JP3051193B2 - Manufacturing method of coke for metallurgy - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a method for producing a coke of a desired quality even when a fine or non-caking coal having a low fluidity or exhibiting no fluidity such as a slightly caking coal or weathered coal is blended. The present invention relates to a method for producing a metallurgical coke that can be obtained.

[0002]

2. Description of the Related Art Metallurgical coke is used as an auxiliary material in blast furnace ironmaking. Since the properties of metallurgical coke directly affect the operation of the blast furnace, great care is taken to ensure the required properties. Among the above properties, its hardness is particularly important. There are various indices indicating the hardness of coke, but when coke is used for metallurgy, the drum strength DI is usually used.

[0003] The drum strength DI is determined by charging a predetermined drum with a predetermined amount of coke, rotating the drum at a predetermined speed a predetermined number of times, taking it out, and sieving it with a predetermined sieve. It is expressed by the ratio of the weight of coke on the sieve of a predetermined sieve to the weight of coke charged.

Usually, the number of rotations of the drum is 30 or 150, and the sieve is 15 mm, 30 mm, 50 mm, etc. In the following description, the number of rotations is 1 if necessary.
Drum strength DI ¹⁵⁰ _{15 with} 50 rotations and a sieve of 15 mm (JIS
The drum strength and DI specified in K2151 are represented by Drum Index).

[0005] Coke for metallurgy is obtained by blending a plurality of types of raw coal (simple coal) into charged coal, charging the charged coal into a coke oven and carbonizing.

[0006] The above-mentioned drum strength DI, which is an important factor for metallurgical coke, depends on the properties of the charged coal. That is, when other operating conditions are constant, the drum strength DI of the obtained coke is determined according to the properties of the charged coal. Since the charged coal is a mixture of plain coals, blending control is important in producing metallurgical coke.

Therefore, various tests are performed in advance to determine the relationship between the characteristics representing the properties of various raw coals and the drum strength DI of the obtained coke. From the known properties of the char, it is possible to estimate the drum strength DI of the resulting coke.

[0008] The characteristics of such raw coal are conventionally known as its volatile matter content and various properties obtained from a test using a Giesera plastometer (a heat softening and melting property test of coal specified in JIS M8801). Numerical values have been used, but in addition to these, various properties of coal microstructure components have been recently adopted.

The above-mentioned coal microstructure components are roughly classified into vitrinite, ecginite and inertinite, of which vitrinite and ecginite are important. Vitrinite is a microstructure mainly derived from the woody part of a plant, while ecginite is a leaf, twig, spore,
It is a fine tissue derived from pollen, seeds, water algae and the like. Both of them are softened and melted when heated, and when the raw coal becomes coke, it reacts itself and plays a role of bonding with each other. Inertia knit is, as the name implies, an inactive component in the raw coal and becomes an aggregate of coke. Thus, when evaluating raw coal, consideration has been given to the values of these coal microstructure components.

For example, in Japanese Patent Application Laid-Open No. 2-97589, the reflectance of vitrinite expressing the amount of vitrinite Ro (reflectance of vitrinite portion under a microscope)
The combination of the maximum flow rate MF (the logarithmic value of the value indicating the maximum rotation speed of the rotor in the test using the above Giesra-Plastometer) is used for the composition control. Here, the reflectance Ro of vitrinite is a parameter indicating the degree of coalification of the raw coal, and it can be said that the larger this value is, the more advanced the degree of coalification is. Further, the maximum flow rate MF of the ghee cellar is an index representing the caking property of the raw coal, and it can be said that the larger the value, the greater the caking property.

If the reflectance Ro of vitrinite and the maximum flow rate MF of the ghee cellar are determined in advance by test for plain coal, the characteristic values of the plain coal are determined by weighting the characteristic values of the plain coal in a blending ratio. The characteristic value of the charged coal is calculated, and the drum strength DI corresponding to the calculated characteristic value is obtained.

This system is based on the fact that the characteristics of the charged coal will be proportional to the blending ratio of the plain coal, that is, the additive properties will be established. Japanese Patent Application Laid-Open No. Sho 62-119291, which is filed by the present applicant, is known as a method for evaluating raw coal that does not use the reflectance Ro of vitrinite and the maximum flowability MF of the gee cellar.
Here, the hydrogen to carbon atom ratio H / C of the active component of the coal is
And the caking property and / or coking property of the coal are evaluated using the oxygen / carbon atom ratio O / C of the component as an index.

[0013]

In recent years, due to changes in raw material circumstances and improvement in operating techniques, fine and non-coking coal, which has been rarely used as a raw material for coke for metallurgy, has been used as plain coal for blending. It is becoming.

Since such fine and non-coking coals show little or no flowability of the gee cellar (that is, they hardly soften and melt even when heated), it is necessary to conduct a test on them using a Giesla plastometer beforehand. Cannot be performed, resulting in the highest flow rate
Will not be obtained. Even if the data is obtained, since it is far from the normal value, the data is additive (the data of plain coal is weighted average weighted by the blending ratio of the charged coal). It is not guaranteed that the above-mentioned method will satisfy the characteristics applicable to charged coal data).

The method disclosed in Japanese Patent Application Laid-Open No. Sho 62-119291 is notable in that the ratio of the hydrogen to carbon atom ratio H / C of the active component of the raw coal and the oxygen to carbon atom ratio O / C of the component are used as indices. However, when fine and non-coking coal is blended, there is a limit that additivity is not established.

In view of such circumstances, the present invention is based on the use of the hydrogen / carbon atom ratio H / C of the active component of the raw coal and the oxygen / carbon atom ratio O / C of the component as indices. However, an object of the present invention is to provide a method of producing coke while accurately estimating the drum strength DI of coke even when fine or non-caking coal having low or no fluidity is blended. Is what you do.

