JP2005090774A - Device for estimating supply amount of garbage for garbage incinerator - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for estimating a supply amount of garbage, capable of properly controlling the combustion of garbage by grasping a weight of garbage supplied into the incinerator and the kind of garbage in advance. <P>SOLUTION: This device comprises a plurality of photographing means 6 for color-photographing the whole upper face of a surface layer part of the garbage in a hopper of the garbage incinerator, a garbage volume estimating means 31 for deriving a three-dimensional coordinates of the upper face of the surface layer part from a plurality of images taken by the photographing means 6 and estimating the volume of garbage in the hopper, a weight measuring means 7 for measuring a weight of the garbage charged into the hopper, a garbage supply weight estimating means 32 for estimating the weight of garbage supplied into the incinerator from the hopper on the basis of the estimated volume of garbage and the measured charging weight, and a heat generation estimating means 33 for estimating the kind of garbage on the basis of the shape and color of the garbage of the surface layer part from the color images taken by the photographing means 6, and estimating the heat generation of the garbage supplied into the incinerator from the hopper on the basis of the estimated kind and the weight of the garbage supplied of the surface layer garbage. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to control of a waste incinerator, and more particularly, to a waste supply amount estimation device that estimates the supply weight and supply heat value of dust supplied from a hopper of a waste incinerator into the waste incinerator.

  If the amount of waste supplied to the waste incinerator or the amount of heat generated is unstable or suddenly fluctuates, the waste combustion in the waste incinerator becomes unstable, which may cause fluctuations in the furnace temperature or local fluctuations. Generation of a high temperature region, generation (or increase) of CO (carbon monoxide), HC (hydrocarbon), NOx (nitrogen oxide), dioxins, etc., and further generation of unburned dust.

  Incinerators that incinerate municipal waste, industrial waste, etc., such as stoker-type waste incinerators, have a hopper laid as a receiving port for carrying waste into the furnace. Garbage thrown into the hopper from above and fed into and accumulated in the hopper is fed into the garbage incinerator from the lower part of the hopper by a pusher type or screw conveyor type feeding means. In a stalker-type waste incinerator, waste from the hopper is supplied to the upstream side of the stalker, which is the garbage transporting means laid at the bottom of the furnace, dried at the upstream part of the stalker, and then burned at the midstream part. The incinerated ash after-treated in is discharged to the outside of the furnace through the outlet from the downstream side of the stoker. Here, in order to stabilize the combustion of dust in the furnace, it is necessary to appropriately control the amount of dust supplied to the furnace and the amount of heat generated by the dust in order to enable stable combustion control.

  Garbage thrown into the hopper is compacted by its own weight, and the specific gravity of the garbage deposited on the bottom increases, and the specific gravity fluctuates depending on the type of garbage and the weight of garbage in the hopper. Even if dust is supplied into the furnace in a certain volume, there is a risk that the supplied weight may fluctuate, and it is necessary to appropriately control the weight of dust supplied into the furnace. In order to solve such problems, for example, in Patent Document 1 below, an image of the upper surface of dust in the hopper is captured by a television camera, and the dust volume in the hopper is calculated from the captured image. Disclosed is a technique for calculating the supply weight of dust per unit time supplied into the furnace by the supply means based on the volume and the weight of the waste put into the hopper, and controlling the supply means based on the calculation result. Has been.

Furthermore, as a technique for detecting in advance the amount of heat generated by the dust supplied into the furnace and reflecting it in combustion control, for example, there is one disclosed in Patent Document 2 below. In Patent Document 2, an infrared detector for detecting the emission temperature of dust in the dust ignition start region in the upstream portion of the stoker is provided in the furnace, and the dust quality for estimating the lower heating value of the dust based on the average value of the detected temperatures. A determination method has been proposed.
JP 2001-355819 A Japanese Patent Laid-Open No. 10-185157 Ito et al., “Study on surface shape measurement of waste in hopper using stereo images”, Environmental Measurement System Control Society Journal “EICA”, Vol. 6, No. 1, pp. 37-43, 2001 Ito et al., “Prediction of Waste Supply in Municipal Waste Incinerators Using Stereo Images”, Proceedings of the Japan Society of Waste Science Research Presentation, pp. 567-569, 2001

  In the above-mentioned patent document 1, the volume of dust in the hopper is calculated using an image of the upper surface of the dust in the hopper, and the supply of dust per unit time supplied into the furnace together with the weight of the dust thrown into the hopper. Although a technique for calculating the weight and controlling the supply unit based on the calculation result is disclosed, it is necessary to calculate the dust volume in the hopper with a predetermined accuracy in order to improve the control accuracy. Therefore, although there are reports that two or more TV cameras to be used can be provided and the volume can be calculated by three-dimensional analysis of the shape of the upper surface of the dust, further improvements for practical use have also been pointed out (for example, the above non-disclosures). (See Patent Documents 1 and 2).

  Patent Document 2 discloses a method for estimating the quality of garbage that has already been put into the furnace and started to burn, and in order to appropriately deal with sudden fluctuations in garbage quality, It is desirable to determine the garbage quality in advance before being thrown in.

  The present invention has been made in view of the above-described problems, and its purpose is to accurately calculate the volume of dust in the hopper so as to accurately grasp the weight of dust supplied into the furnace, and to control dust combustion. It is an object of the present invention to provide a dust supply amount estimation device that can easily and reliably achieve the above. It is another object of the present invention to provide a waste supply amount estimation device that grasps in advance the quality of waste to be supplied into a furnace and that can perform waste combustion control more accurately. Then, by providing a waste incinerator equipped with these waste supply amount estimation devices, stable waste combustion is possible, and the generation of CO, HC, NOx, dioxins and the like is suppressed.

