BLAST FURNACE OPERATION METHOD

Abstract

A blast furnace operation method according to one aspect of the present invention includes: a process of acquiring a correlation between a carbon consumption in reducing gas and a reduction InputΔC in specific carbon consumption caused by blowing the reducing gas into the blast furnace per molar ratio C/H of carbon atoms to hydrogen atoms in the reducing gas; a process of determining a carbon consumption in the reducing gas where the reduction InputΔC in specific carbon consumption is a predetermined target value or higher on the basis of the correlation acquired per C/H; and a process of adjusting the amount of the reducing gas blown into the blast furnace on the basis of the determined carbon consumption in the reducing gas and the carbon proportion in the reducing gas.

Claims

1. A blast furnace operation method in which reducing gas including hydrogen atoms and carbon atoms is blown into a blast furnace, the method comprising: acquiring a correlation between a carbon consumption in the reducing gas and a reduction InputΔC in specific carbon consumption caused by blowing the reducing gas into the blast furnace per molar ratio C/H of carbon atoms to hydrogen atoms in the reducing gas; determining a carbon consumption in the reducing gas where the reduction InputΔC in specific carbon consumption is a predetermined target value or higher on the basis of the correlation acquired per C/H; and adjusting an amount of the reducing gas blown into the blast furnace on the basis of the determined carbon consumption in the reducing gas and a carbon proportion in the reducing gas.

2. The blast furnace operation method according to claim 1, wherein the molar ratio C/H of the carbon atoms to the hydrogen atoms in the reducing gas is 0.15 or higher.

3. The blast furnace operation method according to claim 2, wherein the correlation is represented by a quadratic expression of the carbon consumption in the reducing gas.

4. The blast furnace operation method according to claim 3, wherein the correlation is represented by Y=a1X.sup.2+b1X+c1 (where X represents the carbon consumption in the reducing gas, Y represents the reduction InputΔC in specific carbon consumption, and all the coefficients a1, b1, and c1 represent values that do not depend on the molar ratio C/H).

5. The blast furnace operation method according to claim 4, wherein the carbon consumption in the reducing gas is determined in a range of 21 kg/t-pig to 107 kg/t-pig.

6. The blast furnace operation method according to claim 4, wherein the carbon consumption in the reducing gas is determined in a range of 21 kg/t-pig to 65 kg/t-pig.

7. The blast furnace operation method according to claim 1, wherein the molar ratio C/H of the carbon atoms to the hydrogen atoms in the reducing gas is higher than 0 and lower than 0.15.

8. The blast furnace operation method according to claim 7, wherein the molar ratio C/H of the carbon atoms to the hydrogen atoms in the reducing gas is 0.13 or lower.

9. The blast furnace operation method according to claim 7, wherein the molar ratio C/H of the carbon atoms to the hydrogen atoms in the reducing gas is 0.10 or lower.

10. The blast furnace operation method according to claim 7, wherein the correlation is represented by Y=a2X.sup.2+b2X+c2 (where X represents the carbon consumption in the reducing gas, Y represents the reduction InputΔC in specific carbon consumption, and at least one of the coefficients a2, b2, and c2 represents a function including the molar ratio C/H as a variable).

11. The blast furnace operation method according to claim 1, wherein when the reducing gas is blown into the blast furnace, a flame temperature is adjusted to be 2000° C. or higher.

12. The blast furnace operation method according to claim 11, wherein in order to adjust the flame temperature to be 2000° C. or higher, at least one of an blast volume or an oxygen enrichment in hot blast is adjusted.

13. The blast furnace operation method according to claim 1, wherein the reducing gas is selected from coke oven gas, natural gas, reformed top gas (BFG), city gas, mixed gas thereof, and hydrogen mixed gas obtained by mixing hydrogen gas therewith.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0036] FIG. 1 is a graph showing a correlation between InputΔC and a carbon consumption (kg/t-pig) in reducing gas per C/H in the reducing gas.

[0037] FIG. 2 is a graph showing a relationship between a hydrogen reduction rate and the amount of hydrogen fed (Nm.sup.3/t-pig) into a blast furnace per ton of molten iron.

[0038] FIG. 3 is a graph showing a relationship between a direct reduction rate and the amount of hydrogen fed (Nm.sup.3/t-pig) into a blast furnace per ton of molten iron.

EMBODIMENTS OF THE INVENTION

[0039] Hereinafter, a preferable embodiment of the present invention will be described in detail. In the following embodiment, a numerical limitation range represented using “to” refers to a range including numerical values before and after “to” as a lower limit and an upper limit. A numerical value shown together with “more than” or “less than” is not included in a numerical range.

[0040] <1. Correlation between InputΔC and Carbon Consumption in Reducing Gas>

[0041] First, a correlation between InputΔC and a carbon consumption in reducing gas blown from a tuyere (hereinafter, the correlation will also be referred to as “ΔC-reducing gas correlation”) will be described based on FIG. 1. In FIG. 1 the vertical axis (y axis) represents InputΔC (%) and the horizontal axis (x axis) represents the carbon consumption (kg/t-pig) in the reducing gas.

