JP6554198B1 - Waste quality estimation system and crane operation control system using the waste quality estimation system - Google Patents

Abstract

[PROBLEMS] To grasp the quality of waste in the waste pit more accurately than before, and to control the change in the fuel condition in the incinerator by controlling it so that it becomes uniform at the time of loading by the crane. Provide a waste quality estimation system that can be realized. A waste quality estimation system 500 uses an intelligent information processing technique to input waste quality using input data of input waste generated from image data obtained by imaging a waste pit surface and waste quality teacher data. An input waste quality estimation model generation unit 503 that generates an input waste quality estimation model to be estimated is provided. [Selection] Figure 1

Description

  The present invention relates to a waste quality estimation system in a waste pit. The present invention also relates to a crane operation control system using the waste quality estimation system.

  Conventionally, waste combustion control has controlled the amount of combustion air and the amount of waste supply based on the conditions in the incinerator, but the reaction to changes in the quality (composition, calorific value) of waste introduced into the incinerator There was a limit or a delay by all means.

Patent Document 1 describes the color, shape, and crushing of individual garbage bags from image information in the vicinity of the furnace inlet for the purpose of enabling stable combustion control even when the composition of combustion is subject to severe composition fluctuations. A method is disclosed to identify waste, shear waste, and other waste types, and estimate the weight and heat value of each waste input based on the waste heat value database.
Patent Document 2 discloses a method aiming at performing continuous combustion control of waste with no time delay with respect to the current fuel state by continuously and accurately acquiring information related to the calorific value of the waste in real time. .

  Further, as a method of detecting the agitation state of the dust in order to equalize the quality of the dust in the dust pit, there is, for example, Patent Document 3. According to Patent Document 3, the image data of the captured dust is tone-converted to luminance value to extract an edge, the fractal dimension of the extracted edge is calculated, and the agitation state of the dust is detected based on the calculated fractal dimension. It discloses the configuration.

Patent No. 5361595 Patent No. 5996762 Patent No. 6188571

The present invention is a different approach from the above-mentioned patent documents etc., and grasps the refuse quality in a refuse pit more precisely than before and controls the fuel in the incinerator by controlling it to be uniform at the time of throwing by a crane. The object is to provide a waste quality estimation system capable of suppressing changes in the situation and realizing stable combustion.
Another object of the present invention is to provide a crane operation control system using a waste quality estimation system.

The garbage quality estimation system of the present invention includes:
An input waste that generates an input waste quality estimation model that estimates waste quality by intelligent information processing technology using input data of input waste created from image data obtained by imaging the surface of the waste pit and waste quality teacher data A quality estimation model generation unit;
The dust quality estimation system may have an imaging unit that images the surface of the dust pit.
The garbage quality estimation system may include an input data creation unit that creates input data of input garbage from image data obtained by imaging a dust pit surface.
The waste quality estimation system may have a storage unit that stores input data of input waste.
The waste quality estimation system may have a receiving unit that receives input data of input waste calculated by another system.

In the above invention, the input waste quality estimation model generation unit is based on a time delay amount from when waste is input to the input port of the combustion apparatus until combustion is started, or a record of input waste amount and supplied waste amount, The waste quality teacher data may be temporally associated with the input data to generate the input waste quality estimation model.
The time delay amount may be set in advance, and may be set by, for example, an empirical rule or an experiment. When the combustion apparatus is, for example, a stoker incinerator, for example, 30 to 40 minutes may be set as the time delay amount.
The waste quality estimation system may further include a time delay amount calculation unit that calculates a time delay amount from the time when waste is introduced into the inlet of the combustion device to the start of combustion.

The “garbage quality teacher data” includes, for example, the concentration or amount of waste composition (water, carbon, hydrogen, hydrogen chloride (HCl) and sulfur dioxide (SO ₂ ), etc. in exhaust gas), calorific value, and the like.
The waste quality estimation system may have a waste quality teacher data creation unit that creates waste quality teacher data.
The composition concentration of the waste can be obtained, for example, by analyzing the exhaust gas with an analyzer corresponding to various components such as an exhaust gas analyzer. Further, other components (for example, carbon dioxide) may be calculated from predetermined components (for example, oxygen and moisture).
Further, the calorific value may be calculated, for example, using the method of Japanese Patent No. 5996762, or a measured value may be used.

In the above invention, the input data may further include one or more types of information from among the waste in-car, the date and the weather.
The input data creation unit may create the input data so as to further include one or more types of information from among the waste in-car, the date, and the weather.
In this case, the waste quality teacher data may further include one or more types of information from the waste in-vehicle, the date, and the weather.

In the above invention, the imaging unit is, for example, a visible camera, a spectrum camera, an infrared camera, or the like.
When the imaging unit is a spectral camera, the input data generation unit (or the area ratio calculation system for absorbance) is
A spectral binarized image generation unit that generates a spectral binarized image based on the absorbance for each wavelength separated from the spectral image data;
An area ratio calculation unit may be provided which calculates an area ratio equal to or greater than a threshold value for each wavelength of each evaluation area and calculates an area ratio for the evaluation area. In this case, the input data is an area ratio.
When the imaging unit is a spectral camera, the input data generation unit (or absorbance average value calculation system)
You may have the light absorbency average value calculation part which calculates the average value of the light absorbency in each evaluation area for every wavelength divided from spectrum image data. In this case, the input data is “absorbance average value”.
When the imaging unit is a visible camera, the input data creation unit (or mixing degree evaluation system)
You may have a mixing degree evaluation part which processes the image data predetermined and evaluates the mixing degree of the garbage of each evaluation area. In this case, the input data is the degree of mixing.
The input data may be one or more of an area ratio, an average value of absorbance, and a degree of mixing.
The input data may include an input time at which the waste is input to the combustion apparatus (for example, an input hopper).

