JPH09316516A - Method of regulating components of molten steel - Google Patents

Abstract

PROBLEM TO BE SOLVED: To optimally select the brand of alloy material and to accurately compute its charge by determining a charge of alloy material from the cost of the alloy material and the deviation of the estimated ultimate amount of each element after the charging of the alloy material from the desired amount by means of linear programming at the time of component regulation of a molten steel. SOLUTION: At the time of regulating components in a molten steel, the computing of components is performed by using a computer. A charge of alloy material is computed by means of linear programming, based on the desired components of the molten steel, the values in the actual conditions of the molten steel, the unit price of an alloy material to be charged, respective contents of alloying elements in the alloy material, the yield of the alloy material into the molten steel, and the amount of the molten steel. The charge of the alloy material, capable of minimizing the objective function showing the cost of the alloy material and the sum of the deviations of the estimated ultimate amounts of respective alloying elements after the charging of the alloy material from the desired components, is determined. By this method, the optimum brand of the alloy material can be selected and its charge can be accurately computed even in the case where the content of a certain component in the molten steel exceeds the desired value in the stage of the conclusion of refining, and the automation of alloy computing and the reduction of alloy material cost can be attained.

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to adjusting the composition of molten steel by adding an alloy material such as ferroalloy or a deoxidizing agent before tapping molten steel refined in a refining furnace such as a converter or an electric furnace. The present invention relates to a method capable of accurately adjusting the composition closer to the target composition at a lower cost in a short time by taking into consideration the unit price of the alloy material brand, the content of the alloy element, and the like.

[0002]

2. Description of the Related Art When tapping molten steel refined in a refining furnace such as a converter or an electric furnace, the composition of the molten steel is adjusted by adding an alloying component and a deoxidizing agent as a final step. This component adjustment work must be performed within an extremely short time from the end of refining to tapping, and in the meantime, a series of work such as calculation of the addition amount of alloy material, weighing, and charging must be performed quickly. I won't. Such work has been performed by a skilled person using a calculator, a slide rule, or the like according to the analysis value of the molten steel component after the blowing is stopped.

On the other hand, there are many brands of alloy materials delivered for composition adjustment (usually mainly ferroalloys are used), and the unit price and content of alloying elements vary depending on the brand, and blow-stopping is performed. Selecting the optimum brand of alloy material according to the current value (analytical value) of molten steel and the target component amount reduces the unit cost of the alloy material added for component adjustment and promotes cost reduction of steel materials. It becomes a big factor above.

However, as described above, it is difficult to manually calculate the optimum alloy material brand and the input amount according to the target component amount in an extremely short time before tapping. Therefore, a method of performing those calculations in a short time using a computer that has been rapidly developed recently, that is, the current value and target content of each alloying element in molten steel, and the unit price of each brand of alloying material to be added Based on the content of each alloy element and the amount of molten steel, the following composition adjustment formula (1 ') and the non-negative formula (2') of the alloy material input amount are used, and the alloy material cost formula (3 ')
A method has been established for determining the amount of alloy material input that can minimize the above, and the cost required for the alloy material has been reduced. T ⁱ _E −T ⁱ _S = [Σ _{m = 1} (W _m · C ⁱ _m · Y _i ) / W _Fe ] ... (1 ′) W _m ≧ 0 ... (2 ′) Σ _{m = 1} (A _m · W _m · P _m ) ... (3 ′) (where T ⁱ _E is the target content of alloying element i in the molten steel, T ⁱ _S is the current value of alloying element i in the molten steel, W _m yield W _Fe is the molten steel amount a _m is a coefficient P _m the alloy material of the content Y _i of input amount C ⁱ _m is the alloy element i in the alloy material stocks m of alloy material stocks m in molten steel alloying elements i Represents the unit price of the brand m).

[0005] For example, "Steel Association Lecture Preliminary Collection CP
MP-ISI "Vol. 3 (1990) -288,
Unit price and alloy element content of each added alloy material by brand,
From the six items of yield in molten steel, upper and lower limits of target component value, blowing temperature, blowing condition and oxygen concentration in blowing steel, the total alloy material cost is minimized by taking into consideration the effect of temperature drop. Therefore, a method of determining the brand of additive alloy material and the addition amount according to the linear programming method has also been proposed.

However, among the steel types to be melted,
Due to operational variations, the target component may exceed the upper limit at the end of melting in the refining furnace, or the alloy composition of any brand may not be adjusted to the target component.
With the above calculation method, it is not possible to obtain an effective solution and complete automation cannot be achieved. That is, even in the method using the linear programming method, when the molten steel component value obtained by calculation does not satisfy the constraint equation such as deviating from the upper limit value of the molten steel component in the refining furnace, the execution solution cannot be obtained.

Moreover, in the prior art including the above method,
The adjustment method that takes into account even the impurities contained in the input alloy material has not been established, and when selecting an alloy material with a high impurity content brand, the impurity content often exceeds the specified range. Have been experienced.

Further, in the conventional method, the amount of alloy material input is calculated from the component analysis value (including the estimated value) after refining. Therefore, when the amount of alloy material input is large, the calculated amount of alloy material is weighed. Since it cannot be performed within a short time after the refining is completed, there is a problem that the tapping work is delayed and the subsequent operation line is disturbed, or the composition cannot be adjusted as intended.
When the number of input hoppers installed in the refining furnace is limited, it may be necessary to input in bag-filling units. In this case, the input unit is the weight of one bag. Cannot be solved by conventional linear programming.

