JP2021110012A - Estimation apparatus of phosphorus concentration in molten steel, statistical model construction apparatus, estimation method of phosphorus concentration in molten steel, statistical model construction method, and program - Google Patents

Abstract

To accurately estimate phosphorus concentration in molten steel by operation of an MURC method.SOLUTION: An estimation apparatus of phosphorus concentration in molten steel includes a dephosphorization rate constant calculation part and a phosphorus concentration estimation part. The dephosphorization rate constant calculation part calculates an estimated value of the dephosphorization rate constant of the molten steel by using a statistical model using molten steel data relating to the molten steel to be blown in a converter and data extracted from operation data acquired during blowing treatment as explanatory variables. The phosphorus concentration estimation part estimates the phosphorus concentration in the molten steel during the blowing treatment based on the estimated value of the phosphorus concentration and the dephosphorization rate constant before the blowing treatment of the molten steel. The blowing treatment includes a preliminary treatment step including dephosphorization treatment of molten iron, an intermediate slag discharge step performed after the preliminary treatment step, and a decarburization treatment step performed after the intermediate slag discharge step. The explanatory variable includes at least one data representing the amount of slag discharged from the converter in the intermediate slag discharge step.SELECTED DRAWING: Figure 1

Description

The present invention relates to a molten steel medium phosphorus concentration estimation device, a statistical model construction device, a molten steel medium phosphorus concentration estimation method, a statistical model construction method, and a program.

In converter smelting, control of the components in molten steel at the time of blow-off, especially control of phosphorus concentration in molten steel, is important for quality control of steel. In order to control the phosphorus concentration in molten steel, for example, the top-blown lance height, top-blown oxygen flow rate, bottom-blown gas flow rate, and the amount and timing of input of auxiliary materials such as quicklime or scale are used as the operation amount. .. These manipulated quantities are often determined based on information obtained prior to the start of smelting, such as target phosphorus concentration, hot metal data and criteria created based on past operational performance.

However, when the operation amount is determined based only on the information obtained before the start of blowing, the reproducibility of the dephosphorization behavior in actual blowing is low even under the same operating conditions, and the molten steel at the time of blowing is stopped. There was a problem that the variation in phosphorus concentration became large. If the phosphorus concentration in molten steel can be estimated sequentially with high accuracy using the operation data during the blowing operation, it is possible to suppress the variation in the phosphorus concentration by performing additional operations to reach the target phosphorus concentration during the blowing operation. become.

A technique for sequentially estimating the phosphorus concentration in molten steel using operation data during a blowing operation is described in, for example, Patent Document 1. Patent Document 1 describes a technique for estimating the dephosphorization rate constant using the operating conditions related to smelting and measured values related to exhaust gas, and estimating the phosphorus concentration in molten steel during smelting using the estimated dephosphorization rate constant. Is described. Further, in Patent Document 1, the estimated phosphorus concentration in molten steel is compared with the target phosphorus concentration in molten steel, and the phosphorus concentration in molten steel is controlled by changing the operating conditions related to blowing based on the comparison result. The technology is also described.

Japanese Unexamined Patent Publication No. 2013-23696

In recent years, the development of a technology called the multi-function converter (MURC: MULti Refining Converter) method, which enables the hot metal pretreatment and decarburization treatment to be performed consistently in the same converter in the primary refining, has been promoted. .. In the MURC method, a hot metal pretreatment step (second step, Blow1) including a step of charging the hot metal into a converter (first step) and a dephosphorization treatment by adding flux and blowing oxygen with a top blowing lance, An intermediate discharge step (third step) for discharging the slag generated in the second step, and then a decarburization treatment step (fourth step, Blow2) performed in the same converter are carried out. The MURC method has less heat loss and shorter lead time than the conventional simple refining process (SRP), in which dephosphorization and decarburization are performed in different converters. Due to its short length, it has the advantage of high production efficiency in the steelmaking process.

As described above, in the MURC method, the slag generated in the hot metal pretreatment step (second step) including the dephosphorization treatment is discharged in the intermediate slag removal step (third step). At this time, the amount of slag discharged differs depending on the operation depending on the situation of the dephosphorization treatment. In the decarburization treatment step (fourth step), phosphorus contained in the hot metal after the intermediate slag removal step (third step) is desorbed from the hot metal by the dephosphorization reaction represented by the following formula and taken into the slag, or vice versa. It is separated from the slag and taken into the hot metal again. In the following formula, () means the substance in the slag, and [] means the substance in the hot metal.
3 (CaO) + 5 (FeO) + 2 [P] = (3CaO · P ₂ O ₅ ) + 5 [Fe]

The direction and speed at which the reaction of the above formula proceeds varies depending on the amount of slag remaining in the converter in the decarburization treatment step (fourth step). Here, the amount of slag remaining in the converter changes according to the amount of slag discharged in the intermediate slag step (third step). Therefore, in the operation by the MURC method, it is considered that the amount of slag discharged in the intermediate discharge step (third step) affects the phosphorus concentration in the molten steel during the decarburization treatment. However, in Patent Document 1 described above, the amount of slag discharged in the intermediate slag removal step is not taken into consideration in estimating the phosphorus concentration in the molten steel.

Therefore, the present invention is an apparatus for estimating the phosphorus concentration in molten steel, which can accurately estimate the phosphorus concentration in molten steel by the operation by the MURC method by considering the amount of slag discharged in the intermediate slag removal step. It is an object of the present invention to provide a statistical model construction device, a method for estimating the concentration of phosphorus in molten steel, a method for constructing a statistical model, and a program.

