JP7256365B2 - Slag quantification method - Google Patents

Description

The present invention relates to a slag quantification method.

Slag floats on the surface of molten metal such as molten iron taken out from a blast furnace into a hot metal ladle or molten steel taken out from a converter into a ladle. The slag on the hot metal contained in the hot metal ladle may cause the components to come off in the converter process, which is the post-process. In addition, slag floating on the surface of the molten steel accommodated in the ladle may also cause component deviation in the subsequent secondary refining process. In this way, the slag floating on the surface of the molten metal contained in the container may adversely affect the subsequent processes. It is common to perform slag removal work to remove slag using a scraper (slag removal device).

In general, a slag discharger includes a molten metal container in which molten metal is stored, a tilting carriage that tilts the molten metal container while discharging the slag, a crane device that tilts the molten metal container while it is suspended, It is composed of a slag removing device body for removing slag from the hot water surface, a control device for controlling this slag removing device body, and an operation panel for operating this control device.
The body of the slag discharger includes a scraping arm having degrees of freedom in forward/backward, vertical and rotational movement, a scraping plate attached to the tip of the scraping arm for scraping out slag from the hot water surface, and the scraping arm. It is equipped with a hydraulic unit for driving, and the operation of the scraping arm is controlled by a signal from the control device.

In order to remove slag using a slag removal machine, after tilting the molten metal container, the surface of the molten metal container is imaged by an imaging device and the surface of the molten metal is displayed on the screen of the monitor device, which the operator can view. While watching, it is common to operate the slag extractor via the control panel to scrape out the slag.
In addition, as described in Patent Document 1 below, the hot water surface image is divided into areas, the distribution of slag in each area is determined, and scraping is performed with a scraping pattern according to the distribution, so that the amount of slag is a constant amount. A known technique is to scrape out only a certain area when it is determined that the area has been reduced to a certain level.
In addition, as described in Patent Document 2 below, by utilizing the fact that the surface image of the molten metal is binarized and the dust generation area moves, the dust generation area and the slag area are distinguished, and the luminance with a threshold value is used. Techniques for detecting slag are known.

JP-A-7-159053 Japanese Patent Application Laid-Open No. 2003-013129

However, in the conventional slag removal method performed by the operator using the above-mentioned slag removal machine, it is difficult to determine the completion of slag removal. If this is done, there is a problem that the outflow of molten metal due to scraping increases and the yield decreases.
Therefore, conventionally, it was extremely difficult for a non-skilled operator to properly ascertain the completion of slag removal.
As a technique for solving this problem, as described in Patent Document 2, the surface of the molten metal is imaged, the maximum luminance within a certain period of time is obtained for each pixel of the imaged image, and the obtained maximum luminance is formed. A technique has been proposed for separating molten metal and slag in a molten metal container and detecting slag by processing image data obtained from the molten metal.

However, since the method described in Patent Document 2 is a technique for evaluating the slag occupied area ratio on the surface of the hot water, the amount of slag remaining on the surface of the hot water cannot be quantified. If the thickness is non-uniform, there is a problem that the slag residual amount varies even if the slag occupied area is the same.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method for quantifying slag on molten metal.

The present invention has been made in view of the above circumstances, and has the following features.
"1" The slag quantification method according to the present invention is such that slag is present on the molten metal contained in a container, the slag covers the light emission of the molten metal according to its thickness, and the molten metal is emitted to the outside as transmitted light that has passed through the slag, the surface of the slag is imaged by an imaging device as a hot water surface, and the luminance value is obtained by image processing in each pixel region of the imaged image. In a value calculation step, the relationship between the luminance value and the radiance is obtained in advance, and from the relationship and the temperature conversion formula, the luminance value obtained in the luminance value calculation step is used to obtain the apparent slag in each pixel region. and the apparent slag surface temperature in each pixel region obtained by the slag surface temperature calculation step and the slag thickness estimation formula ( heat conduction equation ) in each pixel region. and a slag calculation step of obtaining the slag volume in the container based on the slag thickness of each pixel region obtained by the slag thickness calculation step and the melt surface area of each pixel region. It is characterized by
"2" The slag quantification method according to the present invention is such that slag is present on the molten metal contained in a container, the slag covers the light emission of the molten metal according to its thickness, and the molten metal is radiated to the outside as transmitted light that has passed through the slag, a thermal image is captured by an imaging device using the surface of the slag as a hot water surface, and the apparent slag in each pixel region is captured from the captured thermal image. A slag surface temperature calculation step for obtaining the surface temperature, and an apparent slag surface temperature in each pixel region obtained by the slag surface temperature calculation step and a slag thickness estimation formula (heat conduction equation) to calculate the slag thickness of each pixel region. and a slag calculating step of calculating the slag volume in the container based on the slag thickness of each pixel region and the melt surface area of each pixel region obtained by the slag thickness calculating step. Characterized by
[3] In the slag surface temperature calculation step, it is preferable to use the following equations (1) and (2).

[4] Preferably , the slag thickness estimation formula ( heat conduction equation ) for obtaining the thickness of the slag is the following formula (3), and the formula for obtaining the total slag weight is the following formula (4).

[5] In the method for quantifying slag according to the present invention, it is preferable that the slag volume or the total weight of slag in the container is determined with respect to the passage of time in the slag surface temperature calculation step.
(6) In the slag quantification method according to the present invention, when the slag volume or total slag weight in the container obtained by the slag calculation step satisfies the slag volume allowable range or the slag weight allowable range defined as the allowable range, It is preferable to have a slag removal completion judgment step for judging completion of treatment or completion of slag removal .
(7) In the method for quantifying slag according to the present invention, it is preferable to use observation light in a wavelength range from visible light to near-infrared region when imaging the surface of the molten metal.

