JPH09108801A - Method for predicting and preventing breakout in continuous casting - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately predict the breakout by arranging a plurality of temp. measuring elements to a mold in the casting direction to measure the temps., obtaining deviations from normal values when the measured values change, using the product of the deviations as a covariance value and monitoring. SOLUTION: The temp. detecting points 3a, 3b on the mold 1 are arranged in line in the casting direction. When the broken part 4 of a solidified shell passes through near the temp. detecting points 3a, 3b, the temps. are raised, and after passing through, the temps. are returned back to the original temp. changed as the hill-shape. In the case of parallel shifting the variation with time of the temp. detecting point 3a in the future direction on the time axis so that the delay of time becomes zero, similar temp. variation appears on the detecting points 3a, 3b in the same timing. The deviations ΔTA and ΔTB are obtd. from the temps. C, D at the normal time. The product of the deviations ΔTA and ΔTB is obtd. and made to the covariance value N. The covariance N becomes the large value only when the broken part 4 of the solidified shell and inclusion pass through. The breakout caused by constraint and the breakout caused by inclusion can surely be predicted by observing the covariance value N.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for predicting breakout caused by solidified shell breakage and inclusions occurring during casting by a continuous casting method and a method for preventing the breakout.

[0002]

2. Description of the Related Art In the process of continuous casting, a molten steel is poured into a mold and then continuously drawn from below the mold to produce a slab. Further, not only the quality of the slab but also the productivity, safety and maintenance of facility functions of the continuous casting are largely influenced by the initial solidification state of the molten steel.

In the initial solidification of the molten steel in the mold, when the molten steel directly contacts the mold to form a solidified shell due to depletion of powder between the mold and the molten steel, the solidified shell is adsorbed to the mold. Be detained. When the adsorbed portion is pulled downward by pulling out the slab, the solidified shell breaks in a V shape starting from the adsorbed portion. After that, the molten steel flowing into the gap of the broken solidified shell forms a solidified shell that is newly adsorbed in the mold, and while repeating the process of further breaking, the solidified shell breaking position gradually moves downward by withdrawal, and finally After leaving the mold, molten steel in the unsolidified portion inside the slab may flow out, so-called restraint breakout may occur.

Further, in the initial solidification of molten steel in the mold, when powder is abnormally flowed between the mold and molten steel or a large inclusion is caught in the surface of the solidified shell,
Only in that part, heat removal by cooling the mold is not sufficient and the thickness of the solidified shell becomes thin. Then, the thin portion of the solidified shell, after exiting the mold by pulling out the slab, could not withstand the molten steel static pressure of the internal non-solidified portion, the inclusions fell off and the slab surface solidified shell broke at the same time. However, so-called inclusion breakout may occur in which molten steel flows out.

When these breakouts occur, it is only a direct damage immediately after the occurrence that various signal lines in the vicinity of the continuous casting machine are cut or the equipment is damaged by the high temperature molten steel flowing out. However, it is impossible to remove the metal itself due to the solidified metal after flowing out, or even if this work is possible, the slab remaining in the mold will be completely solidified and cannot be withdrawn. In addition, there is also the damage that it takes a long time for the recovery work after the occurrence, such as the replacement of the casting roll damaged during the smelting of the outflow metal, and the productivity, safety and safety in continuous casting. It greatly hinders maintenance of facility functions. Furthermore, regarding the inclusion breakout, since the inclusions fall out and the molten steel flows out only after it cannot withstand the molten steel static pressure after passing through the mold, the solidified shell ruptures in the mold and the molten steel immediately after passing the mold. The amount of molten steel that flows out is large compared to the restrictive breakout that flows out.

Therefore, JP-A-57-115959, JP-A-57-115960, and JP-A-57-1
No. 15961, JP-A-57-115962, etc. predict the occurrence of restraint or inclusion breakout by detecting breakage or abnormality of solidified shell due to inclusion while it is in the mold. There has been proposed a method of immediately instructing to stop or decelerate withdrawal to promote the growth of a solidified shell in a mold and prevent these breakouts. All of these are for predicting the occurrence of breakout by capturing a certain abnormal temperature change pattern that appears in the time series change in temperature of a plurality of temperature measuring elements attached to the mold.

[0007]

However, these prediction methods have a problem in terms of accuracy. This is because the change in the mold temperature is to be captured by the temperature deviation from the reference temperature, the time during which the temperature deviation continues, and the rate of temperature change, and specifically, two adverse effects occur. One is that since the time-series smoothed value is used as the reference temperature when calculating the temperature deviation, the temperature deviation cannot be accurately grasped when the mold temperature is not stable due to disturbances or vertical cracks caused by casting conditions. Another point is that if a large number of temperature measuring elements are installed in the casting direction of the mold and prediction is made based on the measured mold temperature, the judgment constant is very large and the adjustment load becomes high. Therefore, it is difficult to adjust it sufficiently.

For this purpose, it is possible to accurately and promptly predict the occurrence of restraint or inclusion breakout while the solidified shell rupture portion and inclusions are in the mold, and based on this, breakage can be predicted. It is necessary to make sure to prevent out. And the prediction method must be easily adjustable. That is, it is necessary to reduce the constant used for the determination as much as possible.

[0009]

Means for Solving the Problems The present invention has been made to solve the above problems. Means 1 is to install a plurality of temperature measuring elements in a casting direction of a mold of a continuous casting machine. The mold temperature is measured, and when the measured value rises, the first-order lag temperature from the rising start point is obtained, and the temperature difference is obtained from the measured mold temperature and the first-order lag temperature for each temperature measuring element in the casting direction. The temperature difference in the temperature measuring element position on the upstream side in the casting direction, the distance between the upstream temperature measuring element position and the downstream temperature measuring element position, and the moving time of the slab obtained from the slab drawing speed In the method, an integrated value with the temperature difference at the position of the downstream temperature measuring element is calculated, and when the integrated value becomes a predetermined value or more, the occurrence of a restrictive breakout due to the fracture of the solidified shell is predicted.

