JPS63210218A - Prediction of furnace heat in blast furnace - Google Patents

Abstract

PURPOSE:To accurately predict furnace heat, especially lowering of the furnace heat in a blast furnace without any delay, by adopting ARMA model introducing a variable correcting difference of iron tapping holes into a MA term only in case the temp. deviation by different tapping holes is judged to be the prescribed value or higher. CONSTITUTION:At the time of predicting the variation with time of the furnace heat level in the blast furnace by the ARMA model using the operational result in the blast furnace, in case the temp. deviation by different tapping holes during the prescribed time exceeds the beforehand fixed value, the variable correcting the difference of the tapping holes is introduced into the MA term, to predict the furnace heat. Then, the above ARMA model is desirable to introduce the variable related to the temp. difference at the inner wall measured in every the prescribed time intervals by inner wall thermometers set at the prescribed positions in the blast furnace and the max. value of solution loss C quantity and the min. value of N content in the top furnace gas component during the prescribed time into the MA term.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to a blast furnace furnace heat prediction method for stable operation of a blast furnace.

(Prior art and its problems) It has been known for a long time that in order to maintain stable operation of a blast furnace, it is necessary to keep the temperature of hot metal constant. For this reason, blast furnace operators have always needed to predict changes in blast furnace heat.

Prediction of temperature drop is extremely important in blast furnace furnace thermal changes, especially since hot metal may solidify due to temperature drop and may no longer flow out of the blast furnace.

A method for predicting blast furnace heat is to consider the blast furnace as a black box and obtain a model through statistical analysis.
There is a method based on the AR (Auto-Regressive) model.

Although the AR model described above has excellent characteristics as a model for predicting temporal fluctuations in the actual furnace heat level of a blast furnace, it has the following problems.

FIG. 16 is a diagram showing changes over time in the actual value and predicted value of hot metal temperature, and in the same figure, 11 is a curve based on the actual value x 112 is the predicted value. As shown in the figure, the forecast 11t! The change in I 1Ffi is determined by the τ sampling time (i.e., the sampling interval is Δt and τXΔ
There is a tendency for the amplitude to be small, with a delay of some time (time). For this reason, there was a problem in that the prediction function could not be fully fulfilled, and the accuracy of the predicted value was also not sufficiently accurate.

In order to solve this problem, ARM A (Auto-
Realistic Moving Average
) (autoregressive moving average) model is published in JP-A-60-248804゜1 Tetsu-to-Hagane J 1984, No. 1,
Although it is disclosed on page 54, etc., the current situation is that no definitive variable to be introduced into the MA term has been found yet.

(Objective of the Invention) The object of the present invention is to provide a blast furnace furnace heat prediction method that eliminates the problems of the above-mentioned prior art and can accurately predict hot metal temperature changes without delay in prediction. .

(Means for Achieving the Object) In order to achieve the above object, the blast furnace furnace heat prediction method according to the present invention uses the blast furnace operation results to predict temporal fluctuations in the blast furnace furnace heat level using the ARMA model. When the temperature deviation due to the difference in the taphole within a predetermined period exceeds a predetermined value, a variable for correcting the taphole is introduced into the MA term.

(Example) 1. Introduced into MA term. The following is considered to be the first reason for the decrease in furnace heat in the variable blast furnace.

High-temperature air (gas flow) blowing up from the blast furnace lining to adjust the hot metal temperature and hot metal m is usually blown into the center of the furnace.
I'm here. However, due to reasons such as raw material charging conditions and charge distribution, a large amount of gas may suddenly flow to the periphery of the furnace. As a result, the endothermic reaction of Fe O+C-+ Fe +GO is promoted and the furnace heat decreases.

By the way, when a large amount of gas flows around the furnace, Na
, when the deposits in the furnace such as Pb and the stagnation layer peel off and fall off the wall, the temperature of the furnace wall in that part increases rapidly. After that, the furnace heat will decrease sooner or later.

Moreover, the following can be considered as the second reason for the decrease in furnace heat of the blast furnace.

When the unloading speed in the blast furnace increases due to raw material charging conditions, charge distribution, etc., the level of the cohesive zone in the blast furnace decreases due to so-called raw ore unloading, resulting in a decrease in furnace heat.

By the way, when the cohesive zone level decreases, the furnace wall temperature at the corresponding portion also decreases rapidly. After that, the furnace heat will decrease sooner or later.

