JPH0424404B2 - - Google Patents

Description

[Detailed Description of the Invention] Technical Field The technical content described in this specification regarding a method for estimating the layer thickness distribution of blast furnace charge is to accurately estimate the layer thickness distribution based on the dense correlation with the gas distribution in the furnace. It is related to the development results aimed at stabilizing blast furnace operation and extending the life of the furnace through the management of furnace heat load, and is located in the field of technology to which blast furnace operation control belongs.

Technical background Regarding charge distribution In general, during blast furnace operation, the distribution of the blast furnace charge at the top of the furnace is affected by a complex interplay of various factors, and the main factors are as follows.

1. Physical properties of the charge - density, particle size, coefficient of internal friction, etc. 2. Charge charging speed 3. Charging conditions - coke base, ore/coke, stock line 4. Fall trajectory of the charge flow - movable in the Bell blast furnace Notch position of the armor, inclination angle of distribution chute in bellless blast furnace 5 Charging sequence - charging time schedule 6 Gas flow in the furnace Now, the blast furnace charge is charged into the furnace through the charging device at the top of the furnace. The distribution of blast furnace charge in the furnace is determined under the influence of each of the above factors.

The most important point in reducing the fuel ratio and stabilizing operation is the layer thickness distribution of the blast furnace charge in the radial direction within the furnace (hereinafter simply referred to as charge layer thickness distribution).

Since this charge layer thickness distribution is closely linked to the gas distribution within the furnace, it is extremely important in blast furnace operation, and great efforts have been made to control it.

Regarding gas distribution This gas distribution must be managed to meet the requirements of both operation and equipment regarding the following points.

1. Reducing the fuel ratio by improving the gas utilization rate 2. Stable operation based on smooth unloading 3. Extending the life of the furnace body by managing the heat load on the furnace body By the way, in order to improve the gas utilization rate and obtain stable unloading, it is necessary to On the other hand, in order to reduce the heat load on the furnace body, a gas distribution oriented toward the center flow is required. , the charge layer thickness distribution is controlled.

Control operations for charge layer thickness distribution Now, methods for controlling charge layer thickness distribution include: 1. Change of charge material 2. Coke base, sinter ratio, pellet ratio, etc. 3. Change of charging method 4. Stock line, These include the loading sequence, the movable armor notch in a bell blast furnace, or the distribution chute inclination angle in a bellless blast furnace.

Among these, the most common changes to the charging method are by adjusting the movable arm in a bell blast furnace and by adjusting the distribution chute in a bellless blast furnace. This method attempts to change the charge layer thickness distribution and, as a result, contribute to matching the gas distribution in the furnace to the target.

The gas distribution can be measured using a shaft sonde (used for component analysis of the gas sampled at each position in the radial direction) and a fixed sonde at the furnace throat (used to measure gas temperature at several points in the radial direction). Common.

However, the quantitative relationship between the control amount of the movable armor and distribution chute on the gas distribution obtained by these sonde measurements is still not fully understood, and it is necessary to analyze the gas distribution based on the above gas distribution information. The charge layer thickness distribution is controlled by trial and error, in which the movable armor or distribution chute is changed and further changes are considered depending on the subsequent change trend in the gas distribution. Even if such a charging method is changed, the expected gas distribution does not easily match, and the current situation is that long-term adjustments are required to obtain the target gas distribution.

Of course, in order to quantitatively understand the relationship between the charge layer thickness distribution and gas distribution and to improve charge control technology, it is necessary to more accurately understand the situation of the charge inside the furnace. Various types of furnace top profile meters have been developed as sensors for measuring the surface profile of the charge, especially among the conditions of the charge in the furnace.

These sensors are broadly classified into contact type (mechanical type) and non-contact type (μ wave type, laser type, etc.). The problems will be explained using as a representative example, but the problems are common to all of the above types.