Further, the means 2 measures the mold temperature by installing a plurality of temperature measuring elements in the casting direction in the mold of the continuous casting machine, and when the measured value drops, the primary from the descent start point The delay temperature is obtained, and the temperature difference is obtained for each temperature measuring element in the casting direction from the primary delay temperature and the measured mold temperature, and the temperature difference at the temperature measuring element position on the upstream side in the casting direction and the upstream side temperature measurement Calculate the integrated value of the temperature difference at the downstream temperature measuring element position after the moving time of the cast piece obtained from the distance between the element position and the downstream temperature measuring element position and the withdrawal speed of the cast piece, and this integrated value Is a method of predicting the occurrence of inclusion breakout caused by inclusions when the value exceeds a predetermined value.

Then, when the integrated value, which is the index of the constraint breakout obtained by the means 1 and the index of the intervening property breakout obtained by the means 2, becomes a predetermined value or more, the magnitude of this integrated value According to the above, there is provided a method for preventing the breakout of restraint properties and inclusions in continuous casting, which is characterized in that the occurrence of breakout is prevented by adjusting the drawing speed of the slab.

[0012]

First, a method for predicting a constrained breakout by detecting a solidified shell rupture occurring during initial solidification in a mold will be described with reference to FIGS.
FIG. 2 shows an example of the temperature detection line 2 installed in the mold 1, which has a plurality of temperature detection points 3a and 3b in the casting direction.
Since the molten steel directly contacts the mold at the solidified shell rupture portion 4, when the molten steel passes near the temperature detection row 2b, the temperature detection point 3a
As shown in FIG. 1 (a), the time series temperature changes of 3 and 3b are
There is a similar temperature change pattern of 3a, in which the temperature greatly increases from the normal temperature and then returns to the original value (so-called mountain shape).
It appears in time with the passage of 3b, that is, with a time delay due to the drawing speed.

With respect to this temperature change pattern, when the time series change of the temperature detection point 3a is translated in the future direction on the time axis so that the time delay due to extraction becomes 0,
As shown in FIG. 1 (b), similar temperature changes appear at the same timing, and these temperature changes are expressed as normal temperature C (t), D
The increase from (t), that is, the deviations ΔTA and ΔTB are captured. The so-called covariance value N (t) calculated by the product of the deviations ΔTA and ΔTB becomes large only when the solidified shell fracture portion passes, as shown in FIG. 1 (c).
It is very suitable as an index for accurately predicting the occurrence of restraint breakout.

Next, a method of predicting the breakout of inclusions by detecting inclusions caught in the solidified shell during initial solidification in the mold will be described with reference to FIGS. FIG. 11 shows an example of the temperature detection line 2 installed in the mold 1, which has a plurality of temperature detection points 3a and 3b in the casting direction. Inclusion 5 caught in the solidified shell
Is a so-called heat removal failure portion, and when this passes through the vicinity of the temperature detection row 2a, the time-series temperature change at the temperature detection points 3a and 3b changes from the normal temperature as shown in FIG. A similar temperature change pattern of 3a and 3b, in which the temperature greatly descends, then rises, and then returns to the original (so-called valley type)
It appears at the timing of passing in the order of, that is, with a time delay due to the drawing speed.

With respect to this temperature change pattern, when the time series change of the temperature detection point 3a is moved in parallel in the future direction on the time axis so that the time delay due to extraction becomes 0,
As shown in FIG. 10 (b), similar temperature changes appear at the same timing, and this temperature change is represented by the normal temperature C (t),
The amount of decrease from D (t), that is, the deviations ΔTA and ΔTB are captured. And calculated by the product of the deviation ΔTA and ΔTB,
The so-called covariance value N (t) becomes a large value only when the inclusions caught in the solidified shell pass, as shown in FIG. 10 (c), so that the occurrence of inclusion breakout can be accurately predicted. It is very suitable as an index for the occasion.

Further, it is considered that the thickness of the solidified shell becomes thinner as the peak-shaped or valley-shaped temperature change is larger, and the larger the temperature rise is, the more restrictive breakout is caused, and the larger the temperature decrease is, the more the intervening property breakout is caused. It has been empirically confirmed that the probability of each occurrence is high. To prevent these breakouts, the breakage of the solidified shell or the inclusions in the mold should be extended to promote sufficient growth of the solidified shell. There is a need. Since the covariance value increases with a large temperature change, it is possible to estimate to some extent how thin the solidified shell is from the covariance value. Therefore, it is possible to minimize the influence on the quality and productivity of the slab.

[0017]

EXAMPLES Examples of the present invention will be described below with reference to equations (1) to (7) and FIGS. First, a method for predicting and preventing a restrictive breakout will be described, and then a method for predicting and preventing an intervening property breakout will be described focusing on differences from the restrictive breakout.

First, the method of calculating the covariance value, which is used as an index of solidified shell rupture occurring during initial solidification in the mold, which causes the restraint breakout, will be specifically described. This is represented by the covariance value N (t) at time t, and is expressed by the following equation (1).

N (t) = ΔTA (t−v (t)) × ΔTB (t) (1-1) ΔTA (t) = A (t) −C (t) (1− 2) ΔTB (t) = B (t) -D (t) (1-3) where ΔTA (t): temperature detection point 3a at time t.
Temperature deviation (° C.) ΔTB (t): Temperature deviation of temperature detection point 3b at time t (° C.) v (t): Time delay between the temperature detection points 3a and 3b due to extraction speed at time t (second) A ( t): temperature of temperature detection point 3a at time t (° C) B (t): temperature of temperature detection point 3b at time t (° C) C (t): temperature of temperature detection point 3a at time t at normal time ( C) D (t): Temperature at normal temperature of temperature detection point 3b at time t (° C) tv (t) in the formula indicates that the time is v (t) seconds before t, This corresponds to the parallel movement on the time axis. Then, the time delay v (t) between the temperature detection points 3a and 3b depending on the drawing speed is expressed by the following formula (2).

V (t) = L / W (t) (2) where v (t): time delay (seconds) between the temperature detection points 3a and 3b due to the drawing speed at time t L: temperature detection Distance between points 3a and 3b (m) W (t): Extraction speed at time t (m / sec) Further, the deviations ΔTA and ΔTB are set so as to increase only when the temperature suddenly rises. The temperature is set to be the lower of the temperature at that time and the first-order lag temperature. When this is expressed by a mathematical expression, the following expressions (3) and (4) are obtained.

C (t) = min {A (t), E (t)} (3) D (t) = min {B (t), F (t)} (4) , E (t): 1 of the temperature detection point 3a at time t
Next delay temperature (° C) F (t): First delay temperature (° C) at temperature detection point 3b at time t The first delay temperature in the formula is the current sampling temperature A (t), B (of temperature detection point 3). t) and the temperatures C (t-Δt) and D (t-Δ) at the normal time calculated from the previous sampling temperature.
t) and the first-order lag coefficient are obtained using the following equations (5) and (6).

