JP2016050209A - Slag monitor device and method for coal gasifier - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve slag evaluation accuracy.SOLUTION: A slag monitor device comprises: an imaging part 101 for imaging the vicinity of a slag hole 3 from which molten slag exits, and the vicinity of a water face 5H of cooling water where the slag exited from the slag hole 3 falls, and acquires respective images; a ridge line detection part 102 for using a water jet algorithm to the respective images, and detecting a ridge line estimated as a slag falling line; an illuminance information acquisition part 103 for acquiring illumination information from plural detection lines provided in a direction crossing the slag falling line, in a predetermined region of the respective images; and a falling line determination part 104 for determining that the ridge line is the slag falling line, and outputting a number of the slug falling lines in the predetermined region, when illumination difference between the illumination information on a cross point between the ridge line and respective detection lines, and illumination information in the vicinity of the cross point along the respective detection lines is equal to or more than a first threshold.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a slag monitoring apparatus and method for a coal gasifier.

For example, an integrated coal gasification combined cycle (IGCC) system that generates power by driving a gas turbine with coal gas obtained by gasification of coal is used by a coal gasification furnace to gasify coal. Is done.
The coal gasifier has a two-stage structure consisting of a lower combustor section (combustion chamber) and an upper reductor section (gasification chamber). In the combustor section, the pulverized coal and char are combusted while burning. The ash component contained in the pulverized coal and char is melted by high-temperature heat and adheres to the water-cooled wall of the furnace by centrifugal force, forming molten slag. It flows down to the bottom of the furnace. The molten slag falls from the slag hole at the bottom of the combustor portion into the slag hopper in which water is stored and is rapidly cooled to be discharged into the outside as a glassy slag.

Avoiding slag hole clogging and slag flow destabilization due to slag solidification is important in the operation of coal gasifiers, and monitoring the slag discharge status in order to operate coal gasifiers normally. Has been done.
The following Patent Document 1 describes a technique for diagnosing a slag dripping state and a furnace state by capturing slag to be dripped with an image, measuring slag dripping frequency and slag drop accumulation by image processing, and calculating a slag flow amount. Has been.
In the following Patent Document 2, it is described that determination is performed by a combination of a visual sensor and an acoustic sensor, and slag dynamic monitoring evaluation of the entire furnace is performed.

JP-A-1-24894 JP 7-34075 A

By the way, when monitoring slag with images, etc., long-term operation of the demonstrator will cause changes in the combustion status in the furnace (overall bright, dark, etc.), aging of the equipment, dirt on the monitoring window, etc. There is a problem that the level of the image is lowered or the image is unclear, the slag recognition rate is deteriorated, and the detection accuracy is lowered.
However, in the conventional method, no countermeasure is taken when the image level is lowered or the image is blurred, and the problem that the slag recognition rate is deteriorated cannot be solved.

  This invention is made | formed in view of such a situation, Comprising: It aims at providing the slag monitoring apparatus and method of a coal gasification furnace which improve slag evaluation precision.

In order to solve the above problems, the present invention employs the following means.
The present invention images the vicinity of the slag hole where the molten slag flows out, and the vicinity of the water surface of the cooling water where the slag that flows out of the slag hole falls, and obtains each captured image, On the other hand, using a watershed algorithm, a ridge line detecting means for detecting a ridge line estimated as a drop line of the slag, and a predetermined region of each captured image are provided in a direction crossing the estimated drop line of the slag. Luminance information acquisition means for acquiring luminance information from a plurality of detection lines, luminance information of intersections between the ridge lines and the detection lines, and luminance information of the luminance information in the vicinity of the intersections along the detection lines When the difference is greater than or equal to the first threshold value, the ridgeline is provided with a falling streak determining means for determining that the ridgeline is a falling streak of the slag. To provide a grayed monitoring device.

  According to the configuration of the present invention, the watershed algorithm is used for each captured image in which the vicinity of the slag hole where the molten slag flows out and the vicinity of the cooling water where the slag flowing out of the slag hole falls are imaged. The ridge line estimated as the slag fall streak is detected, and brightness information on the detection line is acquired from a plurality of detection lines provided in a direction intersecting the estimated slag fall streak in a predetermined region of the captured image. The If the luminance difference between the luminance information at the intersection of the ridge line detected from the watershed algorithm and the detection line and the luminance information in the vicinity of the intersection is equal to or greater than the first threshold value, the ridge line is a slag falling line. It is determined.

  Since the watershed algorithm divides the region according to the change in luminance, the region is divided even for minute changes such as noise and shadows, and the ridgeline may be detected in places other than the slag falling streaks. In addition to edge detection by the watershed algorithm, determination is made by combining luminance information acquired for each detection line in a predetermined area, so that slag can be determined more accurately.

  The falling streak determining means of the slag monitoring device for the coal gasification furnace is configured to obtain the luminance information in the vicinity of the intersection from a first coordinate that is separated from the intersection by a first predetermined interval along the detection line. It is good also as setting it as the average value of the brightness | luminance of the area from the coordinate to the 2nd coordinate which left | separated 2nd predetermined spacing along the said detection line.

  Thus, the luminance information in the vicinity of the intersection can be obtained as a more accurate luminance difference from the intersection by using the average luminance value of the second predetermined interval.

  The present invention captures an image of the vicinity of a slag hole through which molten slag flows out and the vicinity of the water surface of the cooling water from which the slag that has flowed out of the slag hole falls, In a predetermined area, luminance information acquisition means for acquiring luminance information from a plurality of detection lines provided in a direction intersecting with a direction in which the falling stripe of the slag is estimated to fall, and at least one end of each of the detection lines from the other end Peak detection means for comparing the luminance information of adjacent pixels on the detection line in the direction of the side and detecting the pixel at which the value of the luminance information starts to fall as a peak position; and a predetermined number of pixels from the peak position Designated dot detection means for detecting designated dots that are the pixels in which the value of the luminance information has decreased continuously; the luminance information at the peak position; Provided is a slag monitoring device for a coal gasification furnace, comprising fall streak judging means for judging the peak position as a fall streak of the slag when a brightness difference between the designated dot and the brightness information is equal to or greater than a second threshold value. To do.

According to the configuration of the present invention, estimation is performed in a predetermined region of a captured image obtained by imaging the vicinity of the slag hole where the molten slag flows out and the vicinity of the water surface of the cooling water where the slag flowing out of the slag hole falls. Luminance information is acquired from a plurality of detection lines provided in a direction intersecting with the falling stripe of the slag. The luminance information of adjacent pixels on the detection line is compared from at least one end side to the other end side of the detection line, and a pixel whose luminance information value starts to decrease is detected as a peak position. When a designated dot is detected that is a pixel whose luminance information value has fallen continuously for a predetermined number of pixels from the detected peak position, the luminance difference between the luminance information of the designated dot and the luminance information of the peak position is compared. Is equal to or greater than the second threshold value, the peak position is determined as a slag drop streak.
In this way, when there is a designated dot with a predetermined number of pixels continuously decreasing from the peak position and the luminance difference between the peak position and the designated dot is equal to or greater than the second threshold, the peak position is a slag streak. Therefore, it is possible to more accurately determine the slag falling streaks.

  The falling streak determining means of the slag monitoring device of the coal gasification furnace comprises counting means for counting the number of slag falling streaks determined as the slag falling streaks for each detection line, and in the predetermined region The maximum value among the number of falling stripes of the slag counted for each detection line may be output as the number of falling stripes of the slag in the predetermined area.

  By determining the maximum number as the number of slag fallen streaks, the number of slags is not underestimated, so the safety side can be dealt with.

  The present invention captures an image of the vicinity of a slag hole through which molten slag flows out and the vicinity of the water surface of the cooling water from which the slag that has flowed out of the slag hole falls, A binarization level determining unit that determines a binarization level set according to an average value of luminance, and binarizing each pixel of a predetermined monitoring target of the captured image by the determined binarization level, There is provided a slag monitoring device for a coal gasification furnace comprising an area calculating means for calculating an area of the slag region, wherein the pixel that is equal to or higher than the binarization level is used as a slag region.