For the above two reasons, it can be said that the difference in furnace wall temperature has a predictive value on the temperature change in the blast furnace. Therefore, an ARMA-E model can be considered in which a variable related to the furnace wall temperature difference is introduced into the MA term.

FIGS. 1(a) and 1(b) are a side sectional view and a plan sectional view, respectively, showing the arrangement of the inner wall temperature reducer used in an embodiment of the present invention. As shown in the figure (a), the inner wall thermometer 3
7 pieces in the height direction of blast furnace 1 (3 pieces on the back).

(2 abdominal parts, 2 morning glory parts), and 4 pieces are installed in the circumferential direction of the blast furnace 1 as shown in FIG. In other words, out of a total of 28 units, 1mrff;it3 will be installed in 4 directions and 7 levels.

The inner wall thermometer is, for example, disclosed in Japanese Utility Model Application No. 57-81 by the present applicant.
531. The one disclosed in Japanese Utility Model Publication No. 59-16816 may be used, and FIG. 2 is a conceptual diagram showing the inner wall thermometer (hereinafter referred to as "rFM sensor") disclosed in the latter.

In the figure, reference numeral 4 denotes a sheath-type heating element in which two conductive wires 5 are insulatively buried in parallel and having a temperature-sensing part 6 on the front end side. The temperature-sensing parts 6 are arranged in parallel so that they are arranged at different lengthwise positions, and a sheath-type dummy rod 7 is connected to the tips of the temperature-sensing parts 6 to align the leading ends. The sheath type dummy rod 7 has two conductive wires 5 embedded insulatingly in parallel, and has substantially the same thermal conductivity as the sheath type side heating body 4. The FM sensor 3 is formed by keeping the sheath type side heating bodies 4 in a non-contact manner with an insulating material 8 and storing them in a sheath tube 9.

FIG. 3 is an explanatory diagram of the configuration of the M sensor 3. In the figure, 10 to 13 are the walls of the blast furnace, 10 is a brick, and 1
1 is a stave, 12 is a stamp, and 13 is an iron skin. F
As shown in the figure, the M sensor 3 is installed inside the furnace wall by welding to a packing 14 and a welded portion 15. Note that 16 is a filler, and 17 is a milk injection port into which the filler 16 is poured.

Note that due to the installation and structure of the FM sensor 3 described here, the FM sensor 3 itself is eroded along with the erosion of the furnace wall, and the sheath type temperature sensing element 4 may be exposed inside the furnace near the furnace wall. In addition to "furnace wall temperature", "furnace temperature near the furnace wall"
is being measured. Hereinafter, a concept including both will be described as "furnace wall temperature." As mentioned above, compared to conventional thermometers such as sheathed thermocouples, the FM sensor 3 has a large number of measurement points, satisfies rapid temperature measurement response, is capable of long-term continuous temperature measurement, and is highly reliable. Improvements in properties, durability, workability, etc. are π1.

Each FM sensor 3 measures the inner wall temperature of the blast furnace 1 at every predetermined sampling time Δt, as shown in FIG. Here, the inner wall temperature of the i-th FM sensor 3 at time j is 14
．． If Δ between 1 sump J and 1 ring 115 at time j and the previous inner wall temperature is T, then
Inner wall temperature difference (difference value) ΔTJ between J-1,I T... and Tj-1,i
, 1 j, i is Δ”j, i””j, i ”j−i, + .
・It becomes (1). This state is shown in FIG.

These six differential values... are multiplied by a weight wl taking into consideration the height J, 1 circumference direction, etc. of each FM sensor 3. Furthermore, V for the difference value ΔTj,i is negative.

=01 For other values, v is also multiplied by a positive/negative parameter V indicating =1 to obtain a corrected difference value (positive difference value) CTj at time j.

CT--=w・-v, -ΔT-3...(2) J, I
l + Next, the sum of the corrected difference values CTj, , for all FM sensors 3 is taken, and this is set as 5TIj.

5T1j-=, E, CTj,, ・=(3
) and the sum of the difference values ST1 according to the following equation (4).

If the value of ε1 becomes larger than the predetermined threshold value ε1, it can be assumed that there has been a rapid temperature rise of the furnace wall due to the wall deposits inside the furnace, and by quantifying this event, MA
Can be introduced as a variable in a term.

ST1, ≧ε...(4) Also, in equation (2), the positive and negative parameters V・ are the difference values Δ
For positive Tj, , V,=Q.

