Method for creating a strength estimation model for sintered ore, method for estimating the strength of sintered ore, and method for manufacturing sintered ore.

Abstract

To provide a creation method of a strength estimation model of sintered ore capable of estimating the strength of the sintered ore more accurately than ever.SOLUTION: A method for creating a strength estimation model of sintered ore includes a strength acquisition process for obtaining the strength of the sintered ore, a domain fraction calculation process for dividing the texture images of the sintered ore into multiple domains and calculating the area fraction of the domains for each first texture, and a model creation process for creating a strength estimation model for the sintered ore using the area fraction as an explanatory variable and the strength as a target variable, wherein the domain is defined as a region consisting of the continuous first texture and one or both of the textures and voids other than the first texture dispersed within the first texture.SELECTED DRAWING: None

Description

Technical Field

[0001] The present invention relates to a method for creating a strength estimation model of sintered ore, a method for estimating the strength of sintered ore, and a method for manufacturing sintered ore.

Background Art

[0002] Sintered ore is a charge for a blast furnace in blast furnace operation. The raw material of sintered ore is a sintered ore blending raw material in which multiple grades of powdered ore are blended with appropriate amounts of auxiliary raw material powder, miscellaneous raw materials, and solid fuel. Water as a binder is added to the sintered ore blending raw material, mixed, and granulated. The obtained granulated raw material is charged into a Dwight Lloyd type (DL) sintering machine or the like and fired to produce sintered ore. The produced sintered ore is discharged, crushed by a crusher, and sized by a sieve (crushing and screening process). In order to maintain good air permeability in the blast furnace, the product sintered ore needs to be somewhat coarse. Therefore, the undersize is returned to the raw material as recycled ore, and the oversize is sent to the blast furnace as product sintered ore. Therefore, in the production of sintered ore, in order to increase the ratio of product sintered ore to the production amount of sintered ore (yield of product sintered ore), it is necessary to increase the ratio of somewhat coarse sintered ore. For this purpose, the produced sintered ore needs to have high strength.

[0003] Generally, the strength of sintered ore can be evaluated using the drop strength index obtained by a drop strength test. The drop strength index is a value corresponding to the sieve particle size after applying a predetermined drop impact to the sintered ore. However, even when the drop strength index is measured, the cause of improving or deteriorating the strength is often unknown. That is, in order to efficiently improve the yield, it is not sufficient to directly measure the strength of sintered ore using a method such as drop strength measurement. Therefore, it is required to be able to explain the reason for the change in strength using an index other than strength.

[0004] Based on the above discussion, an index that can more accurately estimate the strength of sintered ore, and an accurate strength estimation method using the index have been studied.

[0005] Patent Document 1 discloses a method for estimating the strength of sintered ore from the phase fraction of the mineral phases. Specifically, the phase fraction of the mineral phases is determined by applying Rietveld analysis to the diffraction pattern obtained by X-ray diffraction. Then, among the phase fractions, SFCA (Silico-ferrite of calcium and aluminum, Ca2(Fe,Ca)6(Fe,Al,Si)6O 20 A method for estimating intensity from the phase fraction of ) is disclosed.

[0006] On the other hand, Patent Document 2 proposes a learning model for estimating the microstructure of a composite structure formed by mineral phases from cross-sectional images of sintered ore. In this document, examples are disclosed in which the microstructure of the composite structure is defined as follows: a first composite structure in which either hematite or magnetite is surrounded by calcium ferrite; a second composite structure of calcium ferrite and silicate slag; a third composite structure of either hematite or magnetite and calcium ferrite; and a fourth composite structure of either hematite or magnetite and silicate slag. [Prior art documents] [Patent Documents]

[0007] [Patent Document 1] Japanese Patent Publication No. 2018-179690 [Patent Document 2] Japanese Patent Publication No. 2021-166002 [Overview of the Initiative] [Problems that the invention aims to solve]

[0008] However, it was found that the method for evaluating the strength of sintered ore disclosed in Patent Document 1 has low accuracy.

[0009] Furthermore, the method described in Patent Document 2 could not be used to accurately estimate the strength of the sintered ore.

