Model for predicting weight of oxygen per unit area after decarburization annealing, method for constructing said model, and method for producing grain-oriented electrical steel sheet

Abstract

The present invention pertains to a model for predicting the weight of oxygen per unit area after decarburization annealing, which is used when manufacturing a grain-oriented electrical steel sheet by hot-rolling a steel material for the grain-oriented electrical steel sheet, annealing the hot-rolled sheet as necessary, cold-rolling once or cold-rolling twice or more with intermediate annealing interposed therebetween, decarburizing and annealing, and predicting the weight of oxygen per unit area after decarburization annealing. A grain-oriented electrical steel sheet is produced by applying an annealing separation agent containing MgO as the main component to the surface of a steel sheet and then performing final annealing, the model being obtained by applying, as explanatory variables, at least three of a steel raw material component, annealing conditions before final cold rolling, and decarburization annealing conditions. And target variables corresponding to the explanatory variables are constructed as the oxygen weight per unit area after the decarburization annealing. Further, a grain-oriented electrical steel sheet is manufactured by setting decarburization annealing conditions using the constructed model for predicting the weight of oxygen per unit area so that the weight of oxygen per unit area after decarburization annealing reaches a target value.

Description

Technical Field

[0001] The present invention relates to a model for predicting the oxygen weight per unit area of the surface of a steel sheet after decarburizing annealing, which has a great influence on the film properties and magnetic properties of a grain-oriented electrical steel sheet, a method for constructing the model, and a method for manufacturing a grain-oriented electrical steel sheet using the above prediction model. Background Art

[0002] Grain-oriented electrical steel sheets are soft magnetic materials mainly used for cores of transformers, etc., and excellent magnetic properties are strongly required. Specifically, low iron loss and high magnetic flux density are strongly required. Such grain-oriented electrical steel sheets are usually manufactured as follows: hot-rolling a steel raw material containing components of formation inhibitors such as MnS, MnSe, and AlN, performing hot-rolled sheet annealing as needed, performing one-time cold rolling or two or more times of cold rolling with intermediate annealing to form a cold-rolled sheet with a final sheet thickness, and after decarburizing annealing of the cold-rolled sheet, performing final annealing to cause secondary recrystallization. It should be noted that in recent years, manufacturing technologies for grain-oriented electrical steel sheets that exhibit secondary recrystallization without using inhibitors have also been developed and put into practical use.

[0003] In the above final annealing, since annealing is performed in a state where the steel sheets are wound into a coil shape, in order to prevent sintering adhesion between the steel sheets, an annealing separating agent mainly composed of MgO is usually coated on the surface of the steel sheet after decarburizing annealing. In addition to the function as the above annealing separating agent, the above MgO also has a function of reacting with an oxide film mainly composed of SiO2 formed on the surface of the steel sheet during decarburizing annealing to form a forsterite film. In addition to imparting insulation to the steel sheet surface, this forsterite film also imparts tensile stress to the steel sheet surface by virtue of the low thermal expansion rate of the film, thereby also contributing to the reduction of iron loss.

[0004] The above forsterite film is formed as follows. First, decarburizing annealing is performed on a cold-rolled sheet that has become the final sheet thickness by cold rolling. Through this decarburizing annealing, C in the steel sheet can be reduced to 0.003 mass% or less that can prevent magnetic aging of the product sheet, and an oxide film mainly composed of SiO2 can be formed on the surface of the steel sheet. Then, after coating an annealing separating agent mainly composed of MgO on the surface of the steel sheet, final annealing is performed at a high temperature. At this time, a forsterite film is formed according to the following reaction formula.

[0005] SiO2 + 2MgO → Mg2SiO4

[0006] Thus, the forsterite film is formed using, as one of the raw materials, the SiO2-based oxide film formed on the surface of the steel sheet during decarburizing annealing. Therefore, in order to form a forsterite film with excellent film properties, it is necessary to control the amount of this oxide film within an appropriate range. It should be noted that since the amount of the above-mentioned oxide film is basically proportional to the amount of oxygen on the surface of the steel sheet, hereinafter, "oxygen weight per unit area" is used instead.

[0007] However, the formation behavior of the forsterite film also has a great influence on the formation of inhibitors such as MnS, MnSe, and AlN. In addition, it is known that the forsterite film incorporates the inhibitors that become unnecessary after secondary recrystallization is completed, purifying the steel sheet itself, and thus also contributing to the improvement of magnetic properties. Therefore, forming a forsterite film with excellent film properties, that is, excellent uniformity and adhesion, is extremely important in manufacturing grain-oriented electrical steel sheets with excellent magnetic properties.

[0008] Regarding methods for manufacturing grain-oriented electrical steel sheets with excellent film properties, many techniques have been disclosed. For example, Patent Document 1 discloses the following technique: by containing either or both of Sb and Cu in the steel raw material in a manner that satisfies a given relationship with Cr, an oxide film with excellent adhesion is formed on the surface of the steel sheet, thereby manufacturing grain-oriented electrical steel sheets with excellent magnetic properties. In addition, Patent Document 2 discloses the following technique: by controlling the depth of the desiliconized layer before the final cold rolling within a given range, grain-oriented electrical steel sheets that can achieve a reduction in iron loss and have a uniform surface condition over the entire coil can be manufactured.