E (t) = ALFA × A (t) + (1-ALFA) × C (t−Δt) (5) F (t) = ALFA × B (t) + (1-ALFA) × D (t−Δt) (6) However, ALFA: First-order delay coefficient at time t Δt: Sampling period (second) Then, the first-order delay coefficient is calculated from the following equation (7).

ALFA = 1 / {1 + TAU / Δt} (7) However, TAU: first-order lag time constant at time t (seconds) This time constant is the portion where the temperature at the temperature detection point 3 is increasing. If it is longer than the duration of, the temperature change due to passage through the solidified shell rupture portion can be easily captured. When obtaining the covariance value from the above seven equations, the only constant that needs to be adjusted is the first-order lag time constant, and this value can also be easily determined from the temperature change pattern, so there is virtually no need for adjustment. .

The time-series transition of the covariance value obtained by the above method is recognized as a constraint breakout for the first time when it is larger than the preset limit value of constraint breakout occurrence, which is set in advance. However, it is possible to predict the occurrence of the constrained breakout by predicting that this solidified shell rupture portion will be a breakout immediately after passing through the mold. This restraint breakout occurrence limit value is a constant that requires adjustment, but since it can be easily estimated from the temperature change pattern, it does not substantially need to be adjusted.

Further, the influence of this covariance value due to operating disturbance will be described, and the presence or absence of erroneous detection will be described. Regarding changes in molten steel temperature and artificial changes in the molten metal level at the end of ladle replacement and artificial pouring, when the molten steel temperature decreases or the molten metal level rises, the temperature at the temperature detection point decreases and the normal temperature follows this. Therefore, the deviation becomes 0 and erroneous detection can be avoided. Also, when the molten steel temperature rises or the molten metal surface falls, the temperature at the temperature detection point rises but the slope is gentle, so However, following this, the deviation becomes 0 and erroneous detection can be avoided.

Even for unnatural fluctuations in the surface of the molten metal, when the surface of the molten metal rises, the temperature at the temperature detection point decreases and the normal temperature follows this, so the deviation becomes 0 and erroneous detection can be avoided. Also, when the level of the molten metal drops, a deviation occurs because the temperature at the temperature detection point rises sharply, but this deviation is a small value for the temperature detection point that is a distance from the molten metal surface, so the product of the deviations is calculated. The covariance value to be taken will be a small value and erroneous detection can be avoided.

Regarding the fluctuation of the drawing speed, when the drawing speed is decreased, the temperature at the temperature detection point is decreased and the temperature in the normal state follows this, so that the deviation becomes 0 and erroneous detection can be avoided. In this case, deviations occur because the temperature at all temperature detection points rises, but a sudden increase in temperature is not possible for operational safety, so the normal temperature follows this and the deviation becomes 0, and erroneous detection should be avoided. You can The covariance value thus obtained becomes large only when the solidified shell rupture portion passes, and it can be said that it is very suitable as an index of the restraint breakout.

A processing flow for predicting and preventing the occurrence of the restrictive breakout will be described with reference to FIG. In the figure, 100 is the mold temperature detected by the temperature detection rows 2a to 2d of the mold 1 and the slab drawing speed W detected by a pinch roll (not shown) of the continuous casting machine. Further, a covariance value calculation unit for calculating the covariance value N (t) every moment, 101 is the covariance value N (t) calculated by the covariance value calculation unit 100 as an index of solidified shell rupture, and the operation monitoring screen CRT To the covariance value N by comparing with a preset constraint breakout occurrence limit value T0.
When (t) is larger than the restrictive breakout occurrence limit value T0, the restrictive breakout occurrence prediction determination unit predicts the occurrence of restrictive breakout, and 102 is the covariance from the restrictive breakout occurrence prediction determination unit 101. Value N
When (t) is input, a restraint breakout prevention control unit 103 for instructing deceleration or stop of pulling out as necessary to prevent occurrence of restraint breakout, 103 is the restraint breakout occurrence prediction determination. It is an alarm device that sounds an alarm when necessary when a constraint breakout occurrence prediction result is input from the unit 101.

The processing flow of the covariance value calculation unit 100 will be described with reference to the flowchart shown in FIG. First, the mold temperature A detected at the temperature detection points 3a and 3b of the mold 1
(T), B (t) and the drawing speed W (t) of the slab measured by the pinch roll are read (S41). The time delay between the temperature detection points 3a and 3b due to the drawing speed W (t) at the read time t, that is, the time v at which a position of the slab reaches the temperature detection point 3b after passing through the temperature detection point 3a.
(T) is calculated by the equation (2) (S42).

Then, preset temperature detection points 3a, 3
Based on the sampling period Δt of b and the first-order delay constant TAU at time t, the first-order delay coefficient ALFA is calculated by the equation (7) (S43). Furthermore, the normal temperature (C
(T-Δt), D (t-Δt)), mold temperatures A (t), B at the temperature detection points 3a, 3b at the time of this sampling.
(T), based on the calculated first-order lag coefficient ALFA, the first-order lag temperature E (t) at the temperature detection point 3a at the time t and the temperature detection at the time t by the equations (5) and (6). The first-order lag temperature F (t) at the point 3b is calculated (S44,
45). Then, both of these first-order lag temperatures E (t) and F
(T) and the temperature detection points 3a and 3 read in S41.
At the mold temperatures A (t) and B (t) of b, the above (3),
The temperatures C (t) and D (t) at the normal time are calculated by the equation (4) (S46).

The normal temperature thus obtained is stored (set) (S47), and S4 is set at the next sampling.
Used in 4.

Then, the covariance value N is calculated by the above equation (1).
(T) is calculated (S48, S49), and a schematic diagram thereof is shown in (a) and (b) of FIG. That is, a time delay v between the temperature detection points 3a and 3b from the time t
(T) Mold temperature A measured at the temperature detection point 3a before
(T-v (t)) and the temperature C (t-v) in the normal state.
The deviation ΔTA (t-v (t)) of (t)) is calculated as (1-
2) the mold temperature B (t) measured at the temperature detection point 3b at time t and the temperature D in the normal state
The deviation ΔTB (t) of (t) is calculated by the equation (1-3) (S48). Next, the deviation ΔTA (t-v (t)) and ΔTB (t) are integrated, that is, the covariance value N (t) is calculated by the equation (1-1) (S49) and set in S4A. It is stored (stored) and is output to the operation status monitoring screen CRT to prompt the operator to recognize the operation status, and is output to the restrictive breakout occurrence prediction determination unit 101 and the restrictive breakout prevention control unit 102.

