Information processing device and information processing method

Abstract

To accurately understand the condition of the waste on the grate. [Solution] The information processing device includes a control unit with hardware, which acquires image data generated from thermal image information of an area containing waste in a waste incinerator equipped with a grate, acquires process data of the waste incinerator, virtually divides the grate into multiple areas along the furnace width direction, virtually divides each of the divided areas into multiple areas along the furnace length direction, and derives the amount of waste present on the grate, the amount of waste supplied to the grate, and the temperature of the waste for each of the multiple areas based on at least one of the acquired image data and process data, derives state variables of the waste for each of the multiple areas based on the acquired process data and the derived amount of waste, amount of waste supplied, and temperature of the waste, and smooths the state variables with the state variables of the selected area and the area adjacent to the selected area.

Description

[Technical Field]

[0001] The present invention relates to an information processing device and an information processing method. [Background technology]

[0002] Conventionally, methods have been proposed to measure or estimate the state of waste present on the grate inside a stoker-type incinerator in order to improve the operational efficiency of such incinerators. For example, Patent Document 1 describes estimating the waste layer thickness for each zone of the stoker. Patent Document 2 describes dividing a thermal image of the waste inside the incinerator into predetermined intervals in the depth direction of the incinerator, and calculating the estimated amount of waste in a designated area based on the average height from the division point in a designated area inside the incinerator to the boundary line between the waste and the inner wall, and the area of ​​the designated area. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Patent No. 3030614 [Patent Document 2] Patent No. 6472035 [Overview of the project] [Problems that the invention aims to solve]

[0004] If the amount of waste inside the incinerator can be estimated accurately, the state of the waste can be estimated based on the estimation results, and combustion can be controlled efficiently based on the state of the waste. However, in the conventional technology described above, the amount of waste is estimated only by dividing the inside of the incinerator in the depth direction of the incinerator. Therefore, if there is an uneven distribution of waste in the width direction of the furnace, the amount of waste cannot be estimated accurately.

[0005] This invention has been made in view of the above, and aims to accurately grasp the state of waste on the grate. [Means for solving the problem]

[0006] In order to solve the above-described problems and achieve the object, an information processing apparatus according to an aspect of the present invention includes a control unit having hardware. The control unit acquires image data generated from thermal image information obtained by imaging a region including the waste in a waste incinerator provided with a grate for moving the waste, acquires process data in the waste incinerator, virtually divides the grate into a plurality of regions along the furnace width direction of the waste incinerator, and virtually divides each of the regions virtually divided into the plurality of regions along the furnace width direction into the plurality of regions along the furnace length direction of the waste incinerator. Based on at least one of the acquired image data and the process data, for each of the plurality of regions, the amount of the waste existing on the grate, the supply amount of the waste supplied onto the grate, and the temperature of the waste are derived. Based on the acquired process data, the amount of the waste existing on the derived grate, the supply amount of the waste supplied onto the grate, and the temperature of the waste, a state variable of the waste is derived for each of the plurality of regions, and the state variable for each of the plurality of regions is smoothed using the state variable of a selected region and the state variables of the regions adjacent to the selected region.

[0007] The information processing apparatus according to an aspect of the present invention adjusts each of the smoothed state variables based on the total of the state variables after smoothing for each of the plurality of regions.

[0008] In the information processing apparatus according to an aspect of the present invention, the state variable is the amount of volatile components contained in the waste, the amount of moisture contained in the waste, the amount of fixed carbon contained in the waste, and the amount of ash contained in the waste.

[0009] An information processing method according to one aspect of the present invention is an information processing method executed by an information processing apparatus including a control unit having hardware, the method including: obtaining image data generated from thermal image information obtained by imaging a region including the waste in a waste incinerator including a grate for moving the waste; obtaining process data in the waste incinerator; virtually dividing a plurality of regions along the furnace width direction on the grate, and virtually dividing each of the regions virtually divided into the plurality of regions along the furnace width direction into the plurality of regions along the furnace length direction of the waste incinerator; a first deriving step of deriving, for each of the plurality of regions, the amount of the waste existing on the grate, the amount of the waste supplied onto the grate, and the temperature of the waste based on at least one of the obtained image data and the process data; a second deriving step of deriving a state variable of the waste for each of the plurality of regions based on the obtained process data and the amount of the waste existing on the grate, the amount of the waste supplied onto the grate, and the temperature of the waste derived; and a smoothing step of smoothing the state variable for each of the plurality of regions with the state variable of a selected region and a region adjacent to the selected region.

Advantages of the Invention

[0010] According to the present invention, the state of the waste on the grate can be accurately grasped.

Brief Description of the Drawings

[0011] [Figure 1] FIG. 1 is an overall configuration diagram schematically showing a waste incineration facility to which a waste state estimation apparatus according to an embodiment of the present invention is applied. [Figure 2] FIG. 2 is a side view showing waste in an incinerator, a supply portion of the waste onto the grate, and an imaging unit according to an embodiment of the present invention. [Figure 3] FIG. 3 is a block diagram showing a configuration of a combustion control apparatus according to an embodiment of the present invention. [Figure 4]Figure 4 is a block diagram showing the configuration of a waste condition estimation device in a management system according to one embodiment of the present invention. [Figure 5] Figure 5 shows an example of transmission image data of burning waste captured by the imaging unit according to one embodiment of the present invention. [Figure 6] Figure 6 shows an example in which a longitudinal section of waste is set on boundary image data, which is obtained by setting a boundary line on transmission image data captured by an imaging unit according to one embodiment of the present invention. [Figure 7] Figure 7 is a flowchart illustrating a waste disposal method according to one embodiment of the present invention. [Figure 8] Figure 8 is a flowchart detailing the calculation steps of the waste reaction model in a waste treatment method according to one embodiment of the present invention. [Figure 9] Figure 9 illustrates the division regions and configuration of waste on the grate in a waste incinerator according to one embodiment of the present invention. [Figure 10] Figure 10 shows an example of a filter applied to the smoothing process. [Figure 11] Figure 11 shows a divided region on a fire grate. [Figure 12] Figure 12 is a graph showing the derived values ​​of the waste state in a waste incinerator according to one embodiment of the present invention, for each divided region on the grate. [Figure 13] Figure 13 is a graph showing the derived values ​​of the amount of water evaporated from the waste state in a waste incinerator according to one embodiment of the present invention, for each divided region on the grate. [Figure 14] Figure 14 is a graph showing the derived values ​​of volatile matter emissions from waste in a waste incinerator according to one embodiment of the present invention, for each divided region on the grate. [Figure 15] Figure 15 is a graph showing the total amount of waste on the grate, the amount of waste dropped and supplied onto the grate, and the state of the waste in each region in a waste incinerator according to one embodiment of the present invention. [Modes for carrying out the invention]

[0012] Embodiments of the present invention will be described in detail below with reference to the attached drawings. However, the present invention is not limited to the embodiments described below. Furthermore, in the drawings, the same or corresponding elements are appropriately denoted by the same reference numerals. It should also be noted that the drawings are schematic, and the dimensional relationships of each element may differ from those in reality. Even between drawings, there may be parts where the dimensional relationships and ratios differ.

[0013] First, in order to describe embodiments of the present invention, the inventors will explain the studies they conducted to solve the above-mentioned problems. In their studies regarding the state of waste, the inventors identified the problems in the prior art described above. According to the inventors' studies, the technology described in Patent Document 1 has the problem that even if the amount of waste can be estimated, information regarding the state of the waste cannot be obtained. Furthermore, the technology described in Patent Document 2 has the problem that even if the amount of waste can be estimated, no information regarding the state of the waste can be obtained.

[0014] Therefore, the inventors conducted research and conceived of a method to derive various states of waste inside an incinerator (hereinafter collectively referred to as "waste state") using thermal image data captured inside the incinerator and process data related to the incinerator. Specifically, they devised the following configuration to solve the problems in the prior art described above.

[0015] Specifically, firstly, a theoretical model is constructed that includes the supply of waste to the grate in the incinerator, the movement of waste on the grate, evaporation of water from the waste, thermal decomposition of the waste, and combustion of the thermal decomposition products generated by the thermal decomposition. Secondly, thermal image data of the grate inside the incinerator is acquired to derive the amount of waste supplied to the grate, the temperature of the waste, and the amount of waste, thereby obtaining process data on the grate. Thirdly, process data of the incinerator is acquired by measuring the temperature and gas flow rate of various parts inside the incinerator, as well as the heat absorbed by the steam boiler. Fourthly, a state-space model is constructed using the constructed waste reaction model, the acquired process data on the grate, and the process data of the incinerator. Using data assimilation techniques with a Kalman filter, a waste state estimation process is performed to estimate the amount of waste, the moisture content of the waste, the combustible content of the waste, and the calorific value of the waste at each of the multiple divided positions on the grate. Fifth, a smoothing process is performed on the state of each of the multiple divided positions on the grate to prevent discontinuities from occurring between adjacent positions. Furthermore, sixth, a waste state estimation process is performed immediately (in real time), and combustion in the incinerator is controlled based on the outputted estimate.

[0016] Furthermore, the inventors conducted studies and devised a method for deriving the amount of waste on the grate of an incinerator based on thermal image data. Specifically, in a stoker-type incinerator (hereinafter referred to as a grate-type incinerator), waste supplied by dropping it down a stepped wall from the supply port is burned while being moved by the grate. Therefore, the inventors devised a method to approximate the cross-sectional shape of the waste accumulated on the grate using a simple closed curve (also called a single closed curve or Jordan curve) with vertices at the intersection of the stepped wall and the grate, the top point of the waste, and the combustion point of the waste on the grate, respectively, obtained from thermal image data. In this specification, various triangles and simple closed curves with three vertices are referred to as three-vertex closed curves. In addition, by detecting the boundary line of the outer edge of the waste accumulated on the grate using image signal processing such as deep learning, and applying a closed curve model, the distribution shape of the waste can be estimated with high accuracy. Thus, the inventors devised a method for deriving the amount of waste by detecting the boundary line of the waste on the grate. One embodiment described below was devised based on the inventor's considerations described above.

[0017] [Embodiment] (Waste disposal system) Figure 1 is an overall configuration diagram showing a waste incineration facility according to one embodiment of the present invention. As shown in Figure 1, the waste treatment system 1 according to the embodiment has a combustion control device 30 and a waste state estimation device 40 that can communicate with each other via a network 2. The combustion control device 30 is configured to control the waste incinerator 100.

