JP4845451B2 - Method and apparatus for detecting combustion region in combustion furnace - Google Patents

Description

  The present invention relates to a detection method and a detection device for a combustion region when, for example, garbage burns in a combustion furnace.

  In equipment for burning the combustibles, such as a garbage combustion furnace, it is preferable to grasp the current combustion state, for example, the burn-out point as accurately as possible in order to perform appropriate combustion.

By the way, as a method of grasping such burn-out points, the vicinity of the boundary of the combustion region is photographed with a television camera, and the relationship between the number of pixels on the photographed image and the luminance distribution processing, specifically, In some cases, binarization processing is performed in a direction orthogonal to the direction of movement of dust, and a point where the sum of the number of pixels exceeds a threshold is set as a burnout point (see, for example, Patent Document 1).
JP 2000-274675 (see paragraph [0017])

  According to the process disclosed in the above-mentioned Patent Document 1, when the fluctuation of the bright flame remains in the pixel, it may be erroneously determined that it is the main combustion region despite the noise. Even when the waste of calories is dispersed and burned away from the main combustion region, there is still a possibility that the main combustion region is erroneously determined.

  Accordingly, in order to solve the above-described problems, the present invention has an object to provide a detection method and a detection device for a combustion region in a combustion furnace that can eliminate adverse effects due to uneven combustion due to fluctuations in luminous flame, dispersion of dust, and the like. And

  In order to solve the above problems, a combustion region detection method in a combustion furnace according to the present invention includes a measurement step of measuring radiant energy emitted from a combustion surface of a combusted object with a sensor, and measurement data measured in the measurement step. A data accumulation step for accumulating at least one image of measurement data at each pixel in the measurement area, and inputting the measurement data accumulated at this data accumulation step to binarize the measurement data at each pixel A binarization processing step for processing to extract a high-temperature portion, and a high-temperature portion data in a plurality of images extracted in this binarization processing step are overlapped by performing overlapping processing between images and extracting overlapping portions. A high temperature region recognition that recognizes a high temperature region by performing a labeling process of the overlapping portion extracted in the partial extraction step and the overlapping portion extraction step And steps, the sought after area of the recognized high-temperature region at a high temperature region recognition step, a method and a combustion region determining step of determining the area of the largest area and the combustion region.

  In addition, the detection device for the combustion region in the combustion furnace of the present invention inputs the measurement data measured by the sensor unit that measures the radiant energy released from the combustion surface of the combustion object, and the measurement data measured by the sensor unit. A data accumulating unit capable of accumulating at least one image of measurement data at each pixel in the region, and smoothing for inputting the measurement data accumulated in the data accumulating unit and performing a smoothing process on the measurement data at each pixel A processing unit, a binarization processing unit that binarizes the measurement data in each pixel smoothed by the smoothing processing unit with a predetermined threshold and extracts a high-temperature part, and is extracted by the binarization processing unit The overlapping part extraction unit that extracts the overlapping part by superimposing the high temperature part data in the plurality of images between each image, and the labeling process of the overlapping part extracted by the overlapping part extraction part A high temperature region recognition unit for recognizing a high temperature region, an area calculation unit for obtaining an area of the high temperature region recognized by the high temperature region recognition unit, and determining a region of the maximum area obtained by the area calculation unit as a combustion region And a combustion region determination unit.

  According to the detection method and the detection device for the combustion region, the radiant energy from the combusted object is measured, the obtained measurement data is binarized with a predetermined threshold value, and a plurality of binarized processes are performed. By superimposing pixels of image data, the high temperature region can be made clearer than the low temperature region, and the high temperature region having the maximum area is selected from the multiple high temperature regions that are usually obtained. The remaining small high-temperature area is not taken into account, so the area that continues to burn at high temperatures, that is, the main main combustion area that is not affected by fluctuations such as bright flames and fluctuations, should be detected. Therefore, the combustion state can be accurately controlled, and the stable ignition point position can be detected from this combustion region.