In the above invention, the intelligent information processing technology is, for example, machine learning or deep learning.
The configurations of machine learning and deep learning are not particularly limited, and conventional algorithm configurations may be used.
The input waste quality estimation model is not limited to what is configured as a so-called software program, and may be configured as a dedicated electronic circuit or embedded firmware.
The software program may be stored in memory and the program procedure may be executed by the processor.
The input waste quality estimation model may be updated using different waste quality teacher data.

Another crane operation control system according to the present invention comprises a model storage unit storing an input waste quality estimation model generated by the above-described waste quality estimation system;
A waste quality estimation unit that inputs waste input data to the input waste quality estimation model and estimates waste quality,
The automatic operation control unit automatically controls the crane according to the estimated waste quality output from the waste quality estimation model.
The input data is data including one or more of the input data used in the garbage quality estimation system.
The estimated waste quality is data including one or more of the above-mentioned waste quality teacher data.

The automatic operation control unit may control the crane so that the waste quality is uniform in each evaluation area.
The automatic operation control unit may control the crane so that waste having a high calorific value is thrown when the calorific value of the combustion device is lower than an expected value.
The above-mentioned automatic operation control part may control a crane to throw in refuse with a low calorific value, when calorific value of a combustion device is higher than an expected value.

The following configuration may be employed as a mixing degree evaluation system for calculating the mixing degree.
The mixing degree evaluation system
An imaging unit installed so as to image the garbage in the garbage pit from above;
A laser range finder installed movably above the dust so that the distance to the dust can be detected in the dust pit, and detecting the distance (A) to the measurement point (P);
A measurement point dust height calculation unit that calculates the measurement point dust deposition height (Z) based on the distance (A) and the reference distance (B) from the laser distance meter to the bottom surface of the dust pit;
A three-dimensional dust height calculation unit that calculates three-dimensional height information of the dust based on the plane coordinates in the dust pit and the information of the measurement point dust deposition height;
An image conversion unit configured to convert an image captured by the imaging unit into an aerial viewpoint image based on installation information of the imaging unit;
An image correction unit for obtaining a corrected image corrected so that all the areas of the sky viewpoint image are on the same height plane based on the three-dimensional height information of the dust;
A binarization processing unit that gradations the corrected image and binarizes with a predetermined threshold to obtain a binarized image;
The binarized image may be divided into two or more evaluation areas having a plurality of divided areas, and a mixing degree evaluation unit for evaluating the degree of mixing of the dust in each evaluation area may be included.
The mixing degree evaluation unit is
Calculate the extraction area of the bright or dark part of each evaluation area, calculate the variation of the extraction area of the bright or dark part of the divided area for all evaluation areas or for each evaluation area, and evaluate the degree of mixing by the variation It may have a variation evaluation part.
Here, the variation of the extraction area includes, for example, variance σ ² , standard deviation σ, variation coefficient CV, and the like.
In the evaluation of the above-mentioned fluctuation evaluation unit, as the fluctuation, for example, dispersion σ ² , standard deviation σ, coefficient of variation CV, etc., is larger than a predetermined value, various dusts are present in the area, that is, agitation is sufficiently It shows that the waste is highly homogenized.
The mixing degree evaluation unit
For example, discriminant analysis (binarization of Otsu) is applied to each divided area, or binarization is performed by dynamic threshold method, and each particle is extracted by labeling bright portions (or dark portions), etc. And an area average evaluation unit that calculates the particle area average of each evaluation area and evaluates the degree of mixing by the area average.
In the evaluation of the area average evaluation unit, for example, the more the particle area average is smaller than a predetermined value, the more the waste is being broken, that is, the stirring is sufficiently performed, indicating that the homogenization of the waste is high. .
The mixing degree evaluation unit
You may have an area ratio evaluation part which calculates the area ratio of the bright part of each evaluation area, and a dark part, and evaluates a mixing degree by the said area ratio.
Here, the area ratio is, for example, dark part / bright part, bright part / dark part.
In the evaluation of the area ratio evaluation unit, the larger the area ratio (for example, a value less than 1), the larger the predetermined value, the more the breakage of the waste proceeds, that is, the stirring is sufficiently performed. Garbage bags are brighter than wastes in many cases, so it is estimated that the larger the area of the dark part, the more the garbage bags are crushed.

According to the above-described waste quality estimation system, since the intelligent information processing technology is used, the waste quality can be suitably estimated.
According to the crane operation control system, it is possible to achieve uniform combustion waste quality, that is, stabilization of combustion.
Further, by realizing stabilization of the combustion, advantages such as reduction of auxiliary fuel, reduction of the amount of medicine used by stabilization of exhaust gas properties, saving of operation condition monitoring and the like can be mentioned.
In addition, it is possible to optimize the stirring for the purpose of uniforming the quality of waste in the pit, and it is also possible to reduce the power consumption of the crane and save energy.