[0010] This invention has been made in view of the above circumstances, and aims to provide a method for creating a sintered ore strength estimation model that can estimate the strength of sintered ore with greater accuracy than conventional methods. [Means for solving the problem]

[0011] As a result of diligent research, the inventors have found that the above objective can be achieved by adopting the following configuration.

[0012] 1. A strength acquisition process to obtain the strength of the sintered ore, A domain fraction calculation step involves dividing the microstructure image of the sintered ore into multiple domains and calculating the area fraction of each domain for each first microstructure of the domain. The process includes a model creation step of creating a model for estimating the strength of sintered ore, using the area fraction as an explanatory variable and the strength as an objective variable. A method for creating a strength estimation model for sintered ore, wherein the domain is defined as a region consisting of a continuous first structure and one or both of other structures and voids dispersed within the first structure.

[0013] 2. A method for creating a strength estimation model for a sintered ore as described in 1 above, wherein the domain consists of a domain having primary hematite as the first structure, a domain having secondary hematite as the first structure, a domain having magnetite as the first structure, and a domain having calcium ferrite as the first structure.

[0014] 3. A method for estimating the strength of sintered ore, wherein the strength of the sintered ore is estimated using a sintered ore strength estimation model created by the method for creating a sintered ore strength estimation model described in 1 or 2 above.

[0015] 4. A method for manufacturing sintered ore, comprising changing the manufacturing conditions of the sintered ore based on the strength of the sintered ore estimated by the method for estimating the strength of the sintered ore described in 3 above.

[0016] 5. The method for producing sintered ore according to item 4 above, wherein the change in the manufacturing conditions is a change in the ratio of the components constituting the sintered ore blending raw materials.

Advantages of the Invention

[0017] According to the present invention, it is possible to provide a method for creating a strength estimation model of sintered ore that can estimate the strength of sintered ore with higher accuracy than before.

Modes for Carrying Out the Invention

[0018] The inventors of the present invention have examined in detail the reason why the strength of sintered ore cannot be accurately obtained from the phase fraction of the mineral phase, and obtained the following findings.

[0019] Generally, sintered ore is mainly composed of four mineral phases, hematite (Fe2O3), magnetite (Fe3O4), calcium ferrite (CaO·Fe2O3), and silicate slag (CaO·SiO2), as well as voids. Among them, it has been conventionally known that the amount of the calcium ferrite phase affects the drop strength of sintered ore. Furthermore, due to differences in composition and crystal structure, the calcium ferrite phase is mainly classified into the SFCA phase and the SFCA-I phase, and it has been reported that the phase fraction of the SFCA phase shows a stronger correlation with the strength of sintered ore than other calcium ferrite phases (Patent Document 1). However, in sintered ore, a plurality of mineral phases are not distributed as single phases, but a large number of mixed phases in which the mineral phases are complexly mixed are formed. Therefore, it is expected that the strength of the region of the mixed phase changes depending on the types and amounts of the mineral phases constituting the mixed phase, and the strength of the sintered ore also changes depending on the ratio that the mixed phase occupies in the sintered ore. For these reasons, when a method for predicting the strength of sintered ore from the phase fraction for each single phase is used, the prediction accuracy of the strength of sintered ore becomes low.

[0020] Therefore, as a concept corresponding to the region of the mixed phase, it is important to define a domain that is a region including a continuous first structure and obtain the fraction of the domain for each first structure.

[0021] Hereinafter, an embodiment of the present disclosure will be described.

[0022] The method for creating a strength estimation model of sintered ore according to an embodiment of the present disclosure includes a strength acquisition step, a domain fraction calculation step, and a model creation step.

[0023] (Sintered ore) As the sintered ore according to this embodiment, any sintered ore can be used. For example, the sintered ore can be manufactured by any method. Hereinafter, examples of the manufacturing method of sintered ore will be described.

[0024] In one embodiment of the present disclosure, sintered ore can be manufactured by adding water as a binder to the sintered ore blending raw material, mixing and granulating, and firing the obtained granulated raw material.

