Method for determining decarburization treatment end, vacuum degassing equipment, method for operating vacuum degassing equipment, and method for producing molten steel

Abstract

There are provided a method for determining a decarburization treatment end, vacuum degassing equipment, a method for operating vacuum degassing equipment, and a method for producing molten steel capable of estimating the carbon content in molten steel with high accuracy by properly utilizing exhaust gas measurement values to end the decarburization treatment at appropriate timing. A method for determining a decarburization treatment end for determining an end of decarburization treatment when vacuum degassing treatment including the decarburization treatment is performed by placing molten steel in a reduced-pressure environment includes: dividing the decarburization treatment into two stages of first half decarburization treatment M1 and latter half decarburization treatment M2, and estimating the carbon contents in molten steel in the first half decarburization treatment M1 and the latter half decarburization treatment M2 using different estimation models for the carbon content in molten steel for the first half decarburization treatment M1 and the latter half decarburization treatment M2, in the first half decarburization treatment M1, using the estimation model constructed based on machine learning for actual operation result data (S3), and, in the latter half decarburization treatment M2, determining the end of the decarburization treatment when an estimated value of the carbon content in molten steel has reached a target value (S7).

Description

Technical Field

[0001] The present disclosure relates to a method for determining a decarburization treatment end, vacuum degassing equipment, a method for operating vacuum degassing equipment, and a method for producing molten steel.Background Art

[0002] In a steel making process, molten steel components are adjusted by removing impurities in hot metal, such as carbon, and adding useful alloying components. Particularly for carbon, decarburization is promoted by placing molten steel in a vacuum environment using vacuum degassing equipment, and ultra-low carbon steel can be produced in which the carbon content in molten steel is less than 10 ppm.

[0003] Herein, in vacuum degassing treatment, the carbon content in molten steel is not directly measured, but is only indirectly estimated from the contents of carbon monoxide and carbon dioxide in exhaust gas. In the actual production of the ultra-low carbon steel, the decarburization treatment tends to be performed for an excessively long period due to concern that the carbon content deviates from the standard.

[0004] To solve the prolongation of treatment time by the excessive decarburization treatment, it is effective to estimate the carbon content in molten steel during the treatment with high accuracy, and various methods have been proposed so far. Methods for estimating the carbon content in molten steel can be roughly divided into two methods of estimation methods using exhaust gas measurement values and methods for constructing a decarburization model.

[0005] In estimation methods using the flow rate and measurement values of exhaust gas, the decarburization amount is calculated from measurement values of exhaust gas (e.g., measurement values of the flow rate and the component contents) emitted from vacuum degassing equipment during treatment, and the carbon content in molten steel is estimated. In methods for constructing a decarburization model, the details of decarburization in vacuum degassing equipment are physically considered, and a decarburization model is constructed, thereby estimating the carbon content in molten steel (e.g., NPL 1).

[0006] Further, as a combination of the two methods, methods have been proposed in which the parameters of the decarburization model are determined from the exhaust gas measurement values, and the carbon content in molten steel is estimated using the decarburization model having the determined parameters (e.g., PTL 1 and PTL 2).Citation ListPatent Literatures

[0007] PTL 1: JP 2005-330512 A PTL 2: JP 2015-101742 A Non Patent Literature

[0008] NPL 1: Shinya Kitamura and three others, "Decarburization Model for Vacuum Degasser", Tetsu-to-Hagane, Vol. 80 (1994), No. 3, p. 213-218.Summary of InventionTechnical Problem

[0009] The flow rate and content measurement values of the exhaust gas emitted from the vacuum degassing equipment during the treatment contain information on the actual decarburization results, and the utilization of the information is considered to contribute to the realization of high-accuracy carbon content estimation. However, on the other hand, the exhaust gas measurement values contain errors, and the use of the exhaust gas measurement values for the determination of the model parameters as in the technology of PTL 1 and PTL 2 causes a decrease in carbon content estimation accuracy due to the exhaust gas measurement value errors. In particular, errors allowable in an estimated value of the carbon content in molten steel at the last stage of the decarburization treatment are about several ppm, and the influence of the exhaust gas measurement value errors can be fatal. Thus, to utilize the exhaust gas measurement values for the carbon content estimation, it is important to reduce the adverse effects of the exhaust gas measurement value errors.

[0010] It is an object of the present disclosure made in view of such circumstances to provide a method for determining a decarburization treatment end, vacuum degassing equipment, a method for operating vacuum degassing equipment, and a method for producing molten steel capable of estimating the carbon content in molten steel with high accuracy by properly utilizing the exhaust gas measurement values to end the decarburization treatment at appropriate timing.Solution to Problem

[0011] (1) One aspect of the present disclosure provides a method for determining a decarburization treatment end for determining an end of decarburization treatment when vacuum degassing treatment including the decarburization treatment is performed by placing molten steel in a reduced-pressure environment, the method including: dividing the decarburization treatment into two stages of first half decarburization treatment and latter half decarburization treatment, and estimating the carbon contents in molten steel in the first half decarburization treatment and the latter half decarburization treatment using different estimation models for the carbon content in molten steel for the first half decarburization treatment and the latter half decarburization treatment, in the first half decarburization treatment, using the estimation model constructed based on machine learning for actual operation result data, and in the latter half decarburization treatment, determining the end of the decarburization treatment when an estimated value of the carbon content in molten steel has reached the target value. (2) A method for determining decarburization treatment end is provided in which pressure in the reduced-pressure environment is used for a division reference for the first half decarburization treatment and the latter half decarburization treatment in the configuration according to (1) above. (3) The method for determining decarburization treatment end is provided in which, in the estimation model of the first half decarburization treatment, an explanatory variable is constructed by operation information containing molten steel information and actual operation result values and an object variable is constructed by the carbon content in molten steel at the end of the first half decarburization treatment, the molten steel information is information on the weight and the component content of the molten steel before the vacuum degassing treatment, and the actual operation result values contain the flow rate of exhaust gas emitted during the vacuum degassing treatment and a measurement result of the component content in the exhaust gas in the configuration according to (2) above. (4) One aspect of the present disclosure provides vacuum degassing equipment including: a vacuum degassing device configured to perform vacuum degassing treatment including decarburization treatment by placing molten steel in a reduced-pressure environment; a control device configured to control the vacuum degassing device; and a decarburization treatment end determination device configured to determine an end of the decarburization treatment, in which the decarburization treatment end determination device has a carbon content calculation unit configured to divide the decarburization treatment into two stages of first half decarburization treatment and latter half decarburization treatment, estimate the carbon contents in molten steel in the first half decarburization treatment and the latter half decarburization treatment using different estimation models for the carbon content in molten steel for the first half decarburization treatment and the latter half decarburization treatment, and use the estimation model constructed based on machine learning for actual operation result data in the first half decarburization treatment, and a decarburization treatment end determination unit configured to determine the end of the decarburization treatment when an estimated value of the carbon content in molten steel has reached the target value in the latter half decarburization treatment. (5) The vacuum degassing equipment is provided in which the carbon content calculation unit uses pressure in the reduced-pressure environment for a division reference for the first half decarburization treatment and the latter half decarburization treatment in the configuration of (4) above. (6) The vacuum degassing equipment is provided in which, in the estimation model of the first half decarburization treatment, an explanatory variable is constructed by operation information containing molten steel information and actual operation result values and an object variable is constructed by the carbon content in molten steel at the end of the first half decarburization treatment, the molten steel information is information on the weight and the component content of the molten steel before the treatment of the vacuum degassing treatment, and the actual operation result values contain the flow rate of exhaust gas emitted during the vacuum degassing treatment and a measurement result of the component content in the exhaust gas in the configuration of (5) above. (7) One aspect of the present disclosure provides a method for operating vacuum degassing equipment performing vacuum degassing treatment including decarburization treatment by placing molten steel in a reduced-pressure environment, the method including: using the decarburization treatment determination method according to any one of the configurations (1) to (3) above when an end of the decarburization treatment is determined. (8) One aspect of the present disclosure provides a method for producing molten steel for producing the molten steel refined by applying vacuum degassing treatment to the molten steel in vacuum degassing equipment operated by the operation method according to the configuration of (7) above. Advantageous Effects of Invention

[0012] One aspect of the present disclosure can provide a method for determining a decarburization treatment end, vacuum degassing equipment, a method for operating vacuum degassing equipment, and a method for producing molten steel capable of estimating the carbon content in molten steel with high accuracy by properly utilizing the exhaust gas measurement values to end the decarburization treatment at appropriate timing.Brief Description of Drawings

[0013] FIG. 1 is a structural view of vacuum degassing equipment and a block diagram of a determination device according to one embodiment of the present disclosure; FIG. 2 is a flowchart illustrating a method for determining a decarburization treatment end according to one embodiment of the present disclosure; and FIG. 3 is a graph showing an example of a change with time in vacuum chamber-inside pressure in vacuum degassing treatment charge in which ultra-low carbon steel grade production was performed. Description of Embodiments

[0014] A detailed description below describes an embodiment of the present disclosure with reference to the drawings. In the description of the drawings, the same or similar reference numerals are attached to the same or similar parts, and duplicated descriptions are omitted. The drawings are schematic and are sometimes different from the actual ones. The embodiment described below exemplifies devices and methods for embodying the technical idea of the present disclosure. The technical idea of the present disclosure does not specify materials, structures, arrangement, and the like of constituent components to the materials, structures, arrangement, and the like described below. The technical idea of the present disclosure can be variously altered within the technical range defined by Claims.<Vacuum degassing device>

[0015] Hereinafter, vacuum degassing equipment in one embodiment of the present disclosure is described with reference to the drawings. As illustrated in FIG. 1, vacuum degassing equipment 1 includes a vacuum degassing device 2, a control device 3, and a decarburization treatment end determination device (hereinafter also simply referred to as a "determination device") 4.

[0016] The vacuum degassing device 2 is an RH vacuum degassing device, and applies vacuum degassing treatment to molten steel 6 contained in a ladle 5. The vacuum degassing device 2 includes a vacuum chamber 20, a pair of immersion tubes 21, piping 22, an injection lance 23, an exhaust duct 24, an exhaust gas flow rate meter 25, and an exhaust gas component content meter 26.

