JP5361595B2 - Combustion object processing system, processing method, and combustion furnace combustion control system using the same, supplied to combustion furnace - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a treatment system and a treatment method and a combustion control system for a combustion furnace using the treatment system and treatment method, enabling stable combustion control and supply of a combustion target even when the composition of the combustion target supplied to the combustion furnace is fluctuated sharply. <P>SOLUTION: Input object information input to a pit 2 and image information within the pit 2 are input, and binarization processing etc. is performed with respect to instantaneous image information to estimate instantaneous combustion target information. Based on continuous image information, a dropping trajectory of the combustion target G is estimated. Based on these, an accumulated combustion target distribution state A is estimated. Based on a quality distribution state A, the combustion target G within the pit 2 is agitated, and based on a quantity distribution state A, the combustion target G is transferred to the combustion furnace. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to a processing system and a processing method for a combustion object supplied to a combustion furnace, and a combustion control system for the combustion furnace using these, and particularly to a combustion object having a large composition fluctuation such as waste supplied to a waste incinerator. The present invention relates to a processing system, a processing method, and a combustion control system for a combustion furnace using the same, such as agitation processing in a pit stored before being supplied to a combustion furnace.

  Wastes processed at waste disposal facilities and the like are mixed with various kinds of materials, so the properties (waste quality) of the wastes are not stable due to various factors, and the fluctuations are severe. Therefore, if such a combustible material subject to severe composition fluctuations is supplied to the combustion furnace as it is, stable combustion control becomes difficult, which causes generation of dioxins, carbon monoxide (CO), or nitrogen oxides (NOx). In addition, the combustion furnace may be greatly damaged by high heat in the combustion of a combustion object having a large calorific value or a corrosive substance generated by combustion.

In order to realize stable combustion control of such a combustion furnace,
(I) A method of grasping the quantity and quality of the combustion object supplied to the combustion furnace and controlling the operating conditions of the combustion furnace (conditions such as combustion temperature and air supply amount) corresponding thereto, or (ii) combustion in advance A method for supplying a stable quantity and quality of a combustion object to a combustion furnace by storing a predetermined amount of the object to be supplied to the furnace and stirring and homogenizing it has been proposed and put into practical use.

  For example, for the former, a fluid incinerator as illustrated in FIG. 8 can be cited. It has a calculation mechanism that can accurately predict and measure the amount of dust supply, and can also determine the properties of the dust, so that the amount of dust supply, the amount of secondary air, etc. can be controlled. It has a chute 102 that connects the dust outlet 105 of the duster and the dust supply port of the fluidized bed incinerator 103, and it is possible to observe how the dust falls from the dust outlet 105 of the duster 101. The image of the TV camera 104 attached to the position is processed in real time, and the amount of falling dust 113 or the amount of generated heat and the amount of generated heat 114 are calculated. Based on the calculated signals, the amount of dust supplied, the amount of secondary air, and the amount of flowing air are calculated. Control 115 can be performed (see, for example, Patent Document 1). Here, 106 is a boiler, 107 is a hopper, 108 is a fluidized bed, 109 is a motor, 110 is a blower, 111 is forced air, 112 is secondary air, 116 is motor rotation speed control, 117 is secondary air amount control, Reference numeral 118 denotes flowing air amount control.

  In addition, the latter is illustrated in FIG. 9 so that the waste thrown into the waste pit can be made uniform by controlling the crane by capturing the waste quality in color tone. A control device for an automatic crane for a waste disposal factory that stirs using such a crane can be given. Specifically, it is measured by a garbage height measuring means for measuring the height of the garbage accumulated in the garbage pit P using the parallax of the two cameras 202a and 202b in stereo view, and a garbage height measuring means. A garbage height storage means for storing garbage height information as a garbage height map, and a garbage color tone storage means for discriminating and storing the color tone of the garbage from the garbage height information, based on the garbage color map The crane 201 is controlled so that the foreign waste Ga in the waste pit P is specified and the foreign waste Ga is stirred (see, for example, Patent Document 2). Here, D is a distance between two cameras, G is garbage, H is a hopper, R is a crane operation room, 211 is a crane girder, 212 is a bucket, 213 is a club, and 214 is a traveling rail of the crane girder. .

Japanese Patent Application Laid-Open No. 09-060842 JP 2007-126246 A

  However, the above prior art has the following problems. That is, in the method of accurately predicting and measuring the amount of dust immediately before being put into the incinerator, and determining the properties of the dust, and controlling the amount of dust, secondary air, etc., the properties of the dust are If it is uniform to some extent, it can exert a high control function.However, if the waste properties are very unstable and have large fluctuations, sufficient control cannot be achieved and the combustion state in the incinerator is poor. It became stable and stable combustion control was difficult.

  Also, in the method of monitoring and stirring operations based on the waste height and color tone at the waste incinerator pit, the operation is performed based on only the dust height and color tone information at a specific time, that is, the surface information. Therefore, it has been difficult to ensure sufficient uniformity. Even if the garbage height is controlled to be constant, the uniformity of the garbage quality is not guaranteed, and the dust quality in the entire pit cannot be uniformly agitated only by the garbage quality information obtained from the color tone of the surface alone. This problem could not be solved sufficiently.

  Further, conventionally, since it was not possible to grasp information on the garbage dumped into the pit from the garbage truck, the stirring operation was dependent on the skill of the operator. In many cases, the operation is not sufficiently agitated by the operator alone, and is often put into the incinerator, the combustion state in the incinerator becomes unstable, the generation of dioxins, CO, NOx, etc. increases, and the unburned substances increase. This may cause abnormal fluctuations in furnace temperature.

  An object of the present invention is to provide a combustion object that can be stably controlled even when the combustion object supplied to the combustion furnace has a large compositional variation, in view of the above-described problems of the prior art. It is an object of the present invention to provide a possible processing system, a processing method, and a combustion control system for a combustion furnace using these.

