JPH0620627B2 - Slag outflow prevention method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a method for preventing the outflow of molten slag from a ladle in a continuous casting facility or the like.

[Conventional technology]

It is important for quality assurance of the slab to avoid mixing of slag in the continuous casting operation.

In other words, slag (slag) floats on the molten steel surface in the ladle to keep the molten steel warm and prevent air oxidation, but at the end of pouring molten metal from the ladle to the tundish, The slag flows down following the outflow. If this slag is poured into the tundish, it will not only deteriorate the quality of the cast slab as a non-metallic inclusion, but also cause damage to the refractory lining of the tundish.

Therefore, various attempts have conventionally been made to automatically detect outflow of slag in order to prevent mixing of molten slag into the tundish.

For example, in addition to the visual determination method, an impedance change measuring method (Japanese Patent Laid-Open No. 55-97847, etc.), a vibration measuring method (Japanese Patent Laid-Open No. 57-112964), a method of detecting radiant energy in the microwave band from a nozzle (special The method for measuring the brightness of Japanese Patent Application No. Sho 59-80692, which is related to the present applicant, or the eddy current method disclosed in Japanese Patent Laid-Open No. 53-7359 is known.

The eddy current method (Japanese Patent Laid-Open No. 53-7359) is, as shown in FIG. 1, between the brick 1a and the iron shell 1b on the side wall 1A of the ladle 1,
An exciting coil Ca and a receiving coil (detecting coil) Cb are provided at almost the same level, an alternating current of about 100 Hz is applied to the exciting coil Ca to generate an eddy current, and on the other hand, a change in the eddy current is detected as a voltage change in the receiving coil Cb. Then, after the received signal is amplified and phase detected, the molten steel level is captured.

To further explain this in detail, when the molten steel level gradually decreases as shown in Fig. 3 (a) → (d), initially (Fig. (A)), the eddy current flows along the side wall of the ladle, Surface is coil Ca, C
At b level or below (Figs. b, c),
Eddy current is reflected in the air or on the surface of the slag and the receiving coil Cb
At the time when the molten steel has almost disappeared (Fig. D), the ratio toward the receiving coil Cb is small and the molten steel is mainly transmitted through the bottom wall. Examining the voltage change in the receiving coil Cb based on the change in the flow mode of the eddy current associated with such a level change is as shown in the right side of FIG. In this example, the width in the height direction is 500.
A coil of mm is installed at the center of the pan at a height of 400 mm from the bottom of the pan (hence, the upper end of the coil is 650 mm and the lower end is h = 150 mm).
It can be seen that an inflection point where the voltage sharply decreases near mm is generated. As a result, the residual steel level within the range of l = 0 to 150 mm can be detected, and the range of l = 0 to 50 mm is particularly excellent in detection accuracy.

[Problems to be solved by the invention]

However, when the present inventors conducted a charge test 61 times using a 250-ton ladle, the following was found. That is, the accuracy of detecting that the residual steel level has reached 50 mm above the sliding nozzle based on the above eddy current method is 5 times as shown in FIG. In terms of accuracy, the remaining steel amount is 0.55 tons on average, 3σ
(Probability 99.7%) = 1.65 tons. The residual steel level accuracy was ± 12.5 mm. This is unsatisfactory in terms of yield rate, and it is desirable that the residual steel amount <± 0.5 ton and the residual steel level <± 4 mm.

When the above test results were examined, it was found that the cause of the deterioration of accuracy was the metal attached to the side wall or the corner portion between the side wall and the bottom wall of the ladle 1 shown by the symbol M in FIG. . In fact, as shown in FIG. 6, there is no adherence of metal in the number of ladles (charges) of 9 times, and the detection signal shows a normal change. If there is adhesion, the voltage-remaining steel weight correlation deviates to the heavier weight side, and as a result, an erroneous determination of, for example, 0.32 ton is caused by the detection delay.

Therefore, an object of the present invention is to provide a molten metal outflow prevention method capable of appropriately determining the outflow of molten sludge (slag) and thus reliably preventing the outflow of molten sludge.