According to a certain aspect of the present invention, the hot metal is prepared by using a statistical model in which the hot metal data related to the hot metal blown in the converter and the data extracted from the operation data acquired during the hot metal blowing process are used as explanatory variables. The phosphorus dephosphorization rate constant calculation unit that calculates the estimated value of the dephosphorization rate constant and the phosphorus concentration in the molten steel during the blowing process are estimated based on the phosphorus concentration before the hot metal blowing process and the estimated value of the dephosphorization rate constant. A phosphorus concentration estimation unit is provided, and the blowing treatment includes a pretreatment step including a dephosphorization treatment of hot metal, an intermediate discharge step carried out after the preliminary treatment step, and a decarburization treatment step carried out after the intermediate discharge step. An explanatory variable including at least one data representing the amount of slag discharged from the converter in the intermediate discharge step is provided for a phosphorus concentration estimation device in molten steel.

According to another aspect of the present invention, the hot metal data regarding the hot metal blown by the converter in the past operation, the operation data acquired during the blowing process, and the actual value of the phosphorus concentration in the molten steel are collected. It is equipped with a data collection unit and a statistical model construction unit that builds a statistical model that uses the hot metal data and the data extracted from the operation data as explanatory variables and the hot metal dephosphorization rate constant as the objective variable. The explanatory variables include the pretreatment step including the dephosphorization treatment, the intermediate discharge step carried out after the preliminary treatment step, and the decarburization treatment step carried out after the intermediate discharge step, and the explanatory variables are changed in the intermediate discharge step. A statistical model building device is provided that contains at least one piece of data representing the amount of slag discharged from the furnace.

According to yet another aspect of the present invention, a statistical model is used in which the hot metal data related to the hot metal blown in the converter and the data extracted from the operation data acquired during the blowing process are used as explanatory variables. The step of calculating the estimated value of the dephosphorization rate constant of the hot metal and the step of estimating the phosphorus concentration in the molten steel during the blowing process based on the estimated value of the phosphorus concentration before the hot metal blowing process and the dephosphorization rate constant. The blowing treatment includes a pretreatment step including a dephosphorization treatment of hot metal, an intermediate discharge step carried out after the preliminary treatment step, and a decarburization treatment step carried out after the intermediate discharge step, and includes explanatory variables. Provides a method for estimating phosphorus concentration in molten steel, which comprises at least one data representing the amount of slag discharged from a converter in an intermediate scavenging step.

According to yet another aspect of the present invention, the hot metal data regarding the hot metal blown in the converter in the past operation, the operation data acquired during the blowing process, and the actual value of the phosphorus concentration in the molten steel are collected. The blowing process includes a step of constructing a statistical model using the hot metal data and the data extracted from the operation data as explanatory variables and the hot metal dephosphorization rate constant as the objective variable. The explanatory variables are discharged from the converter in the intermediate discharge step, including the pretreatment step including, the intermediate discharge step carried out after the preliminary treatment step, and the decarburization treatment step carried out after the intermediate discharge step. A method of building a statistical model is provided that includes at least one piece of data that represents the amount of slag.

According to yet another aspect of the present invention, a statistical model is used in which the hot metal data relating to the hot metal blown in the converter and the data extracted from the operation data acquired during the blowing process are used as explanatory variables. The phosphorus dephosphorization rate constant calculation unit that calculates the estimated value of the hot metal dephosphorization rate constant, and the phosphorus concentration in the molten steel during the blowing process based on the phosphorus concentration before the hot metal blowing process and the estimated value of the dephosphorization rate constant. It is equipped with a phosphorus concentration estimation unit for estimating, and the blowing treatment includes a pretreatment step including a dephosphorization treatment of hot metal, an intermediate discharge step carried out after the preliminary treatment step, and decarburization carried out after the intermediate discharge step. Provided by a program to operate the computer as a molten steel phosphorus concentration estimator, including the processing step and the explanatory variables containing at least one data representing the amount of slag discharged from the converter in the intermediate discharge step. Will be done.

According to yet another aspect of the present invention, the hot metal data regarding the hot metal blown by the converter in the past operation, the operation data acquired during the blowing process, and the actual value of the phosphorus concentration in the molten steel are collected. It is equipped with a data collection unit that collects data and a statistical model construction unit that builds a statistical model that uses the hot metal data and the data extracted from the operation data as explanatory variables and the hot metal dephosphorization rate constant as the objective variable. The explanatory variables include the pretreatment step including the dephosphorization treatment of hot metal, the intermediate discharge step carried out after the preliminary treatment step, and the decarburization treatment step carried out after the intermediate discharge step, and the explanatory variables are in the intermediate discharge step. A program is provided to make a computer function as a statistical model building device, including at least one piece of data representing the amount of slag discharged from the converter.

According to the above configuration, the phosphorus concentration in molten steel can be accurately estimated by the operation by the MURC method by considering the amount of slag discharged in the intermediate slag removal step.

It is a figure which shows the schematic structure of the refining equipment including the phosphorus concentration estimation apparatus in molten steel which concerns on one Embodiment of this invention. It is a figure which shows the example of the data structure of hot metal data and operation data. It is a flowchart which shows roughly the process of the converter blowing control method which concerns on one Embodiment of this invention. It is a figure which shows the schematic structure of the refining equipment including the statistical model construction apparatus which concerns on one Embodiment of this invention. It is a flowchart which shows roughly the process of the statistical model construction method which concerns on one Embodiment of this invention. It is a graph which shows the Example and the comparative example of this invention.

Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings. In the present specification and the drawings, components having substantially the same functional configuration are designated by the same reference numerals, so that duplicate description will be omitted.

In the present embodiment, the time change of the phosphorus concentration in molten steel (hereinafter, also referred to as [P] (%)) is represented by the primary reaction formula of the following formula (1). In addition, [P] _ini (%) represents the initial value (phosphorus concentration in hot metal) of [P], and k (sec ^-1 ) represents the dephosphorization rate constant. From the formula (1), [P] t seconds after the start of the blowing process is represented by the following formula (2).