[0018] Network 2 consists of the Internet network, mobile phone network, etc. Network 2 is, for example, a public communication network such as the Internet, and may also include other communication networks such as WANs (Wide Area Networks), mobile communication networks, and wireless communication networks such as Wi-Fi (registered trademark). The data transmitted and received in communication between the combustion control device 30 and the waste state estimation device 40 may include operational management indicators that are important for the operation of the waste incinerator 100. In this case, considering the security of the transmitted and received data, the communication line between the combustion control device 30 and the waste state estimation device 40 can be a dedicated line or a VPN (Virtual Private Network) line. The combustion control device 30 and the waste state estimation device 40 may be configured as a single unit, and the combustion control device 30 and the waste state estimation device 40 may be installed in the same facility as the waste incinerator 100 or in separate facilities. When the waste incinerator 100, the combustion control device 30 and the waste state estimation device 40 are installed in separate facilities, various information and data are communicated via Network 2.

[0019] (Waste incinerator) As shown in Figure 1, the waste incinerator 100 for burning the waste 50 is a grate-type incinerator, comprising a furnace 101 where the waste 50 is burned, a waste input port 102 into which the waste 50 is fed, a waste supply device 103 that supplies the waste 50 to the furnace 101, and n grates 1041-104 n , and a boiler 109. The boiler 109 includes a heat exchanger 109a and a steam drum 109b installed downstream of the furnace outlet 107, which is the outlet of the furnace 101. Grates 1041-104 n Since their configurations are the same, if there is no need to distinguish between them, the subscript is omitted and they are referred to as fire grate 104.

[0020] Waste 50 introduced from the waste input port 102 is supplied into the furnace 101 by a reciprocating waste supply device 103. The grate 104 is driven by a drive device (not shown) to reciprocate, stirring and moving the waste 50 inside the furnace 101. The waste 50 on the grate 104 is burned while being dried by the blowing of combustion air supplied from wind boxes 1301 to 1303 located below the grate 104. For example, wind box 1301 is located below the drying grate, wind box 1302 is located below the combustion grate, and wind box 1303 is located below the post-combustion grate. On the drying grate, drying, ignition, and initial combustion of the waste 50 mainly occur, on the combustion grate, thermal decomposition and partial oxidation of the waste mainly occur, and on the post-combustion grate, post-combustion occurs to completely burn the unburned portion of the remaining waste 50. This post-combustion forms a layer of incinerated ash on the post-combustion grate after complete combustion. The ash generated by the combustion of the waste 50 falls through the ash outlet 105 and is discharged outside the furnace 101. The exhaust gas generated by the combustion of the waste 50 flows from the furnace outlet 107 to the boiler 109.

[0021] The total amount of combustion air supplied from below the grate 104 into the furnace 101 is controlled by a combustion air damper 114 and a bypass damper 126b located immediately adjacent to the combustion air blower 106. The flow rate of combustion air supplied to each of the windboxes 1301 to 1303 is controlled by grate-under combustion air dampers 1141 to 1143, which are located in the piping supplying combustion air to each of the windboxes 1301 to 1303. In other words, the ratio of the flow rates of combustion air supplied to each of the windboxes 1301 to 1303 is controlled by the grate-under combustion air dampers 1141 to 1143. In Figure 1, the area below the grate 104 is divided by three wind boxes along the direction of transport of the waste 50, and combustion air is supplied through each wind box. However, the number of wind boxes below the grate 104 and the number of combustion air dampers 1141-1143 below the grate can be appropriately changed depending on the size and purpose of the waste incinerator 100.

[0022] Furthermore, the combustion air damper 114 is connected, for example, to a preheater 126a connected in series and a bypass damper 126b connected in parallel. The preheater 126a heats the air supplied from the combustion air blower 106. By adjusting the opening of the combustion air damper 114 and the bypass damper 126b, the temperature of the combustion air supplied from below the grate 104 into the furnace 101 is adjusted.

[0023] Cooling air is blown into the furnace 101 by a cooling air blower 111 from cooling air inlets 110 located on the furnace walls and ceiling of the furnace 101. The blowing of cooling air into the furnace 101 further burns unburned components in the combustion gas and prevents the furnace wall temperature from rising excessively. The flow rate of cooling air supplied into the furnace 101 from the cooling air inlets 110 is adjusted by a cooling air damper 115 installed in series with the cooling air blower 111. On the ceiling of the furnace 101, there is a recirculation blower 127 for supplying exhaust gas, from which dust has been removed and moisture has been removed, into the furnace 101, and an exhaust gas recirculation air damper 128 installed in series with the recirculation blower 127, which adjusts the flow rates of exhaust gas and combustion air when the combustion air and the exhaust gas supplied by the recirculation blower 127 are mixed and supplied into the furnace 101. Low air-to-air ratio combustion using the exhaust gas recirculation air damper 128 makes it possible to suppress the generation of NOx during combustion. The exhaust gas and combustion air, whose flow rates have been adjusted by the exhaust gas recirculation air damper 128, are blown into the furnace 101 from the inlet 110a.

[0024] An intermediate ceiling 116 is provided inside the furnace 101 at an upper position along the height direction of the furnace 101. The gas flowing inside the furnace 101 can be separated and discharged by the intermediate ceiling 116 into gas containing a large amount of combustible gas generated in the waste drying process and main combustion process upstream in the direction of transport of the waste 50 in the grates 1041-1043, and combustion exhaust gas generated in the post-combustion process downstream. Specifically, the combustion exhaust gas flows through a flue (main flue) below the intermediate ceiling 116, while the gas containing a large amount of combustible gas flows through a flue (secondary flue) above the intermediate ceiling 116. The combustion exhaust gas and the gas containing a large amount of combustible gas merge in a gas mixing section provided on the furnace outlet 107 side of the furnace 101. The merging of the combustion exhaust gas and the gas containing a large amount of combustible gas in the gas mixing section further promotes the agitation and mixing of the gases in the gas mixing section, and the mixed gas undergoes secondary combustion by supplying secondary combustion air. This stabilizes combustion in the secondary combustion section, suppresses the generation of dioxins during the combustion process, and reduces the generation of unburned waste. Alternatively, the furnace 101 may be configured without an intermediate ceiling 116.

[0025] Thermometers are provided at multiple locations within the furnace 101 to measure the gas temperature inside the furnace 101. Specifically, a combustion chamber gas thermometer 117 is provided along the height of the furnace 101, at a position midway between the grate 104 and the cooling air inlet 110.

[0026] A main flue gas thermometer 118 is provided along the height of the furnace 101, below the furnace outlet 107. A lower furnace outlet gas thermometer 119 is provided along the height of the furnace 101, below the furnace outlet 107. A middle furnace outlet gas thermometer 120 is provided along the height of the furnace 101, in the middle of the furnace outlet 107. A furnace outlet gas thermometer 121 for measuring the combustion control temperature is provided along the height of the furnace 101, downstream of the furnace outlet 107. The temperature measurements taken by the combustion chamber gas thermometer 117, the main flue gas thermometer 118, the lower furnace outlet gas thermometer 119, the middle furnace outlet gas thermometer 120, and the furnace outlet gas thermometer 121 are stored in the combustion control device 30 as combustion process measurement values. The temperature measurement data stored in the combustion control device 30 may be transmitted from the combustion control device 30 to the waste state estimation device 40 as measurement value data.

[0027] An imaging unit 125 is provided downstream of the waste transport direction in the furnace 101. The imaging unit 125 includes, for example, a flame-transmitting camera consisting of an infrared camera, and an image processing unit that processes the captured image data. The imaging unit 125 captures the combustion state of the waste 50 on the grate 104 and stores the transmitted image data, which is thermal image data generated from the captured thermal image information, in the combustion control device 30. Alternatively, the imaging unit 125 may store the transmitted image data in the waste state estimation device 40.

[0028] Figure 2 is a side view showing the installation state of the imaging unit 125. In this embodiment, as shown in Figure 2, the imaging unit 125 is installed in a position that is substantially directly opposite to the waste supply unit 112 and the stepped wall 113. However, the installation of the imaging unit 125 is not limited to a position that is substantially directly opposite to the waste supply unit 112 and the stepped wall 113. The imaging unit 125 can be installed in various positions as long as it is possible to image at least the waste 50 on the grate 104 and the boundary portion between other objects, in this case the stepped wall 113 and the grate 104. The imaging unit 125 may be installed outside the furnace in close proximity to a window provided in the furnace wall 101a, or it may be installed inside the furnace 101 with a water-cooled structure. As shown in Figure 2, the pre-supply waste 51, which is the waste 50 before it is supplied into the furnace 101, falls from the waste supply unit 112 onto the grate 104 at the portion of the stepped wall 113. The waste 52 on the grate, which is the waste 50 that has fallen onto the grate 104, is agitated by the reciprocating motion of the grate 104 and moved forward toward the imaging unit 125.

[0029] The boiler 109 is installed downstream of the secondary combustion section, where secondary combustion takes place, along the direction of transport of the waste 50. The exhaust gas generated by secondary combustion has its thermal energy recovered by the heat exchanger 109a of the boiler 109, and then flows through the flue 11 to a dust removal device 12 consisting of a bag filter and the like. The boiler 109 is equipped with a boiler outlet oxygen concentration meter 122 on the outlet side to measure the oxygen concentration in the exhaust gas.

[0030] In the flue 11, for example, chemicals such as slaked lime and activated carbon are blown into the exhaust gas along with carbon dioxide from the chemical supply device 13. When the chemicals are blown into the exhaust gas, they bind to pollutants such as hydrogen chloride and sulfur oxides contained in the exhaust gas. The dust removal device 12 collects and removes the dust and pollutants contained in the exhaust gas that has flowed through the flue 11, along with the chemicals to which they have bound. An induced draft fan 14 is connected to the dust removal device 12. The induced draft fan 14 draws in the exhaust gas from which dust has been removed by the dust removal device 12. The exhaust gas drawn in by the induced draft fan 14 is sent to the wet scrubber 15.

[0031] The wet scrubber 15 is an exhaust gas cleaning means that cleans the exhaust gas, for example by bringing an aqueous sodium hydroxide solution into contact with the exhaust gas to remove water, sulfides, chlorides, etc. from the exhaust gas. The exhaust gas that has been cleaned by the wet scrubber 15 and contains carbon dioxide is branched and sent to the recirculation blower 127 and the chimney 108. The exhaust gas sent to the chimney 108 is exhausted to the outside.

[0032] The flue 11 is equipped with a gas concentration meter 123 for measuring the concentrations of carbon monoxide and nitrogen oxides in the exhaust gas. The flue 11 is also equipped with an exhaust gas flow meter 124 for measuring the exhaust gas volume. The gas concentration and flow rate measured by the boiler outlet oxygen concentration meter 122, the gas concentration meter 123, and the exhaust gas flow meter 124 are stored in the memory unit 34 of the combustion control device 30 as combustion process measurement values. These combustion process measurement values ​​are also simply referred to as measurement values.