[Means for solving problems]

The first invention method for solving the above problems is a method for preventing the outflow of slag accompanying the injection of molten steel into the tundish from the outflow hole of the ladle by the closing operation of the partition means provided therebetween. , A pair of coils, one of which is an exciting coil and the other of which is a receiving coil, is provided at each of the position of the bottom wall of the ladle on the side of the molten steel outflow hole and the position of the bottom wall on the side opposite to the side of the molten steel outflow hole. , An eddy current is generated from the excitation coil, the eddy current is detected by the receiving coil to detect the eddy current change, and the residual steel level is judged. If the residual steel level reaches a predetermined level, the partition means is operated. It is characterized by closing and stopping the injection into the tundish.

The second invention method is a method for preventing the outflow of slag accompanying the injection of molten steel into the tundish from the outflow hole of the ladle by the closing operation of the partition means provided therebetween, in which the side wall of the ladle is provided. A pair of coils, one of which is an exciting coil and the other of which is a receiving coil, is provided at each of the molten steel outflow hole side position and the position opposite to the molten steel outflow hole side of the bottom wall, and the eddy current from the exciting coil is generated. It is generated, this is detected by the receiving coil and the change in eddy current is captured to determine the residual steel level,
When this residual steel level reaches a predetermined level, the eddy current method that closes the partitioning means to stop the injection into the tundish, and the change in the brightness of the outflow of molten metal from the ladle are detected. When there is a decrease in brightness more than a certain value, the partitioning means is closed and injection into the tundish is stopped together with the brightness measurement method, and the amount of residual steel in the ladle is greater than the specified amount. At the stage, the eddy current method is given priority, and at the stage when the amount is less than a predetermined amount, the luminance measuring method is given priority to prevent spillage.

[Action]

According to the present invention, in the eddy current method, the exciting coil and the receiving coil are conventionally arranged only on the side wall, but in addition to this, the coil is added to the bottom wall, or one of both coils is formed on the side wall and the other is formed on the bottom wall. It is arranged only in the. The inventors of the present invention have investigated the adhesion state of the metal in many charges and found that although the metal frequently adheres to the side wall of the ladle or the corner between the side wall and the bottom wall, the flat part of the bottom wall It was found that the metal was difficult to attach to. Therefore, if one of the exciting and receiving coils is placed in the flat portion of the bottom wall, it is considered that it will not be affected by the inclusion of metal, and when the experiment was actually repeated in that manner, favorable results were obtained as expected. Was obtained.

However, if it is attached to the bottom wall of the pan, the detection sensitivity will be lower than that of the conventional example as described later. However, even if the detection sensitivity decreases, the hot water can be detected accurately and with a sensitivity higher than the minimum required in practice. Therefore, the edge of the hot water can be reliably detected without being affected by the adhesion of the metal.

On the other hand, in the second invention, the Japanese Patent Application No. 59-80 is used together with the eddy current method.
Luminance measuring method represented by No. 692 is used together. Explaining the reasons and advantages of this combination, the luminance measurement method is
It detects the time when the steel flowing out from the ladle disappears and only the slag (Noro) begins to flow out. Therefore,
The realization of non-oxidative casting is not perfect, and the inclusion of slag is inevitable. In this sense, for high-grade steel such as Ca-Si treated material, a residual steel level meter based on the eddy current method that can predict the inclusion of slag is required. Further, as shown in FIG. 7 for reference, when the alumina adhered to the sliding nozzle (S / N) dropped when the sliding nozzle was opened and closed, a change in brightness was caused and the slag flowed out. The rate of judgment and erroneous detection reaches, for example, 12.6%. Furthermore, in the brightness measurement method, the brightness changes greatly for about 4 seconds when the sliding nozzle is opened and closed. Therefore, the measurement site for these 4 seconds is ignored, and the average brightness for the previous 4 seconds and the current brightness are ignored when measuring the brightness. Compare the values with
When there is a decrease in brightness above a certain value, it is determined that the slag has flowed out. Therefore, when the slag flows out within 4 seconds after opening and closing the sliding nozzle, the slag outflow is overlooked, and as shown in FIG.
For example, 11.7% is missed.

