WO2023063107A1

WO2023063107A1 - Control apparatus

Info

Publication number: WO2023063107A1
Application number: PCT/JP2022/036664
Authority: WO
Inventors: 信治岩下; 博幸高木; 稔彦瀬戸口; 潤司今田; 幸司滑澤; 慶一林; 知通江草
Original assignee: 三菱重工環境・化学エンジニアリング株式会社
Priority date: 2021-10-15
Filing date: 2022-09-30
Publication date: 2023-04-20
Also published as: TW202323728A; JP2023059470A; CN118159776A; KR20240073073A; JP7296436B2

Abstract

A control apparatus according to the present disclosure comprises: an information acquisition unit; a steam flowrate prediction unit; and a control unit. The information acquisition unit acquires information relating to a to-be-incinerated object yet to be supplied to a processing space in an incineration facility. The steam flowrate prediction unit predicts, on the basis of prediction information including the information acquired by the information acquisition unit, a main steam flowrate of steam to be generated by a boiler in the incineration facility. The control unit carries out combustion control on the basis of the main steam flowrate predicted by the steam flowrate prediction unit.

Description

Control device

The present disclosure relates to control devices.
This application claims priority to Japanese Patent Application No. 2021-169507 filed on October 15, 2021, the contents of which are incorporated herein.

In Patent Document 1, information indicating the current combustion state is obtained in real time, and based on this information, the amount of combustion heat and the amount of boiler evaporation are estimated, thereby enabling waste combustion control without time delay. A method for controlling combustion is disclosed. This combustion control method obtains the carbon, hydrogen, and moisture in the waste, which are directly related to the calorific value of the waste used as fuel, from the composition of the exhaust gas immediately after combustion, which is obtained in real time. It calculates the amount of combustion heat, the amount of latent heat, and the amount of waste (processed amount).

In Patent Document 2, images of a flame reaching a secondary combustion zone from a primary combustion zone are obtained using a plurality of imaging devices with different viewpoints, and image synthesis processing is performed on the plurality of images obtained from different viewpoints. A method for evaporative control is disclosed that includes creating a three-dimensional image containing the flame of the secondary combustion zone by performing . This evaporation amount control method analyzes the above three-dimensional image to calculate the time change of the flame flow velocity in the direction along the flow path of the combustion gas generated by the primary combustion or secondary combustion. It gives an index of the amount of heat that is being used.

In Patent Document 3, a plurality of infrared cameras are used to observe waste accumulated in at least the drying section and the burning section through a filter that selectively transmits light of wavelengths not emitted by flames, and the viewpoint is A combustion control method is disclosed that includes acquiring a plurality of different thermal images and creating a three-dimensional thermal image based on the plurality of thermal images. In this combustion control method, based on the three-dimensional thermal image, thickness progress information indicating how the thickness of the waste has changed in time series is calculated, and the change in the volumetric flow rate of the waste from the past to the present time is calculated. Based on this, the combustion correction coefficient is determined, and the index of the amount of heat generated from the waste is calculated.

JP 2017-096517 A Japanese Patent Application Laid-Open No. 2019-219108 JP 2021-067381 A

By the way, the main steam flow rate may fluctuate greatly depending on the condition of the incinerated material. Therefore, with the techniques described in Patent Documents 1 to 3, it may be difficult to perform combustion control based on a highly accurate predicted value of the main steam flow rate.

The present disclosure has been made to solve the above problems, and aims to provide a control device capable of performing combustion control based on a highly accurate predicted value of the main steam flow rate.

In order to solve the above problems, the control device according to the present disclosure includes an information acquisition unit, a steam flow rate prediction unit, and a control unit. The information acquisition unit acquires information about the incineration material before it is supplied to the processing space in the incineration facility. The steam flow rate prediction unit predicts the main steam flow rate generated by the boiler of the incineration facility based on the prediction information including the information acquired by the information acquisition unit. The control section performs combustion control based on the main steam flow rate predicted by the steam flow rate prediction section.

According to the control device of the present disclosure, combustion control can be performed based on a highly accurate predicted value of the main steam flow rate.

1 is a schematic configuration diagram showing the entirety of an incineration facility according to an embodiment of the present disclosure; FIG. 1 is a block diagram showing a functional configuration of combustion equipment according to an embodiment of the present disclosure; FIG. 4 is a block diagram showing the functional configuration of a data converter according to the embodiment of the present disclosure; FIG. FIG. 10 is a diagram showing a correlation between an estimated value of waste calorific value based on a detection result of a moisture meter according to an embodiment of the present disclosure and a waste calorific value confirmed in an actual machine; It is a figure which shows an example of the process by the 1st feature-value extraction part which concerns on embodiment of this indication. FIG. 4 is a diagram illustrating an example of processing by an image conversion unit according to the embodiment of the present disclosure; FIG. FIG. 7 is a diagram illustrating an example of processing by a dust layer height detection unit according to the embodiment of the present disclosure; FIG. 4 is a diagram showing an example of correlation between each piece of input information and main steam flow rate according to an embodiment of the present disclosure; FIG. 4 is a diagram showing an example of time delay setting values for each piece of input information according to an embodiment of the present disclosure; FIG. It is a figure which shows an example of the evaluation process by the prediction model determination part which concerns on embodiment of this indication. FIG. 4 is a diagram showing an example of control contents by a control unit according to the embodiment of the present disclosure; FIG. 4 is a flow chart showing the flow of prediction model creation processing according to the embodiment of the present disclosure. 4 is a flow chart showing the process flow of the operation stage of the combustion facility according to the embodiment of the present disclosure; FIG. 4 is a diagram showing an example of a comparison result between a predicted value and an actual measured value of a main steam flow rate according to an embodiment of the present disclosure; 1 is a hardware configuration diagram showing the configuration of a computer according to an embodiment of the present disclosure; FIG.

A control device according to an embodiment of the present disclosure will be described below with reference to the drawings. In the following description, the same reference numerals are given to components having the same or similar functions. Duplicate descriptions of these configurations may be omitted. In the present disclosure, "based on XX" means "based on at least XX", and may include cases based on other elements in addition to XX. Also, "based on XX" is not limited to the case of using XX directly, but may also include the case of being based on what has been calculated or processed with respect to XX. In the present disclosure, "XX or YY" is not limited to either one of XX and YY, but may include both XX and YY. This is also the case when there are three or more selective elements. "XX" and "YY" are arbitrary elements (eg, arbitrary information).

(embodiment)
<1. Overall configuration of incineration facility>
FIG. 1 is a schematic configuration diagram showing the overall configuration of an incineration facility SF according to an embodiment. The incineration facility SF is a stoker furnace that uses, for example, municipal waste, industrial waste, or biomass as the material to be incinerated G. In the following, for convenience of explanation, the “object to be incinerated G” will be referred to as “garbage G”. The incineration facility SF is not limited to the stoker furnace, and may be another type of incineration facility. In this embodiment, the incineration facility SF includes, for example, a crane 1, an incinerator 2, an exhaust heat recovery boiler 3, a cooling tower 4, a dust collector 5, a flue 6, a chimney 7, and a control device 100.

The crane 1 carries the garbage G stored in the garbage pit to the hopper 11 of the incinerator 2, which will be described later, and throws it into the hopper 11. The crane 1 includes a gripping portion 1a that grips the garbage G, and a weight sensor 1b provided in the gripping portion 1a. The weight sensor 1b is, for example, a load cell. The weight sensor 1b detects the weight of the dust G gripped by the gripping portion 1a in a state in which the dust G is gripped and lifted by the gripping portion 1a. A detection result of the weight sensor 1b is transmitted to the control device 100. FIG. The detection result of the weight sensor 1b is an example of "information on the dust G before being supplied to the processing space V" and an example of "information on the properties of the dust G".

In the present disclosure, "information about the properties of the garbage G" means information about the properties or conditions of the garbage G. In addition, in the present disclosure, "information about the properties of the garbage G" is not limited to information that directly indicates the properties of the garbage G, and is information used to specify the properties of the garbage G (for example, combined with other information information that can identify the properties of the dust G) or the like. For example, the weight of the dust G is information that can specify the density of the dust G by being combined with the volume of the dust G, which will be described later. The density of the dust G is an example of the properties of the dust G.