[0033] (Combustion control device) Figure 3 is a block diagram showing the configuration of the combustion control device 30. The combustion control device 30 is an example of an estimation unit according to the present invention. The combustion control device 30 comprises a calculation control unit 31, a manipulated variable reference value adjustment unit 32, a manipulated variable reference value correction unit 33, a storage unit 34, a manipulated variable adjustment unit 35, and a communication unit 36. Specifically, the calculation control unit 31, the manipulated variable reference value adjustment unit 32, the manipulated variable reference value correction unit 33, and the manipulated variable adjustment unit 35 each include a processor such as a CPU (Central Processing Unit), DSP (Digital Signal Processor), or FPGA (Field-Programmable Gate Array) with hardware, and a main memory unit such as RAM (Random Access Memory) or ROM (Read Only Memory) (none of which are shown).

[0034] The storage unit 34 is composed of a storage medium selected from volatile memory such as RAM, non-volatile memory such as ROM, EPROM (Erasable Programmable ROM), hard disk drive (HDD), and removable media. Removable media include, for example, USB (Universal Serial Bus) memory, or disk recording media such as CD (Compact Disc), DVD (Digital Versatile Disc), or BD (Blu-ray® Disc). Alternatively, the storage unit 34 may be configured using a computer-readable recording medium such as an externally insertable memory card.

[0035] The storage unit 34 stores the operating system (OS), various programs, various tables, various databases, etc., for executing the operation of the combustion control device 30. Here, the various programs include information processing programs that realize processing based on models such as the learning model and the trained model according to this embodiment. These various programs can also be recorded on computer-readable recording media such as hard disks, flash memory, CD-ROMs, DVD-ROMs, and flexible disks and widely distributed. The storage unit 34 has a setting value database 341 that stores the incineration amount setting value and evaporation amount setting value input from an external source as information, and a measurement value database 342 that stores the combustion process measurement value (also called process data) related to combustion obtained from the waste incinerator 100 as information. The storage unit 34 may be located on another server that can communicate via various networks.

[0036] The combustion control device 30 controls the combustion air volume, cooling air volume, waste supply device feed speed, and grate feed speed as the control variables at each control end, based on a predetermined control variable reference value setting relational expression. The combustion control device 30 also controls stopping and operating the waste supply device feed speed and grate feed speed. The control variable reference value setting relational expression is a relational expression between the waste incineration amount setting value or waste quality setting value and the control variable reference value (target value of the control variable), and includes control parameters as correction coefficients. The control parameters are adjusted by the control variable reference value adjustment unit 32 to match the waste incineration amount setting value and the waste quality setting value. When at least one of the waste incineration amount setting value and the waste quality setting value is changed, the adjusted control parameters are changed by the control variable reference value adjustment unit 32 in accordance with the changed setting value. The change in control parameters corrects the preset control variable reference value.

[0037] The calculation control unit 31 performs various controls and calculations. Specifically, for example, the calculation control unit 31 can derive the amount of evaporation per unit of time generated by the combustion of waste in the waste incinerator 100 over a predetermined period, for example, one day (24 hours), as an evaporation amount set value. Also, when the calculation control unit 31 functions as a waste calculation unit, for example, it can calculate the waste quality (lower heating value of waste) according to the waste incineration amount set value. The manipulated variable reference value adjustment unit 32 can adjust the manipulated variable reference value by adjusting the control parameters included in the manipulated variable reference value setting relation. The manipulated variable reference value correction unit 33 can correct the manipulated variable reference value adjusted by the manipulated variable reference value adjustment unit 32 based on a predetermined control algorithm (such as PID control or fuzzy calculation). The data referenced by the calculation control unit 31, the manipulated variable reference value adjustment unit 32, and the manipulated variable reference value correction unit 33 are stored in the storage unit 34 in a readable format. The memory unit 34 stores predetermined control variable reference value setting relational expressions and control algorithms, daily evaporation amount setting values ​​and incineration amount setting values ​​transmitted from the waste state estimation device 40, and combustion process measurement values ​​transmitted from the waste incinerator 100 and acquired as combustion state amounts in the furnace 101.

[0038] The control volume adjustment unit 35 adjusts the respective control volume at each control end so that it follows the control volume reference value. Specifically, the control volume adjustment unit 35 includes a combustion air volume adjustment unit 351, an air volume ratio adjustment unit 352, a cooling air volume adjustment unit 353, a waste supply device feed speed adjustment unit 354, a grate feed speed adjustment unit 355, a combustion air temperature adjustment unit 356, and an exhaust gas recirculation air flow rate adjustment unit 357.

[0039] The combustion air volume adjustment unit 351 adjusts the control amount so that the combustion air volume follows the control amount reference value (hereinafter referred to as the corrected control amount reference value) corrected by the control amount reference value correction unit 33. The air volume ratio adjustment unit 352 controls each of the under-grille combustion air dampers 1141 to 1143 to adjust the flow rate and air ratio in each windbox. The cooling air volume adjustment unit 353 adjusts the control amount so that the cooling air volume follows the corrected control amount reference value. Here, the combustion air volume and cooling air volume are adjusted by controlling the opening degrees of the combustion air damper 114, the bypass damper 126b, the under-grille combustion air dampers 1141 to 1143, and the cooling air damper 115, respectively. The waste supply device feed speed adjustment unit 354 adjusts the control amount so that the waste supply device feed speed follows the corrected control amount reference value. The grate feed speed adjustment unit 355 adjusts the control amount so that the grate feed speed follows the corrected control amount reference value. The combustion air temperature adjustment unit 356 controls the opening of the combustion air damper 114 and the bypass damper 126b so that the combustion air temperature follows the corrected control amount reference value. The exhaust gas recirculation air flow rate adjustment unit 357 controls the exhaust gas recirculation air damper 128 so that the flow rate of the recirculated exhaust gas and air follows the corrected control amount reference value. If the control amount reference value has not been corrected by the control amount reference value correction unit 33, the control amount adjustment unit 35 adjusts each control amount based on the uncorrected control amount reference value.

[0040] The communication unit 36 ​​is, for example, a LAN (Local Area Network) interface board or a wireless communication circuit for wireless communication. The LAN interface board and wireless communication circuit are connected to network 2, such as the Internet, which is a public communication network. The communication unit 36 ​​is connected to network 2 and configured to communicate with the waste condition estimation device 40 and other devices and servers.

[0041] (Waste condition estimation device) Figure 4 is a schematic block diagram showing the configuration of a waste state estimation device 40, which is a device for estimating the state of waste. The waste state estimation device 40 is an example of an estimation unit according to the present invention. The waste state estimation device 40 has the configuration of a general computer that can communicate via network 2. The waste state estimation device 40 comprises a control unit 41, a storage unit 42, a communication unit 43, and an input / output unit 44. The control unit 41, storage unit 42, and communication unit 43 are physically and functionally the same as the calculation control unit 31, storage unit 34, and communication unit 36 ​​described above, respectively. The waste state estimation device 40 functions as an information processing device that derives the state of waste 50. The waste state estimation device 40 also functions as a waste volume estimation device that derives or estimates the volume of waste 50, and as a combustion point position measuring device that can measure the position of the combustion point (also called the burnout point).

[0042] First, the input / output unit 44, as an output means, displays images of the waste 50 inside the furnace 101 on a display monitor, such as an organic EL panel or a liquid crystal display panel, or displays characters or figures on the screen of a touch panel display, or outputs sound from a speaker, according to the control unit 41. The input / output unit 44, as an input means, is configured using a user interface such as a keyboard, input buttons, levers, a touch panel for manual input superimposed on a display such as a liquid crystal, or a microphone for voice recognition. The input / output unit 44 is configured to allow a user to input predetermined information to the control unit 41 by operating it. That is, the input / output unit 44 is composed of, for example, a keyboard, a touch panel keyboard built into the display unit to detect touch operations on the display panel, or a voice input device that enables communication with the outside. The input / output unit 44 may also be configured with the output unit and input unit as separate components.

[0043] The control unit 41 loads a program stored in the memory unit 42 into the work area of ​​the main memory unit and executes it. By controlling each component through the execution of the program, it can realize a function that matches a predetermined purpose. In this embodiment, the control unit 41 executes the functions of the boundary generation unit 411, the combustion calculation unit 412, and the learning unit 413 by executing a program stored in the memory unit 42. Specifically, for example, the control unit 41 executes the function of the boundary generation unit 411 by reading a boundary recognition learning model 423, which is a program, from the memory unit 42. The control unit 41 also executes the function of the combustion calculation unit 412 by reading a volume calculation program from the memory unit 42. Details of the functions of the boundary generation unit 411, the combustion calculation unit 412, and the learning unit 413 will be described later.

[0044] The storage unit 42 has the same functional and physical configuration as the storage unit 34 described above, and is composed of a storage medium selected from volatile memory such as RAM, non-volatile memory such as ROM, EPROM, HDD, and removable media. Removable media include, for example, USB memory, or disk recording media such as CD, DVD, or BD. Alternatively, the storage unit 42 may be configured using a computer-readable recording medium such as an externally insertable memory card.

[0045] The storage unit 42 can store an OS for executing the operation of the waste state estimation device 40, various programs, various tables, various databases, etc. Here, the various programs include information processing programs that realize control using the learning model or a pre-trained model according to this embodiment. The storage unit 42 may be installed on another server that can communicate via various networks, or it may be installed on the combustion control device 30.

[0046] Specifically, the storage unit 42 stores the image database 421, the combustion information database 422, the boundary recognition learning model 423, and the waste reaction model 424. In this embodiment, the databases (DBs) described are constructed by a database management system (DBMS) program executed by the aforementioned processor, which manages the data stored in the storage unit 42. The image database 421 and the combustion information database 422 are, for example, relational databases (RDBs). However, the databases stored in the storage unit 42 are not limited to those described above.

[0047] The image database 421 stores searchable image data of the furnace 101, such as transmission image data captured by the imaging unit 125 and boundary image data generated by the boundary generation unit 411. The combustion information database 422 stores searchable combustion information, including various process data, which are sensor values ​​measured by various sensors in the waste treatment system 1, such as combustion process measurements related to the combustion of waste 50 on the grate 104. These image databases 421 and combustion information databases 422 can be constructed using big data, etc. In this case, the image databases 421 and combustion information databases 422 can also be stored in the storage of another server accessible via the network 2.