  According to the configuration of the present invention, the captured images acquired by imaging the vicinity of the slag hole where the molten slag flows out and the vicinity of the water surface of the cooling water where the slag flowing out of the slag hole falls correspond to the average value of the luminance. The binarization level is set, and the captured image is binarized based on the binarization level of each pixel. In a predetermined monitoring target of the binarized captured image, the area is calculated using pixels that are equal to or higher than the binarization level as slag regions.

  In this way, since the binarization level is set according to the average value of the brightness of each captured image, the captured image capturing the vicinity of the slag hole and the captured image capturing the vicinity of the water surface where the slag falls, respectively. A binarization level can be set. In addition, since the binarization level changes according to the situation (background luminance) where each captured image is captured, it corresponds to the captured image compared to the case where a fixed binarization level value is used. Thus, the slag muscle can be determined with higher accuracy.

  The present invention is set for each of the slags in an imaging unit that captures an image of the vicinity of the cooling water surface where the slag that has flowed out of the slag hole that flows out of the molten slag falls and acquires the captured image. Binarization level determining means for determining a binarization level based on the maximum value and the average value of the luminance information in the predetermined area, and binarizing each pixel in the predetermined area based on the determined binarization level The present invention provides a slag monitoring apparatus for a coal gasification furnace, comprising: an area calculating means for calculating an area of the slag region, wherein the pixel that is equal to or higher than the binarization level is used as a slag region.

  According to the configuration of the present invention, in a captured image obtained by capturing an image of the vicinity of the surface of the cooling water from which the slag that has flowed out from the slag hole from which the molten slag flows out, a predetermined region set for each estimated slag is obtained. The binarization level is determined based on the maximum value and the average value of the luminance information, each pixel in the predetermined area is binarized based on the determined binarization level, and pixels that are equal to or higher than the binarization level are defined as slag areas. The area of the slag region is calculated.

  According to the present invention, when a plurality of slag lines are included in one captured image, different predetermined areas are set for the respective slag lines, and a binarization level is obtained for each predetermined area. Binarization is performed for each predetermined area. Thus, for example, one of the two slag streaks can be clearly identified (high brightness) and the other is blurred (low brightness) depending on the brightness of each. Since the binarization level is set, the area of the slug can be calculated correctly.

  When the monitor window is soiled due to deterioration over time due to operating for a long time and the image is blurred, or when there are multiple slags and the distance from the imaging means to each slag is different When focusing on one slag, there was a problem that the recognition rate of the other slag decreased. According to the present invention, since a binarization level is obtained according to a predetermined region including slag streaks, an appropriate binarization level is set for each slag, and slag can be determined more accurately.

  The binarization level determination means of the coal gasification furnace slag monitoring device multiplies a difference between a maximum value of the luminance information in the predetermined area and an average value of the luminance information in the predetermined area by a coefficient. It is preferable to calculate the binarization level by adding the multiplication result and the average value of the luminance information.

  Thereby, the binarization level according to the brightness | luminance condition in a predetermined area | region can be set easily, and the area of slug can be calculated | required correctly.

  The present invention captures the vicinity of the slag hole where the molten slag flows out, and the vicinity of the water surface of the cooling water where the slag that flows out of the slag hole falls, and acquires each captured image, and each of the captured images And a second process of detecting a ridge line estimated as the slag fall streak using a watershed algorithm, and a predetermined region of each captured image provided in a direction intersecting with the estimated slag drop streak The luminance information of the third process of acquiring the luminance information from the plurality of detection lines, the luminance information of the intersection of the ridge line and each of the detection lines, and the luminance information in the vicinity of the intersection along each of the detection lines A slag monitoring method for a coal gasifier having a fourth step of determining that the ridge line is a drop of the slag when the difference is equal to or greater than a first threshold value. .

  The present invention captures the vicinity of the slag hole where the molten slag flows out, and the vicinity of the water surface of the cooling water where the slag that flows out of the slag hole falls, and acquires each captured image, and each of the captured images A second process of acquiring luminance information from a plurality of detection lines provided in a direction intersecting with a direction in which the falling stripe of the slag is estimated to fall in the predetermined region, and at least one end of each of the detection lines from the other end A third step of comparing the luminance information of adjacent pixels on the detection line in the direction of the side and detecting the pixel at which the value of the luminance information starts to decrease as a peak position, and a predetermined number of pixels from the peak position A fourth step of detecting a designated dot that is the pixel in which the value of the luminance information has decreased continuously; the luminance information of the peak position; and the luminance information of the designated dot. When the luminance difference is the second threshold or more and provides a slag monitoring method of the coal gasification furnace and a fifth step determines the peak position as dropping muscle of the slag.

  The present invention captures the vicinity of the slag hole where the molten slag flows out, and the vicinity of the water surface of the cooling water where the slag that flows out of the slag hole falls, and acquires each captured image, and each of the captured images A second step of determining a binarization level set in accordance with an average value of luminance of the image, and binarizing each pixel to be monitored in the captured image according to the determined binarization level; There is provided a slag monitoring method for a coal gasification furnace including a third process of calculating the area of the slag region by setting the pixel that is equal to or higher than a value level as a slag region.

  The present invention is set for each slag in the first process of capturing an image of the vicinity of the surface of the cooling water from which the slag that has flowed out of the slag hole from which the molten slag has flowed out, and acquiring the captured image. A second step of determining a binarization level based on a maximum value and an average value of the luminance information of the predetermined region, and binarizing each pixel of the predetermined region according to the determined binarization level, There is provided a slag monitoring method for a coal gasification furnace, comprising: a third step of calculating an area of the slag region by setting the pixel that is equal to or higher than a binarization level as a slag region.

  The present invention has an effect of improving the slag evaluation accuracy.

It is a whole block diagram of the slag monitoring apparatus which concerns on the 1st Embodiment of this invention. It is an example of the image obtained in the process of calculating | requiring a ridgeline using the watershed algorithm which concerns on the 1st Embodiment of this invention. It is a figure which shows an example of the image obtained by the slag hall camera and the water surface camera. It is a figure which shows an example of the detection line of the region of interest of a slag hole image and a water surface image. It is an example of the figure which showed the luminance information with respect to the coordinate obtained from one detection line. It is an example of the figure which displayed as a list the number of slag muscles detected from each detection line. It is an operation | movement flow of the slag monitoring apparatus which concerns on the 1st Embodiment of this invention. It is a whole block diagram of the slag monitoring apparatus which concerns on the 2nd Embodiment of this invention. It is a figure for demonstrating the peak brightness | luminance tracking of a detection line with the slag monitoring apparatus which concerns on the 2nd Embodiment of this invention. It is an operation | movement flow of the slag monitoring apparatus which concerns on the 2nd Embodiment of this invention. (A) The result of measuring the number of slag bars in the slag hole by the slag monitoring apparatus according to the first embodiment and the result of measuring the number of slag bars by the slag monitoring apparatus according to the second embodiment, (b) FIG. It is an example of the figure of the comparison of the result of having output the number of slag muscles among (a) as the number of slag muscles, and (c) measuring the number of slag muscles by the conventional method. (A) The result of measuring the number of slag muscles on the water surface by the slag monitoring device according to the first embodiment and the result of measuring the number of slag muscles by the slag monitoring device according to the second embodiment, (b) FIG. It is an example of the figure of the comparison of the result of having output the one with many numbers among a) as a slag muscle number, and (c) measuring the number of slag muscles by the conventional method. It is a whole block diagram of the slag monitoring apparatus which concerns on the 3rd Embodiment of this invention. It is an example of the figure of the Fx setting which the binarization level determination part of the slag monitoring apparatus which concerns on the 3rd Embodiment of this invention has. It is an operation | movement flow of the slag monitoring apparatus which concerns on the 3rd Embodiment of this invention. (A) The comparison figure of the result of having calculated the area rate of the slag streak of a slag hole by the slag monitoring device concerning the 3rd embodiment of the present invention, and the result of (b) the area rate calculated by the conventional method It is an example. (A) By the slag monitoring apparatus which concerns on the 3rd Embodiment of this invention, the comparison figure of the result of having calculated the area ratio of the slag streaks of the water surface, and (b) the result of the area ratio calculated by the conventional method It is an example. It is a whole block diagram of the slag monitoring apparatus which concerns on the 4th Embodiment of this invention. It is an operation | movement flow of the slag monitoring apparatus which concerns on the 4th Embodiment of this invention. (A) the result of the area ratio calculated by the conventional method, (b) the result of calculating the area ratio of the slag streaks on the water surface by the slag monitoring device according to the third embodiment of the present invention, and (c) this It is an example of the comparison figure with the result of having calculated the area ratio of the slag streak of the water surface by the slag monitoring apparatus which concerns on the 4th Embodiment of invention.