  In order to achieve this object, the first characteristic configuration of the dust supply amount estimation apparatus according to the present invention is a three or more imaging capable of imaging the entire upper surface of the surface layer portion of the dust supplied and deposited in the hopper of the garbage incinerator. And three-dimensional coordinates at a plurality of positions on the upper surface of the surface layer portion are derived from three or more images picked up by the three or more image pickup means, and the volume of dust in the hopper is estimated based on the three-dimensional coordinates. Based on the waste volume estimating means, the weight measuring means for measuring the input weight of the dust to be put into the hopper, the dust volume estimated by the dust volume estimating means, and the input weight measured by the weight measuring means Garbage supply weight estimation means for estimating the supply weight of garbage supplied from the hopper into the garbage incinerator, wherein the waste volume estimation means is 3 or more captured by the three or more imaging means. Painting Common imaging point extraction means for extracting a plurality of common imaging points obtained by imaging the same position on the upper surface of the surface layer portion for a plurality of sets of two or more arbitrary images selected from the above, and the common imaging point extraction means The three-dimensional coordinate calculation means for calculating the three-dimensional coordinates of the plurality of common imaging points with respect to the plurality of sets of the plurality of images extracted, and adjustment of the calculated three-dimensional coordinates between the plurality of sets of the plurality of images. Three-dimensional coordinate adjusting means for performing, and volume calculating means for calculating the volume of the dust based on the wall surface shape of the hopper, using the adjusted three-dimensional coordinates as three-dimensional coordinates at a plurality of positions on the upper surface of the surface layer portion. It is in the point to have.

  According to the first characteristic configuration, it is possible to capture three or more images of the upper surface portion of the dust using three or more imaging units, and two or more arbitrary images selected from the three or more images can be captured. Multiple sets can be extracted. Specifically, for example, three or more combinations of two images can be extracted. In general, when the same subject is photographed by imaging means arranged at two different positions, a common position (common imaging point) imaged on the two screens is imaged at different positions on the screen. Based on the information, the three-dimensional position of each common imaging point can be specified. Therefore, according to the dust volume estimation means of this feature configuration, the common imaging point extraction means extracts a plurality of common imaging points for any two screens of three or more screens on the upper surface of the dust, and the three-dimensional coordinates. By calculating the three-dimensional coordinates of the plurality of common imaging points by the calculating means, it is possible to roughly grasp the three-dimensional shape of the upper surface portion of the dust. Further, the three-dimensional coordinate adjustment means can correct the three-dimensional coordinate calculation error due to the error at the time of extracting the common imaging point to the three-dimensional shape information on the upper surface of the dust obtained from each set of images. Since common imaging points that could not be extracted by combination may be extracted by another combination, common imaging points can be mutually complemented by multiple sets of images, and more accurate and precise three-dimensional shape can be grasped It becomes. As a result, the volume calculation means can calculate the volume with higher accuracy than the volume that can be calculated based on the two images. Therefore, by using the highly accurate calculated garbage volume, the garbage supply weight estimation means increases the garbage capacity every time the garbage is put into the hopper, and the garbage is carried out from the hopper into the furnace. It is possible to accurately grasp the decrease in the volume of each trash, and from the input weight measured by the weight measuring means, the specific gravity of the trash immediately after the input can be estimated, and the trash in the bottom of the hopper consolidated by the self-weight of the trash Therefore, by multiplying the volume of dust supplied from the hopper into the furnace by the specific gravity, it is possible to estimate the supply weight of the dust supplied from the hopper into the furnace with high accuracy.

  As described above, according to this characteristic configuration, it is possible to easily execute control for keeping the supply weight of the dust supplied into the furnace at a constant value, and as a result, stable combustion control of the dust can be performed.

  In the second feature configuration, in addition to the first feature configuration, the three or more imaging units are trinocular cameras in which relative positional relationships of three imaging units are fixed, and the common imaging A point extraction unit is provided for each of three sets of two images selected from the three images captured by the trinocular camera, or each of the three sets of two images and one set of the three images. Another point is that a plurality of common imaging points obtained by imaging the same position on the upper surface of the surface layer are extracted.

  According to said 2nd characteristic structure, it can replace with a 3 or more image pickup means, and can show | play the same effect as a 1st characteristic structure by using a trinocular camera. In the trinocular camera, since the relative positional relationship between the three imaging units is fixed, the three-dimensional coordinates of the upper surface of the surface layer unit are derived from the two images regardless of the installation location on the hopper. The necessary adjustment work can be prepared in advance, and the same adjustment results can be used in common among a plurality of incinerators. Detailed adjustment work can be omitted.

  The third characteristic configuration includes a plurality of imaging means capable of color imaging the entire upper surface of the surface layer portion of the dust supplied and deposited in the hopper of the garbage incinerator, and a plurality of images captured by the plurality of imaging means. Deriving three-dimensional coordinates at a plurality of positions on the upper surface of the surface layer portion to estimate the volume of the dust in the hopper based on the three-dimensional coordinates, and the input weight of the dust input into the hopper Based on the weight measuring means to measure, the volume of the dust estimated by the dust volume estimating means, and the input weight measured by the weight measuring means, the supply weight of the dust supplied from the hopper into the garbage incinerator is calculated. Estimating dust supply weight, and estimating the dust quality from the shape and color of the surface dust existing in the surface layer from the color image captured by the imaging means, and at least the estimated surface dust A calorific value estimating means for estimating a supply calorific value of dust supplied from the hopper into the garbage incinerator based on the supply quality estimated by the dust quality and the dust supply weight estimating means; It is in the point.

  According to the third characteristic configuration, the dust volume estimating means derives the three-dimensional coordinates from a plurality of images on the upper surface of the dust surface layer, and the dust in the hopper is obtained from the known three-dimensional coordinates of the inner wall surface of the hopper. The waste supply weight estimation means increases the garbage capacity every time the garbage is put into the hopper, and the garbage is carried out from the hopper into the furnace based on the estimated garbage volume. It is possible to accurately grasp the decrease in the volume of each trash, and from the input weight measured by the weight measuring means, it is possible to estimate the specific gravity of the trash immediately after the input, and the trash inside the hopper that has been consolidated by the trash's own weight. Since the specific gravity can also be estimated, by multiplying the volume of dust supplied from the hopper into the furnace by the specific gravity, the supply weight of the dust supplied from the hopper into the furnace can be estimated. Furthermore, the calorific value estimation means estimates the dust quality from the shape and color of the surface dust, so that the distribution of the dust quality of the surface dust can be directly grasped. Assuming this model, the distribution of the garbage quality in the hopper can be grasped, and the amount of heat generated per unit weight of the garbage supplied from the hopper to the furnace can be calculated from the quality of the garbage present at the bottom of the hopper. The supply calorific value of the dust supplied into the furnace from the hopper bottom can be estimated from the supply weight estimated by the dust supply weight estimation means.