[0048] The boundary recognition learning model 423 is an updatable model that includes at least one learning model. If the learning model is not updated, it is stored in the memory unit 42 as a pre-trained model. The boundary recognition learning model 423 may also include a combustion state learning model, which is a learning model capable of extracting combustion regions based on image data within the furnace 101, such as transmission image data and boundary image data. Alternatively, instead of the boundary recognition learning model 423, a rule-based information processing program created without training or other processes may be used to perform predetermined information processing on input data. Furthermore, an automatic decision processing program may be included that realizes decision processing using a combustion image learning model, capable of performing predetermined decisions from combustion image data of the luminous flame itself captured by the imaging unit 125. These various programs can be recorded on computer-readable recording media such as hard disks, flash memory, CD-ROMs, DVD-ROMs, and flexible disks and widely distributed.

[0049] The waste reaction model 424 is a model that employs data assimilation technology. The waste reaction model 424 is constructed based on predetermined relational equations. Note that the following relational equations are merely examples and are not limited to those below. In this embodiment, the waste reaction model 424 is a model relating to the waste 52 on the grate 104. Specifically, the waste reaction model 424 is composed of at least one model from among the relational equations relating to the amount of water evaporation in the waste 52 on the grate (hereinafter referred to as the water evaporation model), the relational equation relating to the amount of volatile matter released by thermal decomposition (hereinafter referred to as the volatile matter release model), and the relational equation relating to the amount of fixed carbon burned (hereinafter referred to as the fixed carbon combustion model).

[0050] (Moisture evaporation model) The moisture evaporation model according to this embodiment is a relational expression for the amount of moisture evaporated per unit time from the waste 52 on the grate (moisture evaporation rate (kg / s)). Through various experiments conducted by the inventors, it was found that drying by heating and drying by ventilation are dominant in the evaporation of moisture from the waste 52 on the grate. In other words, it was found that the evaporation of moisture from the waste 52 on the grate is dominated by evaporation of moisture due to radiant heat and evaporation of moisture due to combustion air supplied to the waste 52 on the grate from below the grate 104. Therefore, in the moisture evaporation model according to this embodiment, the amount of moisture evaporated per unit time from the waste 52 on the grate (kg / s) is the amount of moisture evaporation based on the sum of the amount of moisture evaporated by radiant heat and the amount of moisture evaporated by ventilation of combustion air, as shown in equation (1) below. Water evaporation rate from waste per unit time (kg / s) = (Amount of water evaporation due to radiant heat) + (Amount of water evaporation due to combustion air supply) ... (1)

[0051] (Amount of water evaporation due to thermal radiation per unit time) The amount of moisture evaporated per unit time due to radiant heat from the waste 52 on the grate (kg / s) can be expressed by the following equation (2). Note that when the temperature of the waste reaches the evaporation temperature of water and then the received heat exceeds the heat required for the evaporation of the moisture contained in the waste 52 on the grate, the amount of moisture corresponding to the sensible heat is assumed to evaporate. (Amount of moisture evaporated per unit time due to radiant heat)=q / Q ev (kg / s) =S·φ·ε·σ·T f 4 / Q ev …(2) Here, q (kW): Waste received heat flux, Q ev (J / kg): Latent heat of evaporation of water, S (m 2 ): Representative area, φ (dimensionless): Shape factor, ε (dimensionless): Emissivity, σ (W / (m 2 · K 4 )) = 5.67×10 -8 : Stefan - Boltzmann constant, T f (K): Representative flame temperature.

[0052] (Amount of moisture evaporated per unit time due to the ventilation of combustion air) The amount of moisture evaporated per unit time due to the ventilation of combustion air for the waste 52 on the grate (kg / s) can be expressed by the following equation (3). (Amount of moisture evaporated per unit time due to the ventilation of combustion air) =G m ·(H w - H)·(1 - e -ζ )(kg / s)…(3) (Note that ζ = hG a ·Z·S a / (G m ·C H )) Here, G m (kg / s): Ventilation flow rate, H w (kg / kg′ (water vapor amount / dry air)): Saturated absolute humidity of combustion air, H (kg / kg′): Absolute humidity of combustion air, hGa (W / (m 3 ·K·s)): Heat transfer coefficient, Z (m): Height of waste layer, Sa (m 2 ): Cross-sectional area of ​​the waste layer, C H (J / (kg·K)): This is the specific heat capacity.

[0053] Based on the above, the relationship for the amount of water evaporated from waste per unit time (kg / s) according to this embodiment, i.e., the water evaporation model, can be adopted from the following relationship (4). Water evaporation rate from waste per unit time (kg / s) = (Amount of water evaporation due to radiant heat) + (Amount of water evaporation due to combustion air supply) =S·φ·ε·σ·T f 4 / Q ev +G m ·(H w -H)·(1-e -ζ )……(4)

[0054] (Volatile release model) The volatile matter release model according to this embodiment is a relational expression for the amount of volatile matter released per unit time (kg / s) by thermal decomposition that begins when the evaporation of moisture in the waste 52 on the grate is almost complete. Through various experiments conducted by the inventors, it was found that the following empirical formula (5) holds true for the amount of volatile matter released from the waste 52 on the grate. Therefore, the relational expression for the amount of volatile matter released per unit time (kg / s) from the waste 52 on the grate, as shown in the following formula (5), is adopted as the volatile matter release model according to this embodiment. In this specification, "volatile matter" refers to combustible gases generated when waste (garbage) is heated, and includes components such as carbon monoxide (CO) and methane (CH4). (Amount of volatile matter released from waste per unit time) =A·T n ·exp(-E / (R·T))·W……(5) Here, A(s -1): Frequency factor, E (J / kmol): Activation energy, T (K): Temperature of waste 52 on the grate, R (kJ / (mol·K)): General gas constant, W (kg): Amount of residual volatile matter in waste 52 on the grate. Furthermore, according to experiments conducted by the inventors, in a given furnace 101, A ( / s) = 3.96 × 10 2 ( / s), n=0, E(J / kmol)=6.31×10 7 It was found that the value is (J / kmol). In other words, when the above-mentioned equation (5) is adopted, it is possible to change the values ​​of the frequency factor A, the activation energy E, and the influence of the temperature of the waste on the grate (multiplier n) in each waste treatment system 1 to correspond to the waste incinerator 100. Furthermore, since the flame is mainly caused by the pyrolysis gas released by the thermal decomposition of the waste on the grate 52, it is possible to derive the state of the flame inside the furnace 101 based on the amount of volatile matter released obtained by equation (5).

[0055] (Sealed carbon combustion model) The fixed carbon combustion model according to this embodiment is a relational expression for the rate of combustion of fixed carbon contained in the waste 52 on the grate per unit time (kg / s). Through various experiments conducted by the inventors, it was found that the following empirical formula (6) holds true for the rate of combustion of fixed carbon contained in the waste 52 on the grate. Therefore, the fixed carbon combustion model according to this embodiment adopts the relational expression (6) below for the rate of combustion of fixed carbon contained in the waste 52 on the grate per unit time (kg / s). In this specification, "fixed carbon" refers to combustible solids produced when waste (garbage) is heated, and for example, contains carbon (C) as a component. (Amount of fixed carbon burned in waste per unit time) =A·P O2 n ·exp(-E / (R·T))·W c ...(6) Here, A(s -1 ): frequency factor, P O2(atm): Oxygen partial pressure, E (kJ / kmol): Activation energy, T (K): Temperature of waste 52 on the grate, R (kJ / (mol·K)): General gas constant, W c (kg): This is the residual carbon mass in the waste 52 on the grate. Furthermore, according to experiments conducted by the inventors, in a given furnace 101, A( / s) = 1.36 × 10 6 It was found that ( / s), n=0.68, and E(J / kmol)=130(kJ / kmol). In other words, when the above-mentioned equation (5) is adopted, the frequency factor A, activation energy E, and oxygen partial pressure P are determined in relation to the waste incinerator 100 in each waste treatment system 1. O2 It is possible to change the numerical value of the influence (multiplier n).

[0056] In this embodiment, if the waste reaction model 424 consists of, for example, only a water evaporation model, the waste reaction model 424 can derive the amount of water evaporated from the waste 52 on the grate. This makes it possible to construct a state-space model relating to the state of the waste and to estimate the amount of water contained in the waste 52 on the grate. By being able to estimate the amount of water contained in the waste 52 on the grate, it becomes possible to adjust the amount of air supplied to the waste drying process, especially on the upstream side of the grate 104.

[0057] Furthermore, in this embodiment, if the waste reaction model 424 consists only of, for example, a volatile matter release model, it is possible to derive the amount of volatile matter released by the waste reaction model 424. In this case, the waste 52 on the grate from which the moisture has evaporated is thermally decomposed, releasing volatile matter and burning as a flame. By deriving the amount of volatile matter released, a state-space model relating to the state of the waste can be constructed, and it becomes possible to estimate the amount of volatile matter contained in the waste 52 on the grate. By being able to estimate the mass of volatile matter contained in the waste 52 on the grate, it becomes possible to adjust the amount of combustible gas generated in the upstream waste drying process and the main combustion process, and the amount of combustion exhaust gas generated in the downstream after-combustion process.

[0058] Furthermore, in this embodiment, if the waste reaction model 424 consists only of, for example, a fixed carbon combustion model, the waste reaction model 424 can be used to derive the amount of fixed carbon burned in the waste 52 on the grate. This allows for the construction of a state-space model relating to the state of the waste, and enables the estimation of the mass of fixed carbon contained in the waste 52 on the grate. By being able to estimate the mass of fixed carbon contained in the waste 52 on the grate, combustion control necessary for ash carbonation downstream becomes possible.

[0059] (Derivation of ash content) Furthermore, in this embodiment, if the waste reaction model 424 consists of a water evaporation model, a volatile matter release model, and a fixed carbon combustion model, the waste reaction model 424 can be used to derive the amount of water evaporation, the amount of volatile matter release, and the amount of fixed carbon combustion. This allows for the construction of a state-space model relating to the state of the waste, and enables the estimation of the masses of water, volatile matter, and fixed carbon in the waste 52 on the grate. If the mass and volume of the waste 52 on the grate can then be derived by other methods, the mass of ash can be derived by subtracting the masses of water, volatile matter, and fixed carbon from the mass of the waste 52 on the grate. The ash falls through the ash outlet 105 and is discharged outside the furnace 101. In other words, an ash content model can be constructed as a state-space model capable of deriving the mass of ash from the waste reaction model 424, a model for deriving the mass of the waste 52 on the grate, and the amount of ash falling from the ash outlet 105.

[0060] Next, in order to derive the volume and mass of the waste 52 on the grate, it is necessary to perform boundary identification processing to distinguish the waste 50 on the grate 104 from other areas. Therefore, before describing the method for deriving the volume and mass of the waste 52 on the grate, the boundary identification processing method will be described below.