[0042] Here, InputΔC can be defined as the proportion of reduction in specific carbon consumption caused by blowing the reducing gas into the blast furnace. When Input C of a base operation is represented by A (kg/t-pig) and Input C of an operation where a predetermined amount of reducing gas is blown is represented by B (kg/t-pig), InputΔC is represented by the following Numerical Expression (2). Note that InputΔC of the base operation in FIG. 1 is 0.0. Of course, InputΔC is not limited to being represented by the following Numerical Expression (2). For example, the difference (A−B) between A and B may be acquired as InputΔC.

InputΔC=(A−B)/A×100(%)  (2)

[0043] The carbon consumption in the reducing gas refers to the consumption of carbon fed into the blast furnace by the reducing gas blown from the tuyere as described above, and can be obtained by multiplying the carbon proportion (kg/Nm.sup.3) in the reducing gas by the reducing gas volume (Nm.sup.3/t-pig) per ton of molten iron.

[0044] The reducing gas is blown into the blast furnace from a tuyere provided in the blast furnace. The reducing gas includes reducing components that reduce iron-bearing materials in the blast furnace. Here, the reducing components according to the embodiment include not only a component (for example CO gas or hydrogen gas) that can reduce the iron-bearing materials by itself but also a component (for example, CO.sub.2 gas or hydrocarbon gas) that can produce reducing gas through a reaction in the blast furnace (for example, a reaction with coke, pulverized coal, or the like or decomposition).

[0045] The ΔC-reducing gas correlation shown in FIG. 1 is acquired by performing, for example, a blast furnace operation simulation. As the blast furnace operation simulation, for example, a so-called “Blast Furnace Mathematical Model” Kouji TAKATANI, Takanobu INADA, Yutaka UJISAWA, “Three-dimensional Dynamic Simulator for Blast Furnace”, ISU International, Vol. 39 (1999), No. 1, p. 15 to 22 can be used. In this blast furnace mathematical model, an internal region of the blast furnace is divided in a height direction, a radial direction, and a circumferential direction to define a plurality of meshes (small regions), and the behavior of each of the meshes is simulated. The summary of the blast furnace operation simulation is as follows. That is, the blast furnace operation simulation is performed using various cases where C/H in the reducing gas and the amount of the reducing gas blown (the amount thereof blown per ton of molten iron) are different from each other. The cases also include the base operation (a case where the reducing gas volume is 0). Here, operation conditions are adjusted such that the flame temperature and the molten iron temperature are as constant as possible in the cases. For example, at least one of an blast volume or an oxygen enrichment ratio in hot blast may be adjusted. Here, the flame temperature refers an in-furnace temperature in a tip end portion of the tuyere on the inside of the furnace, and will also be referred to as “tuyere tip temperature”. In the actual operation, the flame temperature is calculated as a tuyere tip theoretical combustion temperature according to Lamm equation described in “Ironmaking Handbook” (Chijinshokan Co., Ltd.), Akitoshi SHIGEMI. The hot blast blown into the blast furnace is gas including air. The hot blast may include hygroscopic moisture and enriched oxygen in addition to air. Schematically, the oxygen enrichment ratio refers to the volume proportion of oxygen in the hot blast with respect to the total volume of the hot blast, and is represented by “Oxygen enrichment ratio (%)={(Blast volume [Nm.sup.3/min]×0.21+Amount of Enriched Oxygen [Nm.sup.3/min]/(Blast volume [Nm.sup.3/min]+Amount of Enriched Oxygen [Nm.sup.3/min])}×100−21. In addition to or instead of the adjustment of the above-described factors, at least one of a coke ratio or a pulverized coal ratio may be adjusted. As a result, InputΔC and the carbon consumption in the reducing gas are acquired per case. Incidentally, a point representing InputΔC and the carbon consumption in the reducing gas of each of the cases is plotted on the xy plane shown in FIG. 1. Points P1 to P8 are examples of the plotted points. A fitted curve of each of the plots is acquired using an approximation method such as a least-squares method or the like. These fitted curves form a graph showing the ΔC-reducing gas correlation. Graphs L1 to L5 are examples of the graph showing the ΔC-reducing gas correlation.

[0046] (1-1. Case where C/H is 0.15 or Higher)

[0047] As described above, the correlation between InputΔC and the carbon consumption in the reducing gas, that is, the ΔC-reducing gas correlation tends to vary between a case where a molar ratio C/H of carbon atoms to hydrogen atoms in the reducing gas is 0.15 or higher and a case where C/H is lower than 0.15. Therefore, first, the ΔC-reducing gas correlation of the case where C/H is 0.15 or higher will be described based on the points P1 to P4 and the graph L1.

[0048] Here, the point P1 represents InputΔC and the carbon consumption in the reducing gas in the base operation (operation where the reducing gas is not blown), the points P2 and P4 represent InputΔC and the carbon consumption in the reducing gas in an operation where coke oven gas (COG, C/H=0.186) as the reducing gas is blown, and the point P3 represents InputΔC and the carbon consumption in the reducing gas in an operation where natural gas (C/H=0.25) as the reducing gas is blown. The points P1 to P3 were acquired using the same method as that of Example 1 described below. The point P4 was acquired using the same method as that of Example 1, except that the flame temperature was set as 2085° C. or 2315° C. The graph L1 is a graph showing the fitted curve of the points of P1 to P4, that is, the ΔC-reducing gas correlation.

