JP2022161090A - Waste information prediction device, combustion control device of incinerator, waste information prediction method, learning method for waste information prediction model, and waste information predication model program - Google Patents

Abstract

To provide a waste information prediction device capable of accurately predicting information on wastes in an incinerator, a combustion control device of an incinerator, a waste information prediction method, a learning method for a waste information prediction model, and a waste information prediction model program.SOLUTION: A waste information prediction device comprises: an input unit capable of inputting captured image information of waste captured outside; feature amount calculation means of dividing the captured image information of the waste input from the input unit into a plurality of blocks, tracking at least one of the plurality of divided blocks for a predetermined time, and calculating a feature amount of the tracked block; a storage unit that stores a relational model that associates the feature amount of the waste with information on the waste; and control means. The control means calculates the feature amount of the waste by the feature amount calculation means based on the captured image information obtained by capturing the waste in an incinerator, which is input from the input unit; and predicts the information on the waste based on the relational model stored in the storage unit.SELECTED DRAWING: Figure 4

Description

The present invention relates to a waste information prediction device, an incinerator combustion control device, a waste information prediction method, a waste information prediction model learning method, and a waste information prediction model program.

In a waste treatment plant such as an incinerator, it is very effective to grasp information (for example, calorie, etc.) of waste to be fed in advance in order to stably burn the waste in the furnace.

As a technology for predicting information about waste, for example, Patent Document 1 describes a prediction model that predicts information about waste by correlating time-series images obtained by imaging the combustion state of waste using optical flow. Techniques for learning are disclosed. In addition, in Patent Document 2, based on a two-dimensional image showing the combustion temperature distribution of the waste near the dry zone and recording only the wavelength in the 3 to 4 μm band, the burning point of the waste on the dry zone A technique for detecting is disclosed.

Japanese Patent Application Laid-Open No. 2020-119407 JP-A-8-49830 JP-A-6-174219

However, the technologies disclosed in Patent Documents 1 and 2 do not extract feature values that are correlated with the state of the waste in the incinerator, so it is difficult to accurately predict information about the waste. . Moreover, even if such low-accuracy feature values are used, it is not possible to efficiently control a waste treatment plant such as an incinerator.

Further, Patent Document 3 discloses a technique of estimating the water content of waste based on the temperature distribution of the cross section in the thickness direction of the waste before the waste is transported to the combustion zone. However, in Patent Document 3, it is necessary to grasp the temperature distribution of the cross section in the thickness direction, and temperature information of a relatively wide area of the dust is required. Therefore, it takes time to grasp the temperature distribution, and since the temperature distribution depends on the retention time in the furnace before the dust falls, the accuracy of grasping the dust quality is not very high.

The present invention has been made in view of the above, and includes a waste information prediction device, an incinerator combustion control device, a waste information prediction method, and a waste information prediction device capable of accurately predicting information on waste in an incinerator. An object of the present invention is to provide a waste information prediction model learning method and a waste information prediction model program.

In order to solve the above-described problems and achieve the object, a waste information prediction device according to the present invention is a waste information prediction device for predicting information about waste in an incinerator, and an input unit capable of inputting photographic image information of the waste; dividing the photographic image information of the waste input from the input unit into a plurality of blocks; feature quantity calculation means for tracking for a predetermined period of time and calculating the feature quantity of the traced block; a storage unit storing a relationship model that associates the feature quantity of the waste with the information on the waste; a control means; and the control means calculates the feature amount of the waste by the feature amount calculation means from the photographed image information obtained by photographing the waste in the incinerator, which is input from the input unit. and predicting the information on the waste based on the relationship model stored in the storage unit.

Further, in the waste information prediction apparatus according to the present invention, in the above invention, the feature amount is a temperature change amount of the waste, a movement amount of the waste, and a division of the waste set in the captured image information. at least one of the total number of a plurality of blocks formed.

Further, in the waste information prediction device according to the present invention, in the above invention, the information on the waste includes the calorie generated when the waste is burned, the amount of steam generated from the boiler of the incinerator when the waste is burned. characterized by including at least one of

Further, the waste information prediction device according to the present invention is characterized in that, in the above invention, it has an output unit for outputting the information related to the waste predicted by the control means.

Further, in the waste information prediction device according to the present invention, in the above invention, the photographed image information of the waste input from the input unit is pre-supply waste before being supplied to the grate of the incinerator. It is characterized by being photographed image information obtained by photographing.

Further, in the waste information prediction device according to the present invention, in the above invention, the relationship model stored in the storage unit learns from the relationship between the feature amount of past waste and the information on the past waste. It is characterized by being a learned waste information prediction model obtained by

Further, in the waste information prediction device according to the present invention, in the above invention, the relationship model stored in the storage unit uses the characteristic amount of past waste as an input value and the information on the past waste as an output value. It is characterized by being a lookup table or a predetermined function.

In order to solve the above-described problems and achieve the object, a combustion control device for an incinerator according to the present invention is a combustion control device comprising a combustion control unit for controlling an incinerator for burning waste, the combustion control device comprising: The control unit is characterized by controlling combustion in the incinerator based on the information about the waste output from the waste information prediction device.

Further, in the incinerator combustion control device according to the present invention, in the above invention, the incinerator includes a fire grate for moving the waste, a waste supply device for supplying the waste onto the fire grate, a blower for blowing air into the incinerator, wherein the combustion control unit controls the movement speed of the waste on the fire grate, the disposal by the waste supply device, and At least one of the article supply speed, the amount of air blown by the blower, and the temperature of the air by the blower is controlled.