Regarding the example of sounding, Fig. 1 shows a sounding-type furnace top profile meter s using a measuring weight. Devices, 3 is a measuring weight, 4 is a wire rope device for raising and lowering the measuring weight, 5 and 5' are guide sheaves, 6 is a lance 1
A movable pulley with a wire rope device 4 fixed inside, 7
is a fixed pulley attached to the rear end of lance 1, 9 is an operating rope wrapped around both pulleys, 10 is its fixed end,
11 is a winding barrel of the operating rope, 12 is a guide sheep, 13 is a tension detection lever of the operating rope 9,
14 is a counter spring.

By rotating the chain mechanism 2, the lance 1 is advanced from the opening 15' in the top furnace wall 15 of the blast furnace toward the furnace core through a seal (not shown for simplicity), and the movable pulley 6 advances together with the lance 1. 1 can be delivered in any radial position. At this time, the measuring weight 3 is slightly lifted.

At this position, for example, the solid line position in the figure, the winding drum 11
When the operating rope 9 is loosened and the measuring weight 3 is lowered, the measuring weight 3 reaches the surface of the charge 16, and the tension of the operating rope 9 decreases rapidly, and the tension is increased by the counter spring 14. Since the detection lever 13 rotates, landing can be detected.

Therefore, at this time, the progress of lance 1 in the radial direction in the furnace x
The measurement point can be displayed in coordinates by the depth y of the measuring weight 3.

After this detection of landing, the lance 1 is further advanced at predetermined distances as shown by the broken line, and the next depth is measured along with the progress. For example, 10 points are obtained for each cycle of each batch.

Each measurement data consists of the measurement time of the j-th point in the charging batch, the radial advance and depth of each measurement point, and this is exactly the same even in the case of a non-contact furnace top profile meter.

About the deposition behavior of the charge By the way, the layer thickness of the charge is determined by the distribution of the ratio of the layer thickness of ore and coke formed as a result of charging the charge into the furnace, the particle size distribution, and the difference between ore and coke. follows a mixed layer distribution.

As shown in FIG. 2, for example, in a bell-type blast furnace, raw material 16' falls into the furnace from a large bell 17.
The particles are deposited at an angle of 16'' according to the inclination angle determined by the characteristic values (particle size, shape, etc.) of the raw material.

After this deposition progresses and the amount of deposition reaches the deposition limit, the deposit 16'' flows toward the center of the furnace, and at the end of charging, the charging profile of the charge 16 as shown in the phantom line. is formed.

Thereafter, redistribution of the charge 16 may occur due to changes in the gas flow during the entire descent of the charge 16, such as inflow due to fluctuations in the stable inclination angle of the charge 16. .

Prior art and its problems Conventionally, the charge layer thickness distribution is controlled by operating a movable armor (hereinafter abbreviated as MA) 18 back and forth to change the falling position of the raw material 16' into the furnace. This is done by changing the distribution state, mainly the radial distribution of the ratio of ore to coke (hereinafter referred to as O/C).

Therefore, in order to control the charge distribution with MA 18, it is necessary to know the layer thickness before redistribution, ie at the time of charging, and the resulting O/C distribution.

This charge layer thickness distribution has conventionally been determined by calculation using the above-mentioned furnace top profile meter according to the following procedure.

(1) As shown in Figure 3a, for each batch (referred to as the n-th batch) of raw material 16' from the large bell 17, the top profile meter measures the charging end time T. The surface profile of the charge P _{o, 1} at o,1 and then the (n+1)th
The surface profile P _{o ,2 of the charge at time T o,} 2 immediately before the second batch charging is measured.

Then, in the radial direction of the furnace, the average rate of descent V _o over time from time T _o,1 to T _o,2
Calculate.