  As described above, according to this feature configuration, since the supply weight and supply heat generation amount of the waste to be supplied into the furnace can be estimated in advance, the waste combustion control in the furnace is performed in advance before the sudden change in the waste quality occurs. The control can be changed to appropriate control, and as a result, stable dust combustion control is possible.

  In the fourth feature configuration, in addition to the third feature configuration, three or more imaging units are provided, and the dust volume estimation unit is configured to select from among three or more images captured by the three or more imaging units. Common imaging point extraction means for extracting a plurality of common imaging points obtained by imaging the same position on the upper surface of the surface layer portion and a plurality of sets of arbitrary two or more selected images are extracted by the common imaging point extraction means. Three-dimensional coordinate calculation means for calculating three-dimensional coordinates of the plurality of common imaging points for the plurality of sets of the plurality of images, and three-dimensional adjustment for adjusting the calculated three-dimensional coordinates between the plurality of sets of the plurality of images. Coordinate adjustment means; and volume calculation means for calculating the volume of the dust based on the shape of the wall surface of the hopper, using the adjusted three-dimensional coordinates as three-dimensional coordinates at a plurality of positions on the upper surface of the surface layer portion. In the point.

  According to the fourth characteristic configuration, it is possible to capture three or more images of the upper surface portion of the dust using three or more imaging units, and two or more arbitrary images selected from the three or more images can be captured. Multiple sets can be extracted. Specifically, for example, three or more combinations of two images can be extracted. In general, when the same subject is photographed by imaging means arranged at two different positions, a common position (common imaging point) imaged on the two screens is imaged at different positions on the screen. Based on the information, the three-dimensional position of each common imaging point can be specified. Therefore, according to the dust volume estimation means of this feature configuration, the common imaging point extraction means extracts a plurality of common imaging points for any two screens of three or more screens on the upper surface of the dust, and the three-dimensional coordinates. By calculating the three-dimensional coordinates of the plurality of common imaging points by the calculating means, it is possible to roughly grasp the three-dimensional shape of the upper surface portion of the dust. Further, the three-dimensional coordinate adjustment means can correct the three-dimensional coordinate calculation error due to the error at the time of extracting the common imaging point to the three-dimensional shape information on the upper surface of the dust obtained from each set of images. Since common imaging points that could not be extracted by combination may be extracted by another combination, common imaging points can be complemented by multiple sets of images, and more accurate and precise three-dimensional shape can be grasped It becomes. As a result, the volume calculation means can calculate the volume with higher accuracy than the volume that can be calculated based on the two images. Therefore, by using the dust volume calculated with high accuracy, it is possible to estimate the supply weight and supply heat amount of the dust supplied from the hopper into the furnace with higher accuracy.

  In the fifth feature configuration, the entire upper surface of the surface portion of the dust that is supplied and deposited in the hopper of the waste incinerator can be imaged in color, and distance information of each pixel of the captured image can be output, and three imaging units A trinocular camera whose relative positional relationship is fixed, a dust volume estimating means for estimating the volume of dust in the hopper based on distance information of each pixel output from the trinocular camera, Based on the weight measuring means for measuring the input weight of the dust put into the hopper, the volume of the dust estimated by the dust volume estimating means, and the input weight measured by the weight measuring means, the waste incineration from the hopper Garbage supply weight estimation means for estimating the supply weight of garbage supplied into the furnace, and estimating the quality of dust from the shape and color of the surface layer garbage present in the surface layer part from the color image captured by the trinocular camera, The estimated A calorific value estimating means for estimating the supply calorific value of the dust supplied from the hopper into the garbage incinerator based on the quality of the garbage on the surface layer and the supply weight estimated by the dust supply weight estimating means; In the point which comprises.

  According to the fifth characteristic configuration, the dust volume estimation means derives, for example, the three-dimensional coordinates from the distance information of each pixel on the upper surface of the dust surface layer output from the trinocular camera, and the known hopper The volume of trash in the hopper can be estimated from the three-dimensional coordinates of the inner wall surface. Based on the estimated volume of trash, the trash supply weight estimation means determines the trash volume for each trash thrown into the hopper. It is possible to accurately grasp the increase and the decrease in the garbage capacity every time the garbage is carried out from the hopper into the furnace, and the specific gravity of the garbage immediately after the introduction can be estimated from the input weight measured by the weight measuring means, and Since the specific gravity of the dust at the bottom of the hopper consolidated by the dead weight of the trash can also be estimated, the waste volume supplied from the hopper to the furnace is multiplied by the specific gravity, so that Supply weight can be estimated. Further, the calorific value estimation means estimates the dust quality from the shape and color of the surface dust in the color image captured by the trinocular camera, so that the distribution of the dust quality of the surface dust can be directly grasped. Assuming a model that gradually accumulates and compacts and moves to the bottom of the hopper, it is possible to grasp the distribution of garbage quality in the hopper, and supply the waste from the hopper to the furnace from the quality of garbage present in the hopper bottom. It is possible to determine the amount of heat generated per unit weight of the waste, and to estimate the amount of heat supplied to the furnace from the bottom of the hopper from the supply weight estimated by the dust supply weight estimation means.

  As described above, according to this feature configuration, since the supply weight and supply heat generation amount of the waste to be supplied into the furnace can be estimated in advance, the waste combustion control in the furnace is performed in advance before the sudden change in the waste quality occurs. The control can be changed to appropriate control, and as a result, stable dust combustion control is possible. In addition, since the relative positional relationship of the three imaging units of the trinocular camera is fixed, the three-dimensional coordinates of the upper surface of the surface layer unit are derived from the two images regardless of the installation location on the upper part of the hopper. Necessary adjustment work can be prepared in advance, and the same adjustment result can be used in common among multiple incinerators. There is an advantage that a simple adjustment work can be omitted.

  In the sixth feature configuration, in addition to any one of the third to fifth feature configurations, the heat generation amount estimation means classifies the shape and color of dust in advance into a plurality of dust types, and sets each dust type. The supply heat generation amount is estimated based on the supply weight estimated by the dust supply weight estimation means, using a dust type database in which the relationship between the heat generation amount per unit weight and the database is made into a database.

  According to the sixth feature configuration, the calorific value estimation means extracts the shape and color of dust from the image of the upper surface of the dust surface layer, and collates it with the dust type database, so that the dust quality of the extracted part That is, the calorific value per unit weight can be easily determined, and the distribution of the dust quality of the entire surface layer of the dust can be grasped.

