JP2020105607A - Continuous casting operation support device, continuous casting operation support method, and program - Google Patents

Abstract

To predict a molten steel temperature drop from an end of the previous step to an end of the continuous casting step with higher accuracy.SOLUTION: A continuous casting operation support device includes a data input section and a temperature drop calculation section. Before the completion of a continuous casting process, the data input section inputs the operation data including predetermined values for the previous process and the continuous casting process. The temperature drop calculation section calculates the first molten steel temperature drop from an end of the previous process to an end of the continuous casting process by using the model formula which includes at least one hinge functions h (r, A) expressed by using a final target value of the molten steel temperature at the end of the continuous casting process, and an elapsed rate r (0≤r≤1) and a threshold value A of the continuous casting process.SELECTED DRAWING: Figure 3

Description

The present invention relates to a continuous casting operation support device, a continuous casting operation support method, and a program.

In the steelmaking process, by using a predetermined operation schedule and the set values of the conditions for alloy injection and secondary refining related to the manufactured material, the converter temperature is adjusted so that the predicted molten steel temperature at the start of continuous casting reaches the specified target temperature. Immediately after the completion of tapping until the molten steel tapped from the ladle to the ladle is cast by continuous casting, the target indicated values of the molten steel temperature at the start of secondary refining and at the end of secondary refining Decide before. The method of determination is based on a detailed condition table for each material and operating method, the changes in molten steel temperature between treatments and during treatment, the ladle transport time between treatments and the operating conditions during secondary refining treatment. There is a method in which the target temperature of the molten steel is determined by an optimization means such as mathematical programming based on the model represented by the regression equation based on the actual data.

However, if the specified value of the target temperature determined in this way is not appropriate and the molten steel temperature in the ladle is too low in actual operation and the molten steel temperature approaches the liquidus temperature at the end of casting or just before the end of casting, Casting may be interrupted prematurely as it is more likely to solidify in the ladle. Therefore, it is desirable to accurately predict the behavior of the molten steel temperature drop with respect to time from the start to the end of casting in continuous casting. For example, Patent Document 1 describes a technique for determining an indicated value of a target temperature by using a model of molten steel temperature change during casting using a linear regression equation.

JP, 2009-7631, A

The technique described in Patent Document 1 above assumes that the molten steel temperature during casting drops linearly with time. However, in the analysis of the operation result data by the present inventors, it is not always appropriate to assume that the temperature of the molten steel decreases linearly with time as the casting time increases and the temperature of the molten steel decreases linearly. I understood.

Therefore, the present invention provides a continuous casting operation support device, a continuous casting operation support method, and a program capable of more accurately predicting the molten steel temperature drop from the end of the previous process to the end of the continuous casting process. The purpose is to do.

According to a certain aspect of the present invention, before the end of the pre-process of the continuous casting process, a data input unit for inputting operation data including planned values for the pre-process and the continuous casting process, and the molten steel temperature at the end of the continuous casting process. And a model formula including at least one hinge function h(r,A) expressed using the progress rate r (0≦r≦1) of the continuous casting process and the threshold value A of A continuous casting operation support device is provided, which includes a temperature drop amount calculation unit that calculates a first molten steel temperature drop amount from the end of the process to the end of the continuous casting process.

According to another aspect of the present invention, before the end of the pre-process of the continuous casting process, a data input step of inputting operation data including scheduled values regarding the pre-process and the continuous casting process, and a molten steel at the end of the continuous casting process. Using a final target value of the temperature and a model formula including at least one hinge function h(r,A) expressed by using the progress rate r (0≦r≦1) of the continuous casting process and the threshold value A, There is provided a continuous casting operation support method, which comprises a temperature drop amount calculating step for calculating a first molten steel temperature drop amount from the end of the previous process to the end of the continuous casting process.

