JPH035257B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a method and apparatus for detecting the level of molten steel in a molding device in continuous casting equipment.

BACKGROUND ART FIG. 1 is a diagram schematically showing continuous casting equipment that is the basis of the present invention. A tundish device 1, which is a heat-resistant container, contains liquid molten steel 3, which is poured into a cylindrical mold device 2 through a pouring section 4. The molding device 2 is cooled with cooling water 6 spouted from a cooling water jetting device 5. The liquid molten steel 3 injected into the molding device 2 is pulled out from below the molding device 2 by a pinch roller 7 while being sequentially cooled and solidified, so that rod-shaped slabs 8 are continuously produced.

Therefore, in continuous casting equipment, mold device 2
It is necessary that the amount of molten steel 3 injected and the amount of slab 8 drawn out are always equal, that is, the height of the molten steel surface (hereinafter referred to as the level) of the molding device 2 is constant. If this balance is disrupted, for example when the injection amount is large, molten steel overflows from above the molding device 2. On the other hand, if a large amount is pulled out, it will fall out from below. In either case, a major accident will occur. For this reason, various methods have been used to detect the hot water level.

In the prior art, on the wall surface of the molding device 2,
There are methods using a large number of thermocouples embedded at equal intervals in the vertical direction, methods using a gamma ray source such as Co ⁶⁶ and a scintillation counter, and methods using a molding device 2.
is imaged by a two-dimensional imaging means, and the video signal is classified into bright signals and dark signals at a predetermined output level.
A method of counting the length of the bright signal for each scanning line forming one screen and detecting the hot water level from the pattern of the counted values (for example, see Japanese Patent Laid-Open No. 119222/1983), etc. has been put into practical use. However, in the method using thermocouples, there is a limit to the spacing between the thermocouples, which results in poor accuracy in detecting the level of the hot water. The molding device had to be made thinner, which shortened the life of the molding device, resulting in a problem of reduced service life. In addition, the method using radiation has a risk of radiation leakage, which poses problems in terms of maintenance and safety. Furthermore, in the method using the imaging means, the brightness of the subject and therefore the image constantly changes due to the environment surrounding the subject, problems with the subject itself, such as the occurrence of smoke or flames, or ambient light. Setting the optimal threshold for detecting the hot water level was unstable and difficult in some cases.

An object of the present invention is to solve the above-mentioned problems and provide a method and device for stably and highly accurately detecting the level of the molten steel inside the molding device.

Means for Solving the Problems The present invention provides an imaging surface whose scanning direction is parallel to the molten steel surface and whose optical axis is at a constant angle with the molten steel surface in a molding device of a continuous casting facility. The image is captured by two-dimensional imaging means installed so as to intersect with each other, and the video signal from the two-dimensional imaging means is divided into at least three stages according to the brightness of the image and converted into numerical information, and the numerical information is converted into numerical information. is integrated for each scanning line in at least one field of the screen, and the molten steel level in the molding device is detected based on a relative change in the integrated value thus obtained. This is a method for detecting the level of molten steel in continuous casting equipment.

The present invention also provides a method for disposing a molten steel surface inside a molding device of a continuous casting facility such that the scanning direction of the imaging surface is parallel to the molten steel surface and the optical axis intersects the molten steel surface at a constant angle. a two-dimensional imaging means; a multi-value conversion circuit that converts the video signal from the image pick-up means into numerical information divided into at least three stages according to the brightness of the image; and a numerical signal from the multi-value conversion circuit. an integrating circuit that integrates each scanning line in at least one field of the imaging screen; a maximum and minimum value detection circuit that detects the maximum and minimum values based on the integrated results of the integrating circuit; , a threshold setting circuit that sets an optimal value between the maximum and minimum values as a threshold based on data obtained from the minimum value detection circuit; an integrated value obtained by the integration circuit; and the threshold setting circuit. a detection circuit that compares the integrated value with a threshold value set in and detects a scanning line portion where the integrated value exceeds the threshold value; and a level that detects the molten steel surface level based on the data obtained by the detection circuit. A device for detecting the level of molten steel in continuous casting equipment, characterized by including a detection circuit.

Effects According to the present invention, the molten steel level inside the molding device of continuous casting equipment can be detected stably and with high precision, thereby improving maintenance and safety, and improving workability. can. That is, the video signal from the two-dimensional imaging means is divided into at least three stages according to the brightness of the image of the molten steel surface, and this numerical information is integrated for each scanning line in at least one field. Since the molten steel surface level is detected based on the relative change in this integrated value, accuracy is improved.