In order to solve the above-described problems and achieve the object, a waste information prediction method according to the present invention is a waste information prediction method for predicting information about waste in an incinerator, an image acquisition step of acquiring photographed image information of the waste to be processed; and dividing the photographed image information of the waste acquired in the image acquisition step into a plurality of blocks, and at least one of the plurality of divided blocks and a waste information prediction step of predicting information about the waste from the feature amount of the waste calculated in the feature amount calculation step. , is characterized by having

Further, in the waste information prediction method according to the present invention, in the above invention, the waste information prediction step predicts using a relationship model in which the characteristic amount of the waste and the information about the waste are associated in advance. It is characterized by

Further, in the waste information prediction method according to the present invention, in the above invention, the relationship model is a learned model obtained by learning from the relationship between the feature amount of the past waste and the information about the past waste. It is characterized by being a waste information prediction model.

Further, in the waste information prediction method according to the present invention, in the above invention, the relationship model is a lookup table or a It is characterized by being a predetermined function.

Further, in the waste information prediction method according to the present invention, in the above invention, the photographed image information of the waste acquired in the image acquisition step is the pre-supply waste before being supplied to the grate of the incinerator. is image information obtained by photographing the .

Further, the waste information prediction method according to the present invention further comprises a combustion control step of controlling combustion of the incinerator based on the waste information predicted in the waste information prediction step. Characterized by

In order to solve the above-described problems and achieve the object, a waste information prediction model learning method according to the present invention is a waste information prediction model learning method for predicting information about waste in an incinerator. , dividing photographed image information of waste in an incinerator for incinerating waste into a plurality of blocks, tracking at least one of the plurality of divided blocks, and inputting a feature value calculated from the tracking block and making the waste information prediction model learn the relationship between the past feature amount and the past information on the waste, using the information on the waste as an output value.

In order to solve the above-described problems and achieve the object, a waste information prediction model program according to the present invention is a waste information prediction model program that can be recorded on a recording medium and predicts information about waste in an incinerator. and learned to output information about the waste when a feature amount calculated from a tracking block that tracks a plurality of blocks set in the captured image information of the waste is input. It is characterized by

According to the waste information prediction device, the incinerator combustion control device, the waste information prediction method, the waste information prediction model learning method, and the waste information prediction model program according to the present invention, the movement of the waste in the incinerator is tracked for each of a number of preset blocks, and the prediction model learns the feature values of each block. This makes it possible to efficiently control a waste treatment plant such as an incinerator.

FIG. 1 is a diagram schematically showing the overall configuration of an incinerator to which a waste information prediction device according to an embodiment of the present invention is applied. FIG. 2 is a side view showing waste, a portion for feeding waste onto a fire grate, and an imaging unit in the incinerator according to the embodiment of the present invention. FIG. 3 is a front view showing the field of view of the imaging section in the incinerator according to the embodiment of the present invention. FIG. 4 is a block diagram showing the configuration of an incinerator combustion control device and a waste information prediction device according to an embodiment of the present invention. FIG. 5 is a diagram showing an example of captured image information captured by the imaging unit in the incinerator according to the embodiment of the present invention. FIG. 6 is a graph summarizing changes in the average temperature of each area of the captured image information, changes in the total average temperature in each area, and changes in the calorie of the pre-supply waste related to the captured image information. FIG. 7 is a diagram showing an example in which an optical flow based on a block matching method is applied to actual photographed image information photographed by an imaging unit in the incinerator according to the embodiment of the present invention. FIG. 8 is a diagram showing an example in which an optical flow by a block matching method is applied to actual captured image information captured by the imaging unit in the incinerator according to the embodiment of the present invention. FIG. 9 is a diagram showing an example in which an optical flow based on a block matching method is applied to actual photographed image information photographed by the imaging unit in the incinerator according to the embodiment of the present invention. FIG. 10 is a diagram for explaining a method of calculating an average temperature rise amount per block as a feature amount in the waste information prediction device according to the embodiment of the present invention. FIG. 11 is a diagram for explaining a method of calculating the movement amount of waste for each block as a feature amount in the waste information prediction device according to the embodiment of the present invention. FIG. 12 is a flow chart showing the processing procedure of the waste information prediction model learning method according to the embodiment of the present invention. FIG. 13 is a flow chart showing the processing procedure of the waste information prediction method according to the embodiment of the present invention.

A waste information prediction device, an incinerator combustion control device, a waste information prediction method, a waste information prediction model learning method, and a waste information prediction model program according to embodiments of the present invention will be described with reference to the drawings. . In addition, in the drawings referred to in the following embodiments, the same reference numerals are given to the same or corresponding parts. Moreover, the present invention is not limited by the following embodiments.

Here, for example, in order to efficiently control a grate (stoker) type incinerator, etc., it is important to grasp information on the waste that is burned in the incinerator (hereinafter referred to as "waste information"). be. Therefore, the present inventor tracks the movement of waste in an incinerator for each of a plurality of preset blocks, makes a prediction model learn the feature values of each block, and uses the prediction model to obtain waste information. Predict (estimate). The embodiments described below have been devised based on the above earnest studies of the inventors.