On the other hand, the eddy current method using a residual steel level meter, which will be described later, is not affected by the fall of alumina due to the opening / closing of the sliding nozzle, does not take the neglected time associated with opening / closing the nozzle, and constantly monitors the residual steel level. Is what you are doing. However, the detection delay due to the adhesion of metal is unavoidable as described above, and the ratio is, for example, 8.
Although it reaches 5% and the residual steel level can be predicted,
Due to its nature, it is not suitable for quick response.

Therefore, in the present invention, both methods are used in combination in order to maximize the advantages of each method while compensating for the difficulties of the above methods. That is, until the amount of residual steel in the ladle reaches a predetermined amount, for example, 0.5 ton, the residual steel level is continuously monitored by prioritizing the eddy current method. The law is given priority. Considering the advantages of this method again with reference to FIG. 7, (1) the eddy current method is prioritized up to a residual steel amount of 0.5 tons.
It is possible to prevent erroneous detection due to alumina falling in the range of up to 4 tons. (2) The slag outflow (11.7%) within 4 seconds after opening and closing the sliding nozzle by the luminance measurement method indicates that the amount of remaining steel is at a specified depth. Once it is detected, it can be prevented by directly closing the sliding nozzle. Further, (3) the detection delay mainly due to the metal adhesion by the residual steel level meter can be prevented by prioritizing the luminance measuring method when the amount becomes 0.5 tons or less.

When performing operations, when casting high-grade steel, the residual steel level meter is used to predict the residual steel level by mixing the slag, and the injection is terminated. Infusion to end. On the other hand, in the casting of ordinary steel, the residual steel level meter continues to be monitored with a residual steel level meter up to 0.5 tons, for example, even if the optical fiber outputs slag, it is ignored until 0.5 ton. After getting below tons,
If it is determined that the optical fiber has flown out of slag, it is determined that the steel has flowed out, and the injection into the tundish is completed.

[Specific Examples of Invention]

 The present invention will be described in more detail below.

FIG. 1 shows the arrangement of coils on the ladle 1 according to the present invention. According to the present invention, for example, the receiving coil Cc is shown.
Is placed on the bottom wall 1B of the ladle 1. Although the illustrated example is an example in which the exciting coil Ca and the receiving coil Cb are arranged on the side wall 1A, the receiving coil Cc is also provided on the bottom wall 1B.
The receiving coil Cb can be omitted. However, in terms of accuracy, it is preferable to leave the receiving coil Cb. Further, an exciting coil may be provided on the bottom wall 1B and a receiving coil may be provided on the side wall 1A. The exciting coil Ca is preferably provided on the sliding nozzle side, and the receiving coil Cc is preferably provided on the opposite side to the sliding nozzle, and its position is d.
= 0.5 to 1.5 m is desirable.

In the present invention, the reason why it is preferable to arrange the receiving (or exciting) coil on the bottom wall 1B of the ladle 1 is as follows. Fig. 8 shows how the depth of the remaining steel on the sliding nozzle and the electromagnetic field strength of the receiving coil change depending on the position of the receiving coil and the presence or absence of adhered metal. Fig. 9 shows the enlarged main part. It is a figure. In this case,
This is an example of attachment at a position of h = 300 mm. First, focusing on FIG. 8, h = 150 mm and h = 400 in the case of Ca + Cb.
Compared with mm, the latter has a higher electromagnetic field strength and a steeper gradient, so the detection accuracy is higher. And at h = 400mm, it doesn't really depend on the presence or absence of adhesion of metal. On the other hand, when h = 150 mm, the detection sensitivity is low, and especially when metal is attached, the gradient is gentle in the state where the remaining steel depth is shallow. Bullion is usually h = 3
Since it adheres to a height range of about 00 mm, in the case of Ca + Cb, it is desirable to dispose at a position of 300 mm or more, preferably 350 mm or more (150 mm in the conventional example).