It is a figure for demonstrating an example of the garbage quality estimation system of Embodiment 1, and a garbage pit side view. It is a figure which shows an example of the luminance image of the whole refuse pit. It is a figure which shows an example of the binarized image which binarized the luminance image of FIG. 2A. It is a figure which shows an example which fractionated the binarized image of FIG. 2B in the evaluation area and the division area. It is a figure which shows the address rule of evaluation area. It is a figure which shows the address rule of a division area. It is a figure which shows the extraction image number of each division area in evaluation area Sa. 6 is a flowchart illustrating processing procedures. It is a figure for demonstrating the structural example which calculates mixing degree. It is a figure for demonstrating the example of another structure which calculates mixing degree. It is a figure for demonstrating the example of another structure which calculates mixing degree. It is a figure for demonstrating the estimation system of refuse quality. It is a figure for demonstrating the structural example which calculates the area ratio of a light absorbency. It is a figure for demonstrating the structural example which calculates a light absorbency average value. It is a figure which shows an example of the binarized image for every divided wavelength. It is a figure which shows an example of the evaluation area of the binarized image for every wavelength division. It is a figure which shows an example of the area ratio (%) for every wavelength which each evaluation area was divided. It is a figure which shows an example of the light absorbency average value for every divided wavelength of each evaluation area. It is a figure which shows an example of the database of input data. It is a figure which shows an example of the input data of the thrown-in refuse. It is a figure which shows an example of waste quality teacher data. It is a figure which shows an example of the accumulation state of the thrown-in waste, and an in-feed state.

(Embodiment 1)
FIG. 1 is an example of a waste quality estimation system 500 and a side view of a waste pit 1.

  FIG. 1 shows the inside of the waste pit 1 in a side view. The waste collection car 9 enters the right side of FIG. 1 and sends waste into the loading area 11 from the loading door 12. The waste D in the input area is transported to the waste stirring area 15 of the waste pit 1 by the crane 22 and the bucket 23. In the stirring area 15, a mixing address divided into a predetermined size on the floor plane is set, and in the first embodiment, the mixing address and the evaluation area coincide. In the mixing area 15, a state where dust is accumulated (D1: accumulated waste) is shown.

(Crane operation control unit)
The crane operation control system 600 moves the crane 22 within or across any mixing address, grabs and lifts the waste with the bucket 23, and moves it to another position on the spot or within the mixing address. Or move to another mixed address and control to drop the waste. In addition, the crane operation control system 600 controls the crane 22 and the bucket 23 so as to move the waste at a plurality of mixing addresses. The crane operation control system 600 has an automatic operation control unit (not shown) and a manual operation control unit 603, and the automatic operation control unit 603 can execute the above control. In addition, the crane operation control system 600 controls the crane 22 and the bucket 23 to grasp and move the waste at an arbitrary mixing address, and controls the feeding of the waste to the input hopper 41 of the incinerator 40. The method of using the input waste quality estimation model will be described later.

  The dust quality estimation system 500 includes the mixing degree, the area ratio of absorbance, and the average absorbance value in each evaluation area from the mixing degree evaluation system 100, the area ratio calculation system 200 for absorbance, and the average absorbance value calculation system 300 described later. The data is received and stored in the storage unit 501. Details of the waste quality estimation system 500 will be described later.

(Example 1 of input data: Configuration for obtaining the degree of mixing)
A mixing degree evaluation system 100 for calculating the mixing degree and its process flow will be described with reference to FIGS. 2 to 4A to 4C. The waste mixture evaluation system 100 shown in FIG. 4A has the following configuration.

The imaging unit 31 images the dust in the dust pit 1 (for example, the stirring area 15) from above. The imaging unit 31 can be fixed and installed on, for example, a pillar, a wall, a ceiling or the like in the dust pit 1, and is installed on a wall below the crane guard 21 in the present embodiment. The imaging unit 31 may be, for example, a CCD camera, a CMOS camera, or an infrared camera that captures a moving image or a still image. In the case of moving image capturing, a still image may be cut out. The image is also associated with the imaging time and stored in a memory (for example, the inside of the imaging unit or the memory of the mixing degree evaluation system 100). The imaging unit 31 may image the dust periodically or at a predetermined timing, for example, a timing at which the degree of mixing is determined.
Moreover, although the imaging part 31 is one in the present Example 1, it is not restrict | limited to this, Multiple imaging parts may be installed. The stirring areas 15 may be divided and imaged by a plurality of imaging units (partly overlapping may be imaged), and each of a plurality of imaging units may be imaged from different angles. The images captured by the plurality of imaging units may be processed by, for example, a combining processing unit (not shown) such that the entire stirring area becomes one image. An image captured by the imaging unit 31 according to the first embodiment is an RGB image. In addition, it is not limited to a RGB image, For example, the surface image by an infrared camera may be sufficient.

The laser distance meter 32 detects the distance to the accumulated dust D1 in the dust pit 1. The laser range finder 32 is fixedly or movably installed at the upper part in the waste pit 1, for example, movably installed in the crane traveling direction in the crane gutter 21 and may move together with the crane 22, and moves separately It may be installed near the crane 22. When fixedly installed vertically downward, the laser range finder 32 may be configured to be adjustable in angle so that the detection angle can be changed and detected.
In the first embodiment, it is fixed to the crane gutter 21, and the distance to the lower garbage (measurement point P) can be detected by changing the detection angle to the left and right in FIG. By moving the crane gutter 21 in the direction perpendicular to the paper surface of FIG. 1, the distance to the dust in the entire stirring area 15 can be detected.
The coordinates (X-Y plane coordinates) of the measurement point P can be calculated based on the position of the laser range finder 32 and the detection angle (measurement point angle θ). The position of the laser range finder 32 can be derived from the position coordinates of the crane gutter 21 and the relative position between the crane gutter 21 and the laser range finder 32.
The coordinates of the measurement point P, the detected distance A, the measurement point angle θ, and the detection time are linked and stored in a memory (for example, a memory in the laser range finder or in the mixing degree evaluation system 1).