[0049] Examples of the reducing gas where C/H is 0.15 or higher include COG, natural gas, city gas, and the like. The reducing gas may be gas obtained by reforming top gas (BFG) (gas obtained by removing water vapor and CO.sup.2 gas from top gas). Among these, the reducing gas including hydrocarbon gas, that is, COG, natural gas, city gas or the like is preferable. When this reducing gas is used, the hydrocarbon gas is combusted in the furnace to generate heat of combustion. Therefore, a further reduction in reducing material ratio can be expected. Further, in an iron mill where a coke furnace is present, the energy can be provided from the iron mill itself by using COG. COG is more preferable to the other reducing gases from the viewpoint of costs. The upper limit of C/H is not particularly limited and, for example, may be 0.3 or lower.

[0050] The composition of COG used to obtain data of the points P2 and P4 is shown in Table 1, and the composition of natural gas used to obtain data of the point P3 is shown in Table 2. These compositions were measured by gas chromatography, a mass spectrometer, or the like. The numerical values of each of the components shown in Table 1 and 2 are the molar ratio (more specifically, the ratio between the molar concentrations (mol/L)). Note that C represents the carbon proportion (kg/Nm.sup.3) in the reducing gas. C/H of the COG having the composition shown in Table 1 below is 0.185. The calculation example is as follows.

(0.065+0.025+0.292+0.02×2+0.008×2)/(0.535×2+0.292×4+0.02×4+0.008×6)=0.185

[0051] In addition, C/H of the natural gas having the composition shown in Table 2 below is 0.271. The calculation example is as follows.

(0.85+0.03×2+0.12×2)/(0.85×4+0.03×4+0.12×6)=0.271

TABLE-US-00001 TABLE 1 Example of COG Composition CO CO.sub.2 H.sub.2 N.sub.2 CH.sub.4 C.sub.2H.sub.4 C.sub.2H.sub.6 C C/H (—) (—) (—) (—) (—) (—) (—) (kg/Nm.sup.3) (—) 0.065 0.025 0.535 0.055 0.292 0.02 0.008 0.23 0.185

TABLE-US-00002 TABLE 2 Example of Natural Gas Composition CH.sub.4 C.sub.2H.sub.4 C.sub.2H.sub.6 C C/H (—) (—) (—) (kg/Nm.sup.3) (—) 0.85 0.03 0.12 0.62 0.271

[0052] As can be seen from FIG. 1, the points P1 to P4 are present on substantially the same graph L1. Accordingly, by acquiring the carbon consumption in the reducing gas irrespective of the type of the reducing gas (in other words, irrespective of C/H in the reducing gas), InputΔC can be uniquely specified. That is, InputΔC and the carbon consumption in the reducing gas have a correlation that does not depend on C/H, this correlation is represented by the graph L1. Further, when the fact that the point P4 is present on the graph L1 is taken into consideration, it can be said that the correlation also does not depend on the flame temperature.

[0053] Since the graph L1 is represented by an upwardly convex graph, the carbon consumption in the reducing gas is represented by a quadratic expression. For example, the graph L1 is represented by the numerical expression Y=a1X.sup.2+b1X+c1. X represents the carbon consumption in the reducing gas, and Y represents InputΔC. All the coefficients a1, b1, and c1 represent values that do not depend on the molar ratio C/H. In the example of FIG. 1, the graph L1 is represented by the numerical expression Y=−0.0014X.sup.2+0.194X (that is, a1=−0.0014, b1=0.194, c1=0). Of course, the graph L1 is not limited to being represented by this numerical expression.

[0054] According to the graph L1, when the carbon consumption in the reducing gas is in a range of 65 kg/t-pig or lower, InputΔC has a positive correlation with the carbon consumption in the reducing gas, and when the carbon consumption in the reducing gas is in a range of higher than 65 kg/t-pig, InputΔC has a negative correlation with the carbon consumption in the reducing gas. In addition, when the carbon consumption in the reducing gas is about 65 kg/t-pig, InputΔC shows a maximum value. Accordingly, when the carbon consumption in the reducing gas is determined such that InputΔC is close to the maximum value, the reducing material ratio can be further reduced.

[0055] More specifically, when the carbon consumption in the reducing gas is in a range of 21 kg/t-pig to 107 kg/t-pig, InputΔC is approximately 4.0% or higher. In this case, for example, assuming that the reducing material ratio of the base operation is 375 kg/t-pig to 500 kg/t-pig, the reducing material ratio is reduced by about 15 kg/t-pig to 20 kg/t-pig or more. This reduction is a significant value in consideration of the daily fluctuation in reducing material ratio, and the effectiveness of the reduction in reducing material ratio can be expected. Accordingly, the carbon consumption in the reducing gas is preferably 21 kg/t-pig to 107 kg/t-pig.

[0056] Here, when the carbon consumption in the reducing gas is about 65 kg/t-pig, InputΔC shows a maximum value, and when the carbon consumption in the reducing gas exceeds 65 kg/t-pig, InputΔC starts to decrease. That is, the effect of reducing Input C is lost. For example, the reason for this is presumed to be that the carbon consumption in the reducing gas is excessively high compared to the amount required for in-furnace reduction such that the gas utilization factor decreases or to be that, although the amount of the reducing gas blown increases together with an increase in the carbon consumption in the reducing gas, under the condition that the flame temperature is constant, the oxygen enrichment ratio increases together with an increase in the amount of the reducing gas blown and the amount of the gas blown into the blast furnace through a hot stove decreases such that the sensible heat of air decreases, or the like. Accordingly, the carbon consumption in the reducing gas is more preferably 65 kg/t-pig or lower, that is, 21 kg/t-pig to 65 kg/t-pig. In this case, InputΔC can be made to be high (specifically 4.0% or higher) with a smaller amount of the reducing gas blown.