Therefore, if an accurate dephosphorization rate constant k is obtained, the phosphorus concentration in the molten steel can be estimated with high accuracy t seconds after the start of blowing. However, in general, the dephosphorization rate constant k in actual blowing is not constant and is considered to fluctuate under the influence of various operating conditions. Therefore, for example, as disclosed in Japanese Patent Application Laid-Open No. 2013-23696, not only static information such as hot metal components and hot metal temperature obtained at the start of blowing and measurement of sublance, but also static information such as hot metal temperature is measured during blowing. Utilizing dynamic information such as exhaust gas flow rate and composition, the dephosphorization rate constant k is sequentially estimated. Hereinafter, a method for sequentially estimating the dephosphorization rate constant k will be described.

First, a statistical model is constructed with the dephosphorization rate constant k as the objective variable and various operating factors as explanatory variables. This statistical model can be appropriately constructed by various statistical methods. In this embodiment, a well-known multiple regression equation is used as the statistical model. The regression equation is constructed as the following equation (3). By substituting the operating factor X at the time of blowing into the equation (3), the dephosphorization rate constant k is estimated, and by applying the dephosphorization rate constant k to the above equation (2), the phosphorus concentration in the molten steel Is estimated.

As the operating factor X in the above formula (3), the factors shown in Table 1 have been conventionally used.

However, in the case of the MURC method, an intermediate slag step is carried out between the pretreatment step including the dephosphorization treatment of hot metal and the decarburization treatment step, and the amount of slag discharged in this intermediate slag removal step is the amount of molten steel. As already mentioned, the estimation accuracy of the phosphorus concentration in molten steel is not sufficient with the conventionally used operating factor X because it affects the concentration of medium phosphorus.

Therefore, in the embodiment of the present invention, by using at least one data representing the amount of slag discharged from the converter in the intermediate slag step as the operating factor X which is an explanatory variable of the statistical model, even in the operation by the MURC method. Accurately estimate the phosphorus concentration in molten steel.

(Estimation of phosphorus concentration in molten steel)
FIG. 1 is a diagram showing a schematic configuration of a refining facility including a phosphorus concentration estimation device in molten steel according to an embodiment of the present invention. As shown in FIG. 1, the refining facility 1 includes a converter facility 10, a measurement control device 20, and a phosphorus concentration estimation device 30 in molten steel. Hereinafter, each part will be further described.

The converter equipment 10 includes a converter 11, a top-blown lance 12, a flue 13, a tuyere 14, and a charging device 15. In the converter equipment 10, the top blowing lance 12 inserted from the furnace mouth of the converter 11 supplies oxygen gas 121 to the hot metal 111 in the converter 11. In the decarburization treatment, the carbon in the hot metal 111 reacts with the oxygen gas 121 _{to become CO gas or CO 2} gas, and these gases are discharged via the flue 13. The hot metal 111 that has undergone the decarburization treatment is sent to the next process as molten steel 112. Further, in the decarburization treatment, phosphorus and silicon in the hot metal 111 also react with the oxygen gas 121 or the auxiliary raw material contained in the slag 113, and are taken into the slag 113 to stabilize. On the other hand, a bottom-blown gas 141 such as nitrogen gas or argon gas is blown from the tuyere 14 to stir the hot metal 111 and promote the above reaction. The charging device 15 charges the auxiliary raw material 151 containing quicklime or limestone constituting the slag 113 and an oxygen-containing auxiliary material such as iron ore for supplying oxygen to the hot metal 111 into the converter 11. When the auxiliary material 151 is a powder, it can be blown together with the oxygen gas 121 by using the top blowing lance 12.

As described above, in the present embodiment, the operation by the MURC method is carried out. That is, in the present embodiment, the blowing treatment in the converter facility 10 is carried out after the pretreatment step including the dephosphorization treatment of the hot metal 111, the intermediate slag step carried out after the pretreatment step, and the intermediate slag step. Includes a decarburization process.

The measurement control device 20 executes various measurements related to the refining process in the converter facility 10 and controls the refining process. Specifically, the measurement control device 20 includes a sublance 21, an exhaust gas analyzer 22, and an exhaust gas flow meter 23 as a measurement system. The sublance 21 is inserted from the furnace port of the converter 11 together with the top blown lance 12, and the measuring device provided at the tip is immersed in the molten steel 112 at a predetermined timing during the decarburization process, whereby the molten steel 112 containing a carbon concentration is contained. (The carbon concentration in the molten steel is referred to as the molten steel carbon concentration), the temperature of the molten steel 112 (also referred to as the molten steel temperature), and the like are measured. Such a measurement using the sublance 21 is also referred to as a sublance measurement in the following description. The exhaust gas analyzer 22 analyzes the components of the gas discharged via the flue 13. Specifically, the exhaust gas analyzer 22 _{measures the concentrations of CO, CO 2} and O ₂ contained in the exhaust gas (the concentration of each component is referred to as the exhaust gas component concentration). On the other hand, the exhaust gas flow meter 23 measures the flow rate of the gas discharged through the flue 13 (referred to as the exhaust gas flow rate). The result of the above sublance measurement and the measurement result of the exhaust gas analyzer 22 and the exhaust gas flow meter 23 are transmitted to the phosphorus concentration estimation device 30 in the molten steel.

On the other hand, the measurement control device 20 includes a lance drive device 24, an oxygen supply device 25, a bottom blowing gas supply device 26, and an input control device 27 as a control system. The lance drive device 24 drives the top-blown lance 12 in the vertical direction. Thereby, the height of the top blowing lance 12, that is, the distance from the hot metal 111 at the position where the oxygen gas 121 is supplied in the converter 11 can be adjusted. The oxygen supply device 25 supplies the oxygen gas 121 to the top blowing lance 12. The flow rate of blown oxygen, that is, the flow rate of the supplied oxygen gas 121 per hour is adjustable. The bottom-blown gas supply device 26 supplies the bottom-blown gas 141 to the tuyere 14. The flow rate of the bottom blown gas 141 to be supplied can also be adjusted. The charging control device 27 controls the charging of the auxiliary raw material 151 by the charging device 15. Specifically, the charging control device 27 controls the charging timing and the charging amount of the auxiliary raw material 151. The operations of the lance drive device 24, the oxygen supply device 25, the bottom blowing gas supply device 26, and the injection control device 27 may be controllable according to a control signal from the molten steel medium phosphorus concentration estimation device 30. The slag level meter 28 measures the level of the slag 113 in the converter 11 from the outside of the converter 11 in a non-contact manner.