  In the seventh feature configuration, in addition to any one of the feature configurations described above, the dust supply weight estimation means uses a dust compression model that models a compression state due to the weight of dust accumulated in the hopper, and The specific gravity of the trash at the bottom of the hopper is obtained, and the supply weight is estimated from the specific gravity and the supply volume of the trash supplied from the hopper into the trash incinerator.

  According to the seventh characteristic configuration, the garbage supply weight estimation means uses the garbage compression model to estimate the specific gravity of the garbage in the bottom portion of the hopper consolidated by the dead weight of the garbage. The specific gravity of the waste at the bottom of the hopper can be estimated by a simple calculation that properly considers the model that gradually accumulates and moves to the bottom of the hopper, and as a result, the weight of the waste that is fed from the hopper into the furnace. Can be estimated with high accuracy.

  An embodiment of a dust supply amount estimation apparatus (hereinafter referred to as “the present invention apparatus”) of a garbage incinerator according to the present invention will be described with reference to the drawings.

<First Embodiment>
First, as shown schematically in FIG. 1, a refuse incinerator 20 to which the apparatus 1 of the present invention is applied includes a hopper 2 that receives refuse that is an incineration object, and a hopper 2 that is provided above the hopper 2. A crane 3 for throwing trash into the hopper, a pusher 4 for pushing the trash thrown into the hopper 2 into the furnace 21 from the bottom of the hopper 2, and trash thrown into the furnace 21 by the pusher 4. This is a stoker-type garbage incinerator configured to include a stoker-type transport means 5 that incinerates and transports the food while stirring. Further, on the upper part of the hopper 2, a trinocular camera 6 (an example in which a plurality of imaging means is integrated) capable of color imaging of the entire upper surface of the surface layer portion of dust supplied and accumulated in the hopper is provided. Yes. Further, the crane 3 is provided with a weigh scale 7 which is a weight measuring means for measuring the input weight of the input garbage.

  The stoker-type transport means 5 is configured by combining a fixed grate and a movable grate (not shown) on the staircase (hereinafter, both grates are simply referred to as “stalker”), and is placed on the reciprocating motion of the movable grate. It is configured to be able to transport the waste to the downstream side sequentially. In addition, the stoker-type transport means 5 is configured to dry the trash and heat it up to the vicinity of the ignition point in the incineration process (or trash combustion process), the combustion process to burn the trash, and the trash after the combustion process. And a drying stage 5a, a combustion stage 5b, and a post-combustion stage 5c that share the three processes of the post-combustion process for completely combusting the unburned components in the incinerated ash, respectively, upstream of the dust conveyance direction Are divided and arranged in order along the downstream side. Further, the drying stage 5a, the combustion stage 5b, and the post-combustion stage 5c are each divided into one or a plurality of compartments 8.

  Each compartment 8 includes a wind box 10 capable of independently supplying primary combustion air from below to the dust layer 9 on the stoker, and each wind box 10 is supplied with a primary combustion supplied by a blower fan 11. Damper mechanisms 12 that can adjust the air supply amount (hereinafter referred to as “primary air amount”) are provided separately. Further, the stoker type conveying means 5 is configured to be able to independently control the stoker conveying speed (stoker speed) for each compartment 8.

  With the above configuration, the dust introduced from the hopper 2 into the furnace 21 is sequentially conveyed on the stoker-type conveying means 5 to the drying stage 5a, the combustion stage 5b, and the post-combustion stage 5c. After being incinerated by the primary combustion air supplied from the box 10, the incineration ash is discharged out of the furnace through the incineration ash discharge port 13.

  Further, as shown in FIG. 1, the refuse incinerator 20 includes the present invention device (garbage supply amount estimation device) 1 and an automatic combustion control device 40 for stable dust combustion control in the furnace 21. . As shown in FIG. 2, the device 1 of the present invention includes a trinocular camera 6 in which three imaging means capable of imaging the entire upper surface of the surface layer portion of dust supplied and deposited in the hopper 2 and a hopper. The waste volume estimating means 31 for estimating the volume of the dust accumulated in the hopper 2, the weight meter 7 which is a weight measuring means for measuring the input weight of the trash put into the hopper 2, and the hopper 2 is supplied to the furnace 21. The waste supply weight estimation means 32 for estimating the supplied weight of the waste to be generated and the heat generation amount estimation means 33 for estimating the supply heat generation amount of the dust supplied from the hopper 2 to the furnace 21 are configured. Three image data captured by the trinocular camera 6 are input to the dust volume estimation means 31 and the heat generation amount estimation means 33, and the input weight measured by the weigh scale 7 is input to the dust supply weight estimation means 32.

  Further, as shown in FIG. 2, the dust volume estimation unit 31 includes a common imaging point extraction unit 34, a three-dimensional coordinate calculation unit 35, a three-dimensional coordinate adjustment unit 36, and a volume calculation unit 37. The Further, the heat generation amount estimation means 33 classifies the shape and color of dust into a plurality of dust types in advance, and connects the relationship with the heat generation amount per unit weight for each dust type to a dust type database 38 which is a database. The garbage type database 38 is configured to be searchable. Here, the waste volume estimation means 31, the waste supply weight estimation means 32, and the calorific value estimation means 33 in the apparatus 1 of the present invention can execute various arithmetic processes described later by software processing based on computer hardware. It has become a structure. The automatic combustion control device 40 is configured to be able to execute various control patterns by software processing based on computer hardware capable of executing automatic combustion control (ACC control) based on PID feedback control.

  Next, estimation processing of the supply weight of the dust supplied to the furnace 21 from the hopper 2 and the supply heat generation amount by the apparatus 1 of the present invention will be described. FIG. 3 shows a flowchart showing the supply weight estimation processing procedure, and FIG. 4 shows a flowchart showing the supply heat generation amount estimation processing procedure. First, a procedure for estimating the supply weight of dust will be described.

First, in step S 1, when new garbage is introduced into the hopper 2 by the crane 3, the input weight ΔW _n of the introduced garbage is measured and output to the garbage supply weight estimation means 32. In step S2, the dust volume estimation means 31 calculates the total dust volume V _n in the hopper based on the three image data P1 to P3 captured by the trinocular camera 6. The calculated dust volume V _n in the hopper 2 is output to the dust supply weight estimation means 32. Detailed volume calculation processing will be described later. Here, since the dust already accumulated in the hopper 2 is compressed and the volume is reduced by the weight ΔW _n of the introduced dust, an image captured after the dust accumulation state of the hopper 2 is stabilized is used.