[0057] In addition, when the reducing gas in which the carbon proportion (kg/Nm.sup.3) is low (in particular, the reducing gas in which the carbon proportion is lower than 0.6 kg/Nm.sup.3) is used, due to the restrictions of the operation, there may be a case where the carbon consumption in the reducing gas is preferably 65 kg/t-pig or lower. Hereinafter, the reason for this will be described in detail.

[0058] In the blast furnace operation, it is necessary that the flame temperature is maintained at a constant value that is higher than or equal to a predetermined value (here, the predetermined value varies depending on various factors but is likely to be a value about 2000° C.) as far as possible. The reason for this is that, when the flame temperature is lower than the predetermined value, the combustibility of pulverized coal decreases and there is a problem, for example, in that unburnt chart is produced and deteriorates in-furnace permeability or in that only a part of pulverized coal fed as the reducing material can be used as the reducing gas (that is produced in the furnace), or the like. When the carbon proportion (kg/Nm.sup.3) in the reducing gas blown from the tuyere is low, it is necessary to blow a large amount of the reducing gas in order to adjust the carbon consumption in the reducing gas to be higher than 65 kg/t-pig. As a result, it is necessary to increase the oxygen enrichment ratio in the hot blast. The reason for this is presumed to be that, unless the oxygen enrichment ratio is increased, there may be a case where the flame temperature cannot be maintained at the predetermined value or higher. Note that, as the oxygen enrichment ratio increases, the oxygen proportion in the hot blast increases such that pure oxygen is blown. At this time, the oxygen enrichment ratio reaches the upper limit, and the oxygen enrichment ratio cannot increase any more.

[0059] For example, when the carbon consumption in the reducing gas is adjusted to 83 kg/t-pig by using the COG having the composition shown in Table 1, it is necessary to blow the COG at 350 Nm.sup.3/t-pig. In this case, by increasing the oxygen enrichment ratio to be close to the upper limit, the flame temperature can be maintained at the predetermined value or higher. However, it is necessary to design operation elements carefully such that the flame temperature is very close to the predetermined value, and it is necessary to monitor the elements carefully during the operation. Accordingly, the operation can be performed, but time and efforts are required for the operation. Further, when the carbon consumption in the reducing gas is 95 kg/t-pig, it is necessary to blow the COG at 400 Nm.sup.3/t-pig. In this case, even when pure oxygen is blown, there may be a case where the flame temperature cannot be maintained at the predetermined value or higher. When the carbon consumption in the reducing gas is 65 kg/t-pig or lower, the amount of the COG blown can be made to be lower than 350 Nm.sup.3/t-pig. Therefore, an allowance can be given to the oxygen enrichment ratio and the flame temperature. Accordingly, when the reducing gas in which the carbon proportion (kg/Nm.sup.3) is low (in particular, the reducing gas in which the carbon proportion is lower than 0.6 kg/Nm.sup.3) is used, the carbon consumption in the reducing gas is preferably 65 kg/t-pig or lower.

[0060] On the other hand, when the natural gas (the carbon proportion is 0.6 kg/Nm.sup.3 or higher) shown in Table 2 is used, basically, the above-described restrictions are not present. For example, even when the carbon consumption in the reducing gas is 100 kg/t-pig which is much higher than 65 kg/t-pig, the amount of the reducing gas blow needs to be only about 170 Nm.sup.3/t-pig. In this case, although a decrease in flame temperature is concerned, the flame temperature can be made to be the predetermined value or higher by increasing the oxygen enrichment ratio. Accordingly, the carbon consumption in the reducing gas can be made to be higher than 65 kg/t-pig.

[0061] Due to the above-described reasons, the point P2 is plotted in a range of 65 kg/t-pig or lower, but the point P3 is plotted in a wider range.

[0062] When operation conditions other than the above-described conditions change, the ΔC-reducing gas correlation may slightly fluctuation from the graph L1. Even in this case, however, it is considered that there is no significant fluctuation in the preferable range of the carbon consumption in the reducing gas.

[0063] (1-2. Case where C/H is Lower than 0.15)

[0064] Incidentally, first, the ΔC-reducing gas correlation of the case where C/H is lower than 0.15 will be described based on the points P1 and P5 to P8 and the graphs L2 to L5. Here, the point P5 represents InputΔC and the carbon consumption in the reducing gas in a case where C/H in the reducing gas is 0.054, the point P6 represents InputΔC and the carbon consumption in the reducing gas in a case where C/H in the reducing gas is 0.097, the point P7 represents InputΔC and the carbon consumption in the reducing gas in a case where C/H in the reducing gas is 0.137, and the point P8 represents InputΔC and the carbon consumption in the reducing gas in a case where C/H in the reducing gas is 0.02. The points P5 to P8 were acquired using the same method as that of Example 2 described below. The graphs L2 to L5 are graphs showing the fitted curves of the points of P5 to P8, that is, the ΔC-reducing gas correlations, respectively.