Next, paying attention to FIG. 9, (I) to (IV) of FIG. 9 are examples in which the receiving coil Cc is arranged on the pan bottom, and the height position h of the exciting coil Ca is changed to 150 mm and 400 mm, and It shows how the sensitivity changes depending on the presence or absence of metal adhesion.

Considering this, it is understood that when the receiving coil Cc is installed on the bottom of the pan, the sensitivity is lowered in any case ((I) to (IV)) as compared with the case of Ca + Cb. In addition, the installation position of the excitation coil Ca is h = 40 compared with the case where h = 150 mm.
It can be seen that the sensitivity is higher when the thickness is 0 mm and the sensitivity is higher when the metal is not included. Therefore, also in this case, the exciting coil Ca has, for example, h = 400.
It is desirable to install it at a high position such as mm. However, in the case of detecting slag outflow at a tilt angle of 1.3 ° in a ladle with a capacity of 250 tons, the average residual steel at the time of detection is about 3.4 tons, so h = 400 mm with a bare metal (I
II) If you pay attention to the line, even if you consider the nozzle range ±
An accuracy of 0.6 ton (3σ-probability 99.7%) is obtained. The residual steel level accuracy is ± 5 mm. This precision is a satisfactory precision when considering practicality.

In this way, it is effective to provide one of a pair of exciting coil or receiving coil on the bottom wall of the ladle in order to deal with the detection delay due to the adhesion of metal, and even if the coil is provided on the bottom wall, It can be seen that sufficient accuracy can be obtained even if gold is attached.

FIG. 2 shows a main control system of the continuous casting equipment. Molten steel from the ladle 1 is sliding gate 2 which constitutes a sliding nozzle which is hydraulically opened and closed through an outflow hole of the ladle 1. Through the ladle nozzle 3, and is made to flow down into the tundish 5 isolated from the outside by the seal cover 4. Molten steel from the tundish 5 is injected into the mold 7 through the immersion nozzle 6 and is drawn downward while being cooled therein.

Here, facing the outflow from the ladle 1, there is provided a slag outflow detector 8 using an optical fiber based on a brightness measuring method described later in detail. Further, in order to operate the optimum residual steel level prediction system 10, in addition to the signal from the slag outflow detector 8, a signal from the residual steel level meter 9 based on the coils Ca, Cb, Cc, a tundish weight signal, and a mold width signal. , A casting length signal is provided from each detector (known per se). Further, the illustrated information is given from the continuous casting process computer (CC pro computer) 11. The optimum residual steel level prediction system 10 gives a ladle tilting instruction and a slide gate opening / closing command based on each signal and information.

Details of the slag outflow detector 8 described above are shown in FIG. That is, facing the molten steel outflow MF, the protective pipe 2
An optical fiber 21 is provided inside the optical fiber 0 to stare at the outflow MF. In addition, an inert gas for purging for purging foreign substances that cause a decrease in accuracy is blown out from the tip through the protective tube 20. In the optical power meter 22, the optical signal from the outflow MF passing through the optical fiber 21 first passes through the filter 22a, is given to the brightness detection sensor 22c via the condenser lens 22b, and is applied to the above-mentioned prediction via the amplifier 22d. Input to the system 10. In this case, the optical power meter 22 measures the average luminance within the circular visual field from the optical fiber. A filter 22a is provided in front of the brightness detection sensor 22c to cut off incident light other than the spectrum of about 0.8 micrometer. In such a brightness measuring method, when the molten steel flow is replaced with the slag outflow, the brightness is increased, whereby the slag outflow can be directly detected.

The brightness measuring method using the slag outflow detector 8 can be combined with the residual steel level meter 9 based on the eddy current method, and the control as described in the above [Operation] is performed, but the residual steel level is measured. The prediction is made as follows.