Next, in step S3, based on the waste volume V _n of Suttepu S1, the waste obtained in step S2 is turned by weight [Delta] W _n and after dust is turned to calculate the specific gravity RB _n of dust at the bottom of the hopper 2. Specifically, the dust accumulated in the hopper 2 is compacted by a dust layer having a minute thickness Δd at a predetermined height h by the weight P (h) of the dust above it. A dust compaction model is assumed in which compression is performed at a compression ratio ε = f (P) that depends on the weight of dust, and becomes ε × Δd. The total volume V _n−1 of the dust before throwing in the waste is calculated using the previous processing, and the volume reduced by compression with the throwing weight ΔW _n is calculated using the dust compaction model. . Here, by _dividing the total volume V _n−1 of the dust before throwing it into a minute thickness Δd and multiplying the product by the compression ratio ε = f (P) at the height h, V ′ _n−1 can be calculated. Assuming that the calculated volume of waste before injection after decrease is V ′ _n−1 , the volume of newly input dust is given as ΔV _n = V _n −V ′ _n−1 , and the newly input surface layer portion the specific gravity RT _n the dust can be calculated as _{RT n = ΔW n / ΔV n} . Save this RT _n in a predetermined memory area for storing as input history of dust. Here, the dust in the surface layer portion introduced in the past is moved to the lower layer sequentially and finally moved to the bottom portion of the hopper 2 every time the dust at the bottom portion of the hopper 2 is supplied into the furnace 21 by the pusher 4. . Therefore, by storing the history of the volume of dust supplied to the furnace 21 from the bottom of the hopper 2 by the pusher 4, it is estimated how many times the dust layer at the bottom of the hopper 2 is waste that has been thrown in before. it can. Here, if the dust layer at the bottom is the waste thrown k times before, the specific gravity RT _n-k when the specific gravity is more compact than the dust thrown k times in the meantime and exists in the surface layer portion. Since it is larger, the specific gravity RB _{n of the} dust layer at the bottom of the hopper 2 at the present time is calculated using the dust compaction model. Note that a model derived beforehand through experiments or the like is used as the compression ratio ε = f (P) used in step S3.

  Next, in step S4, the dust supply volume supplied to the furnace 21 from the bottom of the hopper 2 by the pusher 4 is calculated. The dust supply volume ΔVf is calculated by the difference between the total volumes of dust in the hopper 2 before and after the supply, which is calculated by the dust volume estimation means 31 in the same manner as in step S2.

Next, in step S5, the specific gravity RB _{n of the} dust layer at the bottom of the hopper 2 calculated in step S3 and the dust supply volume ΔVf calculated in step S4 are supplied to the furnace 21 from the bottom of the hopper 2. that the waste feed weight DerutaWf, calculated as ΔWf = ΔVf × RB _n. The calculated dust supply weight ΔWf is output to the automatic combustion control device 40, and the control unit of the pusher 4 in the automatic combustion control device 40 controls the driving of the pusher 4, for example, a unit of dust into the furnace 21. Control is performed so that the supply weight per hour is constant.

  Next, dust volume estimation processing by the dust volume estimation means 31 will be described. First, the principle of deriving the three-dimensional coordinates of the object surface captured in a two-dimensional image from two or more two-dimensional image data will be briefly described.

  As shown in FIG. 5, when the object coordinate system (x, y, z) and the image coordinate system (X, Y) are defined, the coordinates (x, y) of the target object (in this embodiment, dust in the hopper 2). , Z) and the point on the virtual image formation of the camera (one imaging unit of the trinocular camera 6), that is, the corresponding point (Xc, Yc) in the image coordinate system is expressed by the following determinant of Formula 1. Is done. The origin of the object coordinate system (x, y, z) is not particularly limited, but it is convenient in calculating the dust volume if it is set in the hopper 2 or at the installation position of the trinocular camera.

  Here, Hc is a constant obtained from the focal length of the camera (each imaging unit), and the 12 C parameters are amounts determined by the relative position of the two coordinate systems of the object and the camera, the directional relationship, and the focal length of the camera. The C matrix of Equation 1 is called a camera parameter. Camera parameters can be calculated by measuring the position and direction of the camera, but in reality it is difficult to accurately give the focal length and so on, so usually a reference object with a known three-dimensional shape is prepared. Then, it is simple and practical to derive the C parameter by solving Equation 1 from the image obtained by imaging it.

That is, the C parameter can be derived from the point p (x, y, z) on the object surface and the corresponding point P (Xc, Yc) on the image. If the object coordinates x ₁ , y ₁ , z ₁ and the coordinates Xc ₁ , Yc ₁ of the corresponding points on the image are already known for one point p ₁ , Equations 2 and 3 The following two equations are obtained.

  Therefore, to obtain 12 C parameters, 12 simultaneous equations are required. Therefore, Equations 2 and 3 are generated for 6 reference points that are not on the same plane where the three-dimensional coordinates are known. Twelve C parameters can be obtained by solving simultaneously or by using the least square method using six or more reference points. Accordingly, each image is obtained from the coordinates (Xc, Yc) of a point (common imaging point) on the same object in two or more images obtained by imaging the same point using two or more cameras whose camera parameters are known. As shown in the equations (2) and (3), two equations representing the relationship between the common imaging point and the three-dimensional coordinates (x, y, z) in the object coordinate system are obtained. If two or more cameras are used in this way, four or more relational expressions can be obtained for the three unknowns x, y, and z. If this is solved using the least square method, the common imaging point Three-dimensional coordinates (x, y, z) in the object coordinate system can be calculated.

  Therefore, in order to calculate the three-dimensional coordinates of the surface layer of the dust on the hopper 2, the points captured in two or more images are recognized as the same point (common imaging point) in the surface layer of the dust. There is a need. That is, it is necessary to associate pixels pointing to the same point (common imaging point) between images. The common imaging point extraction means 34 first selects candidate points that are common imaging points in any two images of the three images P1 to P3 captured by the trinocular camera 6, first by the following two feature point extraction methods. A plurality are extracted by

1) Extraction of points with large density changes:
A point at which a large change in the shading level is observed in all directions such as the vertex of the object, and a point exhibiting the largest change in the vicinity thereof is extracted as a feature point. Here, the change in the shading level is determined by taking the square of the sum of the differences from the pixels adjacent to the left and right of the target pixel, the square of the sum of the differences from the pixels adjacent to the upper and lower sides, and the pixels adjacent to the upper right and lower left. Change in the four directions of the square of the sum of the differences and the square of the sum of the differences between the pixels adjacent to the lower right and upper left, and the minimum value is defined as the density change of the target pixel. Are extracted as feature points.