[0061] (Boundary discrimination learning model) First, the boundary recognition learning model 423 stored in the memory unit 42 of the waste state estimation device 40 and the method for generating the same will be described. Figure 5 is a diagram showing an example of transparent image data of burning waste captured by the imaging unit 125 of this embodiment. Figure 6 is a diagram showing an example of boundary image data in which a boundary line has been generated for the transparent image data captured by the imaging unit 125 of this embodiment.

[0062] As shown in Figure 5, the imaging unit 125 captures images of the waste supply unit 112 and the waste before it is supplied onto the grate 104, the stepped wall 113, the grate 104 and the waste on the grate 104 52, and the furnace wall 101a within the furnace 101, with the flames passing through, and outputs them as transparent image data. As shown in Figure 6, the boundary recognition learning model 423 performs a process to generate boundary lines 531 for the boundaries between the waste 50 and other objects using the transparent image data captured by the imaging unit 125. In the example shown in Figure 6, boundary lines 531 are drawn at the boundaries between the waste on the grate 52 and the stepped wall 113, the furnace wall 101a, and the grate 104. By drawing boundary lines 531, the boundary image data can define the outer edge of the waste on the grate 52 that is present on the grate 104.

[0063] The data used to generate the boundary recognition learning model 423 consists of transparent image data captured by the imaging unit 125, and processed image data (hereinafter referred to as boundary image data) in which boundaries are identified from the transparent image data and the aforementioned boundary lines 531 are drawn. Preferably, the number of transparent image data and boundary image data are 100 or more each. That is, the boundary image data used for generation is image data in which the boundary lines 531 are drawn by an operator on the transparent image data for each boundary between the waste 50 and the waste supply unit 112, the stepped wall 113, the furnace wall 101a, and the grate 104. When generating the boundary recognition learning model 423, the transparent image data is used as the learning input parameter, and the boundary image data is used as the learning output parameter. The learning unit 413 of the control unit 41 generates the boundary recognition learning model 423 using the above-mentioned learning input parameter and learning output parameter as training data, by deep learning using a hierarchical convolutional neural network (hierarchical CNN) such as U-Net. The control unit 41 generates boundary image data from transparent image data based on the information learned by the learning unit 413. The learning unit 413 also updates the boundary recognition learning model 423 as appropriate using the input transparent image data and boundary image data obtained by the operator correcting or drawing the boundaries.

[0064] Furthermore, when the boundary discrimination learning model 423 includes a combustion state learning model, the input and output datasets used to generate the combustion state learning model are transparent image data or boundary image data as the learning input parameters, and extracted image data from which a predetermined combustion state has been extracted as the learning output parameters.

[0065] In other words, the input parameters for training the combustion state learning model are at least one of the transmission image data captured by the imaging unit 125 and the boundary image data generated by the boundary generation unit 411 based on the boundary recognition learning model 423. The output parameters for training used to generate the combustion state learning model are extracted image data obtained by an operator by drawing the boundary between the waste on the grate 52 and other areas, and the boundary between the waste on the grate 52 with a combustion temperature above a predetermined temperature and other areas containing the waste on the grate 52 with a combustion temperature below a predetermined temperature, on the transmission image data or boundary image data. The imaging unit 125, equipped with a thermal image camera or the like, can output transmission image data based on the temperature distribution of the waste 50. In this case, the output transmission image data can be output as image data in which the brightness is higher as the temperature increases and lower as the temperature decreases. This allows for the extraction of the waste on the grate 52 from the transmission image data, and the generation of extracted image data by extracting the high-temperature areas of the extracted waste on the grate 52 that are above a predetermined temperature as the combustion area. Thus, for transmission image data in which the temperature distribution can be determined by brightness, extracted image data obtained by extracting the waste 52 on the grate and the combustion region may be used as boundary image data.

[0066] The number of transparent image data, boundary image data, and extracted image data used for learning is preferably 100 or more, for example. The learning unit 413 of the control unit 41 generates a combustion state learning model using machine learning, such as deep learning using a neural network, with the above-mentioned learning input parameters and learning output parameters as training data. Based on the content learned by the learning unit 413, the control unit 41 generates extracted image data from the transparent image data or boundary image data. The learning unit 413 can also update the combustion state learning model as appropriate using the input transparent image data or boundary image data, and the extracted image data obtained by the operator correcting or drawing the boundary.

[0067] (Method for estimating the state of waste) Next, a waste state estimation method, which is an information processing method performed by the waste state estimation device 40 of the waste treatment system 1 configured as described above, will be explained. In the waste state estimation method, the volume and mass of the waste 52 on the grate can be derived. Figures 7 and 8 are flowcharts for explaining the waste state estimation method according to this embodiment. In the following explanation, Figures 5 and 6 will also be referred to as appropriate. In this embodiment, step ST1 is the imaging unit 125 in the furnace 101, and steps ST2 to ST6 and ST11 to ST15 are processes performed by the waste state estimation device 40.

[0068] As shown in Figure 7, in step ST1, the imaging unit 125 images the inside of the furnace 101. The imaging unit 125 captures the conditions inside the furnace 101 within its field of view as an image transmitted through the flame. Specifically, as shown in Figure 5, the imaging unit 125 images the waste before supply 51 in the waste supply unit 112, the stepped wall 113, the waste on the grate 52, the grate 104, and the furnace wall 101a as images transmitted through the flame. Note that the waste before supply 51 does not need to be imaged. The imaging unit 125 transmits the captured transmitted image data to the control unit 41 through the input / output unit 44 of the waste state estimation device 40. The control unit 41 stores the received transmitted image data in the image database 421 of the storage unit 42.

[0069] As shown in Figure 6, the imaging unit 125 has a measurement field of view that extends in the vertical direction (height direction: z direction), the horizontal direction (furnace width direction: x direction) of the furnace 101, and the front-to-back direction (conveyance direction: y direction) on the grate 104. In this embodiment, the field of view of the imaging unit 125 includes at least the waste supply unit 112, the stepped wall 113, the grate 104, and the furnace wall 101a. The furnace wall 101a included in the field of view of the imaging unit 125 restricts the outward movement, i.e., spread, of the waste 50 in the horizontal direction. The field of view of the imaging unit 125 only needs to have a field of view that can image the boundary between the waste 52 on the grate 104 and the furnace wall 101a, the grate 104, and the stepped wall 113. Furthermore, it is preferable that the imaging unit 125 can image the waste 50 that has been conveyed to the waste supply unit 112. This makes it possible to image the waste 50 falling at the position of the stepped wall 113.

[0070] Furthermore, the combustion calculation unit 412 of the waste state estimation device 40 sets the coordinates of the transport direction (y direction) of the grate 104 with respect to the position x in the left-right direction (x direction) of the lower end of the stepped wall 113, i.e., the intersection of the stepped wall 113 and the grate 104. That is, coordinates (x,y) are set with respect to the upper surface of the grate 104 in the two-dimensional screen of the captured image data. Here, the x and y directions, as predetermined axes, are usually orthogonal to each other, but are not necessarily limited to being orthogonal. Also, the z direction is set in the vertical direction (height direction) of the furnace 101 with respect to the plane set by the x and y directions. Here, the z direction and y direction are usually orthogonal to each other, but are not necessarily limited to being orthogonal. Similarly, the z direction and x direction are usually orthogonal to each other, but are not necessarily limited to being orthogonal. With these settings, three-dimensional coordinates (x,y,z) can be set for the two-dimensional screen of the captured image data. As a result, the combustion calculation unit 412 can define the height of the waste 50 piled near the stepped wall 113 in a z coordinate corresponding to the x coordinate, and define the downstream end on the grate 104 along the transport direction in a y coordinate corresponding to the x coordinate, in a two-dimensional view of the captured image data.

[0071] Next, returning to Figure 7, we proceed to step ST2. The boundary generation unit 411 of the waste state estimation device 40 reads the boundary identification learning model 423 and determines the boundaries between the waste 50, the waste supply unit 112, the stepped wall 113, the furnace wall 101a, and the grate 104 based on the transparent image data acquired from the imaging unit 125 (see Figure 5). Alternatively, the boundary generation unit 411 may use a rule-based image recognition algorithm to mutually identify the waste 50, the waste supply unit 112, the stepped wall 113, the furnace wall 101a, and the grate 104 and determine their boundaries.

[0072] Next, the boundary generation unit 411 draws boundary lines 531 on the transparent image data for each boundary between the area where the waste 50 exists and other objects, specifically between the waste 50 and the waste supply unit 112, the stepped wall 113, the furnace wall 101a, and the grate 104. As a result, as shown in Figure 6, the boundary generation unit 411 draws boundary lines 531 on the transparent image data for, for example, the boundary between the stepped wall 113, the grate 104, and the furnace wall 101a and the waste 52 on the grate. In other words, the boundary generation unit 411 draws boundary lines 531 so as to surround the outer edge of the waste 52 on the grate. As a result, the boundary generation unit 411 generates boundary image data on which boundary lines 531 have been drawn on the transparent image data. The boundary generation unit 411 outputs the generated boundary image data to the combustion calculation unit 412.

[0073] Next, moving to step ST3, the control unit 41 of the waste state estimation device 40 derives the volume of the waste on the grate 52 from the partial volume. That is, first, the combustion calculation unit 412 calculates the height of the layer of waste on the grate 52 (hereinafter referred to as the waste layer height z). Specifically, the combustion calculation unit 412 of the waste state estimation device 40 calculates the waste layer height z based on the boundary portion between the stepped wall 113 and the waste on the grate 52 at the boundary line 531. Here, the boundary portion between the stepped wall 113 and the waste on the grate 52 at the boundary line 531 is uneven. The combustion calculation unit 412 derives the coordinate z0 of the height of the boundary line 531 corresponding to an arbitrary position coordinate x0 in the x direction along the intersection portion of the grate 104 and the stepped wall 113, which is set in advance in the boundary image data.

[0074] Next, the combustion calculation unit 412 functions as a combustion point position measuring device and measures the position of the combustion point, which is the position of the waste 52 on the grate 104 in the direction of transport. That is, the boundary line 531a at the front position of the boundary line 531 between the waste 52 on the grate and the grate 104 becomes the specific combustion point of the waste 50. Note that the specific combustion point of the waste 50 is usually uneven. The combustion calculation unit 412 derives the coordinate y0 of the position in front of the boundary line 531a corresponding to an arbitrary position coordinate x0 in the x direction along the intersection of the grate 104 and the stepped wall 113, which is set in advance in the boundary image data. Note that the above coordinate (y0, z0) derivation process is not limited to the order described above, and may be performed in parallel, in reverse order, or in any order.