(Incinerator)
An incinerator to which a waste information prediction device according to an embodiment of the present invention is applied will be described with reference to FIG. The incinerator is, for example, a grate-type incinerator, and is installed in a plant or the like that generates power using waste. The incinerator comprises a furnace 1 in which waste is burned, a waste inlet 2 for introducing waste, a waste supply device 3, a grate 4, an ash drop port 5, and air in the furnace 1. A blower (combustion air blower 6, secondary air blower 11) for blowing, a furnace outlet 7, a chimney 8, a boiler 9 equipped with a heat exchanger 9a and a steam drum 9b, a secondary air inlet 10, It has

A combustion air damper 14 is provided in the immediate vicinity of the combustion air blower 6 . In addition, air dampers 14a, 14b, 14c, and 14d for under-grate combustion are provided in the wind box under the grate 4. As shown in FIG. A secondary air damper 15 is provided in the immediate vicinity of the secondary air blower 11 . An intermediate ceiling 16 is provided in the furnace 1 at an upper position along the height direction of the furnace 1 .

Further, in the furnace 1, there are a combustion chamber gas thermometer 17, a main flue gas thermometer 18, a furnace outlet lower gas thermometer 19, a furnace outlet central gas thermometer 20, and a furnace outlet gas thermometer 21. is provided. The boiler 9 is also provided with a boiler outlet oxygen concentration meter 22 and a steam flow meter 25 . A gas concentration meter 23 is provided at the entrance of the chimney 8 . Further, an exhaust gas flow meter 24 is provided on the pipe connecting the outlet of the boiler 9 and the chimney 8 . The measured values measured by these instruments are transmitted to the combustion control device 30 as combustion process measured values and stored in the storage unit 32 (see FIG. 4).

An imaging unit 26 is provided downstream of the furnace 1 in the direction in which the waste is transported. The imaging unit 26 includes a flame transmission camera, for example, an infrared camera, and an image processing unit that processes captured image data.

An installation state of the imaging unit 26 will be described with reference to FIG. As shown in the figure, the imaging unit 26 is arranged outside the furnace 1 and close to a monitoring window provided on the furnace wall 1a. In addition, the imaging unit 26 may be arranged in the furnace 1 by adding a water cooling structure, for example. Waste 50 falls from the waste supply 12 onto the grate 4 at the step wall 13 . The waste 50 that has fallen onto the fire grate 4 is stirred by the reciprocating motion that accompanies the back-and-forth movement of the fire grate 4, and is moved forward, which is toward the imaging unit 26 side.

The imaging unit 26 acquires thermography information of the waste 50 on the grate 4 (hereinafter referred to as “waste on the grate 52”) as captured image information (thermal image information). Here, the wavelength of infrared rays emitted from waste 50 is different from the wavelength of infrared rays emitted from hot gases and flames in space. Therefore, in the image capturing unit 26, by appropriately selecting the infrared wavelength to be measured, captured image information corresponding to the temperature distribution of the layer of the waste on the grate 52 can be obtained even if a flame exists within the measurement field of view. Obtainable. In addition, by setting the measurement range in the furnace length direction, the imaging unit 26 can obtain photographed image information of the layer of the waste on the grate 52 at a position upstream of the combustion area (upstream of the flame). . The photographed image information can be treated as image data in a state through which flames are transmitted, that is, as a plurality of image data.

In other words, the imaging unit 26 captures the waste 50 sent out from the waste supply unit 12 (hereinafter referred to as "pre-supply waste 51"), the stepped wall 13 having a step on which the waste 50 falls, and the grate. The waste 52 and the upper surface of the grate 4 can be imaged in flame transmission. In addition, in the incinerator, a combustion image capturing unit that captures the combustion state of the grate waste 52, that is, the flame itself, may be further provided. The captured image information captured by the imaging unit 26 in a state in which the flame is captured is transmitted to the waste information prediction device 40 immediately or at predetermined time intervals. It should be noted that the captured image information captured by the imaging unit 26 may be stored in the storage unit 32 of the combustion control device 30 and then transmitted from the combustion control device 30 to the waste information prediction device 40 .

The imaging unit 26 is installed at a position substantially facing the waste supply unit 12 and the stepped wall 13, for example. Note that the installation of the imaging unit 26 is not limited to a position substantially facing the waste supply unit 12 and the stepped wall 13 .

An example of the field of view of the imaging section 26 will be described with reference to FIG. As shown in the figure, the imaging unit 26 has a measurement field of view extending, for example, in the vertical direction of the furnace 1 and in the furnace width direction (horizontal direction). The visual field of the imaging unit 26 is the waste supply unit 12, the stepped wall 13, the fire grate 4, and the left and right furnace walls 1a. In addition, the furnace wall 1a included in the field of view of the imaging unit 26 restricts the outward movement of the waste 50 in the left-right direction, that is, the spread. In addition, it is preferable that the imaging unit 26 be configured so as to be able to capture an image of the waste 50 transported to the waste supply unit 12 . Thereby, the image of the waste 50 falling at the position of the stepped wall 13 can be captured.

The configurations of the combustion control device 30 and the waste information prediction device 40 will be described with reference to FIG. The combustion control device 30 and the waste information prediction device 40 are connected to, for example, a dedicated line, a public communication network such as the Internet (for example, a LAN (Local Area Network), a WAN (Wide Area Network), a telephone communication network such as a mobile phone, a public line). , VPN (Virtual Private Network), etc. are connected via a network (not shown) consisting of a combination of one or more.