According to the present invention, the amount of slag remaining on the molten metal can be grasped quantitatively and with less variation than the conventional method by image processing, and the completion of slag removal can be properly determined without relying on an experienced operator. be able to. Therefore, it is possible to provide a technique that can be used as guidance for determining the completion of slag removal by an operator during slag removal work.

BRIEF DESCRIPTION OF THE DRAWINGS It is a schematic diagram which shows the schematic structure of the slag quantification apparatus used for the slag quantification method which concerns on one Embodiment of this invention, and the schematic structure of a slag remover. FIG. 4 is a flowchart diagram showing an example of the slag quantification method; FIG. 2 is a schematic configuration diagram showing an example of a personal computer used when carrying out an example of the slag quantification method; 1. It is a figure which shows an example of the captured image obtained by imaging the surface of hot water in a container using the 1st imaging device shown in FIG. 1. It is a figure which shows an example of the thermal image obtained by imaging the surface of hot water in a container using the 1st imaging device shown in FIG. 2 is a graph showing the correlation between radiance and luminance value used when obtaining radiance with the temperature indicated by a thermal image as a true value. FIG. 10 is a graph showing a correlation when obtaining an approximation formula for calculating temperature from a temperature conversion formula based on previously obtained radiance; FIG. FIG. 3 is an explanatory diagram showing an example of molten iron as an example of molten metal existing in one pixel region and an example of the thickness of slag existing thereon; It is a figure which shows an example of the graph which evaluated the correlation of the slag area ratio in a real image and a thermal image. FIG. 10 is a diagram showing another example of a graph evaluating the correlation of the slag area ratio between a real image and a thermal image obtained when disturbances are removed from the thermal image; FIG. 5 is a graph showing comparative evaluation results of the correlation between molten metal yield loss and return flow slag area ratio (S _cal ) and calculated slag weight (W _cal ).

Hereinafter, a slag quantification method and a quantification device according to the first embodiment will be described based on the drawings.
FIG. 1 is a schematic diagram showing a schematic configuration of a slag quantification device 100 and a schematic configuration of a slag exhauster used in the slag quantification method according to the present embodiment. In addition, in FIG. 1, the container 4 which accommodates the molten metal M is shown in cross section.
As shown in FIG. 1, as an example, the slag quantification device 100 measures the slag S floating on the surface of the molten metal M such as hot metal contained in a container 4 such as a tilted hot metal ladle with a scraping plate. It is used in a slag removal step (slag removal step) in which a slag remover 5 having 51 and an arm 52 is used to scrape the slag out of the container 4 .

The slag removal step is a step of scraping out part of the slag S out of the container 4 using the slag remover 5 while the container 4 is tilted. In the slag-discharging process, part of the slag S remains in the vessel 4 for the purpose of shortening the slag-discharging time and securing the yield of hot metal.

The molten metal M contained in the container 4 has a temperature of, for example, 1200° C. to 1400° C., and emits light by thermal radiation (that is, the molten metal M emits light by itself by thermal radiation). On the other hand, the slag S floating on the surface of the molten metal M contained in the container 4 is cooled by being exposed to the atmosphere and is solidified near the surface. Light emission is blocked according to the thickness of the slag S.

The slag quantification device 100 of the present embodiment includes first imaging means (first imaging device) 1 for imaging the surface of the molten metal in the container 4 from above, and A second imaging means (second imaging device) 3 for imaging, and an image processing means (image processing device) 2 connected to the first imaging means 1 and the second imaging means 3 are provided.
In this specification, the term "surface of molten metal" does not mean only the surface of the molten metal M, but means the surface of the slag when the slag floats on the surface of the molten metal. That is, it means the outermost surface (uppermost surface) of the contents in the container 4 . It is preferable that the first imaging means 1 is installed directly above the container 4, but if it cannot be installed directly above the container 4 due to obstacles, etc., it may be installed diagonally above, so that the surface of the molten metal M can be imaged on a screen as large as possible. location is preferred.

As the first imaging means 1, for example, a device provided with both a CCD camera having a primary sensitivity in the visible light range and a thermal imaging camera (thermography) having a primary sensitivity in the infrared range can be used.
In this embodiment, when the image is captured by the CCD camera, it is possible to calculate the luminance value for each pixel area corresponding to the slag in each pixel area in the captured image. Further, when an image is captured by a thermal imaging camera (thermography), the average temperature value (T:° C.) of the pixel region can be calculated.

The image processing means 2 is composed of, for example, a general-purpose personal computer in which a predetermined program for executing each step described later is installed. The image processing means 2 has an output means such as a monitor for displaying the captured images obtained by the first imaging means 1 and the second imaging means 3 .
FIG. 3 shows an outline of a personal computer for configuring the image processing means 2. As shown in FIG.
A personal computer C shown in FIG.
The input means 13 is, for example, a keyboard for inputting letters and numbers, and various information can be input to the storage means 15 or the control section 14 through this.
The control unit 14 is composed of a so-called CPU (central processing unit), RAM (random access memory), ROM (read only memory), etc., and performs various numerical calculations, information processing, device control, etc. by programs. can be done.

The storage means 15 is, for example, an information recording medium such as an HDD (hard disk drive) or an SSD (solid state drive), and stores programs for performing calculations performed in each process described later, calculation means 17, and prediction means 18. It is possible to store and read out various information necessary for , and the results obtained by these, as necessary.
The output means 16 is, for example, a monitor, a printer, etc., and can display or print various information obtained from a program for carrying out the steps described later on a screen or on paper as necessary.