[Embodiment]
Hereinafter, a detection method and a detection device for a combustion region in a combustion furnace according to an embodiment of the present invention will be described.

In the present embodiment, a case will be described in which the combustion object is garbage, that is, the case where the combustion area of the garbage surface in the garbage combustion furnace is detected.
First, a schematic configuration of the refuse combustion furnace will be described with reference to FIG.

  The refuse combustion furnace 1 is, for example, a stoker furnace, and a combustion chamber 5 provided with a combustion bed 4 in which a grate 3 is disposed across a drying stage, a combustion stage, and a post-combustion stage in the furnace body 2. Is provided at one end side of the furnace body 2 and a discharge port 2b for incineration residue is formed at the other end side.

  The waste combustion furnace 1 detects the combustion state of the dust in the combustion chamber 5, that is, the combustion region of the waste (more precisely, the main combustion region) by measuring the radiant energy. A combustion area detection device 7 is provided.

  As shown in FIG. 2, this combustion region detection device 7 is provided on the inclined upper wall 2c from the combustion stage of the furnace body 2 to the post-combustion stage, and inputs radiant energy released from the surface of the combustion waste. A two-dimensional sensor unit 11 in which a plurality of sensors (pixels, specifically, infrared sensors are used) that are converted into electrical energy are arranged vertically and horizontally, and detected by the two-dimensional sensor unit 11 An A / D conversion unit 12 that performs A / D conversion by inputting a detection signal (measurement data, specifically, an electric signal) at predetermined time intervals (so-called sampling), and this Measurement data (sensor output value corresponding to the amount of radiant energy) converted by the A / D conversion unit 12 is input, and measurement data at each pixel in the measurement region (also detection region) is, for example, one image. Min (Also referred to as one frame), a data storage unit 13 that can be stored, a smoothing processing unit 14 that receives measurement data stored in the data storage unit 13 and performs a smoothing process, and the smoothing processing unit 14 performs smoothing. A binarization processing unit (also referred to as a high temperature part extraction unit) 15 that performs binarization processing by comparing the measured data of each pixel and a predetermined threshold value, and extracts the high temperature part, and this binarization processing unit The high-temperature data storage unit 16 that stores the high-temperature part data extracted in 15 for a plurality of frames and the high-temperature part data stored in the high-temperature data storage unit 16 are overlapped between frames, that is, data of pixels An overlap part extraction unit 17 that performs an AND process (superposition process; so-called and process) and extracts an overlap part, and performs a labeling process of the overlap part extracted by the overlap part extraction part 17 and performs a high temperature region (Usually a plurality of large and small regions are recognized), a high temperature region recognition unit 18, an area calculation unit 19 for determining the area of each high temperature region recognized by the high temperature region recognition unit 18, and the area calculation unit Among the areas determined in 19, the combustion area determination unit 20 determines an area having the maximum area and determines the high-temperature area having the maximum area as a combustion area (main combustion area). The image that is the measurement data may be a plurality of images that are continuous in time series, or may be a time series image picked up at predetermined intervals.

Here, the processing in the smoothing processing unit 14 and its usefulness will be described.
In general, a stoker furnace undergoes five combustion processes: waste preheating, drying, pyrolysis / partial combustion, surface combustion, and cooling. Specifically, the surface temperature of the garbage divided by the grate block generally rises from the first stage of the combustion grate, reaches the maximum temperature in the second or third stage of the combustion grate by combustion, After that, after burning, it shows a temperature profile that decreases to about 700 ° C. over the post-combustion grate.

  However, when the surface temperature distribution of the dust is observed in detail, there are locally low and high temperatures, and there is temperature unevenness, that is, there is combustion unevenness, for example, an error factor when detecting a high temperature combustion region ( This also causes an error when detecting the position of the ignition point, burn-out point, etc. on the grate). By applying a smoothing process to the measurement data corresponding to the temperature, the effects of combustion unevenness can be reduced. It is intended to be removed.

  In this smoothing process, when obtaining measurement data about the temperature at a certain pixel, the average value of the measurement data including the peripheral pixels located around the pixel is calculated as the measurement data at the certain pixel. To do.