[0009]

The present invention has been made by paying attention to the problems as described above, and its purpose is to ensure that the content of a certain component in molten steel at the end of melting in a refining furnace is Even if the target value is exceeded, it is possible to accurately select the optimum alloy material brand and its input amount, and to achieve complete automation of alloy calculation and reduction of alloy material cost. It is intended to establish a possible component adjustment method.

Another object of the present invention is to consider the content of impurities contained in the alloy material used for component adjustment, and to select an appropriate alloy material brand without producing defective molten steel due to impurity contamination. It aims to make it possible to calculate the input amount.

Further, even when it is assumed that the amount of the alloy material charged becomes large and the weight of the alloy material cannot be kept in time within a short time before tapping, it is possible to prevent the alloy material from blowing 2
By inputting in batches, it is possible to make a reliable component adjustment without delaying the tapping operation and subsequent disruption of the operation line, and furthermore, the number of hoppers that are permanently equipped in the refining furnace is limited. Even if it is necessary to input the alloy material in a predetermined weight unit, it is possible to determine the most appropriate alloy material brand and input amount by linear programming considering the unit input amount. It is about trying to establish a method.

[0012]

The composition adjusting method according to the present invention, which was able to solve the above-mentioned problems, is the current state of each alloying element in the molten steel when the alloy material is added to the molten steel to adjust the composition of the molten steel. Value and target content, unit price of each brand of the alloy material and content of each alloying element and yield in molten steel,
In addition, based on the molten steel amount, in the method of adjusting the input amount of alloy material by linear programming to adjust the composition of molten steel,
A method for adjusting the composition of molten steel, characterized in that the alloy material cost is determined, and the alloy material input amount is determined in consideration of the balance between the expected arrival amount of each alloy element after the alloy material is input and the deviation of the target content. is there. Further, it is a component adjusting method for molten steel, in which the amount of alloy material to be charged is determined in consideration of both or one of an upper limit constraint value and a lower limit constraint value which are defined in advance for each alloy element.

More specifically, the method for adjusting the composition of molten steel according to the present invention, in which the alloy material is added to the molten steel to adjust the composition of the molten steel, the present value and target content of each alloying element in the molten steel, Based on the unit price of each brand of the alloy material and the content of each alloying element and the yield in molten steel, and the amount of molten steel,
In the method for adjusting the composition of molten steel by setting the composition of the alloy material by the linear programming method, the composition adjustment formula represented by the following formula (1) and the non-negative composition of the alloy material represented by the following formula (2) are used. Using a formula, the linear programming method determines that the target function shown by the following formula (3), which shows the sum of the difference between the alloy material cost and the expected arrival amount for each alloy element after the alloy material is input, and the target content is the minimum. As described above, the feature is that the amount of alloy material input is determined. (T ⁱ _E −T ⁱ _S ) + ΔD _i = [Σ _{m = 1} (W _m · C ⁱ _m · Y _i ) / W _Fe ] (1) W _m ≧ 0 (2) Σ _{m = 1} (A _m · W _m · P _m ) + Σ _{i = 1} (B _i · | ΔD _i |) (3) (where T ⁱ _E is the target content of the alloy element i in the molten steel T ⁱ _S is the current value of the alloy element i in the molten steel W _m is the input amount of the alloy material brand m C ⁱ _m is the content of the alloy element i in the alloy material brand m Y _i is the yield of the alloy element i in the molten steel W _Fe is the amount of molten steel ΔD _i is the amount of alloying material, the content of alloying element i in the alloying material and the yield of alloying element i in the molten steel calculated from the yield in the molten steel, and the alloy Differences from the target content of the element i A _m and B _i are coefficients P _m is the unit price of the alloy material brand m).

Alternatively, a predetermined upper limit constraint value and a lower limit constraint value of each adjustment component, an upper limit allowable value that can be arbitrarily set between the target value and the upper limit constraint value, and the target value and the lower limit constraint. Membership function for each alloying element and the above (1) and (2) determined from the lower limit allowable value that can be set arbitrarily between the values by using the following equations (4) and (5) Using the formula, by linear programming, the following formula (6) indicating the sum of the alloy material cost, the expected arrival amount and target content of each alloy element after the alloy material is input, and the membership function value is shown. The feature is that the input amount of the alloy material is determined so that the target function to be obtained is minimized. f1 = {[Σ _{m = 1} (W _m · C ⁱ _m · Y _i ) / W _Fe ] + T ⁱ _s −U _ki } / (U _i −U _ki ) ... (4) f2 = 1-{[ Σ _{m = 1} (W _m · C ⁱ _m · Y _i ) / W _Fe ] + T ⁱ _s −L _i } / (L _ki −L _i ) ... (5) λ = 0 (f1 ≦ 0 and f2 ≦ 0) λ = f1 (when f1> 0 and f2 ≦ 0) λ = f2 (when f1 ≦ 0 and f2> 0) Σ _{m = 1} (A _m · W _m · P _m ) + Σ _{i = 1} (B _i · | ΔD _i |) + Σ _{i = 1} α _i · λ _i (6) (wherein U _ki is the upper limit allowable value of alloy element i in molten steel U _i is the alloy element i in molten steel) Upper limit constraint value L _ki is a lower limit allowable value of alloy element i in molten steel L _i is a lower limit constraint value of alloy element i in molten steel α _i is a coefficient λ _i is a membership function value of alloy element i in molten steel ).