In the storage means 15, a program for performing calculations to be performed in each step described later, and various information necessary for executing the calculation means 17 and the prediction means 18 described later are stored in advance before the program is executed. These can be arbitrarily read and manipulated. Further, calculation results obtained by executing the program can be stored and read as necessary. Note that the storage means 15 has only a communication function to the Internet or a network, and the calculation means 17 or the prediction means is obtained by using the storage means, the calculation means, or the prediction means provided in another personal computer connected to the Internet or the network. Of course, it may be constructed so as to be able to perform calculations in the same manner as in 18 and calculate the results.
Processing of each step actually performed using the personal computer C shown in FIG. 3 will be described later.

As the second imaging means 3, like the first imaging means 1, for example, either a CCD camera having a main sensitivity in the visible light range or a thermal image camera (thermography) having a main sensitivity in the infrared light range is used. can be used. In this embodiment, a CCD camera can be used as the second imaging means 3 .

The slag quantification method according to this embodiment is executed using the slag quantification device 100 . A slag quantification method according to the present embodiment will be described below.
FIG. 2 is a flow diagram showing the slag quantification method according to the present embodiment. The slag quantification method according to the present embodiment uses the above-described personal computer C to perform various This is a method of quantifying the slag S floating on the surface of the molten metal M by performing calculations. As shown in FIG. 2, the slag quantification method has a brightness value calculation step S1, a slag surface temperature calculation step S2, a slag thickness calculation step S3, and a slag calculation step S4.
Further, the slag S floating on the surface of the molten metal M is scraped out of the container 4 by using the slag scraper 5 having the scraping plate 51 and the arm 52 described above, and the slag is quantified according to the present embodiment. The total slag weight or total slag volume is calculated by sequential calculation based on the conversion method.

Then, the total slag weight (W _cal ) calculated by this sequential calculation is compared with the predetermined limit slag weight (W _limit ) in a comparison step S5, and when the calculated total slag weight is less than the limit slag weight, It is possible to grasp the completion of slag removal and terminate the slag removal work (slag removal completion determination step).
Alternatively, using the total slag volume instead of the total slag weight, the total slag volume (V _cal ) calculated by the previous sequential calculation and the predetermined limit slag volume (V _limit ) are compared in the comparison step S5 to calculate When the total slag volume obtained is less than the limit slag volume, it can be determined that the slag removal is completed, and the slag removal work can be terminated (slag removal completion determination step).
The personal computer C displays the completion of slag discharge on an output means such as a monitor, so that the operator can confirm the completion of the slag discharge work.
Details of each step will be described below.

(Hot water surface image capture step ST1)
In this embodiment, before performing the brightness value calculation step S1, the surface image capturing step ST1 is performed.
In the molten metal surface image capturing step ST1, first, the molten metal surface in which a plurality of slags S having different thicknesses are simultaneously floating on the surface of the molten metal M is imaged using the first imaging means 1, and the molten metal is captured. The surface image is stored in the storage means 15 of the personal computer C. FIG.
At this time, a plurality of hot water surfaces obtained at different times of the slag removing work using the slag removing machine 5 may be imaged, or the hot water surface in the slag removing work may be continuously imaged over time. .
As a specific example of a plurality of hot water surfaces obtained at different times of the slag removal work, for example, the hot water surface before starting the slag removing work (the hot water surface where the thickness of the slag S is considered to be the largest), the slag removal The melt surface in the middle of the work and the melt surface at the end of slag removal (the melt surface where the thickness of the slag S is considered to be the smallest) can be exemplified. In addition, since the slag quantification method of the present embodiment is used to ascertain the completion of slag removal (slag removal completion), the slag quantification method is sequentially implemented with the passage of time after the middle of the slag removal work. It is preferable to quantify and continuously grasp the amount of slag over time. Therefore, in the initial and middle stages of the slag removal work, the slag may be quantified as many times as required at the required time without continuous quantification. It goes without saying that the slag quantification may be performed continuously with the passage of time from the initial stage of slag removal to the end of slag removal.
Since the first imaging means 1 is equipped with both a CCD camera and a thermal pixel camera, the storage means 15 of the image processing device 2 stores an actual image captured by the CCD camera. Both thermal images (thermography) and thermal images (thermography) are recorded.

In addition, the aperture of the lens provided in the first imaging means 1, and The gain of the video signal output from the first imaging means 1 is adjusted.
Further, in this embodiment, the field of view of the first imaging means 1 is set so as to include not only the hot water surface of the container 4 but also the container 4 and the background within the field of view. Note that the field of view of the first imaging means 1 may be set so that only the hot water surface of the container 4 is included in the field of view.

Here, "luminance" in this specification refers to, for example, 255 gradation gradations of the 8-bit 256 gradation image brightness (that is, luminance on the image) excluding the background portion as black.
FIG. 4 is an actual image showing an example of a captured image obtained by capturing an image of the surface of hot water in the container 4 with the CCD camera of the first imaging means 1 . Although the display in FIG. 4 is black and white display, the actual image is a color image.
FIG. 4 shows a case where the field of view of the first imaging means 1 is set to a wide field of view including the container 4 and the background, and a plurality of slugs with different thicknesses float simultaneously on a single hot water surface. ing. In the example of FIG. 4, an obstacle existing between the first imaging means 1 and the surface of the hot water (in FIG. 4, the black belt-shaped area passing left and right near the center of the hot water surface has pipes and cannot be directly imaged The real image is divided into 10 rectangular pixel areas on the top, bottom, left, and right so as to exclude the area. Moreover, the position where the first imaging means 1 is installed is desirably above the center of the hot water surface in order to evenly photograph the hot water surface. This is an example of a case in which the first image capturing means 1 is installed at a position where the image can be captured from obliquely above the surface of the hot water in consideration of the above.

