Collapse detection system and collapse detection method

Abstract

The collapse detection system disclosed herein comprises: an acquisition unit that acquires, at a first prescribed cycle, captured images of an object to be incinerated that is deposited in a feeder of an incineration facility and pushed out toward a combustion chamber; a first calculation unit that calculates a representative value of brightness based on a first image acquired by the acquisition unit; a second calculation unit that calculates a representative value of brightness based on two or more images captured within a period of time for one collapse of the object to be incinerated, among a plurality of images in time series that were acquired by the acquisition unit before the first image; and a determination unit that performs determination related to the collapse on the basis of the representative value of brightness calculated by the first calculation unit and the representative value of brightness calculated by the second calculation unit.

Description

[Technical Field]

[0001] This disclosure relates to a collapse detection system and a collapse detection method. This application claims priority to Japanese Patent Application No. 2022-183989, filed on November 17, 2022, the contents of which are incorporated herein by reference. [Background technology]

[0002] Patent Document 1 discloses a supply amount detection system comprising: an imaging device configured to capture an image of solid fuel accumulated in the feeder section of an incinerator before it falls into the combustion chamber; and a detection device that detects the amount of solid fuel supplied to the combustion chamber based on the image captured by the imaging device. In this supply amount detection system, the amount of solid fuel supplied to the combustion chamber is detected based on the difference between a first brightness, which is the brightness of the image at a first timing, and a second brightness, which is the brightness of the image at a second timing later than the first timing and is lower than the first brightness. Specifically, the detection device divides the image into a plurality of partitioned images, counts the number of partitioned images in which the difference between the first brightness and the second brightness in each of the plurality of partitioned images exceeds a preset threshold, and detects that the amount of solid fuel supplied to the combustion chamber is excessive when the count exceeds a preset number.

[0003] Patent Document 2 discloses a fluidized bed incinerator having a dust feeder and a chute connecting the dust outlet of the dust feeder to the waste supply port of a fluidized bed incinerator, characterized in that a television camera is mounted in a position where the falling of waste from the dust outlet of the dust feeder can be observed, and a calculation means is provided to calculate the amount of waste falling and the amount of heat generated by the waste based on the image from the mounted television camera. In this fluidized bed incinerator, the amount of waste is detected by a processing flow that includes (a) image acquisition, (b) image binarization, (c) contour recognition, (d) calculation of the area within the contour, (e) calculation of the centroid within that area, (f) storage of the centroid and area, and (g) multiplying the area by the distance traveled between the previously calculated centroid and the currently calculated centroid.

[0004] Patent Document 3 discloses a combustion field observation device that can visualize and observe a combustion field, such as the inside of a boiler furnace during operation. This observation device has means for converting an image acquired by photographing the combustion field to grayscale, means for adjusting the contrast of an image acquired by photographing the combustion field, or means for converting an image acquired by photographing the combustion field to grayscale and then adjusting the contrast, and generates information for visualizing the combustion field by performing predetermined processing on a single image taken at a certain time. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Patent No. 6979482 [Patent Document 2] Japanese Patent Application Publication No. 9-060842 [Patent Document 3] Japanese Patent Publication No. 2019-196845 [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] Incidentally, complex events occur within the combustion chamber, making it difficult to accurately detect the collapse of the material being incinerated. However, improving the accuracy of detecting the collapse of the material being incinerated is expected in order to better control the combustion chamber.

[0007] This disclosure is made to solve the above-mentioned problems and aims to provide a collapse detection system and a collapse detection method that can improve the accuracy of detecting collapses of materials to be incinerated. [Means for solving the problem]

[0008] To solve the above problems, the collapse detection system according to this disclosure includes: an acquisition unit that acquires images of the material to be incinerated that is accumulated in the feeder of an incineration facility and pushed toward the combustion chamber at a first predetermined period; a first calculation unit that calculates a representative value of brightness based on the first image acquired by the acquisition unit; a second calculation unit that calculates a representative value of brightness based on two or more images taken within the time required for one collapse of the material to be incinerated, from among a plurality of time-series images acquired by the acquisition unit before the first image; and a determination unit that makes a determination regarding the collapse based on the representative value of brightness calculated by the first calculation unit and the representative value of brightness calculated by the second calculation unit.

[0009] The collapse detection method relating to this disclosure includes one or more computers acquiring images of the material to be incinerated that is accumulated in the feeder of the incineration equipment and pushed toward the combustion chamber at a first predetermined period, calculating a representative value of brightness based on the acquired first image, calculating a representative value of brightness based on two or more images taken within the time required for one collapse of the material to be incinerated from a plurality of time-series images acquired before the first image, and making a determination regarding the collapse based on the representative value of brightness based on the first image and the representative values ​​of brightness based on the two or more images.

[0010] The collapse detection system according to the present disclosure includes an acquisition unit that acquires, at a first predetermined period, an image obtained by imaging waste to be incinerated that accumulates in a feeder of an incineration facility and is pushed out toward a combustion chamber, a first input element that is a first image acquired by the acquisition unit, and a second input element that is two or more images captured within the time taken for one collapse of the waste to be incinerated among a plurality of images in time series acquired by the acquisition unit prior to the first image. The collapse detection system includes a collapse detection unit that makes a determination regarding the collapse based on the first input element and the second input element.

Advantages of the Invention

[0011] According to the present disclosure, it is possible to provide a collapse detection system and a collapse detection method capable of improving the detection accuracy regarding the collapse of waste to be incinerated.

Brief Description of the Drawings

[0012] [Figure 1] It is a schematic configuration diagram showing the whole incineration facility according to an embodiment of the present disclosure. [Figure 2] It is a functional block diagram of an information processing apparatus according to an embodiment of the present disclosure. [Figure 3] It is a diagram showing an example of an image to be determined by a first determination unit according to a first embodiment of the present disclosure. [Figure 4] It is a diagram showing an example of an image to be determined by a third determination unit according to a first embodiment of the present disclosure. [Figure 5] It is a diagram showing an example of a determination result in time series by a first determination unit according to a first embodiment of the present disclosure. [Figure 6] It is a diagram exemplarily showing a list of images used for learning by a second learned model for determination according to a first embodiment of the present disclosure. [Figure 7] It is a diagram showing an example of a determination result in time series by a second determination unit according to a first embodiment of the present disclosure. [Figure 8] It is a diagram showing an example of an image to be determined by a disturbance determination unit according to a first embodiment of the present disclosure. [Figure 9]This figure shows examples of low-deviation and high-deviation images used for training by a pre-trained disturbance detection model according to the first embodiment of this disclosure. [Figure 10] This figure shows a list of images used for training by a pre-trained disturbance detection model according to the first embodiment of this disclosure, corresponding to the magnitude of the brightness deviation. [Figure 11] This figure shows an example of a time-series determination result by the disturbance determination unit according to the first embodiment of this disclosure. [Figure 12] This figure shows an example of a time-series determination result by the third determination unit according to the first embodiment of this disclosure. [Figure 13] This flowchart shows an example of the operation of the information processing device according to the first embodiment of this disclosure. [Figure 14] This figure shows an example of a time-series determination result by the first collapse determination unit and the second collapse determination unit according to the first embodiment of this disclosure. [Figure 15] This is a diagram illustrating the binarized data according to the second embodiment of this disclosure. [Figure 16] This figure conceptually illustrates the method for calculating similarity by the second collapse determination unit according to the second embodiment of this disclosure. [Figure 17] This flowchart shows an example of the operation of the information processing device according to the second embodiment of this disclosure. [Figure 18] This figure shows an example of a time-series determination result by the first collapse determination unit and the second collapse determination unit according to the second embodiment of this disclosure. [Figure 19] This is a functional block diagram of the information processing apparatus according to the third embodiment of this disclosure. [Figure 20] This figure shows an example of an image that the first calculation unit according to the third embodiment of this disclosure is to calculate. [Figure 21] This figure illustrates a list of images used for training by the fourth pre-trained model for decision-making according to the third embodiment of this disclosure. [Figure 22] This table shows the numerical range of the sum of the luminance change amounts used as the basis for judgment by the third collapse judgment unit according to the third embodiment of this disclosure, and the judgment results of the fourth and fifth judgment units. [Figure 23] This flowchart shows an example of the operation of the information processing device according to the third embodiment of this disclosure. [Figure 24] This is a hardware configuration diagram showing the configuration of a computer according to the embodiments of this disclosure. [Modes for carrying out the invention]

