Molten steel temperature estimation method, molten steel temperature control method, and molten steel production method

Abstract

With this molten steel temperature estimation method, a molten steel temperature in a process for processing molten steel in a steelmaking step is estimated. This molten steel temperature estimation method includes: a step for dividing a period from the start to the end of processing into a plurality of sections by at least one molten steel temperature measurement time point in the middle of the processing; a step for measuring the molten steel temperature at the start time point of each section of the plurality of sections; a step for estimating the molten steel temperature change amount in a current section to which the current time point belongs and molten steel temperature change amounts in sections after the current section on the basis of a molten steel temperature estimation model constructed for each section of the plurality of sections; and a step for estimating the molten steel temperature at the end of the processing by combining the estimated molten steel temperature change amount in the current section and the molten steel temperature change amounts in the sections after the current section.

Description

Molten steel temperature estimation method, molten steel temperature control method, and molten steel manufacturing method

[0001] The present disclosure relates to a molten steel temperature estimation method, a molten steel temperature control method, and a molten steel manufacturing method.

[0002] Steelmaking processes include refining and casting. Refining involves controlling the component concentrations of molten steel within desired ranges. Casting involves solidifying the molten steel. To ensure smooth execution of these processes, it is important to control the temperature of the molten steel within an appropriate range.

[0003] The refining process involves primary refining and secondary refining, etc. One of the main roles of secondary refining is to regulate the temperature of molten steel.

[0004] When adjusting the temperature of molten steel in secondary refining, an operator usually measures the temperature of the molten steel during processing as necessary and estimates the subsequent change in the molten steel temperature. If there is a large difference between the estimated molten steel temperature and the target temperature, the operator performs an operation to heat or cool the molten steel.

[0005] Known methods for estimating a change in molten steel temperature include an estimation method based on a physical model and an estimation method based on a statistical model. For example, Patent Document 1 discloses a method for estimating a change in molten steel temperature based on a physical model. Furthermore, Patent Documents 2 and 3 disclose methods for estimating a change in molten steel temperature based on a statistical model.

[0006] JP 2016-180127 A JP 2012-57195 A JP 8-3621 A

[0007] Patent Document 1 discloses a method for estimating a change in the temperature of molten steel based on a physical model. In order to estimate a change in the temperature of molten steel based on a physical model, it is necessary to accurately grasp the amount of heat possessed by an object, such as a ladle, which is in direct contact with the molten steel and exchanges heat with the molten steel. Therefore, Patent Document 1 discloses a method for estimating the temperature distribution of the ladle based on a physical model and estimating the temperature of the molten steel based on the estimated temperature distribution of the ladle.

[0008] However, when the ladle is repeatedly used, an error in estimating the temperature distribution of the ladle accumulates. Also, it is difficult to accurately estimate the temperature distribution of the ladle based on the temperature of the ladle measured at a single point. Therefore, it is difficult to accurately estimate the change in the temperature of molten steel based on the method disclosed in Patent Document 1.

[0009] Patent Documents 2 and 3 disclose methods for estimating changes in molten steel temperature based on a statistical model constructed based on past processing results. Therefore, the methods disclosed in Patent Documents 2 and 3 do not require estimation of the temperature distribution in the ladle, as in the method disclosed in Patent Document 1.

[0010] The methods disclosed in Patent Documents 2 and 3 use the molten steel temperatures at the start and end of the refining process as learning data to construct a model for estimating the amount of change in molten steel temperature throughout the entire refining process. However, the model constructed in this manner has the following two problems.

[0011] The first problem is that, although a refining process includes various steps, the methods disclosed in Patent Documents 2 and 3 only consider the molten steel temperatures at the start and end, and therefore may not quantitatively and correctly distinguish between factors that affect the molten steel temperature. For example, it is possible that the amount of heat exchange between the molten steel and the ladle is overestimated and the amount of heat exchange between the molten steel and the added auxiliary materials is underestimated. In such a case, if the constructed statistical model is used to estimate the molten steel temperature from unknown data, there is a risk that the molten steel temperature estimation error will be large.

[0012] The second problem is that the methods disclosed in Patent Documents 2 and 3 only consider the molten steel temperatures at the start and end of the refining process, and therefore cannot accurately estimate the molten steel temperature during the refining process. For example, the heat exchange rate between the ladle and the molten steel decreases as the temperature difference between the two decreases, so it is clear that the molten steel temperature changes nonlinearly with time. Therefore, if the change in molten steel temperature is estimated based on the assumption that the change in molten steel temperature changes linearly with time, it is not possible to accurately estimate the change in molten steel temperature during the refining process.

[0013] In normal operations, operators measure the temperature of molten steel during a refining process and attempt to estimate the subsequent molten steel temperature based on the measured molten steel temperature. However, the statistical models used in Patent Documents 2 and 3 cannot estimate the subsequent molten steel temperature based on the molten steel temperature measured during the process. Therefore, it is difficult to compare the estimated molten steel temperature with a target temperature and perform operations such as heating or cooling the molten steel so as to reduce the difference between the estimated molten steel temperature and the target temperature.