Further, in the component adjusting method according to the present invention, the membership function is provided with an appropriate upper limit value, and when the membership function value exceeds the upper limit value, the component adjusting formula (1) is used. Using the formula and the non-negative formula of the alloy material input amount shown in the above equation (2), the linear material programming is used to calculate the alloy material cost and the expected arrival amount and target content of each alloy element after the alloy material is input. The input amount of the alloy material is determined so that the target function represented by the equation (3), which indicates the sum of the deviations, is minimized. If the membership function value is equal to or less than the upper limit value, the membership function is determined. Using the function and the above equations (1) and (2), the linear programming method is used to calculate the alloy material cost, the difference between the expected arrival amount and the target content of each alloy element after the alloy material is charged, and the membership function value. Formula (6) indicating the sum
It is also possible to determine the amount of alloy material input so that the target function shown by is minimum.

In carrying out the above component adjusting method, if the following formula (7) indicating the factor of the impurity element is added and the amount of alloy material input is adjusted by the linear programming method, it may be mixed in the molten steel. It is possible to prevent generation of defective molten steel due to impurities.

T ^b _E −T ^b _S ≧ [Σ _{b = 1} (W _m · C ^b _m · Y _b ) / W _Fe ] (7) (where T ^b _E is contained in the molten steel) Limit content of impurity element b T ^b _S is current value of impurity element b in molten steel C ^b _m is content of impurity element b in alloy material brand m Y _b represents yield of impurity element b to molten steel) .

Further, when the upper limit value and the lower limit value of the alloy material input amount are limited due to restrictions on equipment, etc., they are added as the following formulas (8a) and (8b) to add the alloy material input amount. It is preferable to adopt the method of determination because it can be put to practical use according to the equipment without difficulty. W _m ≧ L _m (8a), W _m ≦ U _m (8b) (where L _m is the lower limit of the amount of alloy material brand m input, and U _m is the amount of alloy material brand m input. Represents the upper limit of).

Further, in the case where it is assumed that the amount of alloy material to be used for adjusting the composition becomes large and the weight of the alloy material cannot be met in time within a short time before tapping, the refining furnace is used. In the final stage of refining, if the method of adding alloy material in two steps before and after blowing is adopted, delay of tapping work and disturbance of the operation line after that can be avoided.
Refining equipment If the number of hoppers that are permanently equipped is limited and it is necessary to input alloy material in a predetermined weight unit, combine the branching and limiting method with the linear programming method and set the alloy material input amount. It becomes possible to deal with it easily by making a decision.

[0020]

BEST MODE FOR CARRYING OUT THE INVENTION Unlike the prior art, the method for adjusting the composition of molten steel according to the present invention considers the deviation between the expected arrival amount and the target content of each alloying element to be adjusted after the alloy material is charged. By doing so, even if it is not possible to match the molten steel composition with the target composition due to the alloy material brand you have,
It is intended to adjust the balance between the deviation from the target component and the alloy material cost in a good balance. In this way, by considering the alloy material cost and the deviation of the expected arrival amount and target content after the alloy material is input for each alloy element, if you want to suppress the deviation of the components even if the cost becomes slightly higher, Even if the component deviation is somewhat large, even when the cost is desired to be reduced, by changing the weighting of the two, a well-balanced adjustment becomes possible.

Further, when the upper limit constraint value and the lower limit constraint value are defined in advance for each alloy element to be adjusted, the upper limit constraint value and the lower limit constraint after the alloy material is charged for each alloy element. In addition to taking into account either or both values,
Alloy material cost, deviation of expected arrival amount and target content after alloy material input for each alloy element, and expected arrival amount after alloy material injection for each alloy element and deviation of upper limit value or lower limit value The aim is to determine the amount of alloy material to be added so that the balance of personnel is improved.

In carrying out the present invention specifically,
When controlling the composition calculation when adjusting the composition of molten steel by adding alloy materials and deoxidizers, when controlling using a computer, the target composition of the molten steel to be manufactured, the actual value of the actual molten steel, the alloy material to be input Unit price, alloying element content and yield in molten steel,
Based on the molten steel amount, I calculate the alloy material input amount by linear programming using the above component adjustment formula (1) and the non-negative expression (2) of the alloy material input amount. The amount of alloy material that can minimize the target function (3) indicating the sum of the expected amount of arrival after the alloy material is injected and the deviation of the target component for each alloy element is determined (FIG. 1). Or II, the upper limit constraint value and the lower limit constraint value for each alloy element that are predetermined, the upper limit allowable value that can be arbitrarily set between the target value and the upper limit constraint value, and the target value and the lower limit constraint value Between the lower limit allowable values that can be set arbitrarily between the above, the membership functions for each alloy element determined as described above using the above equations (4) and (5) and the above (1),
The alloy material input amount is calculated by linear programming using the equation (2), and the alloy material cost, the difference between the expected arrival amount of each alloy element after the alloy material input and the target content, and the membership function The input amount of the alloy material that can minimize the target function (6) indicating the sum of the values is determined (FIG. 2).

Here, as the unit price of the alloy material, the alloy price which changes according to the trend of the alloy market may be sequentially input, and the composition of the alloy material may be changed according to the actual condition. Further, if the target function (3) or (6) is calculated in consideration of the expression (7) for preventing the increase of the impurity element, the generation of the non-standard molten steel due to the impurity factor can be avoided in advance.