[0013] The configuration for implementing the incineration facility and the collapse detection system will be described below with reference to the attached drawings.

[0014] <First embodiment of incineration equipment> The incineration equipment 100 is a stoker-type waste incinerator that incinerates, for example, municipal solid waste, industrial waste, or biomass. Hereinafter, the material to be incinerated may be referred to as "waste." In other words, in this embodiment, waste is fuel for generating a combustion reaction within the incineration equipment. As shown in Figure 1, the incineration equipment 100 includes, for example, a hopper 102, a feeder 104, a furnace body 108, an extruder 110, an air supply device 112, a heat recovery boiler 114, a cooling tower 116, a dust collector 118, a chimney 120, and a collapse detection system 1.

[0015] The feeder 104 is a passage extending toward the combustion chamber R of the furnace body 108. Waste Fg fed in from the hopper 102 is introduced into the feeder 104 and temporarily piled up there. The furnace body 108 has a combustion chamber R inside for incinerating the waste Fg. If the direction in which waste Fg is transported within the furnace body 108 is defined as the transport direction W1 (left-right direction in Figure 1), then the downstream end 121 of the feeder 104 on the downstream side of the transport direction W1 is connected to the inlet 122 of the combustion chamber R.

[0016] The extrusion device 110 has an extrusion arm 124 for pushing out the waste Fg accumulated in the feeder 104 into the combustion chamber R via the receiving port 122. The extrusion arm 124 is movable (advancing and retracting) within the feeder 104 from the upstream side to the downstream side and from the downstream side to the upstream side in the conveying direction W1. In this embodiment, the extrusion arm 124 reciprocates within the feeder 104 in the conveying direction W1, intermittently supplying waste Fg into the combustion chamber R. In this embodiment, the extrusion arm 124 is an example of a controlled device S.

[0017] The furnace body 108 includes a grate 126 (stoker) through which waste Fg pushed into the combustion chamber R via the inlet 122 falls. The grate 126 corresponds to the floor of the combustion chamber R. The grate 126 moves the waste Fg on the grate 126 away from the inlet 122 (downstream in the transport direction W1). The grate 126 is an example of the controlled device S. The combustion chamber R also includes a drying region 128, a combustion region 130, and a post-combustion region 132, which are arranged in order from the upstream side to the downstream side in the transport direction W1. The drying region 128 dries the waste Fg with the heat inside the combustion chamber R. The combustion region 130 burns the waste Fg with a flame 131. The post-combustion region 132 completely burns any waste that was not burned in the combustion region 130. The waste Fg that has been dried, burned, and post-burned in the combustion chamber R becomes ash 135, which falls from the ash chute 146 located downstream of the post-burn area 132 and is discharged to the outside of the furnace body 108.

[0018] The air supply device 112 supplies primary air used for burning waste Fg, and secondary air used to reduce the concentration of unburned gases such as carbon monoxide generated by the combustion of waste Fg, to the combustion chamber R. The air supply device 112 includes an air supply pipe 136, a blower 138 provided on the air supply pipe 136, and a first flow control valve 140 and a second flow control valve 142 provided on the air supply pipe 136. A portion of the air pumped from the blower 138 and flowing through the air supply pipe 136 is supplied as primary air from the bottom of the combustion chamber R through the grate 126, with its flow rate adjusted by the first flow control valve 140. The remaining portion of the air flowing through the air supply pipe 136 is supplied as secondary air from the side wall of the combustion chamber R to the upper part of the combustion chamber R, with its flow rate adjusted by the second flow control valve 142. In this embodiment, for example, primary air is supplied to the dry region 128, the combustion region 130, and the post-combustion region 132 of the combustion chamber R, and secondary air is supplied to the upper side of the combustion region 130. In this embodiment, the blower 138, the first flow control valve 140, and the second flow control valve 142 are all examples of controlled devices S.

[0019] The heat recovery boiler 114, the cooling tower 116, the dust collector 118, and the chimney 120 are each located in the flue 144 through which the exhaust gas 143, generated by the combustion of waste Fg in the combustion chamber R, flows. The exhaust gas 143 flows in the order of heat recovery boiler 114, cooling tower 116, dust collector 118, and chimney 120. The heat recovery boiler 114 generates steam from the thermal energy of the exhaust gas 143. The cooling tower 116 lowers the temperature of the exhaust gas 143 that has passed through the heat recovery boiler 114. The dust collector 118 collects fly ash contained in the exhaust gas 143 that has passed through the cooling tower 116. The chimney 120 exhausts the exhaust gas 143 that has passed through the dust collector 118 to the outside of the incineration facility 100. The steam generated by the heat recovery boiler 114 is supplied to, for example, a steam turbine (not shown) located outside the incineration facility 100.

[0020] <Collapse detection system> The collapse detection system 1 detects when the waste Fg accumulated in the feeder 104 is pushed out toward the combustion chamber R by the extrusion arm 124 and supplied to the combustion chamber R. Specifically, the collapse detection system 1 detects the collapse of waste Fg from the feeder 104 into the combustion chamber R. Here, "collapse" means, for example, that a certain amount of waste Fg accumulated in the feeder 104 is supplied to the combustion chamber R at once. In this embodiment, collapses are classified into a first-scale collapse and a second-scale collapse, which is larger in scale than the first-scale collapse.

[0021] A first-scale collapse includes cases where the layer structure of approximately one-third of the total waste Fg accumulated in the feeder 104 in the infrared image captured by the imaging device 2 described later becomes invisible in the width direction of the furnace body 108 (a direction perpendicular to the transport direction W1), and where some of the waste Fg after the collapse scatters inside the combustion chamber R. Furthermore, a first-scale collapse also includes cases where the layer structure of more than one-third but less than two-thirds of the total waste Fg accumulated in the feeder 104 in the infrared image becomes invisible in the width direction of the furnace body 108, and where some of the waste Fg after the collapse does not scatter inside the combustion chamber R.

[0022] On the other hand, a second-scale collapse includes cases where the layered structure of more than two-thirds of the waste Fg accumulated in the feeder 104 in the infrared image becomes invisible in the width direction of the furnace body 108, and where some of the waste Fg after the collapse is scattered inside the combustion chamber R. Furthermore, a second-scale collapse also includes cases where waste Fg falls onto the flame 131 in the visible light image captured by the imaging device 2 described later, causing the flame 131 to extinguish in an area of ​​more than one-third of the width direction of the furnace body 108.