[0009] Prior Art Documents

[0010] Patent Documents

[0011] Patent Document 1: Japanese Patent Application Laid-Open No. 2003-193134

[0012] Patent Document 2: Japanese Patent Application Laid-Open No. 11-152517 Summary of the Invention

[0013] Technical Problem to be Solved by the Invention

[0014] By applying the above prior art, a certain degree of improvement effect of film properties can be obtained. However, according to the research of the inventors, even when applying the above techniques, deterioration of film properties and magnetic properties caused by the deviation of the oxygen weight per unit area on the surface of the steel sheet after decarburizing annealing are still often observed.

[0015] The present invention has been completed in view of the above problems of the prior art, and an object thereof is to provide a model for predicting the oxygen weight per unit area of the surface of a steel sheet after decarburizing annealing for stably manufacturing an oriented electrical steel sheet having excellent film characteristics and magnetic characteristics, and to propose a method for constructing the model and a method for manufacturing an oriented electrical steel sheet using the above model.

[0016] Means for Solving the Problem

[0017] In order to solve the above technical problems, the inventors have repeatedly conducted in-depth research focusing on the reasons for the variation in the oxygen weight per unit area of the surface of the steel sheet after decarburizing annealing. The oxygen weight per unit area varies not only according to the decarburizing annealing conditions but also according to various factors such as the composition of the steel raw material and the manufacturing conditions before the decarburizing annealing process. In the above prior art, any one of these conditions is limited to control the oxygen weight per unit area. However, it is considered that it is necessary to limit all of the above manufacturing conditions to control the oxygen weight per unit area. Therefore, in the present invention, a prediction model for the oxygen weight per unit area was studied, which uses the above-mentioned multiple manufacturing conditions as explanatory variables and the oxygen weight per unit area corresponding to these manufacturing conditions as the target variable. As a result, it was found that in order to construct a high-precision prediction model for the oxygen weight per unit area, it is important to use at least the composition of the steel raw material, the annealing conditions before the final cold rolling, and the decarburizing annealing conditions as the above explanatory variables. It was also found that by setting the decarburizing annealing conditions using the above prediction model for the oxygen weight per unit area, the deviation of the oxygen weight per unit area can be significantly reduced, and thus an oriented electrical steel sheet having excellent film characteristics and magnetic characteristics can be stably manufactured, and the present invention was completed.

[0018] That is, the present invention is a prediction model for the oxygen weight per unit area after decarburizing annealing, which is used when manufacturing an oriented electrical steel sheet by the following method:

[0019] Hot-rolling a steel raw material for an oriented electrical steel sheet,

[0020] Optionally performing hot-rolled sheet annealing,

[0021] Performing one cold rolling or two or more cold rollings with intermediate annealing,

[0022] Performing decarburizing annealing,

[0023] After coating an annealing separating agent mainly composed of MgO on the surface of the steel sheet,

[0024] Performing final annealing to manufacture an oriented electrical steel sheet,

[0025] Among them, at least one of the hot-rolled sheet annealing and the intermediate annealing must be performed,

[0026] The unit area oxygen weight prediction model uses at least the steel raw material composition, the annealing conditions before the final cold rolling, and the decarburizing annealing conditions as explanatory variables, and uses the target variable corresponding to these explanatory variables as the unit area oxygen weight after decarburizing annealing. Here, the above-mentioned final cold rolling refers to cold rolling to the final sheet thickness. In addition, for the annealing before the above-mentioned final cold rolling, when the annealing performed in the process before the final cold rolling is only hot-rolled sheet annealing, it refers to hot-rolled sheet annealing, and when it is only intermediate annealing or both hot-rolled sheet annealing and intermediate annealing, it refers to intermediate annealing.

[0027] The unit area oxygen weight prediction model of the above-mentioned grain-oriented electrical steel sheet of the present invention is characterized in that the following variables are used among the explanatory variables: as the above-mentioned steel raw material composition, at least the concentrations of C, Si, Mn, and Sb are used; as the above-mentioned annealing conditions before the final cold rolling, at least the soaking temperature, the soaking time, and the oxygen potential P H2O / P H2 of the atmosphere gas during soaking are used; as the above-mentioned decarburizing annealing conditions, at least the decarburizing temperature, the decarburizing time, and the oxygen potential P H2O / P H2 .