[0017] In the vacuum degassing device 2, the pair of immersion tubes 21 provided at the bottom of the vacuum chamber 20 is immersed in molten steel 6 in the ladle 5, and gas inside the vacuum chamber 20 is exhausted through the exhaust duct 24, so that the pressure in the vacuum chamber 20 is reduced and the molten steel 6 is sucked up. Then, an inert gas is injected from one of the immersion tubes 21 through the piping 22, so that the molten steel 6 is refluxed between the vacuum chamber 20 and the ladle 5. In the vacuum degassing device 2, oxygen can be supplied to the molten steel 6 in the vacuum chamber 20 by injecting oxygen from the injection lance 23 installed in the vacuum chamber 20. The vacuum chamber 20 is one example of an evacuated region of the vacuum degassing device 2, i.e., a region that is pressure-reduced to be vacuum. In such a vacuum degassing device 2, a degassing reaction and a decarburization of the molten steel 6 are promoted by refluxing the molten steel 6 containing oxygen in the evacuated region. In this embodiment, the vacuum degassing treatment by the vacuum degassing equipment 1 is refining treatment including decarburization treatment (vacuum decarburization treatment) of promoting decarburization by placing the molten steel 6 in a reduced-pressure environment.

[0018] The exhaust gas flow rate meter 25 and the exhaust gas component content meter 26 are provided inside the exhaust duct 24. The exhaust gas flow rate meter 25 measures the flow rate of exhaust gas. The exhaust gas component content meter 26 measures the contents of components in exhaust gas containing a CO gas, a CO 2 gas, and an O 2 gas.

[0019] The control device 3 contains an information processer, such as a computer. The control device 3 controls the operation amounts related to the operation including the exhaust amount of exhaust equipment, the flow rate of an inert gas for reflux, and the injected oxygen flow rate such that the component contents and the temperature of the molten steel 6 fall within the target range after the vacuum degassing treatment from the actual result values before the vacuum degassing treatment. The control device 3 also collects actual operation result value data including the pressure in the vacuum chamber 20 (vacuum chamber-inside pressure), the flow rate of the inert gas for reflux, the injected oxygen flow rate, the exhaust gas flow rate, and the exhaust gas component contents and outputs the actual operation result value data to the determination device 4.

[0020] The determination device 4 is a device performing decarburization treatment determination of determining the end timing of the decarburization treatment in the vacuum degassing treatment and contains an information processer, such as a computer. The determination device 4 functions as an operation information input unit (hereinafter also simply referred to as an "input unit") 40, a carbon content calculation unit (hereinafter also simply referred to as a "calculation unit") 41, a decarburization treatment end determination unit (hereinafter also simply referred to as an "end determination unit") 42, and a decarburization treatment phase determination unit (hereinafter also simply referred to as a "phase determination unit") 43 by execution of a computer program by a processing unit, such as a central processing unit (CPU), in the information processer. The control device 3 and the determination device 4 are also collectively referred to as a vacuum degassing treatment control system.

[0021] The input unit 40 inputs the actual operation results related to the operation amounts and the time-series exhaust gas measurement values during the vacuum degassing treatment as input information. The time-series exhaust gas measurement values include the flow rate of the exhaust gas emitted from the vacuum degassing device 2 and the component contents of a CO gas, a CO 2 gas, and an O 2 gas contained in the exhaust gas.

[0022] The calculation unit 41 estimates the carbon content in molten steel 6 (carbon content in molten steel) based on the input information acquired by the input unit 40. In a method for calculating the carbon content in molten steel, a plurality of estimation models for the carbon content in molten steel are properly used based on the determination reference including the vacuum chamber-inside pressure.

[0023] The determination unit 42 determines the decarburization treatment end based on the estimated value of the carbon content in molten steel calculated by the calculation unit 41. The determination unit 42 outputs the determination result of the decarburization treatment end to the control device 3. The control device 3 may control the operation amounts related to the operation based on the decarburization treatment end determination result obtained from the decarburization treatment end determination unit 42.

[0024] The phase determination unit 43 determines the phase of the decarburization treatment based on the input information acquired by the input unit 40. Specifically, the phase determination unit 43 determines whether the phase of the decarburization treatment is the first half or the latter half by determining the end of the first half of the decarburization treatment described later.

[0025] In the vacuum degassing equipment 1, the component adjustment including the decarburization treatment and the temperature adjustment of the molten steel 6 are performed by applying the vacuum degassing treatment to the molten steel 6, so that the molten steel 6 having the specified components and temperature is produced. Then, in the decarburization treatment, the internal state and the like of the vacuum degassing device 2 are estimated by the determination device 4, so that the timing of the end of the decarburization treatment is determined. Then, the vacuum degassing treatment is performed based on the determination result of the decarburization treatment. For steps of the vacuum degassing treatment other than the determination of the end of the decarburization treatment end, known one can be used.<Method for determining end of decarburization treatment>

[0026] A method for determining an end of the decarburization treatment according to this embodiment is performed according to the flowchart illustrated in FIG. 2. The determination of the end of the decarburization treatment according to this embodiment roughly contains estimation calculation of the carbon content in molten steel in first half decarburization treatment (hereinafter also simply referred to as a "first half") M1 and estimation calculation of the carbon content in molten steel and the determination of the end of the decarburization treatment in latter half decarburization treatment (hereinafter also simply referred to as a "latter half") M2. The switch from the first half M1 to the latter half M2 of the decarburization treatment is performed based on a determination reference including the pressure in the vacuum chamber 20. The flowchart illustrated in FIG. 2 starts at timing when execution instruction of the decarburization treatment is input, and the treatment proceeds to the treatment of Step S1.

[0027] First, in the treatment of Step S1, the input unit 40 acquires molten steel information before the start of the decarburization treatment from a host computer or the like which is not illustrated. The molten steel information may contain, for example, the weight of the molten steel 6 and the component contents which are the measurement and analysis results obtained by component analysis of the molten steel 6. This completes the treatment of Step S1, and the treatment of determining the decarburization treatment end proceeds to the treatment of Step S2.

[0028] In the treatment of Step S2, the input unit 40 acquires the actual operation result values during the decarburization treatment. The actual operation result values are items required for the calculation in the calculation unit 41. In this embodiment, the input unit 40 acquires at least the measurement results of the flow rate and the component contents of the exhaust gas emitted from the vacuum degassing device 2 during the vacuum degassing treatment as the actual operation result values. Further, in this embodiment, the input unit 40 acquires the pressure in the vacuum chamber 20 during the decarburization treatment (pressure in a reduced-pressure environment in the vacuum chamber 20) as the actual operation result value. Further, the actual operation result values may contain information on the flow rate of the inert gas for reflux and the flow rate of oxygen from the injection lance 23 during the decarburization treatment and information on the auxiliary raw materials charged during the decarburization treatment. Specific examples of the information on auxiliary raw materials include the types and the charging amounts of the auxiliary raw materials. This completes the treatment of Step S2, and the treatment of determining the decarburization treatment end proceeds to the treatment of Step S3. Herein, Step S1 and Step S2 correspond to the input steps. The molten steel information and the actual operation result values are also collectively referred to as operation information.

[0029] In the treatment of Step S3, the phase determination unit 43 determines the end of the first half of the decarburization treatment based on the actual operation result values acquired in the treatment of Step S2. The reference for the end determination includes at least the fact whether the pressure of the vacuum chamber 20 has reached a predetermined threshold (or has reached a value equal to or less than a predetermined threshold). When required in view of a condition desirable as the boundary between the first half and the latter half of the decarburization treatment described later, a condition may be added that is required to be satisfied to determine the end of the first half M1 (first half end), in addition to the condition that the pressure of the vacuum chamber 20 is equal to or less than the threshold. When it has been determined that the first half M1 has not ended, the treatment returns to the treatment of Step S2 and repeat the treatment in and after Step S2 using newly input actual operation result values. On the other hand, when it has been determined that the first half M1 has ended, the treatment of Step S3 is completed and the treatment of determining the decarburization treatment end proceeds to the treatment of Step S4.

[0030] In the treatment of Step S4, the calculation unit 41 estimates the carbon content in molten steel at the point in time of the first half M1 end (at the point in time of the first half end). The smaller a variation in the actual result value of an object variable for an estimation model in general without being limited a machine learning model, the easier it will be to construct a high-accuracy estimation model. In the carbon content estimation in the vacuum decarburization treatment, factors causing a variation in carbon content in molten steel at the point in time of the first half end which is the object variable include the carbon content in molten steel before the vacuum decarburization treatment (carbon content before treatment). The variation factors of the carbon content in molten steel also include an airtight condition change of the vacuum chamber 20 and the exhaust system due to a change in connection between the vacuum chamber 20 and the exhaust system and a change in time required for the air pressure in the vacuum chamber 20 to decrease to the ultimate air pressure due to the oxygen injection during the vacuum degassing treatment. The pressure in the vacuum chamber 20 is referred to as a vacuum chamber-inside pressure.

[0031] Herein, the present inventors have obtained finding that determining the first half end based on the vacuum chamber-inside pressure reduces the variation in the actual result value of the carbon content in molten steel at the point in time of the first half end. Among the variation factors of the carbon content in molten steel described above, the following consideration of the carbon content before treatment shows the validity of this finding. The vacuum chamber-inside pressure is determined by the balance between the exhaust rate from the vacuum chamber 20 and the supply rate of gas to the vacuum chamber 20. The former can be regarded to be an approximately constant value according to the exhaust capacity of the vacuum degassing device 2. On the other hand, for the latter, the higher the decarburization rate, i.e., the supply rate of a CO gas to the vacuum chamber 20, the higher the carbon content in molten steel. Thus, the carbon content in molten steel at a predetermined vacuum chamber-inside pressure is considered to have an approximately constant value.

[0032] The vacuum chamber-inside pressure serving as the reference for the first half end is preferably determined considering the following two points. First, as the first point, it is desirable to select pressure having uniqueness of time in the vacuum decarburization treatment. In the vacuum degassing treatment, the vacuum chamber-inside pressure does not monotonously decrease but sometimes temporarily increases when the operation mode of the exhaust equipment is switched, when the decarburization is active, when oxygen is injected, or the like. The first half end is required to be properly determined in the middle of the decarburization treatment. The presence of a plurality of points in time where the same pressure is taken leads to a variation in carbon content in molten steel at the point in time of the first half end. In the foregoing description, the adverse effects where there is a plurality of points in time where the same pressure is taken are described; on the other hand, when the vacuum chamber-inside pressure serving as the reference is excessively low, the vacuum chamber-inside pressure does not decrease to the reference value, and normal determination of the end of the decarburization treatment cannot be performed.

[0033] Next, as the second point, the vacuum chamber-inside pressure serving as the reference is desirably lower to the extent that the variation in the actual result value of the carbon content in molten steel at the point in time of the first half end is not excessively large. For example, setting an extremely high value, such as 300 Torr, to the reference pressure based on the first point described above hardly has an effect on an increase in accuracy of the carbon content estimation. As described later, the exhaust gas measurement value is used to estimate the carbon content in molten steel in the first half M1, but the pressure serving as the reference is desirably lower to maximize the advantage of the exhaust gas measurement values containing information on the actual decarburization result.