That is, as shown in FIG. 11, the residual steel level meter 9 is provided with an exciting power source 91, pre-amplifiers 92B and 92C, and phase-detector amplifiers 93B and 93C for each coil Ca, Cb, Cc. Then, a reference frequency current is supplied from the excitation frequency adjusting circuit 94 to the excitation power source 91, and an AC excitation current of 50 to 200 Hz is applied to the ladle of 150 to 300 tons. The generated eddy currents are detected by the receiving coils Cb and Cc, and after being pre-amplified and phase-detected, they are taken into the optimum residual steel level prediction system 10.

The signals detected by the receiving coils Cb and Cc show the changes over time in FIG. If a large amount of metal remains when the molten steel flows out, the detection error is large at the time point, making it difficult to detect the molten metal. Therefore, the point is detected by the receiving coil Cc attached at a position substantially corresponding to the point where the metal adhesion is small, and the remaining steel depth d ₇ is stored.
On the other hand, the signal from the receiving coil Cb shows a change as shown up to the subsequent point (the point of the optimum remaining steel depth).
The detected signal levels of about 10 points are stored and used as the material for the next signal processing. Among these 10 points, d ₇ output by the receiving coil Cb at the point is included.

When the residual steel depth (level) reaches a point, the subsequent residual steel level is estimated as follows.

In making this assumption, the brick was worn away with the number of times the ladle was used, and the bottom level of the bottom wall in particular fell,
It must be taken into consideration that the residual steel level detection signal changes as shown in FIG. 14 or FIG. Now, in the nth operation, until ...
l ₁ , l ₂ , l ₃ , l ₄ , (cm), and the molten steel weight in the injected tundish changed to W ₁ , W ₂ , W ₃ , W ₀ (ton), and the mold cross-sectional area S (cm ² ), the density of molten steel is γ (ton / cm ² ),
Α is the coefficient of thermal expansion of molten steel and D (c
m) and the rate of brick ladle wear is R (cm / number of times of use), the residual steel depth d _i is given by the following equation (1).

This equation (1) holds for each use,
A similar change should occur in the output of the receiving coil Cb each time. Therefore, if the above-mentioned point is detected at the current n times, it is possible to estimate the subsequent change in the remaining steel depth by the equation (1). In this case, in the current n-th use, the change to the point is the receiving coil Cb.
Since it is given as a real signal by the output change of
In the estimation from the above point (1) in the range from the point to the point where the optimum remaining steel depth is given, the estimation is performed with correction.

Thus, finally, when the current remaining steel depth reaches the optimum remaining steel depth, operations such as ending the injection into the tundish are performed.

Next, the relationship between the control systems will be described with reference to FIG.

First, if the residual steel depth reaches, for example, 500 mm (A
Point), tilt the ladle. Almost at the same time, raising the set level of molten steel weight in tundish weight control (point B),
A time margin for continuous casting is provided, and closed loop control is continued until, for example, the remaining molten steel reaches 10 tons. In this set level increasing section, of course, the opening degree of the ladle nozzle is controlled to be high.

After that, if the molten steel depth becomes less than 10 tons, slag may start to mix into the molten steel flowing into the tundish. Therefore, in order to delay the start of slag mixing as much as possible, do not open or close the ladle nozzle. Is preferred. Therefore, from the point of less than 10 tons, the opening of the ladle nozzle is kept low or high (C) to stabilize the molten metal surface and prevent inclusion of inclusions in the slab. As a result, the weight of molten steel in the tundish begins to increase or decrease (D). When the weight of molten steel in the tundish increases, it overflows, and when it decreases, the time margin for continuous casting decreases, which is not preferable. Therefore, the amount of molten steel in the tundish is adjusted by changing the opening of the ladle nozzle (E). However, this adjustment should not be performed frequently because the molten metal surface becomes unstable for the above reasons, and when the rate of change in the tundish becomes large, the casting speed is changed (F).