[0023] Therefore, in this embodiment, the collapse detection system 1 detects the scale (amount) of waste Fg that has collapsed into the combustion chamber R. As shown in Figures 1 and 2, the collapse detection system 1 includes, for example, an imaging device 2 and an information processing device 4.

[0024] [Imaging device] The imaging device 2 captures an image of the combustion chamber R so as to capture the waste Fg accumulated in the feeder 104. The image of the waste Fg captured by the imaging device 2 is transmitted to the information processing device 4 in real time. The imaging device 2 is positioned on the furnace body 108 to capture an image of the front surface Fr of the waste Fg that faces the combustion chamber R before it collapses into the combustion chamber R. Specifically, the imaging device 2 is located at the furnace tail 145 of the furnace body 108, downstream in the transport direction W1 from the post-combustion region 132 in the combustion chamber R. However, the imaging device 2 may be installed at a location other than the furnace tail 145 of the furnace body 108 as long as it is possible to capture infrared and visible light images of the front surface Fr of the waste Fg.

[0025] In this embodiment, the imaging device 2 includes an infrared camera 5 capable of capturing infrared images and a visible light camera 6 capable of capturing visible light images (see Figure 1). The imaging device 2 is capable of capturing infrared and visible light images of the front surface Fr of the waste Fg that protrudes downstream in the transport direction W1 from the receiving inlet 122 of the combustion chamber R. The infrared camera 5 captures the front surface Fr of the waste Fg in the wavelength band of, for example, 3.8 μm to 4.2 μm and generates an infrared image. Since the infrared camera 5 takes images in the above wavelength band, it can transmit the flame 131, and the appearance of the flame 131 in the generated infrared image is suppressed. The visible light camera 6 captures the front surface Fr of the waste Fg in a predetermined wavelength band in the visible wavelength range and generates a visible light image. Since the visible light camera 6 takes images in the above wavelength band, it cannot transmit the flame 131, and the flame 131 is mainly reflected in the generated visible light image.

[0026] [Information Processing Device] The information processing device 4 detects information about the waste Fg supplied from the feeder 104 to the combustion chamber R based on the image captured by the imaging device 2. As shown in Figure 2, the information processing device 4 includes, for example, an acquisition unit 40, a collapse detection unit 41, a control unit 80, and a storage unit 90.

[0027] (Configuration of the acquisition unit) The acquisition unit 40 acquires images over time by receiving images transmitted from the imaging device 2 in real time. In this embodiment, the acquisition unit 40 acquires infrared images from the imaging device 2 at a first predetermined period and visible light images from the imaging device 2 at a second predetermined period. The first and second predetermined periods are determined, for example, based on the frame rate (fps: frames per second) of the imaging device 2. The acquisition unit 40 sends the acquired infrared and visible light images to the collapse detection unit 41.

[0028] (Configuration of the collapse detection unit) The collapse detection unit 41 detects, based on infrared and visible light images received from the acquisition unit 40, that waste Fg has collapsed from the feeder 104 into the combustion chamber R, and the scale (amount) of waste Fg that has collapsed and been supplied to the combustion chamber R. The collapse detection unit 41 includes, for example, a first calculation unit 50, a second calculation unit 55, a third calculation unit 60, a fourth calculation unit 65, and a determination unit 70.

[0029] (First Calculation Unit) The first calculation unit 50 receives an infrared image from the images received by the acquisition unit 40 and calculates a first feature quantity based on the received infrared image. In this embodiment, the first feature quantity is a representative value of brightness based on one infrared image (for example, the most recent infrared image) received by the first calculation unit 50. Hereinafter, the infrared image used by the first calculation unit 50 to calculate the representative value of brightness will be referred to as the "first infrared image". That is, the first calculation unit 50 calculates a representative value of brightness based on the first infrared image acquired by the acquisition unit 40. The first infrared image is an example of a first image. In this embodiment, the first calculation unit 50 calculates a representative value of brightness in a specific target region, which is a part of the first infrared image. Hereinafter, this target region will be referred to as the "first target region 42".

[0030] As shown in Figure 3, the first calculation unit 50 calculates a representative value of brightness for the area in the first infrared image where dust Fg accumulated in the feeder 104 is mainly visible, designating this area as the first target area 42. In this embodiment, the first target area 42 that the first calculation unit 50 calculates is divided into multiple areas. In Figure 3, one example is shown where the first target area 42 is equally divided into nine 3x3 matrix areas. Note that the first target area 42 is not limited to being equally divided into nine matrix areas, and may be divided into 2 to 8 areas or 10 or more areas. Hereinafter, the first target area 42 shown in Figure 3 will be referred to as "first area 42a", "second area 42b", "third area 42c", "fourth area 42d", "fifth area 42e", "sixth area 42f", "seventh area 42g", "eighth area 42h", ​​and "ninth area 42i" in order from the top left to the bottom right. The first calculation unit 50 calculates, for example, the average brightness of each of the first region 42a to the ninth region 42i in the first target region 42 in the first infrared image, and then calculates the average value of the entire first target region 42 obtained by further averaging the calculated average brightness of the nine regions as a representative value of brightness based on the first infrared image. Note that the representative value calculated by the first calculation unit 50 is not limited to the average value, but may be a statistical quantity such as the median. In other words, the method of calculating the representative value by the first calculation unit 50 is not necessarily limited to the above. The first calculation unit 50 sends the calculated representative value of brightness in the first target region 42 to the determination unit 70.

[0031] (Second Calculation Department) The second calculation unit 55 receives infrared images from the images received by the acquisition unit 40 and calculates a second feature quantity based on the multiple infrared images received. In this embodiment, the second feature quantity is a representative value of brightness based on the multiple infrared images received by the second calculation unit 55. Specifically, the second calculation unit 55 calculates a representative value of brightness based on multiple infrared images captured within the time required for one collapse of waste Fg, from among two or more infrared images in a time series acquired before the first infrared image used by the first calculation unit 50 to calculate the representative value. The "time required for one collapse" here refers to, for example, the time obtained by averaging multiple times after the time at which a human judged that a single collapse of waste Fg had occurred has been acquired. Specifically, the time required for one collapse of waste Fg is defined by calculating the average number of frames from the distribution of the number of frames (number of images) of the imaging device 2 that indicate a single collapse.

[0032] The two or more infrared images used by the second calculation unit 55 to calculate the representative value of brightness are infrared images captured at predetermined time intervals from each other. Hereinafter, the time interval between the two or more infrared images will be referred to as the "first time". In this embodiment, the second calculation unit 55 calculates the representative value of brightness based on each of the infrared images, which are the number of frames acquired per unit time by the acquisition unit 40. Therefore, the first time is the first predetermined period. The first time is, for example, less than 1 second in length.