[0025] Sintered ore is manufactured using a sintered ore blending raw material. Examples of the components constituting the sintered ore blending raw material include powdered ore, secondary raw material powder, miscellaneous raw materials, and solid fuel. As the powdered ore, powdered ores of a plurality of brands may be used. The secondary raw material powder includes, for example, limestone, silica, and serpentine. The miscellaneous raw materials include, for example, dust, scale, and return ore. The solid fuel includes, for example, pulverized coke. The ratio of the above components can be changed as appropriate.

[0026] First, after granulating the sintered ore blending raw material, it is fired using a sintering device to obtain a sintered cake. Examples of the sintering device include Dwight Lloyd type sintering machines and sintering pots. Examples of the sintering pot include Grinawalt type sintering machines.

[0027] Next, a sintered ore sample is collected from the sintered cake. Then, the collected sintered ore sample is pulverized. The pulverization method is not particularly limited as long as it does not affect the mineral phase. Specifically, a method using a pulverization device such as a vibration mill, ball mill, rotary mill, stamp mill, crusher, etc. is common. Also, it may be pulverized by two or more methods. For example, the sintered cake may be coarsely pulverized with a crusher and then further pulverized with a rotary mill.

[0028] Furthermore, the sintered ore can be cooled before or after crushing the sintered cake. For example, when using sintered cake discharged from a full-scale sintering apparatus, it is generally possible to coarsely crush the obtained sintered cake with a crusher and then introduce it into a cooler for cooling. On the other hand, when obtaining sintered cake using a laboratory-scale sintering pot, it is also possible to crush the sintered cake after cooling.

[0029] The method for creating a strength estimation model of sintered ore according to this embodiment may include a sampling step for sampling sintered ore. The sampling step may be a step for sampling sintered ore after crushing. The method for sampling sintered ore is not particularly limited, but for example, a method using a dedicated box-shaped sampler can be used. Alternatively, a method of sampling sintered ore flowing on a belt conveyor belt may be used. For example, a method of manually sampling from the belt conveyor belt using a ladle or the like may be used, or a method of continuously sampling a portion of the sintered ore flowing on the belt conveyor belt using a robotic arm may be used. Furthermore, a method of sampling sintered ore from a sample container containing samples sorted by particle size for strength measurement may be used. For example, a method of continuously sampling sorted samples using an arm may be used, or an operator may periodically and continuously sample from the sample container by hand.

[0030] When sampling sintered ore, if it is necessary to evaluate the variation in its structure, it is preferable to take multiple samples from the same lot in order to improve the accuracy of the variation evaluation. Furthermore, in order to further improve the accuracy of estimating the strength of the sintered ore, it is preferable to take multiple sintered ore samples from the upper and lower layers of the sintered cake from the same lot. Specifically, it is preferable to take at least one sintered ore sample from each of the upper and lower layers of the sintered cake from the same lot. Here, the upper layer of the sintered cake refers to the layer above the halfway point of the sintered cake's height, and the lower layer of the sintered cake refers to the layer below the halfway point of the sintered cake's height.

[0031] [Strength acquisition process] In the strength acquisition process, the strength of the sintered ore is obtained. The method of obtaining the strength is not limited, but it is preferable to measure the strength of the sintered ore.

[0032] As the strength, it is preferable to use the Shatter Index (SI) defined in JIS M 8711. Alternatively, the Tumble Index (TI) defined in JIS M 8712 may be used. Furthermore, an index correlated with any of the above indices may be used.

[0033] [Domain Fraction Calculation Process] In the domain fraction calculation step, the microstructure image of the sintered ore whose strength was obtained in the strength acquisition step is divided into multiple domains, and the area fraction of each domain is calculated for each first structure of the domain. Here, the calculated area fraction of a domain is referred to as the domain fraction. Furthermore, the domain is a region consisting of one or both of the following: a continuous first structure and other structures and voids dispersed within the first structure. Note that the strength acquisition step and the domain fraction calculation step may be performed in either order, simultaneously, or in parallel.

[0034] (Image of the microstructure of sintered ore) The microstructure image of the sintered ore is an image of the microstructure of any facet of the sintered ore.

[0035] Here, we will describe an example of a method for acquiring a microstructure image. First, the sintered ore sample is crushed using the method described above. After crushing, the desired surface of the sintered ore is exposed as the observation surface by, for example, polishing, cutting, or using a laser, and the observation surface is polished until it becomes mirror-like. At this time, since the sintered ore is irregular in shape and easily crumbles, it is preferable to embed the sintered ore in resin beforehand. Next, a microstructure image is acquired of the observation surface of the polished sintered ore.