Since the surface of the hot water 4 is circular in plan view, as shown in FIG. 4, the surface of the hot water is drawn in an elliptical shape in the actual image photographed by the first imaging means 1 from obliquely above. A total of 10 rectangular pixel regions are defined by dividing each column into 5 in the left and right direction, excluding the portion of .
In FIG. 4, each pixel area is arranged so that the ratio of the background (the ratio of the black area) is large in the pixels located at the top, bottom, left, and right ends, and the ratio of the background is the smallest in the two pixels located above and below the center. ing. In FIG. 4, the solid black area indicates the background outside the surface of the hot water. In FIG. 4, the hot water surface is divided into 10 pixel regions for convenience, but the number of pixel regions is not limited to 10 and can be divided into any number of pixel regions.
Of course, in FIG. 4, 10 pixel regions were defined by excluding the black band-shaped region passing left and right near the center of the hot water surface. Any number of pixel regions may be defined, including the region of .

4 by the CCD camera, the image processing means 2 calculates the average luminance value for each of the above ten pixel regions, stores the average luminance value in the storage means 15, and displays it on the monitor. do. FIG. 4 shows a state in which the average luminance value of each pixel region is displayed on the monitor.
In the example of FIG. 4, in the five pixel regions in the upper row, 95.1, 123.6, 123.7, 160.8, and 136.7 are calculated in order from the left, and the five pixel regions in the lower row are calculated. In the pixel area, the calculated values are 41.3, 62.8, 114.3, 114.8, and 84.2 in order from the left. be done. The unit of brightness is (cd/m ² ).

In actual operation, the slag S having a uniform thickness does not uniformly exist on the molten metal M as shown in FIG. are doing.
If slags S with multiple thicknesses are non-uniformly present on the molten metal M, radiation light emitted from the molten metal M by thermal radiation is blocked by the slags S with different thicknesses.
The greater the thickness of the slug S, the greater the absorption of the emitted light by the slug S. Therefore, the transmitted light transmitted through the thick slug S and the transmitted light transmitted through the thin slug S have different intensities. As a result, in the captured image shown in FIG. 4, each pixel region has different brightness according to the thickness of the slag.

Since the first imaging means 1 has a thermal image camera (thermography) in addition to the CCD camera, the same range as in FIG. 4 can also be imaged as a thermal image.
FIG. 5 shows an example of a thermal image obtained by a thermal imaging camera. As in the case of the actual image shown in FIG. 4, the hot water surface is divided into 10 pixel regions. and display it on the monitor.
In the example of FIG. 5, in the five pixel regions in the upper row, 1047.6, 1111.7, 1126.7, 1170.1, and 1117.8 are calculated in order from the left, and the five pixel regions in the lower row are calculated. Since the pixel regions are calculated to be 763.6, 929.1, 1039.1, 1003.6, and 871.9 in order from the left, these average temperatures are recorded in the storage means 15 for each ten pixel regions. be. The unit of average temperature is (°C).

From the brightness of each pixel region obtained from the actual image shown in FIG. 4, the personal computer C calculates the average temperature of each pixel region using the temperature conversion formula described later, and uses it as the surface temperature of the slag in the next process. Alternatively, the average temperature of each pixel region obtained from the thermal image shown in FIG. 5 can be obtained and used as the surface temperature of the slag in the next process.
In a thermal image obtained when smoke or dust is generated, or when the slag S is scraped off by the scraping plate 51 to generate ripples on the hot water surface, each pixel may be affected by the smoke, dust, or ripples. It causes a large error in measuring the average temperature for each region. In other words, the hot water surface is hidden by smoke or dust, or the hot water surface temperature cannot be calculated accurately due to the influence of waves. will result in a large error in
For this reason, the average temperature of the thermal image is applied only when smoke, dust, or ripples do not occur. It is preferable to determine the surface temperature of the slag as the average temperature of each pixel region calculated from the luminance value as described below, and then perform the processing in the next step.

"Brightness value calculation step S1, slag surface temperature calculation step S2"
Let B be the radiance, σ (W/m ² K) be the Stefan-Boltzmann coefficient, T (°C) be the Celsius temperature, and n be the index of the Celsius temperature (generally n = 4). The relational expression of radiation heat transfer shown by the following equation (1) is known. Further, if the relational expression of radiation heat transfer is modified, the following expression (2) holds as a temperature conversion expression. In the formula (2), T _cal is the calculated Celsius temperature (°C), B _cal is the calculated radiance, DN is the brightness value, and a and b are the temperature of the thermal image per pixel region. A constant when satisfying the approximate expression of the relationship B _cal =aDN+b is shown as a value.

Assuming that there is no smoke or dust on the hot water surface, and the average temperature of the hot water surface obtained from the thermal image shown in FIG. The relationship shown in FIG. 6 exists between the radiance B _cal obtained from the equation and the brightness DN of each pixel region obtained from the actual image shown in FIG. 4, and the relationship shown in FIG. 7 exists between T and T _cal . holds. Note that n is calculated by substituting a plurality of values.
In FIG. 6, in addition to the average temperature of each pixel region shown in the real image shown in FIG. 4, a plurality of captured images of real images are actually taken, and their values are plotted and graphed.

From the multiple plots shown in FIG. 6, an approximation of the calculated radiance B _cal =aDN+b relationship can be obtained. In the example shown in FIG. 6, the correlation (approximate straight line) between B _cal =1243.1×DN+33294 and the coefficient of determination R ² =0.8434 is derived. That is, a=1243.1 and b=33294 are obtained.
Substituting these relationships in equation (2) yields the relationship between T and T _cal as shown in FIG. The relationship shown in FIG. 7 is derived as a high correlation with a coefficient of determination R ² =0.8441. These are obtained by substituting various numerical values for n in the equations (1) and (2) and finding the one with the highest approximation in the graph of FIG. Calculation for obtaining this approximation is performed by calculation software stored in the storage means 15 of the personal computer C, which calculates a plurality of relationships shown in FIGS. The surface temperature of slag S can be calculated.