The combustion control device 30 and the waste information prediction device 40 may be configured separately as shown in FIG. 4, or may be configured integrally. Also, the combustion control device 30 and the waste information prediction device 40 may be installed in the same facility as the incinerator, or may be installed in a separate facility from the incinerator. When the combustion control device 30 and the waste information prediction device 40 are installed in different facilities, communication of various information and various data is performed via the network described above.

The combustion control device 30 includes a control section 31 , a storage section 32 and an operation amount adjustment section 33 . The control unit 31 and the operation amount adjustment unit 33 function as a combustion control unit that controls the incinerator. Specifically, the control unit 31 and the operation amount adjustment unit 33 include processors such as a CPU (Central Processing Unit), a DSP (Digital Signal Processor), and an FPGA (Field-Programmable Gate Array) having hardware, and a RAM (Random and a main storage unit such as an Access Memory) and a ROM (Read Only Memory) (none of which are shown).

The storage unit 32 includes storage media such as volatile memory such as RAM, nonvolatile memory such as ROM, EPROM (Erasable Programmable ROM), hard disk drive (HDD), and removable media. Examples of removable media include disk recording media such as USB (Universal Serial Bus) memories, CDs (Compact Discs), DVDs (Digital Versatile Discs), and BDs (Blu-ray (registered trademark) Discs). Alternatively, the storage unit 32 may be configured using a computer-readable recording medium such as a memory card that can be attached from the outside.

The storage unit 32 can store an operating system (OS) for executing the operation of the combustion control device 30, various programs, various tables, various databases, and the like. Here, the various programs include an information processing program that realizes processing based on a model such as a learning model or a trained model according to this embodiment. These various programs can be recorded on computer-readable recording media such as hard disks, flash memories, CD-ROMs, DVD-ROMs, flexible disks, etc., and can be widely distributed.

The combustion control device 30 performs a combustion control process for controlling the combustion of the waste 50 in the incinerator based on the waste information output from the waste information prediction device 40 as the manipulated variable of each operation terminal. Specifically, based on the waste information, the combustion control device 30 controls the movement speed of the waste 50 on the grate 4 (also referred to as the “grate feeding speed”), the speed of the waste 50 by the waste supply device 3 At least one of the supply speed (also referred to as the “waste feeder feed speed”), the amount and temperature of the combustion air blown by the combustion air blower 6, and the amount and temperature of the secondary air blown by the secondary air blower 11. to control.

In addition to the waste information acquired from the waste information prediction device 40, the combustion control device 30 considers a predetermined manipulated variable reference value setting relational expression (hereinafter referred to as "manipulated variable relational expression"). , may control the above manipulated variables. This manipulated variable relational expression is, for example, a relational expression between the waste incineration amount set value and the manipulated variable reference value (target value of the manipulated variable), and includes a control parameter as a correction coefficient. The control parameters are adjusted by the control unit 31 so as to match the waste incineration amount set value. The adjusted control parameter is changed by the control unit 31 corresponding to the changed setting value when the waste incineration amount setting value is changed. By changing the control parameter, the preset manipulated variable reference value is corrected.

The control unit 31 adjusts the manipulated variable reference value based on the waste information acquired from the waste information prediction device 40 and the adjustment of the control parameter in the manipulated variable relational expression that is considered as necessary. The control unit 31 corrects the adjusted manipulated variable reference value based on a predetermined control algorithm such as PID control or fuzzy calculation. The storage unit 32 stores data referred to by the control unit 31 . The storage unit 32 stores, for example, a predetermined manipulated variable relational expression, a control algorithm, a preset waste incineration amount setting value, a combustion process measurement value acquired as a combustion state quantity in the furnace 1, and the like. there is

The operation amount adjustment unit 33 adjusts the operation amount of each operation end so as to follow the operation amount reference value. Specifically, the operation amount adjustment unit 33 includes a combustion air adjustment unit 331, an air amount ratio adjustment unit 332, a secondary air adjustment unit 333, a waste supply device feed speed adjustment unit 334, and a grate feed. A speed adjustment unit 335 is provided.

The combustion air adjustment unit 331 adjusts the manipulated variable so that the blowing amount and temperature of the combustion air follow the manipulated variable reference value corrected by the control unit 31 (hereinafter referred to as the “corrected manipulated variable reference value”). do. Further, the air amount ratio adjusting section 332 controls each of the under-grate combustion air dampers 14a to 14d to adjust the mutual ratio of the flow rates in the respective wind boxes. In addition, the secondary air adjustment unit 333 adjusts the manipulated variable so that the volume and temperature of the secondary air follow the corrected manipulated variable reference value. Here, the amounts of combustion air and secondary air blown are adjusted by controlling the opening degrees of the combustion air damper 14, the under-grate combustion air dampers 14a to 14d, and the secondary air damper 15. be.

The waste supply device feed speed adjustment unit 334 adjusts the manipulated variable so that the waste feed device feed speed follows the corrected manipulated variable reference value. In addition, the grate feed speed adjustment unit 335 adjusts the manipulated variable so that the grate feed speed follows the corrected manipulated variable reference value. If the control unit 31 does not correct the operation amount reference value, the operation amount adjustment unit 33 adjusts each operation amount based on the operation amount reference value that has not been corrected.