According to still another aspect of the present invention, before the end of the pre-process of the continuous casting process, a data input unit for inputting operation data including planned values regarding the pre-process and the continuous casting process, and at the end of the continuous casting process. Using a model formula including the final target value of the molten steel temperature and at least one hinge function h(r,A) expressed by using the progress rate r (0≦r≦1) of the continuous casting process and the threshold value A. A program for causing a computer to function as a temperature drop amount calculation unit that calculates a first molten steel temperature drop amount from the end of the previous process to the end of the continuous casting process is provided.

According to the configuration of the present invention as described above, the molten steel temperature drop amount from the end of the previous process to the end of the continuous casting process is calculated using the model formula including at least one hinge function h(r,A). Since the estimation is performed, the estimation accuracy is improved as compared with the conventional method.

It is a figure which shows the example of an expression of the continuous polygonal line function by a hinge function. It is a graph which shows roughly the time change of the molten steel temperature in a continuous casting process and its preceding process. It is a block diagram showing a schematic system configuration of a continuous casting operation support device concerning one embodiment of the present invention. It is a flowchart which shows the schematic process of the continuous casting operation support method which concerns on one Embodiment of this invention. In the graph showing the actual value of the change over time of the molten steel temperature drop in the tundish based on the molten steel temperature in the ladle at the end of secondary refining, and the predicted value in each of the Examples and Comparative Examples. is there.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In this specification and the drawings, constituent elements having substantially the same functional configuration are designated by the same reference numerals, and duplicate description is omitted.

In one embodiment of the present invention described below, in consideration of the fact that the molten steel temperature does not drop linearly with respect to the casting time as described above, and that the casting time to start deviating from the straight line also varies depending on the casting completion time, The relationship between the molten steel temperature drop amount and the elapsed rate of the continuous casting process from the start to the end is approximated by a continuous polygonal line. At this time, for the actual value data of the temperature drop amount, a model formula including a hinge function as shown in Formula (1), that is, a function in which the value linearly increases from 0 at the threshold A is used.
h(x,A)=0 (x<A), x−A (x≧A) (1)

FIG. 1 is a diagram showing an example of representation of a continuous polygonal line function by a hinge function. Expression (2) divides the polygonal line function shown in FIG. 1 into intervals and expresses each by a linear expression. In Expression (3), it is expressed by a linear combination of hinge functions having two different thresholds. .. As described above, by using the hinge function, continuous line functions can be expressed by a single mathematical expression.

(Target process)
The process targeted in this embodiment includes at least a continuous casting process, and includes a primary refining process and a secondary refining process as pre-processes of the continuous casting process. Specifically, the primary refining process is a process for converting hot metal or iron scrap into molten steel in a converter or an electric furnace, and the secondary refining process is for refining the molten steel in a ladle with an RH degasser or an LF furnace. It is a process to do. In the following description, for convenience of explanation, there may be a case where a converter refining process is used as the primary refining process and an RH degassing device is used as the secondary refining process. In this case, the molten steel is tapped from the converter to the ladle at the end of the primary refining process, the secondary refining process is carried out with the RH degassing device while holding the molten steel in the ladle, and then the molten steel is removed from the ladle. It is poured into a tundish of a continuous casting machine and cast as a slab in the continuous casting machine.

(Modeling of molten steel temperature drop)
FIG. 2 is a graph schematically showing changes in molten steel temperature with time in the continuous casting process and its pre-process. In FIG. 2, the vertical axis represents the molten steel temperature T and the horizontal axis represents the time t. The thin line shows the molten steel temperature in the ladle, and the thick line shows the molten steel temperature in the tundish. As shown in the figure, the molten steel temperature in the ladle falls after the steel is taken out of the converter and before the start of the secondary refining process. The amount of molten steel temperature drop during this period is ΔT ₁ . Further, the molten steel temperature in the ladle falls during the period from the start to the end of the secondary refining process. The amount of molten steel temperature drop during this period is ΔT ₂ . Further, the temperature of the molten steel in the ladle drops from the end of the secondary refining process to the end of the continuous casting process, and the temperature of the molten steel injected from the ladle into the tundish further drops. The amount of temperature drop of molten steel in the tundish based on the temperature of molten steel in the ladle at the end of the secondary refining process is ΔT ₃ . In the present embodiment, a model formula is defined for each of the molten steel temperature drop amounts ΔT ₁ , ΔT ₂ , and ΔT ₃ .

