Silicon concentration prediction method, operation guidance method, silicon concentration prediction device, operation guidance device, operation guidance system, and terminal device

Abstract

A Si concentration prediction method including: a prediction step (S1) of predicting a future predicted value and a current value of Si concentration using a statistical model and a physical model with operational variables of a blast furnace as input and the Si concentration as output; and an adjustment amount calculation step (S3) of calculating, when the predicted value or the current value significantly deviates from a target value, an adjustment amount of an amount of de-siliconizing agent added in a control action for lowering the Si concentration by adding the de-siliconizing agent, or an adjustment amount of the operational variables in a control action of controlling a thermal level in the blast furnace in order to increase the Si concentration at tapping before de-siliconizing.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a Si concentration prediction method, an operation guidance method, a Si concentration prediction device, an operation guidance device, an operation guidance system, and a terminal device.BACKGROUND

[0002] In blast furnace processes and converter processes in the steel manufacturing, Si concentration in hot metal (hot metal Si concentration) is an index of the product for blast furnaces and a component of raw material for converters, and is therefore an important variable. In response to the recent social demand for CO 2 emission reduction, efforts are being made to decrease the hot metal blending ratio (pig iron blending ratio) in converters. Therefore, Si in hot metal is actively utilized as a heat source in converters.

[0003] When Si concentration in hot metal is too low, an extra heat source (such as FeSi) needs to be added into a converter, leading to increased raw material costs. On the other hand, when Si concentration in hot metal is too high, slopping may occur in a pre-treatment stage, decreasing productivity, or adding extra quicklime for basicity adjustment may become necessary. As described, there is an optimum point for Si concentration, and therefore it is necessary to adjust the Si concentration on the blast furnace side to match the optimum point.

[0004] Various techniques have been proposed for predicting Si concentration in blast furnaces. For example, in Patent Literature (PTL) 1, a method is proposed for predicting future hot metal Si concentration by calculating a predicted value of future hot metal temperature (HMT) when an operation amount is maintained, constructing a regression equation based on the calculated predicted value, and inputting the predicted value into the regression equation.CITATION LISTPatent Literature

[0005] PTL 1: JP 2021-017607 ASUMMARY(Technical Problem)

[0006] To adjust Si concentration, control actions such as lowering the Si concentration by adding a de-siliconizing agent to the casting bed or changing the Si concentration at the time of tapping before de-siliconizing by controlling a thermal level in the blast furnace may be considered. Here, the thermal level is a classification of temperature in the furnace into temperature ranges. The temperature ranges may be constant or variable. In current blast furnace operations, Si concentration takes time to analyze, and therefore an actual performance value is obtained 1 to 2 hours after tapping. Therefore, implementation of an action of adding a de-siliconizing agent to the casting bed requires an accurate estimation (prediction) of current Si concentration. Further, when implementing an action to change Si concentration by controlling the thermal level in a blast furnace, the thermal inertia of the blast furnace process has a large impact and the immediacy of the control action is low, and therefore predicting 2 hours or more in advance is essential. Therefore, there is a need for a high precision prediction method that can handle both of these control actions.

[0007] Further, there is a need for coordination between ironmaking (blast furnace side) and steelmaking processes. For example, when an operator of a steelmaking process can be appropriately informed whether or not to add de-siliconizing agent after tapping, according to an amount of heat source required in the converter blowing charge on the steelmaking side, further decreases in steelmaking costs may result. However, conventional methods are limited to predicting Si concentration in the blast furnace process alone.

[0008] In view of the above problems, it would be helpful to provide a Si concentration prediction method, an operation guidance method, a Si concentration prediction device, an operation guidance device, an operation guidance system, and a terminal device that can predict Si concentration with high precision.(Solution to Problem)