  As a result of intensive studies, the present inventors have found that the above object can be achieved by a treatment system, a treatment method, and a combustion control system for a combustion furnace using these, which are supplied to a combustion furnace shown below. The present invention has been completed.

The present invention relates to a pit for storing a combustion object to be supplied to a combustion furnace, an inlet for introducing the combustion object into the pit, a stirring means for stirring the stored combustion object, and a stored combustion object A processing system for a combustion object to be supplied to a combustion furnace, comprising transfer means for transferring the object to the combustion furnace, and an image monitor for photographing the accumulated combustion object from the inlet to the pit,
(A) Input the input information including the whole combustion object to be put into the pit or individual quantity and quality and the image information from the image monitor,
(B) Binarization processing, mask processing, noise removal processing, and labeling processing are performed on the instantaneous image information in the vicinity of the input port, and individual combustion objects that have been input are individually determined based on the input information. Estimate instantaneous combustion object information including size or quantity and quality,
(C) detecting the position of the identified combustion object on the basis of image information for a predetermined period of time along the path from the inlet to the pit, and estimating the fall trajectory of the combustion object;
(D) The inside of the pit is divided into predetermined areas, and based on the input information, instantaneous combustion object information, and the fall trajectory of the combustion object, the accumulated quantity and quality distribution state A of the combustion object is determined. Estimate
(E) In the quality distribution state A, when the difference between the divided areas is out of a predetermined range, the stirring means is driven to stir the combustion object in the pit,
(F) In the distribution state A of the quantity, when the quantity of the accumulated combustion object or the quantity and quality of a specific region are within a predetermined range, the transfer means is driven to burn the combustion object It is provided with the arithmetic processing part transferred to a furnace.

The present invention also relates to a method of processing a combustion object that is stored in a pit in advance and then supplied to a combustion furnace, and includes an entire combustion object that is input to the pit or an input that includes individual quantities and qualities. Based on the information and image information of the combustion object photographed from the entrance to the pit to the inside of the pit,
(B) Binarization processing, mask processing, noise removal processing, and labeling processing are performed on the instantaneous image information in the vicinity of the input port, and individual combustion objects that have been input are individually determined based on the input information. Estimate instantaneous combustion object information including size or quantity and quality,
(C) detecting the position of the identified combustion object on the basis of image information for a predetermined period of time along the path from the inlet to the pit, and estimating the fall trajectory of the combustion object;
(D) The inside of the pit is divided into predetermined areas, and based on the input information, instantaneous combustion object information, and the fall trajectory of the combustion object, the accumulated quantity and quality distribution state A of the combustion object is determined. Estimate
(E) In the distribution state A of the quality, when the difference between the divided areas is out of the predetermined range, the combustion object in the pit is stirred,
(F) In the distribution state A of the quantity, the combustion object is transferred to a combustion furnace when the accumulated combustion object or the quantity and quality of a specific region are within a predetermined range. .

  A variety of substances and components are mixed in the combustion object supplied to the combustion furnace and the composition thereof may fluctuate greatly, and it is difficult to form stable combustion conditions. The present invention stores such a combustion object in the pit in advance, and based on the information on the input material that is input to the pit and the image information from the input port of the combustion object to be stored to the inside of the pit, Estimate the distribution of quantity and quality of materials, perform appropriate agitation operations based on the estimation results, and supply uniform combustion objects to the combustion furnace to ensure stable combustion conditions for the combustion furnace It is possible to do that. Specifically, for example, when waste is used as a combustion target, the quality of waste can be roughly classified by collected household waste, office waste, or commercial waste, which is input from a collection vehicle. The quantity and quality of the combustion object to be identified can be specified as input information. Therefore, based on such input information and image information, the movement / storage state of each input combustion object is estimated, and the distribution state of the entire combustion object in the pit is grasped and a uniform distribution state is formed. A stirring operation can be set. As a result, it is possible to form a processing system or a processing method capable of supplying a combustion target capable of stable combustion control even when the combustion target supplied to the combustion furnace has a severe composition fluctuation. became.

Here, “binarization” with respect to image information means that each pixel is compared with a fixed threshold value and an image is created only with white and black depending on the size thereof. For example, waste is defined as a combustion object. In this case, it is possible to obtain image information in which other elements are excluded by distinguishing them from colors specific to household garbage bags and colors from other garbage and pit walls. “Mask processing” for image information means that when only a change in a partial area of an image or the like is to be seen, the partial area to be targeted is cut out from the image or the like, and binarization processing of the cut-out image is performed. Thus, image information of only a specific combustion object in the target region can be obtained. “Noise removal processing” for image information refers to performing “expansion” and “degeneration” as typical arithmetic processing mainly applied in binary image processing. In order to connect the components, the black part indicating the waste is expanded and then contracted, and when cutting the fine band-shaped part, the white area as the background is expanded and then contracted, and the light In order to remove fine components that can be adjusted, noise can be removed by repeatedly performing “expansion” and “degeneration”. The “labeling process” for image information is a process of assigning a number to a set of pixels on the image and classifying the pixels. The labeling process can be performed separately for each labeled area. By gathering information (number, area, shape), image information can be reduced and work can be performed efficiently. The “divided area” refers to an area divided in the pit at a constant interval on a plane, and is used as a unit for judging the quantity and quality non-uniformity of accumulated combustion objects.

  The present invention also provides a processing system or a processing method for a combustion object supplied to the combustion furnace, wherein binarization processing, mask processing, noise removal processing, and image information at a specific time in a pit are provided. A labeling process is performed to estimate the quantity and quality distribution state B of the accumulated combustion object, and the difference in quantity or / and quality between the divided areas is compared with the distribution state A at the specific time. When it falls outside the predetermined range, the fall trajectory of the combustion object is corrected, and the corrected distribution state A is estimated.