Here, in FIG. 2, the progress rate r of the continuous casting process described later is shown as a discrete value r(n). Since the elapsed rate r is a time elapsed rate when the time from the start to the end of the continuous casting step is 1, the remaining time of the continuous casting step is 1-r( when the n-th elapsed rate r(n). n). Further, in FIG. 2, the molten steel temperature drop ΔT ₃ (n in the tundish based on the molten steel temperature in the ladle at the end of the secondary refining process at the time of the n-th elapsed rate r(n). ) Is also shown.

Among the above steps, the molten steel temperature can be controlled in two ways, the converter in the primary refining step and the secondary refining step such as the RH degassing device or the LF furnace. Therefore, the molten steel temperature at the end of these steps can be controlled. Use as a control variable. Since there are three control variables to which the target value of molten steel temperature at the end of continuous casting is added, it is rational to create model equations in the above three sections and make simultaneous equations.

(Explanation of symbols)
The symbols of the explanatory variables in the model formula are defined as follows.
u ₁ : Molten steel temperature at the end of converter tapping u ₂ : Molten steel temperature at the end of secondary refining process x _i : Operation data item r: Continuous casting process progress rate

Here, the operation result data variable x _i is, for example, the amount of input alloy in the converter, the time required from the steel output from the converter to the start of the secondary refining process, the processing time of the secondary refining process, and the secondary refining process. Input amount of alloy, time required from the end of the secondary refining process to the start of the continuous casting process, the elapsed time from the start of the continuous casting process, the number of times the ladle is used, and the operation this time since the end of the continuous casting process of the previous operation It includes the time when the ladle was empty before the start of tapping.

A model formula representing the relationship between the molten steel temperature drop ΔT ₁ , ΔT ₂ , ΔT ₃ and the operating conditions in each process is defined as the following 1) to 3) by the linear combination of the above variable data.

1) A model formula expressing the molten steel temperature drop ΔT ₁ from the end of converter tapping to the start of secondary refining process is represented by F ₁ and a _i (i=0,..., N) as regression coefficients and constant terms. ) Is used to define as a formula (4) by linear combination of explanatory variables.

2) a model expression representing the molten steel temperature drop [Delta] T ₂ from the start in the secondary refining process, b _{i (i} = 0 as the regression coefficients and constant term, · · ·, N) using, explanatory variables linear It is defined by equation (5) by combining.

3) The model formula representing the molten steel temperature drop amount ΔT ₃ from the end of the secondary refining process to the end of the continuous casting process is calculated by using the progress rate r (0≦r≦1) of the continuous casting process and a plurality of different thresholds A. A plurality of hinge functions h(r,A _k ) represented by using _k (k=1,...,K) and a hinge function h(r,A _k ) are determined recursively. The coefficient α _k and the regression coefficient and the constant terms F ₂ and c _i (i=0,..., M), α _k are used to linearly combine the explanatory variables as shown in Expression (6). Define.

Here, the elapsed rate r is a time elapsed rate when the time from the start to the end of the continuous casting step is 1. The above-mentioned hinge function h(r,A _k ) is an extension of the single hinge function h(r,A) expressed by using the elapsed rate r and the threshold value A (0<A<1). .. Here, K represents the maximum number of hinge functions h(r, A _k ) included in the model formula. Since the molten steel temperature drop amount ΔT ₃ changes so that the drop rate per unit time increases with the progress of casting, it is preferably set in advance to about K=2 to 5, for example. However, the number of hinge functions h(r, A _k ) included in the actual model formula is equal to the number of non-zero coefficients α _k . Note that, hereinafter, A _k is also referred to as a progress rate threshold group.