[0028] In addition, the present invention provides a method for constructing a unit area oxygen weight prediction model after decarburizing annealing, which is used when manufacturing a grain-oriented electrical steel sheet by the following method:

[0029] Hot-rolling the steel raw material for the grain-oriented electrical steel sheet,

[0030] Optionally performing hot-rolled sheet annealing,

[0031] Performing 1-time cold rolling or 2 or more times of cold rolling with intermediate annealing,

[0032] Performing decarburizing annealing,

[0033] After coating an annealing release agent mainly composed of MgO on the steel sheet surface,

[0034] Performing final annealing to manufacture the grain-oriented electrical steel sheet,

[0035] wherein at least one of the above-mentioned hot-rolled sheet annealing and intermediate annealing must be performed,

[0036] The method includes: using at least the steel raw material composition, the annealing conditions before the final cold rolling, and the decarburizing annealing conditions as explanatory variables, and using the target variable corresponding to these explanatory variables as the oxygen weight per unit area after decarburizing annealing to construct a model for predicting the oxygen weight per unit area. Here, the above-mentioned final cold rolling refers to cold rolling to the final sheet thickness. In addition, for the annealing before the above-mentioned final cold rolling, when the annealing carried out in the process before the final cold rolling is only hot-rolled sheet annealing, it refers to hot-rolled sheet annealing; when it is only intermediate annealing or both hot-rolled sheet annealing and intermediate annealing, it refers to intermediate annealing.

[0037] The method for constructing the prediction model of the oxygen weight per unit area of the above-mentioned grain-oriented electrical steel sheet of the present invention is characterized in that the following variables are used in the explanatory variables: as the above-mentioned steel raw material composition, at least the concentrations of C, Si, Mn, and Sb are used; as the above-mentioned annealing conditions before the final cold rolling, at least the soaking temperature, the soaking time, and the oxygen potential P H2O / P H2 of the atmosphere gas during soaking are used; as the above-mentioned decarburizing annealing conditions, at least the decarburizing temperature, the decarburizing time, and the oxygen potential P H2O / P H2 .

[0038] The method for constructing the prediction model of the oxygen weight per unit area of the above-mentioned grain-oriented electrical steel sheet of the present invention is characterized in that the above-mentioned prediction model of the oxygen weight per unit area is constructed by multiple regression analysis or machine learning.

[0039] In addition, the present invention provides a method for manufacturing a grain-oriented electrical steel sheet, which includes the following series of processes:

[0040] Hot-rolling the steel raw material for the grain-oriented electrical steel sheet,

[0041] Optionally performing hot-rolled sheet annealing,

[0042] Performing 1-time cold rolling or 2 or more times of cold rolling with intermediate annealing,

[0043] Performing decarburizing annealing,

[0044] After coating the annealing separating agent mainly composed of MgO on the steel sheet surface,

[0045] Performing final annealing,

[0046] Among them, at least one of the above-mentioned hot-rolled sheet annealing and intermediate annealing must be performed,

[0047] Using the above-mentioned prediction model of the oxygen weight per unit area, setting the decarburizing annealing conditions based on at least the steel raw material composition and the annealing conditions before the final cold rolling, so that the oxygen weight per unit area after decarburizing annealing reaches the target value.

[0048] Effects of the Invention

[0049] According to the present invention, it is possible to control the oxygen weight per unit area on the surface of a steel sheet after decarburizing annealing within an appropriate range and to significantly reduce its deviation. Therefore, it is possible to stably manufacture an oriented electrical steel sheet having excellent film properties and magnetic properties. Detailed Description of the Invention

[0050] First, the basic technical idea of the present invention will be described.

[0051] As described above, it is considered that the amount of the oxide film on the surface of the steel sheet after decarburizing annealing, that is, the oxygen weight per unit area, varies not only according to the decarburizing annealing conditions but also according to various conditions such as the composition of the steel raw material as the starting material and the manufacturing conditions before the decarburizing annealing process. For example, regarding the influence of the decarburizing annealing conditions, when the decarburizing annealing temperature (decarburizing temperature) is high, the diffusion of oxygen is promoted, and thus the oxygen weight per unit area becomes large. The same applies when the decarburizing annealing time (decarburizing time) is long or when the oxygen potential (P H2O / P H2 ) of the atmosphere gas during decarburizing is high.

[0052] In addition, regarding the influence of the composition of the steel raw material, when the C concentration is high, more oxygen is consumed due to decarburization, and thus the oxygen weight per unit area decreases. On the contrary, when the Si concentration is high, the amount of diffusion of Si to the surface of the steel sheet increases, and the Si oxide increases, so the oxygen weight per unit area becomes large. When the Mn concentration is high, the formation of higher-order oxides such as (Fe,Mn)2SiO4 is also promoted, and thus the oxygen weight per unit area becomes large. On the other hand, when the Sb concentration is high, the diffusion of oxygen is inhibited due to the surface segregation of Sb, and thus the oxygen weight per unit area becomes small.

[0053] In addition, in the annealing process before cold rolling, a desiliconized layer is formed on the surface layer of the steel sheet. The desiliconized layer has the effect of suppressing the initial oxidation during decarburizing annealing and preventing the oxygen weight per unit area from increasing excessively. In addition, when the annealing temperature is high, the above-mentioned desiliconized layer becomes thick, and thus the oxygen weight per unit area becomes small. The same applies when the annealing time is long or when the oxygen potential (P H2O / P H2 ) of the atmosphere gas is high. In particular, this tendency is very significant in the annealing close to the final cold rolling (cold rolling to the final sheet thickness).

[0054] As described above, many manufacturing conditions affect the amount of the oxide film (oxygen weight per unit area) formed on the surface of the steel sheet by decarburizing annealing. However, as described above, in the prior art, any one of these manufacturing conditions is limited to control the oxygen weight per unit area. However, it is considered that, in essence, all manufacturing conditions must be considered to control the oxygen weight per unit area.