  Hereinafter, embodiments of a slag monitoring apparatus and method for a coal gasifier according to the present invention will be described with reference to the drawings. In the description of the embodiment, the slag monitoring device for a coal gasification furnace of the present invention will be described assuming that it is applied to a combined coal gasification combined power plant. The monitoring device is not limited to this.

[First Embodiment]
FIG. 1 is an overall configuration diagram of a slag monitoring device 100 for a coal gasification furnace according to the present embodiment.
The slag monitoring device for coal gasification furnace (hereinafter referred to as “slag monitoring device”) 100 is a device that monitors the flow state of slag generated in the process of gasifying coal in the coal gasification furnace 1. The coal gasification furnace 1 is supplied with coal and a gasifying agent (air, oxygen-enriched air, O 2, etc.) and combustor 1C for burning the coal, a reductor 1R for charging and gasifying the coal, and a combustor 1C. And a slag discharge cylinder 4 for collecting the molten slag discharged from the slag.
The combustor 1 </ b> C is provided with a slag tap 2 and a slag hole 3.

The combustor 1 </ b> C burns pulverized coal and char introduced therein to generate combustible gas from the coal. Furthermore, the molten slag in which the ash content after combustion is melted is generated in the combustor 1C. Since a swirl flow is formed inside the combustor 1C, the molten slag adheres to the inner peripheral surface of the combustor 1C main body and flows down toward the lower slag tap 2.
The slag tap 2 accumulates the ash after combustion as molten slag. A slag hole 3 that guides the molten slag to cooling water of a slag hopper (described later) 5 is disposed in the approximate center of the slag tap 2.

The slag tap 2 is a conical member disposed below the combustor 1C, and has a surface inclined downward toward the center of the combustor 1C main body. By comprising in this way, the molten slag which flowed down from combustor 1C is guide | induced to the slag hole 3 of the center of the combustor 1C main body.
The slag hole 3 guides molten slag from the slag tap 2 to the cooling water of the slag hopper 5, and at the edge of the slag hole 3 is a weir and a groove (outflow guide groove) for guiding the outflow of discharged slag. A plurality (for example, two, positions facing each other at an interval of 180 degrees) are formed. The cross-sectional area of the outflow guide groove is designed so that two slag flows steadily flow down.

The molten slag continuously falls into the cooling water or falls intermittently depending on the operation state of the coal gasification furnace 1, that is, the internal conditions of the combustor 1C.
As shown in FIG. 1, the slag discharge cylinder 4 is provided below the coal gasification furnace 1 (on the vertical direction side). Below the slag discharge cylinder 4, a slag hopper 5 is provided, and cooling water for cooling the molten slag discharged from the slag hole 3 is stored.
A slag reservoir 7 is provided at the lower portion of the slag discharge cylinder 4, and slag (solidified slag) 8 </ b> R that has fallen into the slag hopper 5 and solidified is accumulated.
The molten slag is discharged from the slag hole 3 and accumulated in the cooling water of the slag hopper 5, and when the amount of accumulation increases, the molten slag is accumulated near the water surface 5H and beyond the water surface 5H.

The slag monitoring device 100 includes, for example, a CPU (Central Processing Unit) (not shown), a RAM (Random Access Memory), an auxiliary storage device such as a computer-readable recording medium, an input / output device such as a keyboard, a display, and a speaker, and an external device. A communication device that exchanges information by communicating with a device is provided. A series of processing steps for realizing various functions to be described later are recorded in a recording medium or the like in the form of a program, and the CPU reads the program into a RAM or the like to execute information processing / arithmetic processing. Thus, various functions described later are realized.
Specifically, the slag monitoring device 100 includes an imaging unit (imaging unit) 101, a ridge line detection unit (ridge line detection unit) 102, a luminance information acquisition unit (luminance information acquisition unit) 103, and a falling line determination unit (falling). Muscle determination means) 104.

  The imaging unit 101 captures the vicinity of the slag hole where the molten slag flows out and the vicinity of the cooling water where the slag flowing out of the slag hole 3 falls, and acquires captured images. The imaging unit 101 includes, for example, a camera. Specifically, the imaging unit 101 includes a slag hole camera 11 for acquiring a slag hole image obtained by imaging the vicinity of the slag hole from which the molten slag flows, and the water surface of the cooling water from which the slag flowing from the slag hole 3 falls. And a water surface camera 12 for acquiring a water surface image of the vicinity.

  The slag hall camera 11 is provided outside the side wall of the slag discharge tube 4. The slag hole camera 11 images the slag hole 3 and the vicinity (periphery) of the slag hole 3 through the slag hole monitoring window provided on the side wall of the slag discharge cylinder 4 to generate a slag hole image. The slag hole image includes luminance distribution data, and the luminance distribution data of the slag hole 3 includes data indicating the luminance of each pixel included in the slag hole image.

  The water surface camera 12 is provided outside the side wall of the slag discharge cylinder 4. The water surface camera 12 images the water surface 5H of the cooling water through a water surface monitoring window provided on the side wall of the slag discharge cylinder 4, and generates a water surface image. The water surface image includes luminance distribution data, and the luminance distribution data of the water surface 5H is composed of the luminance of each pixel included in the water surface image.

  The ridge line detection unit 102 detects a ridge line estimated as a drop line of the slag by using a watershed algorithm (a watershed method, a watershed method) for each captured image (slag hole image and water surface image). The watershed algorithm is one of the region segmentation methods. The brightness gradient of the image is regarded as a mountain, the brightness value of the image (pixel) is regarded as the altitude of the mountain, the minimum value of the brightness value, and the maximum value as the valley bottom and the ridge, respectively. This is an algorithm that divides an area where water flows from the position of the ridge into one area, and is a well-known technique.

FIG. 2 is an example of an image including a ridge line X obtained from a slag hole image using the watershed algorithm.
Specifically, when the ridge line detection unit 102 acquires images (for example, color moving images) acquired from the slag hall camera 11 and the water surface camera 12, the ridge line detection unit 102 continues from the acquired moving images a predetermined number of times (for example, 10 times). Capture to create an average image. The ridge line detection unit 102 performs monochrome conversion on a predetermined number (for example, 10) of the captured average images to generate one monochrome image, and smoothes the monochrome image with a smoothing function or the like.

Since the watershed algorithm divides a region by a change in luminance, the region is divided even by a minute change such as noise or shadow. As a result, since excessive division occurs, smoothing processing is performed on the input image in advance using a smoothing function or the like so as to reduce the influence of noise and shadow in order to prevent excessive division.
The ridge line detection unit 102 performs water shed processing on the slag hole image and the water surface image. In the watershed process, the setting parameter is an Mv (Min variation) value, and the default is “1”.

In the watershed algorithm, when the Mv value is increased, the ridge line X of the region of interest (described later) set in the slag hole image and the water surface image gradually decreases, but when the Mv value is incremented by 1, Then, a value when the number of ridge lines of the region of interest does not change twice is determined as a ridge line, and the process is terminated.
By applying such a watershed algorithm, a ridgeline X (a line shown as a crack) is detected as shown in FIG.

  Here, the region of interest is an image region to be processed set in the slag hole image and the water surface image.

  FIG. 3 shows a slag hole image 9H acquired by the slag hole camera 11, a water surface image 9W acquired by the water surface camera 12, and a region of interest ROI (Region Of Interest). The slag hole image 9H includes the slag hole 3 and its vicinity (periphery), and the water surface image 9W includes the vicinity of the water surface 5H where slag lines (hereinafter referred to as “slag lines”) flow down.