(Parameter determination method for each temperature drop model formula)
Next, based on the operation result data, the parameters of the temperature drop amount model equation defined above, specifically, the linear combination coefficients a _i , b _i , c _i , α _k , F ₁ , F ₂ and the hinge function Determine the number k. Also, the elapsed rate threshold value group A _k is given in advance as a constant. In the following example, N sets of operation result data samples are {u ₁ (n), u ₂ (n), x _i (n), r(n), ΔT ₁ (n), ΔT ₂ (n). , ΔT ₃ (n)} (n=1,..., N) is obtained.

In the model formula expressing the molten steel temperature drop amount ΔT ₁ from the end of the converter tapping to the start of the secondary refining process, defined as the above equation (4), the parameters a ₀ , a _i (i=1,... , M) and F ₁ are coefficients to be obtained, and the sum of squares S ₁ of the deviations of the left side and the right side of the equation (4) in the data sample is defined by the equation (7) to determine the optimization problem to be minimized. The coefficient that minimizes S ₁ can be calculated by the method of least squares.

The parameters b ₀ , b _i (i=1,..., M) should be obtained in the model formula expressing the molten steel temperature drop ΔT ₂ from the start in the secondary refining process defined as the above formula (5). and coefficients, by defining the optimization problem for minimizing defined in formula left and right sides of the deviation of the sum of squares S ₂ the formula (5) (8) in the data samples, minimizing the coefficient of the S ₂ to minimize It can be calculated by the square method.

In the model formula expressing the molten steel temperature drop amount ΔT ₃ from the end of the secondary refining process at the time of the progress rate r of the continuous casting process defined as the above formula (6), the parameters F ₂ , c ₀ , c _i ( i=1,..., M), α _k (k=1,..., K) are coefficients to be obtained, and the sum of squares S ₃ of the deviations of the left side and the right side of the equation (6) in the data sample is given by the equation By defining the optimization problem to be defined and minimized in (9), the coefficient that minimizes S ₃ can be calculated by the least square method.

(Optimal combination of hinge functions)
Selection of explanatory variables in the above equation (9), that is, among linear coupling coefficients F ₂ , c _i (i=0,..., M) and α _k (k=1,..., K) By performing what is called "model selection", in which what is made zero is appropriately selected, it is possible to improve the prediction accuracy of the temperature drop amount for new data that is not used in the regression calculation of the parameters by the least squares method. In the present embodiment, in particular, by adding α _k to the model selection target, the number and positions of the bending points of the polygonal line approximation function are selected within a range of preselected positions.

As an index used for model selection, for example, the Akaike's Information Criterion (AIC) is known. The value of AIC is defined as −2×(maximum log likelihood for the actual data of the probabilistic model)+2×(number of model parameters). The model selection by AIC is a method of selecting a good model, specifically, a model closer to a true model, by selecting an explanatory variable so as to minimize the value of AIC in regression calculation.

In the present embodiment, the “maximum log-likelihood for the actual data of the probabilistic model” used for calculating the AIC value is S ₃ calculated by the equation (9), and the “number of model parameters” is not set to zero at the time of regression calculation. The number of coefficients. For the search for a combination of explanatory variables, for example, a set of n explanatory variables selected from a set of operation data items is prepared, and n−1 explanatory variables of the n explanatory variables excluding the explanatory variable are prepared. A step of selecting an explanatory variable having the largest difference between the AIC value in the variable set and the original AIC value, or n+1 in which an operation data item not included in the explanatory variable set is added as an explanatory variable. Performing the step of selecting the operation data item having the largest difference between the AIC value in the set of individual explanatory variables and the AIC value of the original set by the so-called greedy method, which is repeated until the AIC value becomes the minimum You can

(System configuration)
FIG. 3 is a block diagram showing a schematic system configuration of a continuous casting operation support device according to an embodiment of the present invention. As shown in FIG. 3, the continuous casting operation support device 100 includes a data input unit 110, a final target value setting unit 120, a temperature drop amount calculation unit 130, an intermediate target value calculation unit 140, and a target value output unit. Includes 150 and an instruction value determining unit 160. In the description with reference to FIG. 3, specific means for mounting each part of the continuous casting operation support device 100 will be described, and details of the function of each part will be described later.