The incinerator 2 is a furnace that burns while transporting the garbage G thrown into the hopper 11 described later. Exhaust gas is generated in the incinerator 2 as the garbage G is burned in the incinerator 2 . The generated exhaust gas is sent to an exhaust heat recovery boiler 3 provided above the incinerator 2 . The exhaust heat recovery boiler 3 heats the water by exchanging heat between the exhaust gas generated in the incinerator 2 and the water to generate steam.

The exhaust gas that has passed through the heat recovery boiler 3 is cooled by the temperature reduction tower 4 and then sent to the dust collector 5 . After soot and dust are removed by the dust collector 5, the exhaust gas is discharged into the atmosphere through the flue 6 and the chimney 7. The flue 6 is provided with a gas concentration sensor 6a. The gas concentration sensor 6a detects concentrations of various gases (for example, oxygen concentration) contained in the exhaust gas flowing through the flue 6. As shown in FIG. The detection result of the gas concentration sensor 6a may include one or more of the CO concentration, NOx concentration, and SOx concentration instead of/in addition to the oxygen concentration. A detection result of the gas concentration sensor 6 a is transmitted to the control device 100 .

<2. Incinerator>
Next, the incinerator 2 will be described in detail. The incinerator 2 has, for example, a supply mechanism 10, a furnace body 20, a stoker 30, an air box 41, a discharge chute 42, a furnace 43, and a blower mechanism 50.

<2.1 Supply Mechanism>
The supply mechanism 10 is a mechanism that temporarily stores the waste G carried by the crane 1 and sequentially supplies the waste G toward the processing space V of the furnace main body 20, which will be described later. The supply mechanism 10 has, for example, a hopper 11, a feeder 12, an extrusion device 13 (see FIG. 2), an object measuring instrument 14, and a moisture measuring instrument 15.

The hopper 11 is a reservoir provided to supply the refuse G to the interior of the furnace body 20 . Garbage G carried by the crane 1 is put into the hopper 11 . The hopper 11 has an inlet portion 11a and an outlet portion 11b. The entrance part 11a is an entrance part for throwing in the garbage G from the outside. The inlet portion 11a extends, for example, in the vertical direction. Garbage G thrown into the inlet portion 11a moves downward due to gravity. The outlet portion 11b is provided below the inlet portion 11a. The outlet portion 11b is an outlet portion that guides the refuse G supplied from the inlet portion 11a toward a processing space V inside the furnace body 20, which will be described later. The outlet portion 11b extends horizontally, for example.

The feeder 12 is provided at the outlet portion 11b of the hopper 11. The feeder 12 has a plate shape along the bottom of the outlet 11 b of the hopper 11 and is arranged along the bottom of the outlet 11 b of the hopper 11 . The feeder 12 can reciprocate along the direction from the outlet 11 b of the hopper 11 toward the processing space V of the furnace body 20 . The feeder 12 is driven by an extrusion device 13 to push out the refuse G accumulated inside the hopper 11 (for example, the exit portion 11 b of the hopper 11 ) toward the processing space V of the furnace body 20 .

The object measuring instrument 14 is a measuring instrument that detects the height of the garbage G thrown into the hopper 11 by the crane 1 . The object measuring device 14 is, for example, LiDAR (Light Detection and Ranging). The object measuring device 14 is provided, for example, at the entrance portion 11a of the hopper 11 and detects the height of the refuse GM passing through the entrance portion 11a of the hopper 11 . Note that the object measuring instrument 14 may directly detect the volume of the dust G by three-dimensional measurement instead of the height of the dust G. FIG. A detection result of the object measuring device 14 is transmitted to the control device 100 . The detection result of the object measuring device 14 is an example of "information about the dust G before being supplied to the processing space V" and an example of "information about the properties of the dust G".

The moisture measuring instrument 15 is a measuring instrument that detects a value related to moisture contained in the garbage G thrown into the hopper 11 (for example, moisture content or moisture content). In this embodiment, the moisture meter 15 has an irradiation section and a detection section provided in the hopper 11, and an analysis section. The irradiation unit irradiates the garbage G accumulated in the hopper 11 with an electromagnetic wave of a predetermined frequency band. The detection unit receives electromagnetic waves emitted from the irradiation unit and transmitted through the dust G or reflected by the dust G. The analysis unit stores, for example, correlation information in advance that indicates the relationship between changes in electromagnetic wave characteristics (for example, changes in amplitude or changes in phase) and moisture content. The analysis unit detects the moisture content contained in the dust G based on the characteristic change of the electromagnetic wave between the irradiation unit and the detection unit and the correlation information.

In this embodiment, the irradiation unit and the detection unit of the moisture meter 15 are provided slightly above the feeder 12 and detect the moisture content of the garbage G deposited on the upper surface of the feeder 12 . A detection result of the moisture meter 15 is transmitted to the control device 100 . The detection result of the moisture measuring device 15 is an example of "information on the garbage G before being supplied to the processing space V", an example of "information on the properties of the garbage G", and an example of "moisture measurement in the hopper 11". This is an example of "Results".

<2.2 Furnace body>
The furnace main body 20 is provided adjacent to the hopper 11 and is a facility for burning the garbage G while conveying it. Below, the conveying direction of the refuse G in the combustion facility F is called "conveying direction D". The furnace body 20 has a drying stage 20a, a combustion stage 20b, and a post-combustion stage 20c in this order from the upstream side to the downstream side in the transport direction D. As shown in FIG. The drying stage 20a is located upstream of the combustion stage 20b and the post-combustion stage 20c, and is an area for drying the refuse G supplied from the hopper 11 prior to combustion on the stoker 30. The combustion stage 20b and the post-combustion stage 20c are areas where the refuse G in a dried state after passing through the drying stage 20a is burned on the stoker 30. As shown in FIG. In the combustion stage 20b, the pyrolysis gas generated from the dust G causes diffusion combustion, and a luminous flame F is generated. In the post-combustion stage 20c, fixed carbon combustion occurs after diffusion combustion of the dust G, so the luminous flame F does not occur. The combustion stage 20b and the post-combustion stage 20c are examples of a processing space V in which the garbage G is burned. The drying stage 20a is an example of a region on the upstream side of the processing space V in the transport direction D. As shown in FIG.

In this embodiment, the furnace body 20 has a visible light camera 21 and an infrared camera 22 . The visible light camera 21 and the infrared camera 22 are arranged on the downstream side of the processing space V in the transport direction D, and capture images of the upstream side in the transport direction D from the downstream side. In this embodiment, the visible light camera 21 and the infrared camera 22 are provided at the downstream end of the furnace main body 20 in the transport direction D (hereinafter referred to as "furnace bottom"). For example, the visible light camera 21 and the infrared camera 22 capture an image of the upstream side in the conveying direction D from the downstream side through a window provided at the bottom of the furnace body 20 . For example, the visible light camera 21 and the infrared camera 22 are arranged vertically or horizontally adjacent to each other.

The visible light camera 21 captures the luminous flame F from the bottom of the furnace body 20 . The imaging result of the visible light camera 21 is transmitted to the control device 100 .

The infrared camera 22 captures an image of the dust G deposited on the drying stage 20a of the furnace body 20 (that is, upstream of the processing space V) from the bottom of the furnace body 20 through the luminous flame F. Further, in the present embodiment, the infrared camera 22 captures an image of the exit portion 11b of the hopper 11 from the bottom of the furnace body 20 through the luminous flame F. For example, the infrared camera 22 captures an image including the garbage G accumulated on the feeder 12 (an image showing the accumulated state of the garbage G) at the exit portion 11b of the hopper 11 . The imaging result of the infrared camera 22 is transmitted to the control device 100 . The imaging result of the infrared camera 22 is an example of "information about the garbage G before being supplied to the processing space V" and an example of "accumulation state information indicating the accumulation state of the garbage G in the hopper 11". .

Note that in this embodiment, one infrared camera 22 captures an image including both the drying stage 20a of the furnace body 20 and the outlet 11b of the hopper 11 (for example, the dust G accumulated on the feeder 12). Instead of this, the furnace body 20 has a first infrared camera that images the drying stage 20a of the furnace body 20 and a second infrared camera that images the outlet portion 11b of the hopper 11 (for example, the waste G accumulated on the feeder 12). You may provide an outside camera separately. Also, the infrared camera 22 may be provided at another position instead of the bottom of the furnace body 20 .