[0034] When there is no vacuum chamber-inside pressure satisfying the above-described two points and the timing of the first half end cannot be properly determined with only the pressure serving as the reference, other conditions may be added. For example, the presence or absence of oxygen injection, the treatment time, and the like are considered to be additional conditions.

[0035] FIG. 3 illustrates an example of a change with time in the vacuum chamber-inside pressure in a vacuum degassing treatment charge in which ultra-low charcoal steel grade production was performed. From FIG. 3, when the pressure serving as the reference for the first half end of the decarburization treatment is selected, it is desirable to select the pressure in the range of 3 Torr or more and 6 Torr or less or 60 Torr or more, where the time is uniquely determined from the viewpoint of the above-described first point. In consideration of the above-described second point, it is more preferable to select the pressure serving as the reference in a lower value range of 3 Torr or more and 6 Torr or less. For example, the pressure serving as the reference may be set to 5 Torr. When the pressure serving as the reference for the first half end of the decarburization treatment is selected in a single point, it is desirable to consider the variation in the change with time in the vacuum chamber-inside pressure per charge and the estimation accuracy of the carbon content in molten steel at the point in time of the decarburization treatment end.

[0036] As is clear from FIG. 3, pressure suitable as the reference for the first half end of the decarburization treatment is not always a distinct inflection point with respect to time. In the vacuum degassing treatment of performing the ultra-low charcoal steel grade production, the decarburization rate is finally rate-controlled by the carbon supply. However, high-accuracy estimation of the carbon content in molten steel is possible even when the boundary between the first half and the latter half of the decarburization treatment is set in the middle of the stage of rate-controlling the carbon supply rate.

[0037] In this embodiment, the carbon content in molten steel at the point in time of the first half end is estimated by a machine learning model with the operation information acquired by the input unit 40 as an explanatory variable. For the form of the machine learning model, one capable of estimating the carbon content in molten steel at the point in time of the first half end with high accuracy is selected. Examples of candidates thereof include a linear regression model, a neural network model, a support vector regression model, and a decision tree model. In this embodiment, a linear regression model is used. At this time, the machine learning model is represented by Equation (1) below. [Math. 1] C 0 = ∑ j = 1 N α j X j + α 0

[0038] Herein, C 0 is the estimated value of the carbon content in molten steel at the point in time of the first half end, which is the object variable. X j (j = 1, 2,···, N) is the explanatory variable. a j is a parameter corresponding to the explanatory variable X j . a 0 is the constant term of the machine learning model. The parameter α j and the constant term α 0 can be determined by preparing a large number of data sets of previously collected object variables and explanatory variables (actual operation result data) and applying a known method.

[0039] The explanatory variable X j of the machine learning model includes, for example, the information on the weight and the component contents of the molten steel before the treatment and a plurality of actual operation result values including the measurement results of the flow rate of exhaust gas emitted from the vacuum degassing equipment during the treatment and the component contents in the exhaust gas as shown in Table 1 below. [Table 1]Explanatory variableUnitX 1 Carbon content in molten steel before decarburization treatmentppmX 2 Molten steel weightton (metric ton)X 3 Integrated decarburization amount based on exhaust gas information of first half of decarburization treatmentkgX 4 Integrated injected oxygen amount of first half of decarburization treatmentNm 3< X 5 Treatment time of first half of decarburization treatmentsX 6 Exhaust gas flow rate at the point in time of first half end of decarburization treatmentNm 3< / hX 7 Exhaust gas CO content at the point in time of first half end of decarburization treatmentvol%X 8 Exhaust gas CO 2 content at the point in time of first half end of decarburization treatmentvol%X 9 Exhaust gas O 2 content at the point in time of first half end of decarburization treatmentvol%

[0040] Herein, the integrated decarburization amount based on the exhaust gas information of the first half M1 contained as the explanatory variable in Table 1 is calculated by a calculation process described later.

[0041] In consideration of the fact that carbon emitted from the molten steel takes the form of CO or CO 2 , the carbon amount in the exhaust gas per unit time is given by Equation (2) below. The cumulative emitted carbon amount from the start of the treatment (at time 0) to the first half end (at time t 0 ) is given by Equation (3) below. q C , OG t = m C V off t 22.4 ⋅ r CO t + r CO 2 t 100 2 Q C , OG t 0 = ∫ 0 t 0 q C , OG t d t 3

[0042] Herein, q C,OG (t) is the carbon amount in the exhaust gas per unit time [kg / s] at time t. m C is the molar mass of carbon [g / mol]. V off (t) is the volume flow rate [Nm 3< / s] of the exhaust gas at time t. r CO (t) is the CO content [vol%] in the exhaust gas at time t. r CO2 (t) is the CO 2 content in the exhaust gas at time t [vol%]. Q C,OG (t 0 ) is the cumulative emitted carbon amount [kg] from time 0 to t 0 .

[0043] As described above, the form of the machine learning model is not limited to the linear regression model described in this embodiment. Other model forms, such as a neural network model, may be adopted when a higher effect can be confirmed by the enhancement of the estimation accuracy of the carbon content in molten steel. However, the machine learning model is carefully selected in the following points. It is an ultimate object of the present disclosure to reduce the vacuum degassing treatment time. To realize this, directly, the estimation accuracy of the carbon content in molten steel at the point in time of the decarburization treatment end is required to be high. When the estimation accuracy of the carbon content in molten steel at the point in time of the first half end is high, the estimation accuracy of the carbon content in molten steel at the point in time of the decarburization treatment end can be estimated to be high, but there is no direct relation between the two. Thus, the model is required to be carefully selected such that the estimation accuracy of not the former but the latter is enhanced.

[0044] By applying Equation (1), the carbon content in molten steel at the point in time of the first half end can be estimated. This completes the treatment of Step S4, and the treatment proceeds to the treatment of Step S5 of the latter half M2.

[0045] In the treatment of Step S5, the operation information input unit 40 acquires the actual operation result values during the decarburization treatment. The actual operation result values to be acquired may be the same as those acquired in the treatment of Step S2 or may be controlled according to information required for the subsequent estimation of the carbon content in molten steel. For example, the measurement results of the flow rate and the component contents of the exhaust gas emitted from the vacuum degassing device 2 are not always necessary actual operation result values, unlike the estimation calculation of the carbon content in molten steel in the first half M1. Upon the completion of the treatment of Step S5, the treatment of determining the decarburization treatment end proceeds to the treatment of Step S6. The molten steel information and the actual operation result values are also collectively referred to as operation information.

[0046] In the treatment of Step S6, the carbon content calculation unit 41 estimates the carbon content in molten steel. In the treatment of Step S6, the requirement for the estimation model for the carbon content in molten steel used by the carbon content calculation unit 41 is that the carbon content in molten steel can be estimated at every predetermined period or successively. Thus, the carbon content estimation is performed by an estimation model different from the estimation model for the carbon content in molten steel used by the carbon content calculation unit 41 in the treatment of Step S4.

[0047] In this embodiment, the decarburization models of Equations (4) and (5) below are used in Step S6. w L d C L d t = Q C V − C L 4 w V dC V dt = Q C L − C V − ∑ i ak i C V − C E , i + C alloy 5

[0048] Herein, w is the molten steel mass [kg]. C is the carbon content in molten steel [ppm]. Q is the molten steel reflux rate [kg / s]. ak is the decarburization capacity coefficient [kg / s]. C E is the equilibrium value [ppm] of the carbon content in molten steel in the vacuum chamber 20. C alloy is the carbon weight [ppm] in terms of the carbon content in molten steel in a charged auxiliary raw material. The subscript L indicates that one attached by the subscript V indicates the physical quantity of the molten steel 6 in the ladle 5. The subscript V indicates that one attached by the subscript V indicates the physical quantity of the molten steel 6 in the vacuum chamber 20. For example, C V indicates the carbon content in molten steel [ppm] of the vacuum chamber 20. The subscript i is used to identify specific decarburization sites. Examples of the specific decarburization sites include a molten steel surface, bubbles of inert gas for reflux, and the like.

[0049] The carbon content in molten steel after an infinitesimal time is calculated by calculating the change amount of the carbon content in molten steel per infinitesimal time from Equations (4) and (5), and subtracting the change amount from the current carbon content in molten steel. This completes the treatment of Step S6, and the treatment of determining the decarburization treatment end proceeds to the treatment of Step S7.

[0050] In the treatment of Step S7, the end determination unit 42 determines whether the carbon content in molten steel determined in Step S6 (estimated value of the carbonic acid content in molten steel) has reached the predetermined target value (or has reached a value equal to or less than the target value). When the estimated value of the carbon content in molten steel is higher than the target value, the treatment returns to the treatment of Step S5, and the treatment in and after Step S6 is repeated using the newly input actual operation result values. On the other hand, when the estimated value of the carbon content in molten steel has reached a value equal to or less than the target value, it has been determined that the decarburization treatment has ended.

[0051] As described above, according to the method for determining a decarburization treatment end of this embodiment, the decarburization treatment is divided into the two stages of the first half M1 and the latter half M2. Then, the carbon contents in molten steel in the first half M1 and the latter half M2 are estimated using different estimation models for the carbon content in molten steel for the first half M1 and the latter half M2. In the first half M1, the estimation model constructed based on machine learning for the actual operation result data is used. In the latter half M2, the end of the decarburization treatment is determined when the estimated value of the carbon content in molten steel has reached the target value.

[0052] According to the method for determining a decarburization treatment end of this embodiment, the vacuum chamber-inside pressure, i.e., pressure in a reduced-pressure environment of the molten steel 6, may be used for the division reference between the first half M1 and the latter half M2. In the estimation model of the first half M1, the explanatory variable may be constructed by the operation information containing the molten steel information and the actual operation result values, and the object variable may be constructed by the carbon content in molten steel at the end of the first half of the decarburization treatment. The molten steel information is the information on the weight and the component contents of the molten steel before the treatment of the vacuum degassing treatment. The actual operation result values may contain the measurement results of the flow rate of exhaust gas emitted during the vacuum degassing treatment and the component contents in the exhaust gas.

[0053] The above-described configurations and steps minimize the adverse effects of the exhaust gas measurement value errors, and make it possible to reflect the actual results of the decarburization treatment in the estimation of the carbon content in molten steel. Therefore, the carbon content in molten steel can be estimated with high accuracy and the decarburization treatment can be ended at appropriate timing with respect to the carbon content standard. Further, the high-accuracy estimation is performed, so that it can be avoided that the decarburization treatment is performed for an excessively long period due to concern that the carbon content deviates from the standard, and, as a result, the decarburization treatment time can be shortened.<Modifications>

[0054] As described above, the present disclosure is described with reference to the specific embodiment, but it is not intended to limit the invention by the description. Not only the disclosed embodiment but the other embodiments of the present disclosure including various modifications will be apparent to those skilled in the art by reference to the description of the present disclosure. Therefore, the embodiment of the invention described in Claims should be construed to cover embodiments including modifications thereof described in this specification alone or in combination.