According to the example of the present invention described above, as shown in FIG. 5, there is no delay in detection of slag outflow due to clogging of nozzles due to adhesion of metal and low heat, and there is no chance of large residual steel generation, and the average hit rate is 0.57. ton, standard deviation σ (68) = 0.3
9t, 3σ (99.7%) is 1.17t, which is an extremely high result.

This is summarized in Table 1.

[Effects of the Invention] As described above, according to the present invention, not only can slag be surely prevented from flowing out, the quality of the slab can be improved, but the amount of residual hot water (steel) can be greatly reduced, and the yield can be improved. Can be improved.

[Brief description of drawings]

Fig. 1 is a front view of a partial cross section showing an example of coil arrangement in a ladle when the ladle is tilted, Fig. 2 is an explanatory diagram of a main control system of a continuous casting facility, and Figs. 3 (a) to (d) are Fig. 4 is a flow diagram of eddy currents accompanying changes in molten steel level, Fig. 4 is a relational diagram between the ladle coil position and its output voltage, Fig. 5 is a histogram of residual steel amount control accuracy according to the method of the present invention, and Fig. 6 is Fig. 7 shows the detected voltage change with the number of times of use of the pot, Fig. 7 shows the frequency diagram of erroneous detection elements by the eddy current method and luminance measurement method, Fig. 8 shows the residual steel depth-reception coil electromagnetic field intensity correlation diagram, and Fig. 9 shows that. Fig. 10 is a characteristic view of a receiving coil placed on the bottom wall of the ladle in an enlarged view of the main part, Fig. 10 is a measurement mode diagram of the luminance measurement method, Fig. 11 is a block diagram of the residual steel level meter, and Fig. 12 is a side wall. Change diagram of the detection signals of the receiver coil and the bottom wall receiver coil, and FIG. 13 is a schematic diagram showing the relationship between the bottom wall receiver coil and the residual steel, FIG detection signal variation diagram associated with the number of uses of the ladle, FIG. 15 is a time chart showing a control method of the injection end of the tundish. 1 ... Ladle, 2 ... Sliding gate, 3 ... Ladle nozzle, 5 ... Tundish, 7 ... Mold, 8 ... Slag outflow detector, 9 ... Steel level gauge, 21 ... Optical fiber, 22 ... Optical power meter, Ca ... Excitation coil,
Cb, Cc ... Reception coil.

Claims (2)

[Claims] 1. A method for preventing outflow of slag due to injection of molten steel into a tundish from an outflow hole of a ladle by a closing operation of a partition means provided therebetween, wherein a bottom wall of a side wall of the ladle is provided. A pair of coils, one of which is an exciting coil and the other of which is a receiving coil, is provided at each of the molten steel outflow hole side position and the position opposite to the molten steel outflow hole side of the bottom wall, and an eddy current is generated from the exciting coil. Then, this is detected by the receiving coil and the change in eddy current is detected to determine the residual steel level. When the residual steel level reaches a predetermined level, the partitioning means is closed and injection into the tundish is stopped. A method for preventing slag outflow, which is characterized by: 2. A method for preventing outflow of slag due to pouring of molten steel into a tundish from an outflow hole of a ladle by closing a partition means provided therebetween, which is a molten steel outflow hole of a side wall of the ladle. A pair of coils, one of which is an exciting coil and the other of which is a receiving coil, is provided at each of the side position and the position opposite to the molten steel outflow hole side of the bottom wall, and an eddy current is generated from the exciting coil, Is detected by the receiving coil and the change in eddy current is captured to determine the residual steel level,
When this residual steel level reaches a predetermined level, the eddy current method that closes the partitioning means to stop the injection into the tundish, and the change in the brightness of the outflow of molten metal from the ladle are detected. When there is a decrease in brightness more than a certain value, the partitioning means is closed and injection into the tundish is stopped together with the brightness measurement method, and the amount of residual steel in the ladle is greater than the specified amount. A method for preventing molten metal outflow, wherein the eddy current method is prioritized in the step, and the luminance measuring method is prioritized in the step when the amount is less than a predetermined amount to prevent the slag outflow.