ROI (1), ROI (2), ROI (s), ROI (SL) are set as the regions of interest in the slag hole image 9H, and ROI (3), ROI (4) as the regions of interest in the water surface image 9W. ), ROI (5), ROI (w), and ROI (WL).
When monitoring the flow state of the slag, the slag muscles 8A and 8B flowing down from the slag hole 3 are detected and focused.

ROI (s) is defined as a region including the slag hole 3 and the slag downflow bars 8A and 8B. Specifically, ROI (s) is an area including ROI (1), ROI (2), and ROI (SL).
ROI (1) is a rectangular region including the slag hole 3 and the slag muscles 8A and 8B in the region set by ROI (s). By calculating the high-luminance area and the low-luminance area of ROI (1) as evaluation parameters, it is possible to evaluate slag flow determination and slag hole monitoring camera window water cleaning guidance. The high-luminance area of ROI (1) is the area of a region where the luminance is higher than a predetermined value in ROI (1). The low luminance area of ROI (1) is the area of a region where the luminance is lower than a predetermined value in ROI (1).

ROI (2) is a rectangular area including the monitoring area of the opening of the slag hole 3. By calculating the high brightness area of the ROI (2) as an evaluation parameter, it is possible to evaluate slag hole blockage (slag drop caution) and slag melting burner ignition guidance. The opening high luminance area of ROI (2) is the area of a region where the luminance is higher than a predetermined value in ROI (2).
In addition, since the slag hole camera 11 which images the slag hole 3 images the slag hole 3 from obliquely below, the slag hole 3 is shown as an ellipse in the slag hole image 9H.

  ROI (SL) is a predetermined area for detecting slag streaks flowing down from the slag hole 3. The evaluation parameter used when monitoring the flow state of the slag is the number of slag lines flowing down from the slag hole 3.

ROI (w) is a region including slag streaks falling on the water surface. Specifically, ROI (w) is an area including ROI (3), ROI (4), ROI (5), and ROI (WL).
ROI (3) is a rectangular region including the slag falling situation (for example, two lines) on the water surface, and the evaluation parameters used when monitoring the slag flow situation are the time-domain luminance fluctuation amount and the low luminance part. It is an area. The time domain luminance fluctuation amount is a luminance fluctuation amount for each processing cycle in ROI (3). Further, the low luminance area of ROI (3) is the area of a region where the luminance is lower than a predetermined value in ROI (3).
ROI (4) is an area including the entry area of one of the two slug muscles falling on the water surface 5H (left side of the paper; 8A) into the water surface 5H, and ROI (5) falls on the water surface 5H. This is an area including a rush area of one of the two slug bars (right side of the paper; 8B) into the water surface 5H.

By calculating the high-luminance area of ROI (4) and ROI (5) as an evaluation parameter, water surface slag accumulation attention can be determined. The high luminance areas of ROI (4) and ROI (5) are areas where the luminance is higher than a predetermined value in ROI (4) and ROI (5) indicating areas where slag muscles 8A and 8B fall to the water surface 5H. It is an area.
ROI (WL) is a predetermined area for detecting slag streaks falling on the water surface.
Note that the number of slag bars basically depends on the number of outflow guide grooves formed at the edge of the slag hole 3.

  The luminance information acquisition unit 103 acquires luminance information from a plurality of detection lines provided in a direction intersecting with the estimated drop of the slag in a predetermined region of each captured image (slag hole image and water surface image). For example, as illustrated in FIG. 4, the luminance information acquisition unit 103 sets the ROI (SL) as a predetermined area for the slag hole image 9H, sets the ROI (WL) as a predetermined area for the water surface image 9W, When slug lines are falling in the vertical direction (vertical direction on the paper surface), a plurality of (for example, ten) detection lines P are provided in the horizontal direction, and luminance information is obtained from each of the plurality of detection lines P provided. To get.

The slag streak is detected when the average luminance in the ROI is the following equation (1).
B1 (= 20) <detection range <B2 (= 200) (1)
Here, the values of B1 and B2 are set with respect to the luminance level (value of 0 to 255 at 8 bit resolution).
B1 and B2 are parameters and are set arbitrarily. The number in parentheses is the default value. The B1 setting is intended to prevent erroneous detection by assuming that the camera window is saturated when the monitor window is soiled.

  When the luminance difference between the luminance information at the intersection of the ridge line X and each detection line P and the luminance information near the intersection along each detection line P is equal to or greater than the first threshold, It is determined that the ridge line X is a drop line of the slag. Further, the falling line determination unit 104 obtains luminance information in the vicinity of the intersection from the first coordinate e1 that is separated from the intersection by the detection line P along the detection line P by the first predetermined interval, and from the first coordinate e1 along the detection line P. 2 It is set as the average value of the brightness | luminance of the area to the 2nd coordinate e2 separated by predetermined spacing.

The fall line determination unit 104 will be described with reference to FIGS.
The video signals acquired from the slag hall camera 11 and the water surface camera 12 have a value of 0 to 255 with A / D conversion 8-bit resolution, and all the detection lines P in ROI (SL) and ROI (WL). Thus, luminance information (luminance distribution) can be obtained.
FIG. 5 shows an example of luminance information obtained from one detection line P in the region of interest ROI (SL), with the horizontal axis representing the coordinates of the detection line P and the vertical axis representing the luminance level. (See solid line 5). The intersection of the coordinate of the ridge line X detected by the ridge line detection unit 102 by the watershed algorithm and the luminance information of the detection line P is marked with a point (·). In FIG. 5, it overlaps with two vertices of the luminance information graph.

Here, the first predetermined interval is L, and the second predetermined interval is ΔL. From the intersection of the coordinates of the ridge line X and the luminance information of the detection line P, the first coordinate e1 that is a first predetermined distance L along the detection line P is opposite to the intersection along the detection line P from the first coordinate e1. The average value of the luminance of the section up to the second coordinate e2 that is separated by the second predetermined interval ΔL is obtained.
A luminance difference between the luminance information of the intersection and the above-obtained average value is calculated, and if the luminance difference is equal to or greater than the first threshold value, it is determined that this intersection (position marked with a dot) is a slag streak. If there are a plurality of intersections, a luminance difference is calculated for each intersection, and it is determined whether or not all the intersections are slag lines.
It should be noted that the first predetermined interval L and the second predetermined interval ΔL are values set for each plant by the supervisor while observing the situation such as the size of the plant and the size of the slag muscle on site.

Further, the falling stripe determination unit 104 includes a counting unit (counting unit) 110 that counts the number of falling slag stripes determined as slag falling stripes for each detection line P. The maximum value among the number of slag falling streaks counted every time is output as the number of slag falling streaks in a predetermined area.
FIG. 6 shows an example of the number of slag lines determined in each of the ten detection lines P in the region of interest ROI (SL). The maximum number of slag lines shown here is determined to be the number of slag lines in the region of interest ROI. In FIG. 6, since the maximum number is “2”, the detected region of interest ROI (SL) outputs “2” slag streaks.
The slag monitoring device 100 manages the number of slag lines detected in this way as data, and uses it as one parameter for in-furnace slag evaluation.

The slag monitoring device 100 further includes a spectrometer 10 and a falling sound sensor 13.
The spectrometer 10 is provided outside the side wall of the slag discharge cylinder 4. The spectrometer 10 measures the temperature of the central portion of the slag hole 3 through the slag hole observation window, with the central portion (small region) of the slag hole 3 as a visual field.
The falling sound sensor 13 observes the sound of the slag falling on the cooling water surface 5H.
The falling sound sensor 13 is provided in the cooling water of the slag hopper 5. As the falling sound sensor 13, for example, a hydrophone can be used. The falling sound sensor 13 converts the sound input to itself into an electrical signal and outputs it.

  The spectrometer 10 is connected to a dedicated IF board (not shown), and temperature data indicating the measured center temperature of the slag hole 3 is generated by the dedicated IF.