[0017]

SUMMARY OF THE INVENTION A method for producing metallurgical coke according to the present invention is characterized in that a charged coal obtained by blending several types of plain coals, including fine and non-coking coals, is carbonized to a predetermined drum. A method for producing coke having a strength DI, wherein the hydrogen / carbon atom ratio H / C and the oxygen / carbon atom ratio O / C of the active ingredient of the plain coal to be previously blended and the drum strength of the plain coal The relationship with DI is determined, and the hydrogen / carbon atom ratio H / C and the oxygen / carbon atom ratio O / C of the active component of the charged coal are calculated by weighted average using the blending ratio of the plain coal as a weight. In estimating the drum strength DI of the charged coal, for the fine and non-coking coal in the plain coal to be blended, the hydrogen / carbon atom ratio H / C and the oxygen / carbon atom number of the active component are determined. The ratio O / C is an index corrected so as to show additivity even when blended with caking coal. That apparent hydrogen / carbon atomic ratio (H /
C) 'and the apparent oxygen / carbon atom ratio (O / C)', and these apparent hydrogen / carbon atom ratio (H / C) 'and apparent oxygen / carbon atom ratio (O / C)' Indices I _{H / C} , I
The drum strength DI of the charged coal is estimated using each of the _{O / C} , and based on the estimation, the type and blending ratio of the plain coal are selected to be charged, and the charged coal is loaded into a coke oven. And then carbonized.

Hereinafter, the present invention will be described in detail.

The method of the present invention is applied to the case where the charged coal obtained by blending several kinds of plain coals including fine and non-coking coals is carbonized.

Coal is a carbide obtained by fossilization of plants, and contains aromatic organic compounds. Therefore, the main constituent elements are mainly carbon, hydrogen, and oxygen, and the constituent ratio changes according to the kind and part of the original plant, and the degree of carbonization. Of the above three elements, carbon is naturally the largest. That is, carbon is a substrate of coal, and it is considered that the properties of the coal are determined to some extent by the amounts of hydrogen and oxygen bonded thereto.

Therefore, in the present invention, basically, the hydrogen / carbon atom ratio H /
The relationship between C and the oxygen / carbon atomic ratio O / C and the drum strength DI of the plain coal is determined in advance, and the hydrogen content of the active component of the charged coal is determined by a weighted average using the blending ratio of the plain coal as a weight. / Carbon atom ratio H / C and oxygen / carbon atom ratio O /
A method of estimating the drum strength DI of the charged coal by obtaining C is adopted.

When the plain coal is caking coal, each plain coal is
Ratio of hydrogen / carbon atoms in active ingredient H / C and oxygen / carbon
The atomic ratio O / C is H / C, and the O / C is vertical and horizontal.
When plotted on a graph with axes, H / C and O / C
Drum strength DI determined by^{15 0} ₁₅ Is obtained.
(See FIG. 2). This is due to the fact that plain coal water
Of the elemental / carbon atom ratio H / C and the oxygen / carbon atom ratio O / C
The value indicates that the drum strength DI is determined.
You.

In the case where the plain coal is caking coal, when a plurality of types of caking coal are blended to form a charged coal, the hydrogen / carbon atom ratio H / C and the oxygen / carbon atom ratio It is confirmed that additivity is satisfied for both O / C. Therefore, the hydrogen / carbon atom ratio H / C and the oxygen / carbon atom ratio O / C of the charged coal are obtained by performing a weighted average using the blending ratio of the plain coal as a weight. The drum strength DI of the charged coal can be accurately estimated from the relationship between the hydrogen / carbon atom number ratio H / C and the oxygen / carbon atom number ratio O / C and the drum strength DI. Based on this estimation, the type and blending ratio of the plain coal are selected to be charged coal, and the charged coal is charged into a coke oven and carbonized to obtain coke having a desired drum strength DI.

On the other hand, when a fine or non-caking coal having a low or no caking property is blended into the charged coal,
Hydrogen / carbon atom ratio H / C and oxygen / carbon atom ratio O
Even if the weighted average of the mixing ratio based on the value of / C is used, it does not match the actually obtained coke drum strength DI. This is because the hydrogen / carbon atom ratio H / C and the oxygen / carbon atom ratio O / C of the fine and non-coking coal show different behaviors from those of the caking coal, and the additive properties are not always established. Because it does not.

Therefore, in the present invention, the fine / non-caking coal among the plain coals to be blended has a hydrogen / carbon atom ratio H / C and an oxygen / carbon atom ratio O / C of the active component.
The apparent hydrogen / carbon atom ratio (H / C) ', which is an index corrected so as to show additivity even by blending with caking coal, and the apparent oxygen / carbon atom ratio (O / C) ) '.

This correction is carried out by preparing charged coal in which the blending ratio is changed for each brand of fine and non-coking coal (for each plain coal), carbonizing them to form coke, and the drum strength DI And by comparing with the values calculated by the weighted average and correcting the values of the hydrogen / carbon atom ratio H / C and the oxygen / carbon atom ratio O / C so that those values match. . The hydrogen / carbon atom ratio H / C of the plain coal thus corrected is apparently hydrogen / carbon atom ratio (H / C) ',
The corrected oxygen / carbon atomic ratio O / C is defined as an apparent oxygen / carbon atomic ratio (O / C) '. In the case of caking coal, (H / C) and (H / C) 'naturally have the same value even if such correction is performed, and O / C and (O / C)' Has the same value as

The indices I _{H / C} and I _{O / C} taking into account the apparent hydrogen / carbon atom ratio (H / C) ′ and the apparent oxygen / carbon atom ratio (O / C) ′ are used, respectively. The drum strength DI of the charged coal is estimated, and based on the estimation, the type and blending ratio of the plain coal are selected to be charged, and the charged coal is charged into a coke oven and carbonized. Such an index I _{H / C} ,
By using _{IO / C} , the drum strength DI of charged coal can be estimated with the same accuracy as when only fine coal is used, even when fine and non-coking coal is used. Become.

