JPH07150227A - Method for automatically observing vacuum degassing furnace - Google Patents

Abstract

PURPOSE:To automatically detect the charging of alloy and the dropping of stuck metal and to prevent the overlooking operation by observing a vacuum degassing vessel with an image pickup device and treating the obtd. picture. CONSTITUTION:The image pickup device 13 is arranged at the position being possible to take the picture to the exposed surface of molten steel in the vacuum degassing vessel 12. The picture area of the molten steel part exceeding the preset brightness level is obtd. from the brightness information of the molten steel temp., stuck metal, dropped metal and charged alloy shown in the figure in the vessel picking up the image by this method. The possessed ratio of the picture area at the part exceeding this brightness level in the whole figure area in the vessel 12 is calculated by an image processor 17. When this ratio is less than the preset area ratio value, it is judged to charge the alloy or to drop the metal. By this method, miss charge of the alloy or the dropping of the metal can quickly be detected and the countermeasure is quickly N, and the components in the molten steel can be stabilized.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for automatically monitoring the inside of a vacuum degassing furnace.

[0002]

2. Description of the Related Art Vacuum degassing furnaces are used to remove unnecessary oxygen and the like in molten steel, but at the same time, alloys such as Fe-Mn are added to adjust the composition of molten steel. The monitoring items in the operation of the vacuum degassing furnace include checking the dipping depth of the dipping pipe and the state of molten steel recirculation, checking the presence or absence of the addition of alloys, and checking the presence or absence of the ingots adhering to the degassing tank.

A technique for confirming and monitoring the depth of immersion of the immersion pipe and the state of reflux of molten steel is disclosed in Japanese Patent Laid-Open No. 63-190114 "Dip tube monitoring device for vacuum degassing furnace", Japanese Patent Laid-Open No. 58-167718.
"Vacuum degassing method and apparatus for molten metal" has been introduced and has already been automated.

However, it is the current situation that a person is monitoring the image taken by the surveillance camera to confirm whether or not the alloy is thrown in and whether or not the metal is dropped, and no other detecting means has been established.

[0005]

During the operation of the vacuum degassing furnace, the alloy is charged surely at a predetermined amount at the scheduled time,
It is necessary to secure the required molten steel composition.

Further, when the metal adhered to the inside of the vacuum degassing furnace falls into the molten steel flowing back into the tank, the molten steel composition changes and the temperature of the molten steel lowers. Don't

Normally, the alloy is charged based on the automatic sequence signal, but due to a malfunction of the automatic sequence or the alloy being caught in the valve seat of the charging port, the alloy is not charged as planned (time of charging, amount of charging). Or an unscheduled alloy may be thrown in.

[0008] On the other hand, it is impossible to predict the fall of a bare metal, and no preventive means has been established.

Therefore, at present, the operator constantly observes the camera image showing the molten steel surface in the tank to confirm whether or not the alloy is charged as planned and at the same time monitors whether or not the metal is dropped.

There is no particular problem as long as it is confirmed that only the alloy is charged at the scheduled time.

However, there is a problem in that it is necessary to carefully monitor the presence or absence of unscheduled alloy injection and to monitor the presence or absence of ingots, and there may be individual differences or oversights.

Therefore, in order to solve these problems, there has been a strong demand for a monitoring method capable of always performing reliable automatic monitoring and confirmation.

[0013]

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and in a vacuum degassing furnace, an image of a molten steel exposed surface in a tank obtained by installing an image pickup device in the tank is provided. The present invention provides a monitoring method for automatically checking the presence or absence of metal drop and metal drop by processing the brightness level of.

Specific technical conditions are as follows: Molten steel temperature and adhesion ground indicated by the image taken in the tank by an image pickup device provided at a position where the exposed surface of the molten steel in the tank of the vacuum degassing furnace can be photographed. From the brightness information of gold, drop metal and input alloy, the image area of the molten steel part that exceeds the preset brightness level is obtained, and the image area of the part that exceeds this brightness level is calculated as the ratio of the area of the entire molten steel image in the tank. An automatic monitoring method for a vacuum degassing furnace is characterized in that it is calculated, and when this ratio falls below a preset area ratio set value, alloy injection or metal drop is determined.

[0015]

FIG. 1 is an overall configuration diagram of a vacuum degassing apparatus to which the present invention is applied, and 12 in the figure shows a vacuum degassing tank main body. 2 (a), (b), and (c) are image views (transverse cross-sectional views in the tank) of images taken by an imaging device of a state in which molten steel is refluxed in the vacuum degassing tank.