The electric signal output from the falling sound sensor 13 is amplified by an amplifier (not shown), and a signal in a predetermined band including a band component of the falling sound generated when the slag falls into the cooling water of the slag hopper 5 is passed through. The signal is converted to a digital signal.
The slag monitoring device 100 determines from the falling sound detected by the falling sound sensor 13 whether the slag is continuously falling, intermittently falling, or not falling from the slag hole 3.

  An underwater slag observation unit 14 for observing solidified slag (solidified slag) 8R existing in the cooling water of the slag reservoir 7 is provided around the slag reservoir 7. The underwater slag observation unit 14 is disposed below the falling sound sensor 13. In the present embodiment, the underwater slag observation unit 14 receives a plurality of (four in the present embodiment) transmission sensors 14T that transmit detection waves and a plurality of (this embodiment) that receives the detection waves transmitted by the transmission sensors 14T. It is composed of four receiving sensors 14R. The underwater slag observation unit 14 observes the solidified slag 8R in the slag reservoir 7 by detecting the attenuation degree of the detection wave transmitted from the transmission sensor 14T by the reception sensor 14R. When there is a reception sensor 14R in which the detection wave transmitted from the transmission sensor 14T is greatly attenuated by utilizing the attenuation of the detection wave due to the presence of the solidified slag 8R, the reception sensor 14R, It can be determined that the solidified slag 8R exists between the transmission sensor 14T that has transmitted the detection wave.

Next, the operation of the slag monitoring apparatus 100 according to the present embodiment will be described with reference to FIGS. FIG. 7 is an operation flow of the slag monitoring apparatus 100 according to the present embodiment.
The slag hall camera 11 images the vicinity of the slag hole with a color moving image, and the water surface camera 12 images the vicinity of the water surface 5H with a color moving image. When the analysis start command is acquired, ten image captures are performed from the moving images acquired from the slag hall camera 11 and the water surface camera 12 (step SA1 in FIG. 7). One black and white image converted from black and white from the 10 captured images is generated (step SA2 in FIG. 7).

  The generated black and white image is smoothed by the smoothing function (step SA3 in FIG. 7). Edge lines are detected by the watershed process (step SA4 in FIG. 7). Luminance information of intersections determined based on the edge line X detected by the watershed process and the luminance information detected from each detection line P of the region of interest, and luminance information in the vicinity of the intersections separated from the intersections by a predetermined distance The slag muscle is determined based on (step SA5 in FIG. 7). When the slag muscles of the region of interest are determined, the number of slag muscles is evaluated (step SA6 in FIG. 7).

  As described above, according to the slag monitoring device 100 and the method of the coal gasification furnace 1 according to the present embodiment, the vicinity of the slag hole 3 through which the molten slag flows out, and the slag that flows out from the slag hole 3 fall. For each captured image in which the vicinity of the water surface 5H of the cooling water is imaged, a ridge line X estimated as a slag drop streak is detected using the watershed algorithm, and the estimated slag in a predetermined region of the captured image is detected. Luminance information on the detection line P is acquired from a plurality of detection lines P provided in a direction intersecting with the falling stripe. When the luminance difference between the luminance information at the intersection of the ridge line X detected from the watershed algorithm and the detection line P and the luminance information in the vicinity of the intersection is equal to or greater than the first threshold, the ridge line X falls off the slag. It is determined that it is a streak, and the number of falling streaks is output.

  In the watershed algorithm, the region is divided by a change in luminance, so the region is divided even for minute changes such as noise and shadow, and it is estimated that the ridge line X is detected at a place other than the falling line of the slag, In the present invention, in addition to the detection of the ridge line X by the watershed algorithm, the luminance information acquired for each detection line in the predetermined region is determined in combination, so that the slag can be determined more accurately, and the number of slag falling stripes can be more accurately Is output.

As a result, even if the level of the acquired image is reduced or blurring occurs due to changes in the combustion status in the furnace (brighter or darker overall) or changes in the equipment status (such as aging or dirt on the monitoring window) The slag muscle can be reliably identified.
Regarding the diagnosis of slag dynamics in a gasification furnace, a system that comprehensively monitors and evaluates the slag dynamics in the gasification furnace in real time by continuously monitoring the slag status in the slag discharge tower and in the slag discharge tower. Internal evaluation accuracy is improved.

[Second Embodiment]
Hereinafter, a second embodiment of the present invention will be described with reference to FIGS. The slag monitoring apparatus according to the second embodiment is different from the first embodiment in that slag lines are determined by following the value of luminance information detected from the detection line of the region of interest. Hereinafter, description of points common to the first embodiment will be omitted, and different points will be mainly described.

FIG. 8 shows an overall configuration diagram of the slag monitoring apparatus 100a according to the present embodiment. The slag monitoring device 100a includes an imaging unit (imaging unit) 101, a luminance information acquisition unit (luminance information acquisition unit) 103, a peak detection unit (peak detection unit) 105, and a designated dot detection unit (designated dot detection unit) 106. And a falling line determination unit (falling line determination means) 104a.
The imaging unit 101 acquires a slag hole image obtained by imaging the vicinity of the slag hole from which the molten slag flows out, and a water surface image obtained by imaging the vicinity of the cooling water from which the slag flowing out from the slag hole falls.

  The luminance information acquisition unit 103 acquires luminance information from a plurality of detection lines provided in a direction intersecting with a direction in which the falling stripe of the slag is estimated to fall in a predetermined region of the slag hole image and the water surface image. In the present embodiment, the predetermined region of the slag hole image is set to the region of interest ROI (SL), and the predetermined region of the water surface image is set to the region of interest ROI (WL).

The peak detection unit 105 compares the luminance information of adjacent pixels on the detection line in the direction from at least one end side to the other end side of each detection line, and detects a pixel where the value of the luminance information starts to decrease as a peak position. .
The designated dot detection unit 106 detects a designated dot that is a pixel in which the value of the luminance information is lowered continuously for a predetermined number of pixels from the peak position.

  The falling stripe determination unit 104a determines the peak position as the falling stripe of the slag when the luminance difference between the luminance information of the peak position and the luminance information of the designated dot is equal to or greater than the second threshold value. Further, the falling stripe determination unit 104a includes a counting unit (counting unit) 110 that counts the number of falling slag stripes determined as slag falling stripes for each detection line. The maximum value among the number of slag falling streaks counted in step S3 is output as the number of slag falling streaks in a predetermined area.

Next, the operation of the slag monitoring apparatus 100a according to this embodiment will be described with reference to FIGS. Here, a case where slag lines are determined from the region of interest ROI (SL) of the slag hole image 9H will be described as an example. In FIG. 9, the horizontal axis represents the coordinates of the detection line ROI (SL) of the slag hole image 9H, and the vertical axis represents one detection line among the ten detection lines in the ROI (SL). The luminance information obtained with the coordinates is taken as the luminance level, and an example showing the relationship between the coordinates and the luminance level is shown.
The slag hall camera 11 images the vicinity of the slag hole with a color moving image, and the water surface camera 12 images the vicinity of the water surface 5H with a color moving image. When the analysis start command is acquired, ten image captures are performed from the moving images acquired from the slag hall camera 11 and the water surface camera 12 (step SB1 in FIG. 10). One black and white image converted from black and white is generated from the 10 captured images (step SB2 in FIG. 10).

  The generated black and white image is smoothed by the smoothing function (step SB3 in FIG. 10). In the smoothed image, a region of interest ROI (SL) and a region of interest ROI (WL) are set, and luminance information is acquired from each of the plurality of detection lines (step SB4 in FIG. 10). Of the luminance information of the detection line, the luminance level is compared and tracked for each pixel from the left side to the center of the coordinate (step SB5 in FIG. 10). The specified dot is a pixel in which the luminance information of adjacent pixels on the detection line is compared, the pixel whose luminance information value starts to decrease is detected as the peak position, and the luminance information value decreases for a predetermined number of pixels from the peak position. Is detected (step SB6 in FIG. 10). Similarly, the luminance level is compared and tracked for each pixel from the right side of the page toward the center of coordinates, and the peak position and the designated dot are detected (repeating step SB5 and step SB6 in FIG. 10).