When the above indexes I _{H / C} and I _{O / C} are used,
The composition management in the actual operation is represented as shown in FIG. In FIG. 7, for reference, the reflectance Ro of plain coal vitrinite and the tie line of the maximum flow rate MF of the gee cellar are also shown. While the distribution is reduced as much as possible, the change in the coking direction (direction B in the figure) is smaller than the change in the direction of coalification (direction A in the figure) because of the increase in the distribution of low-rank coal as much as possible. Because the operation is limited to a narrow range and there are many operations near the lower limit of the fluidity, the change in the vertical direction of the figure, that is, I _{H / C} -I _{O / C} (hereinafter referred to as ΔI) is more important for the control of the mixture. It becomes. Therefore, in application to the following actual operation, I as an index corresponding to a change in the degree of coalification
_{A combination of O / C} and ΔI as an index corresponding to caking was used.

In the actual operation using a coke oven, factors that affect the drum strength DI include not only the mixing ratio of the plain coal, but also the furnace temperature FT and other operating conditions. If these are not taken into consideration, a more accurate estimated value cannot be obtained. However, it is impossible to consider all the operating conditions as factors for estimating the drum strength DI. Therefore, the furnace temperature FT, which is a factor having a large contribution, is taken up, and the above-mentioned indexes I _{H / C} , I It is desirable to estimate the drum strength DI using _{O / C.}

When the indexes I _{H / C} and I _{O / C} are used, the drum strength DI of coke decreases as I _{O / C} increases.
The value decreases as ΔI departs from the value ΔI _{max of} ΔI that maximizes DI. The value of _{ΔI max} decreases as I _{O / C} decreases. On the other hand, the actual operation data
When the furnace temperature FT is higher, the value of _{ΔI max} decreases, and the drum strength DI decreases sharply due to the deviation of _ΔI from ΔI _max (see FIG. 8).

From the above, the following relational expression can be given as a preferable estimation expression of the drum strength DI.

DI = (DI) _max −F (FT) × (△ I _max − △ I) ² (DI) _max = ab−I _{O / C} + c · FT F (FT) = d−e · FT ΔI _max = f + g · I _{O / C} −h · FT

(DI) _{max in the} first equation is the maximum value of the drum strength DI which can be taken. As shown in the second equation, the index I _{O / C of} the coal blend and the furnace temperature FT of the coke oven at the time of operation are obtained. Is determined centrally.

The first equation is an equation for estimating the drum strength DI which can be actually obtained by subtracting a negative factor from (DI) _max obtained by the second equation. With the correction calculation, it is possible to accurately estimate the drum strength DI in the actual furnace operation.

F (FT) in the first equation is the furnace temperature F
This is a function with T as a variable, and is represented by the third equation.
The capital letter F is used as a function symbol to avoid confusion with the constant f.

ΔI in the first equation is the I of the blended coal.
_{H / C- IO / C.} This value includes the DI
There is a value ΔI _max such that is maximum. Therefore, △
The difference between I _max and ΔI is a negative factor for DI. Such a value of _{ΔI max} is represented by the fourth equation.

A, b, c, d, e, f, g, and h are constants, which are determined by analyzing the actual furnace operation data by a statistical method.

At the time of blending, the drum strength DI is calculated each time using the above-mentioned relational expressions, and the type and blending ratio of the plain coal are selected by trial and error to be charged coal.
By charging the charged coal into a coke oven and carbonizing, coke of desired quality can be obtained.

Can baking test firing for each receiving lot, H / C,
Direct measurement of O / C is actually complicated. Therefore, as a result of diligent studies, there is a high degree of correlation between I _{H / C} and the weight loss of raw coal, and a high degree of correlation with I _{O / C} and the amount of CO generated from raw coal. (See FIGS. 12 and 13). These values can be relatively easily measured simultaneously, for example, by a device combining a thermobalance and a gas chromatograph. Therefore, from the correlation between the heating loss of raw coal and the amount of coke oven gas generated, I
It is also preferable to obtain _{H / C} and I _{O / C.}

[0040]

[Function] When carbonizing coal obtained by blending several types of plain coals, including fine and non-coking coal, by carbonization, the hydrogen / carbon atomic ratio of caking coal H / C and oxygen / carbon atomic number ratio O / C, fine / non-coking coal, corrected hydrogen / carbon atomic number ratio (H / C) '
And apparent oxygen / carbon atom ratio (O / C) '
An index I _{H / C} taking into account these apparent hydrogen / carbon atom ratio (H / C) ′, apparent oxygen / carbon atom ratio (O / C) ′,
The drum strength DI of the charged coal is estimated using each of I _{O / C.} Then, a combination of plain coals giving a drum strength DI equal to or higher than a predetermined value is selected to be charged coal, and the charged coal is charged into a coke oven and carbonized. The combination of plain coals is often performed by a try-and-error method, but the optimal blending may be determined centrally by an optimal programming method.

[0041]

The present invention will be further described with reference to the following examples.

Example 1 FIG. 1 is a scatter diagram showing the relationship between H / C and O / C of active and inactive components of plain coal. Figure 2 is a scatter diagram showing the values of DI ^0.99 ₁₅ corresponding to the H / C and O / C of the active ingredient Tan'ajisumi. Figure 3 is a correlation diagram showing the relationship between the readings from the measured values and Figure 2 of the DI ^0.99 ₁₅ in the case of mutually blended caking coal. FIG. 4 is a relational diagram showing a comparison between measured and calculated values of DI ¹⁵⁰ ₁₅ of the charged coal when fine and non-coking coal is blended with ordinary caking coal. is there. FIG. 5 shows the H / C of the active ingredient of the plain coal and the apparent H.
FIG. 7 is a correlation diagram showing a relationship with / C [that is, (H / C) ′]. FIG. 6 shows the O / C of the active ingredient of the plain coal and the apparent O / C.
FIG. 9 is a correlation diagram showing a relationship with C [that is, (O / C) ′].

According to the present invention, basically, the drum strength DI of coke obtained by carbonizing the raw coal is used to determine the hydrogen / carbon atomic ratio H / C and oxygen / carbon ratio of the active component of the plain coal to be blended. It is determined from the atomic ratio O / C.