[0009] (1) A Si concentration prediction method according to an embodiment of the present disclosure, the Si concentration prediction method comprising: a prediction step of predicting a future predicted value and a current value of Si concentration using a statistical model and a physical model with operational variables of a blast furnace as input and the Si concentration as output; and an adjustment amount calculation step of calculating, when the predicted value or the current value significantly deviates from a target value, an adjustment amount of an amount of de-siliconizing agent added in a control action for lowering the Si concentration by adding the de-siliconizing agent, or an adjustment amount of the operational variables in a control action of controlling a thermal level in the blast furnace in order to increase the Si concentration at tapping before de-siliconizing. (2) The Si concentration prediction method according to (1), as an embodiment of the present disclosure, wherein, in the prediction step, the statistical model is used to predict the current value and the predicted value up to a defined time, and the physical model is used to predict the predicted value beyond the defined time, and the defined time is determined based on the time at which prediction precision of the statistical model becomes lower than prediction precision of the physical model. (3) The Si concentration prediction method according to (1) or (2), as an embodiment of the present disclosure, wherein the operational variables that are input to the statistical model include at least one of: tuyere temperature, furnace heat index, solution loss carbon amount, furnace top gas temperature, tuyere tip gas temperature, or blast volume per iron. (4) The Si concentration prediction method according to any one of (1) to (3), as an embodiment of the present disclosure, wherein the operational variables that are input to the physical model include at least one of: blast volume (blast flow rate), blast volume oxygen, pulverized coal injection (PCI) rate, coke ratio, blast moisture, or blast temperature. (5) An operation guidance method according to an embodiment of the present disclosure, the operation guidance method comprising: a step of causing the predicted value and the current value predicted by the Si concentration prediction method according to any one of (1) to (4) to be presented to an operator. (6) The operation guidance method according to (5), as an embodiment of the present disclosure, further comprises: an input step of inputting a corrected target value of the Si concentration, wherein the adjustment calculation step calculates the adjustment amount of the operational variables using the corrected target value input in the input step as the target value. (7) The operation guidance method according to (6), as an embodiment of the present disclosure, wherein the corrected target value is set based on input information from an operator other than an ironmaking operator. (8) A Si concentration prediction device according to an embodiment of the present disclosure, the Si concentration prediction device comprising: a storage configured to store a statistical model and a physical model with operational variables of a blast furnace as input and Si concentration as output; a predictor configured to predict a future predicted value and a current value of the Si concentration using the statistical model and the physical model; and an adjustment amount calculator configured to calculate, when the predicted value or the current value significantly deviates from a target value, an adjustment amount of an amount of de-siliconizing agent added in a control action for lowering the Si concentration by adding the de-siliconizing agent, or an adjustment amount of the operational variables in a control action of controlling a thermal level in the blast furnace in order to increase the Si concentration at tapping before de-siliconizing. (9) An operation guidance device according to an embodiment of the present disclosure, the operation guidance device comprising: a storage configured to store a statistical model and a physical model with operational variables of a blast furnace as input and Si concentration as output; a predictor configured to predict a future predicted value and a current value of the Si concentration using the statistical model and the physical model, and to cause the predicted value and the current value to be presented to an operator; and an adjustment amount calculator configured to calculate, when the predicted value or the current value significantly deviates from a target value, an adjustment amount of an amount of de-siliconizing agent added in a control action for lowering the Si concentration by adding the de-siliconizing agent, or an adjustment amount of the operational variables in a control action of controlling a thermal level in the blast furnace in order to increase the Si concentration at tapping before de-siliconizing. (10) An operation guidance system according to an embodiment of the present disclosure, the operation guidance system comprising: an operation guidance device; and a terminal device, wherein the operation guidance device or the terminal device comprise: a storage configured to store a statistical model and a physical model with operational variables of a blast furnace as input and Si concentration as output; a predictor configured to predict a future predicted value and a current value of the Si concentration using the statistical model and the physical model, and to output the predicted value and the current value; an adjustment amount calculator configured to calculate, when the predicted value or the current value significantly deviates from a target value, an adjustment amount of an amount of de-siliconizing agent added in a control action for lowering the Si concentration by adding the de-siliconizing agent, or an adjustment amount of the operational variables in a control action of controlling a thermal level in the blast furnace in order to increase the Si concentration at tapping before de-siliconizing, and to output the adjustment amount; an acquisition unit configured to acquire the predicted value, the current value, and the adjustment amount; an input interface configured to accept input of a corrected target value of the Si concentration and a future time or time period for correction; a transmitter configured to transmit the corrected target value and the future time or time period for correction that have been input; and a display configured to display the predicted value, the current value, the adjustment amount, the corrected target value, and the future time or time period for correction. (11) A terminal device according to an embodiment of the present disclosure, the terminal device comprising: an acquisition unit configured to acquire a predicted value and a current value of Si concentration, and an adjustment amount, wherein the predicted value and the current value are predicted by a statistical model and a physical model with operational variables of a blast furnace as input and the Si concentration as output, and the adjustment amount is calculated when the predicted value or the current value significantly deviates from a target value, and indicates an adjustment of de-siliconizing agent added in a control action for lowering the Si concentration by adding de-siliconizing agent, or an adjustment of operational variables in a control action of controlling a thermal level in a blast furnace in order to increase the Si concentration at tapping before de-siliconizing; an input interface configured to accept input of a corrected target value of the Si concentration and a future time or time period for correction; a transmitter configured to transmit the corrected target value and the future time or time period for correction that have been input; and a display configured to display the predicted value, the current value, the adjustment amount, the corrected target value, and the future time or time period for correction. (12) The information terminal according to (11), as an embodiment of the present disclosure, further comprising: an inter-terminal communicator configured to transmit and receive information about the corrected target value to and from another terminal device. (Advantageous Effect)