[0033] In this embodiment, the two or more infrared images are infrared images captured before the first infrared image over a second time period that is at least twice the length of the first time period. The second calculation unit 55 calculates a representative value of brightness based on a plurality of infrared images, for example, three or more (more specifically, four or more) infrared images as the two or more infrared images. Here, S is the time required for one collapse of waste Fg, and T is the first predetermined period. i Let N be the number of infrared images (frames) used in the calculation. i In that case, the following equation (i) holds true. S / 2 <T i ×N i …(i) In other words, the multiple infrared images are multiple images acquired over a period of time longer than at least half the time S. In this embodiment, the multiple infrared images are acquired over a period of time S. Hereinafter, each of the multiple infrared images used by the second calculation unit 55 to calculate a representative value of brightness will be referred to as the "second infrared image". The second infrared image is an example of a second image.

[0034] The second calculation unit 55 calculates a representative value of the brightness of a specific target region, which is a part of the second infrared image. The target region that the second calculation unit 55 calculates is the same region as the first target region 42 described above. The second calculation unit 55 calculates the average brightness of each of the first region 42a to the ninth region 42i of the first target region 42, and then calculates the average value of the entire first target region 42 by further averaging the average values ​​of the nine brightness values ​​that have been calculated. Furthermore, the second calculation unit 55 calculates the average value of the entire first target region 42 as a representative value based on the multiple second infrared images. Note that the representative value calculated by the second calculation unit 55 is not limited to the average value, but may be a statistical quantity such as the median. In other words, the method of calculating the representative value by the second calculation unit 55 is not necessarily limited to the above. The second calculation unit 55 sends the representative values ​​of the entire brightness of the multiple first target regions 42 based on the multiple second infrared images to the determination unit 70.

[0035] (Third Calculation Department) The third calculation unit 60 receives a visible light image from the images received by the acquisition unit 40 and calculates a third feature quantity based on the received visible light image. In this embodiment, the third feature quantity is a representative value of luminance based on one visible light image (for example, the most recent visible light image) received by the third calculation unit 60. Hereinafter, the visible light image used by the third calculation unit 60 to calculate the representative value of luminance will be referred to as the "first visible light image". That is, the third calculation unit 60 calculates a representative value of luminance based on the first visible light image acquired by the acquisition unit 40. The first visible light image is another example of the first image. An example of a visible light image received by the third calculation unit 60 is shown in Figure 4. In Figure 4, for illustrative purposes, the first visible light image (a) is shown in monochrome (black and white). As shown in Figure 4, in this embodiment, the third calculation unit 60 generates an image (b) in which only the red component of the monochromatic components is extracted from the first visible light image (a), and calculates a representative value of the brightness in a specific target region, which is a part of the image (b) in which only the red component is extracted. Hereinafter, the image in which only the red component is extracted from the first visible light image by the third calculation unit 60 will be referred to as the "monochromatic component image". The target region will be referred to as the "second target region 43". The monochromatic component image is an example of a visible light image. Note that the monochromatic component extracted by the third calculation unit 60 from the first visible light image is not limited to the red component; for example, only the green component or only the blue component may be used as the monochromatic component to generate a monochromatic component image from the first visible light image.

[0036] The third calculation unit 60 calculates a representative value of luminance in the second target region 43, which is the area in the monochromatic component image where the flame 131 is mainly reflected. Here, the second target region 43 that the third calculation unit 60 calculates is divided into multiple equal regions. In Figure 4, an example is shown in which the second target region 43 is divided into three regions in the width direction of the furnace body 108 (a direction perpendicular to the transport direction W1). Note that the second target region 43 is not limited to being divided into three equal regions, but may be divided into two regions or four or more regions. Hereinafter, the second target region 43 shown in Figure 4 will be referred to as the "left region 43l," the "central region 43c," and the "right region 43r" from left to right. The third calculation unit 60 calculates the median value of the luminance of the left region 43l, the central region 43c, and the right region 43r in the second target region 43 in the monochromatic component image as a representative value. The representative value calculated by the third calculation unit 60 is not limited to the median, but may be a statistical quantity such as the mean. In other words, the method of calculating the representative value by the third calculation unit 60 is not necessarily limited to the above. The third calculation unit 60 sends the representative values ​​of the brightness of the left region 43l, the central region 43c, and the right region 43r in the calculated monochromatic component image to the determination unit 70.

[0037] (Fourth Calculation Section) The fourth calculation unit 65 receives visible light images from the images received by the acquisition unit 40 and calculates a fourth feature quantity based on the multiple visible light images received. In this embodiment, the fourth feature quantity is a representative value of luminance based on the multiple visible light images received by the fourth calculation unit 65. Specifically, the fourth calculation unit 65 calculates a representative value of luminance based on multiple visible light images captured within the time required for one collapse of waste Fg, among two or more visible light images in a time series acquired before the first visible light image used by the third calculation unit 60 to calculate the representative value.

[0038] The four calculation units 65 use two or more visible light images to calculate a representative value of luminance, which are visible light images captured at predetermined time intervals from each other. Hereinafter, the time interval between the two or more visible light images will be referred to as the "third time interval". In this embodiment, the four calculation units 65 calculate a representative value of luminance based on each of the visible light images, which are the number of frames acquired per unit time by the acquisition unit 40. Therefore, the third time interval is the second predetermined period described above. The third time interval is, for example, less than one second in length.

[0039] In this embodiment, the two or more visible light images are images taken before the first visible light image over a fourth period of time that is at least twice the third period. The fourth calculation unit 65 calculates a representative value of brightness based on a plurality of visible light images, for example, three or more (more specifically, four or more) visible light images as the two or more visible light images. In this embodiment, the fourth calculation unit 65 calculates a representative value of brightness based on each of the number of images acquired per unit time by the acquisition unit 40, which is the number of frames acquired per unit time. Here, S is the time required for one collapse of waste Fg, and T is the second predetermined period. ii Let N be the number of visible light images (frames) used in the calculation. ii In this case, equation (ii) below holds true. S / 2 <T ii ×N ii …(ii) In other words, the multiple visible light images are multiple images acquired over a period of time longer than at least half the time S. In this embodiment, the multiple visible light images are acquired over a period of time S. Hereinafter, each of the multiple images used by the fourth calculation unit 65 to calculate a representative value of luminance will be referred to as the "second visible light image". The second visible light image is an example of a second image.

[0040] The fourth calculation unit 65 generates a monochromatic component image from the second infrared image, and calculates a representative value of the luminance of a specific target region that is a part of the monochromatic component image. The target region for which the fourth calculation unit 65 performs the calculation is the same region as the second target region 43 described above. The fourth calculation unit 65 calculates the median value of the luminance of each of the left region 43l, the central region 43c, and the right region 43r of the second target region 43. Further, the fourth calculation unit 65 calculates the median value of the entire left region 43l, the entire central region 43c, and the entire right region 43r as the representative value based on a plurality of monochromatic component images. Note that the representative value calculated by the fourth calculation unit 65 is not limited to the median value, and may be a statistical quantity such as an average value, for example. That is, the method for calculating the representative value by the fourth calculation unit 65 is not necessarily limited to the above. The fourth calculation unit 65 sends the representative values of the luminance of each of the plurality of entire left regions 43l, the plurality of entire central regions 43c, and the plurality of entire right regions 43r based on the calculated plurality of monochromatic component images to the determination unit 70.

[0041] (Determination unit) Returning to FIG. 2, the configuration of the determination unit 70 will be described. The determination unit 70 performs various determination processes (described later) based on the image received from the acquisition unit 40 and the feature amounts (for example, representative values of luminance) received from each of the first calculation unit 50, the second calculation unit 55, the third calculation unit 60, and the fourth calculation unit 65. In the present embodiment, the determination unit 70 includes, for example, a first determination unit 71, a second determination unit 72, a disturbance determination unit 74, a third determination unit 73, a first collapse determination unit 75, and a second collapse determination unit 76.