Next, the processing flow of the restraint breakout occurrence prediction determination unit 101 will be described with reference to the flowchart shown in FIG. This schematic diagram is shown in FIG. First, the covariance value N (t) calculated by the covariance value calculation unit 100 is read and recognized as an index of solidified shell rupture (S51),
The value is compared with a preset restrictive breakout occurrence limit value T0, and the restrictive breakout occurrence limit value T0 is compared.
It is determined whether or not it is within the constraint breakout occurrence limit value (S52), and if it is within the constraint breakout occurrence limit value, it is set (memorized) as no constraint breakout occurrence prediction (S53) and is larger than the constraint breakout occurrence limit value T0. In this case, it is set (stored) with the predictive occurrence of restrictive breakout (S54).
Then, the restraint breakout occurrence prediction determination result is output to the operation status monitoring screen CRT and is also output to the restraint breakout prevention control unit 102. Further, when it is determined that the restraint breakout occurrence prediction is made, an alarm device 103 is output.
Output to

Further, the processing flow of the restrictive breakout prevention control unit 102 will be described with reference to the flowchart shown in FIG. First, the restraint breakout occurrence prediction determination result set in the restraint breakout occurrence prediction determination unit 101 is read (S61) to determine whether there is no restraint breakout occurrence prediction information or restraint breakout occurrence prediction information. If it is determined (S62) and the information indicates that there is no predictive occurrence of a restrictive breakout, nothing is done. But,
If the information is predictive of occurrence of restraint breakout, the covariance value set in the covariance value calculation unit 100 is read, and a preset drawing speed is selected and instructed according to the magnitude of the numerical value. To do.

FIG. 7 shows an example of setting the drawing speed according to the magnitude of the covariance value. In the time series transition of the covariance value shown in FIG. 1C, the covariance value increases. It is a time series enlarged view about the inside part. In the figure, the time at which the restrictive breakout occurrence limit value T0 is reached is t0, and the times at which the covariance values are greater than T0 are T1, T2, and T3 are t1, t2, and t3, respectively. In order to prevent the occurrence of the restraint breakout, the set values for instructing the pulling speed reduction or stop depending on the magnitude of the covariance value are shown on the right side of the figure. When the covariance value N (t) at time t changes as shown in the figure when the solidified shell rupture portion passes through, an instruction to reduce or stop the drawing speed is given as follows.

When the time t is 0 ≦ 1 ≦ t0, the covariance value N
Since (t) is N (t) ≦ T0, no instruction is given regarding the drawing speed, and when t0 <t ≦ t1, T0 <N (t).
Since ≦ T1, it is instructed to reduce the drawing speed to W1, and when t1 <t ≦ t2, T1 <N (t) ≦ T2, so it is instructed to decrease the drawing speed to W2. t
When 2 <t ≦ t3, T2 <N (t) ≦ T3. Therefore, the pulling speed is instructed to decrease to W3, and t> t3.
In the case of, since N (t)> T3, an instruction to stop the withdrawal is given. The relationship between the set values W1, W2 and W3 of the withdrawal speed is set only when W1 ≧ W2 ≧ W3 ≧ 0 and each is smaller than the withdrawal speed W (t) at time t. That is, when the broken portion of the solidified shell is detected, the increase in the drawing speed is not instructed to prevent the occurrence of the constrained breakout.

There is also a method of estimating the preset covariance value and the set value of the drawing rate by the growth time of the solidified shell, but this time, based on the past experience of restraint breakout. The values are as shown below.

FIG. 8 is a vertical bending type continuous casting equipment for low carbon aluminum killed steel, casting width: 1830 mm, casting thickness: 2
82 mm, drawing speed: 1.30 m / min, an example when a slight solidified shell rupture portion passed during casting, FIG. 9 shows a casting width:
1800mm, casting thickness: 282mm, drawing speed: 1.3
This is an example when a severely solidified shell rupture portion passes during casting at 0 m / min. The upper part shows the temperature of the temperature detection points 3a and 3b and the temperature at the normal time, the middle part shows the covariance value N (t) calculated by the covariance value calculation unit 100, and the lower part shows the drawing speed. The time-series transition is shown with the horizontal axis representing the elapsed time from the start.

Then, a straight line parallel to the horizontal axis is drawn from the interruption covariance value of 500, and this numerical value of 500 is the restrictive breakout occurrence limit value T0. The mold temperature detection row is shown in FIG. 8 for the long side end of the mold (2b in FIG. 11) and for the center of the long side of the mold (2c in FIG. 11), and the distance between the temperature detection points 3a and 3b. Are all 130 mm
Met. Then, as parameters for obtaining the covariance value, the time constant TAU of the first-order lag was 50 seconds and the sampling period ΔT was 0.5 seconds. Further, the instruction of the drawing speed is based on the embodiment of the patent publication introduced in the section of the above-mentioned conventional technique.

FIG. 8 is a time-series transition when the solidified shell was fractured and the solidified shell was actually fractured, but it was dealt with by decelerating the drawing. In the figure, the temperature was around 163.0 seconds. The temperature at the detection point 3a is 168 ° C and the temperature at the normal time (about 140
C.) and a deviation of about 28.degree. C. and the time delay due to the drawing speed is 6.0 seconds, so it was delayed by 6.0 seconds to 169.0.
The temperature at the temperature detection point 3b in the vicinity of the second is 149 ° C., and there is a deviation of about 46 ° C. from the normal temperature (about 103 ° C.). The covariance value is a value of about 1288, which is obtained by multiplying the deviation between the two by about 28 ° C. and about 46 ° C., and this covariance value is 169.0.
This covariance value is maximum around the second. This temperature change is due to the fracture of the solidified shell, and when passing through the temperature detection train, the covariance value is the limit value of the constraint breakout occurrence of 50.
It was higher than 0, and it was judged that the restraint breakout occurrence was predicted.