[0065] The present inventors conducted an investigation on the reducing gas (for example, COG, natural gas, city gas, and the like) in the related art, and C/H in most of the reducing gas was 0.15 or higher. Therefore, the reducing gas in which C/H is lower than 0.15 may be produced, for example, by mixing hydrogen gas with reducing gas in which C/H is 0.15 or higher. The reducing gas in which hydrogen gas is mixed may be any one as long as it is reducing gas in which C/H is 0.15 or higher, and examples thereof include COG, natural gas, top gas, city gas, and the like. In addition, the method of producing the reducing gas is not necessarily limited to this method. For example, the reducing gas in which C/H is lower than 0.15 may be produced by mixing reducing gases having different C/H's (specifically, reducing gas in which C/H is 0.15 or higher and reducing gas in which C/H is lower than 0.02) with each other.

[0066] As clearly seen from FIG. 1, the points P5 to P8 are present in the graphs L2 to L5 that are different from each other, respectively. Accordingly, when C/H in the reducing gas is lower than 0.15, the ΔC-reducing gas correlation varies depending on C/H in the reducing gas. That is, by acquiring C/H in the reducing gas and the carbon consumption in the reducing gas, InputΔC can be uniquely specified. This way, InputΔC and the carbon consumption in the reducing gas have a correlation that depends on C/H, and the correlation per C/H is represented by, for example, the graphs L2 to L5. Note that all the correlations are represented by the upwardly convex graphs (that is, when the carbon consumption in the reducing gas is a given value, a maximum value is shown). It is presumed that, even when the flame temperature fluctuates, there is substantially no effect on the correlation as in the case where C/H is 0.15 or higher.

[0067] Since the graphs L2 to L5 are represented by an upwardly convex graph, the graphs L2 to L5 are represented by a quadratic expression of carbon consumption in the reducing gas. For example, the graphs L2 to L5 are represented by the numerical expression Y=a2X.sup.2+b2X+c2. In order to simplify the drawing, FIG. 1 does not show a curve of a portion where InputΔC starts to decrease. X represents the carbon consumption in the reducing gas, and Y represents InputΔC. Since the shapes of the graphs L2 to L5 vary depending on C/H in the reducing gas, at least one of the coefficients a2, b2, and c2 represents a function including C/H in the reducing gas as a variable. Accordingly, when the carbon consumption in the reducing gas is determined such that InputΔC is close to the maximum value, the reducing material ratio can be further reduced. As described above, the reason why the graphs L2 to L5 start to decrease from the maximum value is presumed to be that the carbon consumption in the reducing gas is excessively high compared to the amount required for in-furnace reduction such that the gas utilization factor decreases or to be that, although the amount of the reducing gas blown increases together with an increase in the carbon consumption in the reducing gas, under the condition that the flame temperature is constant, the oxygen enrichment ratio increases together with an increase in the amount of the reducing gas blown and the amount of the gas blown into the blast furnace through a hot stove decreases such that the sensible heat of air decreases, and the like.

[0068] The graphs L2 to L5 will be described in more detail. In the range of the carbon consumption in the reducing gas where InputΔC does not reach the maximum value, as C/H decreases, the slopes of the graphs L2 to L5 increase. That is, an increase in InputΔC relative to a unit increase in the carbon consumption in the reducing gas increases. Accordingly, as C/H in the reducing gas decreases, InputΔC can be efficiently increased. More specifically, C/H in the reducing gas is preferably 0.13 or lower, more preferably 0.10 or lower, and still more preferably 0.05 or lower. The lower limit of C/H is not particularly limited as long as it is higher than 0.

[0069] As described above, the correlation is present between InputΔC and the carbon consumption in the reducing gas. This correlation, that is, the ΔC-reducing gas correlation tends to vary between a case where the molar ratio C/H of carbon atoms to hydrogen atoms in the reducing gas is 0.15 or higher and a case where C/H is lower than 0.15. That is, when C/H in the reducing gas is 0.15 or higher, the ΔC-reducing gas correlation is uniquely determined irrespective of the type of the reducing gas (in other words, irrespective of C/H in the reducing gas). On the other hand, when C/H in the reducing gas is lower than 0.15, the ΔC-reducing gas correlation varies depending on C/H in the reducing gas. Note that, in all the cases, the correlation between InputΔC and the carbon consumption in the reducing gas is represented by the upwardly convex graph (that is, when the carbon consumption in the reducing gas is a given value, a maximum value is shown).

[0070] Accordingly, by acquiring the ΔC-reducing gas correlation per C/H in the reducing gas in advance, the carbon consumption in the reducing gas can be determined such that InputΔC is a predetermined target value or higher based on the correlation. In addition, the amount of the reducing gas blown into the blast furnace can be adjusted based on the determined carbon consumption in the reducing gas and the carbon proportion in the reducing gas. As a result, desired InputΔC (that is, InputΔC that is the target value or higher) can be obtained. That is, the desired reducing material ratio reduction effect can be obtained, and further the reducing material ratio can be reduced more reliably. The blast furnace operation method according to the embodiment is made based on the above-described findings.

[0071] In the above-described example, the ΔC-reducing gas correlation is acquired by performing the blast furnace operation simulation, but the method of acquiring the ΔC-reducing gas correlation is not limited thereto. Likewise, in an operation in an actual furnace (including a real operation and a test operation) or a test operation in a test blast furnace, the ΔC-reducing gas correlation can be acquired by calculating InputΔC while changing the carbon consumption in the reducing gas.