Embodiment FIG. 2 is a sectional view showing the configuration of an embodiment of the present invention. The molten steel 3 in the tundishing device 1 is
The molten metal is poured into the molding device 2 through the pouring section 4. In the present invention, a two-dimensional imaging means (in this embodiment, an industrial television camera) 9 having a field of view as shown by reference numeral 12 is provided obliquely above the molding device 2, and its optical axis 13 is a cylindrical lens. Central axis line 11 of hot water section 4
The inside of the molding device 2 is imaged by installing the imaging surface so that the scanning direction of the imaging surface is parallel to the hot water level 10 inside the molding device 2. The video signal from the imaging means 9 is processed by the image processing means 14 as will be described later, and thereby the level of the hot water surface 10 is detected.

FIG. 3 is a diagram showing an imaging screen in the molding device 2 captured by the two-dimensional imaging means 9, and corresponding parts in FIG.
Attach. FIG. 4 is a diagram showing the waveform of the video signal of the imaging means 9. With reference to FIGS. 3 and 4,
The imaging means 9 generally has γ lines (262.5 lines per field in Japan's television standard system, 1 field).
One screen is formed by scanning video signals on 525 scanning lines per frame, but first, before scanning one screen, a one screen start signal (hereinafter referred to as vertical synchronization signal) VBL ( (on the left in Fig. 4) is output, and then the first scan start signal (hereinafter referred to as horizontal synchronization signal) HBL is output, and then the video signal S is scanned on the first scan line S1. . When the scanning of the scanning line S1 is completed, the horizontal synchronizing signal HBL is output again, the video signal is scanned to the scanning line S2, and the process is repeated sequentially up to the last scanning line Sr.

Figure 5 shows the video signal included in one scanning line.
It is a figure which shows the waveform of SK1 and SK2. video signal
The levels of SK1 and SK2 are proportional to the brightness of the subject, and for example, the scanning line Sm crossing the image 4a of the pouring part as shown in FIG.
In the video signal corresponding to the scanning line Sn that crosses the image 3a of the hot water surface, the level l1 of the pouring part with high brightness and the part corresponding to the hot water surface is higher than the level l2 of other parts with low brightness. It's summery.

FIG. 6 is a block diagram showing an electrical signal system of the image processing means 14 for detecting the level of the hot water surface from the video signal of the imaging means 9. As shown in FIG. 5, the video signal S proportional to the luminance of the object is input to the multi-value conversion circuit 20, and is divided into at least three or more stages (for example, eight stages in this embodiment) according to the luminance. It is quantified. Referring to FIG. 5, the digitized values are, for example, Di
2,...,7), and the lowest brightness level is D ₀ =0,
D1, D2, ..., D7 in descending order of brightness.
D7 is the highest brightness level. Classifying the output level of the video signal according to the luminance and converting it into numerical values in this way is called multi-value processing. Note that the multi-value processing circuit 20 is supplied with sampling pulses output from the clock generation circuit 21 at intervals of time Δt1 (for example, W1/256), which is sufficiently smaller than the period W1 of the horizontal synchronizing signal HBL. The multi-value processing in the circuit 20 is executed in synchronization with the sampling pulse.

The integrating circuit 22 integrates the multivalued numerical value Di outputted every Δt1 time in synchronization with the sampling pulse. For convenience of explanation, this integrated value will be referred to as D, and D is expressed by the following equation: D=ΣDi= _r 〓 ^j=1 Δtj (1). Here, r is the number of scanning lines forming one field of the screen or one frame corresponding to a plurality of fields (for example, two), and for one field, r=256 (lines), for example.