When it is determined that the luminance difference between the luminance information of the peak position and the luminance information of the designated dot is greater than or equal to the second threshold value, it is determined that the peak position is a slag line (step SB7 in FIG. 10). Steps SB5 to SB7 are repeated for all detection lines in the ROI, and the number of slag lines is determined for each line. The largest value in the ROI is determined as the number of slags in the ROI (step in FIG. 10). SB8). When the slag stripes of the region of interest are determined, the number of slag stripes is evaluated (step SB9 in FIG. 10).
This process is performed for each of a plurality of detection lines in the region of interest ROI (SL) and ROI (WL).

  As described above, according to the slag monitoring device 100a and the method of the coal gasification furnace 1 according to the present embodiment, the vicinity of the slag hole where the molten slag flows out, and the slag flowing out of the slag hole 3 falls. Luminance information is acquired from a plurality of detection lines provided in a direction intersecting with the direction in which the falling stripe of the slag is estimated to fall in a predetermined region of the captured image acquired by imaging the vicinity of the water surface of the cooling water. The luminance information of adjacent pixels on the detection line is compared from at least one end side to the other end side of the detection line, and a pixel whose luminance information value starts to decrease is detected as a peak position.

When a designated dot is detected that is a pixel whose luminance information value has fallen continuously for a predetermined number of pixels from the detected peak position, the luminance difference between the luminance information of the designated dot and the luminance information of the peak position is compared. Is equal to or greater than the second threshold value, the peak position is determined as a slag drop streak, and the number of slag drop streaks in a predetermined region is output.
In this way, when there is a designated dot with a predetermined number of pixels continuously decreasing from the peak position and the luminance difference between the peak position and the designated dot is equal to or greater than the second threshold, the peak position is a slag streak. Therefore, it is possible to accurately determine the slag falling stripes and output the exact number of slag falling stripes.

  In addition, the method of 1st Embodiment and this 2nd Embodiment may be combined, and the one with many slag muscles calculated | required by both may be determined as final slag muscles, and, thereby, more accurate slag muscles Can be determined.

  Here, in Fig.11 (a), the result (solid line; peak luminance tracking) which observed the slag hole by the slag monitoring apparatus 100a which concerns on the 2nd embodiment, and the slag monitoring apparatus 100 which concerns on 1st Embodiment. The results of slag hole observation (dotted line; WTS + luminance) are superimposed and displayed. Based on FIG. 11A, FIG. 11B is a graph showing a case where the number of slug lines is larger as a final slug line. FIG. 11C shows the result of observation by the conventional spatial filter method.

11 (a), 11 (b), and 11 (c) show an off-delay operation function (for example, “0” is set to “0” three times in succession to avoid false detection due to sudden event changes, and “3 times "Is a set value). Comparing FIG. 11 (b) and FIG. 11 (c), in the case of FIG. 11 (b) (detected by combining the first embodiment and the second embodiment), all of them are one (one or more) slugs. A streak is detected. On the other hand, in the conventional method, the number of slag muscles changes frequently, the number of slag muscles also changes from 0 to 2, and there are many cases where the slag muscles cannot be recognized.
As described above, stable detection is possible even in an event that could not be detected in the past (in the case where a decrease in image level, blur, or the like occurs).

  In FIG. 12A, the water surface was observed by the slag monitoring device 100a according to the second embodiment (solid line; peak luminance tracking) and the water surface was observed by the slag monitoring device 100 according to the first embodiment. The result (dotted line; WTS + luminance) is superimposed and displayed. Based on FIG. 12 (a), FIG. 12 (b) is a graph showing the case where the number of slug lines is larger as the final slug line. FIG. 12C shows a result observed by the conventional spatial filter method. In the water surface image, the same tendency as in FIG. 11 was observed.

FIGS. 12A, 12B, and 12C show an off-delay operation function (for example, “0” is set to “0” three times in succession in order to avoid false detection due to sudden event changes, and “3 times "Is a set value).
Comparing FIG. 12 (b) and FIG. 12 (c), in the case of FIG. 12 (b) (detected by combining the first embodiment and the second embodiment), all of the leading (one or more) slags Streaks are detected and stable. On the other hand, in the conventional method, the number of slag muscles changes frequently, and the number of slag muscles also changes from 0 to 3, and there are many cases where the slag muscles cannot be recognized.
Similarly to FIG. 11, stable detection is possible even in an event that could not be detected in the past (when a decrease in image level, blur, or the like occurs).

[Third Embodiment]
Hereinafter, a third embodiment of the present invention will be described with reference to FIGS. The slag monitoring apparatus according to the third embodiment determines the binarization level from the luminance state of the background of the image to be monitored and obtains the area of the slag streak in the first and second embodiments. Different from the embodiment. Hereinafter, description of points common to the first embodiment and the second embodiment will be omitted, and different points will be mainly described.

In FIG. 13, the whole block diagram of the slag monitoring apparatus 100b which concerns on this embodiment is shown.
As illustrated in FIG. 13, the slag monitoring device 100b includes an imaging unit (imaging unit) 101, a binarization level determination unit (binarization level determination unit) 107, and an area calculation unit (area calculation unit) 108. It has.
The imaging unit 101 acquires a slag hole image obtained by imaging the vicinity of the slag hole from which the molten slag flows out, and a water surface image obtained by imaging the vicinity of the cooling water from which the slag flowing out from the slag hole falls.

The binarization level determination unit 107 determines a binarization level that is set according to the average value (average luminance level) of the luminance of each captured image. The binarization level is a threshold value for binarizing the captured image. Specifically, the binarization level determination unit 107 determines the binarization level corresponding to the average value of the luminance of the region of interest ROI (s) for the slag hole image, and for the water surface image. A binarization level corresponding to the average value of the luminance of the region of interest ROI (w) is determined.
In FIG. 14, the horizontal axis represents the ROI average luminance level, and the vertical axis represents the binarization level.
As shown in FIG. 14, the binarization level determination unit 107 has an Fx setting (for example, a table, a function, etc.) for outputting the binarization level with respect to the average luminance of the background of the image. The binarization level determination unit 107 outputs the binarization level when the calculated average luminance is input.

The area calculation unit 108 binarizes each pixel that is a predetermined monitoring target of the captured image based on the determined binarization level, sets a pixel that is equal to or higher than the binarization level as a slag region, and calculates the area of the slag region. The predetermined monitoring target is a region of interest ROI (s) for which the binarization level is obtained and a range smaller than the region of interest ROI (w). Specifically, in the slag hole image, the predetermined monitoring objects are ROI (1) and ROI (2), and in the water surface image, the predetermined monitoring objects are ROI (4) and ROI (5).
Here, the area of the slag region is the area of the slag streaks that flowed out from the slag hole 3 in the slag hole image, and in the water surface image, the area of the accumulated slag that has accumulated in the cooling water and has reached the water surface 5H. including.

Next, the operation of the slag monitoring apparatus 100b according to this embodiment will be described with reference to FIGS. FIG. 15 shows an operation flow of the slag monitoring device 100.
The slag hall camera 11 images the vicinity of the slag hole with a color moving image, and the water surface camera 12 images the vicinity of the water surface 5H with a color moving image. When the analysis start command is acquired, ten image captures are performed from the moving images acquired from the slag hall camera 11 and the water surface camera 12 (step SC1 in FIG. 15). One black and white image converted from black and white is generated from the 10 captured images (step SC2 in FIG. 15).

  The generated black and white image is smoothed by the smoothing function (step SC3 in FIG. 15). In the smoothed image, the average luminance level of each region of interest ROI (s) and region of interest ROI (w) is calculated (step SC4 in FIG. 15). Based on the Fx setting and the average luminance level, a binarization level is calculated (step SC5 in FIG. 15). Based on the binarization level calculated from the slag hole image, the binarization processing is performed on ROI (1) and ROI (2), and on the basis of the binarization level calculated from the water surface image, ROI (4), ROI (5) is binarized (step SC6 in FIG. 15). An area ratio corresponding to the slag region of the binarization level or higher is calculated (step SC7 in FIG. 15). Data management of the calculated area ratio is performed and evaluated. (Step SC8 in FIG. 15).