[0036] The aforementioned tissue images can be acquired, for example, using a microscope. Examples of such microscopes include optical microscopes, scanning electron microscopes, and stereomicroscopes. From the viewpoint that the color tone of the image can be used as a criterion when classifying domains, it is preferable to acquire the images using an optical microscope or a stereomicroscope. On the other hand, when acquiring images using a scanning electron microscope, the type of image is not limited, and images such as secondary electron images or backscattered electron images can be acquired. From the viewpoint of improving the accuracy of classifying domains, multiple types of images may be acquired. Also, from a similar viewpoint, when using a scanning electron microscope, maps of crystal orientation, etc., may be acquired using EBSD (Electron Backscatter Diffraction), and elemental maps may be acquired using EDX (Energy Dispersive X-ray Spectroscopy). For convenience, maps of crystal orientation, etc., and elemental maps will also be referred to as tissue images hereafter.

[0037] In addition to the methods described above, the microstructure images can also be acquired using non-destructive methods, such as performing X-ray CT (Computed Tomography) on a desired surface of the sintered ore.

[0038] The upper and lower limits of the observation magnification of the tissue image are not particularly limited. However, increasing the observation magnification makes it easier to identify the tissue. For this reason, the observation magnification of the tissue image is preferably 50 times or higher, and more preferably 100 times or higher. On the other hand, if the observation magnification is excessively high, the measurement time for acquiring the tissue image becomes longer, and it becomes more troublesome to ensure the representativeness of the sintered ore structure. For this reason, from the viewpoint of measurement efficiency, the observation magnification of the tissue image is preferably 200 times or lower.

[0039] Furthermore, the observation magnification affects how finely the domains are divided, as will be discussed later. As will be discussed later, a domain includes continuous tissue. However, by increasing the observation magnification, a region that was originally observed as a single continuous tissue may be observed as a region containing multiple continuous tissues. In this case, the region of continuous tissue is subdivided, and the domain is divided more finely. Therefore, the fineness of the domain division can be controlled by appropriately setting the observation magnification. As mentioned above, the intensity is determined for each region of the mixed phase, so it is good practice to divide the domains to a fineness that corresponds to the mixed phase, and this can further improve the accuracy of intensity estimation. From these viewpoints as well, the observation magnification of the tissue image is preferably 200x or less.

[0040] The upper and lower limits of the size of the observation area of ​​the microstructure image are not particularly limited, but in order to ensure the representativeness of the sintered mineral structure, it is preferable that the observation area of ​​the microstructure image covers the entire observation surface. Also, from a similar viewpoint, the size of the observation area of ​​the microstructure image may be 25 mm. 2 It is preferable to have a minimum of 100 mm 2 It is more preferable to have the above. The upper limit of the size of the observation area of ​​the tissue image is, for example, 225 mm. 2 It can be done this way.

[0041] The tissue image can be acquired by continuously changing the field of view. More specifically, it is preferable that the tissue image is an image acquired by continuously changing the field of view of the microscope while taking images and then stitching together the acquired images. By acquiring a single tissue image while changing the field of view in this way, it becomes easy to acquire a tissue image that achieves both high magnification and a wide observation area.

[0042] (domain) In the present invention, a domain is defined as a region consisting of a continuous first structure and one or both of other structures and voids dispersed within the first structure. By dividing the microstructure image of sintered ore into these defined domains, the state of the mixed phase of the sintered ore can be evaluated, and an index with a strong correlation to strength can be obtained. For example, in the prior art described in Patent Document 2, the microstructure image was not divided into such domains, and therefore it was not possible to obtain an index suitable for estimating the strength of sintered ore. Note that the same position in the microstructure image is not divided into two or more domains, so there are no overlapping domains.

