Molten iron component estimation device, component estimation method, and desulfurization method

Abstract

Provided are a molten iron component estimation device, a component estimation method, and a desulfurization method, whereby it is possible to estimate the component of molten iron after desulfurization with high accuracy in the desulfurization of molten iron using a mechanical stirring facility. A component estimation device (14) that estimates the S concentration in molten iron (3) during desulfurization of the molten iron (3) using a mechanical agitation desulfurization facility (1) is provided with: a data input unit (140) that inputs current operation data during the desulfurization; the current operation data at least comprises the temperature of the molten iron (3) measured by a non-contact temperature measuring device (12) after the stirring of the molten iron (3) is started and before the input of the desulfurizing agent (4) is started, namely the temperature of the molten iron in the desulfurization treatment; and a model calculation unit (142) that estimates the S concentration in the molten iron (3) after desulfurization on the basis of current operation data and a machine learning model that represents the relationship between the temperature of the molten iron during desulfurization and the amount of the desulfurizer (4) introduced and the reaction rate constant during desulfurization on the basis of past operation data during desulfurization.

Description

TECHNICAL FIELD

[0001] The present application relates to a molten iron composition estimation device, a composition estimation method, and a desulfurization method. BACKGROUND

[0002] S contained in steel causes deterioration of hot shortness, corrosion resistance, toughness, and workability of the steel, and thus it is required to reduce the S concentration in the steel. Furthermore, in recent years, the demand for high purity and high cleanliness of steel has increased, and countermeasures for further reduction of sulfur in steel, that is, promotion of desulfurization and suppression of resulfurization in molten iron refining treatment are required.

[0003] Desulfurization treatment of molten iron is performed in a pretreatment process of molten iron tapped from a blast furnace and a secondary refining process of molten steel. As for the desulfurization treatment in the pretreatment process of molten iron, a KR process using a mechanical stirring device is widely used, and various methods have been proposed in order to efficiently perform the desulfurization treatment.

[0004] For example, in Patent Literature 1, a composition concentration calculation device and a composition concentration calculation method are proposed which estimate the S concentration in molten iron after refining treatment based on past operation data in the refining treatment of molten iron. This method is a method of estimating a model parameter difficult to measure, such as a reaction rate constant, included in a first-order reaction formula, using a composition concentration estimation model, based on past operation data and the composition of molten iron before the current desulfurization treatment, assuming that the desulfurization reaction of molten iron follows the first-order reaction formula.

[0005] PRIOR ART DOCUMENTS

[0006] PATENT LITERATURE

[0007] Patent Literature 1: Japanese Patent Application Publication No. 2020-15959 SUMMARY

[0008] PROBLEMS TO BE SOLVED BY THE INVENTION

[0009] However, in the method of Patent Literature 1, the S concentration in molten iron is estimated based only on information before the current desulfurization treatment, and a method of controlling the operation conditions using information obtained in the desulfurization treatment is not disclosed. Therefore, it can be impossible to estimate the S concentration in molten iron after the desulfurization treatment with high accuracy. For example, in the case where unexpected changes in the desulfurization treatment, such as a sharp change in the temperature of molten iron, a change in the entrainment state of desulfurizer and molten slag into molten iron, etc., it can be possible that the desulfurization capacity decreases to cause poor desulfurization of molten iron.

[0010] The present application was made in view of the above-described problems, and aims to provide a molten iron composition estimation device, a composition estimation method, and a desulfurization method which can estimate the S concentration as a composition of molten iron after desulfurization treatment with high accuracy in desulfurization treatment of molten iron using a mechanical stirring device.

[0011] Method for solving a problem

[0012] (1) According to one embodiment of the present application, there is provided a molten iron composition estimation device that estimates the S concentration in molten iron in a desulfurization treatment of the molten iron using a mechanical stirring type desulfurization apparatus, the molten iron composition estimation device including: a data input unit that inputs current operation data in the desulfurization treatment, the current operation data including at least a temperature of the molten iron in the desulfurization treatment, i.e., a molten iron temperature, measured using a non-contact temperature measuring device after stirring of the molten iron is started and before addition of a desulfurizer is started; and a model calculation unit that estimates the S concentration in the molten iron after the desulfurization treatment based on a mechanical learning model that represents a relationship between the molten iron temperature in the desulfurization treatment, the amount of the desulfurizer added, and a reaction rate constant in the desulfurization treatment, and the current operation data based on past operation data in the desulfurization treatment.

[0013] (2) The molten iron composition estimation device according to the above (1), wherein the data input unit inputs at least the molten iron temperature in the desulfurization treatment, the amount of the desulfurizer added, and the reaction rate constant as a constant in a desulfurization reaction represented by formula (1) as the past operation data, the molten iron composition estimation device further including a model parameter determination unit that creates the mechanical learning model based on the past operation data input by the data input unit.

[0014] [S] = [S]0·exp{K(t-t0)} …(1)

[0015] Here, [S]: S concentration in the molten iron (mass %), [S]0: S concentration in the molten iron before the start of the desulfurization treatment (mass %), K: reaction rate constant (s -1 ), t: desulfurization treatment time (s), t0: time from the start of stirring to completion of entrainment of the slag into the molten iron (s).

[0016] (3) According to one embodiment of the present application, there is provided a molten iron composition estimation method that estimates the S concentration in molten iron in a desulfurization treatment of the molten iron using a mechanical stirring type desulfurization apparatus, the molten iron composition estimation method including: a step of measuring a temperature of the molten iron in the desulfurization treatment, i.e., a molten iron temperature, after stirring of the molten iron is started and before addition of a desulfurizer is started, using a non-contact temperature measuring device; a step of inputting current operation data in the desulfurization treatment that includes at least the measured molten iron temperature in the desulfurization treatment; and a step of estimating the S concentration in the molten iron after the desulfurization treatment based on a mechanical learning model that represents a relationship between the molten iron temperature in the desulfurization treatment, the amount of the desulfurizer added, and a reaction rate constant in the desulfurization treatment, and the current operation data based on past operation data in the desulfurization treatment.

[0017] (4) The molten iron composition estimation method according to the above (3), wherein, as the past operation data, the molten iron temperature in the desulfurization treatment, the desulfurizer input amount, and the reaction rate constant as a constant in a desulfurization reaction formula represented by formula (1) in the past operation are used, and the process of creating the machine learning model based on the input past operation data is further included.

[0018] [S] = [S]0•exp{K(t-t0)} (1)

[0019] Here, [S]: S concentration in molten iron (mass %), [S]0: S concentration in molten iron before the start of desulfurization treatment (mass %), K: reaction rate constant (s -1 ), t: desulfurization treatment time (s), t0: time from the start of stirring to completion of the inclusion of slag into molten iron (s).

[0020] (5) According to one embodiment of the present application, a molten iron desulfurization method is provided, which is a molten iron desulfurization method using a mechanical stirring type desulfurization apparatus, wherein, in the desulfurization treatment, the S concentration in the molten iron after the desulfurization treatment is estimated using the molten iron composition estimation method according to the above (3) or (4).

[0021] (6) The molten iron desulfurization method according to the above (5), further including a process of determining whether the S concentration in the molten iron after the desulfurization treatment is within a prescribed range after the process of estimating the S concentration, and when the S concentration in the molten iron after the desulfurization treatment is not within the prescribed range, changing the desulfurizer input amount, and again performing the process of estimating the S concentration, and again estimating the S concentration in the molten iron.