The waste information prediction device 40 includes a control section (control means) 41 , an output section 42 , an input section 43 and a storage section 44 . The control unit 41 has the same functional and physical configuration as the control unit 31 described above, and includes a processor such as a CPU, DSP, and FPGA having hardware, and a main storage unit such as a RAM and a ROM. (also not shown). Specifically, the control unit 41 functions as a feature amount calculation unit (feature amount calculation means) 411 , a learning unit 412 , and a waste information prediction unit 413 .

The feature quantity calculation unit 411 calculates a feature quantity that characterizes the state of the waste 50 from the image information of the waste 50 in the furnace 1 . Specifically, the feature amount calculation unit 411 tracks the movement of the waste 50 in a plurality of time-continuous captured image information captured by the imaging unit 26, and performs learning input in the learning unit 412, which will be described later. Calculate the feature amount that will be the data. That is, the feature amount calculation unit 411 divides the photographed image information of the waste 50 input from the input unit 43 into a plurality of blocks, traces at least one of the plurality of divided blocks for a predetermined time, and traces the tracked block. is calculated.

The feature quantity calculator 411 captures an image of the waste 50 before it is supplied to the grate 4, that is, the pre-supply waste 51 located near the waste supply unit 12 (hereinafter also referred to as “dust inlet region”). The amount of temperature change of the pre-supply waste 51 is calculated as a feature value based on the photographed image information obtained by the above. This "temperature change amount" is, specifically, an average temperature rise amount per block set in the vicinity of the waste supply unit 12 of the photographed image information, as will be described later.

Here, FIG. 5 shows an example of captured image information inside the furnace 1 captured by the imaging unit 26 . The photographed image information shown in FIG. It was divided into a region D on the side, a region E on the left furnace wall 1a, and a region F on the right furnace wall 1a, and the amount of change in average temperature in each region A to F was calculated. Then, the transition of the average temperature of each area A to F, the transition of the total average temperature of each area A to F, and the transition of the calorie measured after burning the pre-supply waste 51 related to the photographed image information are shown. A summary is shown in the graph of FIG.

As shown in FIG. 6, among the regions A to F, region A, that is, the average temperature of the pre-supply waste 51 near the waste supply section 12 and the calorie have a relatively higher correlation than other regions. I know you are. Based on such new knowledge, the feature amount calculation unit 411 uses the temperature rise amount of the pre-supply waste 51 near the waste supply unit 12, which is highly correlated with the calorie of the pre-supply waste 51, as a feature amount. calculate.

Note that, as will be described later, the feature amount calculation unit 411 is set when the learning unit 412 learns the waste information prediction model 44a (see step S3 in FIG. 12) and when the waste information prediction unit 413 predicts the waste information. (see step S13 in FIG. 13), the feature amount of the waste 50 is calculated.

A specific calculation method of the feature amount in the feature amount calculation unit 411 will be described below with reference to FIGS. 7 to 9. FIG. As shown in the figure, the feature amount calculator 411 divides the captured image information into a plurality of blocks Bl, and tracks at least one of the plurality of divided blocks Bl over a plurality of frames. As a technique for tracking the movement of the waste 50, for example, an optical flow based on a block matching method can be used. In this method, an image is divided into blocks Bl of appropriate size, and a difference between image frames is detected to search for a destination for each block Bl.

As for the blocks to be divided, the entire obtained captured image information may be divided into blocks of the same size, or only a part of the area may be divided into blocks of the same size. Blocking and tracking of captured image information, for example, does not confuse specific waste (for example, the above-mentioned pre-supply waste before combustion 51) with other waste (burning waste, etc.). However, if the block size is too small, processing will take a long time, and if the block size is too large, the tracking accuracy of the block Bl will decrease. Therefore, it is desirable to set the block size with the best processing time and tracking accuracy based on, for example, the resolution of the captured image information.

For example, as shown in FIG. 7, the feature amount calculation unit 411 arranges a plurality of blocks B1 in an area near the waste supply unit 12 in the captured image information. These blocks Bl move along with the movement of the pre-supply waste 51, as shown in FIG. In addition, the dropped block that has fallen out of the tracking range set in the captured image information (the area near the waste supply unit 12 in the figure) is also deleted, and the tracking ends. If the pixel value difference of the block Bl between frames is equal to or less than a predetermined threshold value, it is determined that the pre-supply waste 51 included in the block Bl has fallen onto the fire grate 4, and tracking ends.

Subsequently, the feature amount calculation unit 411 analyzes the captured image information to calculate the amount of change in average temperature for a predetermined time for the tracked block Bl (tracking block), as shown in FIG. 10 . Subsequently, the sum of the average temperatures of the blocks Bl with positive average temperature changes (that is, the temperature is rising) is calculated. The block Bl having a positive average temperature change means that the pre-supply waste 51 included in the block Bl is likely to burn in the future.

Subsequently, the feature amount calculation unit 411 calculates the average temperature rise amount per block as a feature amount by dividing the total sum of the average temperatures of the blocks Bl having positive average temperature changes by the total number of the blocks Bl. In this way, by observing the temperature change of the pre-supply waste 51 tracked for each preset block Bl instead of the entire area near the waste supply unit 12 of the photographed image information, the pre-supply waste 51 moves. However, it is possible to accurately grasp the temperature change of the waste itself.