The molten steel medium phosphorus concentration estimation device 30 is implemented by a device including a communication unit 31, a storage unit 32, a calculation unit 33, and an input / output unit 34, for example, a computer. The communication unit 31 is various communication devices that communicate with each element of the measurement control device 20 by wire or wirelessly, receives the measurement result obtained by the measurement control device 20, and transmits a control signal to the measurement control device 20. do. The storage unit 32 is a storage capable of storing various types of data. The storage unit 32 contains, for example, hot metal data 321 for hot metal 111 to be blown in the converter 11, operation data 322 acquired during the hot metal, target data 323 including a target value of phosphorus concentration in molten steel, and the like. The parameter 324 of the statistical model for calculating the estimated value of the dephosphorization rate constant k is stored. The hot metal data 321 and the operation data 322 are stored as a database in which the data related to the charge becomes one record by using the charge number (CHNO) as a key, for example, as in the data structure shown in FIG. As described above, the hot metal data 321 and the operation data 322 may include the data in the past operation. As shown, in the present embodiment, the operation data 322 shows the tilt angle of the converter 11 when the discharge of the slag 113 is started in the intermediate discharge process, and the slag level before and after the intermediate discharge process. Includes difference data.

The arithmetic unit 33 is an arithmetic unit that executes various arithmetic operations according to a program. The arithmetic unit includes, for example, a CPU (Central Processing Unit), a RAM (Random Access Memory), and a ROM (Read Only Memory). The program is stored in the ROM of the arithmetic unit or the storage unit 32. The calculation unit 33 functions as a slag level difference calculation unit 331, a latent variable calculation unit 332, a phosphorus removal rate constant calculation unit 333, and a phosphorus concentration estimation unit 334 by operating according to the program. The calculation unit 33 may be a PLC (Programmable Logic Controller) or may be realized by dedicated hardware such as an ASIC (Application Specific Integrated Circuit). Hereinafter, the functions of each unit realized by the calculation unit 33 will be further described.

The slag level difference calculation unit 331 calculates the slag level difference before and after the intermediate slag removal step based on the output value of the slag level meter 28 included in the operation data 322 and the blown oxygen flow rate. Specifically, the slag level difference calculation unit 331 is a slag level meter at the time when the blown oxygen flow rate becomes 0 (or a value that can be regarded as 0; the same applies hereinafter) at the end of the pretreatment process including the dephosphorization treatment. The slag level difference is calculated based on the output value of 28 and the output value of the slag level meter 28 at the time when the blown oxygen flow rate becomes a value larger than 0 at the initial stage of the decarburization step. In the example shown in FIG. 2, the output value of the slag level meter used to calculate the slag level difference (level at the end of Blow1 and level before the start of Blow2), and the calculated slag level before and after the intermediate slag removal process. The difference is stored in the storage unit 32 as operation data 322.

The latent variable calculation unit 332 extracts data on the tilt angle of the converter 11 when the slag discharge is started in the intermediate discharge process and the slag level difference before and after the intermediate discharge process from the operation data 322. Latent variables are calculated from these data. As will be described later, both the tilt angle and the slag level difference are data expressing the amount of slag discharged from the converter 11 in the intermediate slag removal step, but each of them contains noise. Therefore, in the present embodiment, the influence of the noise component contained in each data is reduced by calculating the latent variable that best explains the dephosphorization rate constant k from the tilt angle and the slag level difference included in the operation data 322. do. The coefficient for calculating the latent variable uses the partial least squares (PLS) method with the tilt angle and slag level difference in past operations as explanatory variables and the actual value of the dephosphorization rate constant k as the objective variable. By executing it, it can be determined in advance.

The calculation of the latent variable using the PLS method is an example, and another method may be used as long as the influence of the noise component can be similarly reduced. For example, using principal component analysis, the actual value of the dephosphorization rate constant k is not considered as the objective variable, and the amount of information expressed by the two variables of the tilt angle and the slag level difference in the past operation is maximized. , The coefficient for calculating the latent variable may be determined. Alternatively, when the above two variables are dimensionally compressed into one variable (latent variable) by an autoencoder, the dimensional compression parameters are determined so that the original two variables are best reconstructed. May be good.

The dephosphorization rate constant calculation unit 333 calculates an estimated value of the dephosphorization rate constant k using, for example, a statistical model as shown in the above equation (3), using the hot metal data 321 and the operation data 322 as explanatory variables. do. Specifically, the dephosphorization rate constant calculation unit 333 reads the parameter 324 of the statistical model from the storage unit 32, and further substitutes the explanatory variable (operation factor X) of the statistical model extracted from the hot metal data 321 and the operation data 322. By doing so, the estimated value of the dephosphorization rate constant k is calculated. Here, the explanatory variables of the statistical model used by the dephosphorization rate constant calculation unit 333 include the latent variables calculated by the latent variable calculation unit 332. As described above, the latent variable is calculated from the tilt angle of the converter 11 when the slag discharge is started in the intermediate discharge process and the data of the slag level difference before and after the intermediate discharge process, and the intermediate discharge is performed. This is an example of data representing the amount of slag discharged from the converter 11 in the process.