The imaging time of the imaging unit 31 and the distance detection time of the laser range finder 32 may not be completely coincident with each other. The configuration may be such that an image substantially corresponding to the detection time of the laser range finder 32 is captured. When at least one or both are being processed (distance detection, imaging), it is preferable that stirring by a crane is not performed.
Image capturing and distance detection may be performed at a predetermined timing at which the crane is automatically operated for stirring, or at a timing at which the refuse is put into the incinerator 40.

The measurement point dust height calculation unit 101 calculates the measurement point dust deposition height Z based on the distance A, the measurement point angle θ, and the reference distance B from the laser distance meter 32 to the bottom surface of the dust pit 1. The reference distance B is the distance from the laser distance meter 32 to the vertically downward floor (here, the floor of the stirring area 15).
Z = B−A × cos θ (1)

The three-dimensional dust height calculation unit 102 calculates three-dimensional height information W of the dust based on the plane coordinates (X-Y) in the dust pit 1 and the information of the measurement point dust deposition height Z.
The three-dimensional height information W may include information on imaging time and detection time.
The three-dimensional dust height calculation unit 102 calculates three-dimensional height information W at a predetermined timing.
In the present embodiment, the “predetermined timing” may be a fixed interval, a fixed time, or a timing based on an instruction operation of a worker.

  The image conversion unit 111 sets the image captured by the imaging unit 31 to the sky viewpoint image based on the installation information of the imaging unit 31 (for example, the position coordinates installed in the dust pit, the imaging angle, the angle of view of the lens, etc.) Convert to In the first embodiment, since the imaging unit 31 images the dust in the stirring area 15 at a predetermined angle from above obliquely, the captured image is converted into the sky viewpoint image.

Based on the three-dimensional height information W of the garbage, the image correction unit 112 corrects all the areas of the above-described sky viewpoint image so as to be on the same height plane. Let the image obtained here be a correction image.
Thereby, the planar image of the same height is obtained.

The binarization processing unit 113 converts a corrected image, which is an RGB image, into an HSI image, and tone-converts the image with one of hue H, saturation S, and luminance I to obtain a tone-ized image. Next, the binarization processing unit 113 binarizes this toned image with a predetermined threshold to obtain a binarized image. Here, the “predetermined threshold value” may be, for example, an average value of elements (hue H, saturation S, luminance I) when gradation is performed in the entire pit. In the first embodiment, gradation is performed according to the luminance, and the threshold uses the average of the luminance.
Note that as another embodiment, “gradation” is not limited to hue, saturation, and luminance, and gradation may be performed using any of RGB. In the case where the imaging unit is an infrared camera, the binarization processing unit 113 may perform gradation processing using, for example, a wavelength, a frequency, or the like after performing the image conversion processing and the image correction processing.

The mixing degree evaluation unit 130 evaluates the mixing degree of dust using the binarized image. In the first embodiment, the mixing degree evaluation unit 130 has a variation evaluation unit 131.
The variation evaluation unit 131 divides the binarized image into two or more evaluation areas having a plurality of divided areas, and calculates an extraction area of a bright part or a dark part of each divided area.
For example, it is presumed that the bright part is an image with many waste bags as it is, the stirring is insufficient, and the dark part is an image in a state where the waste bags are crushed, and the mixing is sufficient.
Further, the variation evaluation unit 131 calculates the variance σ ² of the extraction area of the bright part or the dark part of the divided area with respect to the entire evaluation area. Although the method of counting the number of pixels is used in the area calculation in the present embodiment, the present invention is not particularly limited thereto, and the area of the bright portion or the dark portion may be calculated by another method.
Along with or instead of the calculation of the variance, the variance evaluation unit 131 can calculate the variance σ ² of the extraction area of the bright part or the dark part of the divided area with respect to each evaluation area.
Further, the variation evaluation unit 131 can calculate the standard deviation σ and / or the coefficient of variation (CV) together with or instead of the variance σ ² .

The variation evaluation unit 131 evaluates the degree of mixing of the dust based on the variation. In the present embodiment, the variation evaluating unit 131 determines whether or not the variance σ ² exceeds a predetermined value. The predetermined value is set in advance, and the setting can be changed for each waste pit (locality, weather). The predetermined value may be set based on experiments and long-term driving rule of thumb.
In the variation evaluation unit 131, as the variance σ ² exceeds the predetermined value and becomes a large value, it indicates that various dusts exist in the area, that is, the stirring is sufficiently performed.

(Processing flow)
In step S1, the imaging unit 31 captures an image in the dust pit. The laser distance meter 32 measures the distance A to the dust (measurement point P) in the dust pit.
In step S2, the dust accumulation height Z at each measurement point is calculated from the following equation.
Z = B−A × cos θ (1)
Next, the three-dimensional height information W of the dust is calculated based on the plane coordinates (XY) in the dust pit 1 and the information of the dust accumulation height Z.

In step S3, the sky viewpoint image is created, and based on the three-dimensional height information W, a correction image is created so that all the areas of the sky viewpoint image are on the same height plane.
In step S4, the corrected image which is an RGB image is subjected to HIS conversion of hue H, saturation S, and luminance I to create a luminance image based on the luminance. The luminance image is gradationized with 256 gradations. FIG. 2A shows an example of a toned image in a dust pit. Here, the number of pixels corresponds to the actual area (agitation area), and the brightness of light and dark is indicated for each pixel.
Note that as another embodiment, “gradation” is not limited to hue, saturation, and luminance, and gradation may be performed using any of RGB. In the case where the imaging unit is an infrared camera, the binarization processing unit 113 may perform gradation processing using, for example, a wavelength, a frequency, or the like after performing the image conversion processing and the image correction processing.