Here, the feature amount calculation unit 411 may calculate the movement amount of the pre-supply waste 51 for each block Bl as the feature amount, in addition to the temperature change amount of the pre-supply waste 51 for each block Bl described above. . In this case, the feature amount calculation unit 411 calculates by totalizing statistics such as the average and variance of the motion vectors of each block Bl. As shown in FIG. 11, when the amount of movement of the block Bl is large, it means that the amount of waste 50 put into the furnace 1 is large. Conversely, when the movement amount of the block Bl is small, it means that the amount of the waste 50 introduced into the furnace 1 is small. Therefore, by calculating the amount of movement of the pre-supply waste 51 for each block Bl as a feature amount, the amount of the waste 50 introduced into the furnace 1 can be grasped.

In addition to the temperature change amount of the pre-supply waste 51 for each block Bl and the movement amount of the pre-supply waste 51 for each block Bl described above, the feature amount calculation unit 411 calculates the pre-supply waste set in the captured image information. The total number of divided blocks Bl of the waste 51 may be calculated as the feature amount. In addition, since the number of dropped blocks that have fallen from a preset tracking range (for example, the area near the waste supply unit 12) in the captured image information indicates the amount of garbage that will be burned from now on, this is also calculated as a feature amount. good too.

The learning unit 412 uses the feature amount calculated by the feature amount calculation unit 411, that is, the feature amount that can be obtained from each block Bl tracked by the feature amount calculation unit 411 as an input variable, and the waste information as an output variable, and uses the past feature amount as an output variable. The waste information prediction model 44a is made to learn the relationship between the amount and past waste information. Then, the learning unit 412 creates a waste information prediction model (learned waste information prediction model) 44a, which is a relationship model that associates the feature amount of the pre-supply waste 51 with the waste information of the pre-supply waste 51. is stored in the storage unit 44 .

The learning method of the waste information prediction model 44a is not particularly limited, but machine learning methods suitable for learning time-series data, such as RNN (Recurrent Neural Network), LSTM (Long Short Term Memory), and sequential linear prediction, are used. be able to.

Moreover, the waste information prediction model 44a is not particularly limited to a learned model. That is, the feature values of the past pre-supply waste 51 in the incinerator whose waste information is to be predicted and controlled from now on are used as input values, and the waste information of the pre-supply waste 51 is used as output values, and the correspondence between them is shown. It may be a LUT (lookup table) or a predetermined function. When predicting, the waste information may be calculated by appropriately approximating or interpolating the input feature amount of the pre-supply waste 51 using these LUTs or a predetermined function.

Here, as described above, the "feature amount" learned by the waste information prediction model 44a includes the amount of temperature change of the waste 50 for each block Bl, the amount of movement of the waste 50 for each block Bl, and the photographed image information. at least one of the total number of blocks Bl set to , and the number of dropped blocks dropped from the tracking range (for example, the area near the waste supply unit 12) set in advance in the captured image information.

The “waste information” learned by the waste information prediction model 44 a includes the quality of the waste 50 . The “quality” includes at least one of the calories (calorific value) generated when the waste 50 is burned and the amount of steam generated from the boiler 9 when the waste 50 is burned.

The waste information prediction unit 413 predicts the waste information by inputting the feature amount calculated by the feature amount calculation unit 411 from the photographed image information of the inside of the furnace 1 to the waste information prediction model 44a. For example, the waste information prediction unit 413 inputs the amount of temperature rise of the pre-supply waste 51 for each block Bl in the vicinity of the waste supply unit 12 to the waste information prediction model 44a as input data. A predicted value of the quality (calorie, water vapor content) of the pre-supply waste 51 is obtained as an output.

The output unit 42 outputs predetermined information to the outside. Under the control of the control unit 41, the output unit 42 displays an image of the waste 50 in the furnace 1, waste information (e.g. quality prediction results), etc. on the display monitor, or displays characters and figures on the screen of the touch panel display. etc., and output audio from the speaker. Specifically, the output unit 42 outputs the waste information of the pre-supply waste 51 predicted by the waste information prediction unit 413 .

The input unit 43 is configured using a user interface such as a keyboard, input buttons, a lever, a touch panel for manual input superimposed on a display such as a liquid crystal, and a microphone for voice recognition. Predetermined information is input to the control unit 41 by a user or the like operating the input unit 43 . The output unit 42 and the input unit 43 may be integrated as an input/output unit, and the input/output unit may be configured by a touch panel display, a speaker microphone, or the like. Specifically, the input unit 43 is configured to be able to input photographed image information of the waste 50 photographed outside.

The storage unit 44 has the same functional and physical configuration as the storage unit 32 described above, and includes storage such as volatile memory such as RAM, nonvolatile memory such as ROM, EPROM, HDD, and removable media. Consists of a medium.

The storage unit 44 can store an OS, various programs, various tables, various databases, and the like for executing the operations of the waste information prediction device 40 . Here, the various programs include an information processing program that realizes control using a learning model or a learned model according to this embodiment. The storage unit 44 may be provided in another server that can communicate via various networks, or may be provided in the combustion control device 30 . Specifically, the storage unit 44 stores a waste information prediction model 44a.

The waste information prediction model 44a is a model that has been trained by the learning unit 412. From the feature amount (for example, the amount of temperature rise of the pre-supply waste 51) obtained from each block Bl of the captured image information, the waste information prediction model 44a It is a model that predicts information (eg quality of pre-supply waste 51). Further, the waste information prediction model 44a predicts the waste information when the feature amount calculated from the tracking block that traces the plurality of blocks Bl set in the photographed image information of the pre-supply waste 51 is input. It is learned to output.

