JPH0229419B2 - RENZOKUCHUZOIGATANIOKERUCHUZOKONOHADANKENSHUTSUHOHO - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to continuous casting, and more particularly to detection of fractures in cast steel in molds.

A major factor that impedes productivity, safety, and equipment maintenance in continuous casting is breakage of the solidified shell (hereinafter simply referred to as shell) of steel cast in a mold or breakout due to large inclusions on the surface of the shell.

Conventionally, the surface temperature of the steel billet that exits the mold is detected, and if the temperature is abnormally high, there is a risk of breakout, so the cooling of that part is strengthened. Since the problem is downstream, breakout prevention measures may not be taken in time. On the other hand, if a fracture in the cast steel that could lead to a breakout could be detected while it is still in the mold, the surface of the fracture would be sufficiently cooled to form a shell thick enough to prevent breakout. Until then, withdrawal can be stopped or slowed down to prevent breakout.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a fracture detection method for detecting fractures in a cast steel shell while it is in the mold, which may result in breakout.

In order to achieve the above object, in the present invention, the temperature T of the surface of the cast steel inside the mold wall is measured at at least two points above and below in the drawing direction.

The temperature T inside the mold wall, which causes a temperature change corresponding to the change in temperature Tc on the surface of the cast steel inside the mold, showed fluctuations as shown in FIG. 1a when measured with a thermocouple. The embedded position of the thermocouple is 20 mm or more and 700 mm or less below the molten metal level in the mold, and 1 mm or more and 30 mm or less on the mold inner wall surface. Peak P ₁ in Fig. 1a indicates that the generated shell seizes at the surface of the mold and ruptures due to the pulling force from the existing shell below and vibration of the mold, and the molten steel flows directly into the mold inner surface from this fracture opening. The temperature T rises rapidly due to contact, and if this is left unattended, the fractured part will be formed by the lower end of the fractured baked shell and the upper end of the separated shell below it, and as a result, even if shell formation progresses, the relative positions will always remain the same. The separated shell moves downward, and the baked shell remains in the mold together with the vibration of the mold, so although it moves downward, the shell formation at the opening does not progress and breaks after leaving the mold. Since there was a high possibility that this would occur, the pinch roll was stopped at that point, and the extraction was stopped or decelerated for about 30 seconds. When drawing was then carried out at the original steady speed, no breakout occurred.

In addition, the temperature drop shown in Figure 1b may occur, but this temperature drop _P2 is due to the fact that the powder entrained from the mold surface or the entrained slag from the tundish cannot float to the surface and solidify as they are on the shell surface. This is due to so-called large inclusions.
At this time as well, when the withdrawal was stopped or decelerated for about 30 seconds and then the withdrawal was continued at the original steady speed, no breakout occurred.

In addition, when measuring the temperatures Tu (upper) and T _D (lower) at two points, upper and lower, inside the wall of a predetermined area below the molten metal level in the mold in the drawing direction of the cast steel, it is possible to detect if there is a break in the shell or large inclusions in the mold. If there is, it moves downward as it is pulled out, so that Tu and T _D change as shown in Figure 1c. That is, temperature fluctuations P ₁₁ and P ₂₁ first appear in the upper high temperature part (Tu), and as the drawing progresses (time passes), corresponding fluctuations appear in the lower low temperature part (T _D ), and the temperature fluctuations P 11 and P 21 appear in the lower temperature part (T D ) for a given period of time. ) and the lower temperature (T _D ).

When the hot water level in the mold falls below the thermocouple position, as shown in Figure 1d, as the hot water level falls, the temperature T first rises and then falls, causing a reversal in Tu and T _D. .

Based on this knowledge, the present invention
Detect abnormalities in cast steel by reversing Tu and T _D. In the case of shell rupture (P ₁₁ , P ₁₂ in Fig. 1c), the fluctuation
Both P ₁₁ and P ₁₂ are in the high temperature direction, and if there are large inclusions on the shell surface, the fluctuations P ₂₁ and P ₂₂ are both in the low temperature direction, so the smooth value of Tu or T _D (simple average value, weighted average value, envelope value, etc.)
Alternatively, it is possible to distinguish between abnormalities in the shell fracture and shell surface inclusions based on whether T _D is high or low.

In shell fracture of cast steel in the mold, 1a
As shown in the figure, the temperature T rises rapidly and drops sharply at the shell surface inclusions. On the other hand, in level fluctuations, the temperature changes are gradual. Therefore, in order to distinguish from this, it is possible to detect an abnormality by at least detecting a sudden rise or fall in temperature. In addition to determining the reversal of Tu and T _D , the present invention detects the temperature change rate ΔT/Δt or the amount of change. to judge the breakage. In a preferred embodiment, the determination is made using the rate of change ΔT/Δt. If the rate of change ΔT/Δt is out of the predetermined range and each change is positive, it is a shell fracture of the in-mold cast steel, and if the change is negative, a large intervening part exists on the surface of the shell. In general, rises and falls in the molten metal level are caused by changing the casting speed or replacing the ladle, and since this is a known part of the operation and can be detected separately, it is not necessary to distinguish it from level fluctuations. Fracture can also be determined from temperature fluctuation characteristics and operating conditions.