FIG. 2 (a) is an image image drawing of a normal state, in which the Xa region around the tank and the Y at the center of the tank are shown.
It is roughly classified into a region.

The Xa area around the tank is a dark image with low brightness because it is a photographed image of a portion to which a low temperature metal is attached.

On the other hand, the Ya area at the center of the tank is about 1500-
The molten steel at a high temperature of 1600 ° C. is refluxing, and the photographed image of this portion is a high-brightness bright image because it is the exposed high-temperature molten steel.

Here, the alloy charged for adjusting the molten steel composition is at room temperature and has a large temperature difference from the refluxed molten steel.

Therefore, when the alloy at room temperature is put into the high temperature molten steel in the vacuum degassing tank, the image obtained by the image pickup device has a low brightness point 21 as shown in FIG. 2 (b).
Many (alloys) appear. That is, a portion having a low brightness level increases in the image.

According to the present invention, based on the brightness level data appearing in such an image, the alloy to be put into the molten steel and the ingot to be dropped are detected.

Here, based on the brightness level data appearing in the image, in order to detect the alloy to be poured into the molten steel and the dropping metal, the division of the picked-up image by the binarization process and the calculation of the image area ratio are performed. Two processes are required: a process (the ratio of the image area of the molten steel to the image area of the entire in-tank image).

First, binarization processing is performed.

Since there is a clear difference in the brightness level between the brightness of the molten steel image and the brightness of the alloy image or the metal image, the value of the intermediate brightness level of the brightness values of the molten steel and the alloy or the metal is By setting and comparing the set brightness value (called P value) with each brightness in the image, the molten steel image and the alloy image,
Or it is to distinguish the images of the falling bullion.

In this way, distinguishing substances by a certain set value (here, distinguishing molten steel from alloy or molten steel from falling metal based on brightness) is called binarization processing.

Next, an image area ratio calculation process is performed.

For example, the binarization process based on the P value obtained previously is applied to the in-tank image at the time of alloy injection shown in FIG. 2 (b), and the steady temperature is calculated from the brightness distribution of the in-tank image taken. Ratio C of the image area of the molten steel to the total area of the image inside the tank
The binary value is obtained by the equation (1).

[0028]

[Equation 1]

Here, the Xb region is a bare metal attached to the center of the tank, the Yb region is molten steel flowing back to the center of the chamber, Zb.
The area represents the alloy that was charged into the bath.

When the C2 value thus obtained is lower than the preset area ratio C1 value (C1> C2)
It can be judged that the alloy is thrown in or the metal is dropped.

The preset value of the brightness level for binarization (the above-mentioned P value) and the preset value for the area ratio determination (C1
The value) can be set by statistically processing the brightness of the image when the alloy is charged, when the metal is dropped, and during normal operation.

Further, since the image pickup device is installed in the degassing tank and the average luminance of the entire image obtained by the adhesion of dust or the like changes with time, the set value of the luminance level for binarization (P
Value) and the set value (C1 value) for determining the area ratio can be corrected more accurately by using the average brightness of the molten steel image and the entire image of the inside of the tank during the previous or the last several degassing processes. It is possible to detect injection and drop of metal.

This concept will be described in detail with reference to the flowchart of FIG.

In FIG. 3, an image is captured from the image pickup device in A, and the brightness level of the image captured in FIG. 3B is divided into a part exceeding a preset brightness level (P value) and a part below the brightness level and binarized. To do.

From the result, the area of the portion exceeding the brightness level (P value) preset in FIG. 3C is calculated,
The bright area ratio (same as C2) defined by equation (2) is calculated.

[0036]

[Equation 2]

In FIG. 3D, it is judged whether or not the bright area ratio is below a preset value (C1), and if it is below, it is judged that the alloy is added, and the result is shown in FIG. , And return to FIG. 3A to capture the next image and repeat the above process.

Further, when the base metal drops into the molten steel, the temperature of the base metal is much lower than the temperature of the molten steel, so that the area of the low brightness portion increases.

However, while the charged alloy drops in the form of particles with a diameter of 1 to 10 cm, the metal ingot drops into a large lump with a diameter of about 30 to 60 cm, and as shown in FIG. Gold has a lower brightness level than alloys and appears as large dots 22.

Therefore, if the set value of the brightness level for binarization is made smaller than that for the detection of alloy injection, only the drop of the metal can be detected.