It is also extremely effective to minimize the target function equation (3) or (6) by inserting the equations (8a) and (8b) for the upper limit value and the lower limit value of the amount of alloy material input. In this case, the upper limit value of the alloy material input amount may be determined by predicting the capacity of the alloy material hopper and the inclusion prevention amount, and the lower limit value of the alloy material input amount is the minimum weighable amount of the alloy material hopper. It may be determined from the cut amount. The yield of each alloy component should be set for each steel type.

As the current value of the molten steel component, it is preferable to adopt an actual analysis value, but in addition to this, an estimated value estimated by a combination of the operating conditions of the refining furnace, the temperature at the end of refining, free oxygen, etc. is used as a substitute. It is also possible to do so. The coefficient used in the target function formula (3) or (6) may be a different value for each item, or may be changed for each steel type to be melted and each charge. By appropriately changing these coefficients, it is possible to change the weighting of each item (cost, deviation from the target component, deviation from the upper limit constraint / lower limit constraint) that is going to be balanced, according to the substance of the operation. It is desirable to determine these coefficients.

Further, in the present invention, it is necessary to determine the amount of the alloy material that can be weighed in a short time at the time of blowing stop, and this amount is required when several kinds of alloy materials are calculated by one weighing hopper. May be a total value of several kinds of alloy materials determined from the cutting speed of the weighing hopper and the time required for cutting, or may be set for each individual alloy material, or of all the alloy materials to be charged. The amount may be determined.

The following equation (9) shows the case of setting for each alloy material. In this case, of all the alloy materials to be charged, the amount of alloy material that can be weighed in a short time at the time of blowing is subtracted to obtain the amount of alloy material to be weighed in advance before blowing. Then, when calculating the amount of alloy material input from the molten steel component value (measured value or estimated value) at the final component preparation stage after blowing, the change in the molten steel component that is increased by the amount of alloy material input that is weighed in advance before blowing The following formula (10) in which the minute is added to the formula (1) is adopted, and the amount of the injected alloy material at the time of blowing stop calculated by the linear programming method is calculated. W _m1 = 0 (when W _m ≦ W _mU ) = W _m −W _mU (when W _m > W _mU ) (9) ΔT ⁱ _E −ΔT ⁱ _S −Σ _{m = 1} (W _m1 · C ⁱ _m · Y _i ) / W _Fe = Σ _{m = 1} (W _m2 · C ⁱ _m · Y _i ) / W _Fe + ΔD _i …… (10) W _m1 : Input of alloy brand m to be weighed in advance before stopping blowing Amount W _mU : Amount of alloy brand m to be added at the time of blowout The meaning of other symbols is the same as above.

When the alloy material is packed in a bag, the alloy material to be packed in the bag is determined by the branching and limiting method. The branch and bound method takes a depth-first search. The relation between the branch and bound method and the linear programming method is performed according to the procedure shown in FIGS.

In the above, the component adjustment formula (1) is a formula showing the relationship between the adjustment amount of the molten steel component and the amount of each alloy material input,
Further, the non-negative expression (2) of the alloy material input amount has a physical meaning that the alloy material input amount is 0 or more. Then, the target function formula (3) is a function aiming at minimizing the sum of the alloy material cost, the expected arrival amount after the alloy material is charged for each alloy element, and the deviation of the target component. By using the linear programming method so as to minimize, the alloy material cost and the deviation from the target component after the alloy material can be set in a well-balanced manner.

On the other hand, when the components are adjusted so as to minimize the target function formula (3), the components can be adjusted so that the deviation from the target component and the alloy material cost are most balanced. However, when the upper limit constraint value and the lower limit constraint value are determined for each alloy element in advance, such constraint conditions are not taken into consideration, so that the brand of alloy material and the input amount are outside the range of the constraint. May decide. In such a case, even if it is not possible to adjust within the above-mentioned constraint range, the brand name of the alloy material should be set so that all elements should be as close to the target values as possible
It is desirable to determine the input amount. Further, even when all the elements can be adjusted within the range of the constraint, if some elements cannot be adjusted to the target value, only a certain element may be adjusted near the upper limit value or the lower limit value. . If it is adjusted near the upper limit value or the lower limit value, it may be out of the range due to operational variations, etc. It is desirable that the value be adjusted to a value close to.

In the above-described case, the upper limit constraint value and the lower limit constraint value of each alloy element which are determined in advance, the upper limit allowable value arbitrarily set between the target value and the upper limit constraint value, and the target value Introducing the membership function for each alloying element determined from the above equations (4) and (5) as described above using the lower limit allowable value arbitrarily set between the lower limit constraint values,
By determining the amount of alloy material input so that the target function equation represented by the equation (6) is minimized, each alloy element can be brought closer to the target component value. (4),
As shown in FIG. 5, the membership function defined by the equation (5) is 0 when the component value of each alloy element is between the lower limit allowable value and the upper limit allowable value, and is greater than or equal to the upper limit allowable value or the lower limit allowable value. When it is less than the value, it is a linear function that increases from 0 in the positive direction. Therefore, the target function formula (6) in which the membership function is introduced is the weight of the constraint condition of each alloy element in addition to the alloy material cost, the expected arrival amount after the alloy material is input for each alloy element, and the deviation of the target component. Is a function containing
By solving with linear programming to minimize this equation,
The cost of the alloy material to be charged, the deviation from the target component after the alloy material is charged, and the deviation from the constraint condition of each alloy element can be set in a well-balanced manner.

