JPH0635605B2 - Blast furnace furnace heat drop prediction method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a blast furnace heat drop prediction method for performing stable operation of a blast furnace.

(Prior Art and Problems Thereof) Conventionally, the heat of the blast furnace is predicted by increasing or decreasing the amount of solution loss carbon (hereinafter referred to as “sol loss C amount”) indicating the quality of the reduced state of the blast furnace. Sollos C
The increase in the amount indicates that the so-called Sollos reaction shown below is promoted.

C + CO ₂ → 2CO Since this reaction is an endothermic reaction, it can be predicted that the heat of the blast furnace decreases.

Sorurosu C amount, in each analysis cycle of the gas chromatography analysis of the composition of the normal furnace top gas (about 3 minutes), CO in the furnace top gas, CO _2, the ratio of N ₂ like and blowing conditions and the raw material charging It is calculated based on the conditions, and in the past it was Solros C every 1 hour.
The furnace heat drop was controlled by the average amount.

7 (a) and 7 (b), FIG. 7 (a) shows the amount of sol loss C every 3 minutes (l1) and the average value of sol loss C every 1 hour (l
2) shows the change with time, and FIG. 6B is a graph showing the change with time of the hot metal temperature. In the figure, the threshold value ε ₁ is exceeded at the time 17 o'clock, but no heat-raising action is taken, and the hot metal temperature is drastically lowered after that. FIG. 8 shows the changes over time when the heat raising action A was initiated at time 13:00 at which the threshold value ε ₂ was exceeded. Similar to FIG. 7, in the figure, l1 indicates an instantaneous value every 3 minutes, and l2 indicates an hourly average sol-loss C amount. By comparing FIG. 7 and FIG. 8, it is understood that the hot metal action A can avoid the decrease in the hot metal temperature as shown in FIG. 7 (b) to some extent as shown in FIG.

However, in the prediction of the average value of the sol-loss C amount for each hour, when there is a sudden increase in the sol-loss amount, in the worst case, there is a delay of about 1 hour in the prediction of the furnace heat drop. was there. For example, even in the case of FIG. 8, due to the prediction delay, the heat-raising action A is taken only after the hot metal temperature falls below the control temperature _Tc to some extent. Therefore, in order to avoid this problem, when the furnace heat drop is predicted by the instantaneous value of the amount of sol loss C measured at intervals of about 3 minutes, the individual variations as shown by l1 in FIG. 7 and FIG. It is large and the noise component is large, so there is no data persistence. Therefore, it is almost impossible to predict the decrease in furnace heat with the instantaneous value of the amount of sol loss C.

(Object of the Invention) An object of the present invention is to solve the above-mentioned problems of the prior art, and to provide a blast furnace heat drop prediction method capable of predicting as quickly as possible and accurately predicting a drop in hot metal temperature. That is.

(Means for Achieving the Purpose) In order to achieve the above-mentioned object, the method for predicting heat loss in a blast furnace according to the present invention is such that during operation of the blast furnace, the amount of solution loss carbon and the amount of nitrogen in the gas component of the furnace top are set at predetermined time intervals. Then, a moving average of the calculated value in a predetermined time width is calculated, and each moving average value of the solution loss carbon amount and the nitrogen amount is compared with a plurality of threshold values to give an evaluation point. , The blast furnace heat input decline is predicted according to the comprehensive evaluation based on the evaluation points.

(Example) Solros C amount (kg / tp) depending on the top gas component analysis by gas chromatography, blowing conditions, raw material charging conditions, etc.
) Is calculated for each sampling time Δt. here,
Let x _j be the amount of Sollos C at the current time j, and time j
If the amount of sol loss C before k sampling time (that is, Δt × k time before) is x _jk , the moving average x _M of the predetermined time width nΔt at the current time j is Can be calculated by

X _M based on the equation (1) is calculated for each sampling time Δt, and the threshold ε for which x _M is predetermined is calculated by the following equation (2).
_{When x} is exceeded, an alarm is issued and the furnace heat drop is predicted.

x _M > ε _x (2) In addition, the amount of N ₂ (%) in the furnace top gas detected by gas chromatography (hereinafter, referred to as “gas N ₂ amount”) has a strong negative value with the amount of Solros C. There is a correlation, and it is possible to predict a decrease in blast furnace heat due to a decrease in the amount of gas black N ₂ instead of the amount of Solros C.

Thus, the gas chromatography N ₂ amount at the current time j and y _j, than the time j k sampling time before the gas chromatographic N ₂ amount (i.e. Delta] t × k times before) When y _jk, at the current time j The moving average y _M of the predetermined time width nΔt is Can be calculated by

Y _M based on equation (3) is calculated for each sampling time Δt, and y _M is set to a predetermined threshold value ε according to equation (4) below.
_{When y} falls below _y , an alarm is issued to predict a decrease in furnace heat.

y _M <ε _y (4) In FIG. 1, (a) shows the change with time of the moving average value of the Solloss C amount, and Δt = 3 (min), n in the equation (1).
= About 7. FIG. 3B shows the change in hot metal temperature with time, and T _c shows the control temperature. As shown in the figure, in order to satisfy the equation (2) after 11 o'clock, it is understood that the furnace heat drop can be predicted and the hot metal action A can be performed to avoid the drop in the hot metal temperature. Moreover, Δt = 3
Since a 21-minute moving average is calculated for each (min),
The prediction of furnace heat drop will not be delayed and the accuracy will be sufficiently reliable.

