an iron ore pellet at a cross section in which voids (pores) have been extracted.

The threshold value Ψth is the boundary value of the sphericity Ψm used to determine whether a void portion corresponds to a pore or a crack, and can be set to a value smaller than 1.0 [-]. Here, the threshold value Ψth is preferably set to 0.5 [-]. If the threshold value Ψth is set to a value smaller than 0.5, the evaluation to identify cracks is more likely to be underestimated, and void portions that are actually recognized as cracks may not be identified as cracks. On the other hand, if the threshold value Ψth is set to a value larger than 0.5, the evaluation to identify pores is more likely to be underestimated, and void portions that are actually recognized as pores may be identified as cracks.

In step S105, the crack volume fraction Rvc of the iron ore pellet is calculated. The crack volume fraction Rvc is calculated using the above formula (2), and the total volume V2 shown in the above formula (2) is the sum of the volume Vv of voids identified as cracks within the volume Vv shown in the above formula (3). Note that the apparent volume V1 shown in the above formula (2) can be calculated from the surface shape of the three-dimensional image obtained by the X-ray CT scan in step S101, as described above.

In step S106, the powdering rate Rp is calculated from the crack volume rate Rvc calculated in step S105. It has been found that there is a correlation between the powdering rate Rp and the crack volume rate Rvc, expressed as a linear function in the following equation (4). If this correlation is calculated in advance, the powdering rate Rp can be calculated from the crack volume rate Rvc.

In the above equation (4), a and b are coefficients that can be determined in advance. Coefficient a is a positive value, and coefficient b is a positive or negative value. Since coefficients a and b depend on the transport route and type of iron ore pellets, they can be determined in advance for each of these dependent factors.

The powdering rate Rp and crack volume rate Rvc can be measured in advance by using iron ore pellets of the same type as the iron ore pellets for which the powdering rate Rp is to be estimated and transporting them along the same transport route as the iron ore pellets for which the powdering rate Rp is to be estimated. Based on these measurements, the correlation between the powdering rate Rp and the crack volume rate Rvc (the above formula (4)) can be determined in advance.

As described above, estimating the powdering rate Rp assumes that the iron ore pellets are transported along the same transport route, and that the impacts acting on the iron ore pellets during transport are equivalent. Therefore, if the iron ore pellets are transported along different routes, the correlation between the powdering rate Rp and the crack volume rate Rvc is calculated for each transport route, and this correlation is used to estimate the powdering rate Rp from the crack volume rate Rvc. On the other hand, since the degree of powdering due to impact may differ depending on the type of iron ore pellet, the correlation between the powdering rate Rp and the crack volume rate Rvc is calculated for each type of iron ore pellet, and this correlation is used to estimate the powdering rate Rp from the crack volume rate Rvc.

According to this embodiment, by measuring the crack volume fraction Rvc of iron ore pellets before pulverization from three-dimensional images obtained by X-ray CT scanning, the pulverization fraction Rp can be estimated based on the correlation (linear function) between the pulverization fraction Rp and the crack volume fraction Rvc. This makes it possible to determine the mass ratio (pulverization fraction Rp) at which iron ore pellets pulverize during transport.

In addition, the mass of the pulverized iron ore pellets (excluding the pulverized material) can be determined, allowing the pellet ratio when charging the iron ore pellets into the blast furnace to be estimated in advance. If the amount of iron ore pellets charged does not reach the target amount, the amount of iron ore pellets charged into the blast furnace can be increased in advance to make up for this shortfall.

First, the correlation between the powdering rate Rp and the crack volume fraction Rvc was determined using eight iron ore pellets (No. 1 to No. 8) that differed only in their lots. Specifically, the eight iron ore pellets were transported along a predetermined transport route, and the powdering rate Rp and the crack volume fraction Rvc were determined for each iron ore pellet. Based on these results, the correlation (linear function) between the powdering rate Rp and the crack volume fraction Rvc was determined. This is explained in detail below.

50 kg of iron ore pellets were collected after loading (corresponding to the mass M1 shown in formula (1) above) and dried at 105°C for 24 hours to facilitate separation of pellet chips (powdered material) adhering to the iron ore pellets. The pellets were then sieved using a sieve with 6.3 mm openings, and the mass of pellet chips with particle sizes of 6.3 mm or less (corresponding to the mass M2 shown in formula (1) above) was measured.

Then, based on the above formula (1), the pulverization rate Rp was calculated for each of the eight iron ore pellets. Note that although pulverization has already occurred in the iron ore pellets after loading, it is assumed that the pellet chips remain attached to the iron ore pellets, and the mass of the iron ore pellets after loading is taken as the mass M1 shown in the above formula (1) (the mass of the iron ore pellets before pulverization).

For each of the eight iron ore pellets mentioned above, 20 pellets were sampled before loading and unloading, and an X-ray CT scan was performed on each individual pellet using a three-dimensional measurement X-ray CT device (TDM3000H-FP, Yamato Scientific Co., Ltd.). The imaging conditions for the X-ray CT scan are shown in Table 1 below. The pixel size was set to a spatial resolution (16.6 μm/pixel) that would not interfere with the detection of voids (including pores and cracks) contained in the iron ore pellets.