[0022] (7) The molten iron desulfurization method according to the above (5) or (6), further including: a process of measuring the temperature of the bath surface of the molten iron after the process of estimating the S concentration; a process of determining the inclusion state of slag on the bath surface of the molten iron based on the measurement results in the process of measuring the temperature of the bath surface of the molten iron; and a process of inputting the desulfurizer into the molten iron when it is determined in the process of determining the inclusion state of the slag that the inclusion of the slag has been completed.

[0023] (8) The molten iron desulfurization method according to any one of (5) to (7) above, further comprising: a step of measuring the temperature of the bath surface of the molten iron after the desulfurizer is charged; and a step of determining the entrainment state of the molten slag on the bath surface of the molten iron based on the measurement result in the step of measuring the temperature of the bath surface of the molten iron, wherein, in the step of determining the entrainment state of the molten slag, when it is determined that the molten slag is properly entrained, stirring by the impeller of the mechanical stirring desulfurization apparatus is continued, and in the step of determining the entrainment state of the molten slag, when it is determined that the molten slag is not properly entrained, at least either the rotation speed or the immersion depth of the impeller is changed.

[0024] Effects of Invention

[0025] According to one embodiment of the present application, a molten iron composition estimation device, a composition estimation method, and a desulfurization method capable of accurately estimating the composition of molten iron after desulfurization treatment using a mechanical stirring device are provided. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a configuration diagram of a mechanical stirring desulfurization apparatus 1 according to one embodiment of the present application.

[0027] Figure 2 is a flowchart of a molten iron desulfurization method according to one embodiment of the present application.

[0028] Figure 3 is a flowchart of a molten iron composition estimation method according to one embodiment of the present application.

[0029] Figure 4 is a flowchart of a method of determining the entrainment state of molten slag after the start of desulfurizer charging and changing the stirring conditions. DETAILED DESCRIPTION

[0030] In the following detailed description, embodiments of the present application are described with reference to the accompanying drawings. In the drawings, the same or similar parts are designated by the same or similar signs, and overlapping descriptions are omitted. Each drawing is schematic, and includes cases different from reality. In addition, the embodiments shown below are examples of devices and methods for embodying the technical idea of the present application, and the technical idea of the present application is not limited to the materials, structures, arrangements, and the like of the constituent parts described below. The technical idea of the present application can be variously changed within the technical scope defined by the claims.

[0031] First, the inventors determined the dosage of desulfurizer based on past and current operating data in the desulfurization treatment of molten iron using a mechanically agitated desulfurization device, added the desulfurizer to the molten iron, and investigated the desulfurization behavior of the molten iron. It was found that by using the temperature of the molten iron before the start of the desulfurizer addition, the S concentration in the molten iron during the desulfurization treatment was estimated based on a machine learning model and a first-order reaction formula that represents the relationship between the temperature of the molten iron, the dosage of the desulfurizer, and the reaction rate constant, thereby estimating the S concentration with high precision. It was also found that by using this estimation method, it was possible to suppress poor desulfurization of the molten iron and excessive addition of the desulfurizer, and to efficiently perform the desulfurization treatment. In the following description, a composition estimation device, a molten iron composition estimation method, and a desulfurization method according to one embodiment of the present invention based on these insights are described.

[0032] <Device Configuration>

[0033] The mechanical stirring desulfurization equipment in one embodiment of the present invention is described. Figure 1 As shown, the mechanical stirring desulfurization equipment 1 is a desulfurization treatment equipment (also called KR desulfurization treatment equipment) that desulfurizes molten iron 3 contained in a molten iron pot 2 by mechanical stirring. The mechanical stirring desulfurization equipment 1 includes an impeller 10, a desulfurizer feeding device 11, a temperature measuring device 12, a control terminal 13, a component estimation device 14, and a determination device 15.

[0034] The impeller 10 is a mechanical agitator having a rotating shaft and a stirring blade made of refractory material at the tip of the rotating shaft. The impeller 10 is raised and lowered in a substantially vertical direction by a lifting device (not shown) and rotated by a rotating device (not shown) consisting of a drive motor and a speed reducer, with the shaft, to which the stirring blade is connected, acting as the rotational axis. The impeller 10 mechanically stirs the molten iron 3 by immersing the stirring blade at the tip of the impeller in the molten iron 3 and rotating it.

[0035] The desulfurizer feeding device 11 is a device for feeding the desulfurizer 4 into the molten iron 3. The desulfurizer 4 is not limited to a specific substance, and a lime flux, a mixture of a lime flux and aluminum oxide, etc. can be used. In addition, the method of feeding the desulfurizer 4 is not limited to a specific method, and a method of feeding the desulfurizer 4 from above the molten iron pot 2 all at once or continuously, a method of spraying the desulfurizer 4 onto the surface of the molten iron 3 together with a carrier gas, etc. can be used.

[0036] The temperature measuring device 12 is a non-contact thermometer disposed above the molten iron pot 2, and measures the temperature of the bath surface of the molten iron 3 contained in the molten iron pot 2. The temperature measuring device 12 is not particularly limited as long as it can measure the temperature of the molten iron 3 and the slag 5 in a non-contact manner. For example, a radiation thermometer, a thermal imager, etc. can be used. In the method of immersing a temperature measuring probe in the molten iron and measuring the temperature of the molten iron 3, there is a risk of contact between the temperature measuring probe and the impeller 10 or the solid phase slag 5 during the desulfurization process, that is, when the impeller 10 is rotating. In this embodiment, by using a non-contact thermometer as the temperature measuring device 12, such a contact risk can be avoided, and the temperature of the molten iron 3 exposed on the bath surface of the molten iron 3 and the slag 5 covering the bath surface of the molten iron 3 can be measured.

[0037] The control terminal 13 controls the desulfurization process in the mechanical stirring desulfurization equipment 1 by controlling the impeller 10 and the desulfurizing agent feeding device 11. Specifically, the control terminal 13 instructs the impeller 10 on stirring conditions such as the rotation speed and immersion depth as the operating conditions of the impeller 10, and obtains the actual results of the stirring conditions. It should be noted that the immersion depth of the impeller 10 refers to the relative Figure 1 The bath surface position of the molten iron 3 indicated by the dotted line in the molten iron pot 2 is such that the impeller 10 is immersed to a depth. In addition, the control terminal 13 instructs the desulfurizer feeding device 11 on the amount and timing of the desulfurizer 4 to be fed, and obtains the actual performance of the amount and timing of the desulfurizer 4. In addition, the control terminal 13 instructs the desulfurizer feeding device 11 on the feeding of the desulfurizer 4 based on the amount of the desulfurizer 4 fed from the component estimation device 14, the determination result of the slag entrainment obtained from the determination device 15, and the like. Furthermore, the control terminal 13 obtains the determination result of the slag entrainment from the determination device 15, and changes the stirring conditions in the desulfurization treatment as needed. Furthermore, past operation data is stored in the control terminal 13. The past operation data is information related to the molten iron 3 from before to after the desulfurization treatment in the past desulfurization treatment and information related to the operating conditions of the mechanical stirring desulfurization equipment 1. Past operation data include, for example, the composition, temperature and weight of the molten iron 3, the weight of the desulfurization slag, the input amount of the desulfurizer 4 and other auxiliary materials, the rotation speed of the impeller 10, the immersion depth and the number of times it is used, the desulfurization treatment time, the time until the slag 5 is completely drawn in, and the number of times the molten iron pot 2 is used. Regarding the temperature of the molten iron 3, in addition to the temperature before the desulfurization treatment, it also includes the molten iron temperature during the desulfurization treatment described later. The time until the slag 5 is completely drawn in refers to the time from the start of stirring (mechanical stirring) of the molten iron 3 by the impeller 10 to the completion of the drawing of the slag 5 into the molten iron 3. Other auxiliary materials are auxiliary materials other than the desulfurizer 4 added to the molten iron 3, for example, iron-containing dust generated in the ironmaking process, used refractory waste, etc.