  As described above, according to the slag monitoring apparatus 100b and the method of the coal gasification furnace 1 according to the present embodiment, the vicinity of the slag hole where the molten slag flows out, and the cooling where the slag flowing out from the slag hole 3 falls. Predetermined regions of interest ROI (s) and ROI (w) are set in the captured images acquired by imaging the vicinity of the water surface, and the average value of the luminance in the predetermined regions of interest ROI (s) and ROI (w) In accordance with the binarization level, each pixel is binarized into ROI (1), ROI (2), ROI (4), and ROI (5) as predetermined monitoring areas based on the binarization level. The In the binarized captured image, the area is calculated using pixels that are equal to or higher than the binarization level as slag regions.

  As described above, since the binarization level is obtained according to the average value of the luminances of the predetermined regions of interest ROI (w) and ROI (s) provided in each captured image, the captured image obtained by capturing the vicinity of the slag hole. The binarization level can be set for each of the captured image obtained by capturing the vicinity of the water surface where the slag falls. In addition, since the binarization level changes depending on the situation (background luminance) where a predetermined monitoring image is captured, the slag can be more accurately compared with the case where a fixed binarization level value is used. Can determine muscle.

FIG. 16 shows a case where the slag hole is observed by the slag monitoring apparatus 100b according to the present embodiment, and the area ratio of the slag muscle is calculated, and the area ratio is calculated by the conventional method (binarization level: fixed). FIG.
The area ratios of ROI (1) and ROI (2) are smaller than that of the conventional method, and the slag area ratio has a smaller variation in the slag area ratio. It has been.

FIG. 17 shows the case where the water surface is observed by the slag monitoring device 100b according to the present embodiment and the area ratio of the slag muscle is calculated, and the area ratio is calculated by the conventional method (binarization level: fixed). FIG.
The area ratio of ROI (4) and ROI (5) shows that the third embodiment has a stable result corresponding to the in-furnace situation with less fluctuation of the slag area ratio and reasonable behavior compared to the conventional method. Has been obtained.

[Fourth Embodiment]
Hereinafter, a fourth embodiment of the present invention will be described with reference to FIGS. The slag monitoring apparatus according to the fourth embodiment is a first embodiment, a second embodiment, and a third embodiment in that a binarization level is determined based on luminance information of a predetermined area including slag streaks. Different from form. Hereinafter, description of points that are common to the first embodiment, the second embodiment, and the third embodiment will be omitted, and different points will be mainly described. Note that in the fourth embodiment, only the processing of the water surface image is performed.

FIG. 18 shows an overall configuration diagram of the slag monitoring apparatus 100c according to the present embodiment. As illustrated in FIG. 18, the slag monitoring apparatus 100c includes an imaging unit (imaging unit) 101, a binarization level determination unit 107c, and an area calculation unit 108c.
The imaging part 101 acquires the water surface image which imaged the water surface vicinity of the cooling water in which the slag which flowed out from this slag hole falls.

  The binarization level determination unit 107c has a binarization level (binarization) based on a maximum value (maximum luminance) and an average value (average luminance) of luminance information of a predetermined area set for each slag in the captured image. The threshold for). Specifically, the binarization level determination unit 107c sets ROI (4) and ROI (5) that are predetermined regions of interest for each of the slags that enter the water surface 5H of the water surface image, and ROI (4) and In each ROI (5), the binarization level is determined based on the maximum value and the average value of the luminance information.

More specifically, the binarization level determination unit 107c multiplies the difference between the maximum value of the luminance information in the predetermined area and the average value of the luminance information in the predetermined area by a coefficient, and the multiplication result and the luminance information The binarization level is calculated by adding the average value (see equation (2)). Here, k is a coefficient.
Binarization level =
(Maximum brightness within ROI−average brightness of ROI) × k + ROI average brightness (2)

  The area calculation unit 108c binarizes each pixel in the predetermined region based on the determined binarization level, sets a pixel that is equal to or higher than the binarization level as a slag region, and calculates the area of the slag region.

Next, the operation of the slag monitoring apparatus 100c according to this embodiment will be described with reference to FIGS. FIG. 19 shows an operation flow of the slag monitoring device 100.
The water surface camera 12 images the vicinity of the water surface 5H with a color moving image. When the analysis start command is acquired, ten image captures are performed from the moving image acquired from the water surface camera 12 (step SD1 in FIG. 19). One black-and-white image converted from black and white is generated from the 10 captured images (step SD2 in FIG. 19).

  The generated black and white image is smoothed by the smoothing function (step SD3 in FIG. 19). In the smoothed image, the average luminance levels of the region of interest ROI (4) and the region of interest ROI (5) are calculated, and the luminance information of the region of interest ROI (4) and the region of interest ROI (5) is calculated. Is output (step SD4 in FIG. 19). A binarization level is calculated based on the region of interest (4) and the region of interest ROI (5) (step SD5 in FIG. 19). Based on the calculated binarization levels, binarization processing is performed on the region of interest ROI (4) and the region of interest ROI (5) (step SD6 in FIG. 19). An area ratio corresponding to the slag region of the binarization level or higher is calculated (step SD7 in FIG. 19). Data management of the calculated area ratio is performed and evaluated. (Step SD8 in FIG. 19).

  As described above, according to the slag monitoring device 100c and the method of the coal gasification furnace 1 according to the present embodiment, the vicinity of the coolant surface where the slag flowing out from the slag hole 3 through which the molten slag flows out falls. The region of interest ROI (4) and the region of interest ROI (5) including slug muscles are set for the captured image acquired by imaging, and the maximum luminance information in the region of interest ROI (4) and the region of interest ROI (5) The binarization level is determined based on the value and the average value, and each pixel in the predetermined area is binarized based on the determined binarization level, and pixels that are equal to or higher than the binarization level are defined as slag areas. The area is calculated.

  Thus, since the binarization level is set in consideration of the maximum value and the average value of luminance information, it is estimated that ROI (4) and ROI (5) include slag streaks in one captured image, For example, even if the ROI (4) slag streaks have high brightness and the ROI (5) slag streaks have low brightness, the two levels of the captured image are based on the binarization level corresponding to each region. Pricing is performed. Thereby, the binarization level according to the brightness | luminance in a region of interest is set, and the area calculation of slug stripes can be performed correctly.

FIG. 20 shows the case where the water surface is observed and the area ratio of the slag streaks is calculated by the slag monitoring device 100c according to the fourth embodiment (FIG. 20C), and the slag monitoring according to the third embodiment. Comparison between the case where the area ratio of slag muscle is calculated by the apparatus 100b (FIG. 20B) and the case where the area ratio is calculated by the conventional method (binarization level: fixed) (FIG. 20A). It is an example.
The ROI (4) slag accumulation area ratio (%) (dotted line) and the ROI (5) slag accumulation area ratio (%) (solid line) are more slag in the fourth embodiment than in the conventional method. Stable results corresponding to the in-furnace situation are obtained with a reasonable behavior with small fluctuation of the area ratio. Further, the area ratio (FIG. 12C) of ROI (4) and ROI (5) by the slag monitoring device 100c according to the fourth embodiment is the ROI (4) by the slag monitoring device 100b of the third embodiment. ) And the area ratio of ROI (5) (FIG. 12B), the detection result is more accurate.

Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention.
Note that the first embodiment to the fourth embodiment may be combined as appropriate. For example, the configuration of the second embodiment may be combined with the configuration of the first embodiment, or the configuration of the fourth embodiment may be combined with the configuration of the first embodiment.