[0010] According to the present disclosure, a Si concentration prediction method, an operation guidance method, a Si concentration prediction device, an operation guidance device, an operation guidance system, and a terminal device that can predict Si concentration with high precision can be provided.BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In the accompanying drawings: FIG. 1 is a diagram illustrating input / output information of a statistical model used according to the present disclosure; FIG. 2 is a diagram illustrating prediction results of a statistical model; FIG. 3 is a diagram illustrating input / output information of a physical model used according to the present disclosure; FIG. 4 is a diagram illustrating a guidance screen presented to an operator; FIG. 5 is a diagram illustrating an example configuration of a Si concentration prediction device according to an embodiment; and FIG. 6 is a flowchart illustrating a Si concentration prediction method according to an embodiment. DETAILED DESCRIPTION

[0012] The Si concentration prediction method, the operation guidance method, the Si concentration prediction device, the operation guidance device, the operation guidance system, and the terminal device according to an embodiment of the present disclosure are described below, with reference to the drawings. A statistical model and a physical model are used, according to the present disclosure. Use of the statistical model or the physical model is determined according to the time for which a prediction is made, and high precision predictions are combined and presented to, for example, an operator. A mechanism for determining Si concentration, the statistical model, and the physical model are described in turn below. Hereinafter, Si concentration means "Si concentration in hot metal", that is, hot metal Si concentration.

[0013] The following two mechanisms are known to determine Si concentration in blast furnaces. Here, [x] in a reaction equation indicates a component x in hot metal, and (x) indicates a component x in slag. First, as the first mechanism, SiO gas is generated in the combustion zone of a tuyere of a blast furnace (SiO 2 + C → SiO + CO) and is absorbed by hot metal dripping at the lower part of the furnace (SiO + [C] → [Si] + CO). As the second mechanism, Si is incorporated into hot metal via a SiO 2 reduction reaction ((SiO 2 ) + 2C → [Si] + 2CO) at a slag-metal interface at the bottom of the blast furnace.

[0014] In hot metal, the thermodynamic equilibrium of [Si] exceeds 10 %. However, actually measured Si concentration is about 2 % at most. Therefore, [Si] is considered to not have reached equilibrium. Further, both mechanisms are greatly affected by thermal conditions inside the furnace. Therefore, furnace thermal factors are taken into account in the calculation of the predicted value of the Si concentration according to the present embodiment.

[0015] The statistical model is not limited to a specific model, but for example, a locally-weighted partial least squares (LW-PLS) model may be adopted, as in the present embodiment. The LW-PLS model can predict Si concentration with emphasis on historical data similar to current operational conditions and suppresses overfitting caused by multicollinearity.

[0016] As illustrated in FIG. 1, the main input variables provided to the statistical model that vary with time are: the tuyere temperature, the furnace heat index, the solution loss carbon amount, the furnace top gas temperature, the tuyere tip gas temperature, and the blast volume per iron. These input variables are the operational variables of the blast furnace and are measured values taken by sensors installed in the blast furnace, or values obtained from a known conversion expression based on measured values. Here, the blast furnace operational variables are parameters that are set or measured in relation to the operation of the blast furnace. The input to the statistical model may include at least one of these operational variables. The tuyere temperature, the furnace top gas temperature, and the tuyere tip gas temperature are the temperatures of gas at the tuyere embedded section, the furnace top gas, and the tuyere tip, respectively. The furnace heat index is the heat supply to the blast furnace. The blast volume per iron is the blast volume used per tonne of hot metal produced.

[0017] Further, the output variable of the statistical model is the Si concentration. It is possible to calculate the Si concentration that changes over time using the statistical model. The statistical model may be used to calculate (predict) the current value and the predicted value of the Si concentration. FIG. 2 illustrates results of predicting hot metal Si after 2 hours using the statistical model (LW-PLS model according to the present embodiment), where good prediction precision is obtained.

[0018] Here, the number of hours into the future for which the predicted value is to be determined needs to be flexibly changed depending on a control action to control the Si concentration. For example, when a control action is to decrease the Si concentration by adding a de-siliconizing agent in the casting bed, the immediacy of the change is high, and therefore predicting only the current value is sufficient. On the other hand, when a control action is to control the thermal level in the blast furnace to increase the Si concentration at tapping before de-siliconizing, the effect of thermal inertia is significant and the immediacy is low, and therefore predicting more than 2 hours in advance is essential.

[0019] The statistical model makes predictions based on measured values from sensors in the blast furnace, and therefore precision is decreased in the case of long-term predictions. As described above, the statistical model uses as input the sensor data in the furnace, such as the tuyere temperature and the furnace top gas temperature. When controlling the thermal level in a blast furnace, accurate predictions can be made after sensor data begins to reflect changes in the thermal level. Therefore, there is a risk that the prediction precision may decrease when the effect of a change in operation amount in the blast furnace (PCI rate, coke ratio, and the like) has not yet been reflected in the sensor data, especially immediately after the change. That is, in statistical models, obtaining sufficient precision in long-term prediction (predicting beyond 2 hours, as an example) is difficult. Accordingly, the physical model is preferably used for long-term prediction.