Therefore, first, these indices are the drum strength D
A basic test was performed to determine if I was a suitable factor to estimate.

As described in detail above, the raw coal is composed of an active component such as vitrinite and ecdinite that softens and melts when heated and an inactive component that does not soften and melt. The active component and the inactive component are considered to have different effects on the substrate strength and the formation of the pore structure, which are the main controlling factors of the coke drum strength DI.
Each plain coal was divided into an active component and an inactive component, and the amounts of hydrogen, oxygen and carbon were measured by elemental analysis for each component, and the values of H / C and O / C were calculated.

FIG. 1 shows the relationship between them. In this figure, black circles and triangles indicate the active ingredients of each plain coal,
Open circles indicate inactive components. The black circles are the active ingredients that form agglomerates when firing in a can, and the solid triangles are the active ingredients that do not agglomerate when firing in a can. Note that the can baking test firing is, for example, a test in which a sample coal is charged into a petroleum can by a predetermined method and carbonized under a certain condition.
Since the dry distillation is performed under a constant firing condition, data comparison is easy.

As can be seen from FIG. 1, the active component and the inactive component behave differently from each other. In particular, regarding H / C, the active ingredient shows a larger value over almost the entire range of O / C, and the difference is remarkable.

This indicates that when predicting the drum strength DI of coke from the H / C and O / C of the plain coal, it is necessary to consider the active component and the inactive component separately. This is because the active component and the inactive component are considered to have different functions on coking of the raw coal.

Therefore, attention was paid only to the active ingredient, and baking tests of various types of plain charcoal were performed. FIG. 2 shows the relationship between H / C and O / C of the active ingredient in this case. This figure shows the DI of the coke sample obtained by the baking test firing.
^Although the actual measured value of ¹⁵⁰ ₁₅ is entered, an iso-intensity curve of DI ¹⁵⁰ ₁₅ value as shown by a dotted line was obtained.

From the above results, if the values of H / C and O / C of the active ingredient of a certain plain coal are known, the values of the drum strength DI of the coke obtained by carbonization based on the values are shown in FIG. Can be predicted.

As described above, in the actual production of coke, it is usual that a plurality of types of plain coals are blended to form a charged coal, and the charged coal is carbonized. Although actual data can be obtained by actually measuring the value of the drum strength DI of charcoal, it is not realistic in daily operation management because actual measurement requires a lot of labor and time. Usually, it is often adopted as a convenient method to calculate the drum strength DI of charged coal by weighing the known drum strength DI of plain coal by the blending ratio, and the value obtained in this manner is often used. The fact that is often inconsistent with the measured values is from experience.

Therefore, the H / C and O / C of the charged coal are calculated from the known H / C and O / C values of various plain coals by performing weighted averaging based on the blending ratio, and corresponding to the calculated values. It is expected that the method of reading and estimating the drum strength DI of the charged coal from FIG. 2 will be applied.

In order to confirm whether or not such estimation was correct, a test was conducted as to whether or not the actually measured value and the calculated value corresponded with an acceptable accuracy as follows.

That is, many kinds of blended coals were prepared by blending two kinds of plain coals (specifically, the blending ratio was changed), and baking test firing was performed on these blended coals. The resulting coke samples cans baked test firing was measured drum strength DI ^0.99 _15.

On the other hand, H of the plain coal constituting these blended coals
/ C and the weighted average from the values of O / C the compounding ratio as the weighting performed seeking H / C and O / C of blend coal were read drum strength DI ^0.99 ₁₅ based on FIG.

When the plain coal used in the test was caking coal, the measured drum strength DI ¹⁵⁰ obtained as described above was used.
FIG. 3 is a plot of ₁₅ and the calculated drum strength DI ¹⁵⁰ ₁₅ .

As can be seen from this figure, in the case of coking coal, there was a high degree of correlation between the two, and it was confirmed that so-called additive properties were established.

Next, the same test was conducted for the case where the plain coals to be blended were caking coal and fine / non-caking coal which did not show caking properties. Did not agree with the measured values.

[0059] Details of the test results using fine and non viscous coals are omitted, when summarize the results of the test series, relates the measured values and calculated values of the drum strength DI ^0.99 _15, ( 1) When only caking coal is used as plain coal and they are blended with each other, there is no significant difference between the calculated value and the measured value. (2) Use both caking coal and fine / non-caking coal as plain coal,
When they are blended, (a) the fine and non-coking coal is a low-rank coal such as weathered coal,
When the O / C of the active ingredient is larger than 0.10, the difference between the calculated value and the measured value increases. (Refer to Fig. 6) (b) When the fine and non-coking coal is high-carbon coal such as anthracite or semi-anthracite, and the H / C of the active ingredient is smaller than 0.60, the calculated value and the measured value The difference increases. (See Fig. 5)
The result was obtained.

The present invention is based on the finding of a universal calculation method applicable to the above (2) (a) and (b), and the basic concept is as follows. It is. In other words, in the case of ordinary caking coal, whatever the brand, it can be interpreted that the ratio of hydrogen and oxygen in the active ingredient that participate in coking is almost constant. It can be explained that the above (1) holds.

On the other hand, in the case of fine / non-coking coal whose O / C is larger than 0.10 or H / C is smaller than 0.60, the ratio of hydrogen and oxygen in the active component which contribute to coking is small. It is considered to be different and will be different on a case-by-case basis.

Therefore, in the present invention, the measured values of H / C and O / C of the active component of the fine / non-coking coal are corrected so that they can be treated in the same manner as in the case of normal coking coal. That is, by adopting this correction value, O / C becomes 0.10.
Even in the case of blending a fine / non-coking coal having a larger or H / C smaller than 0.60, the charged coal (H / C) 'and (O / C)' are used as universal indices. It can be applied to the estimation of the drum strength DI of blended coal.