The waste information prediction model 44a is assumed to be used, for example, as a program module that is part of artificial intelligence software, and is used in a computer (waste information prediction device 40) having a CPU and a storage device. Also, the waste information prediction model 44a may be recorded on a recording medium such as a CD, DVD, flash memory, magnetic tape, or the like and made readable.

(Learning method of waste information prediction model)
A learning method of the waste information prediction model according to the embodiment of the present invention will be described with reference to FIG. In the learning method of the waste information prediction model, an image acquisition process, a feature value calculation process, and a learning process are performed.

First, in the image acquisition step, the imaging unit 26 images the waste 50 in the furnace 1 of the incinerator to acquire captured image information (step S1). Subsequently, in the feature amount calculation process, the feature amount calculation unit 411 divides the captured image information acquired in step S1 into a plurality of blocks Bl (step S2). Subsequently, the feature quantity calculation unit 411 tracks the movement of the waste 50 for each block Bl, and calculates a feature quantity that characterizes the waste 50 contained in each block Bl (for example, the amount of temperature rise of the pre-supply waste 51). (step S3). Subsequently, in the learning process, the learning unit 412 causes the waste information prediction model 44a to learn the relationship between the feature amount calculated in step S3 and the waste information (for example, the quality of the pre-supply waste 51) (step S4), this flow is completed.

(Waste information prediction method)
A waste information prediction method according to an embodiment of the present invention will be described with reference to FIG. In the waste information prediction method, an image acquisition process, a feature value calculation process, and a waste information prediction process are performed.

First, in the image acquisition step, the imaging unit 26 images the waste 50 in the furnace 1 of the incinerator to acquire captured image information (step S11). Subsequently, in the feature amount calculation process, the feature amount calculation unit 411 divides the captured image information acquired in step S11 into a plurality of blocks Bl (step S12). Subsequently, the feature quantity calculation unit 411 tracks the movement of the waste 50 for each block Bl, and calculates a feature quantity that characterizes the waste 50 contained in each block Bl (for example, the amount of temperature rise of the pre-supply waste 51). (step S13). Subsequently, in the waste information prediction step, the waste information prediction unit 413 inputs the feature amount calculated in step S13 to the waste information prediction model 44a to generate waste information (for example, pre-supply waste 51 quality) is predicted (step S14), and this flow is completed.

According to the waste information prediction device, the incinerator combustion control device, the waste information prediction method, the waste information prediction model learning method, and the waste information prediction model program according to the embodiments described above, the waste in the furnace 1 By tracking the movement of the object 50 for each of a plurality of preset blocks Bl and having the waste information prediction model 44a learn the feature amount of each block Bl, the waste information in the furnace 1 can be accurately predicted. It is possible to efficiently control a waste treatment plant such as an incinerator based on the predicted waste information.

That is, according to the waste information prediction device, the incinerator combustion control device, the waste information prediction method, the waste information prediction model learning method, and the waste information prediction model program according to the embodiments, detailed information can be obtained from captured image information. By tracking the movement of the waste 50 and capturing the temperature rise of the focused waste 50 over time, a feature value highly correlated with the quality of the waste 50 (for example, calories, water vapor content) is extracted. Then, by performing learning and prediction using the feature values as input data and using the prediction results in the control of the waste treatment plant, it is possible to improve the accuracy and efficiency of the control of the waste treatment plant. can.

For example, when the temperature rise amount of the pre-supply waste 51 is small (the temperature rise speed is slow), it is estimated that the pre-supply waste 51 contains a large amount of water and a low calorie content. On the other hand, when the amount of temperature rise of the pre-supply waste 51 is large (the temperature rise speed is high), it is presumed that the pre-supply waste 51 contains less water and has a high calorie content. As described above, in the waste information prediction device, the waste information prediction method, and the waste information prediction model learning method according to the embodiment, the relationship between the amount of temperature rise and the quality of the pre-supply waste 51 is determined in advance by the waste information. By making the prediction model 44a learn, the quality of the pre-supply waste 51 can be predicted with high accuracy.

Further, in the incinerator combustion control device and the incinerator combustion control method according to the embodiment, for example, when the predicted calorie is high, the amount of waste 50 to be placed on the grate 4 (that is, the waste supply device feeding speed) is When the predicted calorie is low, control is performed to increase the amount of the waste 50 placed on the grate 4 . As a result, the combustion of the waste 50 can be stabilized, and the amount of steam generated from the boiler 9 and the amount of power generation can be stabilized.

Further advantages and modifications can be easily derived by those skilled in the art. The broader aspects of the disclosure are not limited to the specific details and representative embodiments shown and described above. Accordingly, various changes may be made without departing from the spirit or scope of the general inventive concept defined by the appended claims and equivalents thereof.