<2.3 Stalker>
The stoker 30 includes a plurality of grates 31 and a grate drive 32 (see FIG. 2). The plurality of fire grates 31 form a stoker surface 30a that serves as the bottom surface of the furnace body 20 (for example, the bottom surface of the processing space V). Garbage G is supplied in layers by the supply mechanism 10 to the stoker surface 30a. The stoker surface 30a is provided over the drying stage 20a, the combustion stage 20b, and the post-combustion stage 20c. The plurality of grates 31 includes fixed grates and movable grates. The fixed grate is fixed to the upper surface of the wind box 41, which will be described later. The movable grate reciprocates along the conveying direction D at a constant speed, thereby conveying the garbage G on the movable grate and the fixed grate (on the stoker surface 30a) to the downstream side while stirring and mixing. .

<2.4 Wind box, discharge chute, furnace>
The wind box 41 is provided below the stoker 30 and supplies combustion air to the interior of the furnace body 20 through the stoker 30 . A plurality of wind boxes 41 are arranged in the transport direction D. As shown in FIG. The windbox 41 has a windbox pressure sensor 41a. The wind box pressure sensor 41 a detects the pressure inside the wind box 41 . The pressure inside the wind box 41 corresponds to the pressure of combustion air supplied to the inside of the furnace body 20 through a primary air line 52, which will be described later. A detection result of the wind box pressure sensor 41 a is transmitted to the control device 100 .

The discharge chute 42 is a device for dropping the garbage G, which has become ash after combustion, to the ash extrusion device located below the furnace body 20. The discharge chute 42 is provided at the bottom of the furnace body 20 .

The furnace 43 extends upward from the top of the furnace body 20 . Exhaust gas generated by burning the garbage G in the processing space V is sent to the exhaust heat recovery boiler 3 through the furnace 43 .

<2.5 Blower Mechanism>
The blower mechanism 50 supplies air (for example, combustion air) to the interior of the furnace body 20 . The blower mechanism 50 has, for example, a blower 51 , a primary air line 52 , an air preheater 53 , a secondary air line 54 , a damper 55 and an air flow rate sensor 56 .

The blower 51 is a forced air blower that forces air (for example, combustion air) into the interior of the furnace body 20 . The blower 51 includes, for example, a first blower 51A and a second blower 51B. The first blower 51A pressurizes combustion air into the interior of the furnace body 20 (for example, the processing space V) through the primary air line 52 and the wind box 41 . The second blower 51B pumps combustion air into the furnace 43 through the secondary air line 54 .

The primary air line 52 connects the first blower 51A and the air box 41. One or more (eg, multiple) primary air dampers 55A are provided in the middle of the primary air line 52 . The primary air damper 55A changes the flow rate of combustion air flowing through the primary air line 52 according to the degree of opening of the primary air damper 55A.

The air preheater 53 is a heat exchanger that preheats the air pressure-fed from the first blower 51A. For example, the air preheater 53 is provided in the middle of the primary air line 52 .

The secondary air line 54 connects the second blower 51B and the furnace 43. The secondary air supplied into the furnace 43 goes toward the garbage G from above the stoker 30 . One or more (eg, multiple) secondary air dampers 55B are provided in the middle of the secondary air line 54 . The secondary air damper 55B changes the flow rate of combustion air flowing through the secondary air line 54 according to the degree of opening of the secondary air damper 55B. For convenience of explanation, the primary air damper 55A and the secondary air damper 55B are hereinafter collectively referred to as "damper 55".

The air flow rate sensor 56 detects the flow rate of air (for example, combustion air) supplied inside the furnace body 20 . The air flow sensor 56 includes, for example, a first air flow sensor 56A and a second air flow sensor 56B. The first air flow rate sensor 56A is provided in the middle of the primary air line 52 and detects the flow rate of air supplied through the primary air line 52 . The second air flow rate sensor 56B is provided in the middle of the secondary air line 54 and detects the flow rate of air supplied through the secondary air line 54 . In the following description, the "detection result of the air flow sensor 56" includes, for example, the detection result of the first air flow sensor 56A and the detection result of the second air flow sensor 56B.

<3. Exhaust heat recovery boiler>
Next, the exhaust heat recovery boiler 3 will be described. The exhaust heat recovery boiler 3 includes, for example, a boiler body 61, a pipeline 62, a radiation temperature sensor (infrared temperature sensor) 63, a furnace pressure sensor 64, a feed water flow sensor 65, and a superheater desuperheater flow sensor (steam flow sensor ) 66.

The boiler body 61 is connected to the furnace 43 of the incinerator 2. Exhaust gas generated in the incinerator 2 flows into the boiler main body 61 . A radiation temperature sensor 63 and an in-furnace pressure sensor 64 are provided in the boiler main body 61 . A radiation temperature sensor 63 detects the temperature inside the boiler body 61 . The furnace pressure sensor 64 detects the pressure inside the boiler body 61 . The detection results of radiation temperature sensor 63 and furnace pressure sensor 64 are transmitted to control device 100 .

The pipeline 62 extends inside the boiler body 61 . Line 62 is provided with multiple superheaters and multiple desuperheaters. Water is supplied from the water supply unit to the inlet of the conduit 62 . At least part of the water flowing through the pipe line 62 is heated by heat exchange inside the boiler body 61 and becomes main steam that flows toward an external device (for example, a turbine). A "main steam flow rate", which will be described later, means a flow rate of steam flowing from the pipeline 62 toward an external device (for example, a turbine).

The water supply flow rate sensor 65 is provided at the inlet of the pipeline 62 and detects the flow rate of water supplied to the pipeline 62 . A superheater desuperheater flow sensor 66 is provided in the middle of the pipeline 62 and detects the flow rate of the fluid (eg, steam) flowing through the pipeline 62 . For example, the superheater desuperheater flow sensor 66 includes a first superheater desuperheater flow sensor 66A that detects the flow rate of fluid passing through the primary desuperheater (primary superheater desuperheater flow rate) and a second superheater desuperheater flow sensor 66B that senses the flow rate of fluid passing through the vessel (secondary superheater desuperheater flow). In the following description, the "detection result of the superheater desuperheater flow sensor 66" means, for example, the detection result of the first superheater desuperheater flow sensor 66A and the detection result of the second superheater desuperheater flow sensor 66B. including. The detection results of the feedwater flow rate sensor 65 and the superheater desuperheater flow rate sensor 66 are sent to the control device 100 .

<4. Control device>
Next, the control device 100 is described.
FIG. 2 is a block diagram showing the functional configuration of the incineration facility SF according to the embodiment. The control device 100 centrally controls the incineration facility SF. For example, the control device 100 performs combustion control of the refuse G in the processing space V of the furnace body 20 . In this embodiment, the control device 100 has, for example, an information acquisition unit 110, a data conversion unit 120, a prediction model creation unit 130, a prediction model determination unit 140, a steam flow rate prediction unit 150, and a control unit 160. A device to be controlled by the control device 100 (hereinafter referred to as a “controlled device S”) includes the extrusion device 13, the blower 51, the damper 55, the grate driving device 32, and the like.

<4.1 Information Acquisition Unit>
The information acquisition unit 110 acquires detection results and the like detected by the above-described various sensors included in the incineration facility SF. For example, the information acquisition unit 110 obtains the detection result of the weight sensor 1b (garbage weight), the detection result of the object measuring instrument 14 (garbage height), the detection result of the moisture measuring instrument 15 (garbage moisture detection result), the visible light camera 21 imaging result (combustion flame image), imaging result of the infrared camera 22 (dust layer image), detection result of the wind box pressure sensor 41a (wind box pressure), detection result of the air flow rate sensor 56 (forced air flow rate), radiation The detection result of the temperature sensor 63 (furnace temperature), the detection result of the furnace pressure sensor 64 (furnace pressure), the detection result of the feedwater flow rate sensor 65 (feedwater flow rate), the detection result of the superheater desuperheater flow rate sensor 66 ( superheater desuperheater flow rate) and the detection result (oxygen concentration etc.) of the gas concentration sensor 6a.