[0075] Subsequently, the combustion calculation unit 412 derives three points based on the derived coordinate z0 of the height of the boundary line 531 and the coordinate y0 of the position in front of the boundary line 531a, which correspond to the x-direction position coordinate x0. Here, the three points corresponding to the x-direction position coordinate z0 of the height of the boundary line 531 and the coordinate y0 of the position in front of the boundary line 531a will be explained.

[0076] Specifically, the combustion calculation unit 412 defines point A(x0,0,0) as a predetermined position x0 in the x-direction along the intersection of the stepped wall 113 and the grate 104 in the waste 52 on the grate 104 in the boundary image data. The combustion calculation unit 412 also defines point B(x0,0,z0) as the position of the waste 52 on the grate 52 near the stepped wall 113 corresponding to point A (the upper end of the boundary line 531). Furthermore, the combustion calculation unit 412 defines point C(x0,y0,z0) as the combustion point corresponding to point A in the boundary image data (a position just before the boundary line 531a).

[0077] The combustion calculation unit 412 sets a three-vertex closed curve in the boundary image data, consisting of a simple closed curve connecting points A, B, and C. The three-vertex closed curve ABC corresponds to the longitudinal section of the waste 52 on the grate 104. The predetermined longitudinal section is usually preferably a plane perpendicular to the x-direction, but is not limited to that. Here, for example, a triangle with sides AB, AC, and BC being straight lines is set as the three-vertex closed curve. The angle of vertex angle A is usually 90°, but is not necessarily limited to that. Side AC can be set based on the shape of the upper surface of the grate 104, and side AB can be set based on the shape of the side of the stepped wall 113. Side BC may be set as an upwardly rising curve (convex upward curve) or as a downwardly concave curve (convex downward curve).

[0078] Furthermore, the combustion calculation unit 412 derives the area of ​​a three-vertex closed curve ABC, such as a triangle, with points A, B, and C as its vertices. Since the area S(x) of the three-vertex closed curve ABC changes along the x-direction, it can be expressed by the following equation (7). Area of ​​a three-vertex closed curve ABC = S(x) …(7)

[0079] The combustion calculation unit 412 performs the derivation of points A, B, and C, and the derivation of the area of ​​the three-vertex closed curve ABC, at predetermined intervals Δx along the x-direction. The predetermined interval Δx can be set to any width in the boundary image data. If the predetermined interval Δx is set to an interval less than or equal to the width of one pixel constituting the boundary image data, an interpolated image using any interpolation method, such as a bicubic filter, can be used. In addition, in the boundary image data, which is two-dimensional data, the unit length in the x-direction on the front side of the grate 104 in the transport direction (y-direction) is greater than the unit length in the x-direction at the intersection of the stepped wall 113. That is, in the boundary image data, the width in the x-direction increases along the y-direction. In this case, the predetermined interval Δx may be set to a minimum of one pixel along the x-direction at the intersection of the stepped wall 113, or to one pixel along the x-direction in the front side of the grate 104 in the transport direction. The size of one pixel in the captured image data and boundary image data can be determined by the imaging optical system, and is, for example, about 1.1 cm to 10 cm.

[0080] Next, the combustion calculation unit 412 derives a partial minute volume ΔV of the waste on the grate 52 based on the area of ​​the three-vertex closed curves derived at predetermined intervals Δx along the x-direction. That is, the combustion calculation unit 412 first sets a three-vertex closed curve, such as a right triangle, using the height z0 (point B) of the boundary line 531, which corresponds to the height of the waste on the grate 52 and is corresponding to the horizontal position x0 (point A), and the position y0 (point C) in front of the boundary line 531a. The combustion calculation unit 412 also sets a three-vertex closed curve using the height z1 of the boundary line 531, which corresponds to the horizontal position x1 in the x-direction, and the position y1 in front of the boundary line 531a. Here, these three-vertex closed curves may be similar to each other or may have different shapes.

[0081] First, if the three-vertex closed curve ABC is a triangle, and the three-vertex closed curve set corresponding to the horizontal position x1 is also a triangle, then the partial volume ΔV can be derived by calculating the volume of the truncated triangular pyramid shown in equation (8) below. Note that S(x) is the area of ​​the triangle corresponding to the horizontal position x.

[0082]

number

[0083] Based on equation (8), the volume V of the grate waste 52 deposited on the grate 104 can be derived by summing the partial volumes ΔV along the x-direction over the entire width of the stepped wall 113, as shown in equation (9) below. This makes it possible to estimate the volume V of the grate waste 52 by approximating it using equations (8) and (9). V = Σ(total width of the stepped wall 113) ΔV …(9)

[0084] Furthermore, regardless of whether the three-vertex closed curve consisting of horizontal position x0 and the corresponding height z0 and foreground position y0 is similar to the three-vertex closed curve consisting of horizontal position x1 and the corresponding height z1 and foreground position y1, the partial volume ΔV can generally be derived by the following equation (10). This is true whether the three-vertex closed curves set at multiple arbitrary horizontal positions x are similar to each other or not. In equation (10), S(x) is the area of ​​the three-vertex closed curve corresponding to horizontal position x.

[0085]

number

[0086] Based on equation (10), the volume V of the grate waste 52 accumulated on the grate 104 can be derived by integrating the partial volume ΔV along the x-direction over the entire width of the stepped wall 113, as shown in equation (11) below. This makes it possible to estimate the volume V of the grate waste 52 by approximating it using equations (10) and (11). L is the total width of the stepped wall 113. The combustion calculation unit 412 stores the derived estimated value of the volume V of the grate waste 52 in the combustion information database 422.

[0087]

number

[0088] Furthermore, by setting a three-vertex closed curve between the height z, which is the upper part of the boundary line 531, and the foreground position y, which is the lower part of the boundary line, the overall shape of the waste 52 on the grate can be approximated. This makes it possible to accurately estimate the distribution shape of the waste 50 on the grate 104. The data of the distribution shape of the waste 52 on the grate is stored in the combustion information database 422. Therefore, the control unit 41 reads the estimated value of the volume V of the waste 52 on the grate and the estimated distribution shape information of the waste 52 on the grate from the combustion information database 422 and transmits it to the combustion control device 30. The combustion control device 30 can control the waste incinerator 100 with high precision based on the acquired estimated value of the volume of the waste 52 on the grate and the estimated distribution shape information. As a result, the combustion calculation unit 412 can estimate an approximate value of the volume V of the waste 52 on the grate.

[0089] As described above, the volume of the waste 52 on the grate being transported on the grate 104 can be derived. Furthermore, the volume of the waste 50 dropped onto the grate 104 from the pre-supply waste 51 can be derived. Thus, the mass of each part of the waste 52 on the grate can be derived from the volumes of the transported waste 52 on the grate and the waste 50 supplied onto the grate 104 from the pre-supply waste 51, and the specific gravity that can be estimated according to the surface temperature of the waste 52 on the grate obtained from thermal image data, or the average value of the bulk specific gravity of the pre-supply waste 51 in the waste supply unit 112, which is an example of process data. In other words, the combustion calculation unit 412 derives the mass of the waste 52 on the grate for each part by multiplying the volume of the waste 52 on the grate in each calculated part by the specific gravity that can be estimated according to the observed surface temperature. For example, as described above, since the grate waste 52 is divided into fine parts by its longitudinal section, the control unit 41 can derive the individual mass of each divided part of the grate waste 52.

[0090] Next, the control unit 41 proceeds to step ST4 shown in Figure 7. In step ST4, the waste reaction model 424 is read and calculations are performed to derive the state of the waste 52 on the grate in the furnace 101. Figure 8 is a flowchart detailing the calculation steps of the waste reaction model in step ST4. Figure 9 is a diagram illustrating the division and setting of the region on the grate 104 in the furnace 101 according to this embodiment.

[0091] As shown in Figure 9, the grate 104 is divided into m regions (m is a natural number) in the furnace width direction, which is perpendicular to the furnace length direction, which is the direction in which the waste 52 on the grate is transported. Furthermore, each of the regions divided in the furnace width direction is divided into N regions (N is a natural number) in the transport direction (forward / backward direction: y direction). For the sake of explanation, in the following description, the transport direction may be referred to as the furnace length direction. In the example shown in Figure 9, with m=10, each of the 10 regions divided in the furnace width direction is divided into 18 regions along the transport direction (forward / backward direction: y direction) with N=18, but the number of regions is not limited to these numbers. The numbers (j: 1~N) written along the furnace length direction in Figure 9 correspond to each region divided in the furnace length direction, and the numbers (i: 1~m) written along the furnace width direction correspond to each region divided in the furnace width direction. The control unit 41 sequentially selects regions in the furnace width direction in a loop process, and for each selected i-th region, it derives the state of the waste 52 on the grate in each region (1st region to Nth region) in step ST4.

[0092] The state of the waste derived in step ST4 is the content of at least one of the following: the amount of moisture contained in the grate waste 52, the amount of volatile components of the pyrolysis gas that can be released from the grate waste 52, and the amount of fixed carbon contained in the grate waste 52. The content of ash as the residue after combustion of the grate waste 52 may also be included. Furthermore, it is desirable to adopt the respective content rates of moisture, volatile components, fixed carbon, and ash contained in the grate waste 52 as the state of the waste.

[0093] In the state shown in Figure 9, first, waste is pushed out by a waste supply device 103 such as a pusher and dropped onto the grate 104 for supply. The waste 52 on the grate is moved by the grate 104. This movement of the grate 104 moves the waste 52 on the grate to the ash discharge side, and the ash is discharged from the ash discharge port 105. Combustion air is also supplied to the grate 104 from below.

[0094] As shown in Figure 9, the combustion calculation unit 412 sets up multiple pre-calculated divided regions and multiple region numbers for the waste 50 (waste on the grate 52) on the grate 104 in the furnace 101, and then starts the calculation of the waste reaction model 424.

[0095] As shown in Figure 8, the combustion calculation unit 412 sets the state of the waste at time (t-1) in step ST11. Specifically, for the i-th region selected in the furnace width direction, the unit sets the percentage content of moisture, volatile matter, fixed carbon, and ash, as well as the waste mass (or volume), for each waste 52 on the grate in the first to nth regions in the furnace length direction. In other words, for the i-th region selected in the furnace width direction, the unit obtains the state of the waste and the amount of waste for each region (first to nth regions) on the grate 104. When performing the initial calculation, the initial values ​​of the waste state in the first to nth regions of the i-th region can be set based on probable values. Furthermore, the amount of waste in each region (1st to Nth regions) of the i-th region on the grate 104 can be derived based on at least one of the thermal image data obtained by imaging the inside of the waste incinerator 100 or process data which is sensor values ​​measured by various sensors.