[0055] For example, in the above-described embodiment, the vacuum degassing device 2 is the RH vacuum degassing device, but the present disclosure is not limited to such an example. For example, the vacuum degassing device 2 may be one of another style, such as a DH type having one immersion tube 21 or a device bringing the surface of the molten steel 6 in the ladle 5 into a vacuum state not using the vacuum chamber 20.EXAMPLES

[0056] Hereinafter, the effects of the present disclosure are specifically described based on Examples. However, the present disclosure is not limited to the contents of Examples.

[0057] In this example, the decarburization treatment was performed with the vacuum degassing equipment 1 using the RH vacuum degassing device as the vacuum degassing device 2 to produce ultra-low carbon molten steel having the upper specification limit of the carbon content of 25 ppm in the same manner as in the above-described embodiment. A portion of the molten steel was collected as a sample before the start of the vacuum degassing treatment and after the end of the vacuum degassing treatment, and the carbon contents in molten steel of the samples were actually measured. Further, the flow rate of exhaust gas during the vacuum degassing treatment, the CO content in the exhaust gas, and a change with time in the CO 2 content change in exhaust gas were measured. The end of the decarburization treatment is determined by an operator.

[0058] When the actual result values of the flow rate of exhaust gas during the vacuum degassing treatment, the CO content in the exhaust gas, and the CO 2 content in the exhaust gas were measured, the decarburization rate and the cumulative decarburization amount during the vacuum degassing treatment can be determined by applying the same calculations as in Equations (2) and (3). On the other hand, the cumulative decarburization amount determined from the exhaust gas measurement value does not always match the actual decarburization amount result calculated from the actual measurement values of the carbon contents in molten steel before and after the vacuum degassing treatment. Therefore, the correction is performed by multiplying the decarburization rate by a constant such that the former matches the latter. By performing such a calculation, a certain value as the carbon content in molten steel at an arbitrary time during the vacuum degassing treatment can be determined from the actual measurement values of the carbon content in molten steel and the exhaust gas measurement values.

[0059] In this Example, the carbon content in molten steel at the first half end calculated by the above-described method is set as the actual result value. The first half M1 and the latter half M2 are divided with the vacuum chamber-inside pressure of 5 torr as the reference. In this example, the number of analyzed data used for analysis was set to 46.

[0060] Table 2 shows the correlation coefficient between the actual measurement value of the carbon content in molten steel before the start of the vacuum degassing treatment (carbon content before treatment) and the treatment time of the first half M1 (first half treatment time) and the actual measurement value of the carbon content in molten steel at the first half end (carbon content after first half). Table 2 shows that the actual measurement value of the carbon content before treatment and the first half treatment time have a high positive correlation. On the other hand, the actual measurement value of the carbon content before treatment and the actual result value of the carbon content after first half have a low correlation. Thus, it is found that determining the first half end based on the vacuum chamber-inside pressure removes the influence of the carbon content before treatment on the variation in the actual result value of the carbon content after first half. [Table 2]First half treatment timeActual result value of carbon content in molten steel at first half endActual measurement value of carbon content in molten steel before vacuum degassing treatment0.680.16

[0061] Table 3 shows standard deviations of the actual result value of the carbon content in molten steel at the point in time of the first half end when the decarburization treatment is divided into the first half and the last half by the inventive method and a comparative method as an inventive example and a comparative example, respectively. The number of analyzed data of each of the inventive example and the comparative example was set to 46. In the comparative example, the first half end was determined based on the vacuum decarburization treatment time, and a specific time was determined such that the average of the actual result values of the carbon content in molten steel matches that of a proposed method. Specifically, in the comparative example, a period from the start of the treatment to 452 s was set as the first half of the decarburization treatment. In the inventive example, it is found that the standard deviation of the actual result value of the carbon content in molten steel at the point in time of the first half end is smaller than that of the comparative example, and the dividing the decarburization treatment into the first half and the last half based on the vacuum chamber-inside pressure can reduce the variation in the actual result value of the carbon content in molten steel. [Table 3]Comparative ExampleInventive ExampleStandard deviation of [C] at the point in time of first half end15.96 ppm8.85 ppm

[0062] Table 4 shows the root mean square error (RMSE) of the estimation error of the carbon content in molten steel at the point in time of the first half end, i.e., at the point in time of the vacuum chamber-inside pressure of 5 torr. As the comparative example, the carbon content in molten steel was estimated by a decarburization model. As the inventive example, the carbon content in molten steel was estimated by a linear regression model (ridge regression model) and a neural network model. The number of analyzed data of each of the inventive example and the comparative example was set to 46. Table 4 shows that the carbon content in molten steel can be estimated with higher accuracy in the two models in the inventive example than in the comparative example. In the inventive example, it has been found that the estimation can be performed with higher accuracy in the neural network model than in the linear regression model. [Table 4]Comparative ExampleInventive ExampleDecarburization modelLinear regression modelNeural network modelCarton content estimation RMSE at first half end12.46 ppm7.89 ppm5.43 ppm

[0063] Table 5 shows, in the first row, the standard deviation of the estimation error of the carbon content in molten steel at the point in time of the decarburization treatment end. For the estimation of the carbon content in molten steel in the first half M1, the model, the result of which is shown in Table 4, was used. In the latter half M2, the carbon content in molten steel was estimated using the same decarburization model in all examples. The number of analyzed data of each of the inventive example and the comparative example was set to 46. Table 5 shows that the carbon content in molten steel can be estimated with higher accuracy in the two models in the inventive example than in the comparative example. On the other hand, for the linear regression model and the neural network model, by which the carbon content in molten steel was estimated as the inventive example, it has been found that a difference in the estimation accuracy is small. Thus, the fact that the estimation accuracy of the carbon content in molten steel at the point in time of the first half end is high does not always mean that the estimation accuracy at the point in time of the decarburization treatment end is high, and therefore the machine learning model is required to be carefully selected.

[0064] It is an object of the present disclosure to suppress excessive decarburization treatment in the vacuum degassing treatment by estimating the carbon content in molten steel with high accuracy and to shorten the treatment time. Table 5 shows, in the second row and the third row, an excessive decarburization suppression effect and a shortened decarburization treatment time when the carbon content in molten steel was estimated by the models of the inventive example in place of the model of the comparative example. When the decarburization treatment end is determined based on the estimated value of the carbon content in molten steel, the target value of the carbon content in molten steel to be referred to in the treatment of Step S7 is required to be set not to exceed the upper specification limit of the carbon content considering an error of the carbon content estimation model. Specifically, a value obtained by subtracting three times the error standard deviation of the carbon content estimation model from the upper specification limit of the carbon content may be set as the target value. When the target value is set as described above, the target value of the determination of the decarburization treatment end can be set to a value three times higher than a difference in the model error standard deviations when the models of the inventive example are used as compared with the comparative example as shown in the second row of Table 5. In this example, the effect of shortening the decarburization treatment time due to an increase in target value is as shown in the third row of Table 5. [Table 5]Comparative ExampleInventive ExampleDecarburization modelLinear regression modelNeural network modelCarton content estimation RMSE at first half end3.93 ppm3.32 ppm3.50 ppmExcessive decarburization suppression amount to Comparative Example1.83 ppm1.29 ppmShortened decarburization treatment time to Comparative Example1.7 min1.3 min

[0065] It should be noted that, although the embodiment of the present disclosure is described based on the drawings and examples, it is easy for persons skilled in the art to make various alternations or modifications based on the present disclosure. Therefore, it should be noted that these alternation or modifications are included in the scope of the present disclosure. For example, the functions included in the constituents or the steps can be rearranged in a logically consistent manner, and a plurality of constituents or steps can be combined into one or divided. The embodiment according to the present disclosure can also be realized as a storage medium storing a program to be executed by a processor provided in a device. It should be understood that the scope of the present disclosure also includes these above.Reference Signs List

[0066] 1: vacuum degassing equipment 2: vacuum degassing device 20: vacuum chamber 21: immersion tube 22: piping 23: injection lance 24: exhaust duct 25: exhaust gas flow rate meter 26: exhaust gas component content meter 3: control device 4: decarburization treatment end determination device (determination device) 40: operation information input unit (input unit) 41: carbon content calculation unit (calculation unit) 42: decarburization treatment end determination unit (end determination unit) 43: decarburization treatment phase determination unit (phase determination unit) 5: ladle 6: molten steel M1: first half decarburization treatment (first half) M2: latter half decarburization treatment (latter half)

Examples

examples

[0056]Hereinafter, the effects of the present disclosure are specifically described based on Examples. However, the present disclosure is not limited to the contents of Examples.

[0057]In this example, the decarburization treatment was performed with the vacuum degassing equipment 1 using the RH vacuum degassing device as the vacuum degassing device 2 to produce ultra-low carbon molten steel having the upper specification limit of the carbon content of 25 ppm in the same manner as in the above-described embodiment. A portion of the molten steel was collected as a sample before the start of the vacuum degassing treatment and after the end of the vacuum degassing treatment, and the carbon contents in molten steel of the samples were actually measured. Further, the flow rate of exhaust gas during the vacuum degassing treatment, the CO content in the exhaust gas, and a change with time in the CO 2 content change in exhaust gas were measured. The end of the decarburization treatment is d...

Technical Field

[0001] The present disclosure relates to a method for determining a decarburization treatment end, vacuum degassing equipment, a method for operating vacuum degassing equipment, and a method for producing molten steel.Background Art

[0002] In a steel making process, molten steel components are adjusted by removing impurities in hot metal, such as carbon, and adding useful alloying components. Particularly for carbon, decarburization is promoted by placing molten steel in a vacuum environment using vacuum degassing equipment, and ultra-low carbon steel can be produced in which the carbon content in molten steel is less than 10 ppm.

[0003] Herein, in vacuum degassing treatment, the carbon content in molten steel is not directly measured, but is only indirectly estimated from the contents of carbon monoxide and carbon dioxide in exhaust gas. In the actual production of the ultra-low carbon steel, the decarburization treatment tends to be performed for an excessively long period due to concern that the carbon content deviates from the standard.

[0004] To solve the prolongation of treatment time by the excessive decarburization treatment, it is effective to estimate the carbon content in molten steel during the treatment with high accuracy, and various methods have been proposed so far. Methods for estimating the carbon content in molten steel can be roughly divided into two methods of estimation methods using exhaust gas measurement values and methods for constructing a decarburization model.