  As described above, in the combustion object processing system or method according to the present invention, several estimation processes are performed based on the input information and the image information. At this time, there is an uncertain element in each information at a predetermined ratio, which becomes an error factor of the estimation result. For example, when waste is to be burned, the specific gravity set as the waste quality in the input information may actually differ from the waste in the bag, and the estimated fall trajectory may be misaligned or already in the pit The accumulation position may move by colliding with certain garbage, and its distribution state may change. Moreover, the difference from the waste quality in the bag actually becomes an error in the set calorific value. According to the present invention, a distribution state closer to the actual situation can be estimated by correcting the influence of such fluctuations and errors from image information at a specific time in a pit.

  The present invention is a combustion control system for a combustion furnace using a processing system or a processing method for a combustion object supplied to any of the above combustion furnaces, wherein the book is applied to the combustion object accumulated in a pit. Based on the quantity and quality distribution state A estimated in the treatment system or treatment method, the calorific value of the combustion object supplied to the combustion furnace is estimated and used as one of the combustion control indices of the combustion furnace. It is characterized by.

  In order to operate the combustion furnace under stable combustion conditions, it is preferable to stabilize the supply amount and heat generation amount of the combustion object, but for example, the combustion object whose supply amount and waste quality greatly vary, such as waste. In the case of goods, it was very difficult to ensure these conditions. In the combustion object processing system or method according to the present invention described above, the supply amount can be arbitrarily set by controlling the transfer means, and the calorific value is also qualitative of the stored combustion object. It became possible to ensure qualitative uniformity by stirring properly after grasping the distribution state. In other words, it is possible to supply a combustion object having a uniform heat generation amount and an estimated heat generation amount. Therefore, by using the information on the quantity and quality of the combustion object obtained from such a processing system or processing method, in particular, the real-time calorific value as one of the indicators of combustion control of the combustion furnace, optimum combustion in the actual operation state is achieved. It became possible to control in the state.

Schematic overall configuration diagram of a combustion object processing system according to the present invention Explanatory drawing which illustrates processing operation of this processing system concerning the present invention Schematic illustrating image information in a pit according to the present invention Explanatory drawing which illustrates processing operation of image information concerning the present invention Explanatory drawing which illustrates the estimation processing operation of the fall locus | trajectory of the combustion target object which concerns on this invention Explanatory drawing which illustrates the fall locus | trajectory of the combustion target object which concerns on this invention Schematic illustrating a combustion furnace using a combustion object processing system according to the present invention. Schematic illustrating a fluid incinerator according to the prior art Schematic illustrating a control device for an automatic crane for a waste disposal factory according to the prior art

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. That is, the case where the processing system of the combustion object supplied to the combustion furnace according to the present invention, the processing method, and the combustion control system of the combustion furnace using these are applied to, for example, a stoker-type waste incinerator as one embodiment. explain.

<Combustion Object Processing System According to the Present Invention>
A combustion object processing system (hereinafter referred to as “the present processing system”) supplied to the combustion furnace according to the present invention is supplied to a stoker-type waste incinerator (hereinafter referred to as a “combustion furnace”). A waste pit (corresponding to a pit) for storing waste (corresponding to a combustion object), an input port for introducing the combustion object into the pit, a stirring means for stirring the stored combustion object, and a storage Means for transferring the waste to the combustion furnace, an image monitor for photographing the accumulated waste from the inlet to the pit, means for stirring the waste in the pit, and transferring the waste to the combustion furnace And an arithmetic processing unit that performs predetermined calculation / estimation based on the input predetermined information and drives the stirring means and the transfer means.

[Configuration of this processing system]
FIG. 1 shows a schematic overall configuration example of the present processing system. Waste G collected and mounted from various places is thrown into the pit 2 from the garbage truck C through the slot 1. At this time, the input information including the input amount and the quality of the waste G input in advance can be obtained for each garbage collection vehicle C. A state in which the input waste G moves from the input port 1 into the pit 2 is photographed in real time by the image monitor 3 provided in the upper part of the pit, and image information of the waste G can be obtained. The stored waste G is stirred by a stirrer 4 (corresponding to a stirring means) so that the waste quality becomes uniform, and then collected by a crane 5 and combined with a hopper 6 (corresponding to a transport means together with the crane 5). ) To a combustion furnace (not shown). At this time, the operations of the agitator 4, the crane 5, and the hopper 6 are controlled by an output signal calculated and estimated based on the input information and the image information input to the arithmetic processing unit 7.

  Here, the collected waste G is the shape of the waste that is thrown away from the image information, such as crushed waste, pruning waste (plant waste such as fallen leaves and pruned branches), sludge, etc. The type of garbage can be determined by identifying the color and color. In addition, in recent years, many types of bags have different colors and surface patterns depending on the quality of garbage, such as household waste, office waste, or commercial waste. It is possible to specify input information such as specific gravity or calorific value. Therefore, it is possible to estimate the waste quality and the calorific value of the waste G that has been input together with the total weight and the total volume of the waste that has been input. Furthermore, by recognizing the color and pattern of the bag as the image information, it is possible to estimate the waste quality classification and distribution state. Therefore, when waste G in which these are mixed is introduced, by estimating the distribution state of the waste G in a bag unit, it is possible to perform an estimation equivalent to the distribution state of waste in a dispersed state, and appropriate mixing processing The uniform waste G can be supplied to the combustion furnace. In addition, it is possible to estimate the waste quality using information such as the attribute information of the garbage truck C (information on whether the object to be collected is specified or whether there is a separate collection capability), loading weight, collection area, etc. it can.

  At this time, household waste that contains a lot of water has a low calorific value, and office waste that contains a lot of paper has a stable, almost constant calorific value, and is a commercial system that contains a lot of plastic with a high calorific value. In the case of trash, the calorific value is high. Therefore, it is possible to estimate the calorific value of the input waste and the calorific value of the waste supplied to the combustion furnace by creating a database of the average calorific value for each type of waste in advance. In addition, it is possible to control the supply amount and waste quality of the waste G supplied to the combustion furnace and realize the combustion conditions under the optimum conditions. In other words, the waste G containing a large amount of household waste preferably has a combustion furnace operating condition of a relatively low temperature and a small amount of air supply. In the case of office waste containing a lot of paper, a stable air supply at a stable temperature is possible. In the case of commercial waste that contains a large amount of plastic with a high calorific value, a large amount of air supply is required at a high temperature. By setting the operating conditions of the combustion furnace in advance based on the information on the calorific value of the supplied waste G, the correction and adjustment of the combustion conditions in the case of actual combustion can be made a very narrow range. it can.