In step S5, the average luminance for all pixels is calculated and set as a threshold.
Average brightness = (brightness 0 × the number of pixels + brightness 1 × the number of pixels + ·· brightness 255 × the number of pixels) / total number of pixels (2)

  In step S6, pixels brighter than the threshold are extracted and binarized. As another example, dark pixels may be extracted and binarized. An example of the binarized image binarized is shown in FIG. 2B.

In step S7, all pixels are fractionated in the evaluation area. Each evaluation area is further divided into divided areas. The number of extracted pixels (area) is calculated for each of the divided areas. Here, the extracted pixel is a bright portion (white). FIG. 2C shows an example in which all pixels are divided in the evaluation area. A thick frame is an evaluation area, and a broken line indicates a pixel of the minimum unit. The “minimum unit pixel” may be one or plural. FIG. 2D shows an example of the addresses of the evaluation areas Sa to Si. FIG. 2E shows an example of the addresses of the divided areas Sa1 to Sa9 of the evaluation area Sa. Addresses are similarly set for the other evaluation areas Sb to Si. In the present embodiment, the evaluation area is the same as the crane address of the stirring area 15.
FIG. 2F shows the number of extracted pixels of each of the divided areas Sa1 to Sa9 of the evaluation area Sa.
Sa1 = 5
Sa2 = 7
Sa3 = 4
Sa4 = 8
Sa5 = 5
Sa6 = 2
Sa7 = 8
Sa8 = 5
Sa9 = 5
The number of extracted pixels (corresponding to the extraction area) is similarly calculated for the other evaluation areas Sb to Si.

Ｓａｉは各分割エリアの抽出面積、Ｓａ＿ａｖｅは評価エリアＳａにおける各分割エリアの抽出面積の平均である。ｎは評価エリア内の画素数であり、９×９である。
他の評価エリアＳｂ〜Ｓｉにおいても同様に算出する。
なお、別実施形態として標準偏差、変動係数を算出してもよい。変動係数ＣＶは式（４）で求めることができる。

σは標準偏差である。 In step S8-1, the variance σ ^{2 of the} extraction area is calculated for each evaluation area.

Sai is the extraction area of each divided area, and Sa_ave is the average of the extraction area of each divided area in the evaluation area Sa. n is the number of pixels in the evaluation area, which is 9 × 9.
The same calculation is performed for the other evaluation areas Sb to Si.
In addition, you may calculate a standard deviation and a coefficient of variation as another embodiment. The coefficient of variation CV can be obtained by equation (4).

σ is a standard deviation.

Ｓａｉは各分割エリアの抽出面積、Ｓｏ＿ａｖｅは全評価エリア（ピット全体）Ｓａ〜Ｓｉにおける各分割エリアの抽出面積の平均である。ｎは評価エリア内の画素数であり、９×９である。
他の評価エリアＳｂ〜Ｓｉにおいても同様に算出する。
なお、別実施形態として、分散に代わり、標準偏差、変動係数を算出してもよい。 In step S8-2, the variance σ ² of the extraction area of each evaluation area is calculated with respect to all the evaluation areas.

Sai is the extraction area of each divided area, and So_ave is the average of the extraction area of each divided area in all evaluation areas (entire pits) Sa to Si. n is the number of pixels in the evaluation area, which is 9 × 9.
The same calculation is performed for the other evaluation areas Sb to Si.
As another embodiment, a standard deviation and a variation coefficient may be calculated instead of the variance.

In step S9, the degree of mixing is evaluated. As the variance σ ² exceeds a predetermined value and becomes a large value, it indicates that various dusts are present in the area, that is, the agitation is sufficiently performed.
The degree of mixing in each evaluation area can be evaluated at the time of dispersion for each evaluation area by step S8-1, and the degree of mixing in the evaluation area compared to the entire waste pit can be evaluated at the time of dispersion for all evaluation areas by step S8-2. it can.

  In step S10, the evaluation result (numerical mixing degree) in step S9 is output. The mixing degree is sent from the mixing degree evaluation system 100 to the storage unit 501 of the waste quality estimation system 500 and stored therein.

(Another configuration example of the first embodiment)
Another configuration of the mixing degree evaluation system 100 according to the first embodiment will be described with reference to FIGS. 4B and 4C. The same reference numerals as those in the first embodiment have the same functions, and thus the description thereof will be omitted and different features will be described.

4B has the following configuration. The degree-of-mixing evaluation unit 130 includes an area average evaluation unit 132.
In step S5, the area average evaluation unit 132 applies the discriminant analysis method to each divided area for each divided area, or performs binarization according to the dynamic threshold method to label a bright part (or a dark part). Etc. to extract each particle.
The area average evaluation unit 132 calculates each particle area, and calculates an area average of each particle.
The area average evaluation unit 132 determines whether or not the area average of each particle is smaller than a predetermined value.
In the determination result of the area average evaluation unit 132, the smaller the area average is, the smaller the predetermined value is, the more the rubbing of garbage progresses, that is, the greater the stirring.
Similar to Example 1, the evaluation can be performed in units of evaluation area.

Another configuration of FIG. 4B may be included in the mixing degree evaluation system 100 together with the first embodiment of FIG. 4A.
The degree of mixing may be evaluated by the respective judgment or comprehensive judgment of the variation evaluation unit 131 and the area average evaluation unit 132.

  The alternative configuration of FIG. 4C has the following configuration. The mixing degree evaluation unit 130 includes an area ratio evaluation unit 133.

