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

Abstract

Provided are 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, with which it is possible to estimate the carbon concentration in molten steel with high accuracy after appropriately utilizing an exhaust gas measurement value and to end a decarburization treatment at the appropriate timing. Provided is a method for determining a decarburization treatment end, in which the end of a decarburization treatment is determined when performing a vacuum degassing treatment including the decarburization treatment by placing molten steel in a reduced-pressure environment, wherein: the decarburization treatment is divided into two stages which are a first half decarburization treatment M1 and a latter half decarburization treatment M2; estimation models for the carbon concentration in the molten steel, which differ between the first half decarburization treatment M1 and the latter half decarburization treatment M2, are used to estimate the carbon concentration in the molten steel in the first half decarburization treatment M1 and the latter half decarburization treatment M2; in the first half decarburization treatment M1, an estimation model constructed on the basis of machine learning with respect to actual operation results data is used (S3); and, in the latter half decarburization treatment M2, the end of the decarburization treatment is determined when the estimated value of the carbon concentration in the molten steel reaches a target value (S7).

Description

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

[0001] The present disclosure relates to a method for determining the completion of decarburization treatment, a vacuum degassing facility, a method for operating a vacuum degassing facility, and a method for producing molten steel.

[0002] In the steelmaking process, the molten steel composition is adjusted by removing impurities such as carbon from the molten iron and adding useful alloying elements. Regarding carbon in particular, vacuum degassing equipment is used to place the molten steel in a vacuum environment, which promotes decarburization and makes it possible to produce ultra-low carbon steel with a carbon concentration of less than 10 ppm in the molten steel.

[0003] In the vacuum degassing process, the carbon concentration in the molten steel is not directly measured but is only indirectly estimated from the concentrations of carbon monoxide and carbon dioxide in the exhaust gas. In addition, in the actual production of ultra-low carbon steel, there is a tendency to carry out the decarburization process for an excessively long period of time due to concerns about the carbon concentration not meeting the specifications.

[0004] In order to solve the problem of excessive decarburization, which results in longer processing times, it is effective to estimate the carbon concentration in molten steel with high accuracy during processing, and various methods have been proposed. Methods for estimating the carbon concentration in molten steel can be broadly divided into two types: methods that use exhaust gas measurements and methods that build a decarburization reaction model.

[0005] In the method of estimation using the flow rate and measurement values ​​of exhaust gas, the amount of decarburization is calculated from measurement values ​​(e.g., measurement values ​​of flow rate and component concentration) of exhaust gas discharged from the vacuum degassing equipment during treatment, and the carbon concentration in molten steel is estimated. In the method of constructing a decarburization reaction model, the carbon concentration in molten steel is estimated by physically considering the details of the decarburization reaction in the vacuum degassing equipment and constructing a decarburization reaction model (for example, Non-Patent Document 1).

[0006] Furthermore, as a combination of both methods, a method has been proposed in which parameters of a decarburization reaction model are determined from exhaust gas measurement values, and the carbon concentration in molten steel is estimated using the decarburization reaction model having the determined parameters (e.g., Patent Documents 1 and 2).

[0007] JP 2005-330512 A JP 2015-101742 A

[0008] Shinya Kitamura and three others, "Decarburization Reaction Model in Vacuum Degassing Furnace," Iron and Steel, Vol. 80 (1994) No. 3, pp. 213-218

[0009] The measured flow rate and concentration of the exhaust gas emitted from the vacuum degassing equipment during the process contain information on the actual decarburization performance, and utilizing this information is believed to contribute to the realization of highly accurate carbon concentration estimation. However, the measured exhaust gas values ​​contain errors, and using the measured exhaust gas values ​​to determine model parameters, as in the techniques of Patent Documents 1 and 2, can lead to a decrease in the accuracy of the carbon concentration estimation due to the errors in the measured exhaust gas values. In particular, at the end of the decarburization process, the allowable error in the estimated carbon concentration in the molten steel is only a few ppm, and the impact of errors in the measured exhaust gas values ​​can be fatal. Thus, reducing the adverse impact of errors in the measured exhaust gas values ​​is important when using the measured exhaust gas values ​​to estimate the carbon concentration.

[0010] The present disclosure has been made in consideration of the above circumstances, and aims to provide a method for determining the completion of decarburization treatment, a vacuum degassing equipment, a method for operating a vacuum degassing equipment, and a method for producing molten steel, which are capable of estimating the carbon concentration in molten steel with high accuracy by appropriately utilizing exhaust gas measurement values ​​and completing the decarburization treatment at an appropriate timing.

[0011] (1) According to one aspect of the present disclosure, there is provided a method for determining the completion of decarburization treatment when a vacuum degassing treatment including decarburization treatment is performed by placing molten steel in a reduced pressure environment, the method comprising: dividing the decarburization treatment into two stages, a first half of the decarburization treatment and a second half of the decarburization treatment; estimating the carbon concentrations in the molten steel in the first half of the decarburization treatment and the second half of the decarburization treatment, using different estimation models for the carbon concentration in the molten steel in the first half of the decarburization treatment and the second half of the decarburization treatment; in the first half of the decarburization treatment, using the estimation model constructed based on statistical machine learning of operational performance data; and in the second half of the decarburization treatment, determining the completion of the decarburization treatment when the estimated value of the carbon concentration in the molten steel has reached a target value.