Therefore, if the setting value (P value) of the binarization luminance level and the setting value (C1 value) for the area ratio determination are changed to detect the drop of the bare metal, the same value can be obtained in one imaging device. The method can alternately detect alloy injection and drop of metal.

That is, an image is taken in from the image pickup device in FIG. 3A, and the brightness level of the image taken in in FIG. 3F is divided into parts above and below the preset brightness level, that is, binarization.

From the result, the area of the portion exceeding the brightness level preset in FIG. 3G is calculated, and the bright area ratio (C2 value) defined by the equation (2) is obtained.

In FIG. 3H, it is determined whether or not the bright area ratio (C2 value) is below a preset area ratio value (C1 value). If it is below (C1> C2), the alloy is used. It is determined to be input, the result is transmitted to the host computer in FIG. 3I, the process returns to FIG. 3A to capture the next image, and the above process is repeated.

As described above, with the same apparatus and processing method according to the present invention, it is only necessary to change the setting value of the brightness level for binarization and the setting value for determining the area ratio, and to throw the alloy and drop the metal. Can be detected.

[0046]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A concrete embodiment according to the present invention will be described with reference to FIGS.

FIG. 1 is an overall explanatory view of an example of an apparatus for carrying out the present invention. An image pickup device 13 installed on the upper part of a vacuum degassing tank 12 and an image of the image pickup device 13 are taken in to detect alloy injection and metal drop. The image processing device 16 and the display device 17 that displays the processing result of the image processing device 16. In the figure, 11 is a molten steel ladle, 14 is an alloy charging port, and 15 is an exhaust port.

The image pickup device used in this example is an optical lens type camera, but there is no particular limitation on the use as long as it is a device such as an optical fiber or a thermoviewer capable of discriminating the brightness of a substance.

Further, the image processing device 16 is also connected to a process computer 18 which performs appropriate control by using signals of alloy injection detection and metal drop detection.

FIG. 4 shows an example of a time-series change 30 of the area ratio (C2 value) of the bright portion when the alloy is charged, which is processed by the automatic monitoring apparatus of the present invention.

FIG. 5 shows processing by the automatic monitoring device of the present invention.
Time series change of the area ratio (C2 value) of the bright part when the metal is dropped 40
It shows an example of.

The image photographed by the image pickup device 13 is read by the image processing device 16. In the read image, one image is converted into a digital amount of 512 × 512 pixels by the image processing device 16, and whether or not the brightness level exceeds the preset value of the binarization brightness level set for each pixel is checked. It is checked (this process is binarization), and if it exceeds the set value, the brightness level of the pixel is set to 1, and if it is below the set value, it is set to 0.

Next, the number of pixels exceeding the set value is integrated, and the value is taken as the area value of the bright portion, and the bright portion area ratio (C2 value) is calculated by the equation (2).

FIG. 4 and FIG. 5 show the results of time-series display of changes in the light area ratio (C2 value).

FIG. 4 shows an example in which the setting value (P value) of the binarization luminance level for detecting the alloy injection is 150 and the setting value (C1 value) for determining the area ratio is 18%. .

The area ratio (C2 value) of the bright portion is approximately 30 to 40% in the normal state 31 in which no alloy is added, but the area ratio (C2 value) of the bright portion is 32 when the alloy is added. Is reduced to 10% or less, and the difference in the light area ratio (C2 value) between the normal time 31 and the time when the alloy is charged 32 is as large as 20 to 30%.

Therefore, the set value 33 (C1
Value is set between 10 and 30% (18% in this example)
, And the bright area ratio (C2 value) at the time of alloy injection 32
Is less than the set value 33 (C1 value) for area ratio determination (C1
> C2), so the alloy injection is detected.

Further, FIG. 5 shows an example in which the setting value of the brightness level for binarization for detecting the fall of the metal is 30 (P value) and the setting value for area ratio determination (C1 value) is 50%. Is.

The area ratio (C2 value) of the bright portion is secured at 55 to 60% in the normal time 41 when the metal is not falling, but the area ratio (C2 value) of the light portion is 42 when the metal is dropped. Is 40
It is reduced to about 50%, and there is a large difference between the light area ratio (C2 value) between the normal state 41 and the bare metal drop 42 (C2 value) of 10 to 15%.