As explained above, according to the present invention, the expected amount of alloying elements in molten steel and the target calculated in the target function equation (3) or (6) are calculated from the alloying material components and the amount thereof. By including the term of increase amount deviation, even if the molten steel composition cannot match the target composition due to the alloy material brand on hand, it is possible to adjust the composition with the best balance between the deviation from the target composition and the alloy material cost. It is possible to calculate the brand and input amount of such alloy materials. Furthermore,
Since the objective function formula (6) includes a term in which the weight of the constraint condition of each adjustment component is added, even when the constraint condition is imposed in advance, the cost of the alloy material to be input and the alloy It becomes possible to calculate the brand and the amount of alloy material that allow the three components, the deviation from the target component after the material has been introduced and the deviation from the constraint condition of each adjustment component, to make the most balanced component adjustment.

Furthermore, by setting an appropriate upper limit value for the membership function, and when the membership function value exceeds the upper limit value, the target function expressed by the equation (3) is minimized. The amount of alloy material input is determined, and when the membership function value is equal to or less than the upper limit value, the amount of alloy material input is determined so that the target function represented by the equation (6) is minimized. Similarly, it is determined in advance by the membership function whether each adjustment component can be adjusted within a certain constraint condition,
It is also convenient in actual operation to properly use the equation (3) or the equation (6) (FIG. 6).

If the expression (7) for preventing impurity element up is incorporated into the above component adjusting method, the impurity element contained in the alloy material to be introduced (any element may be used, such as P, S, B, etc.). It is possible to determine the optimum amount of alloy material to be charged while suppressing the amount of harmful elements in steel that are derived from the input alloy material and mixed into molten steel to be below a certain amount. This formula (7) may employ two or more types for each impurity element.

Further, as described above, the above equations (8a) and (8b) indicating the upper and lower limits of the amount of alloy material input determined by the capacity of the alloy material hopper, the minimum amount of cutout that can be weighed, the amount of inclusions prevention, etc. ) To perform the alloy calculation, it is possible to take into account the restrictions on the equipment of the alloy material charging device in the alloy calculation, and when there are alloy material regulation factors other than alloy component adjustment (for example, Even if the quality of molten steel deteriorates when alloy is added), it is possible to calculate the optimum amount of alloy material input.

Further, in order to minimize the alloy material cost, it is necessary to calculate the alloy from the molten steel components at the time of blowing stop, but the amount of alloy material that can be weighed in a short time at the time of blowing stop is Since the amount is limited, it is necessary to weigh the amount of alloy material charged before blowing is stopped for the amount exceeding the amount of alloy iron that can be weighed within the limited time after blowing. However, before the blowout, the molten steel composition at the time of blowout, such as the C concentration, is not known, so it is not possible to accurately calculate the alloy material input amount, and as a result, in consideration of safety, the C content, etc. In some cases, it is necessary to use less expensive alloy materials. Therefore, in order to more reliably suppress the alloy material cost, it is necessary to reduce the amount of the alloy material to be weighed before the blowing is stopped. On the other hand, the amount of alloy material that can be weighed at the time of blowing is limited due to equipment and operating conditions. Therefore, from the amount of alloy material that can be weighed in a short time at the time of blowing, the amount of the alloy material brand m, which is weighed in advance before blowing, and the amount of the alloy material brand m, which is thrown at the time of blowing, are described above. By using the equations (9) and (10), the alloy material cost can be minimized even when the method of separately charging the alloy material before and after blowing is adopted. As a result, it is possible to realize an alloy component adjustment operation that does not adversely affect the subsequent line while minimizing the cost of the alloy material even when the amount of the alloy material input is large.

[0037]

EXAMPLES Next, the present invention will be described in more detail with reference to examples. However, the present invention is not limited to the following examples, and is not limited to the following examples. It is, of course, possible to modify and implement the present invention, and all of them are included in the technical scope of the present invention.

(Experimental Example 1) As a refining furnace for smelting, a furnace having a molten steel amount of 100 tons / charge was used, and the components of the molten steel were prepared using the alloy materials of the brand, composition and unit price shown in Table 2. . The elements to be adjusted are C, Mn, Si, Cr and B as impurities, and the alloy component amounts are adjusted to the target component values 1 to 4 shown in the table based on the analytical component values of the blow-stop molten steel shown in Table 1. In this case, the method of the present invention is adopted and the formulas (1) to (3), or further formulas (7) and (8a) (8) are used.
The composition adjustment experiment was conducted for an example in which the selection of alloy material brand and its addition amount were calculated in consideration of b) and an example in which the conventional method was adopted, and the results shown in Tables 3 and 4 were obtained. The upper limit of the alloy material weighing value at the time of blowing was set to 500 kg for each brand of alloy material. Further, the yield Y _i of each element in the formulas (1) and (3) is 100%, and the coefficient A _m is 0.
01 and coefficient B _i were calculated as 100.

[0039]

[Table 1]

[0040]

[Table 2]

[0041]

[Table 3]

[0042]

[Table 4]

The target component value of 1 is an example of the case where the components are adjusted without taking the factor of impurities into consideration. In the conventional method in which the operator manually calculates the additive alloy material brand and the input amount, the calculation is complicated. Therefore, there is a tendency to refrain from using the alloy F containing both Si and Mn. Since alloy material brands containing Si and Mn alone are selected, the alloy material cost tends to increase. However, if the method of alloy material calculation by linear programming is adopted, the alloy material brand selection and the alloy calculation including the input amount can be performed in an extremely short time. The input amount can be instantly determined, and the alloy material cost can be reduced to about 93% as compared with the conventional method calculated by the operator.