The phosphorus concentration estimation unit 334 is an arbitrary during the blowing process based on the phosphorus concentration before the blowing process included in the hot metal data 321 and the estimated value of the phosphorus removal rate constant k calculated by the phosphorus removal rate constant calculation unit 333. Estimate the phosphorus concentration in the molten steel at the time of (for example, at the time of blowing off). For example, when the explanatory variable of the statistical model used by the dephosphorization rate constant calculation unit 333 includes the data acquired by the sublance measurement performed in the middle of the decarburization process, the phosphorus concentration estimation unit 334 stops blowing after the sublance measurement. The phosphorus concentration in the molten steel may be estimated at a predetermined cycle up to the time. The estimation result may be output via, for example, a display or a printer included in the input / output unit 34.

The input / output unit 34 includes an output device such as a display or a printer, and an input device such as a keyboard, a mouse, or a touch panel. The output device may output an estimated value of the phosphorus concentration in molten steel calculated by the phosphorus concentration estimation unit 334 at any time during the blowing process, for example, at the time of blowing down. The input device may, for example, acquire an operation input for changing the setting of the processing for estimating the phosphorus concentration in molten steel executed in the calculation unit 33, or an operation for adding or modifying the data stored in the storage unit 32. You may get the input.

FIG. 3 is a flowchart schematically showing a process of a converter blowing control method according to an embodiment of the present invention. First, before the start of blowing, the hot metal data 321 is acquired (step S101). These data are obtained, for example, from a converter smelting database stored in an external device (not shown in FIG. 1). After the start of blowing, the operation data 322 is sequentially acquired during the preliminary treatment step including the dephosphorization treatment of the hot metal 111 (step S102). Specifically, for example, the actual flow rate of oxygen blown during dephosphorization and smelting, the concentration of exhaust gas components, the flow rate of exhaust gas, and the like are acquired. At this time, data of a total of 28 slag levels is also collected.

When the pretreatment step is completed, the intermediate slag step is carried out, and the decarburization process is started, data on the amount of slag discharged from the converter 11 in the intermediate slag step is acquired. Specifically, the tilt angle data of the converter 11 when the slag discharge is started in the intermediate slag discharge process is acquired (step S103), and the slag level difference calculation unit 331 slags before and after the intermediate slag discharge process. The level difference is calculated (step S104). As described above, the slag level difference is calculated based on the output value of the slag level total 28, which collects data after the start of smelting and ends the data collection after the start of decarburization processing, and the actual flow rate of blown oxygen. .. Further, the latent variable calculation unit 332 calculates the latent variable from the tilt angle data acquired in step S103 and the slag level difference before and after the intermediate slag level calculation calculated in step S104 (step S105). On the other hand, during the decarburization process, the operation data 322 is sequentially acquired in parallel with the process of steps S103 to S105 described above (step S106).

In the illustrated example, the data of the operating factor X, which is the explanatory variable of the statistical model, are collected by the sublance measurement (step S107), and the dephosphorization rate constant k and the phosphorus concentration in the molten steel can be estimated using the statistical model. become. In this case, the dephosphorization rate constant calculation unit 333 extracts the data of the operating factor X including the latent variable calculated in step S105 from the hot metal data 321 and the operating data 322, and uses the statistical model to obtain the dephosphorization rate constant k. Estimate (step S108). Further, the phosphorus concentration estimation unit 334 substitutes the estimated phosphorus removal rate constant k into the above equation (2) to estimate the phosphorus concentration in the molten steel (step S109). The estimation result of the phosphorus concentration in the molten steel may be output, for example, via a display or a printer included in the input / output unit 34. The phosphorus concentration estimation unit 334 may estimate the phosphorus concentration in molten steel only once at a predetermined timing during the decarburization treatment step, or may repeat the estimation of the phosphorus concentration in molten steel at a predetermined cycle (step). S110).

According to one embodiment of the present invention as described above, the phosphorus concentration in molten steel can be accurately estimated by the operation by the MURC method. For example, if the phosphorus concentration in molten steel at the time of blow-off can be estimated accurately after the sublance measurement as in the above example, the estimated value of the phosphorus concentration in molten steel and the target value of the phosphorus concentration in molten steel indicated by the target data 323. When there is a difference between the two, the charging control device 27 can be controlled so as to change the charging timing and the charging amount of the auxiliary raw material 151. Such control may be executed according to the operation of the operator via the input / output unit 34, for example, or may be automatically executed by a function separately implemented in the calculation unit 33.

(Building a statistical model)
FIG. 4 is a diagram showing a schematic configuration of a refining facility including a statistical model building apparatus according to an embodiment of the present invention. As shown in FIG. 4, the refining facility 1A includes a converter facility 10, a measurement control device 20, and a statistical model building device 40. Since the converter equipment 10 and the measurement control device 20 are the same as the refining equipment 1 described above, duplicate description will be omitted. Hereinafter, each part of the statistical model building apparatus 40 will be further described.

The statistical model building device 40 is implemented by a device including a communication unit 31, a storage unit 42, a calculation unit 43, and an input / output unit 34, for example, a computer. Since the communication unit 31 and the input / output unit 34 are the same as the molten steel medium phosphorus concentration estimation device 30 described above, duplicate description will be omitted. The storage unit 42 is a storage capable of storing various types of data, and includes hot metal data 421, operation data 422 acquired during the blowing process, an actual value of phosphorus concentration in molten steel 423, and a dephosphorization rate constant k. The parameter 324 of the statistical model for calculating the estimated value of is stored. The hot metal data 421 and the operation data 422 are the same data as the hot metal data 321 and the operation data 322 stored in the storage unit 32 in the molten steel medium phosphorus concentration estimation device 30, for example, but are sufficient for constructing a statistical model. Includes data from past operations. The operation data 422 includes data on the tilt angle of the converter 11 when slag discharge is started in the intermediate slag discharge process, and data on the slag level difference before and after the intermediate slag discharge process.