[0012] (2) In the configuration of (1), there is provided a method for determining the completion of the decarburization process, in which the pressure in the reduced pressure environment is used as a criterion for dividing the decarburization process into the first half and the second half.

[0013] (3) In the configuration of (2) above, the estimation model for the first half of the decarburization treatment is constructed with explanatory variables being operational information including molten steel information and operational performance values, and with a target variable being the carbon concentration in the molten steel at the end of the first half of the decarburization treatment, the molten steel information being information relating to the weight and component concentrations of the molten steel before the vacuum degassing treatment, and the operational performance values ​​including measurement results of the exhaust gas flow rate and component concentrations in the exhaust gas discharged during the vacuum degassing treatment, and a method for determining the end of the decarburization treatment is provided.

[0014] (4) According to one aspect of the present disclosure, there is provided vacuum degassing equipment comprising: a vacuum degassing apparatus that performs vacuum degassing, including decarburization, by placing molten steel in a reduced pressure environment; a control device that controls the vacuum degassing apparatus; and a decarburization completion determination device that determines the completion of the decarburization, wherein the decarburization completion determination device divides the decarburization into two stages, a first half of the decarburization and a second half of the decarburization, and estimates the carbon concentrations in the molten steel in the first half of the decarburization and the second half of the decarburization using different estimation models of the carbon concentration in molten steel for the first half of the decarburization and the second half of the decarburization, and in the first half of the decarburization, uses the estimation model constructed based on machine learning of operational performance data; and a decarburization completion determination unit that determines the completion of the decarburization when the estimated value of the carbon concentration in molten steel reaches a target value in the second half of the decarburization.

[0015] (5) In the configuration of (4) above, a vacuum degassing facility is provided, in which the carbon concentration calculation unit uses the pressure in the reduced pressure environment as a dividing criterion for dividing the decarburization treatment into the first half and the second half.

[0016] (6) In the configuration of (5) above, the estimation model for the first half of the decarburization treatment is constructed with explanatory variables being operational information including molten steel information and operational performance values, and with a target variable being the carbon concentration in the molten steel at the end of the first half of the decarburization treatment, the molten steel information being information relating to the weight and component concentrations of the molten steel before the vacuum degassing treatment, and the operational performance values ​​including measurement results of the exhaust gas flow rate and component concentrations in the exhaust gas discharged during the vacuum degassing treatment.

[0017] (7) According to one aspect of the present disclosure, there is provided a method for operating vacuum degassing equipment that performs vacuum degassing treatment, including decarburization, by placing molten steel in a reduced pressure environment, the method using the method for determining decarburization treatment described in any one of the configurations (1) to (3) above when determining the completion of the decarburization treatment.

[0018] (8) According to one aspect of the present disclosure, there is provided a method for producing molten steel, in which the molten steel is subjected to a vacuum degassing treatment in vacuum degassing equipment operated by the operating method described in the configuration of (7) above, thereby producing refined molten steel.

[0019] According to one aspect of the present disclosure, there are provided a method for determining the completion of decarburization treatment, a vacuum degassing system, a method for operating a vacuum degassing system, and a method for producing molten steel, which are capable of accurately estimating the carbon concentration in molten steel by appropriately utilizing exhaust gas measurement values ​​and completing the decarburization treatment at an appropriate timing.

[0020] 1 is a block diagram of a vacuum degassing facility according to an embodiment of the present disclosure, and a block diagram of a determination device thereof. 2 is a flowchart showing a method for determining the completion of decarburization treatment according to an embodiment of the present disclosure. 3 is a graph showing an example of the change in pressure inside a vacuum chamber over time in a charge subjected to vacuum degassing treatment for producing an ultra-low carbon steel grade.

[0021] In the following detailed description, embodiments of the present disclosure will be described with reference to the drawings. In the description of the drawings, identical or similar parts are designated by identical or similar reference numerals, and redundant explanations will be omitted. The drawings are schematic and may differ from the actual product. Furthermore, the embodiments shown below exemplify devices and methods for embodying the technical ideas of the present disclosure, and the technical ideas of the present disclosure do not specify the materials, structures, arrangements, etc. of component parts as described below. The technical ideas of the present disclosure may be modified in various ways within the technical scope defined by the claims.

[0022] <Vacuum Degassing Apparatus> Hereinafter, a vacuum degassing apparatus according to an embodiment of the present disclosure will be described with reference to the drawings. As shown in Fig. 1, the vacuum degassing apparatus 1 includes a vacuum degassing apparatus 2, a control device 3, and a decarburization treatment completion determination device 4 (hereinafter, also simply referred to as a "determination device").

[0023] The vacuum degassing apparatus 2 is an RH vacuum degassing apparatus, and performs vacuum degassing on the molten steel 6 contained in the ladle 5. The vacuum degassing apparatus 2 includes a vacuum vessel 20, a pair of immersion pipes 21, piping 22, an injection lance 23, an exhaust duct 24, an exhaust gas flow meter 25, and an exhaust gas component concentration meter 26.