[0038] The composition estimating device 14 estimates at least the S concentration in the molten iron 3, which is a component of the molten iron 3 after desulfurization, based on past and current operating data and the temperature measurement result of the molten iron 3 obtained by the temperature measuring device 12. The composition estimating device 14 includes a data input unit 140, a model parameter determination unit 141, and a model calculation unit 142. The composition estimation method and each component of the composition estimating device 14 will be described in detail later.

[0039] The determination device 15 is a device for performing various determinations during the desulfurization process, and includes a temperature distribution image generating unit 150, a slag inclusion determination unit 151, and an S concentration determination unit 152. The temperature distribution image generating unit 150 generates a temperature distribution image based on the measurement results of the temperature measuring device 12. The slag inclusion determination unit 151 determines whether the slag 5 is included based on the temperature distribution image generated by the temperature distribution image generating unit 150. The S concentration determination unit 152 determines whether the estimated value of the S concentration in the molten iron 3 after the desulfurization process, estimated by the composition estimation device 14, is within a specified range described later. The determination method and each component of the determination device 15 will be described in detail later.

[0040] It should be noted that the control terminal 13, the component estimation device 14, and the determination device 15 may also be provided in the form of a program executed by a computer, a PLC, or the like.

[0041] <Methods for desulfurizing molten iron>

[0042] The desulfurization method of the molten iron 3 in this embodiment is described. Figure 2 The desulfurization treatment is carried out according to the flowchart shown. It should be noted that, in the present embodiment, the molten iron 3 is tapped from the blast furnace and is received by molten iron holding / transporting containers such as a molten iron pot and a torpedo car. Then, the molten iron pretreatment such as desiliconization and dephosphorization is carried out as needed. Then, the molten iron 3 is transferred to the molten iron pot 2 as needed, and is transported to the mechanical stirring desulfurization equipment 1 while being accommodated in the molten iron pot 2. In addition, before the desulfurization treatment, the composition, weight and temperature of the molten iron 3 are measured, and the measurement results are recorded in the control terminal 13. It should be noted that, in the molten iron pot 2, as needed, auxiliary materials such as desulfurization slag, which is the slag produced in the most recent desulfurization treatment, can also be put into the molten iron pot 2 in advance.

[0043] First, the data input section 140 of the composition estimation device 14 obtains, from the control terminal 13, desulfurization-pre-treatment data that is information of the molten iron 3 before desulfurization treatment (S100). The desulfurization-pre-treatment data is an operation condition related to desulfurization treatment before desulfurization treatment that is measured in advance. The desulfurization-pre-treatment data includes, for example, the composition, temperature, and weight of the molten iron 3, the weight of desulfurization slag, and the number of uses of the iron ladle 2. In addition, the desulfurization-pre-treatment data includes the input amount of the desulfurizer 4 and other sub raw materials, the rotational speed of the impeller 10, the immersion depth and the number of uses, the desulfurization treatment time, and the time until the entrapment of the molten slag 5 is completed as values (initial values) set in advance. Further, it can also be configured such that the desulfurization-pre-treatment data is stored in an upper computer, a server, or the like not shown, and the data input section 140 obtains the desulfurization-pre-treatment data from the upper computer, the server, or the like. Note that step S100 can also be performed after tapping, after pretreatment of the molten iron, or the like before desulfurization treatment.

[0044] Next, the control terminal 13 starts desulfurization treatment by causing the stirring blades of the impeller 10 to be immersed in the molten iron 3 and causing the impeller 10 to rotate (S102). The stirring conditions such as the rotational speed of the impeller 10 and the immersion depth use conditions set in advance.

[0045] Further, the temperature measuring device 12 measures the temperature of the bath surface of the molten iron 3 (S104). Step S104 is performed after a prescribed time elapses from the start of stirring by the impeller 10. The prescribed time is set to a time at which the molten iron 3 sufficiently emerges at the bath surface (upper surface) of the molten iron 3 to an extent that the temperature can be measured, and is appropriately set in accordance with the amount of the molten slag 5, the shape / size of the stirring blades of the impeller 10, the stirring conditions, or the like. In desulfurization treatment of the mechanical stirring type, treatment is usually started in a state in which the bath surface of the molten iron 3 is covered with the molten slag 5 such as desulfurization slag. Therefore, by stirring with the impeller 10 and elapsing of the prescribed time, the molten iron 3 emerges at the bath surface of the molten iron 3, and the temperature of the molten iron 3 can be measured. The temperature measurement result of the molten iron 3 is transmitted to the control terminal 13 and stored. Note that the temperature of the molten iron 3 measured in step S104, that is, the temperature of the molten iron 3 after the start of stirring in desulfurization treatment and before the start of input of the desulfurizer 4 is also referred to as the molten iron temperature in desulfurization treatment.

[0046] Then, the composition estimation device 14 estimates the S concentration in the molten iron 3 after desulfurization treatment (S106). Estimation of the composition (S concentration) of the molten iron 3 by the composition estimation device 14 is performed by the following method.

[0047] (Method of Estimating Composition of Molten Iron)

[0048] In the composition estimation method of molten iron 3 of the present embodiment, as described above, the S concentration in the molten iron 3 after the desulfurization treatment is estimated. In the present embodiment, the S concentration in the molten iron 3 after the desulfurization treatment is estimated on the basis of the temperature of the molten iron 3 measured by the temperature measuring device 12, the data before the desulfurization treatment of the molten iron 3, and the past operation data. Note that hereinafter, the temperature of the molten iron 3 measured by the temperature measuring device 12 (the molten iron temperature during the desulfurization treatment) and the data before the desulfurization treatment of the molten iron 3 are collectively referred to as the current operation data.

[0049] In the composition estimation method of molten iron 3 of the present embodiment, the estimation of the composition of the molten iron 3 is performed in accordance with the flowchart shown in FIG. 8. First, the data input section 140 obtains the current operation data and the past operation data from the control terminal 13 (S200). Figure 3

[0050] Next, the model parameter determination section 141 creates a mechanical learning model representing the relationship between the temperature of the molten iron 3 and the input amount of the desulfurizer 4 and the reaction rate constant K on the basis of the past operation data (S202). Here, the reaction rate constant K is a constant in the desulfurization reaction formula shown in the following (1).