The arithmetic unit 43 is an arithmetic unit that executes various arithmetic operations according to a program. The arithmetic unit includes, for example, a CPU (Central Processing Unit), a RAM (Random Access Memory), and a ROM (Read Only Memory). The program is stored in the ROM of the arithmetic unit or the storage unit 42. The calculation unit 43 functions as a data collection unit 431, a slag level difference calculation unit 432, a latent variable calculation unit 433, and a statistical model construction unit 434 by operating according to the program. The calculation unit 43 may be a PLC (Programmable Logic Controller) or may be realized by dedicated hardware such as an ASIC (Application Specific Integrated Circuit). Hereinafter, the functions of each unit realized by the calculation unit 43 will be further described.

The data collection unit 431 collects the hot metal data 321 and the operation data 422 in the past operation and the actual value 423 of the phosphorus concentration in the molten steel from the storage unit 42. In the collected operation data 422, when the slag level difference before and after the intermediate slag removal step is not calculated, the slag level difference calculation unit 432 and the slag level difference calculation unit 331 in the molten steel medium phosphorus concentration estimation device 30 described above. The slag level difference is calculated in the same manner. Further, the latent variable calculation unit 433 calculates the latent variable from the slag level difference and the tilt angle in the same manner as the latent variable calculation unit 332 in the molten steel medium phosphorus concentration estimation device 30. Since the coefficient for calculating the latent variable changes depending on the data of the past operation on which it is based, the latent variable calculation unit 433 may or may not calculate the latent variable in the collected operation data 422. Latent variables may be calculated for each of the past operations.

The statistical model construction unit 434 constructs a statistical model based on the hot metal data 421 and the operation data 422 in the past operation collected by the data collection unit 431, and the actual value 423 of the phosphorus concentration in the molten steel. The statistical model to be constructed uses the data extracted from the hot metal data 421 and the operation data 422 as explanatory variables, and the dephosphorization rate constant k of the hot metal 111 as the objective variable. The statistical model construction unit 434 calculates the actual value of the dephosphorization rate constant k from, for example, the actual value of the phosphorus concentration before the blowing process and the actual value of the phosphorus concentration in the molten steel at the time of blowing down 423 included in the hot metal data 421. Can be done. The constructed statistical model is stored in the storage unit 42 as a parameter 324. FIG. 5 is a flowchart showing an example of processing of the statistical model building apparatus 40 as described above.

By the function of the statistical model building apparatus 40 as described above, a statistical model that can be used in the molten steel medium phosphorus concentration estimation apparatus 30 described above is constructed. The statistical model construction device 40 may be the same device as the molten steel medium phosphorus concentration estimation device 30, or may be a different device. When these devices are different, for example, by storing the parameter 324 in the removable media and reading it, or by transmitting the parameter 324 via the communication line, the statistical model is in a device different from the device in which the statistical model is constructed. It is possible to estimate the phosphorus concentration in molten steel using.

(About data expressing slag emissions)
As described above, in the embodiment of the present invention, in estimating the phosphorus concentration in molten steel, at least one data representing the amount of slag discharged from the converter in the intermediate discharge step, specifically, the slag level. A statistical model is used that includes latent variables calculated from the difference and the tilt angle of the converter as explanatory variables. As already explained, in order to accurately estimate the phosphorus concentration in molten steel in the operation by the MURC method, the amount of slag discharged from the converter in the intermediate slag process (hereinafter, also referred to as the intermediate slag amount) is considered. That is valid. However, it is difficult to directly measure the amount of intermediate slag because it is a measurement of heavy objects in a high temperature and dust environment.

Therefore, in order to indirectly express the amount of intermediate slag, it is conceivable to add the forming status of slag to the explanatory variables. Although the slag forming status up to the target slag height can be grasped from the transition of the output value of the slag level meter, the slag forming status alone explains the amount of slag actually discharged in the subsequent intermediate slag. It's difficult to do. However, since the amount of hot metal charged is almost constant in normal operation, it is considered that the transition of the output value of the slag level meter indirectly expresses the amount of slag in the converter. Therefore, in the embodiment of the present invention, the slag level difference calculated from the output value of the slag level meter at the end of the pretreatment process and the output value of the slag level meter at the start of the decarburization process is directly or. Indirectly used as an explanatory variable. Conventionally, the slag level meter has been intended to grasp the slag forming status during the pretreatment process, and the measurement is completed at the end of the pretreatment process. In the embodiment of the present invention, in order to enable the calculation of the slag level difference as described above, the slag level meter continues the measurement from the end of the pretreatment step to the start of the decarburization treatment step.

In addition, in order to indirectly explain the amount of intermediate slag, it is conceivable to add the tilt angle transition of the converter during intermediate slag to the explanatory variables. However, since the angle at the start of slag discharge and the angle at the end of discharge of slag cannot be known only from the transition of the tilt angle, it is difficult to explain the intermediate discharge amount only from the transition of the tilt angle. On the other hand, it is considered that the intermediate slag amount can be explained to some extent from the tilt angle of the converter when the slag discharge is started in the intermediate slag discharge process. Specifically, when the tilt angle when the slag starts to be discharged is small, that is, when the slag starts to be discharged without tilting the converter too much, the slag is easily discharged, and as a result, the intermediate discharge is performed. It is estimated that the amount of slag was large. On the other hand, if the tilt angle at the start of slag discharge in the intermediate slag discharge process is large, that is, if the slag does not start to be discharged unless the converter is tilted significantly, the slag is difficult to discharge, resulting in intermediate slag. It is estimated that the amount of slag was small. Therefore, in the embodiment of the present invention, the tilt angle of the converter when the slag discharge is started in the intermediate slag discharge step is directly or indirectly used as an explanatory variable. Conventionally, the tilt angle of the converter when the slag starts to be discharged is recorded by hand on the form, but in the embodiment of the present invention, it is used as an explanatory variable of a statistical model for estimating the phosphorus concentration in molten steel. In order to use it, the tilt angle of the converter when the slag starts to be discharged is input as electronic data via the input / output unit 34.

