CA1164096A - Process for estimating particle size segregation of burden layer in blast furnace top - Google Patents

Abstract

Abstract of the Disclosure A process for estimating the particle size segregation in a burden layer at the top of a blast furnace is disclosed, which comprises measuring the particle size distribution of a burden material before charging to the furnace and a layer thickness distribution of the burden layer after charging, and estimating the particle size distribution at every position in the burden layer charged at the furnace top on the basis of the above measured values, charging conditions and furnace operating conditions according to a simulation model of particle size segregation.

Description

~164C~96 This invention relates to a process for estimating the state of the particle size segregation in burden layer at a top portion of a blast furnace. More particularly, this invention relates to a process for estimating a particle size distribution of a burden layer charged in the top portion of the blast furnace at each positior toward the radial direction of the furnace throat. The estimation is based on the particle size distribution of the burden material before charging, the charging conditions, and the furnace operating conditions accord-ing to a particle size segregation model.
In order to achieve the reduction of fuel rate andthe stabilization of b1ast furnace operation, it is important to optimize the radial distribution of gas flow in a blast furnace by controlling the burden distribution in the furnace top portion. The term "burden distribution in the furnace top portion" used herein mainly means a layer thickness distri-bution for an ore layer and a coke layer, together with the particle size distribution in each layer. In general, the gas flow in the furnace is distributed according to the radial distribution of gas flow resistance of the burden layer, which is determined from the layer thickness distribution and particle size distribution, so that it is necessary to know both of these distributions. In this connection, there are many measure-ments for the layer thickness distribution, but no methods estimating the particle size distribution based on the measured particle size ranges have been developed.

il640~6 In general, the burden distribution at the top of the blast furnace are influenced by various interconnected factors. The main factors are as follows:
1) Physical properties of the burden material such as density, particle size, coefficient of internal friction and so on;

2) Charging speed;

3) Charging conditions such as coke base, ore/coke ratio (hereinafter referred to as O/C), stock line level and so on;

4) Falling trajectory of burden flow, which fundament-ally depends on a notch position of a movable armor in bell-type blast furnace charging gear or the tilting angle of a distributing chute in a bell-less blast furnace charging gear;

5) Charging sequence; and

6) Gas flow rate in the furnace.
Further the geometrical arrangement between the throat of the furnace and the charging equipment is considered to be one of the fundamental factors in the formation of burden distribution, but it is not an operational factor in any given blast furnace. Therefore, when the burden is charged into a specified blast furnace through the available charging equip-ment, the burden distribution is determined by the influence of the above mentioned factors. Particularly, layer thickness distribution and particle size distribution of the burden in 1164~

the radial direction of the furnace are significant in order to achieve a reduction of fuel rate and a stabilization of furnace operation.
In the conventional operation of blast furnaces, the procedure for controlling the burden distribution is based on the control of the layer thickness distribution and lies in optimizing the radial distribu-tion of the thickness ratio of ore layer to coke layer (Lo/LC) or of 0/C, or by means of a product of this ratio with a bulk density ratio (Po/pc)-For instance, it is experientially known that when the horizont-ally sectional area of the throat in the blast furnace is equally divided into a central part (C), a middle part (M) and a peri-pheral part (P), if the relation of the layer thickness ratio (Lo/LC) in these parts is given by the following equation (1):

o/ c)M ~ (L0/Lc)p > (Lo/L )C . . . . . (1), then stable operation with a low fuel rate can be achieved.
However, the control of burden distribution aims at optimizing the radial distribution of gas flow resistance of the burden layer and of the radial gas flow distribution accompanied there-with. For this purpose, there must be known the particle sizedistribution of burden material at each position in the radial direction of the furnace in addition to the above layer thickness distribution. The thickness of the burden layer can be measured directly or indirectly. The techniques of direct measurement are based on the use of an electrode or a magnetic sensor.
The indirect method is based on the procedure of determining :1~64~96 the layer thickness from the difference of the burden surface level measured before and after charging the said burden materials by rneans of a transversely movable sounding device or microwave dev:ice or a layer-measuring system. On the other hand, a method of measurement of particle size distribution is not established at all because the quantity required for exactly determining the particle size distribution of the burden cannot be sampled from the inclined burden surface at given local positions in the radial direction within the operating furnace. In order to optimize the gas flow distribution in the blast furnace, it is essential and i.mportant to know the particle size dis-tribution of the burden at given positions in the radial direc-tion of the furnace.
According to this invention, there is provided a process for estirnating the particle size segregation in a burden layer stacked at a top portion of a blast furnace, which com-prises measuring the particle size distribution of a burden material before charging to the furnace and a layer thickness distribution of said burden layer after charging to the furnace, and estimating a particle size distribution at every position in said burden layer charged in the furnace top on the basis of said measured values, charging conditions and furnace oper-ating conditions according to a simulation model of particle size segregation given by the following equation:
log{Xn/(1-Xn)} = - ~-Q +log{Xn/1-X)}
wherein Xn is a cumulative weight fraction o:E particles having i~64~