[0005] In estimation methods using the flow rate and measurement values of exhaust gas, the decarburization amount is calculated from measurement values of exhaust gas (e.g., measurement values of the flow rate and the component contents) emitted from vacuum degassing equipment during treatment, and the carbon content in molten steel is estimated. In methods for constructing a decarburization model, the details of decarburization in vacuum degassing equipment are physically considered, and a decarburization model is constructed, thereby estimating the carbon content in molten steel (e.g., NPL 1).

[0006] Further, as a combination of the two methods, methods have been proposed in which the parameters of the decarburization model are determined from the exhaust gas measurement values, and the carbon content in molten steel is estimated using the decarburization model having the determined parameters (e.g., PTL 1 and PTL 2).Citation ListPatent Literatures

[0007] PTL 1: JP 2005-330512 A PTL 2: JP 2015-101742 A Non Patent Literature

[0008] NPL 1: Shinya Kitamura and three others, "Decarburization Model for Vacuum Degasser", Tetsu-to-Hagane, Vol. 80 (1994), No. 3, p. 213-218.Summary of InventionTechnical Problem

[0009] The flow rate and content measurement values of the exhaust gas emitted from the vacuum degassing equipment during the treatment contain information on the actual decarburization results, and the utilization of the information is considered to contribute to the realization of high-accuracy carbon content estimation. However, on the other hand, the exhaust gas measurement values contain errors, and the use of the exhaust gas measurement values for the determination of the model parameters as in the technology of PTL 1 and PTL 2 causes a decrease in carbon content estimation accuracy due to the exhaust gas measurement value errors. In particular, errors allowable in an estimated value of the carbon content in molten steel at the last stage of the decarburization treatment are about several ppm, and the influence of the exhaust gas measurement value errors can be fatal. Thus, to utilize the exhaust gas measurement values for the carbon content estimation, it is important to reduce the adverse effects of the exhaust gas measurement value errors.

[0010] It is an object of the present disclosure made in view of such circumstances to provide a method for determining a decarburization treatment end, vacuum degassing equipment, a method for operating vacuum degassing equipment, and a method for producing molten steel capable of estimating the carbon content in molten steel with high accuracy by properly utilizing the exhaust gas measurement values to end the decarburization treatment at appropriate timing.Solution to Problem

[0011] (1) One aspect of the present disclosure provides a method for determining a decarburization treatment end for determining an end of decarburization treatment when vacuum degassing treatment including the decarburization treatment is performed by placing molten steel in a reduced-pressure environment, the method including: dividing the decarburization treatment into two stages of first half decarburization treatment and latter half decarburization treatment, and estimating the carbon contents in molten steel in the first half decarburization treatment and the latter half decarburization treatment using different estimation models for the carbon content in molten steel for the first half decarburization treatment and the latter half decarburization treatment, in the first half decarburization treatment, using the estimation model constructed based on machine learning for actual operation result data, and in the latter half decarburization treatment, determining the end of the decarburization treatment when an estimated value of the carbon content in molten steel has reached the target value. (2) A method for determining decarburization treatment end is provided in which pressure in the reduced-pressure environment is used for a division reference for the first half decarburization treatment and the latter half decarburization treatment in the configuration according to (1) above. (3) The method for determining decarburization treatment end is provided in which, in the estimation model of the first half decarburization treatment, an explanatory variable is constructed by operation information containing molten steel information and actual operation result values and an object variable is constructed by the carbon content in molten steel at the end of the first half decarburization treatment, the molten steel information is information on the weight and the component content of the molten steel before the vacuum degassing treatment, and the actual operation result values contain the flow rate of exhaust gas emitted during the vacuum degassing treatment and a measurement result of the component content in the exhaust gas in the configuration according to (2) above. (4) One aspect of the present disclosure provides vacuum degassing equipment including: a vacuum degassing device configured to perform vacuum degassing treatment including decarburization treatment by placing molten steel in a reduced-pressure environment; a control device configured to control the vacuum degassing device; and a decarburization treatment end determination device configured to determine an end of the decarburization treatment, in which the decarburization treatment end determination device has a carbon content calculation unit configured to divide the decarburization treatment into two stages of first half decarburization treatment and latter half decarburization treatment, estimate the carbon contents in molten steel in the first half decarburization treatment and the latter half decarburization treatment using different estimation models for the carbon content in molten steel for the first half decarburization treatment and the latter half decarburization treatment, and use the estimation model constructed based on machine learning for actual operation result data in the first half decarburization treatment, and a decarburization treatment end determination unit configured to determine the end of the decarburization treatment when an estimated value of the carbon content in molten steel has reached the target value in the latter half decarburization treatment. (5) The vacuum degassing equipment is provided in which the carbon content calculation unit uses pressure in the reduced-pressure environment for a division reference for the first half decarburization treatment and the latter half decarburization treatment in the configuration of (4) above. (6) The vacuum degassing equipment is provided in which, in the estimation model of the first half decarburization treatment, an explanatory variable is constructed by operation information containing molten steel information and actual operation result values and an object variable is constructed by the carbon content in molten steel at the end of the first half decarburization treatment, the molten steel information is information on the weight and the component content of the molten steel before the treatment of the vacuum degassing treatment, and the actual operation result values contain the flow rate of exhaust gas emitted during the vacuum degassing treatment and a measurement result of the component content in the exhaust gas in the configuration of (5) above. (7) One aspect of the present disclosure provides a method for operating vacuum degassing equipment performing vacuum degassing treatment including decarburization treatment by placing molten steel in a reduced-pressure environment, the method including: using the decarburization treatment determination method according to any one of the configurations (1) to (3) above when an end of the decarburization treatment is determined. (8) One aspect of the present disclosure provides a method for producing molten steel for producing the molten steel refined by applying vacuum degassing treatment to the molten steel in vacuum degassing equipment operated by the operation method according to the configuration of (7) above. Advantageous Effects of Invention

[0012] One aspect of the present disclosure can provide a method for determining a decarburization treatment end, vacuum degassing equipment, a method for operating vacuum degassing equipment, and a method for producing molten steel capable of estimating the carbon content in molten steel with high accuracy by properly utilizing the exhaust gas measurement values to end the decarburization treatment at appropriate timing.Brief Description of Drawings

[0013] FIG. 1 is a structural view of vacuum degassing equipment and a block diagram of a determination device according to one embodiment of the present disclosure; FIG. 2 is a flowchart illustrating a method for determining a decarburization treatment end according to one embodiment of the present disclosure; and FIG. 3 is a graph showing an example of a change with time in vacuum chamber-inside pressure in vacuum degassing treatment charge in which ultra-low carbon steel grade production was performed. Description of Embodiments

[0014] A detailed description below describes an embodiment of the present disclosure with reference to the drawings. In the description of the drawings, the same or similar reference numerals are attached to the same or similar parts, and duplicated descriptions are omitted. The drawings are schematic and are sometimes different from the actual ones. The embodiment described below exemplifies devices and methods for embodying the technical idea of the present disclosure. The technical idea of the present disclosure does not specify materials, structures, arrangement, and the like of constituent components to the materials, structures, arrangement, and the like described below. The technical idea of the present disclosure can be variously altered within the technical range defined by Claims.<Vacuum degassing device>

[0015] Hereinafter, vacuum degassing equipment in one embodiment of the present disclosure is described with reference to the drawings. As illustrated in FIG. 1, vacuum degassing equipment 1 includes a vacuum degassing device 2, a control device 3, and a decarburization treatment end determination device (hereinafter also simply referred to as a "determination device") 4.

[0016] The vacuum degassing device 2 is an RH vacuum degassing device, and applies vacuum degassing treatment to molten steel 6 contained in a ladle 5. The vacuum degassing device 2 includes a vacuum chamber 20, a pair of immersion tubes 21, piping 22, an injection lance 23, an exhaust duct 24, an exhaust gas flow rate meter 25, and an exhaust gas component content meter 26.

[0017] In the vacuum degassing device 2, the pair of immersion tubes 21 provided at the bottom of the vacuum chamber 20 is immersed in molten steel 6 in the ladle 5, and gas inside the vacuum chamber 20 is exhausted through the exhaust duct 24, so that the pressure in the vacuum chamber 20 is reduced and the molten steel 6 is sucked up. Then, an inert gas is injected from one of the immersion tubes 21 through the piping 22, so that the molten steel 6 is refluxed between the vacuum chamber 20 and the ladle 5. In the vacuum degassing device 2, oxygen can be supplied to the molten steel 6 in the vacuum chamber 20 by injecting oxygen from the injection lance 23 installed in the vacuum chamber 20. The vacuum chamber 20 is one example of an evacuated region of the vacuum degassing device 2, i.e., a region that is pressure-reduced to be vacuum. In such a vacuum degassing device 2, a degassing reaction and a decarburization of the molten steel 6 are promoted by refluxing the molten steel 6 containing oxygen in the evacuated region. In this embodiment, the vacuum degassing treatment by the vacuum degassing equipment 1 is refining treatment including decarburization treatment (vacuum decarburization treatment) of promoting decarburization by placing the molten steel 6 in a reduced-pressure environment.

[0018] The exhaust gas flow rate meter 25 and the exhaust gas component content meter 26 are provided inside the exhaust duct 24. The exhaust gas flow rate meter 25 measures the flow rate of exhaust gas. The exhaust gas component content meter 26 measures the contents of components in exhaust gas containing a CO gas, a CO 2 gas, and an O 2 gas.

[0019] The control device 3 contains an information processer, such as a computer. The control device 3 controls the operation amounts related to the operation including the exhaust amount of exhaust equipment, the flow rate of an inert gas for reflux, and the injected oxygen flow rate such that the component contents and the temperature of the molten steel 6 fall within the target range after the vacuum degassing treatment from the actual result values before the vacuum degassing treatment. The control device 3 also collects actual operation result value data including the pressure in the vacuum chamber 20 (vacuum chamber-inside pressure), the flow rate of the inert gas for reflux, the injected oxygen flow rate, the exhaust gas flow rate, and the exhaust gas component contents and outputs the actual operation result value data to the determination device 4.

[0020] The determination device 4 is a device performing decarburization treatment determination of determining the end timing of the decarburization treatment in the vacuum degassing treatment and contains an information processer, such as a computer. The determination device 4 functions as an operation information input unit (hereinafter also simply referred to as an "input unit") 40, a carbon content calculation unit (hereinafter also simply referred to as a "calculation unit") 41, a decarburization treatment end determination unit (hereinafter also simply referred to as an "end determination unit") 42, and a decarburization treatment phase determination unit (hereinafter also simply referred to as a "phase determination unit") 43 by execution of a computer program by a processing unit, such as a central processing unit (CPU), in the information processer. The control device 3 and the determination device 4 are also collectively referred to as a vacuum degassing treatment control system.