[0051] [S] = [S]0·exp{K(t-t0)}... (1)

[0052] Here, [S]: the S concentration in the molten iron (mass %), [S]0: the S concentration in the molten iron before the start of the desulfurization treatment (mass %), K: the reaction rate constant (s -1 ), t: the desulfurization treatment time (s), t0: the time from the start of the stirring to the completion of the entrainment of the slag into the molten iron (s).

[0053] In the case where the desulfurizer 4 is input in a state where the entrainment of the slag 5 into the molten iron 3 is inappropriate and the surface of the molten iron 3 is covered with the slag, it is possible that the desulfurizer 4 is captured by the slag 5 before reacting with the S in the molten iron 3, and the desulfurization capacity is reduced, resulting in poor desulfurization of the molten iron 3. Therefore, it is important to create a model that takes into account the time t0 from the start of the stirring to the completion of the entrainment of the slag 5 into the molten iron 3. Note that the time t0 from the start of the stirring to the completion of the entrainment of the slag 5 into the molten iron 3 can be determined by continuously measuring the temperature of the bath surface of the molten iron 3 using the temperature measuring device 12 and performing the determination of the entrainment state of the slag 5 described later.

[0054] In addition, the reaction rate constant in the (1) formula is difficult to measure, and therefore it is important to determine it on the basis of the past operation data and the current operation data.

[0055] The kind of the mechanical learning model is not particularly limited, and for example, a model using a neural network or the like can be applied. By so doing, it is possible to determine the reaction rate constant K with high accuracy on the basis of various operation data.​

[0056] Further, the higher the temperature of the molten iron, and further, the more the amount of the desulfurizer 4, the more easily the desulfurization reaction of the molten iron 3 proceeds. Therefore, it is important to input the temperature of the molten iron 3 and the amount of the desulfurizer 4 as explanatory variables in determining the reaction rate constant K.

[0057] After the step S202, the model parameter determining section 141 determines the reaction rate constant K from the machine learning model made in the step S202, the temperature of the molten iron in the desulfurization treatment determined in the step S104, and the data of the molten iron 3 before the desulfurization treatment (S204). In the step S204, it is preferable to determine the reaction rate constant K using at least the initial value of the amount of the desulfurizer 4, the weight of the desulfurization slag, and the amount of the other secondary raw material in the data of the molten iron 3 before the desulfurization treatment.

[0058] Then, the model calculating section 142 estimates the S concentration in the molten iron 3 after the desulfurization treatment on the basis of the desulfurization reaction formula shown by the formula (1) from the reaction rate constant K calculated in the step S204 and the data of the molten iron 3 before the desulfurization treatment (S206). The estimated S concentration in the molten iron 3 after the desulfurization treatment is sent to the S concentration determining section 152 of the determination device 15.

[0059] According to the molten iron 3 component estimation method of the present embodiment, the S concentration in the molten iron 3 after the desulfurization treatment is estimated using the temperature of the molten iron in the desulfurization treatment determined in the step S104. In the estimation of the S concentration in the molten iron 3 after the desulfurization treatment, it is important to estimate the reaction rate constant K in the desulfurization treatment with high accuracy. Further, as a factor greatly affecting the reaction rate constant K, the temperature of the molten iron 3 can be listed. Therefore, as the temperature of the molten iron 3, the actual temperature in the desulfurization treatment is used instead of the temperature determined before the desulfurization treatment, whereby the estimation accuracy can be improved. Here, in the conventional understanding, due to various factors such as the usage state of the molten iron pot 2 in which the molten iron 3 is stored, the temperature of the desulfurization slag to be charged, and the amount of the desulfurization slag to be charged, it is difficult to estimate the temperature of the molten iron in the desulfurization treatment on the basis of the temperature before the desulfurization treatment. In contrast, in the present embodiment, the temperature of the molten iron 3 in the desulfurization treatment is determined in the step S104, and the determined temperature of the molten iron in the desulfurization treatment is used, whereby the reaction rate constant K can be estimated with high accuracy, and the S concentration in the molten iron 3 after the desulfurization treatment can also be estimated with high accuracy.

[0060] After the step S106, the S concentration judging section 152 judges whether the S concentration in the molten iron 3 after the desulfurization treatment estimated in the step S106 is within a prescribed range (S108). The prescribed range of the S concentration is appropriately set to a preferable S concentration after the desulfurization treatment in accordance with the target S concentration of the molten iron 3. Specifically, the upper limit of the prescribed range of the S concentration is set from the viewpoint of quality. On the other hand, the lower limit of the prescribed range of the S concentration is set to an allowable value in consideration of resulfurization and the like in the subsequent refining treatment. The result of the judgment of the step S108 is transmitted to the control terminal 13.

[0061] When it is judged in the step S108 that the S concentration in the molten iron 3 after the desulfurization treatment is not within the prescribed range, the control terminal 13 changes the input amount (predetermined amount) of the desulfurizer 4 (S110). In the step S110, when the estimated S concentration in the molten iron 3 after the desulfurization treatment is greater than the upper limit of the prescribed range, the input amount of the desulfurizer 4 is changed to be greater than that used in the estimation of the immediately preceding step S106. On the other hand, when the estimated S concentration in the molten iron 3 after the desulfurization treatment is less than the lower limit of the prescribed range, the input amount of the desulfurizer 4 is changed to be less than that used in the estimation of the immediately preceding step S106. That is, in the step S108, the input amount of the desulfurizer 4 is changed in such a manner that the S concentration in the molten iron 3 after the desulfurization treatment is within the prescribed range. Note that the adjustment amount of the desulfurizer 4 in the step S108 is not particularly limited, but if the adjustment amount is too large, the S concentration after the desulfurization treatment can not fall within the prescribed range. On the other hand, if the adjustment amount is too small, the time of repeated calculation becomes long, and thus it can take time until the input amount of the desulfurizer 4 is determined. Therefore, it is preferable to set the adjustment amount to an appropriate amount in accordance with the specifications, operation performance, and the like of the mechanical stirring desulfurization apparatus 1.

[0062] After the step S110, the processing after the step S106 is performed again. Note that in the case where the processing of the step S106 is performed after the step S110, the input amount of the desulfurizer 4 used in the estimation of the S concentration in the molten iron 3 is the input amount changed in the step S110.

[0063] That is, in the series of repeated processing of the steps S106 to S110, the adjustment of the input amount of the desulfurizer 4 is performed in such a manner that the estimated S concentration in the molten iron 3 after the desulfurization treatment is within the prescribed range. Thereby, the input amount of the desulfurizer 4 is set to be neither too much nor too little. When the input amount of the desulfurizer 4 is small, the S concentration in the molten iron 3 after the desulfurization treatment is higher than the target upper limit value, and it is necessary to perform the desulfurization treatment again or to additionally perform the desulfurization treatment in the subsequent processing. On the other hand, when the input amount of the desulfurizer 4 is large, there is no problem in the quality of the molten iron 3, but the manufacturing cost becomes a problem.

[0064] When it is determined in step S108 that the S concentration in the molten iron 3 after the desulfurization treatment is within the prescribed range, the control terminal 13 determines the amount of the desulfurizer 4 used in the estimation of the S concentration in the molten iron 3 after the desulfurization treatment in the immediately preceding step S106 as the amount of the desulfurizer 4 actually used (S112).