  It is preferable that a plurality of the inlets 1 be provided so that the amount of waste G introduced into the pit 2 or the waste quality becomes relatively uniform by the input of the waste G from different directions. Furthermore, in this processing system, as will be described later, in order to make the waste G in the pit 2 uniform, the input port is based on information such as the waste quality of the waste G loaded on the collection vehicle C. It is also possible to select and control one.

  In addition, in FIG. 1, although the structural example which provided the stirrer 4 separately was shown, when waste quality is comparatively uniform and there is little necessity for stirring, stirring operation is not carried out by the crane 5 without providing the stirrer 4. You may make it perform.

  The image monitor 3 is not particularly limited in position and quantity as long as the state in the pit 2 from the insertion port 1 can be photographed in real time as color information. In FIG. 1, a wide-angle video camera is used. A configuration in which one unit is provided is illustrated. In particular, an arrangement is preferable in which the moving path is inclined so that the moving state of the waste G can be photographed from the front so that the moving state of the waste G from the inlet 1 to the pit 2 can be clearly understood. It is preferable that the camera can be photographed from the upper front.

[About processing operations of this processing system]
In the present processing system, the following processing operations are controlled by the arithmetic processing unit 7.
(A) Input waste information including the whole waste to be put into the pit or individual quantity and quality, and image information from the image monitor.
(B) A binarization process, a mask process, a noise removal process, and a labeling process are performed on the instantaneous image information in the vicinity of the input port, and the individual size of the input waste is determined based on the input information. Alternatively, instantaneous waste information including input and waste quality is estimated.
(C) Based on image information for a predetermined period of time from the insertion port to the pit, the position of the identified waste is detected, and the fall trajectory of the waste is estimated.
(D) The inside of the pit is divided into predetermined areas, and the amount of accumulated waste and the distribution state A of waste quality are estimated based on the input information, instantaneous waste information, and waste fall trajectory. .
(E) In the quality distribution state A, when the difference between the divided areas is out of the predetermined range, the stirring means is driven to stir the waste in the pit.
(F) In the quantity distribution state A, when the quantity of accumulated waste or the quantity and quality of a specific area are within a predetermined range, the transfer means is driven, and the waste is sent to the combustion furnace. Transport.

  Next, each processing operation of this processing system will be described in detail with reference to FIGS. FIG. 2 exemplifies the outline of the processing operation of this processing system from the input information (vehicle information) and the image information that are input to the stirring control in the pit and the combustion control of the combustion furnace. The processing content of the image information includes image processing of the input waste, image processing of the waste accumulation distribution, and image processing of the waste in the pit, and vehicle information data and a waste heat generation database are used for these. Based on the input information and image information, the movement / storage state of each injected combustion object is estimated, the distribution state of the entire combustion object in the pit is grasped, and the stirring operation to form a uniform distribution state is performed. Can be set. As a result, it is possible to form a treatment system capable of supplying a combustion target capable of stable combustion control for the combustion furnace even when the composition of combustion supplied to the combustion furnace is subject to severe composition fluctuations. became.

(A) Input of input information and image information Input information including the entire waste input to the pit or individual quantity and quality, and image information from the image monitor are input to the arithmetic processing unit 7. Here, the input information and the image information include at least the following information.
(A-1) Input information (vehicle information)
Specifically, as shown in FIG. 2, as vehicle information, the name of the trader relating to a specific collection vehicle C, the garbage collection area, whether the type of collected garbage is in a bag, household waste, office waste, Information on the type of commercial waste and the identification method (each color, pattern, etc.), the volume of the loaded garbage, and the weight of the loaded garbage are input. At this time, as described above, it is possible to specify the waste quality in units of bags and set the input information such as specific gravity or calorific value in advance, from the input port 1 specifying the total weight or total volume of the input waste. Input information including the quantity and quality of the input waste can be obtained.
(A-2) Image information of input waste Specifically, as shown in FIG. 3, image information in the vicinity of a specific input port 1 (in the case of a plurality of input ports, an image for each input port) is input. The It is possible to identify the color and shape of individual garbage bags or the type of crushed garbage, pruned garbage or other garbage. In addition, the weight and calorific value of each individual garbage can be estimated based on the garbage calorific value database based on the type of each garbage or the predetermined garbage quality of the bag.
(A-3) Image Information of Garbage in the Pit Specifically, as shown in FIG. 3, image information in the pit 2 where the thrown-in garbage is accumulated is input. The color of the dust layer surface and the distribution of the particle size of the dust layer in individual areas at a specific time can be identified. In addition, from the type of the waste layer or the waste quality of the bag body set in advance, the distribution of the waste quality of the waste G deposited, and the distribution of the calorific value based on the weight of the waste and the waste heat generation database Can be estimated.

(B) Estimation of instantaneous combustion object information based on instantaneous image information in the vicinity of the input port The still image obtained by cutting out a moving image is processed as image information, and the number and color of garbage bags input from the collection vehicle C, Recognize shapes. That is, the instantaneous image information is binarized, masked, noise processed, and labeled as shown below, and the area, number, and color of the garbage bags are grasped. From the recognized area of the garbage bag body, for example, the volume of the garbage bag body is calculated assuming an ellipsoidal shape, and the apparent density of the garbage bag body is estimated using the loaded weight as vehicle information. The type of waste, that is, household waste and office waste is judged from the color of the garbage bag and the surface pattern. Furthermore, plant waste such as fallen leaves and pruned branches is judged from the color tendency of the bag (green). For crushed waste, pruned waste, sludge, etc., which are not bag waste, the shape and color of the waste to be introduced are identified, and the type of waste is determined. Here, the average calorific value for each type of garbage is stored in a database, and the information such as the load weight of the collected garbage truck C and the collection area is also used. Calculate the calorific value of the input waste.