(2) As shown in Figure 3b, the surface profile at the (n+1)th batch charging end time T _(o,1),1
P _(o+1),1 is measured with a furnace top profile meter, and at the same time, at time T _(o+1),1, from time T _o,1 of the previous nth batch to T _(o+1),1 A sedimentation profile is obtained by extrapolating the surface profile PS _o,1 that has descended over time using the above average descent velocity V _o .

(3) The measured surface profile P _(o,1),1 of the (n+1)th batch at time T _(o+1),1 and the estimated profile PS _o+1 of the n-th batch, that is, the (n+
1) Calculate the layer thickness of the (n+1)th batch from the lower surface profile of the second batch.

In this way, the layer thickness of the (n+1)th batch is calculated using the furnace top profile meter three times (2 times for the nth batch, once for the (n+1)th batch). This method of estimating O/C has the following problems.

Pointing out the problem According to experience in actual operations, the burden surface profile obtained by measurement using a furnace top profile meter has a large variation, and therefore the layer thickness distribution estimated by this method also has a large variation, making it difficult to The estimation accuracy of layer thickness distribution is poor.

Therefore, in order to use the charge layer thickness distribution estimated by such a method as a guideline for controlling the charge distribution, it is necessary to improve the accuracy of this estimation.

To do this, considering the above-mentioned variations, it would be necessary to perform measurements on several dozen charges, but this is currently not practical considering the durability of the profile meter and the cost required for the measurement. It is difficult to do so.

Fundamentals of Idea and Purpose of Invention Regarding the above-mentioned problems, the inventors have come up with the idea that the number of measurements made by the furnace top profile meter can be reduced by making effective and advanced use of index finger information as described below. It is an object of the present invention to improve the estimation accuracy of thickness distribution.

The index finger is the depth obtained by almost continuously detecting the descent condition at one point on the surface of the furnace contents 16 near the top furnace wall 15 of the blast furnace, separately from the furnace top profile meter S. It is.

Structure of the Invention The above object is achieved through a procedure consisting mainly of the following matters.

Calculate the charge calculation depth YC _o,i corresponding to the depth of the i-th point of any n-th batch using the top profile meter of the blast furnace charge using the following formula, and repeat this calculation sequentially. A method for estimating the layer thickness distribution of a blast furnace charge, the method comprising: determining the surface profile of the charge and the bottom profile of the charge, and estimating the average layer thickness distribution for each charge. .

YC _o,i = (ax _o,i ³ +bx _o,i ³ +cx _o,i +d) +Z _o +{1-k
(1-R _p -x _o,i /R _s )}・ΔS _o,i a, b, c, d: constant, x _o,i : radial coordinate in the furnace, Z _o : determined for each charging sliding amount, k: descent rate distribution constant, R _p : furnace mouth radius, R _s : radial distance from furnace wall to furnace top profile meter measurement point,
ΔS _o,i : Amount of index finger drop in single-point sounding from the time of charging to the time of measurement Specific explanation of the invention As already mentioned, in order to accurately estimate the layer thickness distribution of the charge, it is necessary to determine the surface profile of the charge. In addition to this, a correct detection of the descent velocity distribution is required for estimating the underside profile of the charge.

When estimating the descending velocity distribution for the surface profile of each batch, we focused on using index information that constantly measures the descending state of the charge for each batch.

First, the depth at a single point as index finger information is determined by using the winding and unwinding device to place the measurement weight 20 attached to the tip of the wire 19 that moves up and down on the surface of the charging material 16, as shown in FIG. This can be obtained by following the descent of the object 16 and detecting its situation.

Here, when the raw material 16' on the large bell 17 is charged into the furnace, detection can be continued at all times, except for retracting the measuring weight 20 from buried, so the furnace top profile meter is used as the basis of the blast furnace raw material charging schedule. The measurement data is much more abundant than that of the conventional method.

In FIG. 2, 21 is a winding and unwinding drum;
22 is a counterweight, and 23 is a drive device with a reduction gear and a brake.