[0042] (First determination unit) The first determination unit 71 performs a determination regarding collapse based on the representative value of luminance (first feature amount) received from the first calculation unit 50 and the representative value of luminance (second feature amount) received from the second calculation unit 55. Hereinafter, the determination by the first determination unit 71 is referred to as "first determination". The first determination unit 71 calculates a feature change amount V1 (absolute value) that occurs between the first infrared image and the plurality of second infrared images that are the targets of the first determination according to the following formula (iii). V1 = |A m (t)-(ΣB m (ts )) / s| …(iii)

[0043] A in the above formula (iii) m (t) is the representative value (first feature quantity) of luminance calculated by the first calculation unit 50 at time t. ΣB m (t s ) is a representative value of luminance (second feature quantity) calculated by the second calculation unit 55. s t are multiple time points at which each second infrared image was acquired, and time t is set for one second infrared image. s Each corresponds to a specific element. s is the number of second infrared images, which is the number of frames acquired per unit time.

[0044] The first determination unit 71 performs a first determination to determine whether the calculated feature change amount V1 is greater than or equal to a predetermined first threshold. The first determination unit 71 determines that there is a possibility of collapse if the calculated feature change amount V1 is greater than or equal to the first threshold, and determines that there is no possibility of collapse if the feature change amount V1 is less than the first threshold. In this embodiment, the first determination unit 71 outputs a flag (i=1) indicating that there is a collapse if the feature change amount V1 is greater than or equal to the first threshold. On the other hand, the first determination unit 71 outputs a flag (i=0) indicating that there is no collapse if the feature change amount V1 is less than the first threshold. Hereinafter, the flag (i=1 or 0) related to the collapse determination output by the first determination unit 71 will be referred to as the "first collapse determination flag". In other words, the first collapse determination flag indicates the result of the first determination and shows a value of 1 or 0. The first threshold is stored in advance in the storage unit 90. The first determination unit 71 performs a first determination by referencing a first threshold value stored in the memory unit 90 in a timely manner. The first determination unit 71 sends the first collapse determination flag, which is the result of the first determination, to the first collapse determination unit 75 and the second collapse determination unit 76.

[0045] Here, Figure 5 shows the time-series change of the feature change amount V1 during a specific time period, and the time-series change of the first collapse judgment flag corresponding to the feature change amount V1 during the same time period. The graph in Figure 5 is the result obtained by the inventors' analysis. In Figure 5(b), the time when the operator (skilled worker) of the incineration equipment 100 determined that there was a collapse of waste Fg by visually inspecting the combustion chamber R is indicated by a circle (〇). Also, the time corresponding to the interval between the vertically extending time scale lines (dotted lines) in the graph shown in Figure 5 is the same for all intervals. From the results shown in Figure 5, it can be seen that the time when the operator of the incineration equipment 100 determined that there was a collapse and the time when the first collapse judgment flag was raised (the time when i=1) roughly coincide.

[0046] (Second judgment section) The second determination unit 72 makes a determination regarding collapse based on one or more infrared images received from the acquisition unit 40. In this embodiment, the second determination unit 72 makes a determination regarding collapse using one first infrared image (the most recent infrared image) from the images received from the acquisition unit 40 that the first calculation unit 50 is targeting for calculation. The second determination unit 72 may also use one image different from the first infrared image from the images received from the acquisition unit 40 when making a collapse determination. Furthermore, the second determination unit 72 may use multiple images when making a collapse determination, and the first infrared image may be included among the multiple images.

[0047] When the second determination unit 72 receives the first infrared image as input, it performs a determination using a trained model that has been trained to output a determination result regarding the possibility of collapse occurring. Hereinafter, the determination by the second determination unit 72 will be referred to as the "second determination," and the trained model used by the second determination unit 72 for the second determination will be referred to as the "trained model 91 for the second determination." The trained model 91 for the second determination is pre-stored in the memory unit 90 (see Figure 2). The second determination unit 72 inputs the first infrared image received from the acquisition unit 40 into the trained model 91 for the second determination stored in the memory unit 90, and obtains the output determination result as the result of the second determination. The trained model 91 for the second determination is, for example, a deep learning model (supervised learning model) such as a convolutional neural network (CNN). The second pre-trained model 91 for judgment is generated (learned) by repeating a learning step multiple times (for example, several thousand times) in which an infrared image captured by the imaging device 2 is input, and the presence or absence of collapse in the infrared image (ground truth data judged correct by a human) is taught. Note that instead of a CNN, a recurrent neural network (RNN) or the like may be used for the second pre-trained model 91 for judgment.

[0048] Here, among the infrared images captured by the imaging device 2, image examples (a,d) showing that debris Fg is scattered during collapse (collapse present), and image examples 1 (b,e) and 2 (c,f) showing that there is no collapse (no collapse), are shown in Figure 6, divided into cases where the inside of the combustion chamber R is easily visible and cases where it is not. As shown by the dashed lines in each image example in Figure 6, in this embodiment, the second determination unit 72 inputs the infrared image of the target region, which is a part of the first infrared image, to the second determination trained model 91. Hereinafter, this target region will be referred to as the "third target region 44". The third target region 44 is, for example, the majority of the upper half of the first infrared image, and is a region in which at least all or most of the debris Fg accumulated in the feeder 104 is captured. As shown in the example image (a) in Figure 6, if waste Fg collapses, some (multiple) of the collapsed waste Fg may temporarily scatter within the combustion chamber R, making it impossible to visually confirm the layered structure of waste Fg accumulated in the feeder 104. Multiple infrared images of the third target region 44, as shown in Figure 6, are input to the second pre-trained model 91 for judgment, and the presence or absence of collapse (ground truth data) for each infrared image is taught to the second pre-trained model 91 for judgment, thereby generating the second pre-trained model 91 for judgment in advance. Once the second pre-trained model 91 for judgment has completed its training, when a new infrared image is input, it outputs a numerical value regarding the possibility of collapse occurring in that infrared image. Hereinafter, the numerical value output by the second pre-trained model 91 for judgment will be referred to as the "judgment score".

[0049] The second determination unit 72 performs a second determination to determine whether the determination score obtained from the second determination trained model 91 is equal to or greater than a predetermined second threshold. The second determination unit 72 determines that there is a possibility of collapse if the obtained determination score is equal to or greater than the second threshold, and determines that there is no possibility of collapse if the determination score is less than the second threshold. In this embodiment, the second determination unit 72 outputs a flag (i=1) indicating that there is a collapse if the determination score is equal to or greater than the second threshold. On the other hand, the second determination unit 72 outputs a flag (i=0) indicating that there is no collapse if the determination score is less than the second threshold. Hereinafter, the flag (i=1 or 0) related to the collapse determination output by the second determination unit 72 will be referred to as the "second collapse determination flag". In other words, the second collapse determination flag indicates the result of the second determination and shows a value of 1 or 0. The second threshold is stored in advance in the storage unit 90. The second determination unit 72 performs a second determination by referencing the second threshold value stored in the memory unit 90 in a timely manner. The second determination unit 72 sends the second collapse determination flag, which is the result of the second determination, to the first collapse determination unit 75.