[0014] An object of the present disclosure is to provide a molten steel temperature estimation method, a molten steel temperature control method, and a molten steel manufacturing method that are capable of accurately estimating the amount of change in molten steel temperature.

[0015] [1] A method for estimating the temperature of molten steel in a process for treating molten steel in a steelmaking process, comprising: a step of dividing a period from the start to the end of the process into a plurality of sections based on at least one time point of measuring the temperature of the molten steel during the process; a step of measuring the temperature of the molten steel at the start of each of the plurality of sections; a step of estimating an amount of change in the temperature of the molten steel in a current section to which the current time belongs and an amount of change in the temperature of the molten steel in sections after the current section, based on a molten steel temperature estimation model constructed for each of the plurality of sections; and a step of estimating the temperature of the molten steel at the end of the process by combining the estimated amount of change in the temperature of the molten steel in the current section and the amount of change in the temperature of the molten steel in sections after the current section.

[0016] [2] The molten steel temperature estimation method according to the above [1], wherein parameters of the molten steel temperature estimation model are determined based on past operational results.

[0017] [3] The molten steel temperature estimation method according to the above [1] or [2], wherein in the molten steel temperature estimation model, inputs are operation information of a process preceding the treatment, operation information up to the immediately previous section in the treatment, operation schedule information of the section for which the molten steel temperature estimation model is to be constructed, and equipment information of the treatment, and output is the amount of change in molten steel temperature from the start to the end of the section for which the molten steel temperature estimation model is to be constructed.

[0018] [4] The molten steel temperature estimation method according to any one of [1] to [3] above, wherein the operational information of a process preceding the treatment includes at least one of molten steel information, operational operation amount information, and treatment time information in the preceding process; the operational information up to one previous section in the treatment includes at least one of molten steel information, operational operation amount information, and treatment time information up to the one previous section; the operation schedule information for the section in which the molten steel temperature estimation model is to be constructed includes at least one of operational operation amount information and treatment time information for the section in which the molten steel temperature estimation model is to be constructed; and the equipment information for the treatment includes information on the usage history of a ladle used in the treatment and information on the actual measured temperature of the ladle.

[0019] [5] The molten steel temperature estimation method according to any one of [1] to [4] above, wherein the process is a secondary refining process, a process preceding the process is a primary refining process, operational information of the process preceding the process includes at least one of a weight of molten steel and a temperature of molten steel at the end of the previous process, operational information up to a previous section in the process includes at least information on the temperature of molten steel measured from the end of the previous section until the start of the current section, operation schedule information for a section in which the molten steel temperature estimation model is constructed includes information on at least one of an amount of auxiliary material input, an amount of oxygen blown in, and a processing time, and equipment information for the process includes at least one of actual temperature information obtained by measuring when a ladle used for the process is empty, a time period during which the ladle is empty after discharging the molten steel received previously and before receiving the molten steel this time, and a time period from the end of the previous secondary refining process to the start of the secondary refining process.

[0020] [6] The molten iron temperature estimation method according to any one of [1] to [5] above, wherein the molten steel temperature estimation model is a neural network model in the sections immediately after the start and immediately before the end of the secondary refining process, and a linear regression model in the other sections.

[0021] [7] A molten steel temperature control method, comprising: calculating a molten steel temperature adjustment manipulated variable for adjusting the molten steel temperature so that a difference between the molten steel temperature at the end of the treatment estimated by the molten steel temperature estimation method according to any one of [1] to [6] above and a target molten steel temperature at the end of the treatment falls within a predetermined range.

[0022] [8] A method for producing molten steel, comprising adjusting the molten steel temperature using the molten steel temperature control method according to [7] above.

[0023] According to the molten steel temperature estimation method, molten steel temperature control method, and molten steel manufacturing method disclosed herein, the amount of change in molten steel temperature can be estimated with high accuracy.

[0024] Fig. 1 is a diagram showing an example of a molten steel temperature control device that executes a molten steel temperature estimation method and a molten steel temperature control method according to an embodiment of the present disclosure. Fig. 2 is a block diagram showing an example of the configuration of a molten steel temperature control device according to an embodiment of the present disclosure. Fig. 3 is a diagram showing an example of how a refining process is divided into two sections. Fig. 4 is a diagram showing an example of how a refining process is divided into three sections. Fig. 4 is a flowchart showing an example of a molten steel temperature estimation method and a molten steel temperature control method according to an embodiment of the present disclosure. Fig. 5 is a diagram comparing a comparative example and this embodiment with respect to estimation errors in the amount of change in molten steel temperature.

[0025] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0026] FIG. 1 is a diagram illustrating an example of a molten steel temperature control device that executes a molten steel temperature estimation method and a molten steel temperature control method according to an embodiment of the present disclosure.

[0027] 1 shows an example of a case where secondary refining is performed in the refining process of a steelmaking process. In FIG. 1, a secondary refining apparatus 1 performs secondary refining on molten steel in a ladle 2.