Concerning the descending velocity distribution, we investigated the relationship between the index descending amount and the descending amount at each point in the radial direction obtained by the furnace top profile meter, and found that the descending velocity distribution of the charge was determined by the difference using the radial distance of the blast furnace as a parameter. It was confirmed that it can be treated as a linear approximation of the finger's descent speed.

Therefore, we investigated a charge distribution model using index finger information from single-point sounding and measurement data from a profile meter. The charge distribution model is created based on the actual measurement data of the charge surface profile using a profile meter and the index finger data of the measurement batches.

This model follows the following assumptions.

(1) First, we consider that there is a standard profile that is uniquely determined by the type of charge (ore or coke) and its MA position, with a bell-type blast furnace as a typical example. is assumed to be that this standard profile is sequentially slid downward in the depth direction by an amount determined for each charge (hereinafter, this amount is referred to as the sliding amount Z _o
).

The calculated depth YC o,i of the charge corresponding to the i-th point of the top profile meter s for the nth batch of charge is the depth YS _{o,i obtained} from the standard profile, the sliding amount Z _o and Using the time-corrected descending speed distribution ΔY _o,i from the time of charging to the time of measurement, it is expressed as the following equation.

YC _o,i = YS _o,i +Z _o +ΔY _o,i ……(1) (2) Next, the descending velocity distribution of the charge is calculated by dividing the distance x _o,i from the furnace wall into 1 Approximate with the following formula.

ΔY _o,i = {1−k(1−R _p −x _o,i /R _s )}ΔS _o,i ……
(2) k: descending rate distribution constant, R _p : radius of the furnace mouth, x _o,i : radial coordinate in the furnace, R _s : radial distance from the furnace wall to the measurement point of the furnace top profile meter,
ΔS _o,i : Difference drop of single-point sounding from the time of charging to the time of measurement Based on these two assumptions, calculate the depth of the charge corresponding to each measurement time of the furnace top profile meter, and Once the standard profile and descending velocity distribution constant are obtained so that the calculated depth best matches the actual value measured by the profile meter, the layer thickness distribution of the charge can be calculated using only index information from constant single-point sounding. You will be able to do that.

Here, first, the standard profile YS _o,i determined by the type of charge and the MA position is approximated by a cubic equation of the distance x _o,i from the furnace wall, for example, as shown in the following equation.

However, x: the furnace wall is taken as 0, the distance from it in the direction of the furnace core is positive, and when Yx=0, Y=0, and the distance vertically downward from it is taken as positive.

YS _o,i =ax _o,i ³ +bx _o,i ² +cx _o,i + ...(3) a, b, c, d: Substituting constants (2) and (3) into equation (1) , The calculated depth YC _o,i of the charge corresponding to the i-th point of the n-th batch is as follows.

YC _o,i = (ax _o,i ³ +bx _o,i ² +cx _o,i +d) +Z _o +{1−k
(1-R _p -x _o,i /R _s )}・ΔS _o,i ...(4) By the way, y of the actual measurement point of the i-th furnace top profile meter s of the n-th batch of charging When the coordinate is YR _o,i , the sum of squares E of the error from the above calculated value YC _o,i is expressed by the following formula.

E= 〓 ⁿⁱ (yR _o,i −YC _o,i ) ² ...(5) Constant terms a, b, c, d of the standard profile, slide amount Z so that the sum of squares E of this error is minimized. _o
(n=1 to n), and determine the descent rate constant k.

After determining the constant term in this way, the layer thickness distribution is estimated using the depth according to the standard profile, YS _o,i =ax _o,i ³ +bx _o,i ² +cx _o,i +d and the descent rate distribution constant k.

In order to calculate the layer thickness of the charge, it is necessary to estimate the profile of the top and bottom surfaces of the charge, and by the following procedure, when charging the nth batch, T =
Figure 4 shows the case of estimating the layer thickness distribution of _Tso .
This will be explained according to FIG.