Since the solidified shell rupture was slight, the drawing speed had to be reduced to 0.55 m / min, but when attention was paid to the instruction, the conventional judgment was 174.5.
In seconds, the solidified shell rupture portion was detected for the first time, and an instruction to reduce the speed to 0.20 m / min was issued. Therefore, the speed was reduced too much, but in the present invention, the transition of the covariance value is 5
Since a speed instruction of 0.82 m / min (W1) is output at 165.5 seconds exceeding 00 (T0) and 0.55 m / min (W2) is output at 168.0 seconds exceeding 1000 (T1), approximately 6. 5 seconds earlier and you can give a proper speed instruction,
The growth of the shell can be further promoted.

FIG. 9 is a time-series transition when the breakage of the solidified shell was detected and the solidified shell was actually broken.
At about 5 seconds, the temperature at the temperature detection point 3a is 183 ° C. and a deviation of about 40 ° C. from the normal temperature (about 143 ° C.) occurs.
Since the time delay due to the drawing speed is 6.0 seconds, the temperature at the temperature detection point 3b in the vicinity of 172.5 seconds delayed by 6.0 seconds is 178 ° C and the normal temperature (about 118 ° C) and about 60 ° C.
There is a deviation of. Deviation between these two is about 40 ℃ and about 60
A value of about 2400 multiplied by ° C is a covariance value, and the covariance value is maximum around 172.5 seconds. This temperature change is due to the fracture of the solidified shell, and when passing through the temperature detection series, the covariance value became higher than the restrictive breakout occurrence limit value 500, and it was judged as predictive occurrence of restrictive breakout. is there.

Since the solidified shell rupture was severe, the drawing speed had to be reduced to 0.20 m / min. However, focusing on the instruction, the conventional judgment was 176.0.
At the second, the solidified shell fracture was detected for the first time, and an instruction to reduce the speed to 0.20 m / min was issued, but in the present invention, the transition of the covariance value exceeds 500 (T0) to 164.5 seconds. 0.82m / min (W1), more than 1000 (T1) 1
0.55 m / min (W2), 1500 (T
Since the speed instruction of 0.20 m / min is output in 172 seconds exceeding 2), it is possible to give a proper speed instruction about 6.5 seconds earlier and further accelerate the growth of the shell.

Regarding the drawing speed instruction,
In the conventional determination method, it is necessary to use a double determination of the logic for instructing the slow speed reduction of the light pullout detected in FIG. 8 and the logic for instructing the slow speed reduction of the heavy pullout detected in FIG. However, in the case of the present invention, as described above, since it is possible to instruct to decrease the speed of pulling out and stop by the magnitude of the covariance value, only one logic is sufficient, and the calculation load of the computer can be reduced. It will be possible. Further, it is possible to set the pulling speed reduction instruction in multiple stages with only one logic.

Next, regarding the method of calculating the covariance value, which is used as an index of the inclusions involved in the initial solidification in the mold, which causes the breakout of inclusions, it is the cause of the constraint breakout. Since there are many similarities with the calculation method of the covariance value used as an index of solidified shell rupture that occurs during initial solidification in the mold, the differences between the two will be mainly described. Formula (1-1), Formula (2), Formula (5),
The equations (6) and (7) are the same, and there are two differences. One is the deviation ΔTA in the equations (1-2) and (1-3),
When calculating ΔTB, the order is changed so that the deviation becomes positive, and the other is to calculate the temperatures C (t) and D (t) at the normal time in the expressions (3) and (4). When
When calculating the deviation, the normal temperature is set to 1 at the time so that it becomes large only when there is a rapid temperature drop.
The point is that the next lag temperature is higher. When these are expressed by mathematical formulas, the following (1-2 ') formula, (1-3')
It becomes like Formula, Formula (3 '), and Formula (4').

ΔTA (t) = C (t) −A (t) ... (1-2 ′) ΔTB (t) = D (t) −B (t) ... (1-3 ′) C (T) = max {A (t), E (t)} ... (3 ') D (t) = max {B (t), F (t)} ... (4')

Also regarding the adjustment of the covariance value calculation method, the first-order lag time constant in the equation (7) is set to be longer than the duration of the portion where the temperature is decreasing at the temperature detection point 3 so that the inclusions are included. The same applies in that the temperature change due to passage can be easily captured, and since this value can also be easily determined from the temperature change pattern, there is no need to adjust it substantially.

The time series transition of the covariance value thus obtained is recognized as the inclusion property breakout for the first time when it is larger than the previously designed limit value of inclusion property breakout occurrence. , It is possible to predict the occurrence of inclusion breakout by predicting that this inclusion will fall off after passing through the mold and the thin part of the solidified shell will not be able to withstand the static pressure of molten steel and will break to breakout. it can. The adjustment of the threshold value for the occurrence of the breakout of inclusions can also be easily estimated from the temperature change pattern, and is substantially the same in that it is not necessary.

Next, the influence of the covariance value from the operating disturbance will be described, and the presence or absence of erroneous detection and the improvement points will be described. Regarding changes in molten steel temperature and artificial changes in molten metal level at the end of ladle replacement and infusion, the temperature at the temperature detection point rises when the molten steel temperature rises or falls, and the normal temperature follows this. Therefore, the deviation becomes 0 and erroneous detection can be avoided. Also, when the molten steel temperature decreases or the molten metal surface rises, the temperature at the temperature detection point decreases, but the slope is gentle, so However, following this, the deviation becomes 0 and erroneous detection can be avoided.

Even for unnatural fluctuations in the surface of the molten metal, when the surface of the molten metal descends, the temperature at the temperature detection point rises and the normal temperature follows this, so the deviation becomes 0 and erroneous detection can be avoided. Also, when the level of the molten metal rises, the temperature at the temperature detection point drops sharply, causing a deviation, but this deviation is a small value at the temperature detection point that is a distance from the molten metal surface, so the product of the deviations Since the covariance value that takes is small, false detection can be avoided.

Regarding the fluctuation of the pulling speed, however, when the pulling speed rises, the temperature at the temperature detection point rises and the temperature at the normal time follows this, so that the deviation becomes 0 and erroneous detection can be avoided. In the case of such a decrease, the temperatures at all the temperature detection points drop abruptly, resulting in deviations and large covariance values, resulting in erroneous detection. Therefore, only when the drawing speed is decreased, the temperature at the normal temperature detection point is improved so as to follow the temperature at that time, and the deviation is set to 0 to avoid this erroneous detection. The covariance value thus obtained becomes large only when inclusions pass, and it can be said that it is very suitable as an index of the breakout of inclusion physicality.