[0050] Here, Figure 7 shows the time-series changes in the judgment score during the time period shown in Figure 5, and the time-series changes in the second collapse judgment flag corresponding to the judgment score during the same time period. The graph in Figure 7 is the result obtained by the inventors' analysis. In Figure 7(b), the time when the operator of the incineration equipment 100 determined that there was a collapse of waste Fg by visually inspecting the combustion chamber R is indicated by a circle (〇). Also, the time corresponding to the interval between the vertically extending time scale lines (dotted lines) in the graph shown in Figure 7 is the same for all intervals. From the results shown in Figure 7, it can be seen that the time when the operator of the incineration equipment 100 determined that there was a collapse and the time when the second collapse judgment flag was raised (the time when i=1) roughly coincide.

[0051] (Disturbance detection unit) The disturbance detection unit 74 determines the presence or absence of disturbances based on the brightness feature quantity of one image (for example, the most recent image) received from the acquisition unit 40. Hereinafter, the determination made by the disturbance detection unit 74 will be referred to as "disturbance determination". In this embodiment, the disturbance detection unit 74 performs disturbance determination using the first infrared image that the first calculation unit 50 is targeting for calculation from among the images received from the acquisition unit 40. The disturbance detection unit 74 may use an image different from the first infrared image from among the images received from the acquisition unit 40 when performing disturbance determination. Furthermore, the disturbance detection unit 74 may use multiple images when performing disturbance determination, and the first infrared image may be included among the multiple images. In addition, when the disturbance detection unit 74 receives an infrared image from the acquisition unit 40, it may convert it from RAW data (16 bits) to, for example, BMP data (8 bits).

[0052] The disturbance detection unit 74 calculates, for example, a feature quantity relating to the overall brightness of the first target region 42 in the first infrared image, a feature quantity relating to the brightness of each of the first region 42a to the ninth region 42i in the first target region 42, and a feature quantity relating to the brightness of multiple target regions that are part of the first infrared image but different from the first target region 42. Hereinafter, these multiple target regions that are different from the first target region 42 will be referred to as the "fourth target region 45". As shown in Figure 8, the fourth target region 45 is divided into three locations in the first infrared image, and these three locations of the fourth target region 45 are independent of each other. Here, "independent" means that each of the fourth target regions 45 is spaced apart from each other in the first infrared image. One of the fourth target regions 45 (45a in Figure 8) is the part above the first target region 42 in the first infrared image, and this fourth target region 45 reflects, for example, the ceiling of the furnace body 108. Furthermore, one of the fourth target regions 45 (45b in Figure 8) is the portion to the right of the first target region 42 in the first infrared image, and this fourth target region 45 includes, for example, the side wall of the furnace body 108. Also, one of the fourth target regions 45 (45c in Figure 8) is the portion below the first target region 42 in the first infrared image, and this fourth target region 45 includes, for example, waste Fg that is post-combusted in the post-combustion region 132. Therefore, waste Fg accumulated in the feeder 104 is hardly captured in the fourth target region 45. In this embodiment, each of the multiple fourth target regions 45 is of a different size from the others.

[0053] The disturbance detection unit 74 calculates two types of statistical quantities as features for the entire first target region 42, the first region 42a to the ninth region 42i, and the multiple fourth target regions 45. In this embodiment, the two types of statistical quantities are the mean value of luminance and the standard deviation of luminance. Hereinafter, the feature quantity relating to the luminance of the entire first target region 42 will be referred to as "Feature Quantity A," the feature quantities relating to luminance in each region of the first region 42a to the ninth region 42i will be referred to as "Feature Quantity B," and the feature quantities relating to luminance in each of the multiple fourth target regions 45 (45a, 45b, 45c) will be referred to as "Feature Quantity C." In this embodiment, the disturbance detection unit 74 determines whether the standard deviation of each calculated feature quantity (Feature Quantity A to Feature Quantity C) falls within a predetermined numerical range, which is a first range. The disturbance determination unit 74 determines that the first infrared image is a high-deviation image if, for example, one or more of the standard deviations of feature A, feature B, and feature C are not included in the first range (are outside the first range). On the other hand, the disturbance determination unit 74 determines that the first infrared image is a low-deviation image if all of the standard deviations of feature A, feature B, and feature C are included in the first range. The disturbance determination unit 74 may also determine that the first infrared image is a high-deviation image if two or more, or all, of the standard deviations of feature A, feature B, and feature C are not included in the first range. The first range is pre-stored in the memory unit 90. The disturbance determination unit 74 makes a determination on the first infrared image by referring to the first range stored in the memory unit 90 in a timely manner.

[0054] The disturbance detection unit 74 performs a determination using a pre-trained model that has been trained to output a determination result indicating the presence or absence of a disturbance when the above-mentioned features are input. Hereinafter, the pre-trained model used by the disturbance detection unit 74 will be referred to as the "pre-trained disturbance detection model 92". The pre-trained disturbance detection model 92 is pre-stored in the memory unit 90 (see Figure 2). The disturbance detection unit 74 inputs features A, B, and C to the pre-trained disturbance detection model 92 stored in the memory unit 90, and obtains the output determination result as the disturbance detection result. In this embodiment, the pre-trained disturbance detection model 92 has a high-deviation model and a low-deviation model. When the disturbance detection unit 74 determines that the first infrared image is a high-deviation image, it inputs the features (features A, B, and C) only to the high-deviation model, and obtains the determination result output from the high-deviation model as the disturbance detection result. On the other hand, if the disturbance detection unit 74 determines that the first infrared image is a low-deviation image, it inputs the feature quantities only into the low-deviation model and obtains the determination result output from the low-deviation model as the disturbance detection result.

[0055] In this embodiment, the high-deviation model and the low-deviation model are, for example, supervised learning models such as SVM (Support Vector Machine). Both the high-deviation model and the low-deviation model are generated (learned) by repeating a learning step multiple times in which the aforementioned features are input and whether or not there is collapse in the image (ground truth data judged to be correct by humans) is taught.

[0056] Here, Figure 9 shows an example of a low-deviation image (a) and an example of a high-deviation image (b) among the infrared images captured by the imaging device 2. Figure 10 also shows a list of infrared image examples corresponding to the magnitude of the brightness deviation. As shown in Figures 9 and 10, it can be seen that the larger or smaller the brightness deviation, which is a feature quantity, the more difficult it is to grasp the conditions inside the combustion chamber R. When feature quantities based on multiple infrared images as shown in Figures 9 and 10 are input to the pre-trained disturbance detection model 92, and the presence or absence of disturbances (ground truth data) for each infrared image is taught, the pre-trained disturbance detection model 92 is generated in advance. When a new feature quantity based on an infrared image is input to the trained disturbance detection model 92, it outputs binary data (for example, 0 and 1) indicating the presence or absence of collapse for that infrared image. The disturbance detection unit 74 determines the presence or absence of a disturbance as a disturbance detection by outputting a flag corresponding to the value of the binary data obtained from the pre-trained disturbance detection model 92. Hereinafter, the flags related to disturbance detection output by the disturbance detection unit 74 will be referred to as "disturbance detection flags." Therefore, the disturbance detection flags indicate the result of the disturbance detection.