smaller size than n-th sieve opening, a is a size sec1regation constant and Q is a distance from a collision point of the main falling trajectory against burden surface to the flowing direction, that is, to center and to the wall. The suffix 'o' means the value of Xn at Q=o.
The invention will now be described in detail with reference to the accompanying drawings, wherein:
Figure 1 is a diagrammatical view illustrating a particle size distribution in a burden layer stacked at a top portion of a blast furnace;
Figure 2 is a diagram illustrating measured values for an ore layer thickness;
Figure 3 is a graph showing the relation between log{Xn/(1-Xn)} and the distance from the furnace center, or the distance from the collision point of the main falling tra-jectory against the burden surface to the flowing direction; and Figure 4 is a graph showing the relation between the gas flow rate and the size segregation constant.
In Figure 1 is schematically shown a typical particle size segregation in a burden layer stacked at the top portion of a blast furnace. A burden flow 2 discharged from a charging equipment 1 falls in a space bordered with an upper side 3 and a lower side 4 of a falling trajectory and comes into colli-sion with a previously charged burden S to form a layer thereon.
In this case, when the profile of burden distribution as shown in Figure 1, the burden flow is divided at the position of ~16~

peak 6 appearing in the burden distribution into a stream directed to the center of the furnace and a stream directed to the wall of the furnace. As the charged burden layer thickens, the position of peak 6 is shifted upward along a main falling trajectory

7 of the burden flow as shown in Figure 1. The main falling trajectory 7 is regarded as the curve passing through the points inside the burden flow 2, at which the cumulative weight fraction of burden materials integrated in a certain horizontal plane from the upper side of the falling burden flow toward the lower side reaches 50%. When each of the two streams directing to the furnace center and furnace wall flows with a certain layer thickness, a void between large-size particles plays the same role as a sieve opening in a sieving operation. Under the influence of such voids, small-size particles in the burden material are percolated into a lower portion having a small flow rate and then left in a position near the falling point, while large-size particles go on rolling downward toward the furnace center. As a result, the particle size in case of profile shown in Figure 1 is a maximum at the central part of the furnace, and becomes smaller toward the furnace wall, and is minimum near the collision portion of the burden flow against the previously charged burden. When the profile of burden distribution is V-shape, there is obtained such a particle size segregation that the particle size gradually increases in a direction of from the furnace wall to the furnace center.
~ow, such a phenomenon of particle size segregation 1~640~?6 in the radial direction of the furnace may be simulated by an equation as expressed below. When a horizontal distance from the position of a peak, or the collision point (R*) of the main falling trajectory against the burden surface to an optional downstream point is Q (m) and the cumulative weight fraction of particles having a smaller size than n-th sieve opening is Xn, if the burden stream flows from Q to Q +dQ , a percolation rate of particles having the above mentioned particle size (-dXn/dQ ) has been found to be given by the following equation (2):
dXn/d = a Xn (1-Xn) . . . . . (2) That is, the equation (2) means that the percolation rate of fine particles is in proportion to not only the weight fraction of fine particles, but also a weight fraction of coarse particles acting as a sieve in the percolation. In this equation, a is a constant indicating a degree of particle size segregation in the flowing direction of the burden, which is called a size segregation constant. The value of a depends upon the pro-perties of the burden material, charging speed and gas flow velocity in the furnace and the like.
The integration of equation (2) gives the following equation (3):
log{ Xn/(1-Xn)} = _a Q +log{ Xn/1-Xn)} . . . . (3) In the equation (3), the second term on the right-hand side means the value of {Xn/(1-Xn)} at Q=0. That is, the equation (3) is a simulation model of particle size segregation for a particle size distribution of the burden layer charged at every position of the furnace top toward the radial direction of 1he furnace.
In order to estimate Xn (i.e. cumulative weight frac-tion of particles having smaller size than n-th sieve opening) at every position in the radial direction, the value of the second term on the right hand side of the equation (3) must first be determined, which may be given as follows. The averaged value of the cumulative weight fraction of particles having a particle size smaller than an n-th sieve opening, which are distributed radially from the furnace center to the furnace wall, should be equal to a value Xn of the burden material before the charging. Assuming that the bulk density of the burden layer is constant at each position, Xn is strictly given by the following equation (4):