FIG. 4 shows an example of this. In this figure, it is compared with the measured value of the coal blend in drum strength DI ^0.99 ₁₅ in the case of a coal blend by blending the finely-non viscous coals normal caking coal and the calculated values. That is, in this figure, a drum strength DI ^{15 0} ₁₅ on the vertical axis, taking the blending ratio of the fine, non-tacky coals on normal caking coal on the horizontal axis, measured in the coal blend in drum strength DI ^0.99 ₁₅ thereto Both the value and the calculated value have been written.

In this case, three kinds of raw coals of coking coal brand A brand B fine and non-coking coal brand C were used as samples used in the test. The properties are as shown in Table 1.

Table 1 の Brand name of canned coke of active ingredient Charcoal type H / CO / C DI ¹⁵⁰ ₁₅ ────────────────────────────────── A Caking coal 0.71 0.051 85.1 B cohesive coal 0.75 0.055 82.7 C fine / non-cohesive coal 0.64 0.152 − ─────────────────────────────── ───

In the above table, the drum strength D of brand C is shown.
There is no description of I ¹⁵⁰ _{15 because} DI ¹⁵⁰
_This is because a lump coke sample capable of measuring ₁₅ was not obtained.

First, brands A and C were blended at the blending ratios shown in Table 2 below, and the weighted average of the blending ratios was used to calculate the H / C and O / C of the active ingredients of the blended coal. The drum strength DI ¹⁵⁰ ₁₅ of the coal blend was read from FIG. 2, and the drum strength DI of the coke sample obtained by the baking test was separately conducted.
¹⁵⁰ ₁₅ was measured. Table 2 also shows the results.

Table 2 ─────────────────────────────────── Mixing ratio Weighted average value DI ¹⁵⁰ ₁₅ AC H / CO / C Fig. 2 Reading Actual measured value difference ─────────────────────────────────── 100% 0 % 0.71 0.051 85.1 85.1 0 95 5 0.71 0.054 85.0 83.8 1.2 90 10 0.70 0.061 82.0 73.8 8.2 85 15 0.70 0.066 75.2 69.2 6.0 ─────────────────────── ────────────

Further, brands B and C were blended, and the same test as above was conducted. The results are shown in Table 3 below.

Table 3 ─────────────────────────────────── Mixing ratio Weighted average value DI ¹⁵⁰ ₁₅ AC H / CO / C Fig. 2 Reading Actual measured value difference ─────────────────────────────────── 100% 0 % 0.75 0.055 82.7 82.7 0 95 5 0.74 0.060 83.0 81.2 1.8 90 10 0.74 0.065 81.5 76.2 5.3 85 15 0.73 0.070 75.0 70.8 4.2 ─────────────────────── ────────────

FIG. 4 is a graph showing the drum strength DI ¹⁵⁰ ₁₅ in Tables 2 and 3, and FIG. 4 is a graph obtained by reading FIG. 2 from the weighted average and obtaining DI ¹⁵⁰ ₁₅ .
In contrast, the measured value of the DI ^0.99 ₁₅ due cans baked test firing was indicated by circles. Open circles are coal blends of coals A and C, and black circles are coal blends of coals B and C. As can be seen from the above table and FIG. 4, a considerable difference is observed between the calculated value and the actually measured value. This makes it impossible to treat fine and non-coking coal in the same rank as coking coal.

Therefore, a correction is made so that they can be handled in the same row, and the O / C of the fine / non-coking coal is set to 0.20. The basis for setting 0.20 is as follows.

As can be seen from Tables 2 and 3, the fluctuation of H / C of the coal blend shows a very small value as compared with the fluctuation of O / C. That is, when the blending ratio is variously changed, the O / C of the blended coal fluctuates considerably, but the H / C shows a substantially constant value.

Therefore, attention was paid to the O / C having a large fluctuation, and the O / C of the fine / non-coking coal was corrected and the back calculation was performed so that the weighted average value matched the actually measured value.

For example, in Table 2, the mixing ratio A: B = 95:
In the case of No. 5, if the O / C of the blended coal is 0.058, the weighted average value will meet the actually measured value of the drum strength DI ¹⁵⁰ ₁₅ of the blended coal (in FIG. 2, the H / C 0.071 line). The value of O / C at which the DI ¹⁵⁰ ₁₅ becomes 83.8, which is the actually measured value, is read to obtain 0.058).

Therefore, the weighted average shows that the O / C of the coal blend is
The following equation is obtained by setting the value of O / C of the fine / non-coking coal to be 0.058 as y. 0.95 × 0.051 + 0.05 × y = 0.588 When this equation is solved to obtain y, y = 0.191 is obtained.

As a result of performing such a calculation for other items in Tables 2 and 3, the brand C as shown in Table 4 was obtained.
Was obtained (O / C) ′.

Table 4 合 Mix with actual measured value Correction of brand C of blended coal DI ¹⁵⁰ ₁₅ O / C (O / C) '─────────────────────────────── ──── A: C 95: 5 83.8 0.058 0.191 A: C 90:10 73.8 0.067 0.211 A: C 85:15 69.2 0.072 0.191 B: C 95: 5 81.2 0.063 0.215 B: C 90:10 76.2 0.070 0.205 B : C 85:15 70.8 0.075 0.188 ─────────────────────────────────── Average 0.200 ───── ──────────────────────────────

This means that the O / C of brand C, which is fine and non-coking coal, is a corrected value instead of the actually measured value of 0.152.
A value of 0.200 indicates that the calculated value (read value) and the measured value match well. Therefore, O of brand C newly
/ C was placed and 0.200, indicating, for example, drum strength DI ^0.99 ₁₅ obtained as a result of re-calculation in Table 5.