[0072] <2. Blast Furnace Operation Method>

[0073] Next, the blast furnace operation method according to the embodiment will be described. The blast furnace operation method according to the embodiment includes first to third processes described below.

[0074] (2-1. First Process)

[0075] In the first process, the ΔC-reducing gas correlation is acquired per C/H in the reducing gas. The method of acquiring the ΔC-reducing gas correlation (acquisition method) is not particularly limited. For example, the ΔC-reducing gas correlation can be acquired by performing the blast furnace operation simulation. As the blast furnace operation simulation, for example, a so-called “Blast Furnace Mathematical Model” Kouji TAKATANI, Takanobu INADA, Yutaka UJISAWA, “Three-dimensional Dynamic Simulator for Blast Furnace”, ISIJ International, Vol. 39 (1999), No. 1, p. 15 to 22 can be used. In this blast furnace mathematical model, an internal region of the blast furnace is divided in a height direction, a radial direction, and a circumferential direction to define a plurality of meshes (small regions), and the behavior of each of the meshes is simulated. Calculation conditions of the blast furnace operation simulation are not particularly limited and are preferably determined depending on actual operation conditions. For example, the flame temperature is preferably 2000° C. or higher. Note that, as described above, even when the flame temperature changes, the ΔC-reducing gas correlation does not substantially fluctuate. By performing the blast furnace operation simulation, the ΔC-reducing gas correlation is acquired per C/H. That is, a graph showing the ΔC-reducing gas correlation is acquired. Here, as described above, the ΔC-reducing gas correlation tends to vary between a case where C/H in the reducing gas is 0.15 or higher and a case where C/H in the reducing gas is lower than 0.15. Therefore, it is preferable that plural types of ΔC-reducing gas correlations of various cases are obtained.

[0076] The method of acquiring the ΔC-reducing gas correlation will be described in more detail. The blast furnace operation simulation is performed using various cases where C/H in the reducing gas and the amount of the reducing gas blown are (the amount thereof blown per ton of molten iron) are different from each other. The cases also include the base operation (operation where the reducing gas volume is 0). Here, it is preferable that the calculation conditions (operation conditions) are adjusted such that the flame temperature and the molten iron temperature are as constant as possible in the cases. In order to make the flame temperature constant, at least one of an blast volume or an oxygen enrichment ratio in hot blast may be adjusted. In addition to or instead of the adjustment of the above-described factors, at least one of the coke ratio and the pulverized coal ratio may be adjusted. As a result, InputΔC and the carbon consumption in the reducing gas are acquired per case. Incidentally, a point representing InputΔC and the carbon consumption in the reducing gas of each of the cases is plotted on, for example, the xy plane shown in FIG. 1. Points P1 to P8 are examples of the plotted points. Next, a fitted curve of each of the plots is acquired using an approximation method such as a least-squares method. These fitted curves form a graph showing the ΔC-reducing gas correlation. The graphs L1 to L5 are examples of the graph showing the ΔC-reducing gas correlation.

[0077] (2-2. Second Process)

[0078] In the second process, the carbon consumption in the reducing gas where InputΔC is a predetermined target value or higher is determined based on the ΔC-reducing gas correlation acquired in the first process. That is, the ΔC-reducing gas correlation corresponding to C/H in reducing gas to be actually used is selected, and the carbon consumption in the reducing gas where InputΔC is a predetermined target value or higher is determined based on the selected ΔC-reducing gas correlation. C/H in the reducing gas may be acquired, for example, by specifying the composition of the reducing gas using the above-described measurement method and acquiring C/H based on the specified composition of the reducing gas.

[0079] Here, as described above, the ΔC-reducing gas correlation is represented by the upwardly convex graph. Accordingly, it is preferable that the carbon consumption in the reducing gas is determined such that InputΔC is close to the maximum value. As a result, the reducing material ratio can be further reduced. For example, when C/H in the reducing gas to be actually used is 0.15 or higher, it is preferable that the carbon consumption in the reducing gas is determined in a range of 21 kg/t-pig to 107 kg/t-pig, and it is more preferable that the carbon consumption in the reducing gas is determined in a range of 21 kg/t-pig to 65 kg/t-pig. The reason for this is as described above. That is, by determining the carbon consumption in the reducing gas in the range of 21 kg/t-pig to 107 kg/t-pig, InputΔC can be made to be 4.0% or higher. Further, by determining the carbon consumption in the reducing gas in the range of 21 kg/t-pig to 65 kg/t-pig, InputΔC can be made to be high (specifically 4.0% or higher) with a smaller amount of the reducing gas blown. Further, even when the carbon proportion in the reducing gas is low (in particular, when the carbon proportion is lower than 0.6 kg/Nm.sup.3), the flame temperature can be stably maintained at the predetermined value or higher while increasing InputΔC.

[0080] When the reducing gas in which C/H is 0.15 or higher is used and the carbon consumption in the reducing gas is higher than 65 kg/t-pig, As described above, the flame temperature tends to decrease. Therefore, it is preferable that operation elements including the oxygen enrichment ratio are adjusted such that the flame temperature is a predetermined value (for example, 2000° C.) or higher. In addition, in the set value range, InputΔC decreases. Therefore, the carbon consumption in the reducing gas is excessively high compared to the amount required for in-furnace reduction such that the gas utilization factor decreases. Therefore, a countermeasure for improving the gas utilization factor may be taken. For example, the iron-bearing material may be changed to materials having excellent reducibility.