[0065] After step S112, the temperature measuring device 12 measures the temperature of the bath surface of the molten iron 3 (S114). In step S114, the temperature measuring device 12 measures the temperature of the prescribed region of the bath surface of the molten iron 3, and transmits the measurement result to the determination device 15. Note that the prescribed region of the bath surface refers to a plurality of portions of the bath surface of the molten iron 3, and is a region in which the degree of the entrapment of the slag 5 on the bath surface of the molten iron 3 described later can be confirmed, and is preferably a region of 80% or more of the bath surface area of the molten iron 3.

[0066] After step S114, the determination device 15 determines the state of the entrapment of the slag 5 based on the measurement result in step S114, and determines whether the entrapment of the slag 5 is completed to a degree in which the desulfurizer 4 can be charged (S116). In step S116, the determination device 15 determines the state of the entrapment of the slag 5 into the molten iron 3 based on the temperature difference between the molten iron 3 and the slag 5 on the bath surface of the molten iron 3 from the measurement result of the temperature of the bath surface of the molten iron 3. Since the temperature of the slag 5 is lower than that of the molten iron 3, the state of the entrapment of the slag can be determined by verifying the temperature distribution of the bath surface of the molten iron 3.

[0067] Specifically, the temperature distribution image producing section 150 of the determination device 15 produces a temperature distribution image in a plurality of portions or a prescribed region on the bath surface of the molten iron 3 based on the measurement result of step S114. Next, the slag entrapment determination section 151 calculates the total area of the molten iron exposed portion and the slag covered portion in the produced temperature distribution image, respectively. At this time, the portion in which the measured temperature is equal to or higher than a threshold value X (°C) is defined as the molten iron exposed portion, and the portion in which the measured temperature is lower than the threshold value X (°C) is defined as the slag covered portion, and the total area A m of the molten iron exposed portion and the total area A s of the slag covered portion are calculated, respectively. Note that, regarding the temperature of the slag covered portion, a lower limit value can be further set in consideration of the influence of the impeller 10, the ladle 2, and the like. Then, the slag entrapment determination section 151 determines the state of the entrapment based on the area ratio A m / A s of the molten iron exposed portion to the slag covered portion. At this time, for example, a threshold value Y1 can be set to the area ratio A m / A s , and the determination of whether the entrapment of the slag 5 is completed is performed using the threshold value Y1. Note that the threshold value X and the threshold value Y1 can be arbitrarily set, but the threshold value Y1 is preferably set to 4.0 or more.

[0068] When it is determined in step S116 that the entrapment of the slag 5 is not completed, the process after step S114 is performed again.

[0069] On the other hand, when it is determined in step S116 that the entrapment of the slag 5 is completed, the control terminal 13 controls the desulfurizer feeding device 11 to feed the desulfurizer 4 into the molten iron 3 in the amount determined in step S112 (S118). As described above, when the slag 5 is not sufficiently entrapped, it is possible that the desulfurizer 4 to be fed is captured by the slag 5 and the desulfurizing ability is reduced. However, by determining the entrapment state of the slag 5 in step S116 and feeding the desulfurizer 4 in step S118 after the slag 5 is sufficiently entrapped, it is possible to prevent desulfurization failure. In addition, as a method of determining the entrapment of the slag 5, a method of determining by visual confirmation by an operator can also be considered, but in this method, the determination criteria are dependent on individuals, and thus even under the same operation conditions, there is a deviation in the time until determination. Therefore, there is a concern that the desulfurizing ability is reduced due to entrapment failure of the slag 5, and the desulfurizing treatment time is extended due to delay in the start timing of the desulfurizer 4 feeding.

[0070] After step S118, that is, after the feeding of the desulfurizer 4 is started, the stirring by the impeller 10 is continued, and mechanical stirring for a prescribed time corresponding to the desulfurizing treatment time is performed, and thus the desulfurizing treatment is completed. Note that after the desulfurizing treatment is completed, in order to suppress S pickup from the slag 5 after the desulfurizing treatment into the molten iron 3, as a post-treatment, the slag 5 in the molten iron pot 2 can be removed. In addition, the molten iron 3 after the smelting by such a process is subjected to primary refining treatment in a converter or the like and secondary refining treatment in an LF or RH vacuum degassing device or the like, and is formed into a steel billet or the like steel raw material by a continuous casting method or a bloom-casting and blooming method or the like.

[0071] Here, after step S118, that is, after the feeding of the desulfurizer 4 is started, further Figure 4 the determination of the slag entrapment and the adjustment of the stirring conditions shown in the flowchart. In Figure 4 In the flowchart shown in FIG. 10, first, the temperature measuring device 12 measures the temperature of the bath surface of the molten iron 3 (S300). The measurement of step S300 can be performed similarly to step S114.

[0072] Next, the determination device 15 determines the slag entrapment state based on the measurement result in step S300, and determines whether the entrapment of the slag 5 is appropriate (S302). The determination of step S302 can be performed similarly to the determination of step S116. That is, based on the measurement result of step S300, the temperature distribution image is created by the temperature distribution image creation section 150, and based on the area ratio A m / A sThe entrainment of the slag 5 is determined by the slag entrainment determination section 151. At this time, the slag entrainment determination section 151 preferably sets a value larger than the threshold value Yl as the area ratio A m / A s the threshold value Y2.

[0073] In step S302, when the entrainment of the slag 5 is determined to be inappropriate, the control terminal 13 obtains the determination result and changes the stirring condition (S304). At this time, as the stirring condition, at least one of the rotational speed of the impeller 10 and the immersion depth is changed in such a manner that the slag 5 is further entrained. Note that in step S304, the stirring condition is changed in such a manner that the slag 5 is further entrained. That is, when the rotational speed of the impeller 10 is changed, the change of the stirring condition is performed in such a manner that the rotational speed is increased, and when the immersion depth of the impeller 10 is changed, the change of the stirring condition is performed in such a manner that the immersion depth is shallower.

[0074] The desulfurizer 4 that is put into the molten iron 3 reacts with S in the molten iron 3 and floats on the surface of the molten iron 3 to be captured by the slag 5. Therefore, as the cumulative amount of the desulfurizer 4 put in increases, the amount of the slag in the molten iron pot 2 also increases. Therefore, even if the stirring condition is kept constant, as the cumulative amount of the desulfurizer 4 put in increases, the entrainment state of the slag 5 into the molten iron 3 sometimes becomes inappropriate. With regard to the control of the rotational speed of the impeller 10, the entrainment of the slag 5 into the molten iron 3 can be promoted by increasing the rotational speed. However, if the rotational speed of the impeller 10 is excessively increased, it can lead to equipment failure caused by overload of the impeller control device, or exacerbation of the refractory material consumption of the impeller 10. With regard to the control of the immersion depth of the impeller 10, it is preferable to control in such a manner that the front end of the vortex formed on the surface of the molten iron 3 by the mechanical stirring is a height below the upper end of the impeller 10.

[0075] On the other hand, in step S302, when the entrainment of the slag 5 is determined to be appropriate, the determination of the slag entrainment and the adjustment of the stirring condition end. In this case, the control terminal 13 obtains the determination result and ends the process. Figure 4 In the process shown in FIG. 6, when the entrainment of the slag 5 is determined to be insufficient, the stirring condition is changed to make the desulfurization efficiency higher. Thereby, it is possible to stably improve the desulfurization efficiency.