[0028] The molten steel temperature control device 10 estimates the temperature of molten steel undergoing secondary refining treatment, and determines a molten steel temperature manipulation variable for adjusting the molten steel temperature based on the estimated molten steel temperature, thereby controlling the molten steel temperature.

[0029] In this embodiment, the molten steel temperature estimation method and the molten steel temperature control method are described using an example in which they are applied to a secondary refining process. However, this is merely an example, and the molten steel temperature estimation method and the molten steel temperature control method according to this embodiment can be applied to any process in which molten steel is processed in a steelmaking process. For example, the molten steel temperature estimation method and the molten steel temperature control method according to this embodiment can also be applied to a primary refining process. Here, in the primary refining process, the molten steel temperature rises by approximately 300°C due to desiliconization and decarburization reactions, whereas in the secondary refining process, the molten steel temperature changes by 20 to 40°C. Therefore, the required accuracy of estimating the molten steel temperature is higher in the secondary refining process than in the primary refining process. Therefore, the molten steel temperature estimation method and the molten steel temperature control method according to this embodiment can be more effective when applied to a secondary refining process.

[0030] 2 is a block diagram showing an example of the configuration of a molten steel temperature control device 10 according to an embodiment of the present disclosure. The molten steel temperature control device 10 may be a general-purpose computer such as a workstation or a personal computer, or may be a dedicated computer configured to function as the molten steel temperature control device 10.

[0031] The molten steel temperature control device 10 includes a control unit 11 , an input unit 12 , an output unit 13 , a memory unit 14 , and a communication unit 15 .

[0032] The control unit 11 includes at least one processor, at least one dedicated circuit, or a combination of these. The processor is a general-purpose processor such as a central processing unit (CPU) or a graphics processing unit (GPU), or a dedicated processor specialized for a specific process. The dedicated circuit is, for example, a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC).

[0033] The control unit 11 reads programs, data, etc. stored in the storage unit 14 and executes various functions.

[0034] The input unit 12 includes one or more input interfaces that detect user input and acquire input information based on the user's operation. The input unit 12 includes, for example, physical keys, capacitive keys, a touch screen that is integrated with the display of the output unit 13, or a microphone that accepts voice input.

[0035] The output unit 13 includes one or more output interfaces that output information to notify the user. The output unit 13 includes, for example, a display that outputs information as an image, a speaker that outputs information as sound, etc. The display included in the output unit 13 may be, for example, an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube) display, etc.

[0036] The storage unit 14 is, for example, a flash memory, a hard disk, an optical memory, etc. A part of the storage unit 14 may be located outside the molten steel temperature control device 10. In this case, the part of the storage unit 14 may be a hard disk, a memory card, etc. connected to the molten steel temperature control device 10 via an arbitrary interface.

[0037] The storage unit 14 stores programs for the control unit 11 to execute each function, data used by the programs, and the like.

[0038] The communication unit 15 includes at least one of a communication module compatible with wired communication and a communication module compatible with wireless communication. The molten steel temperature control device 10 is capable of communicating with other devices via the communication unit 15.

[0039] Next, a molten steel temperature estimation method and a molten steel temperature control method executed by the molten steel temperature control device 10 will be described using the case of a secondary refining process as an example.

[0040] In the secondary refining process, the timing for measuring the molten steel temperature during the secondary refining process may be determined depending on the type of steel used in the secondary refining process. Hereinafter, the timing for measuring the molten steel temperature during the secondary refining process will be referred to as the "molten steel temperature measurement time point." The secondary refining process has at least one molten steel temperature measurement time point.

[0041] The molten steel temperature estimation method according to this embodiment divides the period from the start to the end of the secondary refining process into a plurality of sections based on at least one time point at which the molten steel temperature is measured.

[0042] FIG. 3 shows an example of how the period from the start to the end of the secondary refining process is divided into two sections.

[0043] FIG. 3 shows an example of a process in which molten steel is sampled during the secondary refining process and the component concentrations of the sampled molten steel are measured.

[0044] In the example shown in Fig. 3, an operator starts a secondary refining process at time t0. At time t0 when the secondary refining process starts, the operator measures the temperature of the molten steel. At time t0, the temperature of the molten steel is T0.

[0045] At time ta, the operator samples the molten steel and measures the concentrations of the components in the sampled molten steel.

[0046] At time t1 immediately after time ta, the operator measures the molten steel temperature. Therefore, time t1 is the time point at which the molten steel temperature is measured. At the molten steel temperature measurement time t1, the molten steel temperature is T1.

[0047] At time t2, the operator ends the secondary refining process. At time t2 when the secondary refining process ends, the operator measures the molten steel temperature. At time t2, the molten steel temperature is T2.

[0048] In the example shown in Fig. 3, the period from the start to the end of the secondary refining process is divided into two sections by the time t1 at which the molten steel temperature is measured. In the example shown in Fig. 3, section 1 is from time t0 to time t1, and section 2 is from time t1 to time t2.