[0021] The input unit 40 inputs the actual operation results related to the operation amounts and the time-series exhaust gas measurement values during the vacuum degassing treatment as input information. The time-series exhaust gas measurement values include the flow rate of the exhaust gas emitted from the vacuum degassing device 2 and the component contents of a CO gas, a CO 2 gas, and an O 2 gas contained in the exhaust gas.

[0022] The calculation unit 41 estimates the carbon content in molten steel 6 (carbon content in molten steel) based on the input information acquired by the input unit 40. In a method for calculating the carbon content in molten steel, a plurality of estimation models for the carbon content in molten steel are properly used based on the determination reference including the vacuum chamber-inside pressure.

[0023] The determination unit 42 determines the decarburization treatment end based on the estimated value of the carbon content in molten steel calculated by the calculation unit 41. The determination unit 42 outputs the determination result of the decarburization treatment end to the control device 3. The control device 3 may control the operation amounts related to the operation based on the decarburization treatment end determination result obtained from the decarburization treatment end determination unit 42.

[0024] The phase determination unit 43 determines the phase of the decarburization treatment based on the input information acquired by the input unit 40. Specifically, the phase determination unit 43 determines whether the phase of the decarburization treatment is the first half or the latter half by determining the end of the first half of the decarburization treatment described later.

[0025] In the vacuum degassing equipment 1, the component adjustment including the decarburization treatment and the temperature adjustment of the molten steel 6 are performed by applying the vacuum degassing treatment to the molten steel 6, so that the molten steel 6 having the specified components and temperature is produced. Then, in the decarburization treatment, the internal state and the like of the vacuum degassing device 2 are estimated by the determination device 4, so that the timing of the end of the decarburization treatment is determined. Then, the vacuum degassing treatment is performed based on the determination result of the decarburization treatment. For steps of the vacuum degassing treatment other than the determination of the end of the decarburization treatment end, known one can be used.<Method for determining end of decarburization treatment>

[0026] A method for determining an end of the decarburization treatment according to this embodiment is performed according to the flowchart illustrated in FIG. 2. The determination of the end of the decarburization treatment according to this embodiment roughly contains estimation calculation of the carbon content in molten steel in first half decarburization treatment (hereinafter also simply referred to as a "first half") M1 and estimation calculation of the carbon content in molten steel and the determination of the end of the decarburization treatment in latter half decarburization treatment (hereinafter also simply referred to as a "latter half") M2. The switch from the first half M1 to the latter half M2 of the decarburization treatment is performed based on a determination reference including the pressure in the vacuum chamber 20. The flowchart illustrated in FIG. 2 starts at timing when execution instruction of the decarburization treatment is input, and the treatment proceeds to the treatment of Step S1.

[0027] First, in the treatment of Step S1, the input unit 40 acquires molten steel information before the start of the decarburization treatment from a host computer or the like which is not illustrated. The molten steel information may contain, for example, the weight of the molten steel 6 and the component contents which are the measurement and analysis results obtained by component analysis of the molten steel 6. This completes the treatment of Step S1, and the treatment of determining the decarburization treatment end proceeds to the treatment of Step S2.

[0028] In the treatment of Step S2, the input unit 40 acquires the actual operation result values during the decarburization treatment. The actual operation result values are items required for the calculation in the calculation unit 41. In this embodiment, the input unit 40 acquires at least the measurement results of the flow rate and the component contents of the exhaust gas emitted from the vacuum degassing device 2 during the vacuum degassing treatment as the actual operation result values. Further, in this embodiment, the input unit 40 acquires the pressure in the vacuum chamber 20 during the decarburization treatment (pressure in a reduced-pressure environment in the vacuum chamber 20) as the actual operation result value. Further, the actual operation result values may contain information on the flow rate of the inert gas for reflux and the flow rate of oxygen from the injection lance 23 during the decarburization treatment and information on the auxiliary raw materials charged during the decarburization treatment. Specific examples of the information on auxiliary raw materials include the types and the charging amounts of the auxiliary raw materials. This completes the treatment of Step S2, and the treatment of determining the decarburization treatment end proceeds to the treatment of Step S3. Herein, Step S1 and Step S2 correspond to the input steps. The molten steel information and the actual operation result values are also collectively referred to as operation information.

[0029] In the treatment of Step S3, the phase determination unit 43 determines the end of the first half of the decarburization treatment based on the actual operation result values acquired in the treatment of Step S2. The reference for the end determination includes at least the fact whether the pressure of the vacuum chamber 20 has reached a predetermined threshold (or has reached a value equal to or less than a predetermined threshold). When required in view of a condition desirable as the boundary between the first half and the latter half of the decarburization treatment described later, a condition may be added that is required to be satisfied to determine the end of the first half M1 (first half end), in addition to the condition that the pressure of the vacuum chamber 20 is equal to or less than the threshold. When it has been determined that the first half M1 has not ended, the treatment returns to the treatment of Step S2 and repeat the treatment in and after Step S2 using newly input actual operation result values. On the other hand, when it has been determined that the first half M1 has ended, the treatment of Step S3 is completed and the treatment of determining the decarburization treatment end proceeds to the treatment of Step S4.

[0030] In the treatment of Step S4, the calculation unit 41 estimates the carbon content in molten steel at the point in time of the first half M1 end (at the point in time of the first half end). The smaller a variation in the actual result value of an object variable for an estimation model in general without being limited a machine learning model, the easier it will be to construct a high-accuracy estimation model. In the carbon content estimation in the vacuum decarburization treatment, factors causing a variation in carbon content in molten steel at the point in time of the first half end which is the object variable include the carbon content in molten steel before the vacuum decarburization treatment (carbon content before treatment). The variation factors of the carbon content in molten steel also include an airtight condition change of the vacuum chamber 20 and the exhaust system due to a change in connection between the vacuum chamber 20 and the exhaust system and a change in time required for the air pressure in the vacuum chamber 20 to decrease to the ultimate air pressure due to the oxygen injection during the vacuum degassing treatment. The pressure in the vacuum chamber 20 is referred to as a vacuum chamber-inside pressure.

[0031] Herein, the present inventors have obtained finding that determining the first half end based on the vacuum chamber-inside pressure reduces the variation in the actual result value of the carbon content in molten steel at the point in time of the first half end. Among the variation factors of the carbon content in molten steel described above, the following consideration of the carbon content before treatment shows the validity of this finding. The vacuum chamber-inside pressure is determined by the balance between the exhaust rate from the vacuum chamber 20 and the supply rate of gas to the vacuum chamber 20. The former can be regarded to be an approximately constant value according to the exhaust capacity of the vacuum degassing device 2. On the other hand, for the latter, the higher the decarburization rate, i.e., the supply rate of a CO gas to the vacuum chamber 20, the higher the carbon content in molten steel. Thus, the carbon content in molten steel at a predetermined vacuum chamber-inside pressure is considered to have an approximately constant value.

[0032] The vacuum chamber-inside pressure serving as the reference for the first half end is preferably determined considering the following two points. First, as the first point, it is desirable to select pressure having uniqueness of time in the vacuum decarburization treatment. In the vacuum degassing treatment, the vacuum chamber-inside pressure does not monotonously decrease but sometimes temporarily increases when the operation mode of the exhaust equipment is switched, when the decarburization is active, when oxygen is injected, or the like. The first half end is required to be properly determined in the middle of the decarburization treatment. The presence of a plurality of points in time where the same pressure is taken leads to a variation in carbon content in molten steel at the point in time of the first half end. In the foregoing description, the adverse effects where there is a plurality of points in time where the same pressure is taken are described; on the other hand, when the vacuum chamber-inside pressure serving as the reference is excessively low, the vacuum chamber-inside pressure does not decrease to the reference value, and normal determination of the end of the decarburization treatment cannot be performed.

[0033] Next, as the second point, the vacuum chamber-inside pressure serving as the reference is desirably lower to the extent that the variation in the actual result value of the carbon content in molten steel at the point in time of the first half end is not excessively large. For example, setting an extremely high value, such as 300 Torr, to the reference pressure based on the first point described above hardly has an effect on an increase in accuracy of the carbon content estimation. As described later, the exhaust gas measurement value is used to estimate the carbon content in molten steel in the first half M1, but the pressure serving as the reference is desirably lower to maximize the advantage of the exhaust gas measurement values containing information on the actual decarburization result.

[0034] When there is no vacuum chamber-inside pressure satisfying the above-described two points and the timing of the first half end cannot be properly determined with only the pressure serving as the reference, other conditions may be added. For example, the presence or absence of oxygen injection, the treatment time, and the like are considered to be additional conditions.

[0035] FIG. 3 illustrates an example of a change with time in the vacuum chamber-inside pressure in a vacuum degassing treatment charge in which ultra-low charcoal steel grade production was performed. From FIG. 3, when the pressure serving as the reference for the first half end of the decarburization treatment is selected, it is desirable to select the pressure in the range of 3 Torr or more and 6 Torr or less or 60 Torr or more, where the time is uniquely determined from the viewpoint of the above-described first point. In consideration of the above-described second point, it is more preferable to select the pressure serving as the reference in a lower value range of 3 Torr or more and 6 Torr or less. For example, the pressure serving as the reference may be set to 5 Torr. When the pressure serving as the reference for the first half end of the decarburization treatment is selected in a single point, it is desirable to consider the variation in the change with time in the vacuum chamber-inside pressure per charge and the estimation accuracy of the carbon content in molten steel at the point in time of the decarburization treatment end.

[0036] As is clear from FIG. 3, pressure suitable as the reference for the first half end of the decarburization treatment is not always a distinct inflection point with respect to time. In the vacuum degassing treatment of performing the ultra-low charcoal steel grade production, the decarburization rate is finally rate-controlled by the carbon supply. However, high-accuracy estimation of the carbon content in molten steel is possible even when the boundary between the first half and the latter half of the decarburization treatment is set in the middle of the stage of rate-controlling the carbon supply rate.

[0037] In this embodiment, the carbon content in molten steel at the point in time of the first half end is estimated by a machine learning model with the operation information acquired by the input unit 40 as an explanatory variable. For the form of the machine learning model, one capable of estimating the carbon content in molten steel at the point in time of the first half end with high accuracy is selected. Examples of candidates thereof include a linear regression model, a neural network model, a support vector regression model, and a decision tree model. In this embodiment, a linear regression model is used. At this time, the machine learning model is represented by Equation (1) below. [Math. 1] C 0 = ∑ j = 1 N α j X j + α 0

[0038] Herein, C 0 is the estimated value of the carbon content in molten steel at the point in time of the first half end, which is the object variable. X j (j = 1, 2,···, N) is the explanatory variable. a j is a parameter corresponding to the explanatory variable X j . a 0 is the constant term of the machine learning model. The parameter α j and the constant term α 0 can be determined by preparing a large number of data sets of previously collected object variables and explanatory variables (actual operation result data) and applying a known method.