Table 5 ────────────────────────────────── (a) Measurement Figure 2 Reading DI ¹⁵⁰ ₁₅ Deviation from actual measured value DI ¹⁵⁰ ₁₅ (b) Before correction (c) After correction (b)-(b) (c)-(b) O / C adopted O / C adopted X Y ──────── ────────────────────────── 83.8 85.0 83.6 1.2 -0.2 73.8 82.0 75.0 8.2 1.2 69.2 75.2 68.0 6.0 -1.2 81.2 83.0 82.0 1.8 0.8 76.2 81.5 76.0 5.3 -0.2 70.8 75.0 70.0 4.2 -0.8 ────────────────────────────────── Average 75.8 80.3 75.8 4.45- 0.4 unbiased variance (σ _n-1 ) 2.6395 0.9180 number of data points (n) 6 6 ──────────────────────────────── ──

Whether or not there was a significant difference in the deviations X and Y from the actually measured values was tested as follows. | Average value of X−average value of Y | = | 4.45 − (− 0.4) | = 4.85 2.97 × [(σ _{n-1 (X)} ² / n _X ) + (σ _{n-1 (Y)} ² / n ^{_{Y)] 1/2 = 2.97 × [}} (2.6395 2 /6)+(0.9180 2/6)] 1/2 = 3.39

Therefore, | average value of X−average value of Y | ≧ 2.97 × [(σ _{n-1 (X)} ² / n _X ) + (σ _{n-1 (Y)} ² / n _Y )] ^{1 / 2} That is, there is a significant difference at 0.3% risk (99.7% confidence).

FIG. 4 shows the corrected O / C (that is, (O / C
The drum intensity DI ^0.99 _{1 5} obtained with C) ') was filled in by a solid line, it can be seen that the measured values and the calculated values match very well.

As described above, if the H / C value and the O / C value are corrected in advance for the raw coal other than the normal coking coal, and the values are used, all the raw coals are caking. It can be handled in the same row as charcoal.

Since the fluctuation range of H / C is small, it is not necessary to make any particular correction, but a better result can be obtained if it is corrected.

The apparent H / C and apparent O / C of the active ingredient of the fine / non-coking coal thus obtained are (H / C) 'and (O / C)', respectively.

The relationship between the actually measured H / C and (H / C) ′ of the active ingredient of the fine / non-coking coal determined by a test in advance is as shown in FIG. 5, and the actually measured O / C and (O / C) C) 'is as shown in FIG. From these figures, if the measured H / C and O / C of the active component of the fine / non-coking coal are known, (H / C) ′, (O
/ C) 'can be immediately known.

From the above results, it was found that the plain coal
Index I considering (H / C) 'and (O / C)'_{H / C} , I
_{O / C} The weighted average of
Index I _{H / C} , I_{O / C} Is calculated from FIG.
¹⁵⁰ ₁₅ The coke obtained by carbonization of coal blend
Drum strength DI¹⁵⁰ ₁₅ Can be accurately estimated.
Wear.

Example 2 FIG. 7 is a graph showing the relationship between the ranges of charged coals usually used in producing metallurgical coke when the I _{H / C} and I _{O / C} of the raw coal are considered as indices. FIG. Figure 8 is a scatter diagram showing the relationship between a furnace temperature FT another △ I and the drum strength DI ^0.99 ₁₅ coke oven. Figure 9 is a over time graph showing a comparison of the differences in the measured values of the drum strength DI ^0.99 ₁₅ in actual furnace operation and the calculated values. 10, in the actual furnace operation, it is over time graph showing a comparison of the differences in the measured values of the drum strength DI ^0.99 ₁₅ when using the weathered coal and calculated values. Figure 11 is the actual furnace operation, it is over time graph showing a comparison of the difference between the calculated and measured values of the drum strength DI ^0.99 ₁₅ When using a low volatiles instrumentation Nyusumi.

Although it has been described above that the method described in detail in Embodiment 1 can be applied to actual furnace operation, other factors are involved in actual actual furnace operation. Therefore, studies were repeated to more accurately estimate the drum strength DI of the charged coal in the actual furnace operation.

As in the first embodiment, as in the first embodiment, there is no difference in that I _{H / C} and I _{O / C} are regarded as major factors with respect to the drum strength DI. . Of these, attention is paid particularly to I _{O / C} , and the drum strength DI which is the largest in correlation with it is first estimated.
Thereafter, correction of fine adjustment is performed to estimate the drum strength DI that can be actually obtained. Although I _{O / C} is a negative factor for the drum strength DI, the particular attention was paid to the fact that as described above, the fluctuation was larger than the I _{H / C} and the negative of the drum strength DI. Is highly significant, and the drum strength DI can be accurately estimated by using I _{O / C} as an independent variable. However, correction is performed in order to increase the accuracy of the estimation value thus obtained (so-called rough estimation value, this value is regarded as the maximum value). At the time of this correction, the furnace temperature FT and I _{H / C} which are factors of the operating conditions are used.

[0092] Further, in this study, the charged coal is used instead of plain coal. I _{H / C} and I _{O / C} for coal charge
Is obtained by weighing those of plain coals using the blending ratio as a weight, and there is an additive property between them as described in Example 1.

When the IH _{/ C} and _{IO / C} of the raw coal are considered as indices, in the actual furnace operation, the shaded area in FIG. 7 is used as the generally recognized raw material for metallurgical coke. There is a range of coal charge.

For reference, the average reflectance R ₀ (shown by a two-dot chain line) of the conventionally used vitrinite and the isoline (tie line) of the maximum flow rate MF (shown by a one-dot chain line) of the vitrinite are shown in the figure. I wrote how it would be written.

[0095] The dotted line is the iso-intensity curve of the drum strength DI ¹⁵⁰ _15. In the actual furnace operation, the blending is carried out in a direction in which expensive high-flowability coal is reduced as much as possible due to the cost of the raw coal. Therefore, the fine and non-coking coal is used as much as possible. As a result, the caking direction (B in the figure) is compared with the change in the degree of coalification (A direction in the figure).
Direction) is limited to a narrow range. That is, the actual operation is often performed near the lower limit of the liquidity.

Therefore, the change in the vertical direction of the figure, ΔI = I _{H / C} −I _{O / C,} is rather important in the composition control. Here to minus the I _{O / C} is, I O _{/ C}
Is a negative factor for the drum strength DI, and I _{H / C}
This cancels out the effect of.