FIG. 2a shows an example of the configuration of an apparatus for determining breakage according to the present invention, and FIGS. 2b and 2c show the flow of the determination operation. Hereinafter, a fracture detection method according to an embodiment of the present invention will be explained with reference to the apparatus configuration shown in FIG. 2a. The fracture detection and determination device 10 includes a central processing unit (CPU) 11, a semiconductor read-only memory (ROM) 12, and a semiconductor read/write memory (RAM).
13 and an input/output port 14;
, a recorder 20 for recording the temperature T of each part of the mold, and a casting operation panel computer (higher than 10) 3
0, an alarm display device 40, an input keyboard 50, and an element selector 60 are connected, and the output of the amplifier 70 is given to the A/D conversion input terminal of the CPU 11. An outline of the mutual operation of the combinations of the devices shown in FIG. 2a is as follows.

In other words, in addition to casting speed, speed change, and ladle replacement, casting conditions including judgment constants (described later), operational data, and a set command instructing the start of fracture detection operation, as well as one pulse every time the cast steel moves a predetermined distance, A sampling clock pulse CLs, which is emitted at a certain rate, is applied from an operation panel computer 30 to a break detection/judgment device (hereinafter abbreviated as microcomputer) 10. The ROM 12 of the microcomputer 10 stores program data for executing the breakage detection/judgment operation described in detail below, and based on this data, the microcomputer 10 updates each part of the input/output port 14 and the register (RAM 13) when a set command arrives. Initialize (clear) the temporary data storage address of
However, each time CLs arrive, n/2 upper thermocouples 80 ₁ , 80 ₃ ...80n _- are distributed at the same height relatively above the mold wall and distributed around the mold. ₁ and upper thermocouples 80 ₁ , 80 ₃ ... 80n _-1 circling the mold at the same height relatively below the mold wall.
The detected temperatures Tu (upper thermocouple) and T _D (lower thermocouple) of n/2 lower thermocouples 80 ₂ , 80 ₄ . It is stored in averaging registers (n pairs, which refer to RAM memory addresses) associated with each pair. In addition, the upper thermocouple 80 ₁ ,
80 ₃ ...80n _-1 and the lower thermocouple 80 ₂ , 80 ₄ ...80n
(80 ₁ , 80 ₂ ),
(80 ₃ , 80 ₄ ), ... (80n _-1 , 80n) are considered to be a pair.

Each time CLs arrive, these n (n/2
The temperature detected by the thermocouple of
Break detection is started after m CLs have arrived (that is, after m temperature data have been read and stored in time series for each thermocouple).

The detected temperatures of the n thermocouples are determined by first specifying the element selector 60 to 80 ₁ input connection when CLs arrives, converting the analog signal at the A/D conversion input terminal to digital data and assigning it to 80 ₁ . Store the detected temperature in the averaging register, then designate the element selector 60 to ₈₀₂ input connection, store the detected temperature of ₈₀₂ in the averaging register assigned to 802, and repeat this _process up to 80n. This is done in split scan format.

After storing m pieces of data for each thermocouple in this way, each time CLs arrives, it is determined whether Tu≧ _TD and ΔT/Δt are determined for each pair of thermocouples. is output to the host computer 30 and alarm display device 40, and the oldest data in the averaging register is rewritten to new data, and the average value _i+n-1 〓 ⁱ Tu/m, _i+n-1 〓 ⁱ T _D /m is updated, and the temperature data is provided to the recorder 20 and the operating computer 30 in association with the thermocouple.

Next, the breakage detection operation of the microcomputer 10 will be explained in more detail with reference to FIGS. 2b and 2c. In addition, in these drawings, one pair (for example, 80 ₁ ,
Although the abnormality detection operation is shown only for the thermocouple in 80 ₂ ), the same operation is performed for each thermocouple sequentially in time series starting from the arrival of one pulse of CLs.

First, referring to FIG. 2b, when the set command arrives, the microcomputer 10 initializes (clears all registers & clears input ports), reads casting conditions and operation data, and sets judgment reference data Ku, Ku ₁ to
Takes in Ku ₄ , Kd, K _D1 to K _D4 .