[0039] The explanatory variable X j of the machine learning model includes, for example, the information on the weight and the component contents of the molten steel before the treatment and a plurality of actual operation result values including the measurement results of the flow rate of exhaust gas emitted from the vacuum degassing equipment during the treatment and the component contents in the exhaust gas as shown in Table 1 below. [Table 1]Explanatory variableUnitX 1 Carbon content in molten steel before decarburization treatmentppmX 2 Molten steel weightton (metric ton)X 3 Integrated decarburization amount based on exhaust gas information of first half of decarburization treatmentkgX 4 Integrated injected oxygen amount of first half of decarburization treatmentNm 3< X 5 Treatment time of first half of decarburization treatmentsX 6 Exhaust gas flow rate at the point in time of first half end of decarburization treatmentNm 3< / hX 7 Exhaust gas CO content at the point in time of first half end of decarburization treatmentvol%X 8 Exhaust gas CO 2 content at the point in time of first half end of decarburization treatmentvol%X 9 Exhaust gas O 2 content at the point in time of first half end of decarburization treatmentvol%

[0040] Herein, the integrated decarburization amount based on the exhaust gas information of the first half M1 contained as the explanatory variable in Table 1 is calculated by a calculation process described later.

[0041] In consideration of the fact that carbon emitted from the molten steel takes the form of CO or CO 2 , the carbon amount in the exhaust gas per unit time is given by Equation (2) below. The cumulative emitted carbon amount from the start of the treatment (at time 0) to the first half end (at time t 0 ) is given by Equation (3) below. q C , OG t = m C V off t 22.4 ⋅ r CO t + r CO 2 t 100 2 Q C , OG t 0 = ∫ 0 t 0 q C , OG t d t 3

[0042] Herein, q C,OG (t) is the carbon amount in the exhaust gas per unit time [kg / s] at time t. m C is the molar mass of carbon [g / mol]. V off (t) is the volume flow rate [Nm 3< / s] of the exhaust gas at time t. r CO (t) is the CO content [vol%] in the exhaust gas at time t. r CO2 (t) is the CO 2 content in the exhaust gas at time t [vol%]. Q C,OG (t 0 ) is the cumulative emitted carbon amount [kg] from time 0 to t 0 .

[0043] As described above, the form of the machine learning model is not limited to the linear regression model described in this embodiment. Other model forms, such as a neural network model, may be adopted when a higher effect can be confirmed by the enhancement of the estimation accuracy of the carbon content in molten steel. However, the machine learning model is carefully selected in the following points. It is an ultimate object of the present disclosure to reduce the vacuum degassing treatment time. To realize this, directly, the estimation accuracy of the carbon content in molten steel at the point in time of the decarburization treatment end is required to be high. When the estimation accuracy of the carbon content in molten steel at the point in time of the first half end is high, the estimation accuracy of the carbon content in molten steel at the point in time of the decarburization treatment end can be estimated to be high, but there is no direct relation between the two. Thus, the model is required to be carefully selected such that the estimation accuracy of not the former but the latter is enhanced.

[0044] By applying Equation (1), the carbon content in molten steel at the point in time of the first half end can be estimated. This completes the treatment of Step S4, and the treatment proceeds to the treatment of Step S5 of the latter half M2.

[0045] In the treatment of Step S5, the operation information input unit 40 acquires the actual operation result values during the decarburization treatment. The actual operation result values to be acquired may be the same as those acquired in the treatment of Step S2 or may be controlled according to information required for the subsequent estimation of the carbon content in molten steel. For example, the measurement results of the flow rate and the component contents of the exhaust gas emitted from the vacuum degassing device 2 are not always necessary actual operation result values, unlike the estimation calculation of the carbon content in molten steel in the first half M1. Upon the completion of the treatment of Step S5, the treatment of determining the decarburization treatment end proceeds to the treatment of Step S6. The molten steel information and the actual operation result values are also collectively referred to as operation information.

[0046] In the treatment of Step S6, the carbon content calculation unit 41 estimates the carbon content in molten steel. In the treatment of Step S6, the requirement for the estimation model for the carbon content in molten steel used by the carbon content calculation unit 41 is that the carbon content in molten steel can be estimated at every predetermined period or successively. Thus, the carbon content estimation is performed by an estimation model different from the estimation model for the carbon content in molten steel used by the carbon content calculation unit 41 in the treatment of Step S4.

[0047] In this embodiment, the decarburization models of Equations (4) and (5) below are used in Step S6. w L d C L d t = Q C V − C L 4 w V dC V dt = Q C L − C V − ∑ i ak i C V − C E , i + C alloy 5

[0048] Herein, w is the molten steel mass [kg]. C is the carbon content in molten steel [ppm]. Q is the molten steel reflux rate [kg / s]. ak is the decarburization capacity coefficient [kg / s]. C E is the equilibrium value [ppm] of the carbon content in molten steel in the vacuum chamber 20. C alloy is the carbon weight [ppm] in terms of the carbon content in molten steel in a charged auxiliary raw material. The subscript L indicates that one attached by the subscript V indicates the physical quantity of the molten steel 6 in the ladle 5. The subscript V indicates that one attached by the subscript V indicates the physical quantity of the molten steel 6 in the vacuum chamber 20. For example, C V indicates the carbon content in molten steel [ppm] of the vacuum chamber 20. The subscript i is used to identify specific decarburization sites. Examples of the specific decarburization sites include a molten steel surface, bubbles of inert gas for reflux, and the like.

[0049] The carbon content in molten steel after an infinitesimal time is calculated by calculating the change amount of the carbon content in molten steel per infinitesimal time from Equations (4) and (5), and subtracting the change amount from the current carbon content in molten steel. This completes the treatment of Step S6, and the treatment of determining the decarburization treatment end proceeds to the treatment of Step S7.

[0050] In the treatment of Step S7, the end determination unit 42 determines whether the carbon content in molten steel determined in Step S6 (estimated value of the carbonic acid content in molten steel) has reached the predetermined target value (or has reached a value equal to or less than the target value). When the estimated value of the carbon content in molten steel is higher than the target value, the treatment returns to the treatment of Step S5, and the treatment in and after Step S6 is repeated using the newly input actual operation result values. On the other hand, when the estimated value of the carbon content in molten steel has reached a value equal to or less than the target value, it has been determined that the decarburization treatment has ended.

[0051] As described above, according to the method for determining a decarburization treatment end of this embodiment, the decarburization treatment is divided into the two stages of the first half M1 and the latter half M2. Then, the carbon contents in molten steel in the first half M1 and the latter half M2 are estimated using different estimation models for the carbon content in molten steel for the first half M1 and the latter half M2. In the first half M1, the estimation model constructed based on machine learning for the actual operation result data is used. In the latter half M2, the end of the decarburization treatment is determined when the estimated value of the carbon content in molten steel has reached the target value.

[0052] According to the method for determining a decarburization treatment end of this embodiment, the vacuum chamber-inside pressure, i.e., pressure in a reduced-pressure environment of the molten steel 6, may be used for the division reference between the first half M1 and the latter half M2. In the estimation model of the first half M1, the explanatory variable may be constructed by the operation information containing the molten steel information and the actual operation result values, and the object variable may be constructed by the carbon content in molten steel at the end of the first half of the decarburization treatment. The molten steel information is the information on the weight and the component contents of the molten steel before the treatment of the vacuum degassing treatment. The actual operation result values may contain the measurement results of the flow rate of exhaust gas emitted during the vacuum degassing treatment and the component contents in the exhaust gas.

[0053] The above-described configurations and steps minimize the adverse effects of the exhaust gas measurement value errors, and make it possible to reflect the actual results of the decarburization treatment in the estimation of the carbon content in molten steel. Therefore, the carbon content in molten steel can be estimated with high accuracy and the decarburization treatment can be ended at appropriate timing with respect to the carbon content standard. Further, the high-accuracy estimation is performed, so that it can be avoided that the decarburization treatment is performed for an excessively long period due to concern that the carbon content deviates from the standard, and, as a result, the decarburization treatment time can be shortened.<Modifications>

[0054] As described above, the present disclosure is described with reference to the specific embodiment, but it is not intended to limit the invention by the description. Not only the disclosed embodiment but the other embodiments of the present disclosure including various modifications will be apparent to those skilled in the art by reference to the description of the present disclosure. Therefore, the embodiment of the invention described in Claims should be construed to cover embodiments including modifications thereof described in this specification alone or in combination.

[0055] For example, in the above-described embodiment, the vacuum degassing device 2 is the RH vacuum degassing device, but the present disclosure is not limited to such an example. For example, the vacuum degassing device 2 may be one of another style, such as a DH type having one immersion tube 21 or a device bringing the surface of the molten steel 6 in the ladle 5 into a vacuum state not using the vacuum chamber 20.EXAMPLES

[0056] Hereinafter, the effects of the present disclosure are specifically described based on Examples. However, the present disclosure is not limited to the contents of Examples.

[0057] In this example, the decarburization treatment was performed with the vacuum degassing equipment 1 using the RH vacuum degassing device as the vacuum degassing device 2 to produce ultra-low carbon molten steel having the upper specification limit of the carbon content of 25 ppm in the same manner as in the above-described embodiment. A portion of the molten steel was collected as a sample before the start of the vacuum degassing treatment and after the end of the vacuum degassing treatment, and the carbon contents in molten steel of the samples were actually measured. Further, the flow rate of exhaust gas during the vacuum degassing treatment, the CO content in the exhaust gas, and a change with time in the CO 2 content change in exhaust gas were measured. The end of the decarburization treatment is determined by an operator.

[0058] When the actual result values of the flow rate of exhaust gas during the vacuum degassing treatment, the CO content in the exhaust gas, and the CO 2 content in the exhaust gas were measured, the decarburization rate and the cumulative decarburization amount during the vacuum degassing treatment can be determined by applying the same calculations as in Equations (2) and (3). On the other hand, the cumulative decarburization amount determined from the exhaust gas measurement value does not always match the actual decarburization amount result calculated from the actual measurement values of the carbon contents in molten steel before and after the vacuum degassing treatment. Therefore, the correction is performed by multiplying the decarburization rate by a constant such that the former matches the latter. By performing such a calculation, a certain value as the carbon content in molten steel at an arbitrary time during the vacuum degassing treatment can be determined from the actual measurement values of the carbon content in molten steel and the exhaust gas measurement values.