Specifically, as shown in FIG. 4, binarization processing, mask processing, noise removal processing, and labeling processing are performed on instantaneous image information in the vicinity of the inlet 1 as follows. The individual size or input amount of the waste G is estimated, and instantaneous waste information including waste quality such as calorific value is estimated based on the input information and the waste heat value database.
(B-1) Binarization “Binarization” with respect to image information means that each pixel is compared with a fixed threshold value and an image is created with only white and black depending on the size. Specifically, as shown in FIG. 4 (b-1), for example, image information that distinguishes from the color unique to household garbage bags from the color of other garbage, pit walls, etc., and excludes other elements. Can be obtained.
(B-2) Mask processing “Mask processing” for image information refers to cutting out (extracting) a partial area to be targeted from the image or the like when only a change in the partial area of the image or the like is to be seen. Specifically, as shown in FIG. 4 (b-2), an image of an area immediately after the waste G is input in the vicinity of the input port 1 is extracted, and binarization processing is performed. Image information of only the specific waste G (for example, household waste) can be obtained.
(B-3) Noise Removal Processing “Noise removal processing” for image information refers to performing “expansion” and “degeneration” as typical arithmetic processing mainly applied in binary image processing. Specifically, as shown in FIG. 4B-3, when the black and white image of the waste G subjected to the masking process (for example, household waste in a bag) is disturbed (for example, the peripheral edge portion of the image of the bag) In order to connect a cut component, the black portion indicating the waste G is expanded and then degenerated, and when a fine strip portion is cut, it becomes the background. In order to perform degeneration after expanding the white area and to remove fine components that can be generated by adjusting light, it is possible to remove noise by repeatedly performing “expansion” and “degeneration”.
(B-4) Labeling process The “labeling process” for image information is a process of assigning a number to a set of pixels on the image and classifying the pixels. Specifically, as shown in FIG. 4 (b-4), waste G (for example, household waste) can be distinguished and processed for each labeled area, and information (number, By combining the area and shape) and forming the connected component, the image information can be reduced and efficient work can be performed.

(C) Estimation of the trajectory of the waste falling from the entry port into the pit As shown in FIGS. 5 and 6, it is specified based on image information for a predetermined period of time from the entry port 1 to the pit 2. The position of the waste G thus detected is detected, the fall locus of the waste G is obtained, and the deposition position is estimated. The movement of the garbage individual between the still images cut out from the moving image is detected by the falling locus. Here, waste G is used as garbage bags G1 to G3, and the above-described processing operation (b) is performed on the image information for a plurality of times, and the fall trajectory is estimated as shown in FIGS. D). For the three waste bags G1a to G3a (FIG. 5A) after the initial waste G processing operation, for example, the moving direction and moving speed after 1 sec or 0.5 sec are estimated, and the moving positions G1b to G3b are estimated. (FIG. 5B) is assumed, and the deviation from the movement positions G1c to G3c (FIG. 5C) after performing the same processing operation on the actual image information after the same time is corrected. Fall vectors G1ad to G3ad (FIG. 5D) are set. By repeating this, it is possible to estimate the dropping trajectory for each of the garbage bags G1 to G3 as shown in FIG. 6 (A), and further into the pit 2 from the inlet 1 as shown in FIG. 6 (B). The fall trajectory to the divided areas (refer to areas where the pits 2 are divided at regular intervals on the plane, for example, areas represented by s5-t3, s6-t3, s8-t4, etc.) can be estimated. . That is, it becomes possible to track the place where the waste G is input (input 1) and the spread from there to the pit 2.

In the above description, the waste bags G1 to G3 are captured with the waste G as a dot or an equivalent circle. However, since the waste bags G1 to G3 actually have various shapes, the garbage bags G1 to G1 are obtained from the image information at each time. By determining the position of the center of gravity of G3 and tracking the movement of the position of the center of gravity, it is possible to estimate the fall trajectory from the insertion port 1 into the pit 2 for each garbage bag body G1 to G3. Specifically, as shown in FIG. 5E, the following steps are performed.
(C-1) Setting fall vectors of the garbage bags G1 to G3 (c-2) Cutting out (extracting) a still image from a moving image (image information) (FIG. 5A)
(C-3) The gravity center positions G1a to G3a of the falling garbage bags G1 to G3 are detected (FIG. 5A).
(C-4) A fall vector is added to predict the next center of gravity G1b to G3b (FIG. 5B).
(C-5) The next still image is cut out from the moving image (FIG. 5C).
(C-6) The gravity center positions G1c to G3c of the garbage bag body of the next image are detected (FIG. 5C).
(C-7) The center of gravity in the vicinity of the predicted center of gravity position is searched, and the center positions G1d to G3d of the next garbage bags are set (FIG. 5C), and the falling trajectories of the garbage bags G1 to G3 are set. (C-8) The above steps (c-4) to (c-7) are repeated to set the falling trajectory of the garbage bags G1 to G3 from the insertion port 1 to the inside of the pit 2.

(D) Estimation of the amount of waste accumulated in the pit and the distribution state A of the waste quality The inside of the pit is divided into predetermined areas, and (a) input information, (b) instantaneous waste information and (c ) Based on the fall trajectory of the waste, the amount of accumulated waste and the distribution state A of the waste quality are estimated. That is, by tracking a fall trajectory having a predetermined spread of the waste G from the inlet 1 to the pit 2, the planar (two-dimensional) distribution of the waste G in the pit 2 and the accumulation time are accumulated. A three-dimensional (three-dimensional) distribution can be estimated.
(D-1) The planar position in the pit 2 can be estimated from the instantaneous waste information and its fall trajectory for each garbage bag body of waste G from the collection vehicle C or various single garbages. As a result, an instantaneous two-dimensional distribution state of the waste G can be obtained.
(D-2) By accumulating the two-dimensional distribution state sequentially corresponding to the elapsed time, the three-dimensional distribution state of the accumulated waste G can be obtained. In other words, the deposition position, deposition height, and deposition spread of each waste G can be obtained, and the accumulated waste G quantity and quality distribution state A can be estimated. Elements such as the volume, specific gravity, type, etc., of the input waste constitute the distribution state A.