The synchronization separation circuit 29 separates and extracts the horizontal synchronization signal HBL and the vertical synchronization signal VBL from the output signal of the imaging means 9, and further extracts the horizontal synchronization signal HBL and the vertical synchronization signal VBL at the rising and falling edges of the horizontal synchronization signal HBL, as shown in FIG. As shown in Figure 4, each pulse of time Δt2
Generate HBLU and HBLD. Among these pulses, HBLD is applied to the reset terminal R of the integrator circuit to reset the contents of the integrator circuit to zero prior to scanning a screen. Therefore, integrating circuit 2
2 is to integrate the above-mentioned numerical value Di for one scanning line. On the other hand, the pulse HBLU is applied to the clock terminal CLK of the integration circuit 22. The output of the integration circuit 22 is connected to a storage circuit 23, and the storage circuit 23 stores the value D stored by the integration circuit 22 at the rising edge of the pulse HBLU.
Remember. Therefore, in the memory circuit 23, the integrated value D related to the scanning line Si is stored in the memory circuit 23 as
The data will be stored for a period of time until the end of the scan of , and immediately before the start of the scan of the next scan line Si+2. The contents of the memory circuit 23 are inputted to a processing circuit 30 realized by, for example, a microcomputer, and stored in a memory 31 in sequence for each scanning line, and numerical information regarding all the scanning lines stored in the memory 31 is stored. That is, the hot water level is detected from the change in pattern depending on the value D determined by the first equation.

Referring again to FIG. 5, since the level of the video signal S changes depending on the brightness as described above, the level of the video signal S SK2 changes when ambient light or illumination light is applied.
is the video signal SK1 when no light is irradiated.
The level will be higher. As described in the prior art, in the level detection method using binarization, when the brightness of the subject, that is, the molten steel surface changes due to the influence of ambient light or smoke, and the level of the video signal S changes, the threshold value is changed. Optimal settings will no longer be possible, and it will be impossible to distinguish between bright and dark areas. In binarization using a fixed threshold as shown in Figure 5, the value of Di is logic "1" or logic "0", so the video signal
When threshold value Ka is set for SK1, video signal SK2 whose level has become high due to the influence of ambient light, etc.
All the values of Di become logical "1", and the pattern of the integrated value D for one screen does not change.

FIG. 7 is a diagram showing a pattern of the integrated value D according to an embodiment of the present invention. FIG. 7 corresponds to FIG. 5, and the patterns of integrated values D corresponding to video signals SK1 and SK2 in FIG. 5 are K1 and K2, respectively.
There is a clear change in size between the portion corresponding to the image 4a of the pouring portion and the portion corresponding to the image 3a of the molten metal surface portion. Note that in FIG. 7, reference numeral 10a represents the line of intersection between the molten steel surface and the inner wall of the molding device 2 in FIG. 7th
For example, the following method can be used to detect the boundary between the pouring part and the hot water surface part in the image from the change in the pattern of the integrated value D in the figure.

(1) Find the difference between the integrated value D for Si and the integrated value D for Si+1 in adjacent scanning lines Si and Si+1, and find the difference that is the largest from S1 to Sr (r is about 250 to 500). Time scanning line Si
Let +1 be the said boundary.

(2) Integrated value D for all scanning lines S1 to Sr
Set an appropriate threshold value tn for As a boundary line.
Note that the threshold value tn is, for example, the following (2-1) to (2-
You can set it according to one of 3).

(2-1) Find the sum Q of the integrated values D for all the scanning lines in one screen, and use the value tn obtained by dividing this sum Q by the total sum line r as the threshold value.

(2-2) A value tn obtained by multiplying the maximum value among the integrated values D for all the scanning lines in one screen by a coefficient of an appropriate value between 0 and 1 is set as the threshold value.

(2-3) Appropriate value tn between the maximum value and minimum value of the integrated value D for all scanning lines in one screen
is the threshold value.

Note that the above-mentioned threshold value tn is determined according to the actual situation at the operation site.

Referring again to FIG. 6, the above (2-
The level detection method according to step 3) will be specifically explained. The storage circuit 23 sequentially stores the integrated values D for all the scanning lines in one screen as shown in FIG. 8 from the scanning line S1 to the scanning line Sr. The maximum/minimum value detection circuit 24 is the memory circuit 23
Among these integrated values D, the maximum value Dmax and the minimum value Dmin are extracted and output to the threshold value setting circuit 25. The threshold value setting circuit 25 calculates the value from the integrated value Dmax.
Dmin is subtracted to find the difference, and the difference is set to an appropriate value from 0 to 1 in the setting device 26 depending on the operating situation (for example, a value in the range of 0.25 to 0.4).
The value obtained by adding the result of multiplying by Dmin to Dmin is set as the threshold tn,
It is output to the level calculation circuit 27. In the level calculation circuit 27, first, the 1 stored in the storage circuit 23 is
All integrated values D on the screen are sequentially compared with the threshold value tn from scanning line S1 to Sr, and D
>tn, find the scanning line number i+1. Next, a value A is obtained by subtracting the number r/2 of the scanning line Sr/2 at the center of one screen from the number i+1.