Even when the linear programming method is adopted, in the conventional method, a large amount (1426 kg) of alloy F must be weighed at the time of blowing, and it takes a considerable time to weigh and put it in, so that the converter fire There is concern about adverse effects on things,
By adopting the method of the present invention and adding the alloy F separately before and after spraying, and suppressing the amount of alloy material calculated and input at the time of spraying to 500 kg or less The alloy material cost can be effectively reduced without extending the time required.

The target component value 2 is also an example of the case where the component adjustment is performed without adding the impurity factor. In the alloy calculation method adopting the conventional linear programming method, the element C of the above equation (1 ') is used. Since the right side is negative, a practical solution cannot be obtained and alloy calculation cannot be performed. However, according to the method of the present invention, even if the value of (T ⁱ _E −T ⁱ _S ) in the equation (1) becomes negative, it is absorbed by ΔD _i , so that an execution solution can be obtained, and the above ( By selecting the alloy material brand that minimizes the formula 3) and determining the input amount, it becomes possible to perform alloy calculation such that the deviation between the target component value and the molten steel component after addition of the alloy material is minimized. Moreover, by adopting a method in which a part of the alloy material is added before the blowing is stopped, the addition amount of the alloy material added at the time of the blowing is 500 kg, which is the weighing limit.
It can be suppressed to the following and adverse effects on the converter refractory can be avoided.

That is, in the conventional linear programming method, the solution cannot be obtained depending on the target component value and the analysis component value, and it becomes impossible to select the alloy brand for minimizing the cost and calculate the input amount. However, according to the method of the present invention, in any case, the selection of the optimum alloy brand and the input amount thereof can be reliably obtained, and the alloy material cost can always be minimized.

Next, in the method of the present invention when the target component value is 3,
The alloy calculation by the linear programming method is performed by adding the above equation (7) showing the impurity factor with B as an impurity element, whereas the conventional factor does not consider the impurity factor at all. Therefore, in this conventional method, as in the case of the target component value 1, an inexpensive alloy mainly composed of alloy F is calculated,
Although the alloy material cost is low, the B content is 4.29 ppm, which exceeds the upper limit of the target value, and it is a nonstandard product molten steel. On the other hand, in the method of the present invention, the selection criteria of the alloy material brand are changed and the alloy material cost is slightly higher, but the B content is within the upper limit of the target value, and there is no problem with molten steel quality. It can be seen that the alloy calculation is being performed.

Further, when the target component value is 4, the charging amount when the alloy A is selected by the conventional method is calculated to be 559 kg, but when the alloy material of the brand is 1000 kg in a bag, the charging amount is calculated. I can't. However, according to the method of the present invention, by combining the branch and bound methods,
For bagged alloys, it is possible to perform the operation as calculated because it can be calculated before blowing off for each bagged unit.

7 and 8 show a large number of actual operation examples in which the components were adjusted at a practical level using an actual refining furnace.
The alloy cost index and the minimum cost calculation are performed by collecting the data when the founder manually calculates the alloy and when the conventional linear programming method and the method of the present invention are used to calculate the alloy. Execution rate (rate at which a solution was obtained)
Is a graph in which the values are examined and arranged as average values. As is clear from FIG. 7, the alloy cost index is 4% in the conventional linear programming method when the manually calculated value is 100, but is doubled in the present invention to 8%. You can see that Also, from FIG.
In contrast to the performance rate of 100 according to the present invention, it is often the case that the conventional linear programming method cannot find an execution solution, the performance rate is 80%, and the performance cannot be achieved at a rate of 1/4. In that case, the execution rate is only 60%.

As described above, according to the present invention, the component adjustment formula (1), the non-negative formula (2) of the amount of alloy input, the target function indicating the sum of the cost of the alloy material and the deviation from the target component after the alloy material is supplied. By using the formula (3) and performing the alloy calculation by the linear programming method, the alloy material cost can be further reduced as compared with the conventional alloy programming method using the linear programming method. By incorporating the equation (7), it is possible to avoid defective molten steel quality due to impurity factors. In addition, (9),
If the formula (10) is combined and carried out, the amount of alloy brand m to be weighed in advance before blowing is determined from the amount of alloy material that can be weighed within the limited time at the time of blowing and the alloy brand m to be thrown in at the time of blowing If the method for calculating the input amount of is combined and executed, it is possible to further reduce the cost suitable for the actual operation.

(Experimental Example 2) As in Experimental Example 1, because of melting
Uses a molten steel amount of 100 tons / charge as a refining furnace
However, using the alloy material of the brand, composition and unit price shown in Table 2,
The composition of molten steel was adjusted. The elements to be adjusted are C, Mn,
Si, Cr, P, B and the following equations (11) and (12)
The strength coefficient CE and the weldability coefficient CEQ shown in Table 5 are shown in Table 5.
Based on the current composition values of the blown molten steel shown in the table,
When adjusting to the standard component value, the method of the present invention is adopted,
Formulas (1) to (3), or Formulas (1) and (2),
Selection of alloy material brand and addition amount using (4) to (6)
The component adjustment experiment for calculating
I got a fruit. The upper limit constraint value and the lower limit constraint value are shown in Table 5.
The upper limit and lower limit are given in (13) (14) below.
Shown in the formula. Yield Y of each element in the formula_i Is 1
00%, coefficient A_m As for the
In the case of C, Mn, Si, Cr, P, B, set to 10.0,
CE and CEQ were set to 5.0. Also (3)
B in the formula_i Is 0.01, B in equation (6) _i Is 0.002
And α_i Was set to 1000.0 for all elements. CE = C + Si / 5 + Mn / 6 + P + 1.2V + Nb (11) CEQ = C + Mn / 6 + (Cu + Ni) / 15 + (Cr + Mo + V) / 5 (12) U_ki= 0.8 (U_i -Tⁱ _E) + Tⁱ _E ... (13) L_ki= Tⁱ _E-0.8 (Tⁱ _E-L_i ) ・ ・ ・ (14)