  For example, as shown in FIG. 3, when obtaining measurement data T (I, J) with respect to temperature at a pixel (I, J) located at the center, the following equation (1) is used. However, here, a case where one block including peripheral pixels to be subjected to smoothing processing is 3 pixels × 3 pixels is shown. Note that the number of one block is not limited to 3 pixels × 3 pixels, and an optimal number of pixels (for example, 5 pixels) according to the size of combustion unevenness (furnace width and direction of movement of dust). X5 pixels to 11 pixels x11 pixels) is selected.

Moreover, as a threshold value used in the binarization processing unit 15, for example, a value corresponding to 900 ° C. which is an ignition point (determination index) is set.
Further, the overlapping part extraction unit 17 performs a logical product process on the same pixel values (“0” or “1”) for each frame. That is, by this logical product processing, only the high temperature region is left as the combustion region over all the frames, and therefore, the portion where the combustion state is unstable and the uneven combustion occurs, or the fluctuation of the bright flame. A very short high temperature part caused by the above is determined to be out of the combustion region, and the influence of the bright flame can be removed here as well.

Next, a method for detecting the combustion region of waste in the combustion chamber 5 of the waste combustion furnace 1 by the combustion region detection device 7 will be described step by step.
(Step 1)
First, the amount of radiant energy released from the dust surface in the combustion chamber 5 is measured by the two-dimensional sensor unit 11 (measurement step).
(Step 2)
The detection signal detected in the measurement step is input to the A / D converter 12 to perform A / D conversion (A / D conversion step).
(Step 3)
The measurement data obtained in the A / D conversion step is input and stored in a data storage unit 13 for a plurality of frames (data for several images in the measurement region) (data storage step).
(Step 4)
The measurement data accumulated in the data accumulation step is input to the smoothing processing unit 14, the average value of the measurement data including the predetermined pixel and its surrounding pixels is obtained, and the measurement data for the predetermined pixel (Smoothing step). By this smoothing step, the influence of data fluctuation due to temperature unevenness is reduced. An image obtained by this smoothing process is shown in FIG.
(Step 5)
Next, the measurement data obtained in the smoothing step is input to the binarization processing unit 15, and binarization processing is performed with a preset threshold (for example, a value corresponding to the temperature of the ignition point). The high temperature combustion region (main combustion region) is extracted (binarization step). In FIG. 4 (a), the portion above the ignition point is indicated by diagonal lines, and the lower portion is indicated by dotted lines.

Then, image data indicating the combustion region extracted in the binarization step, that is, high temperature data is accumulated in the high temperature data accumulation unit 16 for several frames (high temperature data accumulation step).
(Step 6)
The image data for several frames accumulated in the high-temperature data accumulation step is input to the overlapping portion extraction unit 17 and logical product processing is performed for each pixel (overlapping portion extraction step).
(Step 7)
The image data obtained in the overlapping part extraction step is input to the high temperature region recognition unit 18 and a labeling process is performed on the high temperature region to recognize the region (usually a plurality of regions) (high temperature region recognition step). An image of the recognized high temperature region is shown in FIG. 4B (the shaded area is the high temperature region).
(Step 8)
Data of a plurality of high temperature regions obtained in the high temperature region recognition step is input to the area calculation unit 19 to calculate each area (area calculation step).
(Step 9)
Each of the areas calculated in the area calculation step is input to the combustion region determination unit 20 to find the maximum one, and the other high temperature regions are not considered (ignored), that is, the maximum Only the high temperature region is selected (detected) and determined as the combustion region (main combustion region) (combustion region detection step). An image of the combustion region selected in this combustion region detection step is shown in FIG. 4 (c) (indicated by the hatched portion).