[0020] The physical model is a transient state model, similar to the method described in reference literature (K. Takatani et al. ISIJ International, Vol. 39 (1999), pp. 15), which comprises a set of partial differential equations that take into account physical phenomena such as ore reduction, heat exchange between ore and coke, and ore melting, and is capable of calculating the internal (in-furnace) state of a blast furnace in a transient state. Transient states include, for example, the occurrence of events such as gas channeling, hanging, and the like.

[0021] As illustrated in FIG. 3, the main input variables provided to the physical model that vary over time are the blast volume, the blast volume oxygen, the PCI rate, the coke ratio, the blast moisture, and the blast temperature. These input variables are the operational variables of the blast furnace. The inputs to the physical model may include at least one of these operational variables. The blast volume, the blast volume oxygen, and the PCI rate are the flow rates of air, oxygen, and pulverized coal delivered to the blast furnace, respectively. The coke ratio is the coke ratio at the furnace top and is the weight of coke used per tonne of hot metal produced. The blast moisture is the humidity of the air delivered to the blast furnace. The blast temperature is the temperature of the air delivered to the blast furnace.

[0022] The main output variables of the physical model are the gas utilization rate, the solution loss carbon amount, the reducing agent ratio (RAR), the production rate, the production temperature, and the Si concentration. It is possible to calculate the Si concentration that changes over time using the physical model. The time interval for this calculation is not particularly limited, but is 1 hour according to the present embodiment. The time difference between "t+1" and "t" in the physical model expressions described below is 1 hour according to the present embodiment. According to the present embodiment, the physical model is a three-dimensional transient-state model capable of estimating three-dimensional temperature distribution and other parameters in the furnace. However, the form of the physical model is not limited to three-dimensional transient state models.

[0023] The physical model may be expressed by the following expressions (1) and (2). x t + 1 = f x t , u t y t = C x t

[0024] Here, x(t) is state variables calculated within the physical model. The state variables are, for example, the temperature of the coke, the temperature of the iron, the oxidation degree of the ore, and the descent speed of raw material. y(t) is the Si concentration, which is the control variable. u(t) is the input variables described above, which may be manipulated by an operator of the blast furnace. That is, the input variables are: blast volume BV (t), blast volume oxygen BVO (t), pulverized coal injection rate PCI (t), coke ratio CR (t), blast moisture BM (t), and blast temperature BT (t). The input variables can be expressed as u(t) = (BV(t), BVO(t), PCI(t), CR(t), BM(t), BT(t)). Expression (1) and expression (2) can be repeatedly calculated to predict future Si concentration.

[0025] According to the present embodiment, the Si concentration is predicted using the statistical model or the physical model, depending on "how many hours ahead the predicted value is required", and high precision prediction is combined and presented to, for example, an operator. FIG. 4 is a screen example illustrating such predicted values of the Si concentration presented to an ironmaking operator and a steelmaking operator. The screen in FIG. 4 is a guidance screen that allows an operator to take an appropriate control action according to the predicted Si concentration. In the example in FIG. 4, the charts from the tuyere temperature to the tuyere tip gas temperature indicate operational results of the blast furnace, and the charts from the blast volume to the blast volume oxygen indicate the operational variables that can be manipulated by an operator in the operation of the blast furnace. The horizontal axis indicates time, with zero corresponding to the current time. Further, in the example in FIG. 4, the target value of the Si concentration is 0.5 [wt%]. According to the present embodiment, the statistical model is used for predicted values up to 2 hours ahead, and the physical model is used for predicted values 3 hours or more ahead. That is, the statistical model is used to predict the current value and the predicted value of the Si concentration up to a defined time, and the physical model is used to predict the predicted value of the Si concentration beyond the defined time. An example of the defined time is 2 hours. Here, the defined time may be determined based on the time at which the prediction precision of the statistical model becomes lower than the prediction precision of the physical model. For example, the defined time may be determined by comparing the prediction precision of the statistical model with that of the physical model through experimentation using actual performance data from operation (past measurement data and the like). For example, in an experiment, when the time to be predicted in the experiment is extended to 2.5 hours ahead and the prediction precision of the statistical model falls below the prediction precision of the physical model, the defined time may be determined to be 2 hours.

[0026] Conventionally, steelmaking operators often obtain only an estimate of the current Si concentration based on measured values. By presenting at least a steelmaking operator with the predicted trends in the Si concentration as in FIG. 4, the steelmaking operator can inform an ironmaking operator of an appropriate control action for adjusting an amount of de-siliconizing agent added and adjusting the Si concentration at an early stage. Although it is the ironmaking operator who actually executes the control action, the steelmaking operator, who is aware of the steel casting plan and the like, can check the screen in FIG. 4 and contact the ironmaking side, thereby enabling cooperation that was not possible with conventional prediction and judgments limited to only the ironmaking side. Here, the ironmaking operator and the steelmaking operator may be able to simultaneously view the same predicted trends in the Si concentration by the statistical model and the physical model, while being in different locations. In other words, the predicted trend of the Si concentration, as illustrated in FIG. 4, may be simultaneously displayed on a display device visible to the ironmaking operator and the steelmaking operator. At this time, strong collaboration is possible by sharing the same information. As a result, steelmaking costs may be lowered.