For other values, set v to −1, then apply the total value of the absolute value of the corrected difference value CTj, to the M sensor 3, and calculate the difference based on the equation (3) ° according to the following equation (4). Value sum ST2
If the value of * becomes larger than the predetermined threshold value ε2, it can be assumed that there has been a rapid temperature drop due to the raw ore descent, and by quantifying this event, it can be introduced as a variable in the MA term.

ST2・≧ε　　　　　　　　　...(4).

If an event that satisfies formula (4) (formula (4)' is also acceptable) exists in the past by a time width NΔ from the current time t, then Fl
(t)=1. It is conceivable to introduce a variable F1 (t), which sets Fl (t)=-1 if it does not exist, as an MA term.

In this case, the predicted hot metal temperature value y(t+Δ12) at one point in time
can be estimated using the following equation.

V(t+Δt2) ``”i'=oa1iy(t −i△t2)+bI Fl
(t) ...(5) The first term on the right side is the AR term, and the second term is the MA term.

Note that all and bl are coefficients.

Figures 6 and 7 show examples of prediction using equation (5), where (a) is the hot metal temperature, (b) is the sum of positive difference values, 5T1
j(C) is Fi (t)+7) t'j rough indicating the change over time.

Also, from the current time t, the time width NΔ is moved backwards into the past, (
4) Variable F2 (
It is also possible to introduce t) as an MA term as shown in the following equation (6).

y (t + △t2) =, X, a 27 y (t - iΔt2) + b2F2
(t) ...(6) In addition, a21. F2 is a coefficient.

Figures 8 and 9 show examples of prediction using equation (6), where (a) is the hot metal temperature and (b) is the positive difference value sum ST1.
, (C) is a graph showing changes in F2(i) over time.

Il, the second variable introduced into the MA term It has been known that the increase or decrease in the amount of solution loss carbon (hereinafter referred to as "Tru loss IJ"), which indicates the quality of the reduction state of the blast furnace, has a strong correlation with the temperature of the blast furnace heat. Are known. In other words, an increase in the amount of Tsuru loss C indicates that the so-called Tsuru loss reaction shown below is promoted. The fever will drop.

Therefore, it can be said that an increase or decrease in the amount of truss loss C can be predicted with respect to changes in blast furnace heat.

Therefore, AR that introduces the maximum value of the amount of truss loss C into the MA term
An MA model can be considered.

FIGS. 10(a) and 10(b) show changes over time in the instantaneous value of the hot metal temperature and the average value of the tumble loss C, respectively. Figure (a)
, Ll is the actual value and F2 is the predicted value. The instantaneous value of the amount of truss C (Ng/1-p) is the c in the top gas component.
o, co2. Based on the ratio of N2, etc., air blowing conditions, raw material charging conditions, etc., it is calculated for each simplification time of gas chromatography, and with reference to FIG. is the same width t1 between predetermined u and y
Tsuru loss C hanging average value u(t- for each interval Δt1)
11) This is calculated as the maximum value A of (k-1 to N) at every sampling time Δt2. On the other hand, the instantaneous value y(t) of the hot metal temperature is determined every simplification time Δt2. The current time is shown here.

With such variables, the predicted value of the hot metal temperature V(t+Δt2) at one point in time can be determined by the following equation in ΔRMA'E del.

'Sl' (t+△j2) -Σ a, y (t-1Δt2) + c x (t
)stomach:. 1...(7) The first term on the right side is the AR term, and the second term is the MA term.

Note that a. and C are coefficients.

1 [[, Two variables to be introduced into the MA term 1) The variables mentioned in LL can be expected to have a considerable effect even if they are introduced individually into the MA term, but in order to further improve prediction accuracy, It is conceivable to introduce two variables into the same poem as MA terms.

In this case, the predicted hot metal temperature at one point in time '57 (t+Δt2
) can be estimated using the following equation.

V(t+Δt2) −Σ a ,y(t−1Δt2)1 ,, 1+ +b F (t)+c1x(t) ...(8) V(t+Δt2) −Σ a ,y(t−iΔt2) , :, 2+ -1-F2F2 (t)+C,,X (t)...(9
) In addition, a , b , c , a , b , c2
1i 1 1 2i
The number is 2.

Third variable to be introduced into the IV and MA terms [tackle port correction variable] By the way, normally in a blast furnace, tapping is performed alternately using two tap ports, so for different tap ports A and B, the difference in Fig. 11 is As shown in the change over time in the average hot metal temperature for each tap, deviations often occur in the hot metal temperature, and in such cases, prediction accuracy is affected. Note that Δt3 is the sampling patent number for each tapping date.