9 and 10 are obtained by normalizing each adjustment component by the upper and lower limit constraint values, the target values, and the upper and lower limit allowable values, and plot the calculation results of each adjustment component. FIG. 9 shows a case where V = 0 of the current components in Table 5 is calculated, and it is a case where the target value can be adjusted, and the same result is obtained regardless of whether the formula (3) or the formula (6) is used. To be FIG. 10 shows a case where the upper and lower limit constraints can be adjusted, but not all of them can be adjusted to the target values, and when Equation (3) is used, CE and CEQ are adjusted to be near the lower limit constraint value and the upper limit constraint value. However, when equation (6) is used, CE, CE
Both Q are adjusted inside the upper limit constraint value and the lower limit constraint value. It can be seen that it is more effective to use the equation (6) when it is desired to prevent the adjustment component values from being adjusted to the vicinity of the upper and lower limit constraint values, even if the adjustment accuracy and cost of the components are sacrificed to some extent.

[0053]

[Table 5]

[0054]

The present invention is configured as described above, and even when the content of a certain component in the molten steel exceeds the target value at the end of melting in the refining furnace, the optimum alloy material is obtained. It is possible to accurately determine the brand selection and the input amount, and it is possible to achieve complete automation of alloy calculation and reduction of alloy material cost with certainty.

In addition to the above-mentioned effects, it is possible to properly select the alloy material brand and to determine the amount to be charged, without generating defective molten steel due to impurities contained in the alloy material used for component adjustment. .

Further, when there is an upper limit and a lower limit of the amount of the alloy material input depending on the situation at the site, or the amount of the alloy material input for adjusting the components is large, the alloy material is added within a short time before tapping. Even if it is assumed that the weighing will not be in time, there will be no problems including delay of tapping work and disturbance of the operating line after that, and reliable composition adjustment is possible.

Further, even in the case where the number of hoppers that are constantly provided in the refining equipment is limited and the alloy material must be charged in a predetermined weight unit, the charging amount is also taken into consideration. The most suitable alloy material brand can be selected and the input amount can be determined.

[Brief description of drawings]

FIG. 1 is a flowchart showing a calculation process of a component adjusting method for molten steel according to claim 3.

FIG. 2 is a flowchart showing a calculation process of a component adjusting method for molten steel according to claim 4;

FIG. 3 is a diagram showing an example of the selection process of an alloy material brand and the calculation process of the input amount when performing the linear programming according to the present invention.

FIG. 4 is an explanatory diagram for performing a branch and bound method.

FIG. 5 is a graph illustrating a membership function.

FIG. 6 is a flowchart showing a calculation process of a component adjusting method for molten steel according to claim 5;

FIG. 7 is a graph showing an alloy material cost index when the present invention is implemented and when a conventional manual calculation method or a conventional linear programming method is adopted.

FIG. 8 is a graph showing a comparison of the performance rate of the minimum alloy material cost calculation when the present invention is implemented and when the conventional manual calculation method or the conventional linear programming method is adopted.

FIG. 9 is a graph showing the V value of the current component value is 0 in Experimental Example 2.
6 is a graph showing an example of the calculation result of the present invention in the case of.

10 is a graph showing an example of the calculation result of the present invention when the current component value is used as it is in Experimental Example 2. FIG.

Claims (9)