2) Extracting intersections of line components:
A line component in the image is extracted, and a point where two or more lines intersect is extracted as a feature point. The intersection of the line components is obtained by the following five processes A to E.

A) Edge enhancement by Sobel filter:
With reference to the pixel density of the 3 × 3 rectangular area with the number of pixels centered on each pixel shown in FIG. Here, in FIG. 6, e is the pixel density to be processed, and a to h around it are the densities of eight neighboring pixels centered on e.

B) Binarization processing:
The density of each pixel is converted to 0 or 1 depending on the threshold value.

C) Proximity line connection processing by expansion processing:
If the pixel of the binarized image after the binarization process is 1 or the pixels in the vicinity of 4 are 1, the pixel is also 1. This process causes adjacent lines to expand and join.

D) Thinning process:
The expanded line is narrowed to a line width of 1.

E) Intersection extraction process by filter:
In a binary image from which line components have been extracted by each of the processes A to D, for each point on the line (value is 1), a 3 × 3 rectangular area centered on that point is examined, and the area A point having four or more pixels of 1 is extracted as an intersection.

  Next, the common imaging point extraction unit 34 corresponds to a feature point indicating the same point from among a plurality of candidate points (feature points) to be common imaging points extracted by the two feature point extraction methods in each image. In addition, a matching process for extracting as a common imaging point is performed. In the present embodiment, the dissimilarity S defined by Equation 5 is calculated using the shading information of each pixel of each image Pi (i = 1 to 3). For two images of the image Pi (i = 1 to 3), for example, P1 and P2, the feature point a (x, y) of the image P1 and the corresponding point candidate b (x ′, y ′) of the image P2 with respect to the feature point a The dissimilarity S is defined as shown in Equation 5 using density information in a rectangular area of one side (2r + 1) pixels centered on each feature point.

  However, f1 (x, y) and f2 (x, y) are functions indicating the density of the pixel at the coordinates (x, y) of the images P1 and P2, respectively. Among the corresponding point candidates b for the feature point a, the smallest point in the corresponding point candidates that is smaller than the threshold value set by the dissimilarity S is extracted as a corresponding point, and a common imaging point between the images P1 and P2 is extracted. As specified. This process is executed for all extracted feature points and three images.

  By the way, between the images P1 and P2, the feature point a of the image P1 and the feature point b of the image P2 are associated as a common imaging point, and between the images P1 and P3, the feature point a of the image P1 and the feature point of the image P3. When the feature point b of the image P2 and the feature point c of the image P3 are not associated as the common imaging point between the images P2 and P3 even though c is associated as the common imaging point, the image P1 , The association between P2, the association between the images P1 and P3, or the association between the images P2 and P3 is wrong, so any one of the associations is performed using another candidate point, Check if wrinkles occur between the three images. In other words, it is effective to associate all the combinations of images with respect to one feature point without associating all the feature points between the pair of images P1 and P2. In addition, when the occurrence of wrinkles between the three images is not eliminated, it is necessary to process them as common imaging points between the two images, assuming that they are not common imaging points common to all three images.

  Next, the three-dimensional coordinate calculation means 35 is obtained from the two images for the common imaging point specified between the two images for each of the images P1 and P2, between the images P1 and P3, and between the images P2 and P3. Since four relational expressions of Formula 2 and Formula 3 are obtained, these are solved using the least square method to calculate the three-dimensional coordinates (x, y, z) of each common imaging point.

  Next, the three-dimensional coordinate adjusting unit 36 adjusts the common imaging point coordinates calculated by the three-dimensional coordinate calculating unit 35 between the two images. That is, in the case of a common imaging point common to the three images, there are three types of three-dimensional coordinates (x, y, z) between the images P1 and P2, between the images P1 and P3, and between the images P2 and P3. Although obtained, the results are not necessarily consistent with each other. Therefore, any one of the following three processes is performed.

  1) When one of the three three-dimensional coordinates is more than a predetermined distance away from the other, and the other two are within a predetermined distance, the two coordinates close to each other are adopted and the average is calculated The three-dimensional coordinates of the common imaging point are used.

  2) When the three three-dimensional coordinates are close to each other within a predetermined distance, the average of the three coordinates is set as the three-dimensional coordinates of the common imaging point.

  3) If the three three-dimensional coordinates are separated by a predetermined distance or more, perform the process 1) above for the two points closer to each other, perform the process 2) above, or For the image, six relational expressions of the two formulas (2) and (3) are derived and solved using the least square method to obtain the three-dimensional coordinates (x, y, z) of the common imaging point. Do any of the processing. Alternatively, assuming that there is a problem in specifying the common imaging point, the matching process may be performed again.

  By executing such adjustment processing, three-dimensional coordinate calculation errors due to matching errors can be reduced, and high-precision processing can be realized.

  As described above, when the three-dimensional coordinates are obtained for the common imaging points that can be associated with each other, the three-dimensional shape of the upper layer portion of the dust of the hopper 3 is specified. Then, the volume calculation means 37 integrates a region surrounded by the three-dimensional shape of the inner wall surface of the hopper 3 and the upper layer portion of the dust specified by the same object coordinate system (x, y, z) by the integration process. The total capacity can be calculated. If the xy plane is a horizontal plane, the total area may be accumulated by multiplying the small area of each xy coordinate by the difference between the z coordinate of the inner wall surface of the hopper 3 and the z coordinate of the dust upper layer. By the above processing, the dust volume estimation processing by the dust volume estimation means 31 is completed.

  Next, an estimation processing procedure of the supply heat generation amount of dust by the device 1 of the present invention will be described with reference to the flowchart of FIG. In FIG. 4, steps S1 to S5 are the same as the corresponding steps S1 to S5 of FIG.