Although not shown, the same effect can be obtained when the furnace heat drop is predicted by satisfying the equation (4) using the moving average y _M of the amount of gas black N ₂ .

By the way, when obtaining the moving average of the amount of C loss, the instantaneous value of the amount of C loss may generate abnormal values E1 and E2 due to noise or the like as shown in FIG. 2 (a). Here, assuming that the amount of sol loss C at time j is x _j and the amount of sol loss C before one sampling time Δt is x _j−1 , the sol loss C is
The absolute value Δx _j of the difference value of the amount is Δx _j = | x _j −x _j−1 | (5). A method of finding out the abnormal values E1 and E2 by comparing this Δx _j with the threshold value ε _z as shown in FIG. 11B, and replacing it with the immediately preceding measured value as shown in FIG. Can be considered. By applying this method, a more accurate moving average of the SolLos C amount can be obtained, and as a result, it is possible to perform prediction with considerably high accuracy.

The method for predicting the furnace heat drop based on the moving average of the amount of sol loss C including the correction of the abnormal value can be realized by using a computer. FIG. 3 is a flowchart showing the flow of the processing. In the figure, in step S1, the instantaneous value x _{j of the} amount of sol loss C is obtained for each sampling time Δt. Then, the absolute value [Delta] x _j of the difference value of Sorurosu C amount in step S2, if the absolute value [Delta] x _j of next difference value at step S3 is larger than a threshold epsilon _z, in step S4, the instantaneous value x _j is regarded as an abnormal value, replaced with the immediately preceding measured value x _j−1 , and step S5
Move to. On the other hand, if it is smaller than the threshold value ε _z , the process proceeds to step S5 without changing the instantaneous value x _j . In step S5, the moving average x _M of the time width nΔt is calculated, and in the next step S6, the moving average x _M is compared with the threshold value ε _x.
If the temperature exceeds the threshold ε _x , it is considered that the blast furnace heat drop occurs,
An alarm is output in step S7. On the other hand, if the moving average x _M is below the threshold value ε _x , it is determined that there is no abnormality, the process returns to step S1 again, and steps S1 to S6 are repeated.

Of course, the above-mentioned abnormal value processing and realization by a computer can also be applied to the case of predicting the furnace heat drop based on the moving average value y _M of the amount of gas black N ₂ .

Here, the furnace heat decrease prediction based on the Solros C moving average value is A, and the furnace heat decrease prediction based on the gas black N ₂ moving average value is B.
Then, in the above description, the prediction A and the prediction B were predicted by using one threshold value ε _x and ε _y , respectively, but by providing a plurality of threshold values ε _{x1 to} ε _xn and ε _{y1 to} ε _ym , the furnace heat It is possible to make predictions including the degree of decline.

The thresholds ε _{x1 to} ε _xn and ε _{y1 to} ε _ym are respectively ε _x1 <ε _x2 … <
ε _xn , ε _y1 > ε _y2 ...> ε _ym and n, m stages are set. The moving average value x _{M of the} amount of SolLos C is at maximum the threshold value ε _xi.
If (i = 1 to n) is exceeded, the evaluation point i of the prediction A is given. On the other hand, the moving average value y _{M of the} amount of gas chromatogram N ₂ is minimum and the threshold value ε _yj
When it is less than (j = 1 to m), the evaluation point j of the prediction B is given.
Based on these evaluation points i and j, the total evaluation C
Ask for.

C = w _A · i + w _B · j (6) where w _A and w _B are weights for the predictions A and B. According to the value of this comprehensive evaluation C, it also indicates the degree of furnace heat drop which will occur soon when an alarm is output. This clearly shows how much heat-raising action should be taken.

The flow of processing when the above-described prediction by the comprehensive evaluation C is performed using a computer is shown in the flowchart of FIG. 4, and the description thereof will be given below. First, in step S11, thresholds ε _{x1 to} ε _xn (ε _x1 <ε _x2 used for prediction A and prediction B are used.
... <ε _xn ), ε _{y1 to} ε _ym (ε _y1 > ε _y2 ...> ε _ym ) are set. Next, in step S12, the moving average value x _{M of} the amount of sol loss C and the moving average value y _M of the amount of gas black N ₂ are calculated for each sampling time Δt. At this time, it is desirable to perform the abnormal value processing as shown in FIG. Then, in step S13, the respective evaluation points i and j of the prediction A and the prediction B are initialized to 0.

Thereafter, in steps S14 to S16, the evaluation point i of the prediction A is obtained. First, in step S14, the moving average value x _M of the Sollos C amount and the threshold value ε _x1 are compared, and x
_{If M} ≧ ε _x1 , i = 1 and the value of i are incremented by 1 in step S15, and i = n is determined in step S16, or x _M <ε _{x (i + 1)} is determined in step S14. Steps S14 to S16 are repeated while gradually increasing the value of the threshold ε _{x (i + 1)} until the evaluation point i is calculated, and the process proceeds to step S17. If x _M <ε _x1 , steps S15 and S16 are not executed even once, the evaluation point i is set to 0, and the process proceeds to step S17.