Here, the detection result of the wind box pressure sensor 41a, the detection result of the air flow sensor 56, the detection result of the radiation temperature sensor 63, the detection result of the furnace pressure sensor 64, the detection result of the feed water flow sensor 65, the superheater One or more of the detection result of the heater flow rate sensor 66 and the detection result of the gas concentration sensor 6a are included in the process data described later. Each of these detection results is, together with the detection result of the weight sensor 1b, the detection result of the object measuring instrument 14, the detection result of the moisture measuring instrument 15, the imaging result of the visible light camera 21, and the imaging result of the infrared camera 22, " It corresponds to an example of "prediction information". Note that "obtaining" in the present disclosure is not limited to active acquisition by outputting a transmission request, and includes acquisition by passively receiving information transmitted from various devices. This definition also applies to the following description.

In addition, the information acquisition unit 110 acquires process values indicating the state of each device included in the control target device S as part of process data described later. For example, the controlled device S includes process values indicating the state of the extrusion device 13 (for example, the stroke length of the feeder 12 and/or the moving speed of the feeder 12, process values indicating the state of the blower 51 (for example, the rotation speed of the blower 51), A process value indicating the state of the damper 55 (for example, the opening of the damper 55) and a process value indicating the state of the grate drive device 32 (for example, the moving speed of the grate 31) are acquired as part of the process data. Each of the process data (process value) is an example of “prediction information.” A process value indicating the state of the extrusion device 13 (for example, the stroke length of the feeder 12 and/or the moving speed of the feeder 12) is “the feeder 12 and an example of "supply state information indicating the supply state of the waste G from the hopper 11 to the processing space V." , to the data conversion unit 120 .

<4.2 Data converter>
The data conversion unit 120 performs predetermined data conversion on information received from the information acquisition unit 110 . For example, the data conversion unit 120 performs predetermined data conversion such as feature amount extraction, time delay adjustment, and averaging.

FIG. 3 is a block diagram showing the functional configuration of the data converter 120 according to the embodiment. The data conversion unit 120 includes, for example, a first calorific value estimating unit 121, a second calorific value estimating unit 122, a first feature amount extracting unit 123, an oxygen concentration estimating unit 124, a flammability coefficient calculating unit 125, an image converting unit ( image processing unit) 126, dust layer height detection unit 127, second feature amount extraction unit 128, feeder supply amount estimation unit 129, and adjustment processing unit PU.

(First calorific value estimation unit)
The detection result (garbage weight) of the weight sensor 1b and the detection result (garbage height) of the object measuring device 14 are input to the first heat generation estimation unit 121 . The first calorific value estimator 121 calculates the volume of the garbage G based on the height of the garbage G (for example, based on the height of the garbage G and the size of the gripping portion 1a of the crane 1). Then, the first calorific value estimation unit 121 calculates the density of the dust G by dividing the weight of the dust G by the volume of the dust G. In addition, the first calorific value estimator 121 has a correlation indicating the correlation between the density of the garbage G and the calorific value of the garbage G (for example, lower heating value LHV) (hereinafter referred to as “garbage calorific value”). have information. The correlation information is, for example, a calorific value estimation formula for calculating an estimated value of the calorific value of the refuse G from the density of the refuse G. FIG. The first calorific value estimator 121 calculates an estimated value of the calorific value of waste based on the calculated density of the waste G and the correlation information. The first calorific value estimator 121 outputs the calculated estimated value of the waste calorific value to the adjustment processor PU.

Here, the density in the present embodiment means, for example, bulk density. The bulk density is not the specific density (true density) of the object, but the density calculated from the "weight per unit volume including voids". However, the first calorific value estimation unit 121 may estimate and use the true density instead of/in addition to the bulk density. The density of the garbage G calculated by the first calorific value estimating unit 121 is based on the weight measured outside the hopper 11, and is equivalent to the density of the garbage G inside the hopper 11. is. Therefore, the density of the garbage G calculated by the first calorific value estimator 121 corresponds to an example of "the density of the garbage G in the hopper 11".

(Second calorific value estimation unit)
The detection result of the moisture meter 15 (garbage moisture detection result) is input to the second calorific value estimation unit 122 . Note that the volume of the dust G calculated by the first calorific value estimating unit 121 may also be input to the second calorific value estimating unit 122 . When the volume of the dust G is input, the second calorific value estimator 122 can calculate the moisture content of the dust G by multiplying the moisture content of the dust G by the volume of the dust G. The second calorific value estimator 122 has correlation information indicating the correlation between the value (moisture content or moisture content) related to the moisture content of the garbage G and the calorific value of the garbage (for example, the lower calorific value). The correlation information is, for example, a calorific value estimation formula for calculating an estimated value of the calorific value of waste G from a value related to moisture of the waste G. FIG. The second calorific value estimating unit 122 calculates an estimated value of the calorific value of the waste based on the value related to the water content of the waste G and the correlation information. The second calorific value estimator 122 outputs the calculated estimated value of the waste calorific value to the adjustment processor PU.

Here, FIG. 4 is a diagram showing the correlation between the estimated value of the heating value of waste based on the detection result of the moisture measuring instrument 15 and the heating value of waste confirmed in the actual machine. As shown in FIG. 4, the inventors of the present invention have found that there is a sufficiently high correlation between the estimated value of the heating value of waste based on the detection result of the moisture measuring device 15 and the heating value of waste confirmed in the actual machine. confirmed by et al. In addition, since the estimated value of the calorific value of the dust based on the detection result of the moisture measuring device 15 is information preceding the calorific value of the dust G confirmed in the actual machine, the moisture measuring device can The inventors of the present invention have confirmed that the correlation between the estimated value of the waste heat generation amount based on the detection result of No. 15 and the waste heat generation amount confirmed in the actual machine can be enhanced.

(First feature quantity extraction unit)
The photographing result (combustion flame image) of the visible light camera 21 is input to the first feature quantity extraction unit 123 . The first feature quantity extraction unit 123 performs clustering processing on the input combustion flame image, thereby converting it into color image data IM (see FIG. 5) divided into a plurality of color regions according to color information. . Then, the first feature quantity extraction unit 123 extracts a feature quantity relating to the flame state based on the color image data IM.

An example of "dividing an image into multiple color regions according to color information by clustering processing" will be described. The color information is RGB color components, and each of the plurality of color regions is set by clustering processing so that the RGB color components do not overlap each other. The first feature amount extraction unit 123 decomposes the combustion flame image into RGB color components for each pixel, and determines a color region including each pixel. Note that the color information is not limited to RGB color components, and may be luminance or saturation.

The algorithm for clustering processing is not particularly limited, and various known clustering algorithms can be used. For example, the clustering process may be performed using an algorithm capable of specifying the number of clusters such as k-means, or may be performed using an algorithm such as flowsom which automatically determines the number of clusters.

FIG. 5 is a diagram showing an example of color image data IM. The color image data IM exemplified in FIG. 5 is divided into seven color regions A by clustering processing. It includes an area A4, a fifth color area A5, a sixth color area A6, and a seventh color area A7. Each of the first color area A1 to the seventh color area A7 is converted to a black and white (gray scale) gradation value, and becomes darker from the first color area A1 to the seventh color area A7.

Next, an example of "extracting feature values from color image data IM" will be described. The first feature amount extraction unit 123 calculates the total number of pixels (that is, the area) divided into the first color area A1, and extracts the total number of pixels as a feature amount. For example, the first feature amount extraction unit 123 extracts the total number of pixels of the first color area A1 at predetermined time intervals (for example, every second). The first feature amount extraction unit 123 also calculates the total number of pixels for each predetermined time period for each of the second color area A2 to the seventh color area A7, and extracts each total number of pixels as a feature amount. Note that in the present embodiment, the feature amount includes the total number of pixels of all color regions (the first color region A1 to the seventh color region A7) among the plurality of color regions, but the present disclosure is not limited to this form. . The feature amount may include the total number of pixels in at least one color area among the plurality of color areas.

The first feature amount extraction unit 123 outputs the extracted feature amount related to the flame state to the oxygen concentration estimation unit 124 and to the adjustment processing unit PU. Note that the method for extracting the feature amount by the first feature amount extraction unit 123 is not limited to clustering, and another method may be used.