[0055] Therefore, in the present invention, by using these numerous manufacturing conditions as explanatory variables and the oxygen weight per unit area corresponding to these manufacturing conditions as the target variable, multiple regression analysis or machine learning is performed to construct a model for predicting the oxygen weight per unit area, and the prediction accuracy of the oxygen weight per unit area after decarburizing annealing is studied. As a result, it is known that by using at least the steel raw material composition, the annealing conditions before final cold rolling, and the decarburizing annealing conditions as explanatory variables among the numerous manufacturing conditions, a prediction model for the oxygen weight per unit area with high accuracy can be obtained. Moreover, it is found that by using this prediction model and setting the decarburizing annealing conditions based on at least the steel raw material composition and the annealing conditions before final cold rolling, the oxygen weight per unit area on the surface of the steel sheet after decarburizing annealing can be controlled within a given range, and the deviation can be significantly reduced. Therefore, an oriented electrical steel sheet with excellent film properties and magnetic properties can be stably manufactured, and thus the present invention is completed.

[0056] Next, the preferred component composition of the steel raw material used for manufacturing the oriented electrical steel sheet of the present invention will be described.

[0057] C: 0.01 to 0.1 mass%

[0058] C (carbon) is an important component for improving the texture of the steel sheet. When the content is less than 0.01 mass%, the above-mentioned improvement effect cannot be fully obtained. On the other hand, when C exceeds 0.1 mass%, it is difficult to reduce it to less than 0.003 mass% where magnetic aging does not occur during decarburizing annealing. Therefore, C is preferably in the range of 0.01 to 0.1 mass%, and more preferably in the range of 0.02 to 0.08 mass%.

[0059] Si: 2.0 to 5.0 mass%

[0060] Si (silicon) is an effective component for increasing the resistivity of the steel and reducing the eddy current loss. However, when the content of Si is less than 2.0 mass%, the above-mentioned effect cannot be fully obtained. On the other hand, when it exceeds 5.0 mass%, the cold rollability is significantly reduced. Therefore, it is preferably contained in the range of 2.0 to 5.0 mass%, and more preferably in the range of 2.5 to 4.5 mass%.

[0061] Mn: 0.01 to 1.0 mass%

[0062] Like Si, Mn (manganese) has the effect of increasing the resistivity of steel and reducing eddy current loss. In addition, it also has the effect of improving hot rolling properties. However, when the content of Mn is less than 0.01 mass%, the above effects cannot be fully obtained. On the other hand, when it exceeds 1.0 mass%, it will induce γ phase transformation after secondary recrystallization, which has an adverse effect on magnetic properties. Therefore, it is preferably contained in the range of 0.01 to 1.0 mass%, and more preferably in the range of 0.01 to 0.5 mass%.

[0063] For components other than the above C, Si, and Mn, the situation of using inhibitors such as AlN, MnS, MnSe, etc. to cause secondary recrystallization is different from the situation of not using them.

[0064] In the case of using inhibitors to cause secondary recrystallization, when using AlN as an inhibitor, it is preferably contained Al: 0.010 to 0.04 mass% and N: 0.005 to 0.01 mass%. On the other hand, when using MnS and / or MnSe as inhibitors, in addition to the above Mn, it is preferably further contained S: 0.005 to 0.03 mass% and / or Se: 0.005 to 0.03 mass%. It should be noted that the above inhibitors can be used alone or in combination.

[0065] On the other hand, in the case of not using inhibitors for secondary recrystallization, it is preferable to minimize Al, N, S, and Se as components forming inhibitors. Specifically, it is preferably Al: less than 0.010 mass%, N: less than 0.005 mass%, S: less than 0.005 mass%, and Se: less than 0.005 mass%.

[0066] In addition, in addition to the above components, in order to improve magnetic properties, the steel raw material used in the present invention may further contain any one or more of B: 0.0001 to 0.005 mass%, Ti: 0.001 to 0.01 mass%, P: 0.005 to 0.1 mass%, Cr: 0.01 to 0.5 mass%, Ni: 0.01 to 1.5 mass%, Cu: 0.01 to 0.5 mass%, Nb: 0.002 to 0.08 mass%, Mo: 0.005 to 0.1 mass%, Sn: 0.005 to 0.5 mass%, Sb: 0.005 to 0.5 mass%, and Bi: 0.001 to 0.05 mass%.

[0067] In the steel raw material used in the present invention, the balance other than the above components is substantially Fe and unavoidable impurities.

[0068] It should be noted that, among the steel raw material components described above, as mentioned above, the contents of C, Si, and Mn, which are essential components, have a great influence on the oxygen weight per unit area of the surface of the steel sheet after decarburizing annealing. Therefore, they are necessary as explanatory variables used in the construction of the prediction model. In addition, among the optional additive elements described above, especially Sb segregates on the surface of the steel sheet during annealing and inhibits the diffusion of oxygen, which is an element that has a great influence on the oxygen weight per unit area after decarburizing annealing. Therefore, although Sb is an optional additive element, even in the case where it is not intentionally added (the case where it is contained as an unavoidable impurity), it is also necessary as an essential explanatory variable used in the construction of the prediction model for the oxygen weight per unit area after decarburizing annealing.