Therefore, if the tapping at the current time t and the tapping at the predicted time Δt2 are not from the same taphole, the accuracy of the predicted hot metal temperature value V(t+Δt2) using equation (8) or equation (9) is considered to be worse, so the following (10), (1
The prediction is made by correcting the difference in the tap hole according to Equation 11.

V(t+Δt2) Z':co31 y(t!Δt2)+b3F1 (t)
+Cx(t) d z(t+Δt2)...(10) y(t+Δt2) -fla4. y (t-1lXt2) 10b4 F2
(t)+C4X (t) 10d4Z (t+Δt2)・
...(11) Here, z (t) is a variable for correcting the difference in the tap hole, and the details will be described later. Note that a31. b3. C3゜d3.
a41. b4. C4 and d4 are coefficients.

Figure 12 shows the hot metal temperature y(t) during intermittent pouring.
, where Ll is the hot metal temperature V(t), LSI is the taphole correction variable z(t), and time ``A1.tA2'' is the taphole correction variable z(t), respectively. The tapping start time, the tapping end time, and the time tB1.tB2 at A indicate the tapping start time and the tapping end time at taphole B, respectively, and are predicted with n=2 using equations (8) or 11). Point P1 is the predicted value at time t2 at time t1, and both time t1 and predicted time t2 are at taphole A.
Since the tap time is within the same tap time tA1 to tA2, there is no need to consider the difference in the tap hole, so the prediction is performed using equation (8) or equation (9).

On the other hand, point P2 is the predicted value of time tB1 (t4) at time tA2 (t3), and time tA2 is the predicted value of taphole A
Since time tB1 is the tapping time by taphole B, it is not within the same tapping time. Therefore, (1
Prediction is performed using equation 0) or equation (11). Furthermore, at time tBl, the value of the taphole correction variable z(t) also changes in order to correct the difference in the tapholes.

The taphole correction variable z(t) is determined as follows.

Focusing on the M hot metal temperatures that have increased backwards from the current time t in the past, find the average value Y, and calculate the average value YA of MA of the M hot metal temperatures related to taphole A and MB regarding taphole B. The average value Y8 of (MA+M8=M) is calculated. Here, if the taphole at time 1o is A, z(t)-YA Y...(12
)B, then z(t)=YBY...(1
3) to find the taphole correction variable z (to).

Further, when tapping the same taphole 8 times (S≧2) consecutively, according to the formula (14) below, the formula (12) or (1
3) Change the correction variable z (to) for pig iron tap 1 obtained by the formula.

z(t)−(1+Σ(k/i)) z(t
n ) 0.2° (14) Here, k is lkl<1 and if z (to)>Q then k>O. If Z (to)<O then k<0. This is done in consideration of the heat retention effect of continuously tapping from the same tap hole. In other words, the continuous tapping warms the tap flute, so the actual hot metal temperature value rises higher than normal. Considering this temperature difference 4, the taphole correction variable z (to) includes a temperature rise equivalent to (Σ(lkl/i)) - I l z (t, ) l than normal. be. moreover,
The control temperature, which is the standard for the hot metal temperature actual value, is changed in consideration of the temperature increase due to continuous tapping as described above.

In this way, by considering the difference in tapholes,
As shown in Figure 2, the predicted value 'Sl' (tBl) at time tB1 (t4) is as follows:
In the case of point P2' and the next taphole B, it becomes point P2. Although we have explained the case of intermittent tapping in which there is a period of time between one tap and the next, the overlap can also be applied in the case of plus tapping in which the taps from tapholes A and B are partially overlapped. By focusing on one of the tapholes,
It can be predicted by the above equations (8) to (11). In this case, it is impossible to use continuous hot metal from the same taphole.

Introducing the above-mentioned taphole compensation variable z(t) into the MA term is a fairly effective means when the temperature deviation due to the difference in tapholes is significant as shown in Figure 11. If the temperature deviation due to the difference in the taphole is small, it is almost meaningless and may have a negative effect on the predicted viscosity. Therefore, it is preferable to set the threshold value to 81, provide a condition such as equation (15), and perform prediction using equation (10) or equation (11) only when equation (15) is satisfied.

IY-YB l>εT...(15)
FIGS. 13(^) and 13(B) are flowcharts showing the processing procedure of the prediction method based on equations (8) and (10).