(Oxygen concentration estimation unit)
The feature amount extracted by the first feature amount extraction section 123 and part or all of the process data acquired by the information acquisition section 110 are input to the oxygen concentration estimation section 124 . The process data input to the oxygen concentration estimator 124 are, for example, the detection result of the wind box pressure sensor 41a, the detection result of the air flow rate sensor 56, the detection result of the radiation temperature sensor 63, the detection result of the furnace pressure sensor 64, and the feed water. One or more of the detection result of the flow sensor 65, the detection result of the superheater desuperheater flow sensor 66, the detection result of the gas concentration sensor 6a, and the like. The oxygen concentration estimation unit 124 derives an estimation formula for estimating the oxygen concentration in the processing space V by performing regression analysis by machine learning based on the input feature amount and process data. Then, the oxygen concentration estimator 124 calculates an estimated value of the oxygen concentration in the processing space V in real time based on the input feature amount and process data and the above estimation formula. The oxygen concentration estimation unit 124 outputs the calculated estimated value of the oxygen concentration to the flammability coefficient calculation unit 125 . Note that the method for deriving the above estimation formula by the oxygen concentration estimating unit 124 is not limited to regression analysis, and may be another method. Also, the machine learning algorithm is not particularly limited, and various known algorithms can be used.

(Combustibility factor calculator)
The estimated value of the oxygen concentration calculated by the oxygen concentration estimating unit 124 and part or all of the process data acquired by the information acquiring unit 110 are input to the flammability coefficient calculating unit 125 . The process data input to the oxygen concentration estimation unit 124 is, for example, one or more of the detection result of the radiation temperature sensor 63, the moving speed of the feeder 12, and the like. In this embodiment, the flammability coefficient calculator 125 quantifies the combustion state of the processing space V based on the estimated value of the oxygen concentration, the detection result of the radiation temperature sensor 63, the amount of change in the moving speed of the feeder 12, and the like. Calculate the flammability factor. The flammability coefficient calculator 125 outputs the calculated flammability coefficient to the adjustment processor PU. In the present disclosure, "flammability" means "combustion condition."

(Image converter)
The imaging result (dust layer image) of the infrared camera 22 is input to the image conversion unit 126 . The image conversion unit 126 performs predetermined image processing on the input dust layer image to simplify the dust layer image. For example, the image conversion unit 126 binarizes the input dust layer image. The binarization method is, for example, the Otsu method, but is not limited to this.

FIG. 6 is a diagram showing an example of processing by the image conversion unit 126. FIG. As shown in FIG. 6, the dust layer image, which is a color image (or monochrome image) captured by the infrared camera 22, is converted into a black and white image by the image conversion unit 126. FIG. An image (for example, a black-and-white image) obtained by the image converter 126 is output to the dust layer height detector 127 .

(Garbage layer height detector)
The image obtained by the image conversion unit 126 is input to the dust layer height detection unit 127 . The dust layer height detection unit 127 detects the height of the dust G (dust layer height) in the drying stage 20a of the furnace body 20 based on the input image.

FIG. 7 is a diagram showing an example of processing by the dust layer height detection unit 127. FIG. The dust layer height detection unit 127 detects one or more predetermined attention areas R (see FIG. 6) that are part of the image obtained by the image conversion unit 126 (in the example shown in FIG. 6, 2 places). Then, the dust layer height detection unit 127 divides the image of the set attention area R into a plurality of divided areas Ra (for example, the attention area R is divided vertically and horizontally into 20 areas, and divided horizontally and horizontally). Divided regions Ra) divided into five in the direction are set (see (a) in FIG. 7). Note that in FIG. 7, the data of the two regions of interest R are shown side by side.

For each divided area Ra, the dust layer height detection unit 127 assigns "1" to the divided area Ra when the black is greater than 50%, and assigns "1" to the divided area Ra when the black is 50% or less. 0” (see (b) in FIG. 7). Then, the dust layer height detection unit 127 calculates the position of the uppermost divided area Ra of "1" as the dust layer height. For example, in the example shown in FIG. 7, the height position of line H is calculated as the dust layer height. The dust layer height detection unit 127 outputs the calculated dust layer height to the feeder supply amount estimation unit 129 .

(Second feature quantity extraction unit)
The imaging result (dust layer image) of the infrared camera 22 is input to the second feature amount extraction unit 128 . The second feature amount extraction unit 128 performs clustering processing on the input dust layer image, thereby converting the dust layer image into color image data divided into a plurality of color regions according to the color information. Then, the second feature quantity extraction unit 128 extracts a feature quantity relating to the supply state of dust based on the color image data. Note that the processing method of "dividing an image into a plurality of color regions according to color information by clustering processing" and the clustering processing algorithm are the same as the processing method and algorithm of the first feature quantity extraction unit 123, for example. but can be different.

In this embodiment, the second feature amount extraction unit 128 classifies the input dust layer image into a plurality of color regions by clustering processing. Then, the second feature amount extraction unit 128 calculates the total number of pixels (that is, the area) of the divided color regions, and extracts the total number of pixels as a feature amount relating to the supply state of the dust G. FIG. The second feature quantity extraction unit 128 extracts the total number of pixels in each color region at predetermined time intervals (for example, every second). Note that in the present embodiment, the feature amount includes the total number of pixels of all color regions among the plurality of color regions, but the present disclosure is not limited to this form. The feature amount may include the total number of pixels in at least one color area among the plurality of color areas. The second feature amount extraction unit 128 outputs the extracted feature amount related to the supply state of the garbage G to the feeder supply amount estimation unit 129 . Note that the method for extracting the feature amount by the second feature amount extraction unit 128 is not limited to clustering, and another method may be used.

(Feeder supply amount estimation unit)
The feeder supply amount estimating unit 129 is provided with information indicating the dust layer height calculated by the dust layer height detecting unit 127 and the feature amount of the supply state of the dust G extracted by the second feature amount extracting unit 128. information is entered. Further, the feeder supply amount estimation unit 129 has correlation information indicating the correlation between the garbage layer height and the supply state of the garbage G, and the correlation between the amount of the garbage G supplied from the feeder 12 . The correlation information is, for example, a supply amount estimation formula for calculating the supply amount of the garbage G from the feeder 12 from the feature values of the garbage layer height and the supply state of the garbage G. The feeder supply amount estimator 129 estimates the supply amount of the garbage G from the feeder 12 based on the input information indicating the garbage layer height, the feature amount of the supply state of the garbage G, and the correlation information. calculate. The feeder supply amount estimation unit 129 outputs the calculated estimated value of the supply amount of the garbage G to the adjustment processing unit PU. The estimated value of the supply amount of the refuse G is another example of "supply state information indicating the supply state of the refuse G from the hopper 11 to the processing space V".

(adjustment processor)
The adjustment processing unit PU includes a first calorific value estimating unit 121, a second calorific value estimating unit 122, a first feature amount extracting unit 123, a non-combustibility coefficient calculating unit 125, and a feeder supply amount estimating unit 129. Information and process data acquired by the information acquisition unit 110 are input. Hereinafter, these are collectively referred to as "input information". In this embodiment, the process data input to the adjustment processing unit PU are, for example, the process value of the feeder 12 (for example, the stroke length of the feeder 12 and/or the moving speed of the feeder 12), the detection result of the wind box pressure sensor 41a, Detection results of the air flow rate sensor 56, detection results of the furnace pressure sensor 64, detection results of the radiation temperature sensor 63, detection results of the feed water flow rate sensor 65, detection results of the superheater desuperheater flow rate sensor 66, and the gas concentration sensor 6a. detection results (eg, oxygen concentration). Part or all of the process data (for example, process values of the feeder 12) may be omitted.

The adjustment processing unit PU converts the input information into data to be input to the main steam flow rate prediction model M, which will be described later, by performing predetermined processing on the input information. The adjustment processing unit PU includes, for example, a preprocessing unit PUa and a time delay adjustment unit PUb.

The preprocessing unit PUa performs preprocessing such as averaging processing on one or more pieces of input information. For example, the preprocessing unit PUa averages values obtained at multiple detection points for one or more pieces of input information. Note that the preprocessing by the preprocessing unit PUa may be differentiation processing or the like instead of/in addition to the averaging processing. The preprocessing unit PUa outputs the preprocessed input information to the time delay adjustment unit PUb.