[0024] In the vacuum degassing apparatus 2, a pair of immersion pipes 21 provided at the bottom of the vacuum chamber 20 are immersed in the molten steel 6 in the ladle 5, and gas inside the vacuum chamber 20 is evacuated through an exhaust duct 24, thereby reducing the pressure in the vacuum chamber 20 and sucking up the molten steel 6. An inert gas is then blown from one of the immersion pipes 21 through a pipe 22, causing the molten steel 6 to reflux between the vacuum chamber 20 and the ladle 5. In the vacuum degassing apparatus 2, oxygen can be supplied to the molten steel 6 in the vacuum chamber 20 by blowing oxygen from an injection lance 23 installed in the vacuum chamber 20. The vacuum chamber 20 is an example of a vacuum zone of the vacuum degassing apparatus 2, i.e., a zone that is depressurized to create a vacuum. In this vacuum degassing apparatus 2, the molten steel 6 containing oxygen is refluxed in the vacuum zone, thereby promoting degassing and decarburization reactions in the molten steel 6. In this embodiment, the vacuum degassing treatment by the vacuum degassing equipment 1 is a refining treatment including a decarburization treatment (vacuum decarburization treatment) in which the molten steel 6 is placed in a reduced pressure environment to promote decarburization.

[0025] The exhaust gas flow meter 25 and the exhaust gas component concentration meter 26 are provided inside the exhaust duct 24. The exhaust gas flow meter 25 measures the flow rate of the exhaust gas. The exhaust gas component concentration meter 26 measures the flow rate of CO gas, CO ２ Gas, O ２ Measure the concentration of components in exhaust gas, including gases.

[0026] The control device 3 is composed of an information processing device such as a computer. The control device 3 controls operation-related manipulated variables, such as the exhaust rate of the exhaust system, the reflux inert gas flow rate, and the blown oxygen flow rate, so that the component concentrations and temperature of the molten steel 6 fall within target ranges after the vacuum degassing process from the actual values ​​before the vacuum degassing process. The control device 3 also collects operation result value data, such as the pressure inside the vacuum chamber 20 (vacuum chamber pressure), the reflux inert gas flow rate, the blown oxygen flow rate, the exhaust gas flow rate, and the exhaust gas component concentrations, and outputs the data to the determination device 4.

[0027] The determination device 4 is a device that performs decarburization treatment determination to determine the timing for ending the decarburization treatment in the vacuum degassing treatment, and is configured with an information processing device such as a computer. The determination device 4 functions as an operation information input unit (hereinafter also simply referred to as the "input unit") 40, a carbon concentration calculation unit (hereinafter also simply referred to as the "calculation unit") 41, a decarburization treatment end determination unit (hereinafter also simply referred to as the "end determination unit") 42, and a decarburization treatment timing determination unit (hereinafter also simply referred to as the "timing determination unit") 43, by executing a computer program with an arithmetic processing unit such as a CPU (Central Processing Unit) in the information processing device. The control device 3 and the determination device 4 are collectively referred to as a vacuum degassing treatment control system.

[0028] The input unit 40 receives as input information operational results and time-series flue gas measurement values ​​relating to operation variables during the vacuum degassing process. The time-series flue gas measurement values ​​include the flow rate of the flue gas discharged from the vacuum degassing apparatus 2, CO gas, CO ２ Gas and O ２ The gas component concentrations are included.

[0029] The calculation unit 41 estimates the carbon concentration in the molten steel 6 (carbon concentration in molten steel) based on the input information acquired by the input unit 40. The method of calculating the carbon concentration in molten steel uses a plurality of estimation models for the carbon concentration in molten steel, based on criteria including the pressure inside the vacuum vessel.

[0030] The determination unit 42 determines whether the decarburization treatment is complete based on the estimated value of the carbon concentration in the molten steel calculated by the calculation unit 41. The determination unit 42 also outputs the determination result of the completion of the decarburization treatment to the control device 3. The control device 3 may control the manipulated variables related to the operation based on the determination result of the completion of the decarburization treatment obtained from the decarburization treatment completion determination unit 42.

[0031] The timing determination unit 43 determines the timing of the decarburization process based on the input information acquired by the input unit 40. Specifically, the timing determination unit 43 determines whether the timing of the decarburization process is the first half or the second half by determining the end of the first half of the decarburization process, which will be described later.

[0032] In the vacuum degassing equipment 1, the molten steel 6 is subjected to vacuum degassing, thereby adjusting the composition and temperature, including decarburization, of the molten steel 6, and producing molten steel 6 with predetermined composition and temperature. During the decarburization, the determination device 4 estimates the internal state of the vacuum degassing equipment 2 and determines the timing for completing the decarburization. The vacuum degassing is then performed based on the determination result of the decarburization. Known processes can be used for the vacuum degassing process steps other than the determination of the completion of the decarburization.

[0033] <Method for Determining the Completion of Decarburization Treatment> The method for determining the completion of decarburization treatment according to this embodiment is performed in accordance with the flowchart shown in Fig. 2. The determination of the completion of decarburization treatment according to this embodiment is broadly composed of an estimated calculation of the carbon concentration in the molten steel in a first half M1 of the decarburization treatment (hereinafter also simply referred to as the "first half"), and an estimated calculation of the carbon concentration in the molten steel in a second half M2 of the decarburization treatment (hereinafter also simply referred to as the "second half") and a determination of the completion of the decarburization treatment. Switching from the first half M1 to the second half M2 of the decarburization treatment is performed based on a determination criterion that includes the pressure inside the vacuum chamber 20. The flowchart shown in Fig. 2 starts when a command to execute the decarburization treatment is input, and proceeds to processing in step S1.

