JP2521512B2 - Detection method of charging material flow section in blast furnace - Google Patents

Description

TECHNICAL FIELD The present invention relates to a method for detecting a fluidized portion on the surface of a raw material in a blast furnace.

[Prior Art] On the surface of the raw material in the furnace of the blast furnace, there is a region (hereinafter referred to as the fluidized portion) in which the raw material of high brightness is moved by the gas flow in the furnace, and the position and area are measured and based on this data The calculated index value (fluidization index) can be used to grasp the gas flow in the furnace and plays an important role in operation.

Conventionally, as a method of measuring the position and area of the raw material flowing part, a method of visually observing the position and area of the raw material flowing part by observing the inside of the blast furnace with an optical fiber scope or a television camera has been developed. As an example of this, Japanese Patent Publication No. 62-33283
There is a method disclosed in the issue public information. In this method, the charged raw material is observed using an optical fiber, the image is recorded on a video tape recorder by a video camera, and the recorded image data is analyzed by an image analysis device to obtain the flow area of the furnace center. It was done like this.

[Problems to be Solved by the Invention] However, the conventional detection method has the following problems.

(1) It is necessary for a human to manually trace the boundary of the fluidized portion with a light pen or the like.

(2) Since the optical fiber scope is used, the field of view is limited to a narrow range.

(3) Real-time processing is impossible because an image is once taken on a video tape and is newly put in an image processing device for analysis.

(4) Since the movement of the optical fiber is restricted to the one-dimensional movement, it is impossible to measure the entire surface of the blast furnace interior, which is a two-dimensional plane.

(5) It is difficult to measure the area of the fluidized portion with high accuracy because the size of the image varies depending on the distance between the tip of the optical fiber and the surface of the charging material even in the fluidized portion having the same area.

Due to the above-mentioned problems, it has been difficult to obtain the fluidized portion by full-time real-time processing particularly in actual operation.

The present invention has been made to solve the above problems, and the raw material in the blast furnace capable of detecting the area of the fluidized portion of the raw material surface in the blast furnace in a fully automatic and accurate manner in real time. The purpose is to obtain a method for detecting a fluidized portion.

[Means for Solving the Problem] The method for detecting the charging raw material flow section in the blast furnace according to the present invention obtains an image signal of the charging raw material surface by a television camera installed at the blast furnace furnace opening, For each pixel of the signal, the maximum value or average value within a fixed time is calculated, and the difference between the maximum value or average value between the pixels is calculated to determine the spatial variation of the two-dimensional frame formed by the pixels. It is calculated, and the portion in which the spatial variation amount is smaller than the first specified value is determined to be the flowing portion (claim 1). Further, when determining the flowing portion, a portion in which not only the spatial fluctuation amount is smaller than the first specified value but also the magnitude of the image signal is larger than the second specified value is determined to be the flowing portion. 2). Further, the spatial variation of the two-dimensional frame is obtained by spatial differentiation in which the difference between the maximum value or the average value between pixels is differentiated based on the distance between pixels (claim 3).

The calculation of the maximum value or the average value is started when it is detected that the variation amount between pixels of the image signal on the surface of the charging raw material obtained by the television camera is larger than the third specified value. , In other cases, the image signal at that time is not used for measurement (claim 4).

Then, after determining the flow part, the ratio of the flow part to the image area for one screen captured by the television camera is calculated, and the area corresponding to the image area for one screen is multiplied by the ratio to calculate the flow. The area of the part is calculated (claim 5). Further, the distance between the TV camera and the surface of the charging raw material in the image pickup direction of the TV camera (direction of the line of sight of the TV camera) is measured, and the ratio of this measured distance to a preset reference distance is calculated as described above. The area of the fluidized portion is corrected by multiplying the area of the fluidized portion (claim 6). Alternatively, the distance between the TV camera and the surface of the charging raw material in the image pickup direction of the TV camera is measured, and the zoom lens of the TV camera is operated so that the magnification becomes large in proportion to the measured distance,
The area corresponding to the image area for one screen is always constant (claim 7).

[Operation] In this invention, since the red hot coke is moving in the raw material flowing part, when the maximum brightness or the average brightness of each part is measured for a certain long time (several seconds), it is averaged in the flowing part. On the other hand, since the position of the red hot coke other than the flowing part does not move much within this time, the fact that the maximum brightness or average brightness of each part is uneven is used to change the maximum brightness or average brightness. A region below a predetermined specified value is detected as a fluidized portion.