The continuous casting operation support device 100 includes, for example, a CPU (Central Processing Unit), a storage device, a communication device, an input/output unit, and the like, and is implemented by a computer that executes various calculations according to a program. The computer may be used exclusively as the continuous casting operation support device 100, or may function as the continuous casting operation support device 100 when a versatile computer operates according to a specific program. The program is stored in a storage device of the computer or stored in a removable storage medium and read by the computer.

Of the components of the continuous casting operation support device 100 described above, the data input unit 110 corresponds to a program module configured to input the planned operation value and the processing result data. Specifically, the data input unit 110 reads the planned operation value and the processing result data stored in the storage device of the computer or the removable storage medium, or accepts the input of the planned operation value and the processing result data. When the data of the scheduled operation value is stored in an apparatus external to the continuous casting operation support apparatus 100 or is input to the external apparatus, the data input unit 110 may execute communication with the apparatus.

On the other hand, the final target value setting unit 120, the temperature drop amount calculation unit 130, the intermediate target value calculation unit 140, the target value output unit 150, and the instruction value determination unit 160 are program modules configured to execute a predetermined calculation. Corresponding to. The temperature drop amount calculation unit 130 refers to the model formula 170 stored in the storage device of the computer or the removable storage medium. The instruction value determined by the instruction value determination unit 160 may be stored in a storage device of a computer or a removable storage medium, or may be transmitted to an external device.

(Processing flow)
FIG. 4 is a flowchart showing schematic steps of the continuous casting operation support method according to the embodiment of the present invention. In the present embodiment, the continuous casting operation support method is executed in the continuous casting operation support device 100 described above with reference to FIG. In the model formula 170, it is assumed that parameters such as a regression coefficient are determined in advance based on operation performance data.

First, before the start of the primary refining process or before the start of the secondary refining process, the data input unit 110 inputs the scheduled operation value and the processing result data (step S110). These data are operation data related to the preceding process (primary refining process and secondary refining process) and continuous casting process, which are explanatory variables in the model formulas of formulas (4), (5), and (6), and are the actual values. Including the planned value. The scheduled operation data will be for operation processes that have not been completed when the operation data is input. The operation data may be input after the start of the secondary refining process, but the temperature of the molten steel can be controlled only during the secondary refining process, so at least before the end of the secondary refining process, input of the operation data and Based on this, calculation of an intermediate target value (described later) and determination of a temperature manipulated variable (described later) instruction value are executed.

Next, the final target value setting unit 120 sets the final target value T ₃ ^* of the molten steel temperature at the end of the continuous casting process (step S120). The final target value T ₃ ^* of the molten steel temperature is the tank temperature immediately before all the molten steel injected from the ladle into the tundish is poured into the continuous casting machine mold (that is, at the end of the continuous casting step in the present embodiment). This is the target value of the molten steel temperature in the dish. Specifically, the final target value T ₃ ^* of the molten steel temperature can be set as a temperature obtained by adding a margin in consideration of operational variations to the liquidus temperature.