~ R-R* X e ) O l-X(l-e~~Q'~ h(R +Q')-(R*+Q')dQ' rR* X e )O l_x(l_e-aQ) h(R -Q)-(R*-Q)dQ

= xn-r ROr h(r)dr . . (4), wherein ~ is a size segregation constant at r=o-R*, ~ is a size segregation constant at r=R*~R, r is a distance from the furnace center, h(r) is a function indicating the layer thickness dis-tribution and requires a found value, R is the radius of the `~64~

furnace throat, and R* is a radial position from the furnacecent:er at Q=0 and corresponds to a collision point of the main falling trajectory against the previously charged burden.
In order to obtain the value of R*, it is necessary to measure the profile of burden distribution.
The equation (4) means that an average value derived from the integration of the equation (3) between the furnace center and the furnace wall is equal to the value before the charging. Therefore, the particle size distribution at Q=0, i.e. the value of the second term on the right-hand side of the equation (3) is calculated from the equation (4) considering the found values for the particle size distribution Xn before the charging and the layer thickness distribution h(r) as well as the position R* of peak of the burden distribution profile, so that the particle size distribution at an optional distance Q
can be arithmetically estimated by the equation (3).
As apparent from the equation (4), the value of Xn cannot be calculated exactly. Now, by using the assumed Xn, the integration on the right hand side of the equation (4) is first performed and then the value of Xn satisfying the equation (4) must be determined by a trial and error method, which can easily be performed by means of an electronic computer.
The equation (4) gives a strict value of Xn, but if this value is accepted to have an error of few percents, Xn can be estimated by the following equation (5):

~164~6 Xn/ ( 1 ~Xn ) Xn/ ( 1 ~Xn ) ~ROr-h(r)dr tRo R (R*~Q')~h(R*+Ql) e dQ' - rRo(R*-Q)~h(R*-Q)e dQ

By using the equation (S), the calculation can somewhat be simpli-fied because it is not necessary to perform the trial and error method as in the equation (4).
In the actual operation, the particle size segregation constant of the equation (3) must first be determined. In this case, the burden material in an actual or laboratory furnace are sampled at two positions spaced only by a distance ~Q(m) in the radial direction of the burden level in the furnace.
Then, the particle size analysis for the two samples is performed to determine a difference ~log{(Xn/(1-Xn)} between two positions, from which a is calculated according to the equation (6) as follows:

~log{Xn/(1-Xn)} . . . . . . . . . (6) Moreover, when the sampling of the burden material is carried out at three or more positions, ~ and log{Xn/(1-Xn)} are calculated by the least squares method using the equation (3).
Using this technique a comparison was made between the found value and the estimated value for particle size distribution in burden layer at every position toward radial 1~6~

direction according to the process of the invention to obtain a re!sult as shown in the following Table 1.
In the blast furnace with the throat radius of 5.25 m, the boundary between the ore layer and the coke layer, and the surface of the ore layer were measured by means of a layer thickness measuring device utilizing electrodes. Both radial profiles are shown in Figure 2. Further, the ore layer thickness h(r) was obtained from the difference of both the levels of radial profile as shown in Figure 2.
The particle size analysis was made with respect to four samples of the ore layer, each of which being sampled at a distance of 2.0, 3.0, 4.0 or 4.62 m from the furnace center, to obtain a result as shown in the column "Found value" of Table 1. From these found values is obtained log{Xn/(1-Xn)}, which is plotted in Figure 3 with respect to the radial position.
As a result, R* is 4 m, a is 0.314 (1/m) on the average and is 0.65 (1/m). In this case, the reason why the average value of 0.314 is selected as a value is due to the fact that the a value is 0.310, 0.314, 0.308, 0.317 and 0.321 for Xl, X2, X3, X4 and X5, respectively, which means that these a values are not substantially dependent upon the particle size.
Further, the particle size distribution of the ore before the charging, Xn(%), is shown in the last right-hand column of Table 1. On the other hand, the particle size dis-tribution at a distance of 2.0, 3.0, 4.0 or 4.62 m from the furnace center is estimated according to the equations (3) 1J~64~96 and (5) using the above mentioned values and also shown in the column "Estimated value'l in Table 1.
As is apparent from Table 1, the estimated value is in good agreement with the found value, which shows that the particle size distribution in the burden layer at every position in the radial direction is adequately estimated by the process according to the invention.
Although this example deals with the case where the value of ~ is not dependent upon the particle size and is sub-stantially constant, the invention is also equally applicablewhen the ~ value varies with the particle size.
According to the invention, the operation of determin-ing the size segregation constant by sampling the burden material may be omitted by measuring beforehand a relationship between the size segregation constant and each factor influencing thereupon.
For instance, a relation between the size segregation constant and the gas flow velocity in a bell-less blast furnace top is shown in Figure 4, wherein the charging speed of the burden material is 0.7 m3/sec. As is apparent from Figure 4, the flow rate of the burden material onto the old burden surface toward the furnace center or wall becomes higher with the increase in gas flow velocity, so that the degree of size segregation becomes smaller. Thus, it is sufficient to measure this relation-ship beforehand in the blast furnaces being controlled under various conditions.
In this way, the invention first makes it possible 4Q~6 not only to estimate a particle size segregation state oE a burden layer in the top portion of the blast furnace, but also to quantitatively examine a charging method for optimizing the burden distribution inclusive of layer thickness distribution and particle size distribution. In the latter case, the burden distribution can be controlled to maintain an optimum state, so that the reduction of fuel rate and the stabilization of furnace operation can effectively be achieved in the blast furnace.
Moreover, the fundamental physical phenomenon of the invention consists in the particle size segregation of the burden layer toward the flowing direction on the inclined burden surface. Similarly, such a phenomenon occurs in the supply of particulate matters, granules or the like into a storing apparatus, reaction vessel or the like. In the iron-making process, there are (a) particle size segregation in the layer thickness direction during the supply of raw material onto a pallet for sintering, (b) particle size segregation in radial direction of a banker for raw sintering material or an ore stock yard, and the like. The estimation technique according to this invention also can be applied to such particle size segregation phenomena.