When the average value of luminance for each region obtained from the actual image of FIG. to calculate the numerical values _necessary for other calculations, obtain the approximation as shown in the graphs shown in FIGS. This means that the case of high approximation was the case of n=4, and from this, the surface temperature T of the slag can be obtained. If it is found in this calculation that the approximation is highest when n=4, it is preferable to always substitute n=4 in subsequent calculations.

From these relationships shown in FIGS. 6 and 7, if the surface temperature T of the slag in each pixel region is calculated, the surface temperature T is used in the next step as the slag surface temperature Ts (° C.) for each pixel region. , the slag thickness can be calculated by performing the following slag thickness calculation step S3 using the following formula.
It is preferred that each pixel area data is tri-valued according to luminance, each pixel area is classified into molten metal M, slag S, and other background, and it is determined that slag S exists in the corresponding pixel area.

"First judgment step DS1"
Note that when the surface temperature T of the specific N-th pixel region is obtained by calculation, the slag S may not exist depending on the pixel region, and in this case the surface temperature is not the temperature of the slag. For this reason, the surface temperature of this pixel region (corresponding N-th pixel) may not be the surface temperature of the slug S, so the temperature range that the slug S can take is defined as a range of Tmin or more and Tmax or less. If the surface temperature T does not fall within this temperature range, the surface temperature T moves to the next pixel range (N+1) in the first determination step DS1 to find the brightness of the next pixel range. The temperature calculation step S2 is repeated. That is, when the surface temperature T is in the range of Tmin or more and Tmax or less, the following steps are performed.
As an example, Tmax can be set at 1000°C as the estimated temperature at which the slag solidifies, and Tmin can be set at 800°C.

"Slag thickness calculation step S3"
FIG. 8 schematically illustrates a state in which a slag having a thickness of d (m) floats on hot metal (molten metal) in one of the ten pixel regions shown in FIG. 3 or 4. It is a sectional view.
where ε is the emissivity, F is the view factor, Ts is the slag surface temperature (°C), _T∞ is the ambient temperature (°C), d is the slag thickness (m), and k is the slag thermal conductivity (W/m・K) and T _F as the molten metal temperature (°C), the following heat conduction equation (3) holds.

Further, where W is the slag weight (kg), ρ is the slag density (kg/m ³ ), N _total is the total number of pixel regions, and ΔA is the molten metal surface area per pixel region, the following equation (4) is obtained. A formula for estimating the amount of residual slag holds. In the following equation (4), W is the slag weight (kg), ρ is the slag density (kg/m ³ ), N _total is the total number of pixel regions, and ΔA is the melt surface area per pixel region.

In the above equations (1) to (4), as an example of actual measurement data during actual operation, the molten metal temperature (T _F ) is 1350° C., the ambient temperature (T _∞ ) is 13° C., The Stefan Boltzmann coefficient (σ) is 5.67 × 10 ^-8 (W/m ² · K), the slag emissivity (ε) is 0.4, and the slag thermal conductivity (k) is 0.5 (W/m・K), slag density (ρ) of 2500 to 2600 (kg/m ³ ), shape factor (F) of 1, etc. can be used for calculation as default values. As for the form factor, 1 can usually be substituted assuming that the surface of the slag is infinitely open to the space.
In equation (3), the left side can be considered as a general equation for radiation heat transfer, and the right side is the amount of heat transferred in the slag, which can be considered to follow Fourier's law. , (3) holds.

Solving equations (3) and (4) as equations reveals that W is the slag weight (kg). The slag weight for each pixel region can be obtained by calculation from this luminance value.

"Second determination process DS2 and slag calculation process S4"
When calculating the slag thickness of each pixel region by calculation, if the slag thickness of all pixel regions can be calculated, the next slag calculation step S4 is performed.
If the thickness of the slag in all pixel regions has not been calculated, in the second determination step DS2, the process returns to before the step S1, and in the next pixel region, the brightness value calculation step S1, the slag surface temperature calculation step S2, and the slag thickness calculation are performed. Repeat step S3. The slag weights of all pixel regions are recorded in the storage means 15 of the personal computer C as information in sequence.

"Slag calculation step S4"
Once the slag weights of all pixel regions have been determined, summing all the slag weights determined for each pixel region gives the total slag weight (W _cal ) as the total amount of all slag present on the molten metal M. can be asked for.
When the slag weight (W _cal ) obtained by the above calculation satisfies the condition (W _cal <W _limit ) of the slag limit weight (W _limit ) defined as the allowable range, it can be determined that the slag removal is completed.
Alternatively, when the slag volume (V _cal ) obtained by the above calculation satisfies the condition (V _cal <V _limit ) of the slag limit volume (V _limit ) defined as the allowable range, it can be determined that the slag removal is completed.