In the same figure, in step S1, the sum of positive difference values ST1 is calculated using equations (1) to (3), and in step S2, it is determined whether equation (4) is satisfied between the current time t and the past time width NΔt1. judge. If the equation (4) is satisfied even with 11v, Fl (t) is set to 1 in step S3, and if the equation (4) is never satisfied, Fl (t) is set to -1 in step S4.

Then, in step S5, a period NΔt1 from the current time t
Going backwards in the past, N average values u (Δt1 at t-) (k-1 to N) of the amount of trailing loss C are calculated for each section Δt1.

Next, in step S6, the maximum value of the average value u (Δt1 for t-) obtained in step S5 is obtained from the equation (16) below, and this is set as x(t).

x (t)=max (u (Δt1 to t-))k-1
,...,N...(16) Note that steps 81 to 84 described above. The processes of steps 85 to S6 are performed every sampling time Δt2.

Then, in step S7, it is checked whether the current time t and the predicted time △t2 are within the same tapping time, and if they are within the same tapping time, in step S8, the actual hot metal temperature value y
(t) and the prediction result V(t), the coefficient a11 of equation (8). Determine b1 and C1. Next step S9
The predicted value V(t+Δt2) of the hot metal temperature one point ahead is calculated by the coefficient a11. calculated in step S8. b1. It is obtained from (8) 2' using C1, and the process moves to step 819.

On the other hand, in step S7, the current time t and the predicted time (t
+Δt2) is not within the same tapping time, step 81
0, going backwards from the current time t to the past, M 1
Paying attention to the hot metal temperature during tapping, the average value Y of the M hot metal temperatures, and each of the M hot metal temperatures for tapping 1'' A and tapping B
MA, M, (M 1 M = M) hot metal temperature average value YA.

8 B Calculate YB. Next, in step S11 (15
) It is determined whether the equation (15) is satisfied, and if the equation (15) is satisfied, the process moves to step S12 because it is necessary to correct the difference in the tap hole. Since there is no need to correct the difference in mouth, the process moves to step S8, and the predicted hot metal temperature value y(t+Δt2) is determined by equation (8) as described above, and the process moves to step S19.

On the other hand, if it is necessary to correct the difference in the taphole, the taphole at the predicted time (t+Δt2) is determined in step S12, and if the taphole is A, the predicted time (is calculated in step 813 by equation (12)). Calculate the correction variable z (for tap iron 1 at 10Δt2), and calculate the
If so, in step 814, the taphole correction variable z (+Δt2) is determined using equation (13).

Furthermore, in step S15, it is checked whether the tap hole at the current time t and the predicted time (with Δt2) are the same. Therefore, when tapping continuously from the same taphole, in step 816, according to equation (14),
The taphole compensation variable z(
change Δt2). Next, in step 817, the coefficient a31 of equation (10) is calculated based on the comparison between the actual hot metal temperature value y(t) and the predicted result V(t). b3. Determine C3, d3. Furthermore, in step 818, the predicted value of the hot metal temperature '57 (t+Δt2) at one point ahead is set to step S17.
The coefficient a31. b3. It is obtained from equation (10) using C3 and d3, and the process moves to step S19.

Heat prediction is performed by outputting the hot metal temperature predicted value y (t+Δt2) obtained by equation (8) or equation (10) in step S19, and thereafter the process returns to step S1 to continue the prediction.

FIGS. 14(A) and 14(B) are 70-diagrams showing the processing procedure of the prediction method based on equations (9) and (11). In the same figure, in step 821, the sum of positive difference values 5TIj is calculated using equations (1) to (3), and in step S22, it is determined whether equation (4) is satisfied while reaching the past time width NΔt1. If formula (4) is satisfied, step S
In 23, if the number of times ST1≧ε1 occurs is p, then F2 (t)=p. On the other hand, if equation (4) is not satisfied even once, in step 824, F2
Let (t)-=O. Then, in steps 825 to 839, step 8 in the flowchart of FIG.
Processing similar to steps 5 to 819 is performed.

In addition to the variable related to the furnace wall temperature difference (difference value) and the maximum value of trunnion loss C, which have a high degree of foresight regarding changes in furnace heat, variables that take into account differences in the tap hole are required.

The predicted value obtained by the ARMA model, which is introduced into the MA term as needed, is quite accurate.

V. Supplementary notes In this example, an example was described in which an event that satisfies equation (4) is digitized and introduced into the MA term, but the same result can be obtained even if the event that satisfies equation (4) is digitized and introduced into the MA term. be effective.