In steps S17 to S19, the evaluation point i of the prediction B is calculated. First, in step S17, the moving average value y _M of the gas chromatograph N ₂ and the threshold value ε _y1 are compared, and x _M ≦ ε
If _y1 , j = 1 and the value of j is 1 in step S18.
If it is determined that j = m in step S19,
While gradually decreasing the value of the threshold value ε _{y (j + 1)} until it is determined that y _M > ε _{y (j + 1)} in step S17, steps S17 to S19 are repeated to calculate the evaluation point j, Control goes to step S20. If y _M > ε _y1 , steps S17 and S18 are never executed, the evaluation point j is set to 0, and the process proceeds to step S20.

In step S20, the comprehensive evaluation C is calculated according to the equation (6). If C> 0 in step S21, at least prediction A and prediction B in step S22.
Assuming that the furnace heat drop is predicted by either of the above, an alarm is output while changing the degree of warning according to the value of the comprehensive evaluation C. On the other hand, if C = 0, it is considered that there is no tendency of the furnace heat decrease according to the predictions A and B, and the process returns to step S12.

FIGS. 5 and 6 are graphs showing an actual prediction example in the case where the weights w _A = w _B = 1 and n = m = 3 and the range of the comprehensive evaluation C is 0 to 6, (a) is a representative value of the hot metal temperature of each tap that is actually tapped, (b) is a moving average value x _M of Solros C amount, (c) is a moving average value Y _M of gas black N ₂ amount,
(d) shows the change with time of the comprehensive evaluation C. In FIG. 5, the maximum value of the comprehensive evaluation C indicates 6, and it is necessary to output a serious alarm. Actually, as shown in FIG. 9A, a typical value of the hot metal temperature which is well below the control temperature T _c is measured. In FIG. 6, the maximum value of the comprehensive evaluation C is 2, and a light alarm is sufficient for this level. Actually the same figure
Only the furnace heat reduction shown in (a) has occurred.

Thus, the degree of the heat raising action can be finely changed by changing the degree of the alarm according to the value of the comprehensive evaluation C. As a result, it becomes possible to select the necessary and sufficient heat-raising action, and it is possible to reliably prevent a decrease in the furnace heat, and to avoid an unnecessary rise in the furnace heat due to an excessive heat-raising action, and to stabilize it. And it became possible to operate the blast furnace economically.

(Effects of the Invention) As described above, according to the present invention, prediction can be quickly obtained by performing prediction by comprehensive evaluation in which the moving averages of the Solros C amount and the gas N ₂ amount are compared with a plurality of threshold values at any time. In addition, it is possible to accurately predict a decrease in the hot metal temperature and, in addition, take a heating action as required.

[Brief description of drawings]

Fig. 1 is a graph showing the temporal changes in the moving average of the Solros C amount and the change in the hot metal temperature, Fig. 2 (a), (b)
, (c) is a graph showing the instantaneous value of the sol-loss C amount including the abnormal value, the absolute value of the difference value of the sol-loss C amount, and the instantaneous value of the sol-loss C amount after removing the abnormal value. FIG. FIG. 4 is a flow chart showing the flow of processing when one embodiment is applied to a computer, FIG. 4 is a flow chart showing the flow of processing for predicting furnace heat drop by comprehensive evaluation, and FIGS. 5 and 6 are actual predictions by comprehensive evaluation. In the graph showing an example, (a),
(b), (c), (d) are the hot metal temperature representative value, gas black N ₂ content,
Fig. 7 is a graph showing the amount of solross C and comprehensive evaluation. Fig. 7 is a graph showing the hourly average value of the amount of solross C and the temporal change of the hot metal temperature, and Fig. 8 is the time when the heating action is taken. 2 is a graph showing the hourly average value of the amount of sol loss and the temporal change of the hot metal temperature. ε _x , ε _y , ε _{x1 to} ε _x3 , ε _{y1 to} ε _y3 ...... threshold value, T _c ...
… Control temperature

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Takeshi Yabata 425-45 Wada, Inami-cho, Kako-gun, Hyogo Prefecture

Claims (2)

[Claims] 1. The amount of solution loss carbon and the amount of nitrogen in the gas components at the top of the furnace during blast furnace operation are calculated at predetermined time intervals, and a moving average of the calculated values in a predetermined time width is calculated. A method for predicting heat drop in a blast furnace, in which each moving average value of the amount of nitrogen and the amount of nitrogen are compared with a plurality of threshold values to give an evaluation point, and the heat decrease in the blast furnace is predicted according to the comprehensive evaluation by the evaluation points. 2. The solution loss carbon amount and the nitrogen amount are obtained by comparing the difference values of adjacent sampling data with a predetermined reference value to detect an abnormal value and removing the abnormal value. Prediction method for blast furnace heat drop according to item 1 of the above.