[Embodiment] A method for detecting a raw material flowing portion in a blast furnace according to an embodiment of the present invention will be described with reference to the drawings.

FIG. 2 is a system configuration diagram for implementing this embodiment. In the figure, 1 is a blast furnace, 2 is a raw material surface in the blast furnace, 3
Is a television camera, 4 is a television camera control device, 5 is a television monitor, 6 is a fluidized portion detection device, 7 is a CRT display attached to the fluidized portion measuring device, 8 is the same printing device, 9 is a raw material level meter in the furnace (for example, Saudi Arabia) Level meter).

The television camera 3 is installed in the window of the top of the blast furnace 1 so that the raw material surface 2 in the furnace can be seen. TV camera 3
With respect to the angle, the angle is changed in the horizontal direction and the vertical direction by the end camera controller 4 so that most of the raw material surface 2 in the furnace can be seen. Also, the television camera control device 4 has a fluidized portion measuring device 6 as will be described later.
The zoom magnification of the image obtained by driving the zoom lens of the television camera 3 is controlled according to the distance signal transmitted from the.

The video signal from the television camera 3 is monitored by the television monitor 5 as an image and is input to the fluidized portion detecting device 6.

FIG. 1 is a block diagram showing the configuration of the fluidized portion detection device 6. In the figure, 10 frame memory, 11 is a measurable image selection device, 12 is a maximum brightness selection device, 13 is a brightness signal uniform part identification device, 14 is a high brightness part identification device, 15 is a fluid part identification device, and 16 is a distance. It is a correction / fluid part indexing device.

In the frame memory 10, the image signal of the surface of the charging material photographed by the television camera 3 is converted into a digital signal indicating the brightness level and stored.

The measurable image selection device 11 reads out the signal stored in the frame memory 10, calculates the variation of the signal line by line or over one screen, for example, and determines whether or not the signal is equal to or more than a specified value. The determination method may be, for example, calculating the standard deviation of the signal to determine whether it is equal to or greater than a specified value,
Alternatively, a power spectrum may be obtained by a real-time Fourier transform, which is a well-known technique, and it may be determined whether a frequency component higher than a predetermined frequency is equal to or higher than a specified value.

The measurable image selection device 11 determines that the measured image signal is reliable only when the fluctuation of the image signal over the entire screen is a specified value or more, and the maximum brightness selection device 12 connected to the next stage. Send data to.

For example, when dust is generated in the furnace, a clear image of the raw material surface cannot be obtained, but the conversion of the signal is determined to be small by the above processing, and the signal is processed by the following maximum brightness selection device 12 etc. Since it is not used for, it is not affected by the disturbance caused by dust. According to this method, not only the dust that lowers the brightness level of the screen but also the disturbance caused by the high-intensity dust blown from the lower part of the furnace can be removed.

A signal from the measurable image selection device 11 is input to the maximum brightness selection device 12, and the maximum value for each pixel is obtained. That is, the maximum brightness selection device 12 has a memory corresponding to each pixel, and at the beginning of the period, this memory is preset to a value corresponding to the minimum brightness. Thereafter, each time new data is input, the new data is compared with the contents of the memory, and if the new data is larger, the contents of the memory are replaced with the new data and updated. If the contents of the memory are read out after a certain period of time, that value corresponds to the maximum luminance signal during that period.

FIG. 3 is a characteristic diagram showing the maximum luminance level in the predetermined direction (X-axis direction) created by the above-described arithmetic processing.

In this embodiment, the maximum brightness other selection method is shown, but it is clear from the basic principle of the present invention that the object can be achieved by a method of calculating an average value for each pixel instead. In this case, the memory for each pixel is reset at the beginning of the period, added to the memory each time new data is input, and the number of data items is stored. The average of each pixel may be obtained from the value of. Therefore, in this case, the average value calculating device is used instead of the maximum brightness selecting device.

The signal from the maximum brightness selection device is input to the brightness signal uniform portion identification device 13, and the spatial variation amount is calculated here. As a method of calculating the amount of spatial variation, a spatial differential amount (two-dimensional or one-dimensional), which is a well-known technique in image processing, may be performed for each pixel, or for example, within a rectangle centered on a certain pixel. You may calculate the standard deviation of the signal value of a pixel. The spatial fluctuation amount calculated in this way is compared with a specified value for each pixel, and a portion equal to or less than the specified value is discriminated as a luminance signal uniform portion. In the characteristic diagram shown in FIG. 3, the area A where the signal change is small is determined to be the luminance signal uniform portion.