The area ratio evaluation unit 133 calculates the area ratio (= dark part / bright part) of the light part and the dark part of each evaluation area.
The area ratio evaluation unit 133 determines whether the area ratio is larger than a predetermined value (for example, less than 1).
In the judgment result of the area ratio evaluation unit 133, it is indicated that the more the area ratio is larger than the predetermined value, the more the waste is being broken, that is, the stirring is sufficiently performed.
Similar to Example 1, evaluation can be performed in units of evaluation areas.

The alternative configuration of FIG. 4C may be included in the degree-of-mixture evaluation system 100 along with the above-described embodiment 1 of FIG. 4A and / or the alternative configuration of FIG. 4B.
The degree of mixing may be evaluated based on each judgment or comprehensive judgment of the variation evaluation unit 131, the area average evaluation unit 132, and the area ratio evaluation unit 133.

(Example 2 of input data: Configuration for determining the area ratio of absorbance)
The area ratio calculation system 200 for absorbance shown in FIG. 6 includes a spectrum binarized image generation unit 201 and an area ratio calculation unit 202.

The imaging unit 31 is a visible camera and a spectrum camera.
The spectrum binarized image generation unit 201 captures the entire dust pit 1 with a spectrum camera, and generates a binarized image for each divided wavelength. FIG. 8A shows an example of a binarized image for each wavelength. FIG. 8A only illustrates three types of binarized images, and may be divided into three or more types.

  The area ratio calculation unit 202 divides the dust pits into evaluation areas, calculates the area above (or below) the threshold for each wavelength, and calculates the ratio to the evaluation area area. FIG. 8B shows an example of a binarized image for each wavelength division. FIG. 8C shows an example of the area ratio for each wavelength in which each evaluation area is divided. The area ratio may be expressed as a percentage. The area ratio is sent from the area ratio calculation system 200 to the storage unit 501 of the dust quality estimation system 500 and stored.

(Example 3 of Input Data: Configuration for Determining Area Ratio of Absorbance)
The absorbance average value calculation system 300 shown in FIG. 7 has an absorbance average value calculation unit 301 that calculates the average value of the absorbance in each evaluation area for each wavelength divided from the spectrum image data. FIG. 8D shows an example of the average absorbance value for each divided wavelength of each evaluation area. The absorbance average value is sent from the absorbance average value calculation system 300 to the storage unit 501 of the dust quality estimation system 500 and stored.

(Generation of input waste quality estimation model)
The input data creation unit 502 acquires data on the degree of mixing, the area ratio of absorbance, and the average absorbance value in each evaluation area, and creates a database as original data of the input data.
FIG. 9A shows an example of a database of input data stored in the storage unit 501.
The input data creation unit 502 mixes the waste, the area ratio of absorbance, the average absorbance value, the input waste weight, the input waste when the waste in the evaluation area Sa (corresponding to the mixing address) is input to the input hopper of the incinerator. The time data is stored in the storage unit 501. When wastes in other evaluation areas are thrown in, data is configured in the same way. FIG. 9B shows an example of input data of input waste.

  In the first embodiment, the mixing degree evaluation system 100, the area ratio calculation system 200 for absorbance, and the average absorbance calculation system 300 have a partial function of the input data creation unit 502. As another embodiment, these components may be incorporated into the dust quality estimation system 500 as the input data creation unit 502.

The input waste quality estimation model generation unit 503 generates an input waste quality estimation model for estimating the waste quality by the intelligent information processing technology using the input data and the waste quality teacher data. The dust quality teacher data may be stored in advance in the storage unit 501, or may be acquired from an external device when the input dust quality estimation model generation unit 503 functions.
In the first embodiment, the waste quality teacher data includes the concentration of the waste composition (water, carbon, hydrogen, hydrogen chloride (HCl) and sulfur dioxide (SO ₂ ), etc. in the exhaust gas), and the calorific value.
In another embodiment, the waste quality estimation system 500 may have a waste quality teacher data creation unit that creates waste quality teacher data.
The waste composition concentration can be obtained, for example, by analyzing the exhaust gas with an analyzer corresponding to various components such as an exhaust gas analyzer. Further, other components (for example, carbon dioxide) may be calculated from predetermined components (for example, oxygen and moisture).
Further, the calorific value may be calculated using the method of Japanese Patent No. 599 6762. The garbage quality teacher data creation unit may include a heat generation amount calculation unit that estimates a heat generation amount.
FIG. 10 shows an example of garbage quality teacher data.

(Calculation of calorific value by the method of Japanese Patent No. 5996762)
The first method is a procedure of R1 to R8.
(R1) Measure the concentration of oxygen and moisture in the exhaust gas.
(R2) The concentration of carbon dioxide in the exhaust gas is estimated based on the following equation from the measured component concentrations of oxygen and moisture.
[CO ₂ ] = Ro × (100- [H ₂ O]) / 100- [O ₂ ]
Here, [] indicates percentage concentration concentration, and Ro indicates a coefficient set by subtracting the amount of oxygen component taken into the ash from the oxygen concentration in the atmosphere.
(R3) The nitrogen concentration in the exhaust gas is calculated using the oxygen concentration, the water concentration and the carbon dioxide concentration.
(R4) Based on the calculated nitrogen concentration, a conversion factor for the nitrogen concentration in the combustion air is calculated, and the conversion component concentration of the oxygen, carbon dioxide and water multiplied by the conversion factor is calculated.
(R5) The oxygen consumption per unit supply of the combustion air used for the combustion process is calculated from the converted component concentrations of oxygen, carbon dioxide and water.
(R6) From the calculated oxygen consumption, the calorific value related to carbon dioxide and moisture generated in the combustion process per unit supply of combustion air, contained in the waste water and the waste generated Calculate the amount of latent heat from the total amount of moisture.
(R7) Calculate the amount of treated waste per unit supply of combustion air from the amount of supply of the burnt-treated waste.
(R8) From the calculated calorific value, the latent heat, and the waste amount, the estimated calorific value A per treated waste amount is calculated.