[0081] (2-3. Third Process)

[0082] In the third process, the amount of the reducing gas blown into the blast furnace (for example, the amount of the reducing gas blown per ton of molten iron) is adjusted based on the carbon consumption in the reducing gas determined in the second process and the carbon proportion in the reducing gas. For example, the amount of the reducing gas blow can be obtained by dividing the carbon consumption in the reducing gas by the carbon proportion in the reducing gas. By blowing the reducing gas into the blast furnace in the determined amount thereof blown, the desired reducing material ratio reduction effect can be obtained. The operation conditions other than the above-described conditions may be the same as those in the related art.

[0083] Schematically, while alternately charging the iron-bearing materials and coke in layers into the blast furnace from the top of the blast furnace, the reducing gas is blown into the blast furnace together with the hot blast from the tuyere provided in the blast furnace. The types of the iron-bearing materials and the coke are not particularly limited, and iron-bearing materials and coke that are used in the blast furnace operation in the related art can also be suitably used in the embodiment. The amount of the reducing gas blown is set to the value determined in the third process. The reducing gas may be, for example, one or more selected from the group consisting of COG, natural gas, reformed top gas (BFG), and city gas. The reducing gas may be mixed gas of the gases or hydrogen mixed gas obtained by mixing hydrogen gas with the gases (including the mixed gas). In particular, the reducing gas in which C/H is lower than 0.15 may be produced by mixing hydrogen gas with COG or the like.

[0084] The reducing gas may be blown into the blast furnace without being heated but is preferably blown into the blast furnace after being heated. By blowing the reducing gas into the blast furnace after being heated, further reduction in reducing material ratio can be expected. The heating temperature is preferably about 300° C. to 350° C.

[0085] The tuyere for blowing the reducing gas into the blast furnace (hereinafter, also referred to as “tuyere for reducing gas”) is provided in, for example, a bosch part. The tuyere for reducing gas may be provided in a shaft part. The tuyere for reducing gas may be provided in both of the shaft part and the bosch part. The reducing gas blown from the shaft part preferably includes a large amount of CO and/or H.sub.2 and is blown while managing C/H.

[0086] As in the blast furnace operation in the related art, the hot blast is blown into the blast furnace. The temperature of the hot blast, the composition thereof, and the amount thereof blown may be the same as those of the blast furnace operation in the related art. For example, the hot blast includes air and may further include hygroscopic moisture and enriched oxygen. The hot blast is blow into the blast furnace, for example, from the tuyere provided in the bosch part. The tuyere for blowing the hot blast into the blast furnace may be common to or different from the tuyere for reducing gas.

[0087] As described above, in the embodiment, the carbon consumption in the reducing gas where InputΔC is a predetermined target value or higher is determined based on the ΔC-reducing gas correlation acquired in advance, and the amount of the reducing gas blown is determined based on the determined carbon consumption in the reducing gas and the carbon proportion in the reducing gas. Accordingly, desired InputΔC can be realized relatively reliably. That is, the desired reducing material ratio reduction effect can be obtained, and further the reducing material ratio can be reduced more reliably. As a result, CO.sub.2 emissions can be reduced. Further, according to the ΔC-reducing gas correlation, when the carbon consumption in the reducing gas is a given value (this value varies depending on C/H), InputΔC shows a maximum value. Accordingly, when the set value of the carbon consumption in the reducing gas is determined such that InputΔC is close to the maximum value, the reducing material ratio can be further reduced. Further, by acquiring the correlation per C/H, the amount of the reducing gas blown can be determined and managed based on the correlation corresponding to C/H in the reducing gas. Accordingly, operation elements required for increasing InputΔC can be appropriately set and managed.

EXAMPLES

[0088] Next, the effects of one aspect of the present invention will be described in more detail using examples. However, conditions of the examples are merely exemplary to confirm the operability and the effects of the present invention, and the present invention is not limited to these condition examples. The present invention can adopt various conditions within a range not departing from the scope of the present invention as long as the object of the present invention can be achieved under the conditions.

1. Example 1

[0089] In Example 1, by performing the blast furnace operation simulation, it was verified that, when C/H was 0.15 or higher, the ΔC-reducing gas correlation was present.

[0090] In the blast furnace operation simulation, the above-described “blast furnace mathematical model” was used. Calculation conditions are shown in Table 3. The iron-bearing materials were all sintered ores. In addition, the composition of the sintered ores was T-Fe: 58.5%, FeO: 7.5%, C/S: 1.9, and Al.sub.2O.sub.3: 1.7%. In addition, regarding coke, a case where coke having a composition of C: 87.2% and Ash: 12.6% was used was assumed (“%” represents “mass %” in all the cases).