[0076] <Varied Examples>

[0077] The present application has been described above with reference to specific embodiments. The description is not intended to limit the application. Those skilled in the art who have the benefit of this description can obviously think of changes and modifications without departing from the spirit of the present application, and it is intended that such changes and modifications be included within the scope of the application encompassing various embodiments disclosed herein and further embodiments to be derived from the present application. Accordingly, the embodiments of the application set forth in the claims should be construed in the broadest sense and not limited to the described embodiments. It is intended that the embodiments encompassing these changes and modifications included in the scope of the claims be encompassed by the present application.

[0078] For example, in the above-described embodiment, as the component estimation of the molten iron 3, the S concentration after the desulfurization treatment is assumed to be estimated, but the present application is not limited to this example. For example, the S concentration in the molten iron 3 during the desulfurization treatment after a prescribed time can also be estimated.

[0079] In addition, in the above-described embodiment, the process of creating a mechanical learning model of step S202 is assumed to be performed after the start of the desulfurization treatment, but the present application is not limited to this example. The process of step S202 can also be performed in advance before the start of the desulfurization treatment.

[0080] Further, in the above-described embodiment, the input amount of the desulfurizer 4 is assumed to be changed in such a manner that the estimated value of the S concentration in the molten iron 3 after the desulfurization treatment is within a prescribed range, but the present application is not limited to this example. For example, the input amount of the desulfurizer 4 can be made constant and the desulfurization treatment time can be changed to adjust in such a manner that the estimated value of the S concentration in the molten iron 3 after the desulfurization treatment is within a prescribed range. In an environment where there is a margin in the operation, such as when there is a margin in time due to a failure, when the production capacity is decreased, and the like, it is considered that there is also a margin in the desulfurization treatment time. In this case, there is a possibility that the desulfurization treatment time can be adjusted, and thus by making the input amount of the desulfurizer 4 constant and changing the desulfurization treatment time, there is a possibility that the input amount of the desulfurizer 4 can be reduced and the production cost can be reduced.

[0081] <Effects of Embodiments>

[0082] (1) The component estimation device 14 of one embodiment of the present application is a component estimation device 14 that estimates the S concentration in the molten iron 3 in the desulfurization treatment of the molten iron 3 using a mechanical stirring type desulfurization apparatus 1, and includes:

[0083] a data input unit 140 that inputs current operation data during the desulfurization treatment, the current operation data including at least the temperature of the molten iron 3, i.e., the molten iron temperature during the desulfurization treatment, measured using a non-contact temperature measuring device 12 after the start of stirring of the molten iron 3 and before the start of the input of the desulfurizer 4; and

[0084] a model calculation unit 142 that estimates the S concentration in the molten iron 3 after the desulfurization treatment from a mechanical learning model representing the relationship between the molten iron temperature during the desulfurization treatment and the input amount of the desulfurizer 4 and the reaction rate constant during the desulfurization treatment based on past operation data during the desulfurization treatment and the current operation data.

[0085] According to the above-described (1), by using the molten iron temperature during the desulfurization treatment actually measured, a mechanical learning model with high precision can be used, and thus the S concentration in the molten iron 3 after the desulfurization treatment can be estimated with high precision. Thus, in the desulfurization treatment using the mechanical stirring type desulfurization apparatus 1, the generation of desulfurization failure of the molten iron 3 can be suppressed and the desulfurization treatment can be efficiently performed.

[0086] (2) The configuration according to the above (1), wherein the data input section 140 inputs, as the past operation data, the temperature of the molten iron 3 measured using the non-contact temperature measuring device 12 before the start of the addition of the desulfurizer 4 in at least the past operation, the amount of the desulfurizer added, and the reaction rate constant as a constant in the desulfurization reaction formula shown by the formula (1),

[0087] The molten iron composition estimation device 14 further has a model parameter determination section 141 that creates a machine learning model based on the past operation data input by the data input section 140.

[0088] According to the configuration according to the above (2), the reaction rate constant K can be estimated with high precision. In addition, by taking into account the time to from the start of the stirring to the completion of the entrainment of the molten slag 5 into the molten iron 3 t0, the influence of the decrease in the desulfurization efficiency due to the insufficient entrainment of the molten slag 5 can be excluded, and the S concentration in the molten iron 3 after the desulfurization treatment can be estimated with high precision.

[0089] (3) A molten iron composition estimation method according to one embodiment of the present application is a method of estimating the S concentration in the molten iron 3 in a desulfurization treatment of the molten iron 3 using a mechanical stirring type desulfurization device 1, and includes:

[0090] a step of measuring the temperature of the molten iron 3 after the start of the stirring of the molten iron 3 and before the start of the addition of the desulfurizer 4, i.e., the molten iron temperature in the desulfurization treatment, using a non-contact temperature measuring device 12 (S104);

[0091] a step of inputting current operation data in the desulfurization treatment including at least the measured molten iron temperature in the desulfurization treatment (S200, S106); and

[0092] a step of estimating the S concentration in the molten iron 3 after the desulfurization treatment based on a machine learning model representing the relationship between the molten iron temperature in the desulfurization treatment and the amount of the desulfurizer added and the reaction rate constant in the desulfurization treatment based on past operation data in the desulfurization treatment and the current operation data (S206, S106).

[0093] According to the configuration according to the above (3), the same effects as those of the configuration according to the above (1) can be obtained.

[0094] (4) The configuration according to the above (3), wherein, as the past operation data, the temperature of the molten iron 3 measured using the non-contact temperature measuring device 12 before the start of the addition of the desulfurizer 4 in at least the past operation, the amount of the desulfurizer added, and the reaction rate constant K as a constant in the desulfurization reaction formula shown by the formula (1) are used,

[0095] In the inputting operation data process, at least the molten iron temperature in the desulfurization treatment, the desulfurizer 4 input amount, and the reaction rate constant K as a constant in the desulfurization reaction formula shown in formula (1) in the past operation are inputted as the past operation data.

[0096] The molten iron component estimation method further includes a process of creating a machine learning model based on the inputted past operation data (S202).

[0097] According to the configuration of the above (4), the same effects as those of the configuration of the above (2) can be obtained.

[0098] (5) The molten iron 3 desulfurization method of one embodiment of the present application is a molten iron 3 desulfurization method using a mechanical stirring type desulfurization apparatus 1, in which, in the desulfurization treatment, the S concentration in the molten iron 3 after the desulfurization treatment is estimated using the molten iron 3 component estimation method described in the above (3) or (4).

[0099] According to the configuration of the above (5), the same effects as those of the configuration of the above (1) can be obtained.

[0100] (6) According to the configuration of the above (5), the process further includes a process of determining whether the S concentration in the molten iron 3 after the desulfurization treatment is within a prescribed range (S108) after the process of estimating the S concentration (S206, S106),

[0101] When the S concentration in the molten iron 3 after the desulfurization treatment is not within the prescribed range, the desulfurizer 4 input amount is changed, and the process of estimating the S concentration is performed again (S110, S106), and the S concentration in the molten iron 3 is estimated again.