The slag limit weight (W _limit ) or slag limit volume (V _limit ) is the volume and weight of the molten metal M contained in the vessel 4 during actual operation, so if the vessel 4 is a hot metal ladle, For example, since slag is generated by the reaction between the molten iron and quicklime, iron oxide, and other materials put into the molten iron, if the components of the molten metal M contained in the container 4 are grasped in advance by sample analysis or the like, the molten iron Since the amount of slag generated on the metal M can be clarified by calculation, the slag critical weight or slag critical volume can be determined from this calculated slag amount.
Although the weight of the slag is calculated in the above calculation, the density of the slag is known as described above, so the volume of the slag can also be calculated from the weight, density and thickness of the slag.
For example, a slag limit weight (W _limit ) or a slag limit volume (V _limit ) is set for the purpose of not lowering the yield of molten metal in the slag discharge process. The slag limit weight is the amount of slag generated estimated from the total amount of input into the hot metal ladle for the volume and weight of the molten metal M stored in the vessel 4 that is known during actual operation, and the amount of slag generated in the past. It can be obtained from the product of the amount of generated slag and the standard slag rate, using a standard slag rate determined as 0.7 or the like investigated by the test.
Similarly, the slag critical volume can be obtained as a value obtained by dividing the slag critical weight by the slag density of 2550 (kg/m ³ ).
The slag limit weight can be defined as W _limit = amount of desulfurization slag x (1 - target slag rate). For example, the target slag removal rate can be set at 0.7 because it is considered that the hot metal yield will deteriorate if the slag removal rate exceeds 0.7 based on the standard for actual operation.
When the amount of desulfurization slag generated is 2000 t/ch, W limit can be obtained from the above formula as W _limit =600 t/ch. Similarly, the slag limit volume can be calculated as V _limit =235 m ³ /ch by dividing the slag limit weight by the slag density of 2550 (kg/m ³ ).

For example, the sulfur component in the molten metal M in the vessel 4 is desulfurized by adding a desulfurizing agent (flux), and the sulfur component in the molten metal is transferred to the generated slag. It is generally known that the residual rate has a correlation with the sulfur component content in the molten metal in the post-process.
If sample analysis is used in the preceding process of the slag operation, the sulfur component content in the molten metal before slag generation and the sulfur content in the molten metal after slag generation (after desulfurization refining) can be determined. Based on the results of both measurements, it is possible to calculate the content of sulfur components in the slag in the vessel before starting the slag removal work (difference in content × amount of molten metal = in slag sulfur content).

According to the present embodiment, the content of sulfur components in the slag in the container after the start of the slag discharge work is calculated from the content of the sulfur components in the slag in the container before the start of the slag discharge work calculated as described above. It is calculable. Then, the calculated content of sulfur components in the slag in the vessel after the start of the slag discharge work and the content of sulfur components in the molten metal after desulfurization refining calculated using the sample analysis as described above are calculated. It is possible to predict the content of sulfur components in the molten metal during treatment in the post-process, based on the resulfurization rate from slag to molten metal during treatment in the post-process.

"Comparison step S5"
The value of the slag limit weight or slag limit volume is entered as a threshold in the storage means 15 of the personal computer C shown in FIG.
The calculated values of the slag limit weight or the slag limit volume described above are sequentially calculated with the lapse of time during the slag discharging operation, and stored in the storage means 15 of the personal computer C as calculation information together with time information.
In the comparison step S5, if the slag weight (W _cal ) obtained by the above calculation does not satisfy the condition (W _cal <W _{limit ) of the slag limit weight (W limit )} defined as the allowable range, the slag discharging operation is continued. .
As the slag discharge operation progresses and the amount of discharged slag increases, at some point the slag weight (W _cal ) obtained by the above calculation reaches the slag limit weight (W _limit ) condition (W _cal < W _limit ) is satisfied.
Here, the prediction means 18 provided in the personal computer C determines that the slag removal is completed when the above conditions are satisfied (slag removal completion judgment step), and the control unit 14 outputs the slag removal completion to the output means 16 such as a monitor. is displayed. This allows the operator of the slag-discharging operation to determine that the slag-discharging is complete.

That is, the computer C captures an image of the surface of the molten metal M contained in the container 4 by the first imaging device 1, and based on the captured image, the slag on the molten metal M contained in the container 4 is detected. It is equipped with a program that sequentially calculates and stores the volume or total slag weight over time. Then, this program comprises a computer, a luminance value calculating means for obtaining a luminance value in each pixel region of the captured real image by image processing, and the above-mentioned ( 1) A slag surface temperature measuring means for obtaining the surface temperature of the slag in each pixel region from the relational expression of radiation heat transfer shown by the above equation and the temperature conversion equation shown by the above equation (2), and the slag surface temperature measuring step slag thickness calculation means for estimating the thickness of the slag for each pixel based on the slag surface temperature for each pixel region obtained by and the heat conduction equation represented by the above equation (3); Based on the area and the physical property values of the slag, the slag volume or the total weight of the slag in the container is calculated sequentially using the slag residual amount estimation formula shown in the above formula (4) with respect to the passage of time. It has the function to function.

In addition, since the thickness of the slag S on the molten metal M can be calculated by the method described above, the vertical position of the scraping plate 51 can be accurately positioned during the slag removing operation using the scraping plate 51. .
For example, the surface of the hot water in the container 4 can be imaged from an oblique direction by the second imaging means 3, and an imaged image showing both the slag S and the scraping plate 51 can be obtained. When the container 4 is tilted, the height of the top surface of the molten metal M and the spout 4a of the container 4 are substantially the same, so that the height of the spout 4a and the height of the lower surface of the scraping plate 51 are substantially the same. By setting the height, the height of the lower surface of the scraping plate 51 can be made to match the height of the upper surface of the molten metal M. Alternatively, since the thickness of the slag S is known, the position of the lower end of the scraping plate 51 when scraping off the slag S while the lower end of the scraping plate 51 is buried in the slag S can be matched with the height of the upper surface of the molten metal M.
Since the distance H0 between the lower surface of the arm 52 and the lower surface of the scraping plate 51 is known in advance, the thickness d of the slag S at the portion where the lower end of the scraping plate 51 is buried is given by the relational expression d=H0-H1. , the scraping work can be performed while the vertical position of the scraping plate 51 is accurately positioned.