By the way, with respect to ΔI, there is an optimum value that maximizes the drum strength DI. The value varies depending on the type of coke oven and the sintering conditions such as oven temperature, but is generally about 0 to 0.3. The method of obtaining this value will be described later. If the value is ΔI _max , the square value of the difference between ΔI and the ΔI obtained each time the composition is used is the drum strength D.
It affect the I ¹⁵⁰ _15. This can be seen from FIG.

[0098] That is, is shown the relationship between △ I and the drum strength DI ^0.99 ₁₅ when changing the furnace temperature in the actual furnace operation in FIG. 8, tend to exhibit the appearance of the secondary curve,
The highest drum strength DI ^0.99 ₁₅ shown such △ I value or △ from I _max △ as I deviates drum strength DI
¹⁵⁰ ₁₅ drops.

On the other hand, as can be seen from FIG. 8, the DI when the difference between _ΔI and ΔI _max is 0 is also shown.
^{The value of 150} _15, that is, (DI ¹⁵⁰ ₁₅ ) _max varies depending on the furnace temperature FT.

This means that when the difference is other than 0,
(△ I _max - △ I) both drums intensity D and ² and RoAtsushi FT
This indicates a negative factor for I ¹⁵⁰ ₁₅ .

Based on the above results, first, the drum strength when there is no negative factor (DI ¹⁵⁰ ₁₅ ) _max
Is set as follows. In this case,
I _{/ C} and FT were selected as independent variables because
This is because the correlation with I ¹⁵⁰ ₁₅ ) _max is highly significant, and (DI ¹⁵⁰ ₁₅ ) _max can be accurately estimated. (DI ¹⁵⁰ ₁₅ ) _max = ab- _{IO / C} + c-FT

Here, a, b, and c are constants determined by the type and operation method of the actual furnace, and can be obtained by statistically analyzing a large number of actual operation data.

Next, (DI ¹⁵⁰ ₁₅ ) _max obtained from the above equation
By subtracting the above-mentioned negative factor from the above, the drum strength DI ¹⁵⁰ ₁₅ in the actual furnace operation is estimated. In this case, the estimation formula was set as follows.

(DI ¹⁵⁰ ₁₅ ) = (DI ¹⁵⁰ ₁₅ ) _max −F (FT) × (ΔI _max −ΔI) ² F (FT) = d−e · FT ΔI _max = f + g · I _{O / C} -H FT

D, e, f, g, and h are constants determined by the type and operating method of the actual furnace, similarly to the above a, b, and c, and are obtained by statistically analyzing a large number of actual operating data. be able to.

[0106] Using the equations obtained in this way, the compute an estimate of the pre drum strength DI ^0.99 _15, was measured actually produced actual measurement values of the coke drum strength DI ^0.99 _15. FIG. 9A shows the difference between the estimated value and the actually measured value. For comparison, the reflectance R ₀ of vitrinite and the maximum flow rate MF of the gee cell, which are conventional estimation methods, are compared.
The difference between the estimated value and the measured value using is shown in FIG. All data used are monthly averages. Also,
The coke ovens used are all large Coppers type ovens with a furnace height of 6500 mm.

As can be seen from FIG. 9, it can be confirmed that the estimation method of the present invention is much more accurate than the conventional estimation method and matches the actually measured values.

Incidentally, the variance of the difference is 0.33
0.30.36, whereas those according to the invention are 0.17-
It is 0.20. As a result of a test for the difference in variance using a statistical method, it was recognized that a significant difference was present.

The drum strength DI according to the method of the present invention is
Is that when a large amount of raw coal other than caking coal such as weathered coal is blended, the effect is remarkably exhibited as compared with the conventional method. Fig. 10 shows the drum strength DI when weathered coal is blended.
Are shown difference from the measured values of ¹⁵⁰ ₁₅ estimates, but towards the estimation method according to the present invention it is found that good obviously accuracy than by the conventional method. The difference is remarkable especially when the blending ratio of the weathered coal is large. All individual data used are daily averages. Each of the coke ovens used is a large Coppers type oven having an oven height of 6500 mm, as described above.

Further, FIG. 11 shows a comparison between the estimation method according to the present invention and the conventional method when low-volatility charged coal is used. Normally, the volatile matter content of charged coal is adjusted to about 28%, but in recent years, it has to be reduced to about 22% in many cases in response to recent raw material circumstances and demands for cost reduction. In such a case, it was performed to confirm whether the estimation method according to the present invention can be applied, but as can be seen from the figure, the method according to the present invention is much more accurate than the conventional method. Drum strength DI ¹⁵⁰
₁₅ can be estimated.

Example 3 FIG. 12 is a correlation diagram showing the relationship between the loss on heating of raw coal and I _{H / C.} Fig. 13 shows the amount of CO generated from raw coal and I
FIG. 4 is a correlation diagram showing a relationship with _{O / C.}

As described in detail above, the present invention uses I _{H / C} and I _{O / C} as indices of raw coal with respect to the coke drum strength DI. This means that the values of I _{H / C} and I _{O / C} of the target plain coal and the charged coal must be measured in advance. These values begin by first separating the sample into active and inactive components, and the resulting active components must be subjected to elemental analysis. Elemental analysis results in H / C and O / C values. When the raw coal is caking coal, these values can be adopted as the values of I _{H / C} and I _{O / C} , but when it is fine / non-caking coal, The obtained values of H / C and O / C must be further corrected to (H / C) 'and (O / C)', which must be I _{H / C} and I _{O / C.}

Such an operation is extremely complicated, and if it is performed every time, a lot of manpower and time must be spent.
Precious methods can be difficult to apply. Therefore, as a result of earnest research, the raw coal I
A method for determining _{H / C} and I _{O / C} has been found.

In this method, the values of I _{H / C} and I _{O / C} are immediately determined from the correlation between the measured values and I _{H / C} or I _{O / C} using other measured values which are practically easy to measure. Is what you want.