  At this time, the binarization process, the mask process, and the noise are performed on the image information (distribution information on the surface of the waste G) of the waste G accumulated in the pit 2 at a specific time, as in (b) above. The removal process and the labeling process are performed, and the quantity and quality distribution state B of the accumulated combustion object can be estimated. Further, the distribution of the accumulated waste G can be obtained from a difference image of still images before and after the introduction of the waste G. When the dispersion state B at this time is compared with the distribution state A at the specific time, when the difference in quantity or / and quality between the divided regions is out of a predetermined range, It is preferable to correct the fall trajectory and estimate the corrected distribution state A. In other words, it is desirable that the two coincide with each other ideally, but in this processing system, since several estimation processes are performed based on the input information and the image information, each information has a predetermined ratio. There are uncertainties, which causes an error in the estimation results. For example, the specific gravity set as the waste quality in the input information may actually differ from the waste in the garbage bag, and the estimated fall trajectory may be displaced, resulting in a deviation in the distribution state. Alternatively, the accumulation position may move by colliding with garbage already in the pit 2 and its distribution state may change. Moreover, the difference from the waste quality in the bag actually becomes an error in the set calorific value. In this processing system, a distribution state closer to the actual situation can be estimated by correcting the influence of such fluctuations and errors from image information at a specific time in the pit.

(E) Agitation of waste in the pit In the quality distribution state A, when the difference between the divided areas is outside the predetermined range, the agitator 4 is driven to agitate the waste G in the pit 2. To do. That is, as described above, in addition to the difference in the quality of the input waste G itself, a non-uniform distribution state in the pit 2 due to the difference in the input place and the spread of the input waste G is formed. At this time, the variation in quality in the pit 2 becomes a variation in the heat generation amount of the waste G supplied to the combustion furnace, and stable combustion conditions cannot be ensured. Therefore, for example, as a dust agitation movement model, a three-dimensional average heat generation amount is estimated for each divided region in the pit 2, and the difference (variation) in the average heat generation amount between the regions is within a predetermined range ( For example, when the distribution state A becomes such a non-uniform state, a driving signal is issued from the arithmetic operation unit 7 to the stirrer 4. Specifically, if the difference between the average calorific value (estimated calorific value per unit volume) of two regions (for example, s3-t2 and s9-t5 in FIG. 6B) exceeds 200 kcal, the quality is not good. Since the uniformity is high, the stirrer 4 is moved to an intermediate position between the two so as to stir a wide area including the two regions to achieve uniformity. Image information from the image monitor 3 in a wide area including two regions is input to the operation unit 7 after performing a uniform operation for a predetermined time or while stirring, and the quality is uniform from the color distribution and particle size distribution on the surface of the deposited layer The quality uniformity can be ensured by estimating the property. However, in special cases where high calorific value waste is required due to the combustion conditions of the combustion furnace, for example, a non-uniform distribution of quality is maintained and high calorific value waste is collected with a crane. It is also possible to supply to the combustion furnace. In this processing system, it is also possible to perform such flexible processing operations.

(F) Transfer of waste in the pit to the combustion furnace In the distribution state A of the amount of accumulated waste G or the amount and quality of a specific region within a predetermined range, the crane 5 and The hopper 6 is driven and the waste G is transferred to the combustion furnace. “When the quantity and quality of a specific area are within a predetermined range” is the reason why there is a quantitative variation between the areas in the pit 2 due to the difference in the input place and the spread of the input waste G. And uneven distribution may be formed. Deposits made of materials with less stickiness will be gradually moved from a region with a high deposition height to a smaller region as the amount of deposition increases, but materials such as waste G will expand on the contrary. There is. At this time, if the uniformity of quality is ensured, the non-uniformity in quantity will not be harmful to the combustion furnace if it is transferred in sequence, so gradually from the region where the deposition height is large without performing the stirring operation. Further, by collecting with the crane 5, the uniformity of the amount can be gradually ensured, and the work amount can be minimized. In other words, the distribution state of the amount of waste G and the distribution state of quality do not necessarily coincide, but for the purpose of the present processing system to ensure stable combustion in the combustion furnace, quality uniformity is mainly used. It is preferable to perform stirring control. Even if there is quality uniformity, even if there is no quantity non-uniformity, the quantity is made uniform by preferential transfer from the area where the quantity and quality are within a predetermined range without performing a stirring operation. Thus, it is possible to substantially secure the sex.

[About uniform waste in the pit]
In the present processing system, the distribution state of the waste G in the pit 2 is grasped as described above, and it is based on uniforming using the stirring means corresponding to the non-uniform distribution state. Here, the distribution state of the waste G in the pit 2 can be estimated as instantaneous distribution state information. Further, the input information can be obtained before the garbage is input from the collection vehicle C. Furthermore, since the correspondence relationship between the part in the pit 2 and the inlet 1 which are in a non-uniform distribution state can be estimated from the fall trajectory of the waste G, it is estimated as distribution state information in order to make the part uniform. By reversing the operation process to be performed, it is possible to specify the input port 1 into which the waste G is input next. Therefore, by introducing the waste G of the collection vehicle C at a specific time from the specific input port 1, it is possible to ensure the uniformity of the best distribution state of the waste G in the pit 2 at that moment. In other words, the present processing system has made it possible to perform an unprecedented waste treatment in which the homogenization process using the stirring means can be performed with a minimum amount of energy.