[0027] A specific operation could be, for example, to change the target value of future Si concentration. This is because an action to control the Si concentration, described below, is determined based on the discrepancy between the target value of the Si concentration and the predicted value of the Si concentration. The target value of the Si concentration is determined in relation to blast furnace operation with reference to the steel casting plan and the like in steelmaking. The steel casting plan is determined in a plurality of stages for different time units. For example, based on a monthly production plan or a weekly production plan, a steel casting plan is made for each day up to about 24 hours ahead, and a final steel casting plan is determined every 6 hours or so. Blast furnaces have a long time constant (for example, 8 hours or more in some cases), and therefore the target value of the Si concentration is determined based on the Si concentration required by the steelmaking plant, which is obtained from the 24-hour steel casting plan, while taking operational conditions into account. The required Si concentration is then determined based on the finally determined steel casting plan. An ironmaking plant can determine the target value of the Si concentration by obtaining information on the changed Si concentration from the steelmaking plant (as a specific example, the target value of the corrected Si concentration and the future time or time period for correction, as described below). Further, steel casting plans in steelmaking may be changed several hours in advance due to emergency steel casting or problems. In such cases, it is possible to respond by changing the target value of the Si concentration, and the like.

[0028] Further, Si in hot metal is a heat source at the steelmaking stage. Therefore, for hot metal with the Si concentration below the target value, additional processing that leads to increased costs, such as a charging process of Si alloys or other auxiliary raw materials that serve as heat sources, is required to secure a heat source. In some cases, the target material may not be available for allocation, which may lead to reconsideration of allocation or inventory increase. As long as it is possible to change the target value of the Si concentration and the like, such a cost increase may be preventable and steelmaking costs could be lowered.

[0029] As a control action to control the Si concentration, details of an adjustment amount are explained below for a case where the Si concentration is decreased by adding a de-siliconizing agent. Based on the sensitivity of the unit consumption of the de-siliconizing agent to an amount of decrease in the Si concentration, when predicted to be excessive relative to the target value of the Si concentration, the amount of the de-siliconizing agent added may be adjusted so that the target value of the Si concentration matches the predicted value. The adjustment amount is Δu in the following expression (3). [Math. 2] Δ u = − α y pre t − y ref S u t

[0030] Here, α is the relaxation coefficient. Y ref is the target value of the Si concentration. y pre (t) is the predicted value of the Si concentration after t hours according to the statistical model or the physical model calculated by the method described above. S u (t) is the amount of change in the Si concentration after t hours of unit operation of a de-siliconizing agent, and is determined by a step response test in an actual furnace. Here, the target value of the Si concentration is assumed to be constant in expression (3), but when the target value is changed, the target value of the Si concentration after t hours may be provided as y ref (t). Here, the control action of adding de-siliconizing agent has high immediacy, and therefore it suffices to set t = 0.

[0031] For example, when the Si concentration does not reach a target even after stopping the adding of the de-siliconizing agent, a control action other than the de-siliconizing agent addition action is required. As a control action to control the Si concentration, details of an adjustment amount are explained below for a case of controlling the thermal level in a blast furnace to increase the Si concentration at the time of tapping before de-siliconizing. For example, when blast moisture is used as an operational variable to be adjusted, the adjustment amount for blast moisture is ΔBM in expression (4) below. [Math. 3] Δ BM = − β y pre t − y ref S BM t

[0032] Here, β is the relaxation coefficient. S BM (t) is the amount of change in the Si concentration after t hours of unit operation of blast moisture, and is determined by a step response test in an actual furnace or by a step response calculation using a physical model. In addition to blast moisture, similar control is possible using, for example, blast temperature or PCI rate. Here, the target value of the Si concentration is assumed to be constant in expression (4), but when the target value is changed, the target value of the Si concentration after t hours may be provided as y ref (t). Here, when these operational variables are used, t is preferably 2 to 8. Further, the response time varies depending on the operational variable, and therefore the value of t is determined according to the selected operating variable.

[0033] The calculation of the adjustment amount according to expression (3) or expression (4) may be executed according to an instruction from an operator, but according to the present embodiment, the calculation is executed automatically when the predicted value or the current value of the Si concentration deviates significantly from the target value. Here, a significant deviation from the target value may be determined by whether the magnitude of the predicted value and a value obtained by subtracting the target value from the current value exceed a threshold value. The threshold value may be determined based on, for example, the magnitude of the deviation from the target value allowed in the production of hot metal.