By these series of steps (operations), the combustion region, that is, the main combustion region is detected.
In this way, the radiant energy from the dust surface is measured, and the obtained measurement data is smoothed to remove temperature unevenness, that is, combustion unevenness as much as possible, and then binarized with a predetermined threshold. By performing AND processing between pixels of a plurality of binarized image data, the high temperature region can be made clearer with respect to the low temperature region, and the highest of the high temperature regions that are usually obtained Since the high temperature region with the area is selected so that the remaining small high temperature region is not taken into consideration, the main combustion that is not affected by the uneven combustion such as the bright flame, that is, the region that continues to burn at high temperature The region can be detected, and therefore the combustion state can be accurately controlled, and the position of the stable ignition point can be detected from this combustion region. For example, since the ignition point can be a point that exceeds 10% of the area of the main combustion region, if the combustion region is accurate, the position of the ignition point can naturally be known accurately. Since the main combustion region can be detected, it can be applied not only to the ignition point but also to the detection of the burnout point.

  Moreover, in the said embodiment, although the smoothing process part (smoothing step) which performs the smoothing process with respect to the measurement data of each pixel was provided, it can also be abbreviate | omitted depending on the case.

It is sectional drawing which shows schematic structure of the combustion furnace provided with the combustion area | region detection apparatus which concerns on embodiment of this invention. It is a block diagram which shows schematic structure of the combustion area | region detection apparatus. It is a figure explaining the processing content in the smoothing process part in the combustion area | region detection apparatus. It is a figure which shows the combustion region for demonstrating the combustion region detection method, (a) shows after smoothing processing, (b) shows after binarization processing, (c) shows the detected combustion region. Show.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Garbage furnace 3 Grate 5 Combustion chamber 7 Combustion area detection apparatus 11 Two-dimensional sensor part 12 A / D conversion part 13 Data storage part 14 Smoothing process part 15 Binarization process part 16 High temperature data storage part 17 Duplicate part extraction Unit 18 high temperature region recognition unit 19 area calculation unit 20 combustion region determination unit

Claims (4)

A measurement step of measuring the radiant energy emitted from the combustion surface of the burned object with a sensor;
A data accumulation step for inputting measurement data measured in this measurement step and accumulating at least one image of measurement data at each pixel in the measurement region;
A binarization processing step for inputting the measurement data accumulated in this data accumulation step and binarizing the measurement data in each pixel to extract a high temperature portion;
Overlapping part extraction step for extracting the overlapping part by performing the overlapping process between the images, the high temperature part data in the plurality of images extracted in this binarization processing step,
A high temperature region recognition step for recognizing a high temperature region by performing a labeling process of the overlapping portion extracted in this overlapping portion extraction step;
A combustion region detection method in a combustion furnace, comprising: a combustion region determination step of determining a region of the maximum area as a combustion region after obtaining an area of the high temperature region recognized in the high temperature region recognition step.   Between the data accumulation step and the binarization processing step, there is provided a smoothing process step for inputting the measurement data accumulated in the data accumulation unit and performing a smoothing process on the measurement data in each pixel. A combustion region detection method in a combustion furnace according to claim 1. A sensor unit for measuring the radiant energy emitted from the combustion surface of the burned object;
A data storage unit that inputs measurement data measured by the sensor unit and stores at least one image of measurement data at each pixel in the measurement region;
A binarization processing unit that inputs measurement data stored in the data storage unit, binarizes the measurement data in each pixel with a predetermined threshold, and extracts a high-temperature part; and
The high-temperature part data in the plurality of images extracted by the binarization processing unit, the overlapping part extraction unit that performs overlapping processing between each image and extracts the overlapping part,
A high temperature region recognition unit that recognizes a high temperature region by performing a labeling process of the overlapping portion extracted by the overlapping portion extraction unit;
An area calculation unit for obtaining the area of the high temperature region recognized by the high temperature region recognition unit;
A combustion region detection apparatus in a combustion furnace, comprising: a combustion region determination unit that determines a region of the maximum area obtained by the area calculation unit as a combustion region. A smoothing processing unit is provided between the data storage unit and the binarization processing unit for inputting the measurement data stored in the data storage unit and performing a smoothing process on the measurement data in each pixel. A combustion region detection apparatus for a combustion furnace according to claim 3.

Images