[0043] The microstructure of sintered ore may consist of any type of microstructure. Examples of sintered ore microstructures include primary hematite, secondary hematite, magnetite, calcium ferrite, and silicate slag, but generally, the microstructure of sintered ore mainly consists of the above five types of microstructures. Therefore, it is preferable that the microstructure of sintered ore consists of the above five types of microstructures and other microstructures. Furthermore, one or more of the above five types of microstructures may be further subdivided. In addition, any sintered ore microstructure can be the primary microstructure of a domain, but it is preferable that primary hematite, secondary hematite, magnetite, or calcium ferrite be the primary microstructure.

[0044] A domain is a region containing a continuous first structure. As mentioned above, various types of structures are distributed in a complex mixture within sintered ore, but among these structures, the structure that is continuous in a certain part of the microstructure image is considered the first structure. The structure that is continuous in the part where the mixed phase exists best reflects the melting state during the sintering process in the region corresponding to the domain (whether or not it underwent melting during the sintering process), and therefore best represents the properties of the mixed phase produced by sintering. Furthermore, among the structures contained in a domain, the first structure occupies the largest area fraction of that domain. From this perspective as well, the first structure is considered to be a structure that represents the properties of the mixed phase.

[0045] Furthermore, based on the differences in their primary structure, domains can be divided into those that have undergone melting during the sintering process (also called molten domains) and those that have not undergone melting (also called unmolten domains).

[0046] It is important that a domain consists of the primary tissue, as well as one or both of the other tissues and voids. If a domain is defined as a region that does not contain any tissues other than the primary tissue and does not contain voids, then the phase fraction of each single phase will be evaluated, and the state of the mixed phase cannot be evaluated. Furthermore, the amount of voids, which is thought to affect the strength, cannot be reflected in the domain fraction value.

[0047] Here, within a domain, tissues and voids other than the first tissue are dispersed within the first tissue. That is, if one or both of the tissues and voids other than those tissues are contained within a contiguous tissue, then those tissues and voids, together with the contiguous tissue, shall form a single domain. Furthermore, if a tissue or void exists in a region sandwiched between two or more contiguous tissues, that tissue or void may be considered dispersed within any of those contiguous tissues. The tissue in which the tissue or void is dispersed can be determined using indicators such as the distance between the tissue or void and the contiguous tissue.

[0048] In the present invention, it is preferable that the domains consist of a domain having primary hematite as the primary structure, a domain having secondary hematite as the primary structure, a domain having magnetite as the primary structure, and a domain having calcium ferrite as the primary structure. Here, the four types of domains are defined assuming that the structure of the sintered ore consists of primary hematite, secondary hematite, magnetite, calcium ferrite, silicate slag, and other structures.

[0049] (1) Domains with primary hematite as the primary tissue During the sintering process, a molten material is formed at a location where the original hematite, calcium oxide, and SiO2 coexist. However, there are regions where the original hematite remains unreacted as primary hematite. This domain corresponds to the aforementioned region and is classified as an unmelted domain. This domain may also contain calcium ferrite and silicate slag.

[0050] (2) Domains with secondary hematite as the primary tissue During the sintering process, a molten state is formed in a location where the original hematite, calcium oxide, and SiO2 coexist, and under high oxygen partial pressure, hematite (secondary hematite) crystallizes in a skeletal state. This domain corresponds to the aforementioned region and is classified as a molten domain. This domain may also contain calcium ferrite and silicate slag.

[0051] (3) Domains with magnetite as the primary organization During the sintering process, the original hematite is reduced to magnetite, and there are regions where this magnetite remains without reacting with the surrounding molten material. This domain corresponds to the aforementioned region and is classified as an unmelted domain. This domain may also contain calcium ferrite and silicate slag.

[0052] (4) Domains with calcium ferrite as the primary tissue During the sintering process, a molten region is formed where the original hematite, calcium oxide, and SiO2 coexist, and this molten region crystallizes as calcium ferrite and silicate slag. This domain corresponds to the aforementioned region and is classified as a molten domain. This domain may also contain silicate slag.

[0053] The domains (1) to (4) described above may include structures other than the first structure in which the sintered ore-containing raw materials remain partially unsintered, for example, they may contain SiO2.

[0054] In addition to the domains (1) to (4) mentioned above, it is also possible to establish a domain in which the primary structure is something other than primary hematite, secondary hematite, magnetite, and calcium ferrite contained in sintered ore.

[0055] The following describes how to categorize organizational images into domains.