[Claims] 1. When the alloy material is added to the molten steel to adjust the composition of the molten steel, the current value and target content of each alloying element in the molten steel, the unit price of each brand of the alloying material and the content of each alloying element In the method of adjusting the composition of molten steel by setting the amount of alloy material input by linear programming based on the amount and yield in molten steel, and the amount of molten steel, alloy material cost and each alloy element after the alloy material is supplied A method for adjusting the composition of molten steel, characterized in that the amount of alloy material input is determined in consideration of the balance between the expected arrival amount and the deviation of the target content. 2. The molten steel according to claim 1, wherein the amount of alloy material to be charged is determined in consideration of both or one of an upper limit constraint value and a lower limit constraint value defined in advance for each alloy element. Composition adjustment method. 3. When the alloy material is added to the molten steel to adjust the composition of the molten steel, the present value and the target content of each alloy element in the molten steel, the unit price of each brand of the alloy material and the content of each alloy element Based on the amount and yield in molten steel, and the amount of molten steel, in the method of adjusting the amount of alloy material input by linear programming to adjust the composition of molten steel, a component adjustment formula represented by the following formula (1), Using the non-negative formula of the alloy material input amount shown in the following equation (2), the linear programming method is used to calculate the sum of the difference between the alloy material cost and the expected arrival amount of each alloy element after the alloy material input and the target content. The method for adjusting the composition of molten steel is characterized in that the amount of alloy material charged is determined so that the target function represented by the following equation (3) is (T ⁱ _E −T ⁱ _S ) + ΔD _i = [Σ _{m = 1} (W _m · C ⁱ _m · Y _i ) / W _Fe ] (1) W _m ≧ 0 (2) Σ _{m = 1} (A _m · W _m · P _m ) + Σ _{i = 1} (B _i · | ΔD _i |) (3) (where T ⁱ _E is the target content of the alloy element i in the molten steel T ⁱ _S is the current value of the alloy element i in the molten steel W _m is the input amount of the alloy material brand m C ⁱ _m is the content of the alloy element i in the alloy material brand m Y _i is the yield of the alloy element i in the molten steel W _Fe is the amount of molten steel ΔD _i is the amount of alloying material, the content of alloying element i in the alloying material and the yield of alloying element i in the molten steel calculated from the yield in the molten steel, and the alloy Differences from the target content of element i A _m and B _i are coefficients P _m is the unit price of alloy material brand m) 4. When the alloy material is added to the molten steel to adjust the composition of the molten steel, the analytical value and the target content of each alloy element in the molten steel, the unit price of each brand of the alloy material and the content of each alloy element Amount and yield in molten steel, and based on the molten steel amount, in the method of adjusting the amount of alloy material input by linear programming to adjust the composition of molten steel, each adjustment component determined in advance for each alloy element Upper limit constraint value and lower limit constraint value, upper limit allowable value that can be arbitrarily set between the target value and the upper limit constraint value, and lower limit allowable value that can be arbitrarily set between the target value and the lower limit constraint value To (4) and (5) below
Using the membership function for each alloying element and the above equations (1) and (2) determined by using the equations, the alloy material cost and each alloying element after the alloying material is introduced by the linear programming method. It is characterized in that the amount of alloy material input is determined so that the target function expressed by the following equation (6) indicating the sum of the expected arrival amount and the target content and the membership function value is minimized. A method for adjusting the composition of molten steel. f1 = {[Σ _{m = 1} (W _m · C ⁱ _m · Y _i ) / W _Fe ] + T ⁱ _s −U _ki } / (U _i −U _ki ) ... (4) f2 = 1-{[ Σ _{m = 1} (W _m · C ⁱ _m · Y _i ) / W _Fe ] + T ⁱ _s −L _i } / (L _ki −L _i ) ... (5) λ = 0 (f1 ≦ 0 and f2 ≦ 0) λ = f1 (when f1> 0 and f2 ≦ 0) λ = f2 (when f1 ≦ 0 and f2> 0) Σ _{m = 1} (A _m · W _m · P _m ) + Σ _{i = 1} (B _i · | ΔD _i |) + Σ _{i = 1} α _i · λ _i (6) (wherein U _ki is the upper limit allowable value of alloy element i in molten steel U _i is the alloy element i in molten steel) Upper limit constraint value L _ki is a lower limit allowable value of alloy element i in molten steel L _i is a lower limit constraint value of alloy element i in molten steel α _i is a coefficient λ _i is a membership function value of alloy element i in molten steel ) 5. When the alloy material is added to the molten steel to adjust the composition of the molten steel, the analytical value and the target content of each alloy element in the molten steel, the unit price of each brand of the alloy material and the content of each alloy element Amount and yield in molten steel, and based on the molten steel amount, in the method of adjusting the amount of alloy material input by linear programming to adjust the composition of molten steel, each adjustment component determined in advance for each alloy element From the upper limit constraint value and lower limit constraint value, the upper limit allowable value that can be arbitrarily set between the target value and the upper limit constraint value, and the lower limit allowable value that can be arbitrarily set between the target value and the lower limit constraint value. When an appropriate upper limit is set for the membership function for each alloy element determined as described above using the equations (4) and (5) and the membership function value exceeds the upper limit, The component adjustment formula shown by the formula (1), Using the non-negative expression of the alloy material input amount given by the equation (2), the sum of the difference between the alloy material cost and the expected arrival amount of each alloy element after the alloy material input and the target content is calculated by the linear programming method. The input amount of the alloy material is determined so that the target function represented by the formula (3) shown below becomes the minimum, and when the membership function value is equal to or less than the upper limit value, the membership function and the ( The sum of the alloy material cost, the difference between the expected arrival amount and the target content of each alloy element after the alloy material is charged, and the sum of the membership function values is calculated by linear programming using the equations 1) and 2). Formula (6)
A method for adjusting the composition of molten steel, characterized in that the amount of alloy material input is determined so that the target function shown by is minimized. 6. The method for adjusting the composition of molten steel according to claim 3, wherein the amount of alloy material input is determined by a linear programming method by adding the following equation (7) indicating the factor of impurity element. T ^b _E −T ^b _S ≧ [Σ _{b = 1} (W _m · C ^b _m · Y _b ) / W _Fe ] ... (7) (In the formula, T ^b _E is an impurity element b contained in molten steel. Content of T ^b _S is the current value of impurity element b in molten steel C ^b _m is the content of impurity element b in alloy material brand m Y _b is the yield of impurity element b in molten steel) 7. The method according to any one of claims 3 to 6, further comprising adding the following equations (8a) and (8b) showing the upper limit value and the lower limit value of the alloy material input amount to determine the alloy material input amount. Method for adjusting composition of molten steel. W _m ≧ L _m (8a), W _m ≦ U _m (8b) (where L _m is the lower limit of the amount of alloy material brand m input, and U _m is the amount of alloy material brand m input. Represents the upper limit of 8. The method for adjusting the composition of molten steel according to claim 3, wherein the alloy material is added in two steps, before and after blowing, at the final stage of refining in the refining furnace. 9. The alloy material input amount is determined by combining the linear programming method with a branching and limiting method when at least one of the alloy material brands has a predetermined input unit. A method for adjusting the composition of molten steel according to Crab.