Then, each time CLs arrives, the temperature T is read and sequentially written to the averaging register, and when m data for each thermocouple has been stored, their average value _i+n-1 〓 ⁱ
Tu/m, _i+n-1 〓 ⁱ T _D /m (simply ΣTu/m, ΣT _D /
m) in the average value register (steps above). Then, wait for the arrival of CLs, and when CLs arrives, read the detected temperature Tu of the upper thermocouple (80 ₁ ) and the detected temperature T _D of the lower thermocouple (80 ₂ ), and check whether Tu≦ _TD (YES). See if it is NO. If YES, "1" indicating that the temperature has been reversed is stored in the reversal register, and a count value 1 is stored in the reversal count register. Then, each time CLs arrives, the inversion count register is counted up by 1, and if Tu≦T _D , it is determined that it is abnormal, and T _D > ΣT _D /
If m, it is regarded as a fracture abnormality, and if T _D ≦T _D /m, it is regarded as an inclusion abnormality. If Tu≦T _D is not satisfied at least once again after detecting Tu≦T _D once and before 3 pulses of CLs arrive, it is assumed to be noise or normal fluctuation, and the inversion register and inversion count register are cleared. (The above is the reversal abnormality determination step). If no abnormality is determined in the step, then check T-ΣTu/m≧Ku and if it is YES, T is abnormally high, so store 1 in the Up register and set T to the first abnormal temperature. T ₁
Memory as UpT ₁ register. in this way
After storing 1 in the Up register, each time CLs arrives, T is sequentially stored in the UpT ₂ register and UpT ₃ register, and the contents of the Up register 1 are changed to 2,
3 (step) and the content of the Up register is 3, it is determined that T ₂ −T ₁ ≧Ku ₁ and T ₃ −T ₂ ≧Ku ₂ . If both of these are YES, it is determined that there is a shell rupture (rupture abnormality) because T rises rapidly over two periods of CLs (step). If either one is NO, at most one period of CLs is T.
has not increased, so we wait for the arrival of CLs,
Store the new T in the UpT ₄ register as T ₄
After storing 4 in the Up register, T ₂ −T ₁ ≧
Check whether at least two of Ku ₁ , T ₃ −T ₂ ≧Ku ₂ , T ₄ −T ₃ ≧Ku ₃ are YES, and if YES, it is determined that the shell has broken, and if NO, T is not a continuous rapid increase (the above is a step), search for Ti where Ti = ₂ to ₄ -ΣTu/m≧Ku, rewrite it to the UpT ₁ register, and reduce the contents of the Up register by this shift. This means that the initial T ₁ is T ₁ −ΣTu/m≧Ku, but
This judgment has not been made for T ₂ to _{T 4} , so
This is because the contents of the UpT ₁ register are rewritten to the one whose T is higher among the newly entered T ₂ to T ₄ and which is temporally close to T ₁ , and from now on, the ΔT/Δt will be determined based on this. (The above steps). When it is determined that the shell is broken, jump to Figure 2c, refer to the input operation data,
When there is no room for a rapid temperature rise, a shell rupture signal is set as an output. An alarm is issued by the device 40, and the operation panel computer 30 sets an instruction to stop or decelerate the extraction on the operation panel if it determines from the information it has that there is a high possibility of shell rupture. In response to this, the operator stops or decelerates the extraction. After a predetermined stop or deceleration time, when the operator sets the original steady speed withdrawal start, a set command is given from 30 to 10. If the input operating condition causes T fluctuation, the microcomputer 10 outputs and sets a shell breakage preliminary signal. When the microcomputer 10 outputs the shell rupture, it waits for a set command.
When the shell breakage preliminary signal is output, the process returns to initialization and automatically returns to the detection/judgment operation.

Detection of inclusions based on ΔT/Δt is also performed in the same manner as shell breakage detection in the step described above.
However, when there are inclusions, the temperature T drops rapidly as shown in FIG. 1b. Therefore, in the inclusion detection judgment (Fig. 2c), the abnormality judgment of ΔT/Δt is started when ΣTu/m−T≧K _D , that is, when T has decreased abnormally, and the same procedure as in the step is performed. at the next step
Every time CLs arrive, T (T ₁ to _{T 4} ) is sequentially memorized, and T ₁ −T ₂ ≧K _D1 , T ₂ −T ₃ ≧K _D2 and T ₃ −T ₄
If at least two of ≧K _D2 are YES, it is determined that there is an inclusion. In other respects, the detection and determination of inclusions is the same as the detection and determination of shell breakage, and differs from the detection and determination of shell breakage in that if ΔT/Δt is negative and is out of a predetermined range, it is considered that an inclusion is present.

If no abnormality (shell breakage, presence of inclusions) is detected, replace the oldest temperature data in the averaging register with the temperature data imported this time, calculate the average values ΣTu/m and ΣT _D /m, and calculate the average value. Rewrite the contents of the register.