[0034] First, in the process of step S1, the input unit 40 acquires molten steel information before the start of the decarburization process from a host computer (not shown) or the like. The molten steel information may include, for example, the weight of the molten steel 6 and component concentrations that are the results of measurements and analyses obtained by component analysis of the molten steel 6. This completes the process of step S1, and the process of determining the end of the decarburization process proceeds to the process of step S2.

[0035] In the processing of step S2, the input unit 40 acquires operational performance values ​​during the decarburization treatment. The operational performance values ​​are items necessary for the calculation by the calculation unit 41. In this embodiment, the input unit 40 acquires, as the operational performance values, at least the measurement results of the flow rate and component concentration of the exhaust gas discharged from the vacuum degassing apparatus 2 during the vacuum degassing treatment. In this embodiment, the input unit 40 also acquires, as the operational performance value, the pressure inside the vacuum chamber 20 during the decarburization treatment (the pressure in the reduced pressure environment inside the vacuum chamber 20). Furthermore, the operational performance values ​​may further include information such as the flow rate of the reflux inert gas and the oxygen flow rate from the injection lance 23 during the decarburization treatment, as well as information about the auxiliary materials added during the decarburization treatment. Specific examples of the information about the auxiliary materials include the type and amount of the auxiliary materials added. This completes the processing of step S2, and the processing for determining the completion of the decarburization treatment proceeds to the processing of step S3. Here, steps S1 and S2 correspond to input steps. The molten steel information and the operational performance values ​​are collectively referred to as operational information.

[0036] In step S3, the timing determination unit 43 determines the completion of the first half of the decarburization process based on the operation performance values ​​acquired in step S2. The criteria for determining the completion of the first half include at least whether the pressure in the vacuum chamber 20 has reached a predetermined threshold (below the threshold). If necessary in consideration of desirable conditions for the boundary between the first and second half of the decarburization process, described below, other conditions that must be met to determine the completion of the first half M1 (first half completion) may be added, in addition to the pressure in the vacuum chamber 20 being below the threshold. If it is determined that the first half M1 has not yet completed, the process returns to step S2, and steps S2 and subsequent steps are repeated using newly input operation performance values. On the other hand, if it is determined that the first half M1 has completed, step S3 is completed, and the process for determining the completion of the decarburization process proceeds to step S4.

[0037] In the processing of step S4, the calculation unit 41 estimates the carbon concentration in the molten steel at the end of the first half M1 (end of the first half). Not only for machine learning models, but also for estimation models in general, the smaller the variance in the actual value of the dependent variable, the easier it is to construct a highly accurate estimation model. In estimating the carbon concentration in the vacuum decarburization process, factors that cause variance in the carbon concentration in the molten steel at the end of the first half, which is the dependent variable, include the carbon concentration in the molten steel before the vacuum decarburization process (pre-process carbon concentration). Other factors that cause variance in the carbon concentration in the molten steel include changes in the airtightness of the vacuum chamber 20 and the exhaust system due to reconnection between the vacuum chamber 20 and the exhaust system, and changes in the time required for the air pressure in the vacuum chamber 20 to decrease to the ultimate air pressure due to oxygen injection during the vacuum degassing process. The pressure in the vacuum chamber 20 is referred to as the vacuum chamber pressure.

[0038] The inventors have found that determining the end of the first half based on the vacuum chamber pressure reduces the variation in the actual carbon concentration in the molten steel at the end of the first half. Among the factors that cause variation in the carbon concentration in the molten steel described above, the validity of this finding can be seen by considering the pre-treatment carbon concentration as follows: The vacuum chamber pressure is determined by the balance between the exhaust rate from the vacuum chamber 20 and the gas supply rate to the vacuum chamber 20. The former can be considered to be a generally constant value depending on the exhaust capacity of the vacuum degassing apparatus 2. On the other hand, the decarburization rate, i.e., the CO gas supply rate to the vacuum chamber 20, increases as the carbon concentration in the molten steel increases. Therefore, it is considered that the carbon concentration in the molten steel at a given vacuum chamber pressure will be a generally constant value.

[0039] The vacuum chamber pressure used as the reference for the end of the first half of the process is preferably determined taking into consideration the following two points. First, it is desirable to select a pressure that is unique at each time point during the vacuum decarburization process. During the vacuum degassing process, the vacuum chamber pressure does not decrease monotonically but may temporarily increase when the operation mode of the exhaust equipment is switched, when the decarburization reaction is active, when oxygen is blown in, and so on. The end of the first half of the process must be determined appropriately during the decarburization process. If the same pressure is observed at multiple times, this will lead to variations in the carbon concentration in the molten steel at the end of the first half of the process. While the above explanation points out the drawbacks of the same pressure being observed at multiple times, if the reference vacuum chamber pressure is too low, the vacuum chamber pressure will not decrease to the reference value, making it impossible to properly determine the end of the decarburization process.

[0040] Second, the reference pressure in the vacuum chamber is desirably low within a range that does not excessively increase the variability in the actual values ​​of the carbon concentration in the molten steel at the end of the first half. For example, even if an extremely high value such as 300 Torr is set as the reference pressure in light of the first point, this has almost no effect on improving the accuracy of the carbon concentration estimation. Furthermore, as will be described later, exhaust gas measurement values ​​are used to estimate the carbon concentration in the molten steel in the first half M1, and in order to make the most of the advantage of the exhaust gas measurement values ​​that they contain information on the actual decarburization performance, it is desirably a low reference pressure.