[0096] Next, the combustion calculation unit 412 moves to step ST12 to derive the state of the waste at time t. That is, for the selected i-th region, the combustion calculation unit 412 applies the waste reaction model (moisture evaporation model) to each of the first to nth regions to derive the amount of moisture evaporated from the waste 52 on the grate. Subsequently, for the selected i-th region, the combustion calculation unit 412 applies the waste reaction model (volatile matter release model) to each of the first to nth regions where moisture evaporation has been completed to derive the amount of volatile matter released. Furthermore, for the i-th region, the combustion calculation unit 412 applies the waste reaction model (fixed carbon combustion model) to each of the first to nth regions to derive the amount of fixed carbon burned.

[0097] Next, the combustion calculation unit 412 moves to step ST13 to derive the movement state of the waste 52 on the grate at time t. That is, the waste 51 before supply is dropped onto the grate 104 by the waste supply device 103. In this case, the combustion calculation unit 412 adds the mass of the supplied waste 50 at the location where the waste 50 was supplied (for example, the first to third regions). The location and mass of the dropped waste 50 can be derived by the combustion calculation unit 412 from the thermal image data based on the method for deriving the volume of the waste 50 described above. Alternatively, the location and mass of the dropped waste 50 can be derived by the combustion calculation unit 412 from the process data. Next, the calculation is performed assuming that a portion of the waste 52 on the grate in the j-th region (j=1 to N-1) of the selected i-th region moves to the (j+1)-th region. A portion of the waste 52 on the grate that is in the N-th region is discharged as ash through the ash outlet 105.

[0098] Next, the combustion calculation unit 412 moves to step ST14 to adjust and correct the waste state parameters (state variables) based on the observed values ​​(observed variables) at time t. Here, examples of observed values ​​used for the waste reaction and data assimilation techniques include, as process data, the temperature and pressure inside the furnace 101 and other parts, the gas flow rates of various gases, and the component concentrations of each gas. Furthermore, as process data, the blown air temperature, pressure, and flow rate of each part, as well as the grate speed v, the speed of the waste supply device 103 (pusher speed), the amount of water vapor evaporated, the amount of waste 50 processed, and the calorific value can also be included. Note that the process data is not necessarily limited to the examples described above.

[0099] In this embodiment, the observed variables can be the amount of grate waste 52 present in each region (1st to Nth regions) of the i-th region on the grate 104, and the distribution of surface temperature, obtained from thermal image data captured by the imaging unit 125, and the sensible heat of the combustion gas at the furnace outlet 107 measured by the furnace outlet gas thermometer 121. The term "amount present" includes the concepts of mass (weight) and volume, and may mean at least one of mass and volume. Furthermore, the combustion calculation unit 412 can derive the amount of grate waste 52 supplied to each region on the grate 104. The moving speed of the grate waste 52 used here can be derived from the grate velocity v of the grate 104 by the combustion calculation unit 412.

[0100] The combustion calculation unit 412 calculates the surface temperature T of the waste on the grate in each of the first to N regions for the selected i-th region, based on the flame conditions formed in the furnace 101 by the release of volatile components from the waste on the grate 52. s The following can be derived. The temperature of the waste 52 on the grate in each of the 1st to Nth regions in the selected i-th region is the surface temperature T based on the information obtained from the thermal image data captured by the imaging unit 125. s And the measurable grate temperature T in the grate 104. gIt can be determined from the following. The temperature of the waste 52 on the grate can be derived, for example, from equation (12) below, but is not necessarily limited to equation (12). Temperature of waste 52 on the grate = (Surface temperature T) s +Grate temperature T g ) / 2 …(12)

[0101] Specifically, in step ST14, the combustion calculation unit 412 constructs a state-space model based on the waste reaction model 424 described above in order to estimate the state of the waste. That is, as state variables for the state of the waste, it adopts the mass (weight) and the mixing ratio (percentage) of moisture content, volatile content, fixed carbon content, and ash content of the waste 52 on the grate in each of the first to ninth regions of the selected i-th region. Alternatively, the mass and moisture content of the waste 52 on the grate in each of the first to ninth regions of the i-th region may be adopted as state variables for the state of the waste. Similarly, the mass and volatile content of the waste 52 on the grate in each of the first to ninth regions of the i-th region may be adopted as state variables for the state of the waste. Furthermore, the mass and fixed carbon content of the waste 52 on the grate in each of the first to ninth regions of the i-th region may be adopted as state variables for the state of the waste. The ash content is derived as the remainder of the moisture content, volatile content, and fixed carbon content in the state of the waste.

[0102] In this embodiment, as a method for estimating state variables in the constructed state-space model, an infinite impulse response filter is employed that can estimate continuously changing quantities, in this case the position of the waste 52 on the grate and the mass of each component contained therein, from observations with discrete errors. Here, it is preferable to employ a Kalman filter as the infinite impulse response filter, and in this embodiment, a UKF (Unscented Kalman Filter) is applied as one of the nonlinear Kalman filters, but other methods may also be applied.

[0103] Specifically, the combustion calculation unit 412 estimates the values ​​of each state variable using data assimilation processing such as a Kalman filter, so as to correct the error between the observed values ​​and the waste weight, volatile matter release distribution, and volatile matter release amount in each of the first to N regions, for the selected i-th region on the grate 104 obtained for a state variable at a certain time t.

[0104] In other words, the combustion calculation unit 412 matches the weight of the waste on the grate 52, derived based on the thermal image data captured by the imaging unit 125 as observed values, in each of the first to nth regions of the selected i-th region on the grate 104, with the mass of the waste on the grate 52 obtained based on the state variable model for a certain time t. Furthermore, the combustion calculation unit 412 determines the surface temperature T of the waste on the grate 52 based on the thermal image data obtained as observed values ​​in each of the first to nth regions of the selected i-th region on the grate 104. s The distribution of the state variables is matched with the distribution of volatile matter release obtained for a given state variable at a certain time t. Furthermore, the combustion calculation unit 412 makes the value obtained by subtracting the sensible heat of the blown air from the sensible heat of the combustion gas at the outlet of the furnace 101, obtained as an observed value, proportional to the amount of volatile matter release obtained for a given state variable at a certain time t. This corrects the discrepancy in the state variables, which are parameters, and allows for the derivation of a more accurate state of the waste at time t in the state-space model. Note that these waste state estimation processes can be performed immediately (in real time) while the furnace 101 is in operation.

[0105] Subsequently, the process moves to step ST15, where the state of the waste at time t is set to the state at time t-1 in order to make the state at the next time t+1 the initial state. This allows the combustion calculation unit 412 to perform the setting of the state of the waste at time (t-1) in step ST11.

[0106] After performing the process in step ST4 for each of the first to mth regions in the furnace width direction, the combustion calculation unit 412 proceeds to step ST5 shown in Figure 7 to smooth the derived waste state and adjust the waste mass. Here, the combustion calculation unit 412 performs a smoothing process, for example, using a Gaussian filter, on the waste mass (weight), moisture content, volatile content, fixed carbon content, and ash content estimated by the state-space model.

[0107] Figure 10 shows an example of a filter applied to the smoothing process. Specifically, the Gaussian filter shown in Figure 10 is applied to each of the first to m regions in the furnace width direction, and to each of the first to N regions in the furnace length direction, to smooth the estimated state variables.

[0108] Figure 11 shows a divided region on the grate 104. For example, when smoothing the moisture content w5 of the waste in the j-th region in the furnace length direction from the i-th region in the furnace width direction shown in Figure 11, a 3×3 Gaussian filter shown in Figure 10 is applied to the (i,j) region, and the moisture content in the (i,j) region is smoothed. In this case, the moisture content w after smoothing in the (i,j) region is obtained by equation (13).

[0109]

number

[0110] The combustion calculation unit 412 applies a Gaussian filter to the volatile matter, fixed carbon, and ash content, similar to the moisture content, to smooth out the volatile matter, fixed carbon, and ash content of the waste. The smoothed moisture, volatile matter, fixed carbon, and ash content of the waste are then set as the state of the waste at time (t-1) when the process in step ST11 is performed next. Note that the filter applied in this smoothing process may be a filter other than a Gaussian filter, for example, a median filter.

[0111] Next, the combustion calculation unit 412 adjusts the total mass of the waste and the masses of the first to m regions in the furnace length direction in each of the first to m regions in the furnace width direction, so that the total mass of the waste is preserved before and after smoothing. Specifically, the combustion calculation unit 412 derives the total mass of the waste before smoothing using equation (14) and adjusts the total mass of the waste after smoothing using equation (15). The combustion calculation unit 412 also adjusts the mass of each region using equation (16).

[0112]

number

[0113]

number

[0114]

number

[0115] Here, M_total is the total mass of waste before smoothing, M_i,j is the mass of the (i,j) region on the grate 104 before smoothing, M_total_smooth is the total mass of waste after smoothing, and M_i,j_smooth is the mass of the (i,j) region on the grate 104 after smoothing. Also, M_i,j_new is the corrected mass of waste in the (i,j) region, N is the number of regions in the furnace width direction, and M is the number of regions in the furnace length direction. Note that the smoothing and mass adjustment processes in step ST5 may be performed at multiple time intervals rather than at one time interval.

[0116] In addition, the combustion calculation unit 412 adjusts the amounts of water, volatile matter, fixed carbon, and ash in the first to m regions in the furnace width direction and the first to N regions in the furnace length direction, in the same manner as the mass.

[0117] Furthermore, simply dividing the grate 104 into the furnace length direction and the furnace width direction and performing calculations for each region does not take into account interactions such as the movement of waste between adjacent regions, which may lead to inconsistencies in the estimation results between adjacent regions. On the other hand, in the present invention, by performing this smoothing process, homogenization is achieved between adjacent regions, so that interactions between adjacent sections can be simulated. In addition, in the present invention, the smoothing process not only smooths regions adjacent in the furnace width direction, but also applies a two-dimensional filter to perform smoothing in the furnace length direction as well as the furnace width direction. Therefore, after smoothing, the occurrence of inconsistencies not only in the furnace width direction but also in the furnace length direction can be suppressed.

[0118] Next, the combustion calculation unit 412 executes the process in step ST5, then proceeds to step ST6 shown in Figure 7, where it outputs the percentages of moisture, volatile matter, fixed carbon, and ash in the waste. This completes the estimation process of the waste state in the waste 52 on the grate inside the furnace 101. Note that the flowcharts in Figures 7 and 8 are executed repeatedly as long as the furnace 101 continues to operate.