Therefore, the set value 43 (C1 for area ratio determination)
Value) between 45 and 50% (50% in this example)
Set), and the bright area ratio (C2 value) when the bullion falls 42
Is less than the set value 43 (C1 value) for area ratio determination (C1
> C2) Therefore, the drop of the metal is detected.

FIG. 6 shows an example in which the operating condition of the vacuum degassing furnace is continuously monitored by the monitoring device according to the method of the present invention.

FIG. 6 (a) is a signal showing that the alloy is charged into the degassing furnace when the alloy charging command signal 53 from the host computer is at the level 1 and the alloy is not charged at the level 0. It is a state diagram.

When the alloy is charged into the degassing furnace, the result shown in FIG.
As shown in (c), the area ratio 50 (C2 value) of the bright portion decreases and falls below the area determination set value 51 (C1 value).

Further, the output signal 52 of the alloy injection detection result of this monitoring device is at the level of 1 as shown in FIG. 6B, which coincides with the states of FIGS. 6A and 6B. Indicates that alloy has been added.

When the alloy input command signal 53 of the host computer shown in FIG. 6A becomes 0, the bright area ratio 50 (C2 value) shown in FIG. 6C recovers and the area ratio judgment set value 51. (C
6), the output signal 52 of the alloy charging detection result shown in FIG. 6B has a level of 0, and it is detected that the alloy has not been charged.

As shown in this example, the alloy injection command of the host computer and the alloy injection detection result of this monitoring device are completely in agreement, and according to this method, there is no oversight by the observer, and the presence or absence of alloy injection is surely performed. Indicates that can be detected.

As described above, the establishment of the detection means in place of the monitoring staff ensures the automation of this process.

[0068]

According to the present invention, the inside of the vacuum degassing tank is monitored by the image pickup device and the obtained image is processed to reliably and automatically detect alloy injection and metal drop. Therefore, it is possible to prevent humans from overlooking.

As a result, it was possible to promptly detect the mistaken injection of the alloy and the drop of the metal, which had occurred several times a month, the countermeasures were speeded up, and the molten steel composition was stabilized.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram of an example of a vacuum degassing apparatus for carrying out the present invention.

2 (a), (b), and (c) are image images of the inside of a vacuum degassing tank obtained by an image pickup device, showing a normal time, an alloy charging time, and a bare metal falling time, respectively.

FIG. 3 is a flowchart showing the concept of the present invention.

FIG. 4 is a diagram showing an example of a signal processing result at the time of charging an alloy, which is obtained according to the present invention.

FIG. 5 is a diagram showing an example of a signal processing result when a metal is similarly dropped.

6 (a), (b) and (c) are diagrams showing an example of data obtained when continuously monitoring the presence / absence of an alloy by the monitoring method according to the present invention.

[Explanation of Codes] 11 ... Molten steel ladle 12 ... Vacuum degassing tank 13 ... Imaging device 14 ... Alloy injection port 15 ... Exhaust port 16 ... Image processing device 17 ... Display device 18 ... Process computer 21 ... Alloy 22 ... Ingot 30 ... Change in area ratio of bright part 31 ... Change in area ratio of bright part during normal operation 32 ... Change in area ratio of bright part when alloy is charged 33 ... Set value for area ratio determination 40 ... Change in area ratio of bright part 41 ... Transition of change in area ratio of bright part at normal time 42 ... Transition of change of area ratio of bright part when metal is dropped 43 ... Set value for area ratio determination 50 ... Change of area ratio of bright part 51 ... For area ratio determination Set value 52 ... Output signal of alloy injection detection result 53 ... Alloy injection command signal from host computer

Continuation of front page (51) Int.Cl. ⁶ Identification number Reference number within the agency FI Technical indication location H04N 7/18 D (72) Inventor Makoto Moriguchi 1st Nishinosu, Oita-shi Oita Steel Co., Ltd. Oita Works

Claims (1)

[Claims] 1. A molten steel temperature, a deposit metal, and a falling metal which are shown by an image taken in the tank by an imaging device provided at a position where the exposed surface of the molten steel in the tank of the vacuum degassing furnace can be photographed. And, from the brightness information of the input alloy, obtain the image area of the molten steel part exceeding the preset brightness level, calculate the ratio of the image area of the part exceeding this brightness level to the total area of the molten steel image in the tank, and calculate this ratio. A method for automatically monitoring a vacuum degassing furnace, characterized in that it is judged that an alloy has been introduced or a metal has fallen when the area ratio has fallen below a preset area ratio set value.