[0049] FIG. 4 shows an example of how the period from the start to the end of the secondary refining process is divided into three sections.

[0050] Figure 4 shows an example of a case where decarburization is performed during secondary refining. The decarburization of molten steel proceeds by reacting carbon in the molten steel with oxygen. When the molten steel is continuously cast, it is necessary to sufficiently reduce the oxygen concentration in the molten steel. Therefore, after performing the decarburization process to reduce the carbon concentration in the molten steel to a target value, the operator adds a deoxidizer to reduce the oxygen concentration in the molten steel.

[0051] In the example shown in Fig. 4, an operator starts a secondary refining process at time t0. At time t0 when the secondary refining process starts, the operator measures the temperature of the molten steel. At time t0, the temperature of the molten steel is T0.

[0052] At time tb, an operator adds a deoxidizer to the molten steel, but at time t1, before the time tb at which the deoxidizer is added, the operator measures the oxygen concentration in the molten steel and determines the amount of deoxidizer to be added. At this time, the operator also measures the temperature of the molten steel at time t1. Therefore, time t1 is the time point at which the temperature of the molten steel is measured. At the time point t1 at which the temperature of the molten steel is measured, the temperature of the molten steel is T1.

[0053] At time tb, the operator adds a deoxidizer and then stirs the molten steel. Then, at time t2, the operator measures the oxygen concentration in the molten steel to confirm whether the oxygen concentration has been sufficiently reduced. At this time, the operator also measures the molten steel temperature at time t2. Therefore, time t2 is the time point at which the molten steel temperature is measured. At the molten steel temperature measurement time t2, the molten steel temperature is T2.

[0054] At time t3, the operator ends the secondary refining process. At time t3 when the secondary refining process ends, the operator measures the molten steel temperature. At time t3, the molten steel temperature is T3.

[0055] In the example shown in Fig. 4, the period from the start to the end of the secondary refining process is divided into three sections by the molten steel temperature measurement times t1 and t2. In the example shown in Fig. 4, section 1 is from time t0 to time t1. Section 2 is from time t1 to time t2. Section 3 is from time t2 to time t3.

[0056] The molten steel temperature estimation method according to this embodiment measures the molten steel temperature at the start of each of a plurality of intervals. Then, the molten steel temperature estimation method according to this embodiment estimates the amount of change in molten steel temperature in the current interval to which the current time belongs and the amount of change in molten steel temperature in intervals after the current interval, based on a molten steel temperature estimation model constructed for each of the plurality of intervals. Details of the molten steel temperature estimation model will be described later.

[0057] For example, in the example shown in Fig. 4 , when the current time is between t0 and t1, the section to which the current time belongs is Section 1. The molten steel temperature estimation method according to this embodiment estimates the amount of change in molten steel temperature from the start to the end of Section 1, which is the current section, based on the molten steel temperature estimation model established for Section 1. Furthermore, the molten steel temperature estimation method according to this embodiment estimates the amount of change in molten steel temperature from the start to the end of Section 2, which is a section subsequent to Section 1, which is the current section, based on the molten steel temperature estimation model established for Section 2. Furthermore, the molten steel temperature estimation method according to this embodiment also estimates the amount of change in molten steel temperature from the start to the end of Section 3, which is a section subsequent to Section 1, which is the current section, based on the molten steel temperature estimation model established for Section 3.

[0058] In this way, by dividing the period from the start to the end of the secondary refining process into a plurality of sections based on the time points at which the molten steel temperature is measured and estimating the amount of change in the molten steel temperature for each section, the molten steel temperature estimation method according to this embodiment can accurately estimate the amount of change in the molten steel temperature for each section. For example, in the example shown in Fig. 4, the temperature of the molten steel rises when a deoxidizer is added at time tb, but such a temperature rise is observed only in section 2. Therefore, it is sufficient that only the molten steel temperature estimation model for section 2 is a model that takes into account the addition of a deoxidizer. By constructing a molten steel temperature estimation model for each section in this way, the molten steel temperature estimation method according to this embodiment can accurately estimate the amount of change in the molten steel temperature for each section.

[0059] As described above, the molten steel temperature estimation model is constructed for each interval. The molten steel temperature estimation model may be constructed in any model format that can accurately estimate the amount of change in molten steel temperature from the start to the end of the interval. For example, the model format of the molten steel temperature estimation model may be a linear regression model, a support vector regression model, a neural network model, or the like.