Next, the temperature drop amount calculation unit 130 calculates the molten steel temperature drop amount ΔT ₃ using the model formula of the above formula (6) (step S130), and the intermediate target value calculation unit 140 determines the final value determined in step S120. An intermediate target value u ₂ ^* of the molten steel temperature at the end of the secondary refining process is calculated based on the target value T ₃ ^* and the molten steel temperature drop ΔT ₃ calculated in step S130 (step S140). Specifically, for the intermediate target value u ₂ ^* , the planned operation value and the actual processing value input in step S110 are set in the explanatory variable item in the above equation (6), and the molten steel at the end of the continuous casting process is set. The temperature drop amount ΔT ₃ is u ₂ ^* −T ₃ ^* , and the continuous casting process progress rate r is set to 1. By solving for u ₂ ^*, it can be calculated as in the following formula (10).

Further, in step S130, the temperature drop amount calculation unit 130 calculates the molten steel temperature drop amounts ΔT ₁ and ΔT ₂ using the model formulas of the above formulas (4) and (5), and in step S140, the intermediate target value calculation unit. 140 may calculate the intermediate target value u ₁ ^* of the molten steel temperature at the end of the primary refining process, that is, at the end of the tapping of the converter, based on the molten steel temperature drop amounts ΔT ₁ and ΔT ₂ . Specifically, the intermediate target value u ₁ ^* is calculated by setting the molten steel temperature drop ΔT ₂ by setting the scheduled operation value and the actual processing value input in step S110 in the explanatory variable item in the above equation (5). The planned operation value and the processing actual value input in step S110 are set in the item of the explanatory variable in the equation (4), and the molten steel temperature drop amount ΔT ₁ is set to u ₁ ^* −(u ₂ ^* +ΔT ₂ ). By solving the above for u ₁ ^* , it is possible to calculate as in the following Expression (11).

Next, the target value output unit 150 outputs the intermediate target values u ₁ ^* , u ₂ ^* calculated in step S140 (step S150). Specifically, for example, the target value output unit 150 may output the intermediate target values u ₁ ^* , u ₂ ^* to a display device such as the display 210 connected to the continuous casting operation support device 100. In this case, the intermediate target values u ₁ ^* , u ₂ ^* are referred to by the operator. Alternatively, when the instruction value determination unit 160 automatically determines the instruction value of the temperature manipulated variable, the target value output unit 150 outputs the intermediate target values u ₁ ^* , u ₂ ^* to the instruction value determination unit 160.

Next, the instruction value determination unit 160 determines the instruction value of the temperature manipulated variable in the previous process of the continuous casting process based on the intermediate target values u ₁ ^* , u ₂ ^* calculated in step S140 (step S160). .. When only the intermediate target value u ₂ ^* is calculated in step S140, the instruction value determination unit 160 determines the instruction value of the temperature manipulated variable in the secondary refining process. Furthermore, when the intermediate target value u ₁ ^* is also calculated in step S140, the instruction value determination unit 160 also determines the instruction value of the temperature manipulated variable in the primary refining process.

Specifically, the temperature manipulated variable includes an input amount of an auxiliary raw material such as graphite or aluminum that raises the molten steel temperature or an auxiliary raw material such as iron scrap that lowers the molten steel temperature. The instruction value determination unit 160 may determine the basic unit for the molten steel weight as the instruction value of the manipulated value, and the auxiliary raw material may be charged according to the basic unit. For example, the instruction value determination unit 160 determines the instruction value of the temperature operation amount according to the input performed by the operator on the input device 220 with reference to the intermediate target values u ₁ ^* and u ₂ ^* output in step S150. Alternatively, the instruction value determination unit 160 may use the intermediate target values u ₁ ^* , u ₂ ^* output from the target value output unit 150 for the input amount of the auxiliary material stored in the storage device of the computer or the removable storage medium. The reference value of the temperature manipulated variable may be automatically determined by referring to the model formula or table of the molten steel temperature change. The determined instruction value of the temperature manipulated variable is transmitted to, for example, an output device such as a display referred to by an operator of the auxiliary raw material charging device, or an automatic raw material charging device.