  Step S6 is processed in parallel with step S2. In step S6, the three image data P1 to P3 input from the trinocular camera 6 to the dust volume estimation unit 31 in step S2 are also input to the heat generation amount estimation unit 33. The calorific value estimation means 33 uses one of the image data P1 to P3 to extract a region recognized as the same color range from the RGB output values, and divides the dust surface layer portion into a plurality of colors. Further, the three-dimensional coordinates corresponding to the extracted two-dimensional image area of each section are extracted from the three-dimensional coordinates of the dust surface layer portion obtained as intermediate data in the process of step S2, and for each section, color data and Assign 3D coordinates.

  Next, in step S7, the dust type database 38 is accessed, and for each category extracted in step S6, dust types having the same or similar color data are first extracted as dust type candidates by searching, and each dust type candidate is selected. Correlation processing is performed between the registered three-dimensional shape pattern and the three-dimensional coordinates of each section, and the degree of coincidence of the three-dimensional shapes is calculated. Here, a known three-dimensional pattern matching process is used. Then, the dust type candidate having the highest degree of coincidence is specified as the dust type of the corresponding category, and the standard value of the calorific value per unit weight and the specific gravity before compression registered in the dust type database 38 is extracted for each category. To do.

Next, in step S8, the volume of each section is obtained using the three-dimensional coordinates of each section and the newly introduced dust volume ΔV _n calculated in step S3, and the pre-compression extracted from the dust type database 38 is obtained. From the specific gravity, the weight ratio of each category of newly introduced garbage is calculated and registered as a history of the introduced garbage. Here, the dust compaction model used in Step S3 is assumed to be the same regardless of the quality of the waste, and a model is used in which the dust in the surface layer that has been thrown k times before exists at the bottom of the hopper 2, Assuming that the distribution ratio of the sections is maintained and the dust on the surface layer has moved to the bottom of the hopper 2, the supply weight ΔWf obtained in step S5 is obtained in steps S7 and S8 at the time k times before charging. Further, using the heat generation amount per unit weight and the weight ratio of each section, it is possible to estimate the supply heat generation amount of the dust supplied to the furnace 21 from the bottom of the hopper 2.

  The estimated supply calorific value of dust is output to the automatic combustion control device 40, and the automatic combustion control device 40 determines the furnace temperature, the stoker speed, the primary air amount, and the pusher 4 based on the input supply calorific value. Change the setting of control values such as dust supply weight per unit time, and set the dust combustion control to a predetermined control pattern based on measurement data such as furnace temperature measurement value, stoker speed measurement value, primary air volume measurement value, etc. Run based on. Or you may make it the control part of the pusher 4 in the automatic combustion control apparatus 40 perform drive control of the pusher 4 so that the supply calorific value per unit time may be maintained constant. Note that the control of the automatic combustion control device 40 is not the gist of the present invention, and thus detailed description thereof is omitted. Further, in FIG. 1, the measurement means for each measurement value is not shown.

Second Embodiment
Next, 2nd Embodiment of this invention apparatus 1 is described with reference to FIG. The difference from the first embodiment is the configuration of the dust volume estimation means 31 that estimates the content of data output from the trinocular camera 6 and the volume of dust accumulated in the hopper 2 using the data. . Since other components are the same as those in the first embodiment, a duplicate description is omitted.

  In the second embodiment, a product incorporating the functions of the common imaging point extraction unit 34, the three-dimensional coordinate calculation unit 35, and the three-dimensional coordinate adjustment unit 36 of the dust volume estimation unit 31 in the first embodiment as the trinocular camera 6. Is assumed. That is, the trinocular camera 6 calculates the distance along the imaging direction from the trinocular camera 6 for each pixel in the captured image, and outputs RGB data and distance data for each pixel, and the dust volume The estimation unit 31 uses distance data, and the heat generation amount estimation unit 33 uses RGB data.

  The dust volume estimating means 31 is composed of only the three-dimensional coordinate calculating means 35 ′ and the volume calculating means 37, and the three-dimensional coordinate calculating means 35 ′ converts the distance data output from the trinocular camera 6 into three-dimensional coordinates. Process. The volume calculation means 37 calculates the total volume of dust in the hopper 2 in the same manner as in the first embodiment, based on the three-dimensional coordinates of the dust surface layer and the three-dimensional coordinates of the inner wall surface of the hopper 2.

  Hereinafter, another embodiment will be described.

  <1> Although the trinocular camera 6 is used in the first embodiment, a plurality of monocular cameras may be used instead.

  <2> In each of the above-described embodiments, the device 1 of the present invention includes the calorific value estimation means 33. However, in the usage form of the incinerator in which there is no great variation in the quality of the input waste, the calorific value is not necessarily estimated The supply heat generation amount may not be estimated by the means 33. Therefore, the device 1 of the present invention may not include the calorific value estimation means 33.

  <3> In each of the above embodiments, the calorific value estimation means 33 extracts the three-dimensional shape of each section of the dust surface layer portion using three-dimensional coordinates in step S6 shown in FIG. It does not matter as a shape. In this case, the shape pattern registered in the dust type database 38 is also two-dimensional data.

  <4> In each of the above embodiments, the pusher 3 is used as a means for supplying dust from the hopper 2 to the furnace 21. However, instead of this, other supply means such as a screw conveyor may be used. Further, the shape of the hopper 2 is not limited to the shape shown in FIG. Further, the installation position of the trinocular camera 6 is not limited to the position shown in FIG.

  <5> In each of the above embodiments, a stalker-type trash incinerator is exemplified as the trash incinerator 20 to be controlled by the apparatus 1 of the present invention. However, the trash incinerator 20 may be other than the stalker type. Absent.

The block diagram explaining the structure of the related part of the garbage supply amount estimation apparatus of a garbage incinerator and a garbage incinerator concerning the present invention The block diagram which shows typically the structure of one Embodiment of the waste supply amount estimation apparatus of the waste incinerator which concerns on this invention The flowchart which shows the estimation processing procedure of the supply weight of refuse by the refuse supply amount estimation apparatus of the refuse incinerator which concerns on this invention The flowchart which shows the estimation process sequence of the waste heat generation amount of the waste by the waste supply amount estimation apparatus of the waste incinerator according to the present invention Explanatory drawing which shows the relationship between the object coordinate system (x, y, z) and image coordinate system (X, Y) which the three-dimensional coordinate calculation means of a garbage volume estimation means uses for three-dimensional coordinate calculation. The figure which shows the pixel density of the 3 * 3 rectangular area for demonstrating the edge emphasis process by a Sobel filter by the common imaging point extraction means of a dust volume estimation means The block diagram which shows typically the structure of another embodiment of the refuse supply amount estimation apparatus of the refuse incinerator which concerns on this invention.