A={(i+1)-r/2}...(2) This A is multiplied by the interval Δz between two adjacent scanning lines.

l=A×Δz {(i+1)−r/2}×Δz (3) l in the third equation is the scanning line Si+1 corresponding to the boundary between the molten metal surface part and the pouring part, and the scanning line at the center of the screen. S2
This is the distance on the screen from r/2. Now, the reference point (zero point) of the level of molten steel in the molding device is
Set an arbitrary point on the inner wall of the molding device within the range that can be imaged by the imaging means 9, for example, point Z0 in FIG. 9,
If the imaging means 9 is installed so that this point Z0 coincides with the scanning line Sr/2 at the center of the image of the imaging means 9, then the distance ZL from Z0 to the level P of molten steel can be determined by the following formula. can.

ZL=l/K×cosθ…(4) Here, l={(i+1)−r/2}×Δz where K: optical magnification θ: angle that the imaging device makes with the horizontal plane Note that in this example, the above Z0 Although the calculation is simplified by making the scanning line Sr/2 the same as the scanning line Sr/2, it is not necessarily limited to this condition.

Effects As described above, according to the present invention, a video signal obtained by a two-dimensional imaging device is multivalued into numerical information divided into at least three stages according to the brightness of the image, and the obtained multivalued The molten steel surface level is detected by integrating the digitized numerical signals for each scanning line of the screen and comparing the results with a preset threshold, which strongly reduces the effects of ambient light and changes in subject brightness. It is possible to stably detect the level of molten steel in a molding device in continuous casting equipment.

[Brief explanation of drawings]

Fig. 1 is a diagram showing an outline of continuous casting equipment, Fig. 2 is a sectional view showing the configuration of an embodiment of the present invention, and Fig. 3 is an image taken by the two-dimensional imaging means 9 of an embodiment of the present invention. FIG. 4 is a diagram showing the waveform of the video signal of the two-dimensional imaging means 9, and FIG. 5 is a diagram showing the waveform of the video signal S included in one scanning line. Figure 6 shows the image processing means 1.
4. FIG. 7 is a diagram showing the pattern of the integrated value D for each horizontal scanning line in the image of FIG. 4. FIG. 8 is a diagram showing the pattern of the integrated value D in one screen. , FIG. 9 is a diagram showing the relative positional relationship between the pouring unit 4 and the imaging device 9. As shown in FIG. 2... Mold device, 3... Molten steel, 9... Two-dimensional imaging means, 10, 10a... Molten steel surface level,
14... Image processing means, 20... Multi-value conversion circuit, 2
2... Integration circuit, 23... Memory circuit, 24... Maximum/minimum value detection circuit, 25... Threshold value setting circuit, 26
... Setting device, 27 ... Level calculation circuit, 30 ...
Processing circuit, 31...Memory, S1-Sr...Scanning lines forming an image screen.

Claims (1)

[Scope of Claims] 1. The molten steel surface inside the molding device of continuous casting equipment is scanned so that the scanning direction of the imaging surface is parallel to the molten steel surface and its optical axis intersects the molten steel surface at a constant angle. An image is captured by an installed two-dimensional imaging means, the video signal from the two-dimensional imaging means is divided into at least three stages according to the brightness of the image, and converted into numerical information, and the numerical information is transmitted to at least one of the screens. Molten steel of continuous casting equipment, characterized in that the molten steel surface level in the molding device is detected based on the relative change in the integrated value obtained in this way by integrating the molten steel for each scanning line in the field. Method for detecting hot water level. 2. Two-dimensional imaging means arranged such that the scanning direction of the imaging surface of the molten steel surface inside the molding device of the continuous casting equipment is parallel to the molten steel surface, and the optical axis thereof intersects the molten steel surface at a constant angle. a multi-value conversion circuit that converts the video signal from the imaging means into numerical information divided into at least three levels according to the brightness of the image; an integrating circuit that integrates for each scanning line in one field; a maximum and minimum value detection circuit that detects the maximum and minimum values based on the integrated results of the integration circuit; and a maximum and minimum value detection circuit. a threshold setting circuit that sets an optimal value between the maximum and minimum values as a threshold based on the obtained data; an integrated value obtained by the integration circuit; and a threshold set by the threshold setting circuit. and a level detection circuit that detects the molten steel surface level based on the data obtained by the detection circuit. Features: A device for detecting the level of molten steel in continuous casting equipment.