Next, the average brightness of the actual image is obtained according to the above-mentioned pixel area, the slag surface temperature is calculated by calculation, and the slag weight is calculated by the above calculation, and the slag surface obtained by the thermal image Robustness evaluation when calculating the slag weight by the above calculation using the temperature will be described.
In FIG. 9, the vertical axis plots the slag area ratio on the actual image shown in FIG. The result of plotting the slag area ratio assuming 1000° C. on the horizontal axis is shown.

One plot shown in FIG. 9 shows the measurement result of the slag area ratio in one hot metal ladle during actual operation. In FIG. 9, the plots indicated by circles are the cases where the slag area ratio was measured during actual operation, and smoke, dust, or ripples on the surface of the hot water were visually confirmed. This is the case when the When smoke or dust occurs, the heat of the smoke or dust is detected and thermally imaged, so there is a high possibility that the temperature lower than the hot metal temperature will be observed. change, it is thought that an error will occur in the measurement result. The rippling of the hot water surface is the rippling generated when the scraping plate 51 scrapes off the slag S during the slag discharging operation.

FIG. 10 is a graph showing only plots excluding the cases of smoke generation or water surface ripples shown in FIG. The coefficient of determination ^R2 obtained from the plot shown in FIG. 9 is 0.3857, indicating a low correlation, while the coefficient of determination ^R2 obtained from the plot illustrated in FIG. is high, it can be seen that eliminating the thermal image disturbance improves the correlation between the area ratio of the actual image and the area ratio of the thermal image.
From this, it can be seen that it is necessary to eliminate the influence of disturbances such as smoke, dust, and water surface ripples when calculating the slag area ratio with a high degree of accuracy using a thermal image. From this result, it can be estimated that the real image has higher robustness than the thermal image, and it can be estimated that the real image is superior to the high-precision calculation of the slag area ratio. In the case of a thermal image, dust generation and smoke generation are also largely captured as heat, which causes temperature measurement errors.

When imaging the surface of the molten metal M, dust may be generated from the surface of the molten metal M. When dust is generated, if an image is captured by an imaging device having a wavelength in the middle to far infrared region, the dust is regarded as a hot object. There is a risk of erroneous detection, and in order to avoid this, it is considered desirable to perform imaging with an imaging device having a wavelength in the visible light to near-infrared region (2.5 μm or less).
In addition, if the surface of the hot water is rippling, the error in the radiance tends to be relatively large when the image is taken with an imaging device having a wavelength in the middle to far infrared region. It is considered desirable to take an image with an imaging device having a wavelength in the visible light to near-infrared region (2.5 μm or less). As described above, when the angle of the surface to be measured changes due to ripples, the emissivity changes accordingly, and it can be seen that there is a possibility that an error may occur in the obtained radiance.

FIG. 11 shows the slag area ratio (S _cal ) for the molten metal yield loss and recovery [S] (S component remaining in the molten steel after operation in the post-process converter) in actual operation. FIG. 10 is a diagram showing comparative evaluation results of respective correlations between a case where the slag discharge work is performed by judging that the slag weight (W _cal ) is judged to be complete and the slag discharge work is carried out.
In FIG. 11, yield loss ratio=yield loss (t/ch)/maximum yield loss (t/ch), return [S] ratio=return [S] amount (%)/maximum return [S] amount (%) , slag area ratio=slag area ratio (%)/maximum slag area ratio (%), and slag weight ratio=slag weight (kg)/maximum slag weight (kg). Each maximum value is shown in relative evaluation based on the maximum value obtained during operation.

In FIG. 11, the graphs arranged vertically on the left side show the yield loss, and the yield loss is compared by the slag area ratio and the slag weight. It became larger than the correlation coefficient when operated by the area ratio.
In FIG. 11, the graphs arranged on the upper and lower sides of the right side are the comparison of the return [S], and the variation of the return [S] when operating by slag weight calculation is the same as the return [S] when operating by the slag area ratio. was smaller than the scatter of As for the return [S], since there is a target value, it is desirable that the variation is as small as possible. Since it is said that the amount of slag remaining after slag discharge affects recovery [S], it can be seen that there is less variation when operating by calculating the slag weight.

In the embodiment described above, the case where the molten iron as the molten metal M is stored in the container 4 is taken as an example, and the case where the slag on the molten iron is quantified has been described.
However, the present invention can also be applied to the case of quantifying the slag generated in the process of refining molten pig iron in a converter into molten steel.
By imaging the slag on the molten steel with the first imaging device 1 while the converter is not blowing, and performing the same image analysis and calculation as in the above-described embodiment, the slag on the molten steel can be quantified. can. Since slag may be discharged by the tilting cylinder in the converter, we will quantify the amount of slag in the converter and grasp the slag weight to develop a technology that contributes to the judgment of the completion of slag removal (judgment of slag removal completion). can provide.

In the embodiment described above, the storage means 15 of the personal computer C stores a slag quantification program as described above.
As an example, this quantification program uses a personal computer C to image the surface of the molten metal M contained in the container 4 by the image capturing means 1 and 2, and the luminance value in each pixel region of the imaged image is imaged. The luminance value calculating means obtained by the process and the relationship between the luminance value and the radiance are obtained in advance as described above, and the luminance value obtained in the luminance value calculating step is used from the relationship and the temperature conversion formula described above. a slag surface temperature calculating means for obtaining the surface temperature of the slag M in each pixel region; and a slag calculating means for obtaining the slag volume or total slag weight in the container based on the melt surface area per pixel area and the physical property values of the slag. can.

Then, when the slag weight or slag volume obtained as described above reaches the slag limit weight or slag limit volume with the lapse of time, the control unit 14 displays completion of the slag discharging work on the output means 16 such as a monitor. . This allows the operator of the slag-discharging operation to determine that the slag-discharging is complete.