As described above, the following excellent technical effects can be obtained by using the processing system and the processing method according to the present invention.
(I) Waste information (garbage type, density, calorific value, distribution, etc.) can be obtained online.
(Ii) A non-uniform area in the pit is detected, an index indicating whether or not the stirring process is necessary can be obtained, and waste can be made uniform efficiently.
(Iii) Waste can be made uniform efficiently, reducing the number of crane operations and the operation time, leading to cost reduction.
(Iv) Since it is possible to grasp the quality of input waste, which has been difficult until now, it is possible to supply stable waste waste that can be reflected in combustion control, and to stabilize the combustion furnace. Combustion operation can be realized.
(V) Information on the type and amount of waste loaded on the collection vehicle can be collected, which can be used for waste carry-in management.
(Vi) However, it is also possible to maintain a non-uniform distribution state of quality, for example, to selectively collect high calorific waste with a crane and supply it to the combustion furnace. It is also possible to perform such flexible processing operations.

<Combustion furnace combustion control system using this treatment system>
A combustion control system for a combustion furnace according to the present invention (hereinafter referred to as “the present combustion system”) receives, as input, waste (combustion target) and auxiliary combustion air supplied from the present processing system to a stoker-type waste incinerator. The amount of heat generated, exhaust gas, and dust ash are output. In addition, any one of the quantity of combustion object to be input into the combustion furnace, the quality of the combustion object, the amount of air, the temperature of the air, and the stoker speed is controlled, and the temperature, gas concentration, gas flow direction, gas in the furnace In addition to either the flow rate or the evaporation amount, the heat generation amount of the combustion object estimated in the present processing system is controlled as a control index. At this time, since the heat generation amount is uniformly controlled in the present processing system, the combustion conditions in the combustion furnace for the supplied combustion object can ensure very high stability.

[Configuration of this combustion system using this treatment system]
FIG. 7 shows a schematic overall configuration example of the present combustion system. In the incinerator 20, a receiving hopper 21 that receives the waste G supplied from the processing system 10 and a stoker 23 that burns the waste are provided in the furnace body 22. The waste in the receiving hopper 21 is transferred to the stoker 23 by the waste supply device 24. Sent. The stalker 23 is composed of a combustion stalker 23a and a post-combustion stalker 23b, and is separately driven to reciprocate to feed garbage. The furnace body 22 includes a primary combustion zone 22a provided at the upper part of the stoker 23, a secondary combustion zone 22b at the upper part thereof, a primary combustion air supply device 25 for supplying air to the stoker 23 and the primary combustion zone 22a, A secondary combustion air supply device 26 that supplies secondary combustion air to the secondary combustion zone 22b, an ash discharge part 27 that discharges dust ash, and an exhaust gas discharge part 28 that discharges exhaust gas in the furnace are provided. The exhaust gas from the exhaust gas discharge unit 28 is discharged from the chimney 30 after dust is separated by the bag filter 29. The dust ash from the ash discharge unit 27 is separated and stored in the dust storage unit 31 together with the dust from the bag filter 29, the chemical solution such as the heavy metal stabilizer from the chemical solution supply unit 32, and the makeup water from the water supply unit 33. Processed by.

The garbage thrown into the receiving hopper 21 is primarily burned by the primary combustion air while being sent in the order of the combustion stalker 23a and the post-combustion stalker 23b. In the combustion stoker 23a, garbage is mainly combusted by the primary combustion air. The primary combustion air supplied to the combustion stoker 23a is an amount necessary and sufficient for the combustion of the waste, and the gas generated from the waste on the combustion stoker 23a contains a high concentration of NOx. In the post-combustion stoker 23b, unburned content in the incineration ash is burned out. In the secondary combustion zone 22b, secondary combustion air is supplied to the lower, middle, and upper portions thereof to completely burn the unburned or incompletely burned products from the primary combustion zone 2a. The dust ash generated by the combustion is discharged from the ash discharge unit 27, and the exhaust gas in the furnace is discharged from the exhaust gas discharge unit 28. At this time, CO, NOx, oxygen (O ₂ ), and the like in the exhaust gas are managed so as to be below a predetermined concentration.

The control mechanism of the combustion system includes a dust supply device 24 that controls the input amount of dust, a stalker drive device (not shown) that controls and operates the stalker speed, a primary combustion air supply device 25, and a secondary combustion air supply device 26. A primary combustion air preheater (not shown) that controls the temperature of the primary combustion air, and a secondary combustion air preheater that controls the temperature of the secondary combustion air. In addition, as a sensor unit (not shown) for measuring the amount and quality of the waste put into the receiving hopper 21 and a sensor unit for detecting the combustion state in the incinerator 20, for example, a NOx concentration meter, an O ₂ concentration meter, a CO 2 A densitometer is provided in at least one of the secondary combustion zone 22b and the primary combustion zone 22a (not shown). Furthermore, a temperature sensor for detecting the temperature distribution and a sensor for measuring the amount of evaporation corresponding to the amount of energy accompanying combustion are provided. Each detection signal (detection information) from these sensor units is input as a process data to a control device (not shown).

In the present combustion system, the above-described components are combined to process / manipulate a plurality of process data and manipulated variables. Specifically, based on process data such as evaporation amount, quantity and quality of combustion object, gas concentration, gas flow rate, furnace temperature, etc., burnout point position, boiler evaporation amount, dust layer level, exhaust gas O ₂ concentration The primary air amount, the secondary air amount, the combustion chamber outlet temperature, etc. are calculated, and these data are taken into the control device, and the optimal feed rate so that the boiler evaporation amount and the excess air ratio become the set values, Operation amounts such as the stalker speed and the air damper opening are calculated.

  In addition to such a normal control mechanism, the control mechanism of the combustion system is connected to the arithmetic processing unit 7 of the processing system 10 and can be a part of the control function of the combustion system. Specifically, as a signal from the arithmetic processing unit 7, information on the total amount (weight) and quality (heat generation amount) of dust stored in the pit 2 and information on the distribution state A is input to the control device. Based on such input, the control device calculates an operation amount for controlling the control mechanism and transmits / receives a control signal to / from the control mechanism. From the control device, the operation amount thus calculated is transmitted to the control mechanism, and the control operation is performed.