[0102] According to the configuration of the above (6), generation of desulfurization failure can be suppressed. In addition, excessive input of the desulfurizer 4 can be prevented, and thus the cost of the desulfurization treatment can be reduced, and the extension of the desulfurization treatment time can be suppressed.

[0103] (7) According to the configuration of the above (5) or (6), the process further includes:

[0104] a process of measuring the temperature of the bath surface of the molten iron 3 (S114) after the process of estimating the S concentration (S206, S106);

[0105] a process of determining the entrapment state of the molten slag 5 on the bath surface of the molten iron 3 (S116) based on the measurement result in the process of measuring the temperature of the bath surface of the molten iron 3 (S114); and

[0106] a process of inputting the desulfurizer 4 into the molten iron 3 (S118) when it is determined in the process of determining the entrapment state of the molten slag 5 (S116) that the entrapment of the molten slag 5 is completed.

[0107] According to the above (7), the desulfurizer 4 can be reliably charged after the slag 5 is entrained, and thus a decrease in desulfurization efficiency can be prevented.

[0108] (8) The configuration according to any one of the above (5) to (7), further comprising:

[0109] a step (S300) of measuring the temperature of the bath surface of the molten iron 3 after the desulfurizer 4 is charged; and

[0110] a step (S302) of determining the entrainment state of the slag 5 on the bath surface of the molten iron 3 based on a measurement result in the step (S300) of measuring the temperature of the bath surface of the molten iron 3,

[0111] in the step (S302) of determining the entrainment state of the slag 5, when it is determined that the slag 5 is properly entrained, the stirring by the impeller 10 of the mechanical stirring desulfurization apparatus 1 is continued,

[0112] in the step (S302) of determining the entrainment state of the slag 5, when it is determined that the slag 5 is not properly entrained, at least either one of the rotational speed and the immersion depth of the impeller 10 is changed (S304).

[0113] According to the above (8), even when the cumulative amount of the desulfurizer 4 charged increases, the entrainment state of the slag 5 can be properly controlled, and a decrease in desulfurization efficiency can be prevented.

[0114] Embodiment

[0115] An embodiment performed by the present inventors will be described. In the embodiment, in a real machine in which the amount of molten iron per one charge is about 200 tons, the molten iron 3 tapped from a blast furnace is accommodated in the molten iron ladle 2, and then desiliconization and dephosphorization treatments are performed as necessary. Here, the S concentration in the molten iron 3 after the desiliconization and dephosphorization treatments is 0.0300 mass% or less. Next, the molten iron ladle 2 in which the molten iron 3 is accommodated is transported to the mechanical stirring desulfurization apparatus 1, and a desulfurization treatment is performed. At this time, various changes are made to the stirring conditions of the impeller 10 and the charging conditions of the desulfurizer 4, and the like, and the operation is performed. Note that, as the desulfurizer 4, burned lime having a particle diameter of 0.5 mm or less is used. In addition, the method of charging the desulfurizer 4 is set to a method in which the desulfurizer 4 is blown together with a carrier gas to the surface of the molten iron 3.

[0116] In the operation of the embodiment, when the desulfurizer 4 is charged into the molten iron 3 of the components and temperature shown in Table 1, the S concentration in the desulfurized molten iron 3 is estimated using the component estimation device 14 of the above-described embodiment, and the amount of the desulfurizer 4 to be charged is determined so as to be below the estimated S concentration. In estimating the S concentration in the desulfurized molten iron 3, the past operation data and the current operation data are used, and the S concentration in the molten iron during the desulfurization treatment is estimated based on the machine learning model and the first-order reaction formula that represent the relationship between the temperature of the molten iron 3 and the amount of the desulfurizer 4 to be charged and the reaction rate constant K. Then, the desulfurizer 4 is charged into the molten iron 3 at the determined amount, and thus the desulfurization treatment is performed.

[0117] In addition, in Comparative Examples 1 to 3, instead of using the machine learning model and the first-order reaction formula, a multiple regression model that represents the relationship between the temperature of the molten iron 3 and the amount of the desulfurizer 4 to be charged and the S concentration in the desulfurized molten iron 3, which is constructed based only on the past operation data, is used to determine the amount of the desulfurizer 4 to be charged. In addition, in Comparative Examples 4 to 9, when the amount of the desulfurizer 4 to be charged is determined similarly to the above-described embodiment, the temperature of the molten iron 3 before the desulfurization treatment is used instead of the temperature of the molten iron during the desulfurization treatment.

[0118] In addition, in Examples 4 to 6 and 10 to 12, the processing of Steps S114 and S116 is performed, the state of the inclusion of the molten slag 5 into the molten iron 3 is determined, and the charging of the desulfurizer 4 is started after the inclusion of the molten slag 5 into the molten iron 3 is completed. Note that, in the other examples and comparative examples, the charging of the desulfurizer 4 is started before the inclusion of the molten slag 5 into the molten iron 3 is completed.

[0119] Further, in Examples 7 to 12, the processing of Steps S300, S302, and S304 is performed, the state of the inclusion of the molten slag 5 into the molten iron 3 is determined after the charging of the desulfurizer 4 is started, and the stirring conditions are changed when the inclusion of the molten slag 5 into the molten iron 3 is inappropriate. The changed stirring conditions are used in the control of the stirring conditions shown in Table 1.

[0120] Further, regarding the components of the obtained molten iron 3, the results of the investigation in the molten iron ladle 2 after the desulfurization treatment are collectively described in Table 1. Note that the contact risk in Table 1 is the contact risk of the temperature measuring probe with the solid-phase molten slag, and the case where a non-contact type thermometer is used is set to "none", and in the case where a contact type thermometer is used, the case where the temperature measuring probe is in contact with the solid-phase molten slag and cannot be used continuously for three times or more is set to "high". Further, the desulfurization rate in Table 1 is a value obtained by expressing the difference between the S concentration in the molten iron 3 before the desulfurization treatment and the S concentration in the molten iron 3 after the desulfurization treatment as a percentage with respect to the S concentration in the molten iron 3 before the desulfurization treatment.

[0121]

[0122] It is confirmed from the results shown in Table 1 that, under the conditions in which the S concentration in the molten iron 3 after the desulfurization treatment is estimated using the component estimation device 14 of the above-described embodiment, and the input amount of the desulfurizer 4 is determined so that the estimated S concentration is below the target value, the desulfurization rate is 80% or more.

[0123] Further, Examples 4 to 6 are conditions in which the temperature during the desulfurization treatment is measured using the temperature measuring device 12, the entrapment state of the slag 5 is determined before the desulfurizer 4 is input, and the desulfurizer 4 is input after the entrapment of the slag 5 into the molten iron 3 is completed. It is confirmed that, under this condition, the desulfurization effect is further improved compared to Examples 1 to 3, and the desulfurization rate becomes 90% or more.

[0124] Further, Examples 7 to 9 are conditions in which the temperature during the desulfurization treatment is measured using the temperature measuring device 12, the entrapment state of the slag 5 is determined after the input of the desulfurizer 4 is started, and the stirring conditions are changed when the entrapment of the slag 5 into the molten iron 3 is inappropriate. It is confirmed that, under this condition, the desulfurization effect is further improved compared to Examples 1 to 3, and the desulfurization rate becomes 90% or more.