[0060] The model format of the molten steel temperature estimation model may vary depending on the dominant factors of the temperature accuracy for each section. For example, the molten steel temperature changes nonlinearly over time due to changes in the heat exchange rate between the ladle 2 and the molten steel. In the early section (immediately after the start) of the secondary refining process, when the temperature of the ladle 2 is relatively low, the heat exchange rate from the molten steel to the refractory on the inner wall of the ladle 2 is high, and the heat exchange rate is significantly affected by the amount of heat stored in the refractory in the ladle 2 immediately before the steel is poured into the refractory. In the later sections, when the temperature of the ladle 2 rises due to heat transfer from the molten steel, the heat exchange rate between the molten steel and the ladle 2 decreases, and the temperature accuracy is dominated by exothermic reactions due to the addition of auxiliary materials and oxygen injection. On the other hand, in the section immediately before the end of the secondary refining process, when the temperature change between the molten steel and the ladle 2 is small, the temperature change itself is small, and the temperature accuracy is dominated by uncertain factors such as wear and tear on the refractory on the inner wall of the ladle 2 and variations in the amount of heat conduction between the molten steel and the ladle due to changes in its properties. Taking this property into consideration, in the section at the beginning of the secondary refining process (immediately after the start) and the section immediately before the end of the secondary refining process, a nonlinear model such as a multi-layer neural network (deep learning), support vector regression, decision tree, or ensemble learning can be used as a model that can reflect the nonlinearity of temperature changes and predict complex temperature changes by incorporating information about the usage history of the ladle 2, and in the section from the beginning of the secondary refining process onwards where the exothermic reaction due to the material balance of the addition of auxiliary materials and the blowing of oxygen dominates the temperature accuracy, a linear model can be used with an emphasis on the robustness of the model (stability against outliers and noise in the training data).

[0061] The inputs to the molten steel temperature estimation model are operational information for the process preceding the secondary refining process, operational information for the previous section in the secondary refining process, operational schedule information for the section for which the molten steel temperature estimation model is to be constructed, and equipment information for the secondary refining process.

[0062] Here, the pre-process of the secondary refining treatment means a treatment that is carried out before the secondary refining treatment, and may be, for example, a primary refining treatment.

[0063] The operational information for the pre-process of the secondary refining process may include at least one of molten steel information, operational control variable information, and processing time information for the pre-process of the secondary refining process. The molten steel information may include, for example, information such as the weight of the molten steel, the temperature of the molten steel at the end of the pre-process, and the component concentration of the molten steel at the end of the pre-process. The weight of the molten steel is correlated with the heat capacity of the molten steel contained in the ladle 2. The molten steel temperature at the end of the pre-process of the secondary refining process is the initial value of the molten steel temperature in the section immediately after the start of the secondary refining process. Information on the component concentration of the molten steel affects the exothermic reaction caused by the addition of auxiliary materials and oxygen blowing in the secondary refining process, and therefore significantly affects the temperature change during the secondary refining process. The operational control variable information may include, for example, the weight of the auxiliary materials added in the pre-process. The amount of slag in the secondary refining process charge affects the molten steel temperature in the form of heat radiation from the top surface of the ladle 2. However, the amount of slag may increase if auxiliary materials are added in the pre-process to modify the slag. The processing time information may include, for example, information on the time from the end of the previous process to the start of the secondary refining process, because this information significantly affects the amount of heat exchange between the molten steel and the ladle before the start of the secondary refining process, and therefore significantly affects the amount of change in the temperature of the molten steel during the secondary refining process.

[0064] Furthermore, the operational information up to the previous section in the secondary refining process means, for example, operational information in sections 1 and 2 when the molten steel temperature estimation model is a molten steel temperature estimation model constructed for section 3.

[0065] The operation information up to the previous section in the secondary refining process may include at least one of molten steel information, equipment information, operational control amount information, and processing time information up to the previous section. The molten steel information may include, for example, information on the molten steel temperature measured after the end of the previous section and until the start of the current section. By using information on the molten steel temperature before the start of the current section as input information (explanatory variable) to the molten steel temperature estimation model as an initial condition, prediction accuracy can be improved. The operational control amount information may include, for example, information on the amount of auxiliary material input, the amount of oxygen blown, etc. This is because temperature changes associated with operational operations in the previous section may be observed with a delay in the next section. The processing time information may include information on the processing time for each section up to the previous section. This is because the processing time for each section up to the previous section significantly affects the amount of heat exchange between the molten steel and the ladle in the previous section, and therefore significantly affects the amount of heat exchange between the molten steel and the ladle in the next section.

[0066] Furthermore, the operation schedule information for the section for which the molten steel temperature estimation model is constructed is, for example, operation schedule information for section 3 if the molten steel temperature estimation model is a molten steel temperature estimation model constructed for section 3 .

[0067] The operation schedule information for the section for which the molten steel temperature estimation model is to be constructed may include at least one of operation control amount information and processing time information for the section for which the molten steel temperature estimation model is to be constructed. The operation control amount information may include, for example, information such as the planned amount of auxiliary material input and the planned amount of oxygen injection. This is because the input of auxiliary material directly affects the heat balance of the molten steel, and oxygen injection promotes exothermic reactions such as oxidation and decarburization of the molten steel, and therefore these are factors that significantly affect the accuracy of the molten steel temperature estimation model for that section. The processing time information may include, for example, information such as the planned processing time for the section for which the molten steel temperature estimation model is to be constructed.