[0069] Next, a method for manufacturing the oriented electromagnetic steel sheet of the present invention will be described.

[0070] First, a steel raw material having the composition of the present invention conforming to the above description is hot-rolled to produce a hot-rolled sheet, and if necessary, the hot-rolled sheet is subjected to hot-rolled sheet annealing. Here, the conditions for the above hot-rolled sheet annealing are preferably in the range of soaking temperature: 800 to 1150 °C, soaking time: 20 to 120 s. When the soaking temperature of the hot-rolled sheet annealing is less than 800 °C and / or the soaking time is less than 20 s, there is a risk that the banded structure formed during hot rolling remains, and a uniformly sized primary recrystallized structure cannot be obtained, which hinders the grain growth during secondary recrystallization. On the other hand, when the soaking temperature of the hot-rolled sheet annealing exceeds 1150 °C and / or the soaking time exceeds 120 s, the particle size after hot-rolled sheet annealing becomes too large, and it is still difficult to obtain a uniformly sized primary recrystallized structure.

[0071] In addition, for the gas atmosphere during soaking in the above hot-rolled sheet annealing, from the viewpoint of promoting the secondary recrystallization of sharp Goss orientation and improving magnetic properties by forming a decarburized layer on the surface layer of the steel sheet after hot-rolled sheet annealing, it is preferably an oxidizing gas atmosphere mainly composed of an inert gas, and the oxygen potential P H2O / P H2 is set in the range of 0.1 to 1.0. When P H2O / P H2 is less than 0.1, the formation of the decarburized layer is not promoted, and the magnetic properties of the product deteriorate. On the other hand, when it exceeds 1.0, a large amount of oxides are formed on the surface layer of the steel sheet, it becomes difficult to remove the scale, and the rolling load during subsequent cold rolling increases, resulting in a decrease in productivity.

[0072] Next, the hot-rolled sheet after the above hot rolling or after annealing of the hot-rolled sheet is descaled by pickling or the like, and then, it is cold-rolled once or cold-rolled two or more times with intermediate annealing to produce a cold-rolled sheet having a final sheet thickness (product sheet thickness). It should be noted that in the case of performing the above intermediate annealing, the conditions are preferably set in the range of soaking temperature: 900 to 1200 °C, soaking time: 20 to 120 s. When the soaking temperature of the intermediate annealing is lower than 900 °C and / or the soaking time is less than 20 s, there is a risk that the grain diameter after the intermediate annealing is too small, the Goss nuclei in the primary recrystallized structure are reduced, and the magnetic properties deteriorate. On the other hand, when the soaking temperature exceeds 1200 °C and / or the soaking time exceeds 120 s, the grain diameter after the intermediate annealing becomes too large, and it is still difficult to obtain a primary recrystallized structure with uniform grain size.

[0073] In addition, regarding the gas atmosphere during soaking in the above intermediate annealing, from the viewpoint of controlling the desiliconized layer and oxide layer formed on the surface layer of the steel sheet after the intermediate annealing, it is preferably set to an oxidizing gas atmosphere containing non-reactive gases such as nitrogen and hydrogen, and the oxygen potential P H2O / P H2 is set in the range of 0.01 to 1.0. When P H2O / P H2 is less than 0.01, no desiliconized layer is formed on the surface of the steel sheet after the intermediate annealing, or it becomes too thin, resulting in poor formation of the forsterite film during the final annealing. On the other hand, when it exceeds 1.0, a large amount of oxides are formed on the surface of the steel sheet after the intermediate annealing, so it is difficult to remove them by pickling or the like, causing an increase in the rolling load during subsequent cold rolling and deterioration of the surface properties of the steel sheet, resulting in a reduction in productivity.

[0074] It should be noted that the manufacturing method of the oriented electrical steel sheet of the present invention must perform at least one of the above-described annealing of the hot-rolled sheet and intermediate annealing.

[0075] Next, decarburization annealing that also serves as primary recrystallization annealing is performed on the above cold-rolled sheet having the final sheet thickness. Preferably, the decarburization temperature of the above decarburization annealing is set in the range of 750 to 950 °C, and the decarburization time is set in the range of 80 to 200 s. When the decarburization temperature is lower than 750 °C and / or the decarburization time is less than 80 s, there is a risk of insufficient decarburization, or the grain size of the primary recrystallized grains is too small, the driving force for secondary recrystallization is too large, and the degree of aggregation towards the Goss orientation after secondary recrystallization decreases. On the other hand, when the decarburization temperature exceeds 950 °C and / or the decarburization time exceeds 200 s, there is a risk that the grain size of the primary recrystallized grains becomes too large, which instead inhibits secondary recrystallization.