Explanation of symbols

1: Waste supply amount estimation device for waste incinerator according to the present invention 2: Hopper 3: Crane 4: Pusher (supply means)
5: Storker type conveying means 6: Trinocular camera (multiple imaging means)
7: Weigh scale (weight measuring means)
8: Compartment 9: Waste layer 10: Wind box 11: Blower fan 12: Damper mechanism 13: Incineration ash discharge port 20: Waste incinerator 21: In-furnace 31: Waste volume estimation means 32: Waste supply weight estimation means 33: Heat generation Quantity estimation means 34: Common imaging point extraction means 35: Three-dimensional coordinate calculation means 35 ': Three-dimensional coordinate calculation means 36: Three-dimensional coordinate adjustment means 37: Volume calculation means 38: Dust type database 40: Automatic combustion control device

Claims (7)

Three or more imaging means capable of imaging the entire top surface of the surface portion of the dust supplied and deposited in the hopper of the garbage incinerator;
Dust volume estimation for deriving three-dimensional coordinates at a plurality of positions on the upper surface of the surface layer portion from three or more images picked up by the three or more image pickup means and estimating the volume of dust in the hopper based on the three-dimensional coordinates. Means,
A weight measuring means for measuring an input weight of dust to be put into the hopper;
Dust supply weight estimation means for estimating the supply weight of garbage supplied from the hopper into the garbage incinerator based on the waste volume estimated by the waste volume estimation means and the input weight measured by the weight measurement means. And comprising
The dust volume estimating means picks up the same position on the upper surface of the surface layer portion for a plurality of sets of arbitrary two or more images selected from three or more images picked up by the three or more image pickup means. Common imaging point extraction means for extracting a plurality of common imaging points;
Three-dimensional coordinate calculation means for calculating three-dimensional coordinates of the plurality of common imaging points for the plurality of sets of images extracted by the common imaging point extraction means;
3D coordinate adjustment means for adjusting the calculated 3D coordinates between the plurality of sets of the plurality of images;
Volume adjustment means for calculating the volume of the dust based on the wall shape of the hopper, using the adjusted three-dimensional coordinates as three-dimensional coordinates at a plurality of positions on the upper surface of the surface layer portion. Waste supply estimation device for garbage incinerators. The three or more imaging units are trinocular cameras in which the relative positional relationship between the three imaging units is fixed,
The common imaging point extraction unit may include three sets of two images selected from three images captured by the trinocular camera, or the three sets of two images and one set of the three images. 2. The apparatus according to claim 1, wherein a plurality of common imaging points are picked up for imaging the same position on the upper surface of the surface layer part. A plurality of imaging means capable of color imaging the entire upper surface of the surface layer portion of the dust supplied and deposited in the hopper of the garbage incinerator;
Dust volume estimation means for deriving three-dimensional coordinates at a plurality of positions on the upper surface of the surface layer portion from a plurality of images captured by the plurality of imaging means, and estimating a volume of dust in the hopper based on the three-dimensional coordinates; ,
A weight measuring means for measuring an input weight of dust to be put into the hopper;
Dust supply weight estimation means for estimating the supply weight of garbage supplied from the hopper into the garbage incinerator based on the waste volume estimated by the waste volume estimation means and the input weight measured by the weight measurement means. When,
The waste quality is estimated from the shape and color of the surface dust existing on the surface layer portion from the color image captured by the imaging means, and at least the estimated dust quality of the surface dust and the dust supply weight estimation means are estimated. And a heat generation amount estimation means for estimating a supply heat generation amount of the dust supplied from the hopper into the waste incinerator based on the supply weight. Estimating device. Three or more imaging means are provided,
The dust volume estimating means picks up the same position on the upper surface of the surface layer portion for a plurality of sets of arbitrary two or more images selected from three or more images picked up by the three or more image pickup means. Common imaging point extraction means for extracting a plurality of common imaging points;
Three-dimensional coordinate calculation means for calculating three-dimensional coordinates of the plurality of common imaging points for the plurality of sets of images extracted by the common imaging point extraction means;
Three-dimensional coordinate adjustment means for adjusting the calculated three-dimensional coordinates between the plurality of sets of the plurality of images;
Volume adjustment means for calculating the volume of the dust based on the wall shape of the hopper, using the adjusted three-dimensional coordinates as three-dimensional coordinates at a plurality of positions on the upper surface of the surface layer portion. The apparatus for estimating a dust supply amount of a refuse incinerator according to claim 3. Color imaging can be performed on the entire top surface of the surface layer of dust that has been supplied and deposited in the hopper of a garbage incinerator, and the distance information of each pixel in the captured image can be output. The relative positional relationship between the three imaging units is fixed. A trinocular camera,
Waste volume estimation means for estimating the volume of dust in the hopper based on distance information of each pixel output from the trinocular camera;
A weight measuring means for measuring an input weight of dust to be put into the hopper;
Dust supply weight estimation means for estimating the supply weight of garbage supplied from the hopper into the garbage incinerator based on the waste volume estimated by the waste volume estimation means and the input weight measured by the weight measurement means. When,
The quality of dust is estimated from the shape and color of surface dust existing on the surface layer portion from the color image captured by the trinocular camera, and the estimated dust quality of the surface dust and the dust supply weight estimation means estimate And a heat generation amount estimating means for estimating a supply heat generation amount of the dust supplied from the hopper into the waste incinerator based on the supplied weight. Quantity estimation device.   The calorific value estimation means classifies the shape and color of garbage into a plurality of garbage types in advance, and uses the garbage type database in which the relationship between the calorific value per unit weight for each of the garbage types is made into a database. 6. The waste supply amount estimation apparatus for a waste incinerator according to claim 3, wherein the supply heat generation amount is estimated based on the supply weight estimated by a weight estimation unit.   The dust supply weight estimation means obtains the specific gravity of the dust at the bottom of the hopper using a dust compression model that models the compression state due to the weight of the dust accumulated in the hopper, and determines the specific gravity and the hopper from the specific gravity and the hopper. The waste supply amount estimation apparatus for a waste incinerator according to any one of claims 1 to 6, wherein the supply weight is estimated from a supply volume of waste supplied into the waste incinerator.

Images