In this case, the heating loss was used as the measured value for I _{H / C} , and the CO generation amount was used as the index for I _{O / C.} As already mentioned, these values can be measured relatively easily by, for example, a measuring device combining a thermobalance and a gas chromatograph.

FIG. 12 shows the loss on heating at 800 ° C. and I
FIG. 13 is a correlation diagram showing a relationship between _{H / C} and FIG. 13 is a correlation diagram showing a relationship between the amount of generated CO and I _{O / C.} It can be seen that there is a highly significant correlation in each case.

[0117] Thus, for example, from I when _{H / C} and I _{O / C} is as to obtain a raw coal is unknown, to measure these heating loss and CO emissions, 12 and 13, immediately I _H The values of _{/ C} and I _{O / C} can be determined.

[0118]

The method of the present invention can be applied to a case where a fine or non-coking coal having low or no flowability is blended, and the estimation accuracy is as high as that of the conventional vitrinite. It is much better than an estimation method using a combination of the maximum flow rate MF and the maximum flow rate MF, or a method using H / C and O / C of raw coal as indices.

Therefore, expensive high-flow coal is reduced as much as possible,
Inexpensive fine and non-coking coal can be increased as much as possible, which is extremely advantageous for operation management, budget management, and evaluation when purchasing a new brand of raw coal, and expanding the range of raw coal for coke production. The place that contributes is great.

[Brief description of the drawings]

FIG. 1 is a scatter diagram showing the relationship between H / C and O / C of active and inactive components of plain coal.

2 is a scatter diagram showing the values of DI ^0.99 ₁₅ corresponding to the H / C and O / C of the active ingredient Tan'ajisumi.

FIG. 3 is a correlation diagram showing the relationship between the measured value of DI ¹⁵⁰ _{15 and} the value read from FIG. 2 when caking coal is blended with each other.

4 is a relational diagram showing a comparison between measured values and calculated values of DI ^0.99 ₁₅ of the coal blend in case of the conventional caking coal in coal blend by blending a fine and non viscous coals.

FIG. 5: H / C of active ingredient and apparent H / C of plain coal
FIG. 9 is a correlation diagram showing a relationship with [ie, (H / C) ′].

FIG. 6: O / C of active ingredient and apparent O / C of plain coal
FIG. 7 is a correlation diagram showing a relationship with [ie, (O / C) ′].

FIG. 7 is a relationship diagram showing a range of charged coal used as a raw material of metallurgical coke in a case where I _{H / C} and I _{O / C} of raw coal are considered as indices.

FIG. 8: ΔI and drum strength D by coke oven temperature FT
It is a scatter diagram showing the relationship between I ^0.99 _15.

9 is a over time graph illustrating a comparison between the measured values of the drum strength DI ^0.99 ₁₅ in actual furnace operation and the calculated values.

[10] In actual furnace operation, it is over time graph illustrating a comparison between the measured values of the drum strength DI ^0.99 _{1 5} in the case of firing the blended partially weathered coal and calculated values.

[11] In actual furnace operation, it is over time graph illustrating a comparison between the calculated and measured values of the drum strength DI ^0.99 ₁₅ when firing low volatiles instrumentation Nyusumi.

FIG. 12 is a correlation diagram showing a relationship between a heating loss of raw coal and I _{H / C.}

FIG. 13 is a correlation diagram showing the relationship between the amount of CO generated from raw coal and I _{O / C.}

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-62-119291 (JP, A) JP-A-2-97589 (JP, A) JP-A-57-175257 (JP, A) JP-A-58-1983 160390 (JP, A) JP-A-58-164686 (JP, A) JP-A-59-179582 (JP, A) JP-A-61-145288 (JP, A) JP-A-1-2522693 (JP, A) (58) Field surveyed (Int. Cl. ⁷ , DB name) C10B 57/04 G01N 33/22

Claims (3)

(57) [Claims] 1. A method for producing coke having a predetermined drum strength DI by carbonizing charged coal obtained by blending several types of plain coals including fine and non-coking coals. The relationship between the hydrogen / carbon atomic ratio H / C and the oxygen / carbon atomic ratio O / C of the active ingredients of the plain coal and the drum strength DI of the plain coal is determined in advance. When estimating the drum strength DI of the charged coal by obtaining the hydrogen / carbon atom ratio H / C and the oxygen / carbon atom ratio O / C of the active component of the charged coal by weighted averaging with the ratio being weighted, Regarding the fine and non-caking coal in the plain coal provided for
The apparent hydrogen / carbon atom ratio (the ratio of the number of carbon atoms H / C and the ratio of oxygen / carbon atoms O / C), which is an index corrected so as to show additivity even when blended with caking coal ( H /
C) 'and the apparent oxygen / carbon atom ratio (O / C)', and these apparent hydrogen / carbon atom ratio (H / C) 'and apparent oxygen / carbon atom ratio (O / C)' Indices I _{H / C} , I
The drum strength DI of the charged coal is estimated using _{O / C} , and the type and blending ratio of the plain coal are selected based on the estimation, and the coal is charged, and the charged coal is loaded into a coke oven. A method for producing metallurgical coke, which is introduced and carbonized. 2. The estimation of the drum strength DI is performed by the following relational expression: DI = (DI) _max -F (FT) × (△ I _max- △ I)
² (DI) _max = ab- _{IO / C} + c-FT F (FT) = d-e-FT △ I _max = f + g- _{IO / C-} h-FT where: (DI) _max : sell DI maximum · F of (FT): △ I of △ I _max DI reduction rate, due to deviation from the △ I _max: DI that is maximized △ I-FT: furnace temperature · △ I: I _{H / 2. The} method according to claim 1, wherein the step is performed based on _C - _{IO /} Ca, b, c, d, e, f, g, h: constant. 3. The method according to claim 1, wherein I _{H / C} and I _{O / C} are obtained by reducing the weight loss of the raw coal by heating.
2. The production method according to claim 1, wherein said method is determined from the correlation with the amount of generated CO.