[0076] In addition, for the gas atmosphere during decarburization in the above decarburizing annealing, from the viewpoints of promoting the decarburization reaction or forming a good internal oxide layer, it is preferably set to a wet hydrogen gas atmosphere such as pure hydrogen alone, or a mixed gas of an inert gas such as nitrogen or argon and hydrogen. The oxygen potential P H2O / P H2 is set in the range of 0.1 to 0.8. When P H2O / P H2 is lower than 0.1, there is a risk of insufficient decarburization, resulting in deterioration of the magnetic properties of the product caused by magnetic aging. On the other hand, when it exceeds 0.8, the oxygen weight per unit area on the surface of the steel sheet after decarburizing annealing becomes excessive, and the forsterite film formed during final annealing becomes too thick, which may cause defects such as pitting peeling.

[0077] It should be noted that the above decarburizing annealing conditions are preferably controlled so that the oxygen weight per unit area (total of both sides) on the surface of the steel sheet after decarburizing annealing reaches 0.70 to 1.50 g / m 2 in the range. When the oxygen weight per unit area is less than 0.70 g / m 2 , there is a risk that the film thickness of the forsterite film formed by final annealing is too thin and is easily damaged by external stress, and the film adhesion becomes poor. On the other hand, when the oxygen weight per unit area exceeds 1.50 g / m 2 , there is a risk that the film thickness of the forsterite film becomes too thick and causes defects such as pitting peeling. More preferably, it is in the range of 0.75 to 1.35 g / m 2 . Here, the above oxygen weight per unit area is obtained by measuring the oxygen content of the total thickness of the steel sheet after decarburizing annealing by JIS G 1239:2014 (Iron and steel - Method for quantitative determination of oxygen - Inert gas fusion - Infrared absorption method) and converting the obtained total oxygen amount into the oxygen amount per unit surface area (both sides) of the steel sheet.

[0078] Here, what is important in the present invention is that it is preferable to construct an oxygen weight per unit area prediction model using the actual data of the manufacturing conditions in the past few years, and use this oxygen weight per unit area prediction model to set the decarburizing annealing conditions based on at least the above steel raw material composition and the annealing conditions before final cold rolling, so that the oxygen weight per unit area on the surface of the steel sheet after decarburizing annealing as the control object reaches a given range. Here, the above final cold rolling refers to cold rolling to the final sheet thickness. In addition, for the annealing conditions before the above final cold rolling, when the annealing performed in the process before the above final cold rolling is only hot rolled sheet annealing, it refers to the conditions of hot rolled sheet annealing, and when it is only intermediate annealing, or when both hot rolled sheet annealing and intermediate annealing are performed, it refers to the conditions of intermediate annealing.

[0079] In addition, the above oxygen weight per unit area prediction model is preferably constructed by multiple regression analysis or machine learning, with at least the steel raw material composition, annealing conditions before final cold rolling, and decarburizing annealing conditions among the actual data of manufacturing conditions in the past few years as explanatory variables, and the oxygen weight per unit area corresponding to these manufacturing conditions as the target variable. However, methods other than the above can also be used.

[0080] Here, importantly, the explanatory variables of the annealing conditions before final cold rolling and the decarburizing annealing conditions used in the construction of the above oxygen weight per unit area prediction model have a significant impact on the oxygen weight per unit area of the steel plate surface after decarburizing annealing. Specifically, for the above annealing conditions before final cold rolling, at least the soaking temperature, soaking time, and oxygen potential P H2O / P H2 of the gas atmosphere during soaking need to be used as the explanatory variables of the oxygen weight per unit area prediction model. In addition, for the above decarburizing annealing conditions, at least the decarburizing temperature, decarburizing time, and oxygen potential P H2O / P H2 of the atmosphere gas during decarburizing need to be used as the explanatory variables of the oxygen weight per unit area prediction model. It should be noted that for the purpose of further improving the accuracy of the oxygen weight per unit area prediction model, manufacturing conditions other than the above can of course be added as explanatory variables.

[0081] In addition, there is no particular limitation on the machine learning algorithm used in the construction of the above oxygen weight per unit area prediction model. For example, random forest can be used, or neural network can also be used.

[0082] Next, after applying an annealing release agent mainly composed of MgO on the surface of the steel plate after the above decarburizing annealing and drying it, final annealing is carried out to cause secondary recrystallization, and a forsterite film is formed. It should be noted that the above "mainly composed of MgO" means that the content of MgO in the whole annealing release agent exceeds 50 mass%.

[0083] Subsequently, after the above steel plate after final annealing removes the unreacted annealing release agent by pickling or the like, flattening annealing that also serves as shape correction of the steel plate, or coating an insulating film, or performing magnetic domain refinement treatment is carried out as needed to produce a product plate.

[0084] [Example 1]

[0085] Using the actual data of the manufacturing conditions of grain-oriented electrical steel sheets in the past two years as explanatory variables and the oxygen weight per unit area of the steel sheet after decarburizing annealing as the target variable, a prediction model for the oxygen weight per unit area of the steel sheet after decarburizing annealing was constructed using random forest, one of the machine learning tools. At this time, the explanatory variables used for constructing the prediction model were set to the following four levels: using only the actual data of the decarburizing annealing conditions in the past two years (level 1); using the actual data of both the steel raw material composition and the decarburizing annealing conditions in the past two years (level 2); using the actual data of both the annealing conditions before the final cold rolling and the decarburizing annealing conditions in the past two years (level 3); and using the actual data of all three of the steel raw material composition, the annealing conditions before the final cold rolling, and the decarburizing annealing conditions in the past two years (level 4).