Void areas (including cracks and holes) were extracted by performing a binarization process on three-dimensional images of iron ore pellets obtained by X-ray CT scanning. Here, the threshold value for the binarization process was set to 110, and areas with a brightness value of less than 110 were identified as void areas. Next, the threshold value Ψth was set to 0.3 (Example 1), 0.5 (Example 2), 0.7 (Example 3), and 1.0 (Comparative Example). Cracks were identified from within the void areas, and the crack volume ratio Rvc was then calculated. The method for calculating the crack volume ratio Rvc is as described above.

In this example, 20 iron ore pellets were collected from each of the eight iron ore pellets. After determining the crack volume fraction Rvc for each iron ore pellet, the average crack volume fraction Rvc of the 20 pellets, Rvc_ave, was calculated.

The measurement results for the above-mentioned powdering rate Rp and crack volume rate Rvc_ave are shown in Table 2 below. Table 2 below also shows the porosity, which is the ratio of voids (including cracks and holes) to the apparent volume of the iron ore pellets.

In each of Examples 1 to 3 and the Comparative Example, the correlation (approximation) between the powdering rate Rp and the crack volume rate Rvc was determined based on the powdering rate Rp and the crack volume rate Rvc_ave in the iron ore pellets (No. 1 to No. 8), and it was possible to express it as a linear function as shown in the following Table 3. The following Table 3 also shows the coefficient of determination ^R2 in the correlation in Examples 1 to 3 and the Comparative Example.

As can be seen from Table 3 above, by setting the threshold value Ψth to be less than 1.0, the coefficient of determination ^R2 was higher than when the threshold value Ψth was set to 1.0. Furthermore, when the threshold value Ψth was set to 0.5, the coefficient of determination ^R2 was highest. Therefore, it can be seen that setting the threshold value Ψth to 0.5 best represents the correlation between the powdering rate Rp and the crack volume rate Rvc.

Next, the crack volume fraction Rvc was measured for the iron ore pellets for which the reduction rate Rp was to be estimated. These iron ore pellets were the same type as the iron ore pellets used when determining the correlation (linear function) between the reduction rate Rp and the crack volume fraction Rvc described above; only the lot was different. The crack volume fraction Rvc was measured using the same method as used to measure the crack volume fraction Rvc_ave for the iron ore pellets (No. 1 to No. 8) described above.

When the crack volume fraction Rvc was measured with the threshold value Ψth set to 0.5, the crack volume fraction Rvc was found to be 2.3 [vol%]. Furthermore, when the correlation when the threshold value Ψth was 0.5 (Example 2 shown in Table 3 above; Rp = 2.72 × Rvc + 3.27) was used to calculate the powder fraction Rp from the measured crack volume fraction Rvc, the powder fraction (estimated value) Rp was found to be 9.53 [wt%].

On the other hand, when the powdering rate Rp of the iron ore pellets (50 kg) for which the powdering rate Rp was to be estimated was calculated by measuring the mass of the pellet chips after loading, the powdering rate (measured value) Rp was 9.50 [wt%]. The estimated and measured results of the above-mentioned powdering rate Rp are shown in Table 4 below. As can be seen from Table 4 below, the powdering rate (estimated value) Rp was almost identical to the powdering rate (measured value) Rp. Therefore, by estimating the powdering rate Rp as in this example, the accuracy of the estimation of the powdering rate Rp can be ensured.

Claims (5)

A method for estimating a powdering rate, which is a ratio of the mass of powdered iron ore pellets to the mass of iron ore pellets before powdering, for iron ore pellets that are powdered due to impact during transportation, comprising:
Extracting voids contained in the iron ore pellets based on a three-dimensional image obtained by X-ray CT scanning of the iron ore pellets before pulverization;
Among the extracted void portions, void portions having a sphericity equal to or less than a threshold value are identified as cracks, and a crack volume ratio, which is the ratio of the total volume of the cracks to the apparent volume of the iron ore pellet, is calculated;
calculating the powdering rate from the calculated crack volume rate based on a correlation between the crack volume rate and the powdering rate, which is expressed by a linear function and which has been calculated in advance;
A method for estimating the powdering rate of iron ore pellets, comprising: The method for estimating the iron ore pellet fineness rate described in claim 1, characterized in that the threshold value is 0.5 [-]. A method for estimating the powdering rate of iron ore pellets according to claim 1 or 2, characterized in that the powdered material has a particle size of 6.3 mm or less. A method for estimating the iron ore pellet fineness rate according to claim 1 or 2, characterized in that the correlation is determined for each iron ore pellet transport route. 3. The method for estimating the powdering rate of iron ore pellets according to claim 1, wherein the sphericity is determined from the volume and surface area of the void portion in the three-dimensional image.