[0068] The equipment information for the secondary refining process may include, for example, at least one of the following: actual temperature information (ladle surface temperature) on the inner wall surface obtained by measuring when the ladle 2 used for the secondary refining process is empty (empty ladle), the number of times the ladle 2 has been used (the number of times the ladle 2 has received molten steel), the time during which the ladle 2 is empty after the previous molten steel is discharged and before the current molten steel is received (empty ladle time), and the time from the end of the previous secondary refining process to the start of the current secondary refining process (RH tank empty time). These factors affect the amount of heat stored in the secondary refining equipment 1 and ladle 2 used in the secondary refining process, and using them as inputs (explanatory variables) for the model contributes to improving the temperature accuracy in the early stage (immediately after the start) and immediately before the end of the secondary refining process.

[0069] Furthermore, the output of the molten steel temperature estimation model is the amount of change in molten steel temperature from the start to the end of the interval for which the molten steel temperature estimation model is constructed. For example, if the molten steel temperature estimation model is constructed for interval 3, the output is the amount of change in molten steel temperature from the start to the end of interval 3.

[0070] The parameters of the molten steel temperature estimation model may be determined based on data obtained from past operational results. When constructing the molten steel temperature estimation model, all or part of the data obtained from the past operational results may be used. When part of the data is used, which data to use may be determined by applying a known method. The known method may be, for example, cross-validation.

[0071] Furthermore, the parameters of the molten steel temperature estimation model may be updated using newly acquired operational performance data every time the secondary refining process is performed.

[0072] The constructed molten steel temperature estimation model for each section may be stored in the storage unit 14.

[0073] The molten steel temperature estimation method and the molten steel temperature control method according to this embodiment will be described with reference to the flowchart shown in Fig. 5. The process shown in Fig. 5 may be executed in response to an instruction from an operator after the start of the secondary refining process, or may be executed at a predetermined timing.

[0074] Step S101: The control unit 11 of the molten steel temperature control device 10 acquires operational information of the previous process. The control unit 11 may acquire, for example, the operational information of the previous process transmitted by another information processing device via the communication unit 15.

[0075] The operational information of the previous process may include at least one of molten steel information, operational operation amount information, and processing time information in the process previous to the secondary refining treatment.

[0076] Step S102: The control unit 11 acquires the target temperature of the molten steel at the end of the secondary refining process. The control unit 11 may acquire the target temperature input by the operator through an input operation to the input unit 12, or may acquire the target temperature transmitted by another information processing device via the communication unit 15.

[0077] Step S103: The control unit 11 acquires the operation information up to the previous section in the secondary refining process as the operation information during the secondary refining process. For example, the control unit 11 may acquire the operation information up to the previous section transmitted by another information processing device via the communication unit 15.

[0078] The operation information up to the previous section in the secondary refining process may include at least one of molten steel information, equipment information, operation amount information, and processing time information up to the previous section.

[0079] Step S104: The control unit 11 determines which section the current time belongs to. The control unit 11 may determine the section to which the current time belongs, for example, based on the operation information acquired in step S103. Alternatively, the control unit 11 may determine the section to which the current time belongs, for example, based on the time elapsed from the start of the secondary refining process to the current time. Alternatively, the control unit 11 may determine the section to which the current time belongs, for example, based on the content of events that have been executed from the start of the secondary refining process to the current time. The event information may include information on molten steel temperature measurement, molten steel oxygen concentration measurement, molten steel sampling, etc., that have been executed up to the previous section. Hereinafter, the section to which the current time belongs may be referred to as the "current section."

[0080] Step S105: The control unit 11 acquires operation schedule information for the current section. For example, the control unit 11 may acquire the operation schedule information for the current section transmitted by another information processing device via the communication unit 15. Alternatively, the control unit 11 may acquire the operation schedule information for the current section stored in the memory unit 14 from the memory unit 14.

[0081] The operation schedule information for the current section may include at least one of operation amount information and processing time information for the current section.

[0082] Step S106: The control unit 11 inputs the operation information of the previous process acquired in step S101, the operation information up to the previous section in the secondary refining process acquired in step S103, the operation schedule information for the current section acquired in step S105, and the equipment information for the secondary refining process into the molten steel temperature estimation model constructed for the current section, and estimates the amount of change in molten steel temperature in the current section.

[0083] Step S107: The control unit 11 determines whether the current section is the final section in the secondary refining process. If the current section is not the final section (No in step S107), the control unit 11 proceeds to step S108. If the current section is the final section (Yes in step S107), the control unit 11 proceeds to step S109.

[0084] Step S108: The control unit 11 proceeds from the current section to the next section, and again executes the processes of steps S105 and S106. For example, if the current section is section 2, the control unit 11 proceeds from section 2 to section 3, and estimates the amount of change in molten steel temperature in section 3, which is the section after the current section, section 2. In this case, the control unit 11 inputs the operation information of the previous process, the operation information up to section 2, and the operation schedule information for section 3 into the molten steel temperature estimation model constructed for section 3, and estimates the amount of change in molten steel temperature in section 3.