In the above embodiment, the sampling period (Δt
The reason that pulses synchronized with the drawing of the cast steel is that the movement of the abnormal point (shell breakage, presence of inclusions) is synchronized with the movement of the cast steel, and therefore the speed becomes large. This is because the speed affects ΔT/Δt because the temperature changes faster.

By using the sampling pulse as the extraction synchronization pulse as in the above embodiment, temperature inversion abnormalities will not be overlooked and ΔT/Δt will not be affected by speed. Moreover, Ku, K _D , Ku ₁ , ~ Ku ₄ , K _D1 ~ K _D4 are
The value corresponds to the casting conditions. For example, if the casting speed differs, the temperature at the same point in the mold changes, so it is preferable to set Ku and K _D to different values.
is provided to the microcomputer 10 in accordance with the casting conditions.

In the above embodiment, even if Tu≦T _D once becomes Tu≦T D, Tu≦T _D again during three pulses of CLs.
Unless this happens, it will not be considered abnormal. In the case of an abnormality, Tu≦ _TD continues for several pulses or more, so even if this is done, the abnormality will not be overlooked, and the probability of erroneously determining noise or the like as an abnormality is reduced.

Furthermore, in the above embodiment, the steps,
Monitoring for temperature inversion abnormality is started when the temperature T deviates from the average value by a predetermined value or more. This is to allow the determination of shell rupture or the presence of inclusions to occur after the temperature rises or falls to some extent.
This prevents erroneously regarding the error (see figure) as an abnormality. The temperature increase or decrease due to shell breakage or inclusions will continue for several tens of seconds at normal casting speeds, so if the sampling period is several hundred milliseconds to several seconds, it will continue for several cycles or several tens of cycles. When referring to four consecutive cycles _of data T ₁ to _{T 4} as shown in _FIG .
T ₃ < T ₄ or T ₁ > T ₂ > T ₃ > T ₄ , but the temperature T
Since the reading is at one point in a very short period of time, if the temperature at the peak or valley of the steady vibration is read by chance, ΔT/Δt will not always be equal to or higher than the predetermined value between T ₁ and T ₄ . Therefore, if ΔT/Δt for at least two periods is above a predetermined range, it is considered to be abnormal. Even if no abnormality determination is made in four cycles, the abnormal temperature rise or fall continues, and the determination is continued with some data updated, so the probability of overlooking an abnormality is extremely low.

As described above, in the present invention, shell rupture detection is started when the temperature at the upper and lower two points inside the mold wall is reversed, and the rate of increase (or the amount of change) of the measured temperature T _D at the lower part is started.
Since breakage of the shell is detected when the shell breakage point exceeds a predetermined value, a breakout can be prevented by taking protective measures such as stopping pulling out before the shell breakage point passes through the mold. By increasing the temperature sensor arrangement density in the circumferential direction of the mold wall, shell fractures on the entire surface of the cast steel can be detected in detail.

[Brief explanation of drawings]

Figures 1a and 1b are graphs showing the results of measuring the temperature change inside a predetermined wall below the molten metal level in the mold at one location, and Figures 1c and 1d at two locations, upper and lower. Figure 1a shows when the shell is broken, Figure 1b shows when large inclusions are mixed in on the shell surface, Figure 1c shows when the shell is broken and large inclusions are mixed in on the shell surface, and Figure 1d shows when the molten metal level drops. show. FIG. 2a is a block diagram showing the configuration of one apparatus for implementing the present invention, and FIGS. 2b and 2c are flowcharts showing the fracture detection operation thereof. 10: Fracture detection and determination device.

Claims (1)

[Claims] 1. The temperature inside the inner wall located below the molten metal level in the continuous casting mold is measured at predetermined positions in the upper and lower parts in the casting steel drawing direction, and the measured temperature in the lower part is determined.
T _D is greater than or equal to the measured temperature Tu of the upper part and greater than a predetermined temperature difference, and the measured temperature T _D of the lower part is
1. A method for detecting fractures in cast steel in a continuous casting mold, comprising detecting fractures in a solidified shell of cast steel at a point in time when an increasing rate of change and/or an amount of change in is equal to or greater than a predetermined value. 2. Measure the temperature inside the inner wall located below the hot water level in the continuous casting mold at predetermined positions in the upper and lower parts in the casting steel drawing direction, and calculate the measured temperature in the lower part.
A point in time when T _D is greater than or equal to the measured temperature Tu of the upper part and greater than a predetermined temperature difference, and the measured temperature Tu of the upper part and the lower part, the rate of increase and/or the amount of change in T _D is greater than a predetermined value. A method for detecting fractures in cast steel in a continuous casting mold, comprising detecting fractures in a solidified shell of cast steel.