In one embodiment of the present invention as described above, the molten steel temperature drop during continuous casting is accurately estimated before the start of converter steel output or before the start of the secondary refining process, and the end of the continuous casting process. It is possible to determine the instruction value of the temperature manipulated variable in the primary and secondary refining processes so that an appropriate molten steel temperature can be achieved at any time. Thereby, for example, it is possible to prevent the end of casting due to the molten steel temperature lowering than the predicted value and decreasing at the end of casting.

Next, examples of the present invention will be described. In the present example, first, regarding the explanatory variables of the model formula shown in the above formula (6), the molten steel temperature u ₂ after the secondary refining, the processing time of the secondary refining process, the amount of alloy input in the secondary refining process, With respect to the time required from the end of the secondary refining process to the start of the continuous casting process, the molten steel weight, and the molten steel temperature in the tundish during the continuous casting process, total operation data of 152 charges was collected. Next, in order to estimate the accuracy of the model formula, the prediction accuracy of the molten steel temperature drop amount by the model formula for new data was estimated by the following cross-validation method.

In the test, first, 152 charges are divided into approximately equal numbers into 10 sets, one set is set aside for prediction accuracy estimation, and parameters such as regression coefficients are calculated using the remaining 9 sets of data. By setting (0.2, 0.4, 0.6, 0.8) in the elapsed rate threshold value group A _k and selecting the model by the AIC as described above, the regression coefficient and the hinge in the model formula of the formula (6) Determine the function. By substituting the explanatory variables in the set of stored charges into the equation (6) including the above regression coefficient and hinge function, the molten steel temperature drop ΔT ₃ was calculated, and the molten steel temperature in the tundish during the continuous casting process was calculated. The prediction deviation was calculated by comparing with the actual measurement value. The above-described processing was repeated while changing the data set aside for estimating the prediction accuracy, and the statistical amount related to the predicted deviation of the obtained molten steel temperature drop ΔT ₃ was evaluated.

In addition, as a comparative example, when the molten steel temperature drop during the continuous casting process is not approximated by a polygonal line, that is, the model _obtained by removing the hinge function h(r, A _k ) from the model formula in the formula (6) is similarly cross-validated. Prediction deviation data was calculated by the method.

Table 1 shows the results of comparison between the examples and the comparative examples with respect to the statistical amount regarding the predicted deviation with respect to the total 56 data in the final stage of continuous casting with the continuous casting process progress rate r≧80%. Since the comparative example does not reflect that the molten steel temperature drop rate during casting increases with time, the predicted value of the temperature drop becomes smaller than the actual value. Therefore, the predicted deviation average of the comparative example has a relatively large negative value. On the other hand, in the example, since the molten steel temperature drop rate during casting increases with the passage of time, the predicted deviation average becomes a value close to zero. Here, the standard deviation of the predicted deviation is larger in the example than in the comparative example, but the difference is as small as 0.18° C., and the hypothesis that the average of the predicted deviations of both is the same is 1% level. It can be rejected by a statistical test (t-test).

FIG. 5 shows the molten steel temperature drop amount in the tundish based on the molten steel temperature in the ladle at the end of secondary refining in the continuous casting process of a certain charge (corresponding to the above-mentioned molten steel temperature drop amount ΔT ₃ ). 5 is a graph showing the actual value of change for each elapsed time and the predicted value in each of the example and the comparative example. The measurement of the actual value and the calculation of the predicted value are carried out at the same elapsed time, and the continuous casting process progress rate r at the final elapsed time is 88.6%. At the actual values, it is observed that the molten steel temperature drop rate during casting increases with the passage of time. On the other hand, the predicted value of the temperature drop amount in the comparative example has a larger deviation from the actual value toward the end of casting because the temperature drop amount is calculated linearly with the passage of time. On the other hand, the predicted value of the temperature drop amount in the example follows the change of the actual value even near the end of casting, that is, it can be confirmed that the molten steel temperature drop rate during casting increases with the passage of time. ..