[0086] Next, steel raw materials (slabs) having the four component compositions shown in Table 1 were hot-rolled, and hot-rolled sheet annealing was performed according to the conditions shown in Table 1 ( soaking temperature, soaking time, P of the atmosphere gas during soaking H2O / P H2 ). Then, one-pass cold rolling (without intermediate annealing) was performed to produce a cold-rolled sheet with a final sheet thickness of 0.23 mm. Then, decarburizing annealing that also served as primary recrystallization annealing was carried out according to the conditions shown in Table 1 (decarburizing temperature, decarburizing time, P of the atmosphere gas during decarburizing H2O / P H2 ). After that, the oxygen weight per unit area of the surface of the steel sheet after the above decarburizing annealing was measured according to JIS G1239:2014, and it was compared with the oxygen weight per unit area obtained by prediction using the prediction models constructed with the explanatory variables of the above levels 1 to 4.

[0087] The above results are collectively shown in Table 1. From these results, it can be seen that in the prediction model for the oxygen weight per unit area of the present invention (level 4) constructed using at least the actual data of all three of the steel raw material composition, the annealing conditions before the final cold rolling, and the decarburizing annealing conditions, the deviation of the measured value of the oxygen weight per unit area from the predicted value is controlled within ±3%, and the oxygen weight per unit area after decarburizing annealing can be predicted with good accuracy.

[0088]

[0089] [Example 2]

[0090] Using the actual data of the steel raw material composition, the annealing conditions before the final cold rolling, and the decarburizing annealing conditions in the past two years as explanatory variables and the oxygen weight per unit area after decarburizing annealing as the target variable, a prediction model for the oxygen weight per unit area was constructed using a neural network, one of the machine learning tools.

[0091] Next, steel raw materials (slabs) having various component compositions shown in Table 2 were hot-rolled, and in PH2O / P H2 : After annealing a hot-rolled sheet with a soaking temperature of 1000 °C and a soaking time of 60 s in a gas atmosphere of 0.30, an intermediate sheet thickness was produced by the first cold rolling. According to the conditions shown in Table 2 (soaking temperature, soaking time, P of the atmosphere gas during soaking) H2O / P H2 ) After intermediate annealing, a cold-rolled sheet with a final sheet thickness of 0.23 mm was produced by the second cold rolling. Then, decarburizing annealing that also serves as primary recrystallization annealing was performed on the above cold-rolled sheet. At this time, as shown in Table 2, the conditions for the above decarburizing annealing were set to the following two conditions: uniformly setting the decarburizing temperature × decarburizing time to 840 °C × 120 s, and setting the oxygen potential P of the atmosphere gas during decarburizing H2O / P H2 to 0.50 (comparative example); and uniformly setting the decarburizing temperature × decarburizing time to 840 °C × 120 s, and using the unit area oxygen weight prediction model constructed above to set the oxygen potential P of the atmosphere gas during decarburizing H2O / P H2 , such that the unit area oxygen weight is 0.90 g / m 2 (invention example). Then, the unit area oxygen weight of the steel sheet after the above decarburizing annealing was measured according to JIS G 1239:2014.

[0092] Next, an annealing release agent containing 2.0 parts by mass of TiO2 added to 100 parts by mass of MgO was made into a slurry, coated on the surface of the steel sheet after the above decarburizing annealing and dried, and then secondary recrystallization was completed under the conditions of 850 °C × 50 hr. Then, a final annealing for purification treatment was performed under the conditions of 1200 °C × 5 hr. Next, after removing the unreacted annealing release agent from the steel sheet after the above final annealing, an insulating film mainly composed of phosphate was coated, and sintering of the film and flattening annealing that also serves as shape correction were performed under the conditions of 850 °C × 1 min to produce a product sheet.

[0093] Next, test pieces were collected from the above product sheet, and the film properties (uniformity, adhesion) and magnetic properties (magnetic flux density B8, iron loss W 17 / 50 ) were evaluated. Specifically, for the uniformity of the film, the film appearance was observed with the naked eye. If it was uniform, it was evaluated as ○; if it was slightly non-uniform, it was evaluated as △; if it was non-uniform, it was evaluated as ×. In addition, for the adhesion of the film, each product sheet was wound around round bars of various diameters, and evaluated by the minimum diameter at which the film did not peel off ("bending peel-off diameter"). In addition, the magnetic flux density B8 and the iron loss W 17 / 50 were measured according to JIS C 2556:2015.