A process flow for predicting and preventing the occurrence of inclusion breakout will be described with reference to FIG.
In the figure, 100 is the temperature detection lines 2a to 2d in the mold 1.
Then, the detected mold temperature and the slab drawing speed W detected by the pinch roll (not shown) of the continuous casting machine are input, and the covariance value N (t) is calculated every moment based on this. The covariance value calculation unit 101 outputs the covariance value N (t) calculated by the covariance value calculation unit 100 to the operation monitoring screen CRT as an index of inclusions, and also sets a preset inclusion physical property breakout limit value. Compared with T0, the covariance value N
When (t) is larger than the inclusion property breakout occurrence limit value T0, an inclusion property breakout occurrence prediction determination unit for predicting the occurrence of an inclusion property breakout occurrence, 102 is from the inclusion property breakout occurrence prediction determination unit 101, the covariance When the value N (t) is input, the inclusion property breakout prevention control unit 103 for instructing the deceleration or stop of the withdrawal as necessary to prevent the occurrence of the inclusion property breakout, 103 is the inclusion property breakout prediction. It is an alarm device that sounds an alarm as necessary when a result of predicting occurrence of an intervening physical property breakout is input from the determination unit 101.

The processing flow of the covariance value calculation unit 100 will be described with reference to the flowchart shown in FIG. First, the mold temperature A detected at the temperature detection points 3a and 3b of the mold 1
(T), B (t) and the drawing speed W (t) of the slab measured by the pinch roll are read (S131). Temperature detection point 3a based on the drawing speed W (t) at the read time t
And 3b, that is, the time v at which a position of the slab reaches the temperature detection point 3b after passing through the temperature detection point 3a.
(T) is calculated by the equation (2) (S132), and the change is determined from the read drawing speed W (t) (S133).

When the drawing speed W (t) does not change or when the drawing speed W (t) increases,
Based on the preset sampling cycle Δt of the temperature detection points 3a and 3b and the first-order delay constant TAU at the time t,
The first-order delay coefficient ALFA is calculated by the equation (7) (S134). Further, the normal temperature (C (t-Δt), D stored and calculated and stored at the previous sampling is stored.
(T-Δt)), temperature detection point 3 at the time of this sampling
Based on the mold temperatures A (t) and B (t) of a and 3b and the calculated first-order lag coefficient ALFA, the first-order lag of the temperature detection point 3a at the time t is calculated by the equations (5) and (6). Temperature E
(T) and the first-order lag temperature F (t) at the temperature detection point 3b at time t are calculated (S135), (S136). Then, both of these first-order lag temperatures E (t), F (t) and the S
Mold temperature A of the temperature detection points 3a and 3b read at 41
At (t) and B (t), the temperatures C (t) and D (t) at the normal time are calculated by the above equations (3 ') and (4') (S13).
7).

On the other hand, when it is determined in S133 that the speed change of the drawing speed is decreasing, the mold temperature read this time is set to the normal temperature (S138). That is, in this case, since no deviation occurs and the covariance value N (t) becomes 0, it is possible to avoid erroneous detection while the drawing speed is decreasing. The normal temperature thus obtained is stored (set) (S139), and the next sampling is performed at S13.
Used in 5.

Then, the covariance value N is calculated by the equation (1).
(T) is calculated (S13A, S13B),
This schematic diagram is shown in FIGS. That is, the mold temperature A (tv (t)) measured at the temperature detection point 3a before the time delay v (t) between the temperature detection points 3a and 3b from the time t and the normal temperature C ( t-
The deviation ΔTA (t-v (t)) of v (t)) is calculated as
-2 ') and the temperature detection point 3 at time t
The mold temperature B (t) measured in b and the temperature D in the normal state
The deviation ΔTB (t) of (t) is obtained by the above equation (1-3 ′) (S13A). Next, the deviation ΔTA (tv)
(T)) and ΔTB (t) are integrated, that is, (1-)
The covariance value N (t) is calculated by the equation 1) (S13B),
Set (memorize) in S13C, operating status monitoring screen C
Output to RT to encourage operators to recognize the operating situation,
The output is output to the inclusion property breakout occurrence prediction determination unit 101 and the inclusion property breakout prevention control unit 102.

Next, the process flow of the inclusion property breakout occurrence prediction determination unit 101 will be described with reference to the flowchart shown in FIG. This schematic diagram is shown in FIG. First,
Covariance value N calculated by the covariance value calculation unit 100
(T) is read and recognized as an index of the inclusion (S14
1), comparing the value with a preset inclusion physical property breakout occurrence limit value T0 to determine whether it is within the inclusion physical property breakout occurrence limit value T0 (S142). In this case, it is set (memorized) as no prediction of occurrence of inclusion physical breakout (S14).
3) If it is at or above the occurrence limit value T0 of occurrence of inclusion breakout, it is set (stored) as prediction of occurrence of breakout of inclusion property (S144). Then, the judgment result of the occurrence prediction of the inclusion breakout is output to the operation status monitoring screen CRT, and also output to the inclusion breakout prevention control unit 102. Further, when it is determined to be the breakout occurrence prediction of the inclusion, the alarm device Output to 103.

Further, the processing flow of the inclusion property breakout prevention control unit 102 will be described with reference to the flowchart shown in FIG. First, the inclusion property breakout occurrence prediction determination result set in the inclusion property breakout occurrence prediction determination unit 101 is read (S151), and it is determined whether there is no inclusion property breakout occurrence prediction information or inclusion property breakout occurrence prediction information. It is determined (S152), and if the information indicates that there is no prediction of the occurrence of inclusion physical breakout, nothing is done. However, when it is the information of occurrence prediction of inclusion breakout, the covariance value set in the covariance value calculation unit 100 is read, and the preset drawing speed is selected according to the magnitude of the numerical value. To instruct.

FIG. 16 shows an example of setting the drawing speed according to the magnitude of the covariance value. In the time series transition of the covariance value shown in FIG. 10C, the covariance value increases. It is a time series enlarged view about the inside part. In the figure, the time at which the inclusion physical property breakout limit value T0 is reached is t0, and the times at which the covariance values are greater than T0 are T1, T2, and T3 are t1, t2, and t3, respectively. In order to prevent the occurrence of a breakout of inclusions, the right side of the figure shows the set value for instructing the speed reduction or stop of the drawing according to the magnitude of the covariance value. Specifically, when the covariance value N (t) at time t changes as shown in the figure when an inclusion passes through, an instruction to reduce or stop the extraction speed is given as follows.