This is because the molten steel in the ladle is filled up to near the pot lid at the start of continuous casting, but the amount of molten steel decreases as the casting progresses, and the surface of molten steel lowers. It is thought to do. Since the amount of molten steel decreases and the heat outflow rate per weight of molten steel increases, the rate of decrease in molten steel temperature is not constant but increases with the progress of casting. In the embodiment of the present invention, such behavior of the molten steel temperature drop can be accurately approximated by a continuous polygonal line function represented by a hinge function.

The preferred embodiments of the present invention have been described above in detail with reference to the accompanying drawings, but the present invention is not limited to these examples. It is obvious that a person having ordinary knowledge in the technical field to which the present invention pertains can come up with various changes or modifications within the scope of the technical idea described in the claims. Of course, it is understood that these also belong to the technical scope of the present invention.

100... Continuous casting operation support device, 110... Data input unit, 120... Final target value setting unit, 140... Intermediate target value calculation unit, 150... Target value output unit, 160... Instructed value determination unit, 170... Model formula, 210 ... display, 220... input device.

Claims (6)

Before the end of the pre-process of the continuous casting process, a data input unit for inputting operation data including scheduled values for the pre-process and the continuous casting process,
A final target value of the molten steel temperature at the end of the continuous casting step, and at least one hinge function h(r, which is expressed using the progress rate r (0≦r≦1) and the threshold value A of the continuous casting step. A temperature drop amount calculation unit that calculates a first molten steel temperature drop amount from the end of the previous process to the end of the continuous casting process using a model formula including A);
A continuous casting operation support device equipped with. An intermediate target value calculation unit that calculates an intermediate target value of the molten steel temperature at the end of the preceding step, based on the first molten steel temperature drop amount calculated by the temperature drop amount calculation unit,
The continuous casting operation support device according to claim 1, further comprising: an instruction value determination unit that determines an instruction value of the temperature manipulated variable in the previous step based on the intermediate target value calculated by the intermediate target value calculation unit. .. The preceding step includes a primary refining step and a secondary refining step,
The temperature drop amount calculation unit includes a second molten steel temperature drop amount from the end of the primary refining process to the start of the secondary refining process, and from the start of the secondary refining process to the secondary refining process. Calculate the third molten steel temperature drop until the end based on the model formula,
The intermediate target value calculator calculates the intermediate target value of the molten steel temperature at the end of the primary refining process and the secondary refining process based on the second molten steel temperature drop amount and the third molten steel temperature drop amount. The continuous casting operation support device according to claim 2, which calculates an intermediate target value of the molten steel temperature at the end. The model formula for calculating the first molten steel temperature drop amount is a plurality of hinge functions represented by using the elapsed rate r and a plurality of different thresholds A _k (k=1,..., K). The continuous casting operation support according to any one of claims 1 to 3, comprising (r, A _k ) and a coefficient α _k recursively determined for each of the plurality of hinge functions. apparatus. Before the end of the pre-process of the continuous casting process, a data input step of inputting operation data including scheduled values for the pre-process and the continuous casting process,
A final target value of the molten steel temperature at the end of the continuous casting step, and at least one hinge function h(r, which is expressed using the progress rate r (0≦r≦1) and the threshold value A of the continuous casting step. A continuous casting operation support method, comprising: using a model formula including A), a temperature drop calculating step of calculating a first molten steel temperature drop from the end of the previous step to the end of the continuous casting step. .. Before the end of the pre-process of the continuous casting process, a data input unit for inputting operation data including scheduled values for the pre-process and the continuous casting process,
A final target value of the molten steel temperature at the end of the continuous casting step, and at least one hinge function h(r, which is expressed using the progress rate r (0≦r≦1) and the threshold value A of the continuous casting step. A program for causing a computer to function as a temperature drop amount calculation unit that calculates a first molten steel temperature drop amount from the end of the previous process to the end of the continuous casting process using a model formula including A). ..

Images