Based on each input information and the time delay set value individually set for each input information, the time delay adjustment unit PUb simultaneously supplies the data to the main steam flow rate prediction model M as one data set (a set of input information). It associates the input information to be input on the time axis. That is, there is a time lag between each input information change and the main steam flow rate change. In other words, each input information becomes a leading signal leading to changes in the main steam flow rate. For example, input information associated with hopper 11 or a position closer to hopper 11 will have a greater leading lead than input information associated with a position closer to process volume V. FIG.

FIG. 8 is a diagram showing an example of the correlation between each piece of input information and the main steam flow rate. In this embodiment, the length of the time delay setting value is changed multiple times for each piece of input information, and the time delay setting value with the highest correlation between the input information and the main steam flow rate is selected.

For example, the correlation between the input information indicating the supply amount from the feeder 12 and the main steam flow rate (see (a) in FIG. 8) is highest when T2 [minute] is set as the time delay setting value. . In other words, the input information indicating the supply amount from the feeder 12 is a preceding signal that precedes the main steam flow rate by T2 [minutes]. Similarly, the correlation between the input information indicating the flammability coefficient and the main steam flow rate (see (b) in FIG. 8) is highest when T3 [minute] is set as the time delay set value. In other words, the input information indicating the flammability coefficient is a preceding signal that precedes the main steam flow rate by T3 [minutes]. For example, T3 [minute] is shorter than T2 [minute].

FIG. 9 is a diagram showing an example of time delay setting values for each piece of input information. In FIG. 9, T1 [minute]>T2 [minute]>T3 [minute]. However, these relationships are not limited. The time delay setting value for each piece of input information can be set as appropriate.

The time delay adjustment unit PUb correlates the input information simultaneously input to the main steam flow rate prediction model M based on the time delay set values for each input information as described above, thereby adjusting the main steam flow rate at a certain point in the future. Generate a data set (ie, a collection of time-aligned input information) for prediction. The adjustment processing unit PU outputs the data set generated by the time delay adjustment unit PUb.

<4.3 Prediction model creation unit>
In the prediction model creation process (learning process), the prediction model creation unit 130 learns the combination of the data set created by the adjustment processing unit PU and the correct data of the predicted value of the main steam flow rate corresponding to the data set. entered as data. The prediction model creation unit 130 performs machine learning based on the input learning data to generate a main steam flow rate prediction model M for predicting the main steam flow rate at a future point in time. The main steam flow rate prediction model M is a learned model that outputs a prediction value of the main steam flow rate at a future point in time when the data set generated by the adjustment processing unit PU is input. The main steam flow rate prediction model M is, for example, LSTM (Long Short Term Memory) or XGBoost (eXtreme Gradient Boosting), but is not limited to these. The machine learning algorithm is not particularly limited, and various known machine learning algorithms can be used.

In this embodiment, the prediction model creation unit 130 generates a plurality of main steam flow rate prediction models M that predict the main steam flow rate at a plurality of different future times. For example, the predictive model generator 130 generates a plurality of main steam flow rate prediction models M that respectively output predicted values of the main steam flow rates 60 seconds ahead, 120 seconds ahead, and 180 seconds ahead. Note that, instead of generating the plurality of main steam flow rate prediction models M, the prediction model creation unit 130 creates one main steam flow rate prediction model M that outputs a plurality of prediction values respectively corresponding to a plurality of future points in time. may be generated.

In addition, the prediction model creation unit 130 varies the learning period (learning data accumulation period) and generates a plurality of main steam flow rate prediction models M based on the learning data of a plurality of learning periods with different lengths. For example, the prediction model creation unit 130 creates the main steam flow rate prediction models M respectively corresponding to one day's worth of learning data, two days' worth of learning data, . . . , seven days' worth of learning data.

<4.4 Prediction model determination unit>
The prediction model determination unit 140 evaluates a plurality of main steam flow rate prediction models M corresponding to a plurality of learning periods generated by the prediction model creation unit 130, and determines the main steam flow rate prediction model M used in the steam flow rate prediction unit 150. Select.

FIG. 10 is a diagram showing an example of evaluation processing by the prediction model determination unit 140. FIG. In the present embodiment, the prediction model determination unit 140 is based on accuracy indicators such as the root mean square error (RMSE: Root Mean Square Error) and the mean absolute scale error (MASE: Mean Absolute scale Error) in a plurality of learning periods. A plurality of corresponding main steam flow rate prediction models M are evaluated. In this embodiment, as the main steam flow rate prediction model M corresponding to each learning period, a set of a plurality of main steam flow rate prediction models M that respectively predict the main steam flow rates 60 seconds ahead, 120 seconds ahead, and 180 seconds ahead are provided. evaluated. Then, among the plurality of main steam flow rate prediction models M corresponding to the plurality of learning periods, the learning that has the highest overall prediction accuracy at a plurality of future points in time (60 seconds ahead, 120 seconds ahead, and 180 seconds ahead) A set of main steam flow prediction models M corresponding to the period is selected. In the example shown in FIG. 10, a set of main steam flow rate prediction models M corresponding to a five-day learning period is selected. The main steam flow rate prediction model M selected by the prediction model determination section 140 is output to the steam flow rate prediction section 150 . Note that the values of S1 to S7 in FIG. 10 are values specifically calculated based on the RMSE or MASE formula, and show an example where S1<S2<S3<S4<S5<S6<S7.

<4.5 Main Steam Flow Rate Predictor>
The steam flow rate prediction unit 150 uses the data set generated by the adjustment processing unit PU and the main steam flow rate prediction model M selected by the prediction model determination unit 140 in the operation stage of the incineration facility SF to predict future Derive the predicted value of the main steam flow rate. In this embodiment, a plurality of main steam flow rate prediction models M are used to predict the main steam flow rates 60 seconds ahead, 120 seconds ahead, and 180 seconds ahead, respectively. A predicted value for the main steam flow is derived. The steam flow rate prediction unit 150 derives a predicted value of the main steam flow rate at a predetermined cycle (for example, every second or every 10 seconds). The steam flow rate prediction unit 150 outputs the derived predicted value of the main steam flow rate to the control unit 160 .

<4.6 Control section>
The control unit 160 performs combustion control of the processing space V based on the predicted value of the main steam flow rate derived by the steam flow rate prediction unit 150 (for example, the predicted value of 60 seconds ahead, 120 seconds ahead, and 180 seconds ahead). . Specifically, the control unit 160 controls the controlled device S so that the amount of fluctuation in the combustion state in the processing space V becomes small.

FIG. 11 is a diagram showing an example of the content of control by the control unit 160. As shown in FIG. If the future predicted value of the main steam flow rate (for example, one of the predicted values for 60 seconds ahead, 120 seconds ahead, and 180 seconds ahead) is below a preset lower threshold TH1, the control unit 160 It is determined that insufficient combustion will occur, and control is performed to promote combustion. In addition, when the future predicted value of the main steam flow rate (for example, one of the predicted values for 60 seconds ahead, 120 seconds ahead, and 180 seconds ahead) exceeds a preset upper threshold value TH2, It determines that excessive combustion will occur in the future, and performs control to suppress combustion.

In this embodiment, since the control instruction is output based on the deviation of the predicted value from the set value (reference value), fluctuations in the main steam flow rate can be suppressed. That is, in the present embodiment, the predicted value of the main steam flow rate at a future point in time is set at the lower threshold TH1 or the upper threshold TH2, not at the time when the actual measured value of the main steam flow reaches the lower threshold TH1 or the upper threshold TH2 (point A in FIG. 11). A control instruction to change the combustion control is output when the threshold TH2 is reached (point B in FIG. 11). Fluctuations in the main steam flow rate when control is performed based on predicted values (see the two-dot chain line in FIG. 11) are compared to fluctuations in the main steam flow rate when control is performed based on actual measurements ( (see solid line).

Specifically, the control unit 160 includes a feeder control unit 161, an air supply control unit 162, and a grate control unit 163. Each controller performs, for example, PI control (proportional integral control). However, the control algorithm is not limited to PI control, and various known control algorithms can be used.