  The control system is based on a PID control system that performs stabilization control on instantaneous data, and as an upper control, an autoregressive model control system (AR control) that performs dynamic optimization control on instantaneous data fluctuations. System) and a fuzzy control system that performs steady-state return control on instantaneous data that deviates from a preset control range or its fluctuation amount, and it is preferable to use the functions of each control system in an effective connection. . It is possible to perform control in an optimal combustion state that is closer to the design conditions in the actual plant operating state, and at the same time, it is possible to perform real-time optimal combustion control in the actual operating state.

  At this time, the supply amount of the waste G, which is normally agitated and the waste quality is made uniform, is set, and the crane 5 and the hopper 6 are driven via the arithmetic processing unit 7 to control the transfer of the waste G in the pit 2. And stable combustion conditions can be secured under optimum conditions. However, as described above, the waste G in the pit 2 which restricts the stirring control of the present processing system so as to selectively supply, for example, waste having a high calorific value as needed, is unevenly distributed. It is also possible to control to maintain the state A. Further, based on the vehicle information (input information) of the collection vehicle C, the waste G loaded on the specific collection vehicle C before being input to the pit 2 is preferentially input, and a desired supply amount or a calorific value is set. It is also possible to control the waste G to be supplied to the combustion furnace.

  The treatment furnace to which the treatment system according to the present invention and the present combustion system using the treatment system can be applied is not limited to an incinerator, and may be an electric ash melting furnace, a gasification melting furnace, or the like. Moreover, although the automatic combustion control apparatus of the stoker type | mold waste incinerator is made into the structural example above, waste incinerators other than the stoker type, an electric ash melting furnace, a gasification melting furnace, etc. may be used.

1 Input port 2 Pit 3 Image monitor 4 Stirrer (stirring means)
5 Crane (transportation means)
6 Hopper (transportation means)
7 Arithmetic processing part C Garbage truck G Waste (combustion target)

Claims (5)

A pit for storing a combustion object to be supplied to the combustion furnace, an inlet for introducing the combustion object into the pit, a stirring means for stirring the stored combustion object, and the stored combustion object for the combustion furnace A combustion object processing system to be supplied to a combustion furnace, comprising a transfer means for transferring to a combustion furnace, and an image monitor for photographing the accumulated combustion object from the inlet to the pit,
(A) Input the input information including the whole combustion object to be put into the pit or individual quantity and quality, and image information from the image monitor,
(B) Binarization processing, mask processing, noise removal processing, and labeling processing are performed on the instantaneous image information in the vicinity of the input port, and individual combustion objects that have been input are individually determined based on the input information. Estimate instantaneous combustion object information including size or quantity and quality,
(C) detecting the position of the identified combustion object on the basis of image information for a predetermined period of time along the path from the inlet to the pit, and estimating the fall trajectory of the combustion object;
(D) The inside of the pit is divided into predetermined areas, and the quantity and quality distribution state A of the accumulated combustion object is determined based on the input information, instantaneous combustion object information, and the fall trajectory of the combustion object. Estimate
(E) In the quality distribution state A, when the difference between the divided areas is out of a predetermined range, the stirring means is driven to stir the combustion object in the pit,
(F) In the distribution state A of the quantity, when the quantity of the accumulated combustion object or the quantity and quality of a specific region are within a predetermined range, the transfer means is driven to burn the combustion object A processing system for an object to be combusted supplied to a combustion furnace, comprising an arithmetic processing unit for transferring to the furnace.   A binarization process, a mask process, a noise removal process, and a labeling process are performed on the image information at a specific time in the pit, and the quantity and quality distribution state B of the accumulated combustion object are estimated, and the specific time When the difference in quantity or / and quality between the divided areas is out of a predetermined range as compared with the distribution state A in FIG. 3, the fall trajectory of the combustion object is corrected, and the corrected distribution The state A is estimated, and the combustion object processing system supplied to the combustion furnace according to claim 1. A method of processing a combustion object that is stored in a pit in advance and then supplied to a combustion furnace, the entire combustion object that is input into the pit, or input information including individual quantity and quality, and Based on the image information of the burning object taken from the inlet to the pit,
(B) Binarization processing, mask processing, noise removal processing, and labeling processing are performed on the instantaneous image information in the vicinity of the input port, and individual combustion objects that have been input are individually determined based on the input information. Estimate instantaneous combustion object information including size or quantity and quality,
(C) detecting the position of the identified combustion object on the basis of image information for a predetermined period of time along the path from the inlet to the pit, and estimating the fall trajectory of the combustion object;
(D) The inside of the pit is divided into predetermined areas, and based on the input information, instantaneous combustion object information, and the fall trajectory of the combustion object, the accumulated quantity and quality distribution state A of the combustion object is determined. Estimate
(E) In the distribution state A of the quality, when the difference between the divided areas is out of the predetermined range, the combustion object in the pit is stirred,
(F) In the distribution state A of the quantity, the combustion object is transferred to a combustion furnace when the accumulated combustion object or the quantity and quality of a specific region are within a predetermined range. A method for treating a combustion object supplied to a combustion furnace.   A binarization process, a mask process, a noise removal process, and a labeling process are performed on the image information at a specific time in the pit, and the quantity and quality distribution state B of the accumulated combustion object are estimated, and the specific time When the difference in quantity or / and quality between the divided areas is out of a predetermined range as compared with the distribution state A in FIG. 3, the fall trajectory of the combustion object is corrected, and the corrected distribution The state A is estimated, The processing method of the combustion target object supplied to a combustion furnace of Claim 3 characterized by the above-mentioned.   It is a combustion control system of the combustion furnace using the processing system or processing method of the combustion target object supplied to the combustion furnace in any one of Claims 1-4, Comprising: With respect to the combustion target object accumulated in the pit Based on the quantity and quality distribution state A estimated in the present processing system or method, the calorific value of the combustion object supplied to the combustion furnace is estimated, and one of the combustion control indices of the combustion furnace A combustion control system for a combustion furnace.

Images