[0085] Step S109: The control unit 11 combines the estimated amount of change in molten steel temperature in the current section with the amount of change in molten steel temperature in sections after the current section to estimate the molten steel temperature at the end of the secondary refining process.

[0086] Step S110: The control unit 11 determines whether the difference between the molten steel temperature at the end of the secondary refining process estimated in Step S109 and the target molten steel temperature acquired in Step S102 is within a predetermined range. If the difference is within the predetermined range (Yes in Step S110), the control unit 11 ends the process. If the difference is not within the predetermined range (No in Step S110), the control unit 11 proceeds to Step S111.

[0087] Step S111: The control unit 11 calculates the molten steel temperature adjustment operation amount for bringing the difference between the estimated molten steel temperature at the end of the secondary refining process and the target molten steel temperature within a predetermined range, and then returns to step S 105. Here, the molten steel temperature adjustment operation amount refers to the amount by which the operation for adjusting the molten steel temperature is to be performed in order to adjust the molten steel temperature.

[0088] For example, if the estimated molten steel temperature at the end of the secondary refining process is higher than the target temperature, the molten steel temperature can be lowered by adding auxiliary materials for cooling to the molten steel. In this case, the control unit 11 calculates the amount of auxiliary materials for cooling that should be added as the molten steel temperature adjustment manipulation variable.

[0089] Furthermore, for example, if the estimated molten steel temperature at the end of the secondary refining process is lower than the target temperature, the molten steel temperature can be increased by injecting oxygen into the molten steel. In this case, the control unit 11 calculates the amount of oxygen that should be injected as the molten steel temperature adjustment manipulated variable.

[0090] The control unit 11 may calculate the molten steel temperature adjustment manipulated variable based on the relationship between the molten steel temperature adjustment manipulated variable and the amount of change in the molten steel temperature, which is stored in advance in the storage unit 14, for example.

[0091] As described above, according to the molten steel temperature estimation method of this embodiment, the amount of change in molten steel temperature in each section is estimated using the molten steel temperature estimation model constructed for each section, so that the amount of change in molten steel temperature for each section can be estimated with high accuracy, and the molten steel temperature at the end of the secondary refining process can be estimated with high accuracy. Furthermore, according to the molten steel temperature control method of this embodiment, the molten steel temperature can be adjusted based on the molten steel temperature adjustment manipulated variable calculated in step S111, so that the molten steel temperature can be controlled with high accuracy.

[0092] (Example) FIG. 6 is a diagram comparing the estimation error of the amount of change in molten steel temperature between a comparative example and this embodiment.

[0093] Fig. 6 shows the results of estimating the molten steel temperature in a vacuum degassing process accompanied by decarburization. In the example shown in Fig. 6, the vacuum degassing process is divided into three sections. Section 1 is the section from the start of the process to the point at which the molten steel temperature is measured immediately before the addition of the deoxidizer. Section 2 is the section from the point at which the molten steel temperature is measured immediately before the addition of the deoxidizer to the point at which the molten steel temperature is measured immediately after the addition of the deoxidizer. Section 3 is the section from the point at which the molten steel temperature is measured immediately after the addition of the deoxidizer to the end of the process.

[0094] The comparative example shows the results of estimating the amount of change in molten steel temperature using a model that estimates the amount of change in molten steel temperature throughout the entire vacuum degassing process. The model used in the comparative example is a linear model. The model used in the comparative example was constructed based on actual data from 590 runs. The amount of change in molten steel temperature at the end of interval 1 and interval 2 was estimated under the assumption that the molten steel temperature changes linearly over time.

[0095] This embodiment shows the results of estimating the amount of change in molten steel temperature using molten steel temperature estimation models constructed for each of Sections 1 to 3. A neural network model was used as the molten steel temperature estimation model for Section 1. A linear model was used as the molten steel temperature estimation model for Section 2. A neural network model was used as the molten steel temperature estimation model for Section 3. The models for Sections 1 to 3 used in this embodiment were constructed based on actual data from 590 runs.

[0096] 6 shows the estimation error between the estimated molten steel temperature and the actually measured molten steel temperature during vacuum degassing treatment. The estimation error is the result of calculating the root mean square error based on the estimated molten steel temperature and the actually measured molten steel temperature for 103 runs.

[0097] 6, it can be seen that the molten steel temperature estimation method according to this embodiment can estimate the amount of change in molten steel temperature with higher accuracy than the comparative example. In particular, at intermediate points such as the end of Section 1 and Section 2, the estimation error in the comparative example is significantly large, and it can be seen that the molten steel temperature estimation method according to this embodiment has significantly higher accuracy than the comparative example.