When the time t is 0≤t≤t0, the covariance value N
Since (t) is N (t) ≦ T0, no instruction is given regarding the drawing speed, and when t0 <t ≦ t1, T0 <N (t).
Since ≦ T1, it is instructed to reduce the drawing speed to W1, and when t1 <t ≦ t2, T1 <N (t) ≦ T2, so it is instructed to decrease the drawing speed to W2. t
When 2 <t ≦ t3, T2 <N (t) ≦ T3. Therefore, the pulling speed is instructed to decrease to W3, and t> t3.
In the case of, since N (t)> T3, an instruction to stop the withdrawal is given. The relationship between the set values W1, W2 and W3 of the withdrawal speed is set only when W1 ≧ W2 ≧ W3 ≧ 0 and each is smaller than the withdrawal speed W (t) at time t. That is, when the inclusion is detected, the increase in the extraction speed is not instructed so as to prevent the occurrence of the inclusion breakout.

There is also a method of estimating the preset covariance value and the set value of the drawing rate by the growth time of the solidified shell, but this time, based on the experience of past breakout of inclusion physical properties, The values are as shown below.

FIG. 17 shows a high-charcoal aluminum killed steel using a curved continuous casting facility, which has a casting width of 1830 mm and a casting thickness of 28.
2 mm, drawing speed: 1.35 m / min, it is an example when a light inclusion passes through during casting. Fig. 18 shows a casting width: 180 using a vertical bending type continuous casting equipment for high carbon aluminum killed steel.
0 mm, casting thickness: 282 mm, drawing speed: 1.20
This is an example when a heavy inclusion passes during casting at 1.40 m / min (during acceleration). The upper part shows the temperatures of the temperature detection columns 3a and 3b and the temperature at the normal time, and the middle part shows the covariance value calculation unit 100.
In the covariance value N (t) calculated in step 1, the lower part shows the withdrawal speed, the vertical axis shows the time series transition, and the horizontal axis shows the elapsed time from the start of measurement.

A straight line parallel to the horizontal axis is drawn from the covariance value of 1000 in the middle row, and this numerical value of 1000 is the limit value T0 for occurrence of inclusion breakout. Note that the mold temperature detection row is for the center of the short side of the mold (2a in FIG. 11), and the distance between the temperature detection points 3a and 3b is 16
It was 0 mm. Then, as parameters for obtaining the covariance value, the time constant TAU of the first-order lag was set to 50 seconds, and the sampling period was set to ΔT of 0.5 seconds. Further, the instruction of the drawing speed is based on the embodiment of the patent publication introduced in the section of the above-mentioned conventional technique.

FIG. 17 is a time-series transition when the inclusion of an inclusion was detected and the inclusion was actually involved, but since it was mild, it was dealt with by deceleration of the withdrawal.
At about 9.0 seconds, the temperature at the temperature detection point 3a has a deviation of about 68 degrees Celsius from the normal temperature (about 219 degrees Celsius) at 151 degrees Celsius, and the time delay due to the drawing temperature is 7.0 seconds. .0
At the temperature detection point 3b in the vicinity of 176.0 seconds which is delayed by 2 seconds, the temperature is 105 ° C., and there is a deviation of about 40 ° C. from the normal temperature (about 145 ° C.). The value of about 2720, which is obtained by multiplying the difference between the two by about 68 ° C. and about 40 ° C., is the covariance value, and the covariance value is maximum around 176.0 seconds. This temperature change is due to the inclusions caught in the solidified shell, and the covariance value becomes higher than the inclusion physical property breakout occurrence limit value 1000 when passing through the temperature detection series, and it is judged that the occurrence of the inclusion property breakout is predicted. It is a thing.

Focusing on the drawing speed instruction, in the conventional judgment, the inclusion was detected for the first time at 180.0 seconds and an instruction to decrease the speed was issued. However, in the present invention, the transition of the covariance value is 1000 ( 0.8 at 168.5 seconds exceeding T0)
2 m / min (W1), exceeding 2000 (T1) 171.
Since the speed instruction of 0.55 m / min (W2) is output in 5 seconds, an appropriate speed instruction can be issued about 11.5 seconds earlier, and the growth of the shell can be further promoted.

FIG. 18 is a time-series transition when the inclusion is detected, the inclusion is actually involved, and the extraction is stopped because it is severe, but a breakout occurs when it is not in time. The temperature of the temperature detection point 3a is 185 ° C. in the vicinity of 171.5 seconds, which is the normal temperature (about 261
℃) and the deviation of about 76 ℃, and the time delay due to the drawing speed is 7.0 seconds, so it was delayed by 7.0 seconds 178.5.
The temperature at the temperature detection point 3b near the second is 133 ° C., which is a deviation of about 58 ° C. from the normal temperature (about 191 ° C.). The covariance value is a value of about 4408, which is obtained by multiplying the deviation of about 68 ° C. and the deviation of about 40 ° C. from each other.
This covariance value is maximum around the second. This temperature change is due to the inclusions caught in the solidified shell, and the covariance value becomes higher than the inclusion physical property breakout occurrence limit value 1000 when passing through the temperature detection series, and it is judged that the occurrence of the inclusion property breakout is predicted. It is a thing.

Focusing on the drawing speed instruction, in the conventional judgment, the inclusion was detected for the first time at 182.0 seconds and an instruction to stop was issued, but the transition of the covariance value was 1000.
0.82 m / min (W at 175.0 seconds exceeding (T0)
1), 0.5 over 171.5 seconds exceeding 2000 (T1)
5 m / min (W2), exceeding 3000 (T2) 172.
Output 0.20 m / min (W3) speed instruction at 0 seconds, and
Since the stop is instructed at 178.0 seconds exceeding 000 (T3), it is possible to give an instruction about 4.0 seconds earlier in the case of stop, and it is possible to further promote the growth of the shell.
Or it could have been possible to prevent the occurrence of breakouts.