As described above, it is considered that the difference in slag level and the tilt angle of the converter when the slag starts to be discharged can be effectively used as data expressing the amount of intermediate waste. However, since slag forming is intense during blowing in the pretreatment step, there is a high possibility that measurement noise is included in the output value of the slag level meter for calculating the slag level difference. On the other hand, regarding the tilt angle of the converter, since the operator determines whether or not the slag discharge has started, it is considered that a noise component due to the variation in the determination is included. That is, these data include a variation (amount of information with respect to k) explaining the dephosphorization rate constant k, which is an objective variable, and a variation (noise component with respect to k) not explaining the dephosphorization rate constant k. Therefore, in the above-described embodiment of the present invention, the influence of the noise component is reduced by, for example, calculating the latent variable using the PLS method. In the PLS method, a latent variable represented by a linear combination of explanatory variables is constructed, and the objective variable is expressed as a linear combination of the latent variables. Since the latent variable is determined so that the inner product of the latent variable and the objective variable is maximized, it is possible to construct a latent variable that best explains the objective variable, that is, a variable that does not explain the objective variable and has a reduced noise component. be. As already described, it is possible to reduce the influence of the noise component by a method other than the PLS method.

FIG. 6 is a graph showing an example and a comparative example of the present invention. In the graph, the dephosphorization rate constant k is estimated in four examples in which the explanatory variables of the statistical model are changed, using the hot metal data for 427 charges, the operation data, and the actual value of the phosphorus concentration in the molten steel at the time of blowing down as a sample. The standard deviation of the error between the estimated value and the measured value is shown. Base is a comparative example in the case where there is no explanatory variable expressing the amount of slag discharged from the converter in the intermediate discharge step, and for example, the explanatory variables shown in Table 1 above are used. Case1 to Case4 are examples of the present invention using explanatory variables as shown in Tables 2 to 5 below, for example.

As shown in the above table, Case 1 is an example in which only the slag level difference (m) before and after the intermediate slag level difference is added to the explanatory variables of the comparative example. Case 2 is an example in which only the tilt angle (°) of the converter when the slag starts to be discharged is added to the explanatory variable of the comparative example. Case 3 is an example in which both the slag level difference (m) before and after the intermediate slag removal step and the tilt angle (°) of the converter when the slag starts to be discharged are added to the explanatory variables of the comparative example. In Case 1 to Case 3, the added explanatory variables used the acquired data as they were, and the latent variables were not calculated. For Case4, the latent variable calculated from the slag level difference (m) before and after the intermediate slag level difference and the tilt angle (°) of the converter when the slag started to be discharged was added to the explanatory variables of the comparative example. This is an example.

With reference to the graph of FIG. 6, the implementation of Case 1 and Case 2 in which only one of the slag level difference before and after the intermediate discharge process or the tilt angle of the converter when the slag starts to be discharged is added to the explanatory variables. In the example as well, the standard deviation of the error between the estimated value and the measured value is lower than that in the comparative example. From this result, as an embodiment of the present invention, when the slag level difference before and after the intermediate slag step or when slag begins to be discharged as data expressing the amount of slag discharged from the converter in the intermediate slag step. It can be seen that the estimation accuracy is improved as compared with the prior art even when only one of the tilt angles of the converter is used. Further, in the example of Case 3 in which both the slag level difference before and after the intermediate discharge step and the tilt angle of the converter when the slag starts to be discharged are added to the explanatory variables, compared with the examples of Case 1 and Case 2. Furthermore, the standard deviation of the error is reduced. From this result, as data expressing the amount of slag discharged from the converter in the intermediate discharge process, the difference in slag level before and after the intermediate discharge process and the tilt angle of the converter when slag begins to be discharged. It can be seen that the estimation accuracy is further improved by using both, and that the calculation of the latent variable is not essential in the embodiment of the present invention. Furthermore, in the example of Case 4 in which the latent variable is added to the explanatory variable, the standard deviation of the error is further reduced as compared with Case 3. From this result, it can be seen that in the embodiment of the present invention, the influence of the noise component contained in the data is reduced by calculating the latent variable, and the estimation accuracy can be further improved.

Although the preferred embodiments of the present invention have been described in detail with reference to the accompanying drawings, the present invention is not limited to such examples. It is clear that a person having ordinary knowledge in the field of technology to which the present invention belongs can come up with various modifications or modifications within the scope of the technical ideas described in the claims. , These are also naturally understood to belong to the technical scope of the present invention.

1,1A ... refining equipment, 10 ... converter equipment, 11 ... converter, 12 ... top blow lance, 13 ... flue, 14 ... tuyere, 15 ... input device, 20 ... measurement control device, 21 ... sublance, 22 ... Exhaust gas analyzer, 23 ... Exhaust flow meter, 24 ... Lance drive device, 25 ... Oxygen supply device, 26 ... Bottom blowing gas supply device, 27 ... Input control device, 28 ... Slug level meter, 30 ... Concentration estimation device, 31 ... communication unit, 32 ... storage unit, 33 ... arithmetic unit, 34 ... input / output unit, 40 ... statistical model construction device, 42 ... storage unit, 43 ... arithmetic unit, 111 ... hot metal, 112 ... molten steel, 113 ... slug, 121 ... Oxygen gas, 141 ... Bottom blown gas, 151 ... Auxiliary raw material, 321 ... Hot metal data, 322 ... Operation data, 323 ... Target data, 324 ... Parameters, 331 ... Slug level difference calculation unit, 332 ... Latent variable calculation unit, 333 ... Dephosphorization rate constant calculation unit, 334 ... Phosphorus concentration estimation unit, 421 ... Hot metal data, 422 ... Operation data, 423 ... Actual value, 431 ... Data collection unit, 432 ... Slug level difference calculation unit, 433 ... Latent variable calculation unit 434 ... Statistical model construction department.