REFERENCE SIGNS LIST 1 furnace 1a furnace wall 2 waste inlet 3 waste feeder 4 grate 5 ash drop outlet 6 combustion air blower 7 furnace outlet 8 chimney 9 boiler 9a heat exchanger 9b steam drum 10 secondary air inlet 11 secondary Air Blower 12 Waste Feed 13 Step Wall 14 Combustion Air Damper 14a, 14b, 14c, 14d Under-Grate Combustion Air Damper 15 Secondary Air Damper 16 Intermediate Ceiling 17 Combustion Chamber Gas Thermometer 18 Main Flue Gas Thermometer 19 Furnace outlet lower gas thermometer 20 Furnace outlet central gas thermometer 21 Furnace outlet gas thermometer 22 Boiler outlet oxygen concentration meter 23 Gas concentration meter 24 Exhaust gas flow meter 25 Steam flow meter 26 Imaging unit 30 Combustion control device 31 Control unit 32 Storage Part 33 Manipulated amount adjustment part 331 Combustion air adjustment part 332 Air amount ratio adjustment part 333 Secondary air adjustment part 334 Waste supply device feed speed adjustment part 335 Grate feed speed adjustment part 40 Waste information prediction device 41 Control part 411 Feature value calculation unit 412 Learning unit 413 Waste information prediction unit 42 Output unit 43 Input unit 44 Storage unit 44a Waste information prediction model 50 Waste 51 Waste before supply 52 Waste on grate

Claims (17)

A waste information prediction device for predicting information about waste in an incinerator,
an input unit capable of inputting captured image information of the waste captured outside;
The captured image information of the waste input from the input unit is divided into a plurality of blocks, and at least one of the plurality of divided blocks is tracked for a predetermined time to calculate the feature amount of the tracked block. a quantity calculating means;
a storage unit that stores a relationship model that associates the characteristic amount of the waste with the information about the waste;
a control means;
has
The control means calculates the feature amount of the waste by the feature amount calculation means from the photographed image information obtained by photographing the waste in the incinerator, which is input from the input unit, and stores the feature amount. and predicting the information about the waste based on the relationship model stored in a unit. The feature amount includes at least one of a temperature change amount of the waste, a movement amount of the waste, and a total number of a plurality of divided blocks of the waste set in the captured image information. The waste information prediction device according to claim 1. 2. The information on the waste includes at least one of the calories generated when the waste is burned and the amount of steam generated from the boiler of the incinerator when the waste is burned. The waste information prediction device according to claim 2. 4. The waste information prediction device according to any one of claims 1 to 3, further comprising an output unit for outputting the information about the waste predicted by the control means. 2. The photographed image information of the waste input from the input unit is photographed image information obtained by photographing the pre-supply waste before it is supplied to the grate of the incinerator. The waste information prediction device according to any one of claims 1 to 4. The relationship model stored in the storage unit is a learned waste information prediction model obtained by learning from the relationship between the past feature amount of the waste and the information on the past waste. The waste information prediction device according to any one of claims 1 to 5. The relational model stored in the storage unit is a lookup table or a predetermined function that takes as an input value a characteristic amount of past waste and has an output value that is information related to the past waste. The waste information prediction device according to any one of claims 1 to 5. A combustion control device comprising a combustion control unit that controls an incinerator that burns waste,
The combustion control unit is
Combustion control of an incinerator, wherein combustion in the incinerator is controlled based on the information about the waste output from the waste information prediction device according to any one of claims 4 to 7. Device. The incinerator comprises a grate for moving the waste, a waste supply device for supplying the waste onto the grate, and a blower for blowing air into the incinerator,
Based on the information about the waste, the combustion control unit controls the speed of movement of the waste on the grate, the speed of supply of the waste by the waste supply device, the amount of air blown by the blower, the blower 9. The combustion control device for an incinerator according to claim 8, wherein at least one of the air temperatures is controlled by A waste information prediction method for predicting information about waste in an incinerator, comprising:
an image acquisition step of acquiring photographed image information of the waste to be put into the incinerator;
Feature quantity calculation for dividing the photographed image information of the waste acquired in the image acquisition step into a plurality of blocks, tracking at least one of the plurality of divided blocks, and calculating a feature quantity of the tracking block. process and
a waste information prediction step of predicting information about the waste from the feature amount of the waste calculated in the feature amount calculation step;
A waste information prediction method, comprising: 11. The waste information prediction method according to claim 10, wherein the waste information prediction step uses a relationship model in which the characteristic amounts of the waste and the information about the waste are associated in advance. 12. The relationship model according to claim 11, wherein the relationship model is a learned waste information prediction model obtained by learning from the relationship between the feature amount of past waste and the information on the past waste. waste information prediction method. 12. The disposal according to claim 11, wherein the relational model is a lookup table or a predetermined function having as an input value a characteristic amount of past waste and having as an output value information relating to the past waste. Object information prediction method. The photographed image information of the waste obtained in the image obtaining step is photographed image information obtained by photographing the pre-supply waste before it is supplied to the grate of the incinerator. The waste information prediction method according to any one of claims 10 to 13. 15. The method according to any one of claims 10 to 14, further comprising a combustion control step of controlling combustion of said incinerator based on the waste information predicted in said waste information prediction step. waste information prediction method. A learning method for a waste information prediction model that predicts information about waste in an incinerator, comprising:
Dividing captured image information of waste in an incinerator for incinerating waste into a plurality of blocks, tracking at least one of the plurality of divided blocks,
Using the feature amount calculated from the tracking block as an input value and the information about the waste as an output value, the relationship between the past feature amount and the past information about the waste is applied to the waste information prediction model. A learning method for a waste information prediction model, characterized by learning. A waste information prediction model program that predicts information about waste in an incinerator that can be recorded on a recording medium,
It is characterized by learning to output information about the waste when a feature amount calculated from a tracking block that tracks a plurality of blocks set in the captured image information of the waste is input. waste information prediction model program.

Images