[0041] If there is no vacuum chamber pressure that satisfies the above two conditions and the timing for the end of the first half cannot be determined appropriately based on pressure alone, other conditions may be added, such as whether or not oxygen is blown in, or the processing time.

[0042] Figure 3 shows an example of the change in pressure over time in the vacuum chamber for a charge subjected to vacuum degassing treatment to produce an ultra-low carbon steel. From Figure 3 , when selecting the reference pressure for the end of the first half of the decarburization treatment, it is desirable to select a pressure from the range of 3 Torr to 6 Torr or 60 Torr or higher, which uniquely determines the time, from the viewpoint of the first point mentioned above. Furthermore, taking into account the viewpoint of the second point mentioned above, it is more desirable to select the reference pressure from the lower range of 3 Torr to 6 Torr. For example, the reference pressure may be 5 Torr. When selecting a single reference pressure for the end of the first half of the decarburization treatment, it is desirable to take into account the variation in the change in pressure over time in the vacuum chamber for each charge and the accuracy of estimating the carbon concentration in the molten steel at the end of the decarburization treatment.

[0043] As is clear from Fig. 3, the pressure suitable as a criterion for the completion of the first half of the decarburization treatment is not necessarily a clear inflection point with respect to time. In the vacuum degassing treatment used to produce ultra-low carbon steel, the decarburization reaction rate is ultimately determined by the carbon supply, but even if the boundary between the first and second half of the decarburization treatment is set during the carbon supply rate-determining stage, it is possible to estimate the carbon concentration in molten steel with high accuracy.

[0044] In this embodiment, the carbon concentration in the molten steel at the end of the first half is estimated by a machine learning model that uses the operational information acquired by the input unit 40 as explanatory variables. The type of machine learning model is selected to be one that can estimate the carbon concentration in the molten steel at the end of the first half with high accuracy. For example, a linear regression model, a neural network model, a support vector regression model, or a decision tree model can be cited as candidates. In this embodiment, a linear regression model is used. In this case, the machine learning model is expressed by the following formula (1).

[0045]

[0046] Here, C ０ is the estimated value of the carbon concentration in the molten steel at the end of the first half, which is the objective variable. ｊ (j=1, 2, ..., N) is an explanatory variable. ｊ is the explanatory variable X ｊ is a parameter corresponding to α ０is the constant term of the machine learning model. ｊ and the constant term α ０ can be determined by preparing a large number of data sets (operational performance data) of previously collected response variables and explanatory variables and applying a known method.

[0047] Explanatory variable X of the machine learning model ｊ The data includes, for example, information on the weight of molten steel and the concentration of components before treatment, as shown in Table 1 below, and a plurality of operational performance values, including measurement results of the flow rate of exhaust gas discharged from the vacuum degassing equipment during treatment and the concentration of components in the exhaust gas.

[0048]

[0049] Here, the cumulative decarbonization amount based on the exhaust gas information of the first half M1 included as an explanatory variable in Table 1 is calculated by the calculation process described below.

[0050] Carbon emitted from molten steel is CO or CO ２ Taking into account that the amount of carbon in the exhaust gas per unit time is expressed by the following formula (2): ０ ) The cumulative amount of carbon emissions is calculated using the following formula (3).

[0051]

[0052] Here, q Ｃ，ＯＧ (t) is the amount of carbon in the exhaust gas per unit time [kg / s] at time t. m Ｃ is the molar mass of carbon [g / mol]. ｏｆｆ (t) is the volumetric flow rate of exhaust gas at time t [Nm ３ / s]. ＣＯ (t) is the CO concentration [vol%] in the exhaust gas at time t. ＣＯ２ (t) is the CO in the exhaust gas at time t ２ Concentration [vol%]. Q Ｃ，ＯＧ (t ０ ) is from time 0 to t ０ This is the cumulative carbon emissions [kg] up to now.

[0053] As described above, the type of machine learning model is not limited to the linear regression model described in this embodiment, and other model types, such as a neural network model, may be adopted as long as a significant effect can be confirmed in improving the estimation accuracy of the carbon concentration in molten steel. However, the following points must be taken into consideration when selecting a machine learning model. The ultimate goal of the present disclosure is to shorten the vacuum degassing treatment time, and to achieve this, the estimation accuracy of the carbon concentration in molten steel at the end of the decarburization treatment must be high. If the estimation accuracy of the carbon concentration in molten steel at the end of the first half is high, it can be expected that the estimation accuracy of the carbon concentration in molten steel at the end of the decarburization treatment will also be high, but there is no direct relationship between the two. Therefore, care must be taken to select a model that improves the estimation accuracy of the latter rather than the former.

[0054] By applying the formula (1), it is possible to estimate the carbon concentration in the molten steel at the end of the first half portion. This completes the processing of step S4, and the process proceeds to step S5 of the second half portion M2.

[0055] In the processing of step S5, the operation information input unit 40 acquires operational performance values ​​during the decarburization treatment. The acquired operational performance values ​​may be the same as those acquired in the processing of step S2, or may be adjusted according to information required for the subsequent estimation of the carbon concentration in the molten steel. For example, the measurement results of the flow rate and component concentrations of the exhaust gas discharged from the vacuum degassing apparatus 2 differ from the estimated calculation of the carbon concentration in the molten steel in the first half M1, and are not necessarily required operational performance values. When the processing of step S5 is completed, the processing for determining the completion of the decarburization treatment proceeds to the processing of step S6. The molten steel information and operational performance values ​​are collectively referred to as operational information.