[0098] As described above, the molten steel temperature estimation method according to this embodiment includes the steps of dividing the period from the start to the end of processing into a plurality of sections, measuring the molten steel temperature at the start of each of the plurality of sections, estimating the amount of change in molten steel temperature in a current section to which the current time belongs and the amount of change in molten steel temperature in sections after the current section based on a molten steel temperature estimation model constructed for each of the plurality of sections, and estimating the molten steel temperature at the end of processing by combining the estimated amount of change in molten steel temperature in the current section and the amount of change in molten steel temperature in sections after the current section. In this way, by dividing the period from the start to the end of processing into a plurality of sections and estimating the amount of change in molten steel temperature for each of the current section and the sections after the current section, the molten steel temperature estimation method according to this embodiment can accurately estimate the amount of change in molten steel temperature for each section and the amount of change in molten steel temperature up to the end of processing.

[0099] Furthermore, the molten steel temperature control method according to this embodiment calculates the molten steel temperature adjustment manipulated variable so that the difference between the molten steel temperature at the end of processing estimated by the molten steel temperature estimation method according to this embodiment and the target molten steel temperature at the end of processing falls within a predetermined range, thereby enabling the molten steel temperature at the end of processing to be controlled with high accuracy.

[0100] The present disclosure is not limited to the above-described embodiments. For example, multiple blocks shown in the block diagrams may be integrated, or one block may be divided. Instead of executing multiple steps shown in the flowcharts in chronological order as described, each step may be executed in parallel or in a different order depending on the processing capacity of the device executing each step, or as needed. Other modifications are possible within the scope of the present disclosure.

[0101] For example, in the above-described embodiment, the molten steel temperature estimation model is constructed as a statistical model based on past operational records, but the present invention is not limited to this. The molten steel temperature estimation model may be constructed as a physical model.

[0102] REFERENCE SIGNS LIST 1 Secondary refining device 2 Ladle 10 Molten steel temperature control device 11 Control unit 12 Input unit 13 Output unit 14 Memory unit 15 Communication unit

Claims

1. A method for estimating the temperature of molten steel in a process for treating molten steel in a steelmaking process, comprising the steps of: dividing a period from the start to the end of the process into a plurality of sections based on at least one time point during the process when the temperature of the molten steel is measured; measuring the temperature of the molten steel at the start of each of the plurality of sections; estimating the amount of change in the temperature of the molten steel in a current section to which the current time belongs and the amount of change in the temperature of the molten steel in sections after the current section, based on a molten steel temperature estimation model constructed for each of the plurality of sections; and estimating the temperature of the molten steel at the end of the process by combining the estimated amount of change in the temperature of the molten steel in the current section and the amount of change in the temperature of the molten steel in sections after the current section.

2. The molten steel temperature estimation method according to claim 1, wherein the parameters of the molten steel temperature estimation model are determined based on past operational results.

3. A molten steel temperature estimation method according to claim 1 or 2, wherein the inputs to the molten steel temperature estimation model are operational information for a process preceding the treatment, operational information for the previous section in the treatment, operation schedule information for the section for which the molten steel temperature estimation model is to be constructed, and equipment information for the treatment, and the output is the amount of change in molten steel temperature from the start to the end of the section for which the molten steel temperature estimation model is to be constructed.

4. A molten steel temperature estimation method as set forth in claim 3, wherein the operational information of a process preceding the treatment includes at least one of molten steel information, operational operation amount information, and treatment time information in the preceding process; the operational information up to one previous section in the treatment includes at least one of molten steel information, operational operation amount information, and treatment time information up to the one previous section; the operational schedule information for the section in which the molten steel temperature estimation model is constructed includes at least one of operational operation amount information and treatment time information for the section in which the molten steel temperature estimation model is constructed; and the equipment information for the treatment includes information on the usage history of a ladle used in the treatment and information on the actual temperature of the ladle.

5. A molten steel temperature estimation method according to claim 4, wherein the process is a secondary refining process, a process preceding the process is a primary refining process, operational information for the process preceding the process includes at least one of the weight of molten steel and the temperature of molten steel at the end of the previous process, operational information for the previous section in the process includes at least information on the temperature of molten steel measured from the end of the previous section until the start of the current section, planned operational information for the section in which the molten steel temperature estimation model is constructed includes information on at least one of the amount of auxiliary material input, the amount of oxygen blown in, and processing time, and equipment information for the process includes at least one of actual temperature information obtained by measuring when a ladle used for the process is empty, the time during which the ladle is empty from the time the molten steel received previously is discharged until the time the molten steel received this time is received, and the time from the end of the previous secondary refining process to the start of the secondary refining process.

6. The molten iron temperature estimation method according to claim 5, wherein the molten steel temperature estimation model is a neural network model in the sections immediately after the start and immediately before the end of the secondary refining process, and a linear regression model in other sections.

7. A molten steel temperature control method, which calculates a molten steel temperature adjustment operation amount for adjusting the molten steel temperature so that the difference between the molten steel temperature at the end of the processing estimated by the molten steel temperature estimation method according to any one of claims 1 to 6 and the target molten steel temperature at the end of the processing falls within a predetermined range.

8. A method for producing molten steel, which comprises adjusting the temperature of molten steel using the method for controlling the temperature of molten steel according to claim 7.

Images