(A) Estimation of the bottom profile of the charge of the n-th batch (1) First, for the (n-1)th batch, which is the previous batch of the n-th batch, at time T as shown in Fig. 4(a), Depth YC _{(o-1 ),i given by the profile of the charge of the (n-1)th batch in s(o-1)} according to its standard profile.
YS _(o-1),i and sliding amount Z determined for each charging
Calculate from.

(2) As shown in Figure 5, the index finger of the n-th batch between the time of charging the (n-1)th batch and the time of charging the n-th batch, that is, the time
The amount of sounding drop ΔS over time from T _s(o-1) to T _so is determined as the index finger.

(3) Calculate the amount of descent at each point using the descent speed distribution constant k and the sounding descent amount ΔS.

(4) Estimate the descent profile of the (n _{-1)th batch, which is the profile of the bottom surface of the charge of the nth batch, from the depth YS (o-1),i} according to the standard profile and the amount of descent at each point. (Figure 4b).

(B) Estimating the surface profile of the n-th batch of charge The profile of the n-th batch at time T _so is determined by one of the following methods using the depth YS _o,i according to its standard profile.

[1] Use the sliding amount at the time of charging (Figure 6a).

[2] Using the descent profile of the (n-1)th batch and the standard profile of the nth batch, calculate the sliding amount back so that the calculated bed volume is equal to the volume of the charge (actual value). (Figure 6b).

By calculating (A) and (B) for each batch, the average layer thickness of coke and ore is calculated.

EXAMPLE FIG. 7 shows the results of the average layer thickness distribution according to the present invention obtained through experiments conducted in a blast furnace having a capacity of 6000 t/d.

The layer thickness of ore is maximum at the furnace wall and decreases toward the furnace core, while the opposite trend was observed for coke. It can be seen that the thickness estimation method is appropriate.

Next, measurement No. 1 (solid line) to measurement No. 2 (dashed line) corresponds to when an action is applied to push the movable armor during ore charging, aiming for central flow gas distribution. .

This shows that the average layer thickness of ore increases at the periphery and decreases at the core.

Effects of the Invention As described above, through the effective use of index information according to the present invention, as shown in the above results, the estimation accuracy of the layer thickness distribution is good, which advantageously contributes to the stable operation of the blast furnace and the extension of the life of the furnace body. be able to.

[Brief explanation of drawings]

Figure 1 is an explanatory diagram of the operation of the furnace top profile meter, Figure 2 is an explanatory diagram of the measurement of the index finger of the charge, and Figures 3 a and b are the estimation of the charge layer thickness distribution using the furnace top profile meter. 4a and 4b are explanatory diagrams of the procedure for estimating the lower surface profile of the charge for the n-th batch, and FIG. 5 is an explanatory diagram for the procedure for measuring the index drop of the n-th batch. FIGS. 6a and 6b are explanatory diagrams for estimating the surface profile of the n-th batch charge, and FIG. 7 is a graph showing a specific example of the layer thickness distribution of ore and coke.

Claims (1)

[Scope of Claims] 1 Calculate the charge calculation depth YC _o,i corresponding to the depth of the i-th point of any n-th batch using the top profile meter of the blast furnace charge using the following formula, Blast furnace charging characterized in that by sequentially repeating this calculation, the surface profile of the charge and the bottom surface profile of the charge are determined, and the average layer thickness distribution for each charge is estimated. Method for estimating layer thickness distribution. YC _o,i = (ax _o,i ³ +bx _o,i ² +cx _o,i +d) +Z _o +{1-k
(1-R _p -x _o,i /R _s )}・ΔS _o,i a, b, c, d: constant x _o,i : radial coordinate in the furnace Z _o : sliding amount determined for each charging k: Drop rate distribution constant R _p : Furnace mouth radius R _s : Radial distance from the furnace wall to the furnace top profile meter measurement point ΔS _o,i : Difference drop in single point sounding from charging to measurement