[0056] To classify an tissue image into domains means to identify which tissue is present at every point in the tissue image, and based on that, to determine which domain each point belongs to, or whether it belongs to any domain. Here, if the tissue image is electronic data, for example, pixels can be used as the points. Two specific examples of a point that does not belong to any domain are given below. (A) When no organization exists at the said location, and the said location is not a void in the definition of a domain as described above. (B) When tissue exists at the aforementioned location, but no other tissue or voids are dispersed within that tissue, nor is the tissue dispersed within other tissues. If the tissue present at the aforementioned location is located continuously over a wide area, it is common for one or both of other tissues and voids to be dispersed within that tissue, and therefore it usually does not fall under (B). On the other hand, if the area occupied by the tissue present at the aforementioned location is narrow, it is highly likely that other tissues and voids are not dispersed within that tissue. In other words, if the area occupied by the tissue present at the location is narrow and it is located far from other tissues (isolated), it is highly likely that it falls under (B).

[0057] The method for dividing tissue images into domains is not particularly limited. However, for example, when acquiring tissue images using an optical microscope and dividing those tissue images into the domains (1) to (4) above, the method can be carried out as follows.

[0058] First, the sintered mineral structures—hematite, magnetite, and calcium ferrite—are identified by their brightness and color in the optical microscope images. Hematite regions appear white, while magnetite regions have a reddish tint.

[0059] Next, a region in which magnetite is continuous is combined with one or both of the other tissues and voids dispersed within the magnetite to form (3) a domain in which magnetite is the primary tissue.

[0060] Next, hematite that retains the shape of the original ore and has a rounded, granular texture is identified as primary hematite. Then, a region where primary hematite is continuous is combined with one or both of the other textures and voids dispersed within the primary hematite to form (1) a domain in which primary hematite is the primary texture.

[0061] Next, among the hematites, angular hematites other than primary hematites are identified as secondary hematites. Then, the regions in which secondary hematites are continuous are combined with one or both of the other structures and voids dispersed within the secondary hematites to form (2) a domain in which secondary hematites are the primary structure.

[0062] Furthermore, any region that does not belong to any of the above (1) to (3), and in which calcium ferrite is continuous, is combined with one or both of the other tissues and voids dispersed within the calcium ferrite to form (4) a domain in which calcium ferrite is the primary tissue.

[0063] In this way, the domains can be divided according to the definition of each domain. Furthermore, for all domains, the primary organization of the domain will be the organization with the highest area fraction within that domain.

[0064] Here, for domain division, it is preferable to use machine learning from the viewpoint of performing the work quickly and simply, and furthermore, eliminating arbitrariness. In this case, as training images, images obtained by dividing the microstructure of sintered ore into multiple domains are used. The aforementioned machine learning includes known methods such as deep learning. Specifically, it is preferable to use U-net, which is one of the FCNs (Fully Convolutional Networks) that applies CNN (Convolutional Neural Network). U-net is a network suitable for image segmentation and has the properties of extracting feature maps by image convolution and performing deconvolution processing while retaining the feature maps.

[0065] (Domain fraction) In the domain fraction calculation process, after dividing the tissue image into domains using the method described above, the domain fraction is calculated for each first tissue within a domain. For example, the area fraction can be calculated as the ratio of the total area of ​​a domain whose first tissue is a certain tissue to the total area of ​​all domains, and this area fraction can be used as the domain fraction for a domain whose first tissue is that particular tissue.

[0066] [Model creation process] In the model creation process, a strength estimation model for sintered ore is created using the domain fractions calculated in the domain fraction calculation process as explanatory variables and the strength of the sintered ore obtained in the strength acquisition process as the dependent variable. Hereinafter, the strength estimation model may simply be referred to as the model.

[0067] The method for creating the model is not limited, but for example, a model can be created using machine learning. In one embodiment of the present invention, a model is created by performing multiple regression analysis as machine learning. Specifically, for sintered ore samples from which intensity has been obtained, multiple regression analysis is performed with intensity as the dependent variable and the domain fraction of the corresponding sintered ore sample as the independent variable, thereby calculating the relationship between the dependent variable and the independent variable as a regression equation. The regression equation is used as the model equation for the intensity estimation model.