In addition, although an FM sensor was used for the inner wall temperature it in this example, a normal temperature sensor (for example, a sheathed thermocouple) can be used as a substitute, although there is a problem in terms of service life. Although the prediction accuracy will be slightly lower due to its reliability and low temperature response, it can be used as a substitute.

Further, in this embodiment, 28 FM sensors 3 are installed in 7 levels and in 4 directions, but it goes without saying that they may be installed appropriately depending on the characteristics of the blast furnace.

Furthermore, when calculating the average of the amount of trailing loss C for each section Δt1, the instantaneous value of trailing loss C1 may generate an abnormal value E1°E2 due to noise or the like, as shown in FIG. 15(a). Here, if the amount of vine loss C at time j is Xj, and the amount of vine loss C before lJ sampling time Δ is xj-1, then the absolute value of the difference value of the amount of vine loss C is ΔX・−1×・−Xj−1) ...(17)
It becomes J. By comparing this ΔX・ with the threshold value ε2 as shown in the same figure (b), abnormal values El and E2 are found, and the abnormal values El and E2 are found in the same figure (C
), a method of smoothing by replacing it with the previous value can be considered. By applying this method, a more accurate maximum value of the truss loss C amount can be determined, and as a result, even more accurate prediction is possible.

In this example, the maximum value of the average value for each interval Δt1 was determined as the maximum value of the crane loss Cf1l within the predetermined period NΔt1, but the maximum value of the instantaneous value of the crane [1 sq. The maximum value may be used. however,
Since it cannot be denied that it is susceptible to the influence of noise components, it is preferable to perform the above-mentioned noise removal process at the same time.

In addition, the nitrogen concentration (%) in the furnace top gas detected by gas chromatography (hereinafter referred to as "Gas chromatography N21J") has a strong negative correlation with the turret loss C;
A similar effect can be expected by using the minimum value of the two quantities in the MA term instead of the maximum value between zero and C.

(Effect of the invention) As explained above, the variables for correcting the difference in the taphole are
By using the ARMA model, which is introduced into the MA term only when a temperature deviation due to a difference in the taphole is found to be more than a predetermined value, it is possible to accurately predict the blast furnace furnace temperature, especially the furnace heat decrease, without delay.

[Brief explanation of the drawing]

FIGS. 1(a) and (b) are a side sectional view and a plan sectional view showing the arrangement of an FM sensor used in an embodiment of the present invention, and FIGS. 2 and 3 are a conceptual diagram and a plan view of the FM sensor, respectively. 4 is a graph showing changes in inner wall temperature over time, FIG. 5 is a graph showing changes in inner wall temperature difference over time, and FIGS. 6 to 9 each show a prediction example in an embodiment of the present invention. (a) is the hot metal temperature, (b) is the sum of positive difference values, (C
) is a graph showing changes over time in variables related to inner wall temperature difference,
Figures 10 (a) and (b) are graphs showing the changes over time in the instantaneous value of hot metal temperature and the average value of trough loss C, respectively. Graphs showing changes in hot metal temperature over time, Figure 12 is a graph showing changes in hot metal temperature and taphole compensation variables over time, Figures 13 (^), (B) and Figures 14 (^), (13 ) are flowcharts showing the processing procedure of the prediction method in the embodiment of the present invention, and FIGS. 15(a), (b), and (c) respectively.
are graphs showing the instantaneous value of the vine [1sc amount including abnormal values, the absolute value of the difference value of the vine loss C amount, and the instantaneous value of the vine loss Cff1 after removing the abnormal value, and Fig. 16 shows the conventional AR
2 is a graph showing prediction results based on actual hot metal temperature values based on the method. 1...Blast furnace, 3...FM sensor, Ll・
... Actual value, [2... Predicted value Fig. 1 (a) (b) Fig. 2 and Fig. 3 Fig. 6 Fig. 8

Claims (2)

[Claims] (1) When predicting temporal fluctuations in the blast furnace heat level using the ARMA model using blast furnace operation results, it is assumed that when the temperature deviation due to different tapholes exceeds a predetermined value within a predetermined period, , a blast furnace furnace heat prediction method characterized in that a variable for correcting the tap hole is introduced into the MA term. (2) The ARMA model includes, in the MA term, a variable related to the inner wall temperature difference measured at predetermined time intervals by an inner wall thermometer installed at a predetermined location in the blast furnace, and the maximum value of the amount of solution loss carbon in a predetermined time width. or the minimum value of the amount of nitrogen in the furnace top gas component in a predetermined time period.