In the basic embodiment of the present invention, the brightness signal uniform part is a fluid part, but in the embodiment shown in FIG. 1, the maximum brightness is obtained by the high brightness part identification device 14 described below, and the brightness is further increased. Among the signal uniform parts, the one whose absolute value of brightness is equal to or more than the specified value is used as the fluid part to improve reliability. This is so that if a part of the visual field is dark and the brightness does not change, this part is not mistakenly determined to be a flowing part.

The brightness signal of the frame memory 10 is input to the high-intensity part identifying device 14, and the signal is compared with a specified value for each pixel, and a signal exceeding the specified value is determined as a high-intensity part. The input signal to the high-luminance part identification device 14 does not necessarily have to be the signal from the frame memory 10, and the signal from the maximum-luminance selection device 12 (or the average value calculation device) may be input.

The output of the brightness signal uniform part identification device 13 and the output of the high brightness part identification device 14 are input to the flowing part identification device 15, and as described above, the absolute value of the brightness of the brightness signal uniform part is not less than the specified value. Find the flow part and the region of the flow part. FIG. 4 is an explanatory view (white portion of the figure) showing an image when the fluidized portion at this time is extracted, and the number of pixels of the fluidized portion is counted,
By the count α determined by the focal length of the camera lens,
The flow section area S ′ is obtained by the following equation (1).

S ′ = α · C (1) Here, S ′ is the fluid area and C is the fluid pixel number.

The above-mentioned flow area S ′ is input to the distance correction / fluid part indexing device 16, and the surface level of the in-reactor raw material from the in-reactor raw material level meter 9 and the camera focal length from the television camera control device 4 are also input. Then, the next arithmetic processing is performed.

Generally, the distance between the camera and the measurement surface of the raw material in the furnace is approximated by a function of the angle θ in the horizontal direction of the camera, the angle φ in the vertical direction, and the surface level H of the raw material in the furnace. The in-furnace raw material surface level H is measured by the in-furnace raw material level meter 9. Generally, the surface of the raw material in the furnace is in the shape of a skirt as shown in FIG. 2, and when the central axis of the blast furnace is the Z axis and the x axis and the y axis are orthogonal to the radial direction, z = Z (x, y, H )… (2) on the surface. The angle θ between this plane and the camera,
The intersection between the line of sight and the field of view of the camera, which is determined by φ and the installation position of the camera, is analytically obtained, and the distance between the point and the installation position of the camera is obtained. The flow section area S is obtained by S = l / l ₀ · S ′ (3) with the camera reference distance as l ₀ .

The distance correction in the distance correction / fluid part indexing device 16 may be performed by the following method in addition to the above method. That is, this is a method of normalizing the size of the image in advance by using the zooming of the television camera 3, and the distance 1 between the camera and the raw material measurement surface in the furnace is calculated by the above-described method, and the television is calculated according to the distance. The zoom lens of the camera is driven so that an image with a constant magnification is always obtained regardless of the distance. By doing so, the flow section area S can be obtained without performing a correction calculation such as the equation (3).

The fluidization index is calculated by the fluidization index defined by the following formula.

F = S / T (4) where F is the fluidization index, S is the area of the fluidized portion, and T is the area of the region of the surface of the raw material taken by the television camera.

In the flow section detecting device 6 configured as described above,
The original image from the TV camera 3 is stored in the memory frame 10 as a binarized signal according to its brightness level, and the measurable image selection 11 is such that the fluctuation of the image signal stored in the memory frame 10 is more than a predetermined specified value. At some point, the measurement data is sent out as reliable image data. At this time, when the fluctuation is less than a predetermined value, the image data is not sent out and is not used for the calculation of the next maximum value (or average value). The maximum brightness selection device 12 obtains the maximum brightness signal (or average value) for each pixel, and the brightness signal uniform part identification device
13 calculates the amount of spatial variation of the maximum luminance signal (or average value) and obtains the luminance signal uniform portion. On the other hand, the high-luminance part identification device 14 obtains a high-luminance part, and the fluid part identification device 15 determines that the part corresponding to the high-luminance part region in the uniform luminance signal part is the fluid part, and the area S '
Is required.