[0056] In the process of step S6, the carbon concentration calculation unit 41 estimates the carbon concentration in molten steel. In the process of step S6, a requirement for the estimation model of the carbon concentration in molten steel used by the carbon concentration calculation unit 41 is that the carbon concentration in molten steel can be estimated at predetermined intervals or continuously. Therefore, the carbon concentration is estimated using an estimation model different from the estimation model of the carbon concentration in molten steel used by the carbon concentration calculation unit 41 in the process of step S4.

[0057] In this embodiment, in step S6, the decarburization reaction models of the following formulas (4) and (5) are used.

[0058]

[0059] Here, w is the mass of molten steel [kg], C is the carbon concentration in the molten steel [ppm], Q is the molten steel reflux rate [kg / s], and ak is the decarburization reaction capacity coefficient [kg / s]. C Ｅ is the equilibrium value [ppm] of the carbon concentration in the molten steel in the vacuum vessel 20. ａｌｌｏｙ is the carbon concentration in molten steel converted from the weight of carbon in the input auxiliary materials [ppm]. The subscript L indicates the physical quantity of the molten steel 6 in the ladle 5. The subscript V indicates the physical quantity of the molten steel 6 in the vacuum vessel 20. For example, C Ｖ indicates the carbon concentration [ppm] in the molten steel in the vacuum vessel 20. The subscript i is used to identify a specific decarburization reaction site. Specific decarburization reaction sites include the surface of the molten steel and the reflux inert gas bubbles.

[0060] The amount of change in the carbon concentration in the molten steel per minute time is calculated using equations (4) and (5), and the carbon concentration in the molten steel after the minute time is calculated by subtracting this amount from the current carbon concentration in the molten steel. This completes the process of step S6, and the process of determining whether the decarburization process has been completed proceeds to step S7.

[0061] In the process of step S7, the completion determination unit 42 determines whether the carbon concentration in the molten steel (estimated value of the carbon dioxide concentration in the molten steel) obtained in step S6 has reached a predetermined target value (whether it is equal to or less than the target value). If the estimated value of the carbon concentration in the molten steel is higher than the target value, the process returns to step S5, and the processes from step S6 onwards are repeated using newly input operational performance values. On the other hand, if the estimated value of the carbon concentration in the molten steel is equal to or less than the target value, it is determined that the decarburization process has been completed.

[0062] As described above, according to the method for determining the completion of decarburization treatment according to this embodiment, the decarburization treatment is divided into two stages: a first half M1 and a second half M2. Then, different estimation models for the carbon concentration in molten steel are used for the first half M1 and the second half M2, and the carbon concentrations in molten steel in the first half M1 and the second half M2 are estimated. Furthermore, in the first half M1, an estimation model constructed based on machine learning of operational performance data is used, and in the second half M2, the completion of the decarburization treatment is determined when the estimated value of the carbon concentration in molten steel has reached a target value.

[0063] Furthermore, according to the method for determining the completion of decarburization treatment according to this embodiment, the pressure inside the vacuum vessel, i.e., the pressure of the molten steel 6 in a reduced pressure environment, may be used as a dividing criterion between the first half M1 and the second half M2. Furthermore, the estimation model for the first half M1 may be constructed using operational information including molten steel information and operational performance values ​​as explanatory variables, and the carbon concentration in the molten steel at the end of the first half of the decarburization treatment as a target variable. Furthermore, the molten steel information may be information regarding the weight and component concentrations of the molten steel before the vacuum degassing treatment, and the operational performance values ​​may include measurement results of the flow rate of the exhaust gas discharged during the vacuum degassing treatment and the component concentrations in the exhaust gas.

[0064] The above-described configuration and process enable the estimation of the carbon concentration in molten steel to reflect the results of the decarburization process while minimizing the adverse effects of errors in exhaust gas measurement values. Therefore, the carbon concentration in molten steel can be estimated with high accuracy, and the decarburization process can be terminated at an appropriate timing relative to the carbon concentration standard. Furthermore, the highly accurate estimation makes it possible to avoid performing the decarburization process for an excessively long time due to concerns about the carbon concentration not meeting the standard, thereby shortening the decarburization process time.

[0065] <Modifications> Although the present disclosure has been described above with reference to specific embodiments, it is not intended that the invention be limited by these descriptions. By referring to the description of the present disclosure, other embodiments of the present disclosure that include various modifications in addition to the disclosed embodiments will be apparent to those skilled in the art. Therefore, it should be understood that the embodiments of the invention described in the claims also cover embodiments that include these modifications described herein, either alone or in combination.

[0066] For example, in the above embodiment, the vacuum degassing apparatus 2 is an RH vacuum degassing apparatus, but the present disclosure is not limited to such an example. For example, the vacuum degassing apparatus 2 may be of another type, such as a DH type having a single immersion tube 21, or an apparatus that creates a vacuum on the surface of the molten steel 6 in the ladle 5 without using a vacuum vessel 20.