[0094] The results of the above measurements are all recorded in Table 2. According to these results, when the oxygen weight per unit area prediction model of the present invention is applied, the deviation of the oxygen weight per unit area is significantly reduced. As a result, the deviations of the film properties and magnetic properties are also greatly reduced, and the difference ΔB8 = 0.005 T between the maximum and minimum values of the magnetic flux density and the difference ΔW 17 / 50 = 0.062 W / kg are achieved in all steel plates. In contrast, when the oxygen weight per unit area prediction model of the present invention is not applied, in the steel plates of No. 6 and 10 where the oxygen potential P H2O / P H2 happens to be close to the set value 0.50 of the comparative example, good film properties and magnetic properties are obtained. However, generally speaking, the deviations of the film properties and magnetic properties are large (ΔB8 = 0.173 T, ΔW 17 / 50 = 0.383 W / kg), and it can be seen that steel plates with excellent film properties and magnetic properties cannot be stably obtained.

[0095]

Claims

1. A prediction model for the oxygen weight per unit area after decarburization annealing, which is used when manufacturing grain-oriented electrical steel sheets by the following method: Hot-rolling a steel raw material for grain-oriented electrical steel sheets, Optionally performing hot-rolled sheet annealing, Performing one cold rolling or two or more cold rollings with intermediate annealing, Performing decarburization annealing, After coating an annealing separating agent mainly composed of MgO on the steel sheet surface, Performing final annealing to manufacture grain-oriented electrical steel sheets, Among them, At least one of the hot-rolled sheet annealing and the intermediate annealing must be performed, The prediction model for the oxygen weight per unit area at least uses the following three as explanatory variables: the steel raw material composition, the annealing conditions before the final cold rolling, and the decarburization annealing conditions, and uses the target variable corresponding to these explanatory variables as the oxygen weight per unit area after decarburization annealing. Here, the final cold rolling refers to the cold rolling to the final sheet thickness. Moreover, for the annealing before the final cold rolling, when the annealing performed in the process before the final cold rolling is only the hot-rolled sheet annealing, it refers to the hot-rolled sheet annealing, and when it is only the intermediate annealing or both the hot-rolled sheet annealing and the intermediate annealing, it refers to the intermediate annealing.

2. The prediction model for the oxygen weight per unit area according to claim 1, wherein The following variables are used in the explanatory variables: As the steel raw material composition, at least the concentrations of C, Si, Mn, and Sb are used; As the annealing conditions before the final cold rolling, at least use the soaking temperature, soaking time, and the oxygen potential P of the atmosphere gas during soaking H2O / P H2 ; As the decarburization annealing conditions, at least use the decarburization temperature, decarburization time, and the oxygen potential P of the atmosphere gas during decarburization H2O / P H2 .

3. A method for constructing a prediction model for the oxygen amount per unit area after decarburization annealing, which is used when manufacturing grain-oriented electrical steel sheets by the following method: Hot-rolling a steel raw material for grain-oriented electrical steel sheets, Optionally performing hot-rolled sheet annealing, Performing one cold rolling or two or more cold rollings with intermediate annealing, Performing decarburization annealing, After coating an annealing separating agent mainly composed of MgO on the steel sheet surface, Performing final annealing to manufacture grain-oriented electrical steel sheets, Among them, At least one of the hot-rolled sheet annealing and the intermediate annealing must be performed, This method includes: constructing a model for predicting the oxygen weight per unit area after decarburization annealing by using at least the following three as explanatory variables: the steel raw material composition, the annealing conditions before the final cold rolling, and the decarburization annealing conditions, and using the target variable corresponding to these explanatory variables as the oxygen weight per unit area after decarburization annealing. Here, the final cold rolling refers to the cold rolling to the final sheet thickness. Moreover, for the annealing before the final cold rolling, when the annealing performed in the process before the final cold rolling is only the hot-rolled sheet annealing, it refers to the hot-rolled sheet annealing, and when it is only the intermediate annealing or both the hot-rolled sheet annealing and the intermediate annealing, it refers to the intermediate annealing.

4. The method for constructing a prediction model for the oxygen weight per unit area according to claim 3, wherein The following variables are used in the explanatory variables: As the steel raw material composition, at least the concentrations of C, Si, Mn, and Sb are used; As annealing conditions before the final cold rolling, at least use soaking temperature, soaking time, and oxygen potential P of the atmosphere gas during soaking H2O / P H2 ; As the decarburization annealing conditions, at least use the decarburization temperature, the decarburization time, and the oxygen potential P of the atmosphere gas during decarburization H2O / P H2 .

5. The method for constructing a prediction model for the oxygen weight per unit area according to claim 3 or 4, wherein The prediction model for the oxygen weight per unit area is constructed by multiple regression analysis or machine learning.

6. A method for manufacturing grain-oriented electrical steel sheets, which includes the following series of processes: Hot-rolling a steel raw material for grain-oriented electrical steel sheets, Optionally, hot-rolled sheet annealing is carried out. One cold rolling is carried out or two or more cold rollings with intermediate annealing are carried out. Decarburizing annealing is carried out. After applying an annealing separating agent mainly composed of MgO to the surface of the steel sheet, Final annealing is carried out. Among them, At least one of the hot-rolled sheet annealing and the intermediate annealing must be carried out. Using the unit area oxygen weight prediction model described in claim 1 or 2 above, the decarburizing annealing conditions are set based on at least the steel raw material composition and the annealing conditions before the final cold rolling, so that the unit area oxygen weight after decarburizing annealing reaches the target value.

Images