230 tons of hot metal is put into a cylindrical hot metal ladle with a diameter of about 4m and a capacity of about ^36m3 . was installed, and the actual image shown in FIG. 4 and the thermal image shown in FIG. 5 were taken.
The actual image and the thermal image were divided into 10 pixel regions as shown in FIGS. 4 and 5, and the conversion temperature (Ts: slag surface temperature:° C.) was obtained by the temperature conversion formula based on the following equation (5). The calculation results are shown in Table 1 below.

Next, based on the conversion temperature Ts, the slag thickness was calculated by the following equation (6) and the heat conduction equation based on the equation (7) obtained by modifying the equation (6). The results are shown in Table 1 below.

From these calculation results, the slag weight was calculated based on the following formula (8) for estimating the total amount of slag. The results are shown in Table 1 below.

As the calculation results shown in Table 1 show, the total slag weight is calculated from the temperature conversion formula (5), the modified heat conduction equation (7), and the total slag weight calculation formula (8). It turns out that it can be calculated. That is, the luminance value DN is obtained from the actual image captured by the first imaging device, and the slag surface temperature (converted temperature: ° C.) can be obtained by calculation from this luminance value DN. From the slag surface temperature, the slag It was found that the thickness (m) of the slag can be calculated, and the weight (kg) of the slag can be calculated from the slag thickness and the physical property values of the slag described above.

DESCRIPTION OF SYMBOLS 1... 1st imaging means, 2... image processing apparatus, 3... 2nd imaging means, 4... container, 4a... spout, 5... slag exhaust machine, 100... slag quantification apparatus, M... molten metal, S... slag , d... Slag thickness, C... Computer, 13... Input means, 14... Control section, 15... Storage means, 16... Output means, 17... Calculation means, 18... Prediction means, 51... Scraping plate, 52... Arm.

Claims (7)

A slag is present on the molten metal contained in the container, and the slag covers the light emitted from the molten metal according to its thickness, and the light emitted from the molten metal is emitted to the outside as transmitted light transmitted through the slag. Take an image of the surface of the slag as a hot water surface with an imaging device,
a luminance value calculation step of obtaining a luminance value in each pixel region of the captured image by image processing;
The relationship between the luminance value and the radiance is obtained in advance, and the apparent surface temperature of the slag in each pixel region is obtained from the relationship and the temperature conversion formula using the luminance value obtained in the luminance value calculation step. A slag surface temperature calculation step;
a slag thickness calculation step of estimating the slag thickness of each pixel region from the apparent slag surface temperature in each pixel region obtained by the slag surface temperature calculation step and a slag thickness estimation formula ( heat conduction equation ) ;
A slag calculating step of calculating the slag volume or total slag weight in the container based on the slag thickness of each pixel region, the surface area of each pixel region , and the density of the slag obtained by the slag thickness calculating step. A method for quantifying slag characterized by: A slag is present on the molten metal contained in the container, and the slag covers the light emitted from the molten metal according to its thickness, and the light emitted from the molten metal is emitted to the outside as transmitted light transmitted through the slag. Take a thermal image with an imaging device with the surface of the slag as the hot water surface,
a slag surface temperature calculating step of obtaining an apparent slag surface temperature in each pixel region from the captured thermal image;
a slag thickness calculation step of estimating the slag thickness of each pixel region from the apparent slag surface temperature in each pixel region obtained by the slag surface temperature calculation step and a slag thickness estimation formula (heat conduction equation);
A slag calculating step of calculating the slag volume or total slag weight in the container based on the slag thickness of each pixel region, the surface area of each pixel region, and the density of the slag obtained by the slag thickness calculating step. A method for quantifying slag characterized by: 2. The slag quantification method according to claim 1, wherein the following equations (1) and (2) are used in the slag surface temperature calculation step.

(However, in formula (1), B is the radiance, σ is the Stefan Boltzmann coefficient (W / m ² K), T is the Celsius temperature (° C.), n is the index of the Celsius temperature, and T in the formula (2) _cal is the Celsius temperature (° C.) obtained by calculation, B _cal is the radiance obtained by calculation, DN is the luminance value, and a and b are thermal images taken with the surface of the slag as the hot water surface, and the one pixel area The surface temperature of the hot water surface obtained from the thermal image of the hit is grasped as the true value of the apparent surface temperature of the slag per pixel area , and the radiance B _cal obtained from the above equation (1) and each pixel area A constant represented by an approximation straight line of B _cal =aDN+b in which a plurality of values are grasped in advance for the relationship between Δ and the luminance DN .) The slag thickness estimation formula (heat conduction equation) for obtaining the thickness of the slag is the following formula (3), and the formula for obtaining the total slag weight is the following formula (4). 4. The slag quantification method according to any one of 3.

(where, in equation (3), ε is the emissivity, F is the view factor, T _s is the apparent surface temperature of the slag (°C), T _∞ is the ambient temperature (°C), d is the slag thickness (m), and k is Slag thermal conductivity (W/m K), T _F is molten metal temperature (°C), W is slag weight (kg), ρ is slag density (kg/m ³ ), N _total is total The number of pixel regions and ΔA indicate the molten metal surface area per pixel region.)
5. The slag quantification method according to any one of claims 1 to 4 , wherein in said slag calculating step, the slag volume or total slag weight in said container is determined over time. When the slag volume or total slag weight in the container obtained by the slag calculation step satisfies the slag volume allowable range or the slag weight allowable range defined as the allowable range, the slag removal completion determination step of determining that the treatment is completed or the slag removal is completed. The method for quantifying slag according to claim 5 , characterized by comprising : The method for quantifying slag according to any one of claims 1 to 6, wherein observation light in a wavelength range from visible light to near-infrared region is used when imaging the surface of the molten metal.

Images