[0067] The effects of the present disclosure will be specifically described below based on examples, but the present disclosure is not limited to the contents of the examples. In this example, as in the above embodiment, decarburization was performed in a vacuum degassing facility 1 using an RH vacuum degassing device as the vacuum degassing device 2, and ultra-low carbon molten steel with a carbon concentration upper limit of 25 ppm was produced. Before and after the start and end of the vacuum degassing process, a portion of the molten steel was taken as a sample, and the carbon concentration in the molten steel of this sample was measured. In addition, the exhaust gas flow rate, CO concentration in the exhaust gas, and CO concentration in the exhaust gas during the vacuum degassing process were measured. ２ The change in concentration over time was measured. The termination of decarburization treatment was at the discretion of the operator.

[0068] Flow rate of exhaust gas during vacuum degassing treatment, CO concentration in exhaust gas, and CO in exhaust gas ２ When the actual concentration values ​​are measured, the decarburization rate and cumulative decarburization amount during vacuum degassing can be determined by performing calculations similar to those of Equations (2) and (3). On the other hand, the cumulative decarburization amount determined from the exhaust gas measurement values ​​does not necessarily match the actual decarburization amount calculated from the actual measured values ​​of the carbon concentration in molten steel before and after vacuum degassing. For this reason, the decarburization rate is corrected by multiplying it by a constant so that the former matches the latter. By performing such calculations, a reliable value for the carbon concentration in molten steel at any time during vacuum degassing can be determined from the actual measured value of the carbon concentration in molten steel and the measured value of the exhaust gas.

[0069] In this example, the carbon concentration in the molten steel at the end of the first half calculated by the above method was used as the actual value. The first half M1 and the second half M2 were divided based on the vacuum chamber pressure of 5 torr. In this example, the number of analysis data used in the analysis was 46.

[0070] Table 2 shows the correlation coefficients between the measured value of the carbon concentration in the molten steel before the start of vacuum degassing treatment (pre-treatment carbon concentration) and the actual values ​​of the treatment time of the first half M1 (first half treatment time) and the carbon concentration in the molten steel at the end of the first half (post-first half carbon concentration). Table 2 shows that there is a high positive correlation between the measured value of the pre-treatment carbon concentration and the first half treatment time. On the other hand, there is a low correlation between the measured value of the pre-treatment carbon concentration and the actual value of the post-first half carbon concentration. Therefore, it can be seen that by determining the end of the first half based on the pressure inside the vacuum chamber, the influence of the pre-treatment carbon concentration on the variability in the actual value of the post-first half carbon concentration can be eliminated.

[0071]

[0072] Table 3 shows the standard deviation of the actual values ​​of the carbon concentration in molten steel at the end of the first half when the decarburization treatment was divided into an early part and an late part by the inventive method and the comparative method, for the inventive example and the comparative example. The number of analysis data points for each of the inventive example and the comparative example was 46. In the comparative example, the end of the first half was determined based on the vacuum decarburization treatment time, and the specific time was determined so that the average actual values ​​of the carbon concentration in molten steel would match those of the proposed method. Specifically, in the comparative example, the period from the start of treatment to 452 seconds was defined as the first half of the decarburization treatment. In the inventive example, the standard deviation of the actual values ​​of the carbon concentration in molten steel at the end of the first half was smaller than in the comparative example, and it can be seen that dividing the decarburization treatment into an early part and a late part based on the pressure inside the vacuum vessel can reduce the variability in the actual values ​​of the carbon concentration in molten steel.

[0073]

[0074] Table 4 shows the root mean square error (RMSE) of the estimation error of the carbon concentration in molten steel at the end of the first half, i.e., when the pressure in the vacuum vessel was 5 torr. As a comparative example, the carbon concentration in molten steel was estimated using a decarburization reaction model. As an inventive example, the carbon concentration in molten steel was estimated using a linear regression model (ridge regression model) and a neural network model. The number of analysis data sets for the inventive example and the comparative example was 46. Table 4 shows that both inventive examples were able to estimate the carbon concentration in molten steel with higher accuracy than the comparative example. It is also clear that, among the inventive examples, the neural network model was able to perform estimation with higher accuracy than the linear regression model.

[0075]

[0076] The first row of Table 5 shows the standard deviation of the estimation error of the carbon concentration in molten steel at the end of the decarburization treatment. The model whose results are shown in Table 4 was used to estimate the carbon concentration in molten steel in the first half M1, and the same decarburization reaction model was used in all examples to estimate the carbon concentration in molten steel in the second half M2. The number of analytical data sets for the inventive examples and comparative examples was 46. Table 5 shows that both inventive examples were able to estimate the carbon concentration in molten steel with higher accuracy than the comparative examples. Meanwhile, it can be seen that the difference in estimation accuracy between the linear regression model and the neural network model used to estimate the carbon concentration in molten steel in the inventive examples was small. Thus, high accuracy in estimating the carbon concentration in molten steel at the end of the first half does not necessarily mean high accuracy in estimating the carbon concentration at the end of the decarburization treatment, so care must be taken when selecting a machine learning model.