Moreover, the information on the number of slug lines obtained by the first embodiment and the second embodiment and the information on the area of the slug lines obtained by the third embodiment and the fourth embodiment are, for example, It may be used to determine whether or not the slag flow muscle is stable and whether or not the slag hole may be blocked, and various guidance may be output as appropriate based on the determination result. For example, when it is determined that there is a possibility that the slag hole may be blocked, a notification is made from the display or the speaker, and when it is determined that slag has accumulated in the slag hole, Informing that the ignition of the molten burner is urged, and in the case where it is determined that it is time to clean the monitoring window of the slag hall camera or the water surface camera, informing that effect on the display or speaker. The worker who has received the notification can take action based on the notification.
Further, the results of the number of slag bars and the area of the slag bars obtained by the first to fourth embodiments may be applied to various determination logics described in Japanese Patent No. 5448669.

DESCRIPTION OF SYMBOLS 1 Coal gasifier 2 Slag tap 3 Slag hole 5 Slag hopper 5H Water surface 100,100a, 100b, 100c Slag monitoring apparatus 101 Imaging part 102 Ridge line detection part 103 Luminance information acquisition part 104,104a Falling line determination part 105 Peak detection part 106 designation | designated Dot detection unit 107, 107c binarization level determination unit 108, 108c area calculation unit P detection line X ridge line e1 first coordinate e2 second coordinate

FIG. 20 shows the case where the water surface is observed and the area ratio of the slag streaks is calculated by the slag monitoring device 100c according to the fourth embodiment (FIG. 20C), and the slag monitoring according to the third embodiment. Comparison between the case where the area ratio of slag muscle is calculated by the apparatus 100b (FIG. 20B) and the case where the area ratio is calculated by the conventional method (binarization level: fixed) (FIG. 20A). It is an example.
The ROI (4) slag accumulation area ratio (%) (dotted line) and the ROI (5) slag accumulation area ratio (%) (solid line) are more slag in the fourth embodiment than in the conventional method. Stable results corresponding to the in-furnace situation are obtained with a reasonable behavior with small fluctuation of the area ratio. Moreover, the area ratio (FIG. 20 (c)) of ROI (4) and ROI (5) by the slag monitoring device 100c according to the fourth embodiment is the ROI (4) by the slag monitoring device 100b of the third embodiment. ) and (area ratio of 5) (FIG. 20 (b) ROI has a) more accurate than the detection results.

Claims (11)

Imaging means for capturing the vicinity of the slag hole where the molten slag flows out, and the vicinity of the water surface of the cooling water where the slag flowing out of the slag hole falls,
Using a watershed algorithm for each of the captured images, ridge line detection means for detecting a ridge line estimated as a drop line of the slag,
Luminance information acquisition means for acquiring luminance information from a plurality of detection lines provided in a direction intersecting with the estimated drop line of the slag in a predetermined region of each captured image;
When the luminance difference between the luminance information at the intersection of the ridge line and each detection line and the luminance information in the vicinity of the intersection along each detection line is equal to or greater than a first threshold, the ridge line is A slag monitoring device for a coal gasification furnace, comprising fallen line determining means for determining that the slag is dropped.   The falling line determination means obtains the luminance information in the vicinity of the intersection from a first coordinate that is separated from the intersection by a first predetermined distance along the detection line, and from the first coordinate along the detection line. The slag monitoring device for a coal gasification furnace according to claim 1, wherein the slag monitoring device is an average value of luminance in a section up to a second coordinate separated by a predetermined interval. Imaging means for capturing the vicinity of the slag hole where the molten slag flows out, and the vicinity of the water surface of the cooling water where the slag flowing out of the slag hole falls,
Luminance information acquisition means for acquiring luminance information from a plurality of detection lines provided in a direction intersecting with a direction in which the falling stripe of the slag is estimated to fall in a predetermined region of each captured image;
A peak for comparing the luminance information of adjacent pixels on the detection line in a direction from at least one end side to the other end side of each detection line, and detecting the pixel at which the value of the luminance information starts to fall as a peak position Detection means;
Designated dot detection means for detecting a designated dot that is the pixel in which the value of the luminance information has fallen continuously for a predetermined number of pixels from the peak position;
Coal comprising fall line determination means for determining the peak position as a drop line of the slag when the luminance difference between the luminance information of the peak position and the luminance information of the designated dot is equal to or greater than a second threshold. Gasifier slag monitoring device. The falling line determining means includes
Comprising a counting means for counting the number of slag falling stripes determined as the slag falling stripes for each detection line;
4. The device according to claim 1, wherein, in the predetermined area, a maximum value among the number of falling stripes of the slag counted for each detection line is output as the number of falling stripes of the slag in the predetermined area. A slag monitoring device for a coal gasifier according to claim 1. Imaging means for capturing the vicinity of the slag hole where the molten slag flows out, and the vicinity of the water surface of the cooling water where the slag flowing out of the slag hole falls,
Binarization level determination means for determining a binarization level set according to an average value of luminance of each of the captured images;
Based on the determined binarization level, each pixel to be monitored in the captured image is binarized, the pixel that is equal to or higher than the binarization level is set as a slag region, and an area calculation for calculating the area of the slag region And a slag monitoring device for a coal gasifier. Imaging means for capturing an image of the vicinity of the surface of the cooling water from which the slag that has flowed out of the slag hole from which the molten slag flows out, and acquiring a captured image;
In the captured image, binarization level determination means for determining a binarization level based on a maximum value and an average value of the luminance information of a predetermined area set for each slag;
Area calculating means for binarizing each pixel in the predetermined area according to the determined binarization level, setting the pixel that is equal to or higher than the binarization level as a slag area, and calculating an area of the slag area; Slag monitoring device for coal gasifier. The binarization level determination means includes
The binarization level is calculated by multiplying a difference between the maximum value and the average value of the luminance information in the predetermined area by a coefficient, and adding the multiplication result and the average value of the luminance information. Item 7. A slag monitoring device for a coal gasifier according to item 6. A first step of imaging the vicinity of the slag hole from which the molten slag flows out, and the vicinity of the water surface of the cooling water from which the slag that has flowed out of the slag hole falls;
Using a watershed algorithm for each of the captured images, a second step of detecting a ridge line estimated as a drop line of the slag;
A third process of acquiring luminance information from a plurality of detection lines provided in a direction intersecting with the estimated drop of the slag in a predetermined region of each captured image;
When the luminance difference between the luminance information at the intersection of the ridge line and each detection line and the luminance information in the vicinity of the intersection along each detection line is equal to or greater than a first threshold, the ridge line is A slag monitoring method for a coal gasification furnace, comprising: a fourth process for determining that the slag falls. A first step of imaging the vicinity of the slag hole from which the molten slag flows out, and the vicinity of the water surface of the cooling water from which the slag that has flowed out of the slag hole falls;
A second process of acquiring luminance information from a plurality of detection lines provided in a direction intersecting with a direction in which a drop line of the slag is estimated to fall in a predetermined region of each captured image;
The luminance information of adjacent pixels on the detection line is compared in at least one end side to the other end side of each detection line, and the pixel at which the value of the luminance information starts to decrease is detected as a peak position. 3 processes,
A fourth step of detecting a designated dot which is the pixel in which the value of the luminance information has decreased continuously from the peak position by a predetermined number of pixels;
Coal gasification comprising: a fifth step of determining the peak position as a drop line of the slag when the luminance difference between the luminance information of the peak position and the luminance information of the designated dot is greater than or equal to a second threshold value How to monitor furnace slag. A first step of imaging the vicinity of the slag hole from which the molten slag flows out, and the vicinity of the water surface of the cooling water from which the slag that has flowed out of the slag hole falls;
A second step of determining a binarization level set according to the average value of the luminance of each captured image;
According to the determined binarization level, each pixel that is a predetermined monitoring target of the captured image is binarized, the pixel that is equal to or higher than the binarization level is set as a slag region, and the area of the slag region is calculated. A method for monitoring slag of a coal gasifier having a process. A first step of capturing an image of the vicinity of the cooling water from which the slag that has flowed out of the slag hole from which the molten slag has flowed falls,
A second step of determining a binarization level based on a maximum value and an average value of the luminance information in a predetermined area set for each slag in the captured image;
And a third step of binarizing each pixel in the predetermined region according to the determined binarization level, setting the pixel that is equal to or higher than the binarization level as a slag region, and calculating an area of the slag region. Gasifier slag monitoring method.

Images