Claims (11)

Estimated value of the dephosphorization rate constant of the hot metal using a statistical model with the hot metal data related to the hot metal blown in the converter and the data extracted from the operation data acquired during the hot metal as explanatory variables. Dephosphorization rate constant calculation unit to calculate
It is provided with a phosphorus concentration estimation unit that estimates the phosphorus concentration in the molten steel during the blowing process based on the phosphorus concentration before the hot metal blowing process and the estimated value of the dephosphorization rate constant.
The blowing treatment includes a pretreatment step including a dephosphorization treatment of the hot metal, an intermediate slag step carried out after the pretreatment step, and a decarburization treatment step carried out after the intermediate slag removal step.
The explanatory variable is a phosphorus concentration estimation device in molten steel containing at least one data representing the amount of slag discharged from the converter in the intermediate slag removal step. The at least one data shows the difference in slag level before and after the intermediate slag removal step measured using a slag level meter installed in the converter, and slag discharge was started in the intermediate slag discharge step. The phosphorus concentration estimation device in molten steel according to claim 1, which includes any or both of the tilt angles of the converter at the time. The operation data is the difference in slag level before and after the intermediate slag removal process measured using a slag level meter installed in the converter, and when slag discharge is started in the intermediate slag discharge process. Including the tilt angle of the converter
The molten steel concentration estimation device further includes a latent variable calculation unit that calculates a latent variable from the slag level difference and the tilt angle.
The phosphorus concentration estimation device in molten steel according to claim 1, wherein the at least one data includes the latent variable. The operation data includes the flow rate of oxygen blown into the converter.
The output value of the slag level meter at the time when the blown oxygen flow rate became 0 at the end of the pretreatment step and the value where the blown oxygen flow rate became larger than 0 at the beginning of the decarburization treatment step. The phosphorus concentration estimation device in molten steel according to claim 2 or 3, further comprising a slag level difference calculation unit that calculates the slag level difference based on the output value of the slag level meter at a time point. A data collection unit that collects hot metal data related to hot metal blown in a converter in past operations, operation data acquired during the hot metal blowing process, and actual values of phosphorus concentration in molten steel.
It is provided with a statistical model construction unit that constructs a statistical model in which the hot metal data and the data extracted from the operation data are used as explanatory variables and the hot metal dephosphorization rate constant is used as an objective variable.
The blowing treatment includes a pretreatment step including a dephosphorization treatment of the hot metal, an intermediate slag step carried out after the pretreatment step, and a decarburization treatment step carried out after the intermediate slag removal step.
The explanatory variable is a statistical model building apparatus including at least one data representing the amount of slag discharged from the converter in the intermediate slag step. The at least one data shows the difference in slag level before and after the intermediate discharge step measured using a slag level meter installed in the converter, and the slag discharge was started in the intermediate discharge step. The statistical model building apparatus according to claim 5, which includes any or both of the tilt angles of the converter at the time. The at least one data shows the difference in slag level before and after the intermediate discharge step measured using a slag level meter installed in the converter, and the slag discharge was started in the intermediate discharge step. The statistical model building apparatus according to claim 5, which includes a latent variable calculated from the tilt angle of the converter at the time. Estimated value of the dephosphorization rate constant of the hot metal using a statistical model with the hot metal data related to the hot metal blown in the converter and the data extracted from the operation data acquired during the hot metal as explanatory variables. And the process of calculating
Including a step of estimating the phosphorus concentration in the molten steel during the blowing process based on the phosphorus concentration of the hot metal before the blowing process and the estimated value of the dephosphorization rate constant.
The blowing treatment includes a pretreatment step including a dephosphorization treatment of the hot metal, an intermediate slag step carried out after the pretreatment step, and a decarburization treatment step carried out after the intermediate slag removal step.
The explanatory variable is a method for estimating the phosphorus concentration in molten steel, which comprises at least one data representing the amount of slag discharged from the converter in the intermediate slag removal step. A process of collecting hot metal data related to hot metal blown in a converter in past operations, operation data acquired during the hot metal blowing process, and a process of collecting actual values of phosphorus concentration in molten steel.
It includes a step of constructing a statistical model in which the hot metal data and the data extracted from the operation data are used as explanatory variables and the dephosphorization rate constant of the hot metal is used as an objective variable.
The blowing treatment includes a pretreatment step including a dephosphorization treatment of the hot metal, an intermediate slag step carried out after the pretreatment step, and a decarburization treatment step carried out after the intermediate slag removal step.
The explanatory variable is a method of constructing a statistical model including at least one data representing the amount of slag discharged from the converter in the intermediate slag step. Estimated value of the dephosphorization rate constant of the hot metal using a statistical model with the hot metal data related to the hot metal blown in the converter and the data extracted from the operation data acquired during the hot metal as explanatory variables. Dephosphorization rate constant calculation unit to calculate
It is provided with a phosphorus concentration estimation unit that estimates the phosphorus concentration in the molten steel during the blowing process based on the phosphorus concentration before the hot metal blowing process and the estimated value of the dephosphorization rate constant.
The blowing treatment includes a pretreatment step including a dephosphorization treatment of the hot metal, an intermediate slag step carried out after the pretreatment step, and a decarburization treatment step carried out after the intermediate slag removal step.
The explanatory variable is a program for operating a computer as a phosphorus concentration estimation device in molten steel, which includes at least one data representing the amount of slag discharged from the converter in the intermediate slag removal step. A data collection unit that collects hot metal data related to hot metal blown in a converter in past operations, operation data acquired during the hot metal blowing process, and actual values of phosphorus concentration in molten steel.
It is provided with a statistical model construction unit that constructs a statistical model in which the hot metal data and the data extracted from the operation data are used as explanatory variables and the hot metal dephosphorization rate constant is used as an objective variable.
The blowing treatment includes a pretreatment step including a dephosphorization treatment of the hot metal, an intermediate slag step carried out after the pretreatment step, and a decarburization treatment step carried out after the intermediate slag removal step.
The explanatory variable is a program for operating a computer as a statistical model construction device, which includes at least one data representing the amount of slag discharged from the converter in the intermediate discharge step.

Images