[0057] Here, Figure 11 shows the time-series changes in the judgment results regarding the presence or absence of disturbances by the operator of the incineration equipment 100 during a specific time period, and the time-series changes in the disturbance judgment flag during the same time period. The graph in Figure 11 is the result obtained by the inventors' analysis. Note that the time periods T0 to T9 shown in Figure 11 are the same time periods T0 to T9 shown in Figures 5 and 7. Also, the time corresponding to the interval between the vertically extending time scale lines (dotted lines) in the graph shown in Figure 11 is the same for all intervals. From the results shown in Figure 11, it can be seen that the time when the operator of the incineration equipment 100 determined that there was a disturbance by visually inspecting the combustion chamber R and the time when the disturbance judgment flag was raised almost coincide.

[0058] (Third Judgment Department) The third determination unit 73 makes a determination regarding collapse based on one or more images received from the acquisition unit 40. In this embodiment, the third determination unit 73 receives a visible light image from the images received from the acquisition unit 40 and calculates a representative value of luminance based on the received visible light image. The third determination unit 73 makes a determination regarding collapse based on the representative value of luminance (third feature) received from the third calculation unit 60 and the representative value of luminance (fourth feature) received from the fourth calculation unit 65. Hereinafter, the determination by the third determination unit 73 will be referred to as the "third determination". The third determination unit 73 calculates the amount of feature change V2 (absolute value) that occurs between the monochromatic component image based on the first visible light image that is the subject of the third determination and the monochromatic component images based on a plurality of second visible light images, for each region (left region 43l, central region 43c, right region 43r), according to the following formula (iv). V2 = |C m (t)-(ΣD m (t u )) / u| …(iv)

[0059] C in equation (iv) above m (t) is the representative value of luminance (third feature) calculated by the third calculation unit 60 at time t. ΣD m (t u ) is the representative value of luminance (fourth feature quantity) calculated by the fourth calculation unit 65. u These are multiple time points at which each second visible light image (monochromatic component image) was acquired, and time t is used for one second visible light image. u Each corresponds to a single element. u is the number of second visible light images (monochromatic component images), which is the number of frames acquired per unit time.

[0060] The third determination unit 73 performs a third determination for each region to determine whether the calculated feature change amount V2 is greater than or equal to a predetermined third threshold. The third determination unit 73 determines that there is a possibility of collapse if the feature change amount V2 of one or more regions among the calculated feature change amounts V2 for each region is greater than or equal to the third threshold, and determines that there is no possibility of collapse if the feature change amount V2 for all regions is less than the third threshold. In this embodiment, the third determination unit 73 outputs a flag (i=1) indicating that there is a collapse if the feature change amount V2 is greater than or equal to the third threshold. On the other hand, the third determination unit 73 outputs a flag (i=0) indicating that there is no collapse if the feature change amount V2 is less than the third threshold. Hereinafter, the flag (i=1 or 0) related to the collapse determination output by the third determination unit 73 will be referred to as the "third collapse determination flag". In other words, the third collapse determination flag indicates the result of the third determination and shows a value of 1 or 0. The third threshold value is pre-stored in the memory unit 90. The third threshold value may be different for each region. The third determination unit 73 performs the third determination by referencing the third threshold value stored in the memory unit 90 in a timely manner. The third determination unit 73 sends the third collapse determination flag, which is the result of the third determination, to the first collapse determination unit 75.

[0061] Here, Figure 12 shows the time-series changes in the feature change amount V2 of each region (left region 43l, central region 43c, right region 43r) during a specific time period, and the time-series changes in the third collapse judgment flag corresponding to the feature change amount V2 during the same time period. The graph shown in Figure 12 is the result obtained by the inventors' analysis. In (d) of Figure 12, the time when the operator of the incineration equipment 100 determined that there was a collapse of waste Fg by visually inspecting the combustion chamber R is indicated by a circle (〇). Also, the time corresponding to the interval between the vertically extending time scale lines (dotted lines) in the graph shown in Figure 12 is the same for all intervals. From the results shown in Figure 12, it can be seen that the time when the operator of the incineration equipment 100 determined that there was a collapse and the time when the third collapse judgment flag was raised (the time when i=1) roughly coincide.

[0062] (1st collapse judgment section) The first collapse determination unit 75 determines the scale of the collapse based on the result of the first determination received from the first determination unit 71, the result of the second determination received from the second determination unit 72, and the result of the third determination received from the third determination unit 73. In this embodiment, the first collapse determination unit 75 determines whether there is a second-scale collapse, which is larger in scale than the first-scale collapse, based on the first collapse determination flag, the second collapse determination flag, and the third collapse determination flag. Specifically, the first collapse determination unit 75 determines that there is a second-scale collapse if the sum of the values ​​of the first collapse determination flag, the second collapse determination flag, and the third collapse determination flag is equal to or greater than the determination threshold. In other words, the first collapse determination unit 75 detects a second-scale collapse. When the first collapse determination unit 75 detects a second-scale collapse, it outputs a flag indicating that a collapse has occurred. On the other hand, the first collapse determination unit 75 determines that there is no second-scale collapse if the sum of the values ​​of the first collapse determination flag, the second collapse determination flag, and the third collapse determination flag is less than the determination threshold. In other words, the first collapse determination unit 75 detects that there is no second-scale collapse. When the first collapse determination unit 75 detects that there is no second-scale collapse, it outputs a flag indicating that there is no second-scale collapse. Hereinafter, the flag output by the first collapse determination unit 75 will be referred to as the "second-scale collapse detection flag". The determination threshold is stored in advance in the storage unit 90. In this embodiment, the determination threshold is an integer, for example, 2. In other words, in this embodiment, the first collapse determination unit 75 determines that there is a second-scale collapse if two-thirds or more of the determination results of the first determination, the second determination, and the third determination described above indicate the possibility of a collapse. The first collapse determination unit 75 determines whether or not a second-scale collapse has occurred by referring in a timely manner to the determination threshold stored in the memory unit 90. The first collapse determination unit 75 sends a second-scale collapse detection flag to the second collapse determination unit 76 and the control unit 80.

[0063] (Second collapse judgment section) The second collapse determination unit 76 determines whether or not there is a collapse of the first scale based on the change in brightness of multiple images received from the acquisition unit 40, when the first collapse determination unit 75 determines that there is no collapse of the second scale. In this embodiment, the second collapse determination unit 76 determines whether or not there is a collapse of the first scale based on the first collapse determination flag received from the first determination unit 71. Specifically, the second collapse determination unit 76 determines that there is a collapse of the first scale when the first collapse determination flag indicates that there is a collapse (i=1). In other words, the second collapse determination unit 76 detects a collapse of the first scale. When the second collapse determination unit 76 detects a collapse of the first scale, it outputs a flag indicating that there is a collapse. On the other hand, the second collapse determination unit 76 determines that there is no collapse when the first collapse determination flag indicates that there is no collapse (i=0). In other words, the second collapse determination unit 76 detects that there is no collapse. The second collapse detection unit 76 outputs a flag indicating no collapse when it detects a collapse of the first magnitude. Hereinafter, the flag output by the second collapse detection unit 76 will be referred to as the "first magnitude collapse detection flag." Also, below, when there is no distinction between the "first magnitude collapse detection flag" and the "second magnitude collapse detection flag" mentioned above, they will simply be referred to as the "collapse detection flag." The second collapse detection unit 76 sends the first collapse detection flag to the control unit 80.