The second method is the procedure of S1 to S7.
(S1) The component concentrations of oxygen, carbon dioxide and moisture in the exhaust gas are measured.
(S2) The nitrogen concentration in the exhaust gas is calculated from the measured component concentrations.
(S3) Based on the calculated nitrogen concentration, a conversion factor for the nitrogen concentration in the combustion air is calculated, and the conversion component concentration of the oxygen, carbon dioxide and water multiplied by the conversion factor is calculated.
(S4) The oxygen consumption per unit supply amount of the combustion air used for the combustion process is calculated from the converted component concentrations of oxygen, carbon dioxide and moisture.
(S5) From the calculated oxygen consumption, the calorific value related to carbon dioxide and moisture generated in the combustion process per unit supply of combustion air, contained in the amount of generated water and the waste Calculate the amount of latent heat from the total amount of moisture.
(S6) The amount of waste treated per unit supply of combustion air is calculated from the amount of supply of waste subjected to the combustion treatment.
(S7) From the calculated calorific value, latent heat quantity and waste quantity, an estimated calorific value B per treated waste quantity is calculated.

(Calculate calorific value by other methods)
The third method is the procedure of A1 to A8.
(A1) The exhaust gas flow rate from the combustion furnace is measured.
(A2) The component concentrations of oxygen, carbon dioxide and moisture in the exhaust gas from the combustion furnace are measured.
(A3) The nitrogen concentration [N ₂ ] in the exhaust gas is calculated from the measured component concentrations.
(A4) Based on the calculated nitrogen concentration [N ₂ ], a conversion factor for the nitrogen concentration An in the atmosphere is calculated, and the conversion component concentrations Go, Gd and Gw of the oxygen, carbon dioxide and water multiplied by the conversion factor Is calculated.
(A5) Based on the component concentration Gd of carbon dioxide converted, the amount of carbon in the combustible substance per combustible substance unit supply amount is calculated.
(A6) The oxygen component concentration Go and the carbon dioxide component concentration Gd converted from the oxygen concentration Ao in the atmosphere are subtracted, and the amount of hydrogen in the combustible material consumed in the combustion process and the water generated by the combustion process Calculate the amount.
(A7) Based on the component concentration Gw of the converted water and the calculated amount of water, the amount of evaporation of water in the material to be burned is calculated.
(A8) Combustion-treated combustible based on the calorific value of reaction heat of carbon and hydrogen in the combustible generated in the combustion treatment and the latent heat of water using the calculated carbon, hydrogen, and moisture The calculated calorific value A per unit supply amount is calculated.

(Calculate calorific value by other methods)
The fourth method is a procedure of B1 to B9.
(B1) The exhaust gas flow rate from the combustion furnace is measured.
(B2) The concentration of oxygen and water in the exhaust gas is measured.
(B3) From the measured component concentrations [O ₂ ] and [H ₂ O] of oxygen and water, the carbon dioxide concentration [CO ₂ ] in the exhaust gas is calculated based on the following equation 1.
[CO ₂ ] = Ro × (100− [H ₂ O]) / 100− [O ₂ ]
Here, [] indicates percentage concentration concentration, and Ro indicates a coefficient set by subtracting the amount of oxygen component taken into the ash from the oxygen concentration in the atmosphere.
(B4) The nitrogen concentration [N2] in the exhaust gas is calculated from the measured oxygen concentration [O ₂ ], the water concentration [H ₂ O], and the calculated carbon dioxide concentration [CO ₂ ].
(B5) A conversion factor for nitrogen concentration An in the atmosphere is calculated based on the calculated nitrogen concentration [N2], and the conversion component concentrations Go, Gd and Gw of the oxygen, carbon dioxide and water multiplied by the conversion factor are calculated. calculate.
(B6) Based on the converted component concentration Gd of carbon dioxide, a carbon amount Ec in the combusted material per combustible unit supply amount is calculated.
(B7) The oxygen component concentration Go and the carbon dioxide component concentration Gd converted from the oxygen concentration Ao in the atmosphere are subtracted, and the amount of hydrogen Eh in the material to be burned consumed in the combustion treatment is generated by the combustion treatment The amount of water Cw is calculated.
(B8) Based on the converted water component concentration Gw and the calculated water content, the amount of water evaporation in the combusted material is calculated.
(B9) Using the calculated amount of carbon, amount of hydrogen, and amount of water, the combusted material subjected to combustion processing based on the calorific value of the reaction heat of carbon and hydrogen in the combusted material generated by the combustion processing and the amount of latent heat of water The calculated calorific value B per unit supply amount is calculated.
In the above, the amount of mixed air and the amount of oxygen in the substance to be burned may be calculated.
Based on the following formula, the amount of mixed air other than the combustion air supplied to the combustion furnace is calculated.
(Amount of mixed air) = (exhaust gas flow rate × [N ₂ ]-combustion air supply amount × An) / An
The amount of oxygen in the combustible substance may be calculated based on the following equation.
(Oxygen amount in the combusted material) = (exhaust gas flow rate × [O ₂ ]) − (combustion air supply amount × Aa−exhaust gas flow rate × [CO ₂ ] −moisture amount Cw) − (mixed air amount × An)

  In the first embodiment, the input waste quality estimation model generation unit 503 generates waste quality teacher data and input data based on the amount of time delay from the start of loading the waste into the inlet of the combustion apparatus and the start of combustion. The input waste quality estimation model is generated by temporally correlating the For example, in the case of a stoker-type incinerator, the amount of time delay is, for example, 30 to 40 minutes.