TABLE-US-00003 TABLE 3 Calculation Conditions Productivity t/d/m.sup.3 2.71~2.81 Blast volume Nm.sup.3/t-pig   4~1035 Oxgen enrichment ratio %  7.6~78.9 Hygroscopic Moisture of Air g/Nm.sup.3 5 Flame temperature ° C. 2175~2225

[0091] In Example 1, by performing the blast furnace operation simulation while changing the type of the reducing gas (that is, the value of C/H) and the amount of the reducing gas blown (the amount of the reducing gas blown per ton of molten iron), it was verified that the ΔC-reducing gas correlation was present. As the reducing gas, the COG having the composition shown in Table 1 or the natural gas having the composition shown in Table 2 was used. The reducing gas was blown into the blast furnace from the tuyere provided in the bosch part. The blast volume and the oxygen enrichment ratio in the hot blast were adjusted such that the flame temperature was as constant as possible (that is, was a value in the range shown in Table 3) when the reducing gas was blown. Note that the oxygen enrichment ratio was adjusted such that the flame temperature was 2085° C. in Case 8 and the oxygen enrichment ratio was adjusted such that the flame temperature was 2315° C. in Case 9. Further, in all the cases, the coke ratio was adjusted such that the molten iron temperature was constant. As fixed conditions, the pulverized coal ratio was 115 kg/t-pig, and the blast temperature was 1000° C. The calculation results are shown in Table 4 and FIG. 1.

TABLE-US-00004 TABLE 4 Calculation Results Type of Reducing Carbon Reducing gas Consumption in Gas volume Reducing Gas InputΔC — Nm.sup.3/t-pig kg/t-pig % Case 0 None 0 0 0.0 Case 1 COG 88 21 4.0 Case 2 COG 191 46 5.6 Case 3 COG 244 59 6.1 Case 4 Natural Gas 55 34 4.9 Case 5 Natural Gas 106 65 6.6 Case 6 Natural Gas 141 87 6.2 Case 7 Natural Gas 173 107 4.2 Case 8 COG 195 46 5.6 Case 9 COG 193 45 5.4

[0092] As shown in Table 4 and FIG. 1, it was able to be verified that the ΔC-reducing gas correlation was present. Further, it was also verified that, when C/H in the reducing gas was 0.15 or higher, the ΔC-reducing gas correlation was uniquely determined irrespective of the type of the reducing gas (in other words, irrespective of C/H in the reducing gas). By determining the carbon consumption in the reducing gas where InputΔC is a predetermined target value or higher using the ΔC-reducing gas correlation and determining the amount of the reducing gas blown based on the determined carbon consumption in the reducing gas, the reducing material ratio can be reduced more reliably, and further CO.sub.2 emissions can be reduced.

2. Example 2

[0093] In Example 2, by performing the blast furnace operation simulation, it was verified that, when C/H was lower than 0.15, the ΔC-reducing gas correlation was present.

[0094] In the blast furnace operation simulation, the above-described “blast furnace mathematical model” was used. Calculation conditions were the same as those of Example 1. In addition, it was assumed that the same iron-bearing materials and the same coke as those of Example 1 were used.

[0095] In Example 2, by performing the blast furnace operation simulation while changing C/H in the reducing gas and the amount of the reducing gas blown (the amount of the reducing gas blown per ton of molten iron), it was verified that the ΔC-reducing gas correlation was present. In the actual operation, C/H in the reducing gas can be adjusted, for example, by mixing the COG having the composition shown in Table 1 with hydrogen gas at a different mixing ratio per case. The reducing gas was blown into the blast furnace from the tuyere provided in the bosch part. The blast volume and the oxygen enrichment ratio in the hot blast were adjusted such that the flame temperature was as constant as possible (that is, was a value in the range shown in Table 3) when the reducing gas was blown. Further, in all the cases, the coke ratio was adjusted such that the molten iron temperature was constant. As fixed conditions, the pulverized coal ratio was 115 kg/t-pig, and the blast temperature was 1000° C. The calculation results are shown in Table 5 and FIG. 1.

TABLE-US-00005 TABLE 5 Calculation Results C/H in Reducing Carbon Reducing gas Consumption in Gas volume Reducing Gas InputΔC — Nm.sup.3/t-pig kg/t-pig % Case 0 — 0 0 0.0 Case 1 0.054 99 6.1 5.7 Case 2 0.054 199 12.1 10.0 Case 3 0.054 294 17.9 11.8 Case 4 0.054 396 24.1 11.9 Case 5 0.097 99 11.1 5.6 Case 6 0.097 199 22.5 8.3 Case 7 0.097 292 33.0 10.2 Case 8 0.097 393 44.4 9.9 Case 9 0.137 98 16.1 4.2 Case 10 0.137 199 32.9 8.0 Case 11 0.137 293 48.4 9.2 Case 12 0.137 340 56.1 9.0 Case 13 0.02 99 2.6 5.1 Case 14 0.02 390 10.2 13.7 Case 15 0.02 473 12.3 13.4

[0096] As shown in Table 5 and FIG. 1, it was able to be verified that the ΔC-reducing gas correlation was present. Further, it was also verified that, when C/H in the reducing gas was lower than 0.15, the ΔC-reducing gas correlation varied depending on C/H in the reducing gas. Accordingly, the correlation corresponding to C/H in the reducing gas is selected, and the carbon consumption in the reducing gas where InputΔC is a predetermined target value or higher is determined using the selected correlation. By determining the amount of the reducing gas blown based on the determined carbon consumption in the reducing gas, the reducing material ratio can be reduced more reliably, and further CO.sub.2 emissions can be reduced.