[0068] The present invention relates to a method for estimating the strength of sintered ore, which involves estimating the strength of sintered ore using a sintered ore strength estimation model created by the method for creating the sintered ore strength estimation model. For a sintered ore sample with unknown strength, the domain fraction obtained from the sintered ore sample is substituted into the model equation as an explanatory variable, and the strength, which is the objective variable, is calculated. According to this strength estimation method, even during blast furnace operation, the strength of sintered ore can be accurately estimated simply by calculating the domain fraction for a sintered ore sample with unknown strength and substituting the domain fraction into a pre-created sintered ore strength estimation model, thereby allowing for the prediction of the yield of the finished sintered ore.

[0069] The method for producing sintered ore according to the present invention is a method for changing the production conditions of sintered ore based on the strength of the sintered ore estimated by the method for estimating the strength of the sintered ore. By reviewing the production conditions of the sintered ore based on the estimated strength, it is possible to improve the yield of the sintered ore product. The production conditions that can be changed are not particularly limited, but include the amount and type of gaseous fuel, whether or not oxygen is blown into the sintering furnace, and negative pressure. For example, if it is found that it is good to reduce the number of domains with magnetite as the primary structure based on the estimated strength of the sintered ore, it is good to blow oxygen into the sintering furnace. Here, it is preferable that the change in production conditions is a change in the ratio of the components constituting the sintered ore blending raw materials. As mentioned above, the components constituting the sintered ore blending raw materials include, for example, powdered ore, auxiliary raw material powder, miscellaneous raw materials, and solid fuel. For example, if it is found that it is good to increase the number of domains with calcium ferrite as the primary structure based on the estimated strength of the sintered ore, it is good to increase the basicity (CaO / SiO2) of the raw materials. [Examples]

[0070] The following describes, based on examples, a method for creating a strength estimation model for sintered ore and a method for estimating the strength of sintered ore according to one embodiment of this disclosure.

[0071] (Example 1) A strength estimation model for sintered ore was created as follows.

[0072] First, by changing the amount of coke in the sintered ore mixture, three types of sintered cakes were obtained using different heat levels. A laboratory-scale sintering pot was used for sintering. Next, sintered ore samples (No. 1-3) were taken from each sintered cake.

[0073] The obtained sintered ore samples were split in half, embedded in resin, and the surface to be evaluated (observation surface) was mirror-polished. Polishing was performed using SiC paper from #120 to #400, followed by polishing with diamond at 9 μm, 3 μm, 1 μm, and 0.25 μm, as well as colloidal silica polishing.

[0074] Microstructure images of the sintered ore were obtained from the observation surface of the polished sintered ore. The microstructure images of the sintered ore were prepared by using an optical microscope equipped with imaging capabilities, capturing images of the entire observation surface of the sintered ore at 50x magnification while continuously changing the field of view, and then stitching together the obtained optical microscope images.

[0075] The aforementioned microstructure images were divided into the domains (1) to (4) described above using the method described, and the domain fraction of each sintered ore sample was calculated. In addition, the drop strength index specified in JIS M 8711 was measured for each sintered ore sample.

[0076] Table 1 shows the measured strength and calculated domain fraction of the aforementioned sintered ore sample.

[0077] [Table 1]

[0078] A model for estimating the strength of sintered ore was created by performing multiple regression analysis using the domain fractions of the sintered ore samples shown in Table 1 as explanatory variables and the strength as the dependent variable, thereby deriving a model equation. Then, the strength of the sintered ore (hereinafter, the estimated strength of the sintered ore will be referred to as "estimated strength") was estimated by substituting the domain fractions of the sintered ore samples shown in Table 1 into the model equation. The coefficient of determination obtained using the measured strength of the sintered ore (hereinafter, "measured strength") and the estimated strength is 0.94. In this way, a model for estimating the strength of sintered ore was created by performing multiple regression analysis.

[0079] (Example 2) Next, the strength of the sintered ore was estimated using the strength estimation model that was created.