The feeder control unit 161 acquires a process value indicating the movement of the feeder 12 from the extrusion device 13, and generates a control instruction value for the feeder 12 based on PI control, for example. The feeder control unit 161 outputs the generated control instruction value to the extrusion device 13 to control the movement of the feeder 12 and control the amount of the refuse G supplied to the processing space V. FIG. For example, the feeder control unit 161 increases the supply amount of the garbage G when promoting combustion. On the other hand, the feeder control unit 161 reduces the supply amount of the refuse G when suppressing combustion.

The air supply control unit 162 acquires a process value relating to the rotational speed of the fan 51 and/or the opening degree of the damper 55 from the fan 51 or the damper 55, and provides a control instruction value for the fan 51 and/or the damper 55 based on, for example, PI control. to generate The air supply control unit 162 controls the blower 51 and/or the damper 55 by outputting the generated control instruction value to the blower 51 and/or the damper 55, and adjusts the supply amount of air (for example, combustion air) to the processing space V. to control. For example, the air supply control unit 162 increases the amount of air supply when promoting combustion. On the other hand, the air supply control unit 162 reduces the amount of air supply when suppressing combustion.

The grate control unit 163 acquires process data about the moving speed of the grate 31 from the grate drive device 32, and generates a control instruction value for the grate 31 based on PI control, for example. The grate control unit 163 outputs the generated control instruction value to the grate driving device 32 to control the grate 31 and control the stirring state of the garbage G. FIG. For example, the grate control unit 163 increases the moving speed of the grate 31 when promoting combustion. On the other hand, the grate control unit 163 reduces the moving speed of the grate 31 when suppressing combustion.

<5 Process flow>
Next, an example of the flow of processing in the control device 100 described above will be described. However, the order of the processes described below is not limited to the following example, and may be changed as appropriate.

<5.1 Prediction model creation>
First, the processing (learning processing) for creating the main steam flow rate prediction model M will be described. The processing for creating the main steam flow rate prediction model M, which will be described below, is also executed in parallel during the operation stage of the incineration facility SF, which will be described later.

FIG. 12 is a flow chart showing the flow of prediction model creation processing. First, the information acquisition unit 110 acquires the detection results of various sensors and process data (S101). Next, the data conversion unit 120 generates a data set to be input to the main steam flow rate prediction model M based on the detection results of various sensors and the process data acquired by the information acquisition unit 110 (S102). That is, the data conversion unit 120 performs calculations and clustering using various estimation formulas, and performs time delay adjustment processing and the like on the input information obtained by these calculations, thereby generating a data set.

Next, the prediction model creation unit 130 accumulates the data sets generated by the data conversion unit 120 over multiple days (S103). Then, the prediction model creation unit 130 varies the learning period (learning data accumulation period) to generate a plurality of main steam flow rate prediction models M based on the learning data of a plurality of learning periods of different lengths (S104). .

Next, the prediction model determination unit 140 evaluates a plurality of main steam flow rate prediction models M with different learning periods generated by the prediction model creation unit 130, and selects the main steam flow rate prediction model M used in the steam flow rate prediction unit 150. Select (S105). In this embodiment, the prediction model determination unit 140 selects the main steam flow rate prediction model M currently used by the steam flow rate prediction unit 150 among the plurality of main steam flow rate prediction models M newly generated by the prediction model creation unit 130. It is determined whether or not there is a main steam flow rate prediction model M with higher prediction accuracy than the

Then, if there is no main steam flow rate prediction model M with higher prediction accuracy than the main steam flow rate prediction model M currently in use among the plurality of newly generated main steam flow rate prediction models M (S105: NO), S103 , and the processes of S103 and S104 are performed again. On the other hand, if there is a main steam flow rate prediction model M with higher prediction accuracy than the main steam flow rate prediction model M currently in use among the plurality of newly generated main steam flow rate prediction models M (S106: YES), The prediction model determination unit 140 outputs the main steam flow rate prediction model M with high prediction accuracy to the steam flow rate prediction unit 150 to update the main steam flow rate prediction model M to be used (S107). The processes from S101 to S107 described above are repeatedly executed in the operating stage of the incineration facility SF.

<5.2 Treatment at the operating stage of combustion equipment>
Next, the processing at the operating stage of the incineration facility SF will be described.
FIG. 13 is a flow chart showing the flow of processing in the operating stage of the combustion equipment. First, the information acquisition unit 110 acquires detection results of various sensors and process data (S201). Next, the data conversion unit 120 generates a data set to be input to the main steam flow rate prediction model M based on the detection results of various sensors and the process data acquired by the information acquisition unit 110 (S202). The data conversion section 120 outputs the generated data set to the steam flow rate prediction section 150 .

Next, the steam flow rate prediction unit 150 derives a predicted value of the main steam flow rate at a future point in time based on the data set received from the data conversion unit 120 and the main steam flow rate prediction model M (S203). The steam flow rate prediction unit 150 outputs the derived predicted value of the main steam flow rate at the future time to the control unit 160 . Next, the control unit 160 derives the control amount of the controlled device S based on the predicted value of the main steam flow rate (S204). Then, the control unit 160 outputs a control instruction value based on the derived control amount to the controlled device S (S205). The processes from S201 to S205 described above are repeatedly executed in the operating stage of the incineration facility SF.

<6. Action effect>
The main steam flow rate may vary greatly depending on the supply state of the dust G and the properties of the dust G. Therefore, when prediction is made from information after the combustion process, it may be difficult to improve the prediction accuracy of the main steam flow rate.

On the other hand, in the present embodiment, the control device 100 includes an information acquisition unit 110 that acquires information about the garbage G before being supplied to the processing space V in the incineration facility SF, and the information acquired by the information acquisition unit 110. A steam flow rate prediction unit 150 that predicts the main steam flow rate generated by the heat recovery boiler 3 of the incineration facility SF based on prediction information including and a control unit 160 that performs

According to such a configuration, the flow rate of the main steam is predicted based on the information about the garbage G before it is supplied to the processing space V, so it is possible to predict the flow rate of the main steam with high accuracy. This enables the control device 100 to perform combustion control based on the highly accurate predicted value of the main steam flow rate. As a result, the fluctuation width of the main steam flow rate can be suppressed.

FIG. 14 is a diagram showing an example of comparison results between predicted values and measured values according to this embodiment. As shown in FIG. 14, it can be confirmed that the predicted value of the main steam flow rate by the control device 100 follows the variation of the measured value of the main steam flow rate with high accuracy. Further, the present inventors have confirmed that the predicted value of the main steam flow rate by the control device 100 of the present embodiment improves the prediction accuracy as compared with the case where the prediction is performed only from the information after the combustion process. .

<7. Variation>
In the above-described embodiment, the time delay setting values individually set for each piece of input information are set after the time delay setting value that maximizes the correlation between each piece of input information and the main steam flow rate is selected and set. , the set time delay setting value is used as a fixed value. However, the time delay adjustment unit PUb recalculates the correlation between each input information and the main steam flow rate at a predetermined cycle, and sets the time delay so that the correlation between each input information and the main steam flow rate becomes higher. You can change the value. According to such a configuration, it may be possible to further improve the prediction accuracy of the main steam flow rate even when the properties of the garbage G change depending on the season or other factors.

(Other embodiments)
As described above, the embodiments of the present disclosure have been described in detail with reference to the drawings, but the specific configuration is not limited to these embodiments, and design changes and the like are included within the scope of the present disclosure.

FIG. 14 is a hardware configuration diagram showing the configuration of the computer 1100 according to this embodiment. Computer 1100 includes, for example, processor 1110 , main memory 1120 , storage 1130 and interface 1140 .

Each functional unit of the control device 100 described above is implemented in the computer 1100 . The operation of each functional unit described above is stored in the storage 1130 in the form of a program. The processor 1110 reads a program from the storage 1130, develops it in the main memory 1120, and executes the above processing according to the program. Also, the processor 1110 secures storage areas in the main memory 1120 to be used by the functional units described above according to the program.

The program may be for realizing part of the functions that the computer 1100 exhibits. For example, the program may function in combination with another program already stored in storage 1130 or in combination with another program installed in another device. Further, the computer 1100 may include a custom LSI (Large Scale Integrated Circuit) such as a PLD (Programmable Logic Device) in addition to or instead of the above configuration. Examples of PLDs include PAL (Programmable Array Logic), GAL (Generic Array Logic), CPLD (Complex Programmable Logic Device), and FPGA (Field Programmable Gate Array). In this case, part or all of the functions implemented by processor 1110 may be implemented by the integrated circuit.