[0059] In this Example, the carbon content in molten steel at the first half end calculated by the above-described method is set as the actual result value. The first half M1 and the latter half M2 are divided with the vacuum chamber-inside pressure of 5 torr as the reference. In this example, the number of analyzed data used for analysis was set to 46.

[0060] Table 2 shows the correlation coefficient between the actual measurement value of the carbon content in molten steel before the start of the vacuum degassing treatment (carbon content before treatment) and the treatment time of the first half M1 (first half treatment time) and the actual measurement value of the carbon content in molten steel at the first half end (carbon content after first half). Table 2 shows that the actual measurement value of the carbon content before treatment and the first half treatment time have a high positive correlation. On the other hand, the actual measurement value of the carbon content before treatment and the actual result value of the carbon content after first half have a low correlation. Thus, it is found that determining the first half end based on the vacuum chamber-inside pressure removes the influence of the carbon content before treatment on the variation in the actual result value of the carbon content after first half. [Table 2]First half treatment timeActual result value of carbon content in molten steel at first half endActual measurement value of carbon content in molten steel before vacuum degassing treatment0.680.16

[0061] Table 3 shows standard deviations of the actual result value of the carbon content in molten steel at the point in time of the first half end when the decarburization treatment is divided into the first half and the last half by the inventive method and a comparative method as an inventive example and a comparative example, respectively. The number of analyzed data of each of the inventive example and the comparative example was set to 46. In the comparative example, the first half end was determined based on the vacuum decarburization treatment time, and a specific time was determined such that the average of the actual result values of the carbon content in molten steel matches that of a proposed method. Specifically, in the comparative example, a period from the start of the treatment to 452 s was set as the first half of the decarburization treatment. In the inventive example, it is found that the standard deviation of the actual result value of the carbon content in molten steel at the point in time of the first half end is smaller than that of the comparative example, and the dividing the decarburization treatment into the first half and the last half based on the vacuum chamber-inside pressure can reduce the variation in the actual result value of the carbon content in molten steel. [Table 3]Comparative ExampleInventive ExampleStandard deviation of [C] at the point in time of first half end15.96 ppm8.85 ppm

[0062] Table 4 shows the root mean square error (RMSE) of the estimation error of the carbon content in molten steel at the point in time of the first half end, i.e., at the point in time of the vacuum chamber-inside pressure of 5 torr. As the comparative example, the carbon content in molten steel was estimated by a decarburization model. As the inventive example, the carbon content in molten steel was estimated by a linear regression model (ridge regression model) and a neural network model. The number of analyzed data of each of the inventive example and the comparative example was set to 46. Table 4 shows that the carbon content in molten steel can be estimated with higher accuracy in the two models in the inventive example than in the comparative example. In the inventive example, it has been found that the estimation can be performed with higher accuracy in the neural network model than in the linear regression model. [Table 4]Comparative ExampleInventive ExampleDecarburization modelLinear regression modelNeural network modelCarton content estimation RMSE at first half end12.46 ppm7.89 ppm5.43 ppm

[0063] Table 5 shows, in the first row, the standard deviation of the estimation error of the carbon content in molten steel at the point in time of the decarburization treatment end. For the estimation of the carbon content in molten steel in the first half M1, the model, the result of which is shown in Table 4, was used. In the latter half M2, the carbon content in molten steel was estimated using the same decarburization model in all examples. The number of analyzed data of each of the inventive example and the comparative example was set to 46. Table 5 shows that the carbon content in molten steel can be estimated with higher accuracy in the two models in the inventive example than in the comparative example. On the other hand, for the linear regression model and the neural network model, by which the carbon content in molten steel was estimated as the inventive example, it has been found that a difference in the estimation accuracy is small. Thus, the fact that the estimation accuracy of the carbon content in molten steel at the point in time of the first half end is high does not always mean that the estimation accuracy at the point in time of the decarburization treatment end is high, and therefore the machine learning model is required to be carefully selected.

[0064] It is an object of the present disclosure to suppress excessive decarburization treatment in the vacuum degassing treatment by estimating the carbon content in molten steel with high accuracy and to shorten the treatment time. Table 5 shows, in the second row and the third row, an excessive decarburization suppression effect and a shortened decarburization treatment time when the carbon content in molten steel was estimated by the models of the inventive example in place of the model of the comparative example. When the decarburization treatment end is determined based on the estimated value of the carbon content in molten steel, the target value of the carbon content in molten steel to be referred to in the treatment of Step S7 is required to be set not to exceed the upper specification limit of the carbon content considering an error of the carbon content estimation model. Specifically, a value obtained by subtracting three times the error standard deviation of the carbon content estimation model from the upper specification limit of the carbon content may be set as the target value. When the target value is set as described above, the target value of the determination of the decarburization treatment end can be set to a value three times higher than a difference in the model error standard deviations when the models of the inventive example are used as compared with the comparative example as shown in the second row of Table 5. In this example, the effect of shortening the decarburization treatment time due to an increase in target value is as shown in the third row of Table 5. [Table 5]Comparative ExampleInventive ExampleDecarburization modelLinear regression modelNeural network modelCarton content estimation RMSE at first half end3.93 ppm3.32 ppm3.50 ppmExcessive decarburization suppression amount to Comparative Example1.83 ppm1.29 ppmShortened decarburization treatment time to Comparative Example1.7 min1.3 min

[0065] It should be noted that, although the embodiment of the present disclosure is described based on the drawings and examples, it is easy for persons skilled in the art to make various alternations or modifications based on the present disclosure. Therefore, it should be noted that these alternation or modifications are included in the scope of the present disclosure. For example, the functions included in the constituents or the steps can be rearranged in a logically consistent manner, and a plurality of constituents or steps can be combined into one or divided. The embodiment according to the present disclosure can also be realized as a storage medium storing a program to be executed by a processor provided in a device. It should be understood that the scope of the present disclosure also includes these above.Reference Signs List

[0066] 1: vacuum degassing equipment 2: vacuum degassing device 20: vacuum chamber 21: immersion tube 22: piping 23: injection lance 24: exhaust duct 25: exhaust gas flow rate meter 26: exhaust gas component content meter 3: control device 4: decarburization treatment end determination device (determination device) 40: operation information input unit (input unit) 41: carbon content calculation unit (calculation unit) 42: decarburization treatment end determination unit (end determination unit) 43: decarburization treatment phase determination unit (phase determination unit) 5: ladle 6: molten steel M1: first half decarburization treatment (first half) M2: latter half decarburization treatment (latter half)

Examples

examples

[0056]Hereinafter, the effects of the present disclosure are specifically described based on Examples. However, the present disclosure is not limited to the contents of Examples.

[0057]In this example, the decarburization treatment was performed with the vacuum degassing equipment 1 using the RH vacuum degassing device as the vacuum degassing device 2 to produce ultra-low carbon molten steel having the upper specification limit of the carbon content of 25 ppm in the same manner as in the above-described embodiment. A portion of the molten steel was collected as a sample before the start of the vacuum degassing treatment and after the end of the vacuum degassing treatment, and the carbon contents in molten steel of the samples were actually measured. Further, the flow rate of exhaust gas during the vacuum degassing treatment, the CO content in the exhaust gas, and a change with time in the CO 2 content change in exhaust gas were measured. The end of the decarburization treatment is d...

Claims

1. A method for determining a decarburization treatment end for determining an end of decarburization treatment when vacuum degassing treatment including the decarburization treatment is performed by placing molten steel in a reduced-pressure environment, the method comprising: dividing the decarburization treatment into two stages of first half decarburization treatment and latter half decarburization treatment, and estimating carbon contents in molten steel in the first half decarburization treatment and the latter half decarburization treatment using different estimation models for the carbon content in molten steel for the first half decarburization treatment and the latter half decarburization treatment, in the first half decarburization treatment, using the estimation model constructed based on machine learning for actual operation result data, and in the latter half decarburization treatment, determining an end of the decarburization treatment when an estimated value of the carbon content in molten steel has reached a target value.

2. The method for determining a decarburization treatment end according to claim 1, wherein pressure in the reduced-pressure environment is used for a division reference for the first half decarburization treatment and the latter half decarburization treatment.

3. The method for determining a decarburization treatment end according to claim 2, wherein in the estimation model of the first half decarburization treatment, an explanatory variable is constructed by operation information containing molten steel information and actual operation result values and an object variable is constructed by the carbon content in molten steel at an end of the first half decarburization treatment, the molten steel information is information on a weight and a component content of the molten steel before the vacuum degassing treatment, and the actual operation result values contain a flow rate of exhaust gas emitted during the vacuum degassing treatment and a measurement result of a component content in the exhaust gas.

4. Vacuum degassing equipment comprising: a vacuum degassing device configured to perform vacuum degassing treatment including decarburization treatment by placing molten steel in a reduced-pressure environment; a control device configured to control the vacuum degassing device; and a decarburization treatment end determination device configured to determine an end of the decarburization treatment, wherein the decarburization treatment end determination device has a carbon content calculation unit configured to divide the decarburization treatment into two stages of first half decarburization treatment and latter half decarburization treatment, estimate carbon contents in molten steel in the first half decarburization treatment and the latter half decarburization treatment using different estimation models for the carbon content in molten steel for the first half decarburization treatment and the latter half decarburization treatment, and use the estimation model constructed based on machine learning for actual operation result data in the first half decarburization treatment, and a decarburization treatment end determination unit configured to determine the end of the decarburization treatment when an estimated value of the carbon content in molten steel has reached a target value in the latter half decarburization treatment.

5. The vacuum degassing equipment according to claim 4, wherein the carbon content calculation unit uses pressure in the reduced-pressure environment for a division reference for the first half decarburization treatment and the latter half decarburization treatment.

6. The vacuum degassing equipment according to claim 5, wherein in the estimation model of the first half decarburization treatment, an explanatory variable is constructed by operation information containing molten steel information and actual operation result values and an object variable is constructed by the carbon content in molten steel at an end of the first half decarburization treatment, the molten steel information is information on a weight and a component content of the molten steel before the treatment of the vacuum degassing treatment, and the actual operation result values contain a flow rate of exhaust gas emitted during the vacuum degassing treatment and a measurement result of the component content in the exhaust gas.

7. A method for operating vacuum degassing equipment for performing vacuum degassing treatment including decarburization treatment by placing molten steel in a reduced-pressure environment, the method comprising: using the decarburization treatment determination method according to any one of claims 1 to 3 when an end of the decarburization treatment is determined.

8. A method for producing molten steel for producing the molten steel refined by applying vacuum degassing treatment to the molten steel in vacuum degassing equipment operated by the operation method according to claim 7.

Images