. .
., ~64~6 _ ~o ~
C h _~ ~ ~) C~ ~ ~ ~ ~ ~ .
U~rJ a)r~ ~ O O O C5 rC ~ h C,) ~ c~ ~ c~~1 ~0 CJ O
.,~
~ 4 l 1~
O O ~ I~
u~ ~ i~ ) ~ ~i ~ O C`' e ~ ~ ~ ~
O , ,~ .~ I ~ ~ a~
CO O C~
s~ I o ~U~ ~ ~ ~ ~ O ~ O O
x~
~ ~ ' N t~ I
~ ~-~ I ~ ~r~ u~ ~ ~ ~ ~ ~ u~
U~ ~ I VV~. . . . . . . . .
tO I U~ 00 ~ ~ ~1~ ~ ~-I O O O
C~ ~1 O . .. _ J_~ I CY~ ~ QJ
h h I - C: 3 ~ ~ ~ O ~ 1 CN
r~ O O ~ O
,.C O I
~ ~O 0~ .
a) I I ~ a) ~d V I ~ ~ r~ oo 1~ ~ ~ u~u~ r_ 4-1 ~1) 1 J~ 1 . . . . . . . .
O 1:: 1 U~ '01~') ~' 1~) 0 0 ~ O O O ~O
~, a ~ a I
O ~ ' O ...
v a l`
tl5 0~ ~ ~ ~ ~ ~`J ~C~lCOU~ ~ ,_1 O
h ~ I O td 1~00 11~CO1~ ~ O O O 00 ~1 l ~ ~' a) I
V ~ .
~ ~ ~ I ~ ~
. v ~ 'v'l v ~ ~I 1` `J O O il r_ co - c~
3 ~1 1 U~ ~ tl) C~ ~ ~ O 0 00 0 0 0 O
~ ~ ~ I ~_~
I

oo u~
. I ~ ~ C~ O O O
I _ _ I
_ . ~ Lr) U~ o o o ' O
........
C~ Lr1 r~ o ~ o u~
,, ~ a , , , , , , , , , ~ ~
v ,l ~ u~ o o o ~
~ ~n ~ . . . .. . . ~ ~ N
I ~a o u~ I~ ~c~oo ~ ~ o I ~~ <1 C~ U~

~ -- ~

Claims

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A process for estimating the particle size segregation in a burden layer stacked at a top portion of a blast furnace, which comprises measuring the particle size distribution of a burden material before charging to the furnace and a layer thickness distribution of said burden layer after charging to the furnace, and estimating a particle size distribution at every position in said burden layer charged in the furnace top on the basis of said measured values, charging conditions and furnace operating conditions according to a simulation model of particle size segregation given by the following equation:
log{Xn/(1-Xn)} = -.alpha.? ?+log{Xn°/1-X°n)}
wherein Xn is a cumulative weight fraction of particles having smaller size than n-th sieve opening, .alpha. is a size segregation constant and ? is a distance from a collision point of the main falling trajectory against burden surface to the flowing direction.