[0034] FIG. 5 is a diagram illustrating an example configuration of a Si concentration prediction device 10 according to an embodiment. As illustrated in FIG. 5, the Si concentration prediction device 10 includes a storage 11, a predictor 12, and an adjustment amount calculator 13. The Si concentration prediction device 10 acquires operational variables including measured values (sensor data) and predicts the Si concentration using the statistical model and the physical model described above. Further, the Si concentration prediction device 10 may acquire various instructions, including, for example, an instruction to start a prediction process. The Si concentration prediction device 10 may present the predicted value and the current value to an operator, in which case the Si concentration prediction device 10 also functions as an operation guidance device. When functioning as an operation guidance device, the Si concentration prediction device 10 causes display of the future predicted value and the current value of the Si concentration and the like on a display 30. The display 30 may be a liquid crystal display (LCD), an organic electroluminescence panel (OLED panel), or other display device. Here, the display 30 may be a display device included in a terminal device 40 used by an operator. Operators include ironmaking operators and steelmaking operators. The display 30 may display, for example, a screen such as illustrated in FIG. 4. The terminal device 40 may be a portable device such as a smartphone or tablet, for example. An operation guidance system may comprise an operation guidance device (the Si concentration prediction device 10) and the terminal device 40. In the operation guidance system, the terminal device 40 may include an acquisition unit 41, an input interface 42, a transmitter 43, and the display 30. An inter-terminal communicator 44 is described later. The acquisition unit 41 may acquire the predicted value, the current value, and the adjustment amount of the operational variables in the control action output from the operation guidance device. The input interface 42 is realized by an input device such as a keyboard or a pointing device, where a correction value of the target value of the Si concentration and the future time or time period for correction are input. For example, an input is made when an operator at a steelmaking plant, after checking the changes in the predicted value of future Si concentration and the status of changes in the steel casting plan, requests that an operator at an ironmaking plant changes the target value of future Si concentration. Requests to change the target value of the Si concentration may be made by another means of communication, not via the system. Hereinafter, the target value that has been changed by a request for change may be referred to as a corrected target value. The transmitter 43 transmits the corrected target value of the Si concentration that has been input and the future time or time period for correction to the operation guidance device (the Si concentration prediction device 10). Further, the display 30 may display the predicted value, the current value, and the adjustment amount acquired by the acquisition unit 41 as described above. The display 30 may further display the corrected target value and the future time or time period for correction. Further, sending and receiving messages between a plurality of the terminal device 40 may be possible. In this case, the display 30 has a function of displaying messages sent and received. As illustrated in FIG. 5, each of a plurality of the terminal devices 40 may include an inter-terminal communicator 44 to send and receive requests to change the target value of the Si concentration, the time period of change, the corrected target value, and the like between a plurality of the terminal devices 40. For example, information about the correction of the target value of the Si concentration may be sent and received between the terminal device 40 used by an operator of ironmaking and the terminal device 40 used by an operator of steelmaking via the respective inter-terminal communicators 44, and information sharing may take place in real time. Further, using such a function, a corrected target value for the Si concentration may be set based on input information (for example, a request to change the target value for the Si concentration as described above) from a steelmaking operator (that is, an operator other than an ironmaking operator). Here, the acquisition unit 41 may be realized, for example, by a processor (arithmetic processing unit) of the terminal device 40. Further, the number of the terminal devices 40 is not limited to one, and may be multiple as in the example above. For example, the terminal device 40 may be disposed at the respective stationing locations of a steelmaking operator and an ironmaking operator, and may be communicatively connected to the operation guidance device by a network such as a local area network (LAN) or the like. Further, the inter-terminal communicator 44 may be a communication means that uses a network, but is not limited to a specific one as long as sending and receiving information about the corrected target value of the Si concentration to and from another terminal device is possible.

[0035] Components of the Si concentration prediction device 10 are described below. The storage 11 stores the statistical model and the physical model with operational variables of a blast furnace as input and the Si concentration as output. Further, the storage 11 stores a program and data related to the prediction of the Si concentration. The storage 11 may include any storage device, such as a semiconductor storage device, an optical storage device, and a magnetic storage device. Semiconductor storage devices may include, for example, semiconductor memory. The storage 11 may include a plurality of types of storage devices.

[0036] The predictor 12 predicts the future predicted value and the current values of the Si concentration using the statistical model and the physical model. When the Si concentration prediction device 10 also functions as an operation guidance device, the predictor 12 presents the predicted value and the current value to an operator by causing display on the display 30.

[0037] The adjustment amount calculator 13 calculates the adjustment amount of the operational variables in a control action when the predicted value or the current value deviates significantly from the target value. The control action may be to decrease the Si concentration by adding de-siliconizing agent, or to control the thermal level in the blast furnace to increase the Si concentration at the time of tapping before de-siliconizing. In other words, the adjustment amount calculated by the adjustment amount calculator 13 includes adjustment amounts for both decreasing and increasing the Si concentration, including, for example, the adjustment amount of an added amount of the de-siliconizing agent. The adjustment amount calculator 13 may, in the case of a control action in which de-siliconizing agent is added, calculate the adjustment amount for the added amount of de-siliconizing agent according to expression (3). Here, when the corrected target value is input, the adjustment amount calculator 13 calculates the adjustment amount using the corrected target value. The adjustment amount calculator 13 may calculate the adjustment amount for blast moisture, blast temperature, or PCI rate as in expression (4) for a control action that controls the thermal level in the blast furnace. For example, a computer that manages the production of hot metal may automatically change conditions for the production of hot metal based on the calculated adjustment amount. Here, when the Si concentration prediction device 10 also functions as an operation guidance device, the adjustment amount calculator 13 may present the calculated adjustment amount to an operator by causing display on the display 30. Such presentation to an operator (operation guidance) may be executed as part of a method of producing hot metal or steel.