[0125] Further, Examples 10 to 12 are conditions in which the temperature during the desulfurization treatment is measured using the temperature measuring device 12, the entrapment state of the slag 5 is determined before the desulfurizer 4 is input, and the entrapment state of the slag 5 is determined after the input of the desulfurizer 4 is started. Further, under this condition, the desulfurizer 4 is input after the entrapment of the slag 5 into the molten iron 3 is completed, and the stirring conditions are changed when the entrapment of the slag 5 into the molten iron 3 is inappropriate. It is confirmed that, under this condition, the desulfurization effect is further improved compared to Examples 1 to 3, and the desulfurization rate becomes 90% or more, and the desulfurization rate is further increased compared to Examples 4 to 9.

[0126] On the other hand, it is confirmed that, under the conditions in which the input amount of the desulfurizer 4 is determined using a multiple regression model constructed based only on past operation data (Comparative Examples 1 to 3), the input amount of the desulfurizer 4 is insufficient, and thus the desulfurization rate is less than 60%.

[0127] Further, it is confirmed that, when the temperature of the molten iron 3 is measured using the non-contact temperature measuring device 12 before the stirring is started (Comparative Examples 4 to 6), although the desulfurization rate is equivalent to the results shown in Examples 1 to 3 of Table 1, the input amount of the desulfurizer 4 becomes excessive. This is because, in the temperature measuring device 12, the temperature of the slag 5 present on the bath surface of the molten iron 3 is actually measured, and the temperature measurement result of the molten iron 3 is a lower temperature than the actual temperature.

[0128] Further, it was confirmed that when the temperature of the molten iron 3 was measured using a contact type thermometer before the stirring was started (Comparative Examples 7 to 9), although the desulfurization rate and the amount of the desulfurizer 4 to be added were equivalent to the results shown in Examples 1 to 3, the risk of contact of the temperature measuring probe with the solid phase slag was high. Therefore, it was known that it was difficult to continuously use the contact type thermometer.

[0129] Explanation of symbols

[0130] 1 Mechanical stirring type desulfurization apparatus

[0131] 10 Impeller

[0132] 11 Desulfurizer adding device

[0133] 12 Temperature measuring device

[0134] 13 Control terminal

[0135] 14 Component estimation device

[0136] 140 Data input section

[0137] 141 Model parameter determination section

[0138] 142 Model calculation section

[0139] 15 Determination device

[0140] 150 Temperature distribution image production section

[0141] 151 Molten slag entrainment determination section

[0142] 152 S concentration determination section

[0143] 2 Molten iron pot

[0144] 3 Molten iron

[0145] 4 Desulfurizer

[0146] 5 Molten slag

Claims

1. A molten iron composition estimating device for estimating the sulfur concentration in molten iron during desulfurization treatment of molten iron using a mechanical stirring desulfurization facility, comprising: a data input unit for inputting current operation data in the desulfurization treatment, wherein the current operation data at least includes the temperature of the molten iron measured by a non-contact temperature measuring device after the start of stirring of the molten iron and before the start of adding a desulfurizer, i.e., the temperature of the molten iron in the desulfurization treatment; and A model calculation unit estimates the S concentration in the molten iron after desulfurization treatment based on a machine learning model that represents the relationship between the molten iron temperature in the desulfurization treatment, the amount of desulfurizer added, and the reaction rate constant in the desulfurization treatment based on past operation data in the desulfurization treatment and the current operation data.

2. The molten iron composition estimating device according to claim 1, wherein: The data input unit inputs at least the molten iron temperature in the desulfurization treatment, the amount of the desulfurizer added, and the reaction rate constant as a constant in the desulfurization reaction formula shown in formula (1) in the past operation as the past operation data, The molten iron composition estimation device further includes a model parameter determination unit that creates the machine learning model based on the past operation data input by the data input unit. [S]=[S]0·exp{K(t-t0)} …(1) Here, [S]: S concentration in molten iron (mass %), [S]0: S concentration in molten iron before desulfurization (mass %), K: reaction rate constant (s -1 ), t: desulfurization treatment time (s), t0: time from the start of stirring to the completion of slag inclusion into molten iron (s).

3. A method for estimating the composition of molten iron, wherein the method is for estimating the sulfur concentration in the molten iron during desulfurization treatment of the molten iron using a mechanical stirring desulfurization device, comprising: The step of measuring the temperature of the molten iron after the start of stirring of the molten iron and before the start of adding the desulfurizer, that is, the temperature of the molten iron during the desulfurization treatment, using a non-contact temperature measuring device; a step of inputting current operation data of the desulfurization process including at least the measured molten iron temperature during the desulfurization process; as well as A process for estimating the S concentration in the molten iron after desulfurization treatment based on a machine learning model that represents the relationship between the molten iron temperature in the desulfurization treatment, the amount of desulfurizer added, and the reaction rate constant in the desulfurization treatment based on past operation data in the desulfurization treatment, and the current operation data.

4. The method for estimating the composition of molten iron according to claim 3, wherein: As the past operation data, at least the molten iron temperature in the desulfurization treatment, the input amount of the desulfurizer, and the reaction rate constant as a constant in the desulfurization reaction formula shown in formula (1) in the past operation are used. The method for estimating the composition of molten iron further comprises the step of creating the machine learning model based on the input past operation data. [S]=[S]0·exp{K(t-t0)} …(1) Here, [S]: S concentration in molten iron (mass %), [S]0: S concentration in molten iron before desulfurization (mass %), K: reaction rate constant (s -1 ), t: desulfurization treatment time (s), t0: time from the start of stirring to the completion of slag inclusion into molten iron (s).

5. A method for desulfurizing molten iron using a mechanical stirring desulfurization device, wherein: During the desulfurization treatment, the method for estimating the composition of molten iron according to claim 3 or 4 is used to estimate the S concentration in the molten iron after the desulfurization treatment.

6. The method for desulfurizing molten iron according to claim 5, wherein: The method further comprises, after the step of estimating the S concentration, a step of determining whether the S concentration in the molten iron after desulfurization is within a predetermined range. When the S concentration in the molten iron after the desulfurization treatment is not within a predetermined range, the amount of the desulfurizer added is changed, and the step of estimating the S concentration is performed again to estimate the S concentration in the molten iron again.

7. The method for desulfurizing molten iron according to claim 5 or 6, wherein: Also features: a step of measuring the temperature of the bath surface of the molten iron after the step of estimating the S concentration; a step of determining a state of slag entanglement on the bath surface of the molten iron based on a measurement result in the step of measuring the temperature of the bath surface of the molten iron; and A step of adding a desulfurizing agent into the molten iron when it is determined in the step of determining the entanglement state of the slag that the entanglement of the slag is completed.

8. The method for desulfurizing molten iron according to any one of claims 5 to 7, wherein: Also features: The step of measuring the temperature of the bath surface of the molten iron after adding the desulfurizer; and a step of determining the entanglement state of slag on the bath surface of the molten iron based on the measurement result in the step of measuring the temperature of the bath surface of the molten iron, In the step of determining the entanglement state of the slag, when it is determined that the slag is properly entangled, stirring is continued by the impeller of the mechanical stirring desulfurization equipment. In the step of determining the entanglement state of the slag, when it is determined that the slag is not appropriately entangled, at least one of the rotation speed and the immersion depth of the impeller is changed.

Images