Examples of the storage 1130 include magnetic disks, magneto-optical disks, and semiconductor memories. The storage 1130 may be an internal medium directly connected to the bus of the computer 1100, or an external medium connected to the computer 1100 via the interface 1140 or communication line. Moreover, when this program is delivered to the computer 1100 via a communication line, the computer 1100 receiving the delivery may develop the program in the main memory 1120 and execute the above process. Also, the program may be for realizing part of the functions described above. Furthermore, the program may be a so-called difference file (difference program) that implements the above-described functions in combination with another program already stored in the storage 1130 .

<Appendix>
For example, the control device 100 described in each embodiment is understood as follows.

(1) The control device 100 according to the first aspect includes an information acquisition unit 110 that acquires information about the incinerator G before being supplied to the processing space V in the incineration facility SF, and the information acquired by the information acquisition unit 110. Based on the prediction information including the above information, a steam flow rate prediction unit 150 that predicts the flow rate of the main steam generated by the heat recovery boiler 3 of the incineration facility SF, and the main steam flow rate predicted by the steam flow rate prediction unit 150 and a control unit 160 that performs combustion control based on the control unit 160 . Note that "prediction information" does not mean information dedicated to prediction, but is used in the broad sense of information that can be used for prediction. That is, the prediction information may be information collected or stored mainly for a purpose different from the prediction of the main steam flow rate.

According to such a configuration, the main steam flow rate is predicted based on the information about the incinerator G before being supplied to the processing space V, so it is possible to predict the main steam flow rate with high accuracy. This enables the control device 100 to perform combustion control based on the highly accurate predicted value of the main steam flow rate.

(2) The control device 100 according to the second aspect is the control device 100 according to the first aspect, and the information includes information on the property of the incineration material G. "Information on the properties of the material to be incinerated G" includes, for example, values related to the moisture content of the material to be incinerated G (moisture content or moisture content), weight of the material to be incinerated G, height of the material to be incinerated G, height of the material to be incinerated G , the density (bulk density or true density) of the material G to be incinerated, and the calorific value of the material G to be incinerated.

According to such a configuration, the main steam flow rate can be predicted by reflecting the properties of the incinerator G that affect the main steam flow rate. This makes it possible to predict the main steam flow rate with even higher accuracy.

(3) The control device 100 according to the third aspect is the control device 100 according to the first or second aspect, and the information is the accumulation of the incinerator G in the hopper 11 of the incineration facility SF. Contains pile state information that indicates the state.

According to such a configuration, it is possible to predict the main steam flow rate by reflecting the accumulated state of the incinerator G in the hopper 11, which affects the main steam flow rate. This makes it possible to predict the main steam flow rate with even higher accuracy.

(4) The control device 100 according to the fourth aspect is the control device 100 according to the third aspect, and the accumulation state information is, from the downstream side in the conveying direction D of the incinerator G in the incineration facility SF, An infrared image of the exit portion 11b of the hopper 11 captured through the flame in the processing space V is included.

According to such a configuration, it is possible to reflect the accumulated state of the incineration materials G in the hopper 11 with higher accuracy based on the infrared image. This makes it possible to predict the main steam flow rate with even higher accuracy.

(5) The control device 100 according to the fifth aspect is the control device 100 according to any one of the first to fourth aspects, and the information is sent from the hopper 11 of the incineration facility SF to the processing space V. supply state information indicating the supply state of the incinerator G of

According to such a configuration, the main steam flow rate can be predicted by reflecting the supply state of the incinerator G from the hopper 11 to the processing space V, which affects the main steam flow rate. This makes it possible to predict the main steam flow rate with even higher accuracy.

(6) The control device 100 according to the sixth aspect is the control device 100 of any one of the first to fifth aspects, and is included in the information or obtained from the information, the incineration facility SF A calorific value estimating unit (first calorific value estimation 121 or a second calorific value estimator 122), and the steam flow rate predictor 150 predicts the main steam flow rate based on the lower calorific value estimated by the calorific value estimator.

According to such a configuration, the main steam flow rate can be predicted by reflecting the lower calorific value of the incinerated material G estimated from the density or moisture content measurement result of the incinerated material G in the hopper 11 . This makes it possible to predict the main steam flow rate with even higher accuracy.

(7) The control method according to the seventh aspect acquires information about the garbage G before being supplied to the processing space V in the incineration facility SF, and based on the prediction information including the acquired information, the incineration facility SF predicting the flow rate of the main steam generated by the heat recovery boiler 3, and performing combustion control based on the predicted flow rate of the main steam. According to such a configuration, as with the control device 100 according to the first aspect, it is possible to perform combustion control based on a highly accurate predicted value of the main steam flow rate.

(8) The program according to the eighth aspect causes the computer to acquire information about the garbage G before being supplied to the processing space V in the incineration facility SF, based on the information for prediction including the acquired information, It includes predicting the main steam flow rate generated by the heat recovery boiler 3 of the incineration facility SF and performing combustion control based on the predicted main steam flow rate. According to such a configuration, as with the control device 100 according to the first aspect, it is possible to perform combustion control based on a highly accurate predicted value of the main steam flow rate.

The present disclosure relates to a control device for controlling an incineration facility for, for example, municipal solid waste, industrial waste, or biomass. According to the control device of the present disclosure, combustion control can be performed based on a highly accurate predicted value of the main steam flow rate.

SF... Incineration facility, G... Matter to be incinerated (garbage), 1... Crane, 2... Incinerator, 3... Exhaust heat recovery boiler, 4... Cooling tower, 5... Dust collector, 6... Flue, 7... Chimney , 10... Supply mechanism, 11... Hopper, 11a... Inlet part, 11b... Outlet part, 12... Feeder, 13... Extrusion device, 14... Object measuring instrument, 15... Moisture measuring instrument, 20... Furnace body, 20a... Drying stage , 20b... Combustion stage, 20c... Post-combustion stage, V... Processing space, 21... Visible light camera, 22... Infrared camera, 30... Stoker, 31... Grate, 32... Grate drive device, 41... Wind box, 41a... wind box pressure sensor, 50... blower mechanism, 51... fan, 52... primary air line, 53... air preheater, 54... secondary air line, 55... damper, 56... air flow rate sensor, 61... boiler body, 62... Pipe line, 63... Radiation temperature sensor, 64... In-furnace pressure sensor, 65... Feed water flow rate sensor, 66... Superheater desuperheater flow rate sensor, 100... Control device, 110... Information acquisition unit, 120... Data conversion unit , 130... Prediction model creation unit, 140... Prediction model determination unit, 150... Steam flow rate prediction unit, 160... Control unit

Claims

an information acquisition unit that acquires information about the incinerated material before it is supplied to the processing space in the incineration facility;
a steam flow rate prediction unit that predicts a main steam flow rate generated by a boiler of the incineration facility based on prediction information including the information acquired by the information acquisition unit;
and a control unit that performs combustion control based on the main steam flow rate predicted by the steam flow rate prediction section.
The control device according to claim 1, wherein the information includes information on the properties of the incinerator.
The control device according to claim 1 or claim 2, wherein the information includes accumulation state information indicating the accumulation state of the incinerator in the hopper of the incineration facility.
4. The method according to claim 3, wherein the accumulation state information includes an infrared image obtained by capturing an exit portion of the hopper through a flame of the processing space from a downstream side of the incineration facility in a conveying direction of the material to be incinerated. Control device as described.
5. The control device according to any one of claims 1 to 4, wherein the information includes supply state information indicating a supply state of the material to be incinerated from a hopper of the incineration facility to the processing space.
Low heat generation of the incinerated matter based on the density of the incinerated matter in the hopper of the incineration facility or the moisture content measurement result of the incinerated matter in the hopper included in or obtained from the information further comprising a calorific value estimator for estimating the amount of
The control device according to any one of claims 1 to 5, wherein the steam flow rate prediction section predicts the main steam flow rate based on the lower heating value estimated by the heating value estimation section.