As another embodiment, “temporal correspondence between garbage quality teacher data and input data” may be performed using the methods (1) to (5).
(1) Store in the storage unit input data and weight data of waste to be fed into the feeding hopper.
(2) The deposition state of the input waste is fractionated, and the weight of each fractionated sediment and the input data are stored in the storage unit. For example, the weights of the first fractionated sediment G1, the second fractionated sediment G2, and the third fractionated sediment G3 are stored.
(3) Every time waste is supplied to the incinerator of the combustion apparatus, the weight of the supplied waste (for example, 50 kg in-furnace supply) is subtracted from the remaining weight of the lowermost first fractionated accumulated waste in the charging hopper. The weight of the supplied garbage may be supplied in a fixed amount (for example, 50 kg) or may be measured each time the weight of the supplied garbage is supplied.
(4) Fraction deposits with zero remaining weight are removed from the record. The weight that could not be pulled out is subtracted from the remaining weight of the nth fraction-deposited debris at the bottom.
(5) Correspond the input data of the supplied waste and the waste quality teacher data at this time.
FIG. 11 shows the accumulation state and the feeding state of the input waste (garbage TAG).

(Another embodiment)
Although three types are used for input data in the first embodiment, the present invention is not limited to this, and one or two or more types out of an area ratio, an average value of absorbance, and a mixing degree may be used.
In addition, as input data, one or more types of information may be further included among the waste in-car, the date, and the weather.
In the first embodiment, the waste quality teacher data is the concentration of the waste composition (water, carbon, hydrogen, hydrogen chloride (HCl) and sulfur dioxide (SO ₂ ), etc. in the exhaust gas), the calorific value, etc. However, it may be one or more of the above.
In the first embodiment, input data is created by the mixture degree evaluation system 100, the absorbance area ratio calculation system 200, and the absorbance average value calculation system 300. However, the present invention is not limited to this, and the input data creation unit of the garbage quality estimation system 500 is used. 502 may have these functions.

(Crane operation control system)
The crane operation control system 600 inputs the input data of the input waste into the input waste quality estimation model, and stores the input waste data of the input waste into the input waste quality estimation model. A waste quality estimation unit 602 that estimates the quality, and an automatic operation control unit 603 that automatically controls the crane according to the estimated waste quality output from the waste quality estimation model.
The above-mentioned automatic operation control part 603 may control a crane so that refuse quality may become uniform in each evaluation area.
The above-mentioned automatic operation control part 603 may control a crane to throw in refuse with a high calorific value, when calorific value of a combustion device is lower than an expected value.
The above-mentioned automatic operation control part 603 may control a crane to throw in refuse with a low calorific value, when calorific value of a combustion device is higher than an expected value.

  The crane operation control system 600 is configured not only to automatic operation control but also to manual operation control. The operator can operate the crane according to the distribution of the estimated waste quality of each evaluation area displayed on the monitor.

  The input data of the input waste is obtained by the functions of the mixing degree evaluation system 100, the area ratio calculation system 200 of absorbance, the average absorbance calculation system 300, and the dust quality estimation system 500 described above.

  Each component of mixing degree evaluation system 100, area ratio calculation system 200 of absorbance, absorbance average value calculation system 300, waste quality estimation system 500, and crane operation control system 600 is an information processing having a memory, a processor, and a software program. It can be configured by a device (for example, a computer), a dedicated circuit, firmware, or the like.

DESCRIPTION OF SYMBOLS 1 Garbage pit 15 Stirring area 21 Crane garter 22 Crane 23 Bucket 31 Image pick-up part 32 Laser distance meter 100 Mixture evaluation system 101 Measurement point dust height calculation part 102 Three-dimensional waste height calculation part 111 Image conversion part 112 Image correction part 113 Binarization processing unit 130 Mixing degree evaluation unit 131 Variation evaluation unit 132 Area average evaluation unit 133 Area rate evaluation unit 200 Area rate calculation system 201 Spectrum binarized image generation unit 202 Area rate calculation unit 300 Absorbance average value calculation system 301 Absorbance Average value calculation unit 500 Waste quality estimation system 501 Storage unit 502 Input data creation unit 503 Input waste quality estimation model generation unit 600 Crane operation control system 601 Model storage unit 602 Garbage quality estimation unit 603 Automatic operation control unit

Claims (3)

An input waste that generates an input waste quality estimation model that estimates waste quality by intelligent information processing technology using input data of input waste created from image data obtained by imaging the surface of the waste pit and waste quality teacher data A quality estimation model generation unit,
The input waste quality estimation model generation unit
Based on a preset amount of time lag from when dust is introduced into the inlet of the combustion apparatus until combustion is started , the garbage quality teacher data and the input data are correlated in time. generating the charged waste matter estimation model, or,
The waste quality teacher data and the input data are calculated by calculating the timing at which a specific input waste is supplied to the combustion furnace from the amount of waste input to the inlet and the amount of waste supplied to the combustion furnace. A waste quality estimation system for generating the input waste quality estimation model by performing temporal correspondence of   The waste quality estimation system according to claim 1, further comprising an input data creation unit that creates input data of input waste from image data obtained by capturing the surface of the waste pit. A model storage unit for storing an input waste quality estimation model generated by the waste quality estimation system according to claim 1 or 2;
A waste quality estimation unit that inputs waste input data to the input waste quality estimation model and estimates waste quality,
A crane operation control system having an automatic operation control unit that automatically controls a crane according to the estimated waste quality output from the input waste quality estimation model.

Images