[0038] The Si concentration prediction device 10 may be realized by a computer, such as a process computer that controls the operation of a blast furnace or the production of hot metal, for example. The computer includes, for example, memory, a hard disk drive (storage device), and a CPU (processor). Various functions are realized through the organic collaboration of hardware such as a CPU and memory, OS, and required application programs. The storage 11 may be realized, for example, by a storage device. The predictor 12 and the adjustment amount calculator 13 may be realized, for example, by a processing device.

[0039] FIG. 6 is a flowchart illustrating the Si concentration prediction method according to an embodiment. The Si concentration prediction device 10 predicts the Si concentration according to the flowchart illustrated in FIG. 6. The Si concentration prediction method illustrated in FIG. 6 may be executed as part of a method of producing hot metal.

[0040] The predictor 12 predicts the future predicted value and the current value of the Si concentration using the statistical model and the physical model (step S1, prediction step). The predictor 12 may cause the predicted value and the current value to be presented to an operator.

[0041] When the predicted value or the current value of the Si concentration deviates significantly from the target value (step S2, Yes), the adjustment amount calculator 13 calculates the adjustment amount of an operational variable in a control action as described above (step S3, adjustment amount calculation step). When the corrected target value is input in the adjustment amount calculation step, the adjustment amount calculator 13 calculates the adjustment amount of the operational variable using the corrected target value as the target value. The processing of the Si concentration prediction method may further include a step in which the corrected target value is input via the input interface 42 and the transmitter 43 (input step). The series of processing in the Si concentration prediction method may be repeated in a predetermined cycle, or may be executed again, for example, when there is a change in the target value. The adjustment amount calculator 13 may cause the calculated adjustment amount to be presented to an operator. Here, when neither the predicted value nor the current value of the Si concentration deviates from the target value (step S2, No), the calculation of the adjustment amount need not be executed.

[0042] As described above, the Si concentration prediction method, the operation guidance method, the Si concentration prediction device 10, the operation guidance device, the operation guidance system, and the terminal device 40 according to the present embodiment can predict Si concentration with high precision by using the statistical model or the physical model according to the time for which a prediction is made. Further, when a steelmaking plant requests a change in the Si concentration, the target value of the Si concentration can be changed to allow the ironmaking plant to appropriately calculate a control action (adjustment amount of operational variables) according to the predicted values of the statistical model and the physical model. Therefore, deviation from the target value of the Si concentration is preventable even in operations involving an ironmaking plant and a steelmaking plant.

[0043] Although an embodiment of the present disclosure has been described based on the drawings and examples, it should be noted that a person skilled in the art may make variations and modifications based on the present disclosure. Therefore, it should be noted that such variations and modifications are included within the scope of the present disclosure. For example, functions and the like included in each component and step may be rearranged, and a plurality of components and steps may be combined into one or divided, as long as no logical inconsistency results. An embodiment according to the present disclosure may be realized as a program executed by a processor provided to a device or as a storage medium on which the program is stored. The scope of the present disclosure should be understood to include these examples.

[0044] The configuration of the Si concentration prediction device 10 illustrated in FIG. 5 is an example. The Si concentration prediction device 10 does not have to include all of the components illustrated in FIG. 5. The Si concentration prediction device 10 may include components other than those illustrated in FIG. 5. The operation guidance system includes an operation guidance device (the Si concentration prediction device 10) and the terminal device 40, but the components included in each are not limited to the example illustrated in FIG. 5. For example, it is possible to configure at least one of the predictor 12 or the adjustment amount calculator 13 to be included in the terminal device 40. Therefore, the operation guidance system may be configured so that the operation guidance device or the terminal device 40 includes the storage 11, the predictor 12, the adjustment amount calculator 13, the acquisition unit 41, the input interface 42, the transmitter 43, and the display 30.REFERENCE SIGNS LIST

[0045] 10Si concentration prediction device 11storage 12predictor 13adjustment amount calculator 30display 40terminal device 41acquisition unit 42input interface 43transmitter 44inter-terminal communicator

Claims

1. A Si concentration prediction method comprising: a prediction step of predicting a future predicted value and a current value of Si concentration using a statistical model and a physical model with operational variables of a blast furnace as input and the Si concentration as output; and an adjustment amount calculation step of calculating, when the predicted value or the current value significantly deviates from a target value, an adjustment amount of an amount of de-siliconizing agent added in a control action for lowering the Si concentration by adding the de-siliconizing agent, or an adjustment amount of the operational variables in a control action of controlling a thermal level in the blast furnace in order to increase the Si concentration at tapping before de-siliconizing.