[0119] Figure 12 is a graph showing the state of the waste 52 on the grate in a certain region in the furnace width direction at a certain time t, as derived as described above. In the example shown in Figure 12, the region on the grate 104 is divided into 18 regions (N=18) in the furnace length direction, and the amount of waste, moisture content, volatile matter, fixed carbon, and ash content are shown for each region from the 1st to the 18th region. The graph shown in Figure 12 will change in real time as time t progresses.

[0120] Furthermore, Figures 13 and 14 are graphs showing the amount of water evaporated and volatile matter released per unit time, respectively. Also, Figure 15 is a graph showing the total amount of waste 52 on the grate of the waste incinerator 100, the amount of waste 50 dropped and supplied onto the grate 104, and the state of the waste in each region corresponding to Figure 12.

[0121] The combustion calculation unit 412 can derive the ratio of water content in the waste shown in Figure 12 based on the amount of water evaporation in each of the 1st to 18th regions shown in Figure 13. Conversely, the combustion calculation unit 412 can derive the amount of water evaporation in each of the 1st to 18th regions shown in Figure 13 by differentiating the ratio of water content in the waste shown in Figure 12 with respect to time. Similarly, the combustion calculation unit 412 can derive the ratio of volatile matter release in the waste shown in Figure 12 based on the amount of volatile matter release in each of the 1st to 18th regions shown in Figure 14. Conversely, the combustion calculation unit 412 can derive the amount of volatile matter release in Figure 14 by differentiating the ratio of volatile matter release in the waste in each of the 1st to 18th regions shown in Figure 12 with respect to time. Furthermore, as shown in Figures 12 and 15, by measuring the total amount of waste 50 and the amount supplied onto the grate 104, the graph of the waste state corresponding to Figure 12 changes over time.

[0122] As described above, according to this embodiment, the waste state estimation device 40 constructs a state-space model using the constructed waste reaction model, acquired process data on the grate, and process data inside the incinerator, and performs a waste state estimation process that estimates the amount of waste, the moisture content of the waste, the combustibility of the waste, and the calorific value of the waste at each of the multiple divided positions on the grate using data assimilation technology with a Kalman filter. The combustion control device 30 can then control the furnace 101 with a more accurate understanding of the state of the waste 50 inside the furnace 101 based on the state estimated by the waste state estimation device 40, thus enabling more appropriate control.

[0123] (Recording medium) In the above-described embodiment, a program capable of executing the management method by the combustion control device 30 and the waste state estimation device 40 can be recorded on a recording medium readable by a computer or other machine or device (hereinafter referred to as "computer, etc."). By having the computer, etc. read and execute the program on the recording medium, the computer functions as the waste state estimation device 40 and the combustion control device 30. Here, a recording medium readable by a computer, etc. refers to a non-temporary recording medium that stores information such as data and programs by electrical, magnetic, optical, mechanical, or chemical action and can be read by a computer, etc. Examples of such recording media that can be removed from a computer, etc. include flexible disks, magneto-optical disks, CD-ROMs, CD-R / Ws, DVDs, BDs, DATs, magnetic tapes, and memory cards such as flash memory. In addition, recording media fixed to a computer, etc. include hard disks and ROMs. Furthermore, SSDs can be used as both a recording medium that can be removed from a computer, etc. and a recording medium that is fixed to a computer, etc.

[0124] Furthermore, the program to be executed by the combustion control device 30 and waste state estimation device 40 according to one embodiment may be configured to be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network.

[0125] [Differentiation] Although embodiments of the present invention have been described above, the present invention is not limited to the embodiments described above and can be implemented in various other forms. For example, the present invention may be implemented by modifying the embodiments described above as follows. The embodiments described above and the following modifications may be combined with each other. The present invention is also included in configurations that appropriately combine the components of each embodiment and each modification described above. Furthermore, further effects and modifications can be easily derived by those skilled in the art. Therefore, broader embodiments of the present invention are not limited to the embodiments and modifications described above, and various modifications are possible.

[0126] (First variation) Next, a modified example of the above-described embodiment will be explained. In the first modified example, the three-vertex closed curve ABC may be an obtuse triangle where the angle θ of the vertex angle A, which is the intersection of the stepped wall 113 and the grate 104, is greater than 90°. In this case as well, an estimated value of the volume V of the waste on the grate can be derived as an approximate value by appropriately using equations (7) to (11) above from the area S(x) of the vertical cross-section of the waste on the grate 52 corresponding to the horizontal position x of the stepped wall 113.

[0127] (Second variation) Furthermore, in the second modified example, the three-vertex closed curve ABC may be a simple closed curve in which the vertex angle A, which is the intersection of the stepped wall 113 and the grate 104, is set to 90°, and the side BC has an upward-curving shape. Here, the side BC can be a curve that follows various upward-convex functions, such as a curve that follows a part of an nth-degree polynomial function (where n is an integer greater than or equal to 2) or a trigonometric function. The choice of which function the curve of side BC follows can be selected based on the properties of the waste 50. In this case as well, an estimated value of the volume V of the waste 52 on the grate can be derived as an approximate value by appropriately using equations (7) to (11) described above, from the area S(x) of the vertical cross-section of the waste 52 on the grate corresponding to the horizontal position x of the stepped wall 113.

[0128] (Third variation) Furthermore, in the third modified example, the three-vertex closed curve ABC may be defined as an acute triangle where the angle θ of the vertex angle A, which is the intersection of the stepped wall 113 and the grate 104, is less than 90°. In this case as well, an estimated value of the volume V of the waste on the grate can be derived as an approximate value by appropriately using equations (7) to (11) described above, from the area S(x) of the vertical cross-section of the waste on the grate 52 corresponding to the horizontal position x of the stepped wall 113.

[0129] (Fourth variation) Furthermore, in the fourth modified example, the three-vertex closed curve ABC is a simple closed curve consisting of a curve with a vertex angle A, which is the intersection of the stepped wall 113 and the grate 104, having an angle θ of 90° and side BC having a downward-curving shape. Here, side BC can be a curve that follows various downward-convex functions, such as an exponential function, an inverse proportional function, an nth-degree function (where n is an integer greater than or equal to 2), or a curve that follows a part of a trigonometric function. The choice of which function side BC follows can be made based on the properties of the waste 50. In this case as well, an estimated value of the volume V of the waste 52 on the grate can be derived as an approximate value by appropriately using equations (7) to (11) above from the area S(x) of the vertical cross-section of the waste 52 on the grate corresponding to the horizontal position x of the stepped wall 113.

[0130] The first to fourth variations described above can be combined as appropriate. That is, the angle θ between sides AB and AC can be applied to the second and fourth variations, or the curved shape of side BC can be applied to the first and third variations.

[0131] Furthermore, in the combustion control device 30 and waste state estimation device 40 according to one embodiment, the above-mentioned "parts" can be read as "circuits" or the like. For example, the communication unit can be read as a communication circuit.

[0132] Further effects and modifications can be readily derived by those skilled in the art. Broader aspects of the present invention are not limited to the specific details and representative embodiments expressed and described above. Accordingly, various modifications are possible without departing from the spirit or scope of the overall concept of the invention as defined by the appended claims and their equivalents. For example, the types of numerical values ​​and information given in the above-described embodiment are merely examples, and different types of numerical values ​​and information may be used as needed, and the present invention is not limited by the descriptions and drawings that constitute part of the disclosure of the present invention in the above-described embodiment.

[0133] In the embodiment described above, the state of the waste was defined as the ratio of moisture content, volatile matter content, fixed carbon content, and ash content. However, it is also possible to define the state of the waste as the ratio of at least two amounts selected from moisture content, volatile matter content, fixed carbon content, and ash content.

[0134] For example, in the embodiment described above, deep learning using a neural network is adopted as an example of machine learning, but machine learning based on other methods may also be performed. For example, other supervised learning methods such as support vector machines, decision trees, Naive Bayes, and k-nearest neighbors may be used. Also, semi-supervised learning may be used instead of supervised learning. [Explanation of symbols]

[0135] 1. Waste disposal system 30 Combustion control device 40. Waste condition estimation device 50 waste 51 Pre-supply waste 52 Waste on the grate 100 waste incinerators 101 Furnace 104, 1041-1043 Fire Grates 106 Combustion Air Blower 112 Waste Supply Department 113 Step wall 114 Combustion air damper 1141-1143 Air damper for combustion below the grate 125 Imaging Unit 126a Preheater 126b Bypass Damper 127 Recirculation Blower 128 Exhaust gas recirculation air damper 1301~1303 Wind box

Claims

1. It includes a control unit with hardware, The control unit, Image data generated from thermal image information captured of the area containing the waste inside a waste incinerator equipped with a grate for moving the waste is acquired. The process data in the aforementioned waste incinerator is acquired, The grate is virtually divided into multiple regions along the width direction of the waste incinerator, and each of the regions virtually divided along the width direction is virtually divided into multiple regions along the length direction of the waste incinerator. Based on at least one of the acquired image data and process data, the amount of waste present on the grate, the amount of waste supplied to the grate, and the temperature of the waste are derived for each of the multiple regions. Based on the acquired process data, the amount of waste present on the grate, the amount of waste supplied to the grate, and the temperature of the waste, state variables of the waste are derived for each of the multiple regions. The state variables for each of the multiple regions are smoothed using the state variables of the selected region and the regions adjacent to the selected region. Information processing device.

2. For each of the multiple regions, each of the smoothed state variables is adjusted based on the sum of the smoothed state variables. The information processing apparatus according to claim 1.

3. The state variables are the amount of volatile matter contained in the waste, the amount of water contained in the waste, the amount of fixed carbon contained in the waste, and the amount of ash contained in the waste. The information processing apparatus according to claim 1.

4. An information processing method performed by an information processing device equipped with a control unit having hardware, The steps include: acquiring image data generated from thermal image information captured of a region containing the waste inside a waste incinerator equipped with a grate for moving the waste; The steps include: acquiring process data in the aforementioned waste incinerator; The steps include: dividing the grate into multiple virtual regions along the width direction of the waste incinerator, and dividing each of the regions virtually divided along the width direction into multiple virtual regions along the length direction of the waste incinerator; A first derivation step, based on at least one of the acquired image data and process data, derives for each of the multiple regions the amount of waste present on the grate, the amount of waste supplied to the grate, and the temperature of the waste; A second derivation step of deriving state variables for the waste for each of the multiple regions based on the acquired process data, the amount of waste present on the grate, the amount of waste supplied to the grate, and the temperature of the waste, A smoothing step in which the state variables for each of the multiple regions are smoothed with the state variables of the selected region and the regions adjacent to the selected region, An information processing method that performs the following.

Images