[0080] First, three types of sintered ore samples (A, B, and C) to be used for strength estimation were prepared as follows. Initially, the coke content of the sintered ore mixture was set to 4.0%, 4.5%, or 5.0%, respectively, and sintered in a laboratory-scale sintering pot to obtain a sintered cake with a diameter of 300 mm and a height of 600 mm. Next, after the sintered cake cooled, it was dropped and crushed to obtain sintered ore samples with a diameter of approximately 20 mm.

[0081] The obtained sintered ore samples were split in half, embedded in resin, and the surface to be evaluated (observation surface) was mirror-polished. Polishing was performed using SiC paper from #120 to #400, followed by polishing with diamond at 9 μm, 3 μm, 1 μm, and 0.25 μm, as well as colloidal silica polishing.

[0082] Microstructure images of the sintered ore were obtained from the observation surface of the polished sintered ore. The microstructure images of the sintered ore were prepared by using an optical microscope equipped with imaging capabilities, capturing images of the entire observation surface of the sintered ore at 50x magnification while continuously changing the field of view, and then stitching together the obtained optical microscope images.

[0083] Subsequently, the tissue image was divided into the domains (1) to (4) above using a trained network (U-net), and the area fraction of each domain was calculated as the domain fraction. As the training image for the trained network, an optical microscope image of sintered ore was divided into the domains (1) to (4) above using the method described above.

[0084] Next, the estimated strength of each sintered ore sample was determined by substituting the domain fraction of each sintered ore sample into the strength estimation model for sintered ore created in Example 1. The actual strength was also determined by measuring the drop strength index as defined in JIS M 8711. A comparison was then made between the estimated and actual strengths for each sintered ore sample.

[0085] On the other hand, as a comparative example, the strength was estimated using the following method. First, XRD-Rietveld analysis was performed on a sintered ore sample separate from sintered ore samples A, B, and C, referring to the method described in Japanese Patent Publication No. 2013-122403, to determine the phase fractions of hematite, magnetite, calcium ferrite, and dicalcium silicate. In addition, the drop strength index specified in JIS M 8711 was measured, and a regression equation was created between the above phase fractions and the drop strength index. Subsequently, the phase fractions were determined for sintered ore samples A, B, and C, and the estimated strength values ​​were obtained by substituting these phase fractions into the above regression equation.

[0086] Table 2 shows the results for estimated and measured intensity. Compared to the intensity estimated by the comparative example method, the intensity estimated by the intensity estimation model according to the present invention showed extremely good agreement with the measured intensity. The coefficient of determination was 0.91, indicating a good fit of the regression equation.

[0087] [Table 2]

[0088] As described above, according to the present invention, by preparing a strength estimation model for sintered ore in advance and determining the domain fraction of the sintered ore, it becomes possible to accurately estimate the strength of the sintered ore. In other words, the domain fraction is an excellent indicator for explaining the strength of the sintered ore. Furthermore, by reviewing the manufacturing conditions of the sintered ore based on the estimated strength, it becomes possible to improve the yield of the finished sintered ore product.

Claims

1. A strength acquisition process to obtain the strength of the sintered ore, A domain fraction calculation step involves dividing the microstructure image of the sintered ore into multiple domains and calculating the area fraction of each domain for each first microstructure of the domain. The process includes a model creation step of creating a model for estimating the strength of sintered ore, using the area fraction as an explanatory variable and the strength as an objective variable. A method for creating a strength estimation model for sintered ore, wherein the domain is defined as a region consisting of a first structure that is continuous and occupies the highest area fraction within the domain, and one or both of other structures and voids dispersed within the first structure.

2. A method for creating a strength estimation model for a sintered ore according to claim 1, wherein the domain comprises a domain having primary hematite as the first structure, a domain having secondary hematite as the first structure, a domain having magnetite as the first structure, and a domain having calcium ferrite as the first structure.

3. A method for estimating the strength of sintered ore, comprising estimating the strength of sintered ore using a strength estimation model for sintered ore prepared by the method for creating a strength estimation model for sintered ore according to claim 1 or 2.

4. A method for manufacturing sintered ore, comprising changing the manufacturing conditions of sintered ore based on the strength of the sintered ore estimated by the method for estimating the strength of sintered ore described in claim 3.

5. The method for producing sintered ore according to claim 4, wherein the change in the manufacturing conditions is a change in the ratio of the components constituting the sintered ore blending raw materials.

Images