[0077] The present disclosure aims to prevent excessive decarburization during vacuum degassing and shorten the processing time by estimating the carbon concentration in molten steel with high accuracy. The effect of preventing excessive decarburization and the shortened decarburization time when the carbon concentration in molten steel is estimated using the model of the invention example instead of the comparative example in this example are shown in the second and third rows of Table 5. When determining the end of decarburization based on the estimated carbon concentration in molten steel, the target value of the carbon concentration in molten steel referenced in step S7 must be set so as not to exceed the upper limit of the carbon concentration specification, taking into account the error of the carbon concentration estimation model. Specifically, the target value can be set by subtracting three times the standard deviation of the error of the carbon concentration estimation model from the upper limit of the carbon concentration specification. When setting the target value in this way, as shown in the second row of Table 5, by using the model of the invention example rather than the comparative example, the target value for determining the end of decarburization can be set to a value three times higher than the difference in the model error standard deviation. In this example, the effect of increasing the target value on the shortened decarburization time is shown in the third row of Table 5.

[0078]

[0079] Although the embodiments of the present disclosure have been described based on the drawings and examples, it should be noted that those skilled in the art would easily be able to make various modifications or alterations based on the present disclosure. Therefore, it should be noted that these modifications and alterations are included within the scope of the present disclosure. For example, the functions included in each component or step can be rearranged so as not to cause logical inconsistencies, and multiple components or steps can be combined or divided into one. The embodiments of the present disclosure can also be realized as a storage medium on which a program executed by a processor included in an apparatus is recorded. It should be understood that these are also included within the scope of the present disclosure.

[0080] REFERENCE SIGNS LIST 1 Vacuum degassing equipment 2 Vacuum degassing device 20 Vacuum tank 21 Immersion tube 22 Piping 23 Injection lance 24 Exhaust duct 25 Exhaust gas flow meter 26 Exhaust gas component concentration meter 3 Control device 4 Decarburization treatment completion determination device (determination device) 40 Operation information input unit (input unit) 41 Carbon concentration calculation unit (calculation unit) 42 Decarburization treatment completion determination unit (completion determination unit) 43 Decarburization treatment timing determination unit (timing determination unit) 5 Ladle 6 Molten steel M1 Decarburization treatment first half (first half) M2 Decarburization treatment second half (second half)

Claims

1. A method for determining the completion of a decarburization process when a vacuum degassing process including a decarburization process is performed by placing molten steel in a reduced pressure environment, the method comprising: dividing the decarburization process into two stages, an early part of the decarburization process and a late part of the decarburization process; estimating the carbon concentrations in the molten steel in the early part of the decarburization process and the late part of the decarburization process using different estimation models of the carbon concentration in the molten steel for the early part of the decarburization process and the late part of the decarburization process; in the early part of the decarburization process, using the estimation model constructed based on machine learning on operational performance data; and in the late part of the decarburization process, determining the completion of the decarburization process when the estimated value of the carbon concentration in the molten steel reaches a target value.

2. A method for determining the completion of decarburization treatment according to claim 1, wherein the pressure in the reduced pressure environment is used as a criterion for dividing the decarburization treatment into the first half and the second half.

3. A method for determining the completion of decarburization treatment as described in claim 2, wherein the estimation model for the first half of the decarburization treatment is constructed with explanatory variables being operational information including molten steel information and operational performance values, and a target variable being the carbon concentration in the molten steel at the end of the first half of the decarburization treatment, the molten steel information being information regarding the weight and component concentrations of the molten steel before the vacuum degassing treatment, and the operational performance values ​​including measurement results of the exhaust gas flow rate and component concentrations in the exhaust gas discharged during the vacuum degassing treatment.

4. Vacuum degassing equipment comprising: a vacuum degassing apparatus that performs vacuum degassing treatment including decarburization treatment by placing molten steel in a reduced pressure environment; a control device that controls the vacuum degassing apparatus; and a decarburization treatment completion determination device that determines the completion of the decarburization treatment, wherein the decarburization treatment completion determination device divides the decarburization treatment into two stages, a first half of the decarburization treatment and a second half of the decarburization treatment, and estimates the carbon concentrations in the molten steel in the first half of the decarburization treatment and the second half of the decarburization treatment using different estimation models of the carbon concentration in molten steel for the first half of the decarburization treatment and the second half of the decarburization treatment, and in the first half of the decarburization treatment, the estimation model is used based on machine learning on operational performance data; and a decarburization treatment completion determination unit that determines the completion of the decarburization treatment when the estimated value of the carbon concentration in molten steel reaches a target value in the second half of the decarburization treatment.

5. The vacuum degassing equipment according to claim 4, wherein the carbon concentration calculation unit uses the pressure in the reduced pressure environment as a dividing criterion for dividing the first half of the decarburization treatment from the second half of the decarburization treatment.

6. The vacuum degassing equipment of claim 5, wherein the estimation model for the first half of the decarburization treatment is constructed with explanatory variables being operational information including molten steel information and operational performance values, and a target variable being the carbon concentration in the molten steel at the end of the first half, the molten steel information being information regarding the weight and component concentrations of the molten steel before the vacuum degassing treatment, and the operational performance values ​​including measurement results of the exhaust gas flow rate and component concentrations in the exhaust gas discharged during the vacuum degassing treatment.

7. A method for operating a vacuum degassing facility which performs vacuum degassing including decarburization by placing molten steel in a reduced pressure environment, comprising the step of determining completion of the decarburization treatment using the method for determining completion of the decarburization treatment according to any one of claims 1 to 3.

8. A method for producing molten steel, comprising the steps of: subjecting molten steel to a vacuum degassing treatment in a vacuum degassing facility operated according to the operating method recited in claim 7, thereby producing refined molten steel.

Images