2. The Si concentration prediction method according to claim 1, wherein, in the prediction step, the statistical model is used to predict the current value and the predicted value up to a defined time, and the physical model is used to predict the predicted value beyond the defined time, and the defined time is determined based on the time at which prediction precision of the statistical model becomes lower than prediction precision of the physical model.

3. The Si concentration prediction method according to claim 1 or 2, wherein the operational variables that are input to the statistical model include at least one of: tuyere temperature, furnace heat index, solution loss carbon amount, furnace top gas temperature, tuyere tip gas temperature, or blast volume per iron.

4. The Si concentration prediction method according to any one of claims 1 to 3, wherein the operational variables that are input to the physical model include at least one of: blast volume, blast volume oxygen, pulverized coal injection rate, coke ratio, blast moisture, or blast temperature.

5. An operation guidance method comprising a step of causing the predicted value and the current value predicted by the Si concentration prediction method according to any one of claims 1 to 4 to be presented to an operator.

6. The operation guidance method according to claim 5, further comprising an input step of inputting a corrected target value of the Si concentration, wherein the adjustment calculation step calculates the adjustment amount of the operational variables using the corrected target value input in the input step as the target value.

7. The operation guidance method according to claim 6, wherein the corrected target value is set based on input information from an operator other than an ironmaking operator.

8. A Si concentration prediction device comprising: a storage configured to store a statistical model and a physical model with operational variables of a blast furnace as input and Si concentration as output; a predictor configured to predict a future predicted value and a current value of the Si concentration using the statistical model and the physical model; and an adjustment amount calculator configured to calculate, when the predicted value or the current value significantly deviates from a target value, an adjustment amount of an amount of de-siliconizing agent added in a control action for lowering the Si concentration by adding the de-siliconizing agent, or an adjustment amount of the operational variables in a control action of controlling a thermal level in the blast furnace in order to increase the Si concentration at tapping before de-siliconizing.

9. An operation guidance device comprising: a storage configured to store a statistical model and a physical model with operational variables of a blast furnace as input and Si concentration as output; a predictor configured to predict a future predicted value and a current value of the Si concentration using the statistical model and the physical model, and to cause the predicted value and the current value to be presented to an operator; and an adjustment amount calculator configured to calculate, when the predicted value or the current value significantly deviates from a target value, an adjustment amount of an amount of de-siliconizing agent added in a control action for lowering the Si concentration by adding the de-siliconizing agent, or an adjustment amount of the operational variables in a control action of controlling a thermal level in the blast furnace in order to increase the Si concentration at tapping before de-siliconizing.

10. An operation guidance system comprising: an operation guidance device; and a terminal device, wherein the operation guidance device or the terminal device comprise: a storage configured to store a statistical model and a physical model with operational variables of a blast furnace as input and Si concentration as output; a predictor configured to predict a future predicted value and a current value of the Si concentration using the statistical model and the physical model, and to output the predicted value and the current value; an adjustment amount calculator configured to calculate, when the predicted value or the current value significantly deviates from a target value, an adjustment amount of an amount of de-siliconizing agent added in a control action for lowering the Si concentration by adding the de-siliconizing agent, or an adjustment amount of the operational variables in a control action of controlling a thermal level in the blast furnace in order to increase the Si concentration at tapping before de-siliconizing, and to output the adjustment amount; an acquisition unit configured to acquire the predicted value, the current value, and the adjustment amount; an input interface configured to accept input of a corrected target value of the Si concentration and a future time or time period for correction; a transmitter configured to transmit the corrected target value and the future time or time period for correction that have been input; and a display configured to display the predicted value, the current value, the adjustment amount, the corrected target value, and the future time or time period for correction.

11. A terminal device comprising: an acquisition unit configured to acquire a predicted value and a current value of Si concentration, and an adjustment amount, wherein the predicted value and the current value are predicted by a statistical model and a physical model with operational variables of a blast furnace as input and the Si concentration as output, and the adjustment amount is calculated when the predicted value or the current value significantly deviates from a target value, and indicates an adjustment of an amount of de-siliconizing agent added in a control action for lowering the Si concentration by adding de-siliconizing agent, or an adjustment of operational variables in a control action of controlling a thermal level in a blast furnace in order to increase the Si concentration at tapping before de-siliconizing; an input interface configured to accept input of a corrected target value of the Si concentration and a future time or time period for correction; a transmitter configured to transmit the corrected target value and the future time or time period for correction that have been input; and a display configured to display the predicted value, the current value, the adjustment amount, the corrected target value, and the future time or time period for correction.

12. The terminal device according to claim 11, further comprising an inter-terminal communicator configured to transmit and receive information about the corrected target value to and from another terminal device.

Images