The distance correction / fluid part indexing device 16 is based on the area S ′, the distance between the intersection of the field of view of the TV camera and the surface of the charging material and the TV camera (proportional to the blast furnace charging level), and the reference distance of the camera. Then, the fluidization index F is obtained using the equations (2), (3) and (4).

The fluidizing section detecting device 6 obtains the fluidization index F in real time, and the output is output to the CRT display 7 and the printing device 8.

[Effects of the Invention] As described above, according to the present invention, since the red hot coke moves in the fluidizing section and the maximum brightness or the average brightness thereof is averaged, when the fluctuation value is equal to or less than the predetermined reference value, Since the area is determined to be the fluidized portion, the fluidized state of the raw material in the blast furnace can be measured in real time.

In addition, when determining the moving part, it is determined that the part where the variation value is less than or equal to the predetermined specified value and the magnitude of the pixel signal is greater than or equal to the predetermined specified value is the flowing part. Reliability has improved.

Further, when the fluctuation of the image signal is smaller than the specified value, the signal is not used for the calculation so that the measurement error due to the dust in the furnace does not occur.

In addition, the area of the flow section is corrected based on the ratio of the distance between the TV camera and the intersection of the line of sight of the TV camera with the charging material surface, and the reference distance. The effect of being able to measure accurately without being affected by is obtained.

And so on.

[Brief description of drawings]

FIG. 1 is a block diagram of a fluidized-bed detecting device for carrying out a method according to an embodiment of the present invention, FIG. 2 is a system configuration diagram of related equipment including the device, and FIG. 3 is a description of a brightness profile. FIG. 4 and FIG. 4 are explanatory diagrams of the fluidized portion extraction image. 1 ... Blast furnace, 2 ... Blast furnace charging surface, 3 ... TV camera, 4 ... TV camera control device, 5 ... TV monitor, 6 ... Flow part detection device, 7 ... CRT display, 8
...... Printer, 9 …… Furnace material level meter

Claims (7)

(57) [Claims] 1. An image signal of a surface of a charging raw material is obtained by a television camera installed at a blast furnace throat, and a maximum value or an average value within a predetermined time is calculated for each pixel of the image signal, and Obtaining the difference between the maximum value or the average value between the pixels,
A device in a blast furnace, characterized in that a spatial variation amount of a two-dimensional frame formed by the pixels is calculated, and a portion where the spatial variation amount is smaller than a first prescribed value is determined as a fluidized portion. Method for detecting the flow part of the input material. 2. An image signal of the surface of the charging raw material is obtained by a television camera installed at the blast furnace throat, and a maximum value or an average value within a fixed time is calculated for each pixel of the image signal, and Obtaining the difference between the maximum value or the average value between the pixels,
A spatial variation amount of a two-dimensional frame formed by the pixels is calculated, the spatial variation amount is smaller than a first prescribed value, and the magnitude of the image signal is larger than a second prescribed value. A method for detecting a charged material flowing part in a blast furnace, wherein a part is determined as a flowing part. 3. A spatial variation of the two-dimensional frame,
The method for detecting a charging material flow part in a blast furnace according to claim 1 or 2, wherein the difference between the maximum value or the average value between the pixels is obtained by spatial differentiation which is differentiated based on the distance between the pixels. 4. When it is detected that the amount of variation between pixels of the image signal on the surface of the charging raw material obtained by a television camera is larger than a third specified value, from that time, each pixel of the image signal is detected. The arithmetic processing for calculating the maximum value or the average value within a fixed time is started every time.
Alternatively, the method for detecting a charging raw material flow section in the blast furnace according to 3 above. 5. After determining the flow part, the ratio of the flow part to the image area for one screen captured by the television camera is calculated, and the area corresponding to the image area for one screen and the ratio are calculated. The method for detecting the charged raw material flowing part in the blast furnace according to claim 1, 2, 3 or 4, wherein the area of the flowing part is calculated by multiplication. 6. The distance between the television camera and the surface of the charging raw material in the image pickup direction of the television camera is measured, and the calculated area of the fluidized portion is multiplied by the ratio of the measured distance to a preset reference distance. The method for detecting a charging material flow part in a blast furnace according to claim 5, wherein the area of the flow part is corrected. 7. The zoom lens of the TV camera is operated such that the distance between the TV camera and the surface of the charging raw material in the image pickup direction of the TV camera is measured, and the magnification is increased in proportion to the measured distance. The method for detecting a raw material flowing part in a blast furnace according to claim 5.