Regarding the drawing speed instruction,
In the conventional determination method, the double determination of the logic for instructing the pulling speed reduction detected in FIG. 17 and the logic for instructing the stop of the extraction detected in FIG. 18 is used. As described above, since it is possible to instruct to reduce the speed of pulling out and to stop the pulling out according to the magnitude of the covariance value, only one logic is sufficient, and the calculation load of the computer can be reduced. Further, it is possible to set the pulling speed reduction instruction in multiple stages with only one logic.

In the above, regarding the break-out property and the inclusion property breakout, the prediction method in the case where there are two temperature detection points in one temperature detection row has been described, but it is considered how many temperature detection points should be. . In the case of one point, it will be predicted only by the deviation from the normal temperature at the temperature detection point, but if the detection point is close to the molten metal surface, false detection due to fluctuations in the molten metal surface, if it is far from the molten metal surface In this case, it becomes difficult to avoid erroneous detection due to voids caused by solidification shrinkage between the solidified shell and the mold, and sufficient detection accuracy cannot be ensured.

When there are three or more points, the deviation from the normal temperature of the third and fourth temperature detection points calculated by correcting the time delay due to the drawing speed and using the primary delay. It is only necessary to newly multiply the equation for calculating the covariance value as the third term and the fourth term. At this time, it can be detected more accurately than in the case of two points. However, there is a problem that the maintenance cost of many temperature detection points increases. In actual operation, one temperature detection point,
When the case of 2 points and 3 points was investigated, there was a problem of accuracy as described above in the case of 1 point, but there was no difference in accuracy between 2 points and 3 points. Therefore, in this predicting method, it has been found that the number of temperature detection points should be at least two for each temperature detection row. There is another method to calculate the sum of the indices instead of the product of the deviations, but in this case it is difficult to identify changes in the molten metal surface, fracture of the solidified shell and inclusions, so this method is not very suitable. Absent.

[0071]

By carrying out the present invention, it is possible to reliably predict and prevent the restrictive breakout due to solidified shell rupture and the inclusion breakout due to inclusions.

[Brief description of the drawings]

FIG. 1 (a) is a diagram showing a change in mold temperature when passing through a fractured portion of a solidified shell, (b) is a diagram showing a time delay of the mold temperature corrected, and (c) is a covariance value showing the temperature change. Figure

FIG. 2 is a diagram showing temperature detection points installed in a mold.

FIG. 3 is a block diagram of an example of a constrained breakout prediction device.

FIG. 4 is a diagram showing an operation flow of a covariance value calculation unit used for restrictive breakout prediction.

FIG. 5 is a diagram showing an operation flow of a predictive determination unit for occurrence of restrictive breakout.

FIG. 6 is a diagram showing an operation flow of a restrictive breakout prevention control unit.

FIG. 7 is an enlarged view of the time series of FIG. 1 (c) and is an explanatory view of the method for preventing the breakout of restraint.

FIG. 8 is a diagram showing changes in mold temperature, solidified shell rupture index, and drawing speed in Examples.

FIG. 9 is a diagram showing changes in mold temperature, solidified shell rupture index, and drawing speed in Examples.

FIG. 10 (a) is a diagram showing a mold temperature change when passing through inclusions, and (b) is a diagram showing a mold temperature with a time delay corrected.
(C) Diagram showing temperature change by covariance value

FIG. 11 is a diagram showing temperature detection points installed in a mold.

FIG. 12 is a block diagram of an example of a breakout prediction device for inclusions.

FIG. 13 is a diagram showing an operation flow of a covariance value calculation unit used for predicting an inclusion physical property breakout.

FIG. 14 is a diagram showing an operation flow of an inclusion property breakout occurrence prediction determination unit.

FIG. 15 is a diagram showing an operation flow of an inclusion property breakout prevention control unit.

FIG. 16 is an enlarged view of the time series of FIG. 10 (c) and is an explanatory diagram of a method for preventing the breakout of inclusions.

FIG. 17 is a diagram showing changes in mold temperature, index of inclusions, and drawing speed in Examples.

FIG. 18 is a graph showing changes in mold temperature, index of inclusions, and drawing speed in Examples.

[Explanation of symbols]

 1 Mold 2a Mold temperature detection row (center of short side of mold) 2b Mold temperature detection row (edge of long edge of mold) 2c Mold temperature detection row (center of long edge of mold) 2d Mold temperature detection row (edge of long edge of mold) 3a Upper mold temperature detection point 3b Lower mold temperature detection point 4 Broken portion of solidified shell 5 Inclusion caught in solidified shell

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Hatano Imasayu, Oita City, Oita Prefecture, Oita City, Nishinosu 1st Nippon Steel Co., Ltd. Oita Steel Works (72) Inventor, Koichi Dobashi, Oita City, Oita City, Nishinozu Made in Japan Oita Works

Claims (3)

[Claims] 1. A mold of a continuous casting machine is provided with a plurality of temperature measuring elements in the casting direction to measure the mold temperature, and when the measured value rises, the first-order lag temperature from the rising start point is measured. The temperature difference is determined for each temperature measuring element in the casting direction from the measured mold temperature and the first-order lag temperature, and the temperature difference at the temperature measuring element position on the upstream side in the casting direction and the upstream temperature measuring element position The integrated value with the temperature difference at the downstream temperature measuring element position after the moving time of the slab obtained from the distance between the downstream temperature measuring element positions and the withdrawing speed of the slab is calculated, and this integrated value is a predetermined value. A predictive method of a constrained breakout in continuous casting, characterized in that when the above is reached, breakout due to solidified shell rupture occurs. 2. A continuous casting machine mold is provided with a plurality of temperature measuring elements in the casting direction to measure the mold temperature, and when the measured value drops, the first-order lag temperature from the starting point of the drop is measured. The temperature difference is obtained for each temperature measuring element in the casting direction from the first-order lag temperature and the measured mold temperature, and the temperature difference at the temperature measuring element position on the upstream side in the casting direction and the upstream temperature measuring element position The integrated value with the temperature difference at the downstream temperature measuring element position after the moving time of the slab obtained from the distance between the downstream temperature measuring element positions and the withdrawing speed of the slab is calculated, and this integrated value is a predetermined value. A method of predicting a breakout of inclusion physical properties in continuous casting, characterized in that when the above is reached, breakout due to inclusions is generated. 3. The continuous process according to claim 1, wherein when the integrated value exceeds a predetermined value, the withdrawal speed of the slab is adjusted according to the size of the integrated value. A method for preventing breakout due to solidified shell rupture and inclusions in casting.