[0064] (Configuration of the control unit) The control unit 80 controls a plurality of controlled devices S based on collapse detection flags received from the first collapse determination unit 75 and the second collapse determination unit 76 (see Figures 1 and 2). In this embodiment, when the control unit 80 receives a second-scale collapse detection flag indicating a collapse from the first collapse determination unit 75, it controls one or more of the plurality of controlled devices S, for example, to reduce the concentration of unburned gas in the combustion chamber R. On the other hand, when the control unit 80 receives a first-scale collapse detection flag from the second collapse determination unit 76, it controls one or more of the plurality of controlled devices S, for example, to operate at rated capacity. The control unit 80 transmits signals to each controlled device S, for example, signals indicating the increase or decrease in the movement speed of the extrusion arm 124, the movement speed of the grate 126, the rotation speed of the blower 138, the valve opening of the first flow control valve 140, and the valve opening of the second flow control valve 142. Furthermore, if the control unit 80 receives a first-scale collapse detection flag from the second collapse determination unit 76, it may control one or more of the multiple controlled devices S so that the concentration of unburned gas in the combustion chamber R is reduced to a smaller extent than when a second-scale collapse detection flag is received.

[0065] (Operation of information processing device) Next, an example of the operation of the information processing device 4 in this embodiment will be described with reference to Figure 13. However, the order of the processes described below is not limited to the following example and may be rearranged as appropriate.

[0066] The acquisition unit 40 acquires an infrared image from the imaging device 2 (step S1). The acquisition unit 40 also acquires a visible light image from the imaging device 2 (step S11). Following the processing in step S1, the first calculation unit 50 calculates a representative value of brightness based on the first infrared image acquired in step S1. The second calculation unit 55 also calculates a representative value of brightness based on the second infrared image acquired in step S1 (step S2). Next, the disturbance determination unit 74 performs disturbance determination based on the first infrared image acquired in step S1 (step S3). If the disturbance determination unit 74 determines that there is a disturbance (step S3: YES), the process returns to step S1. On the other hand, if the disturbance detection unit 74 determines that there is no disturbance (step S3: NO), the first determination unit 71 makes a first determination based on the representative value of luminance calculated in step S2 (step S4), and the second determination unit 72 makes a second determination based on the representative value of luminance calculated in step S2 (step S5).

[0067] Following the processing in step S11, the third calculation unit 60 calculates a representative value of luminance based on the first visible light image acquired in step S11. The fourth calculation unit 65 then calculates a representative value of luminance based on the second visible light image acquired in step S1 (step S12). Next, the third determination unit 73 performs a third determination based on the representative value of luminance calculated in step S12 (step S13).

[0068] Following the processing in steps S4, S5, and S13 above, the first collapse determination unit 75 determines whether or not a second-scale collapse has occurred based on the results of the determination in step S4, the determination in step S5, and the determination in step S6 (step S6). That is, in step S6, the first collapse determination unit 75 determines whether or not the sum of the first collapse determination flag, the second collapse determination flag, and the third collapse determination flag is equal to or greater than the determination threshold. If the first collapse determination unit 75 determines that a second-scale collapse has occurred (step S6: YES), it detects the second-scale collapse (step S7). When the processing in step S7 is completed, the process returns to step S1.

[0069] On the other hand, if the first collapse determination unit 75 determines that there is no collapse of the second magnitude (step S6: NO), the second collapse determination unit 76 determines whether or not there is a change in the representative value of brightness (step S8). That is, in step S8, the second collapse determination unit 76 determines whether or not there is a collapse of the first magnitude based on the first collapse determination flag. If the second collapse determination unit 76 determines that there is a collapse of the first magnitude (step S8: YES), it detects the collapse of the first magnitude (step S9). When the processing of step S9 is completed, the process returns to step S1. On the other hand, if the second collapse determination unit 76 determines that there is no collapse of the first magnitude (step S8: NO), it detects that there is no collapse (step S10). When the processing of step S10 is completed, the process returns to step S1.

[0070] The operation of the information processing device 4 described above is repeatedly performed during the operation phase of the incineration equipment 100.

[0071] (Effects / Actions) Complex phenomena occur within the combustion chamber R, making it difficult to properly detect the collapse of waste Fg. For example, one contributing factor is that ash 135 may be scattered unintentionally within the combustion chamber R, and this scattered ash 135 may briefly appear in the image.

[0072] In this embodiment, the presence or absence of waste Fg collapse is determined based on a representative value of brightness based on a first infrared image and a representative value of brightness based on a plurality of second infrared images taken within the time required for one collapse of waste Fg, from a plurality of time-series infrared images acquired before the first infrared image. Therefore, compared to, for example, determining the presence or absence of collapse based on the brightness of each image acquired at two different timings, the presence or absence of waste Fg collapse can be determined with higher accuracy. In other words, the detection accuracy regarding the collapse of waste Fg can be improved. As a result, the combustion state of waste Fg in the combustion chamber R can be accurately grasped.

[0073] Figure 14 shows an example of the time-series results of the collapse detection flags output by the first collapse determination unit 75 and the second collapse determination unit 76 according to this embodiment. That is, Figure 14 shows the final determination result (detection result) by the information processing device 4 in this embodiment. The graph shown in Figure 14 is the result obtained by the inventors' analysis. In Figure 14, the time when the operator of the incineration equipment 100 determined that there was a collapse of first-scale waste Fg by visually inspecting the combustion chamber R is indicated by a triangle (△), and the time when the collapse of second-scale waste Fg was determined is indicated by a circle (〇). Also, the time corresponding to the interval between the vertically extending time scale lines (dotted lines) in the graph shown in Figure 14 is the same for all intervals. From the results shown in Figure 14, it is possible that the information processing device 4 excessively detected the presence of a collapse of waste Fg (time T -18 and time T -17 Although some discrepancies can be observed, it has been found that the time when the operator of incineration facility 100 determined that a collapse had occurred almost coincides with the time when the collapse determination flag was raised.

[0074] <Second embodiment of the incineration facility> Next, a second embodiment of the incineration equipment 100 according to this disclosure will be described. In the second embodiment described below, components common to the first embodiment described above are denoted by the same reference numerals in the figures and their descriptions are omitted. In the second embodiment, the configuration of the second collapse determination unit 76 of the collapse detection unit 41 differs from the second collapse determination unit 76 described in the first embodiment described above.

[0075] In this embodiment, the second collapse determination unit 76 calculates the similarity between the image acquired by the acquisition unit 40 and one or more images previously acquired by the acquisition unit 40. If the calculated similarity does not meet predetermined conditions, it determines that there is no collapse. The following describes the determination of whether or not there is a collapse by the second collapse determination unit 76, using as an example the case where the second determination unit 72 calculates the similarity between the first infrared image at time t and multiple infrared images acquired in the past prior to the first infrared image.

[0076] As shown in Figure 15, the second collapse determination unit 76 binarizes (turns into 0 / 1) the luminance contained in a specific target region (b), which is a part of the infrared image (a) received from the acquisition unit 40, into data of 0 and 1. Hereinafter, the target region that the second collapse determination unit 76 targets for binarization will be referred to as the "fifth target region 46". The fifth target region 46 is, for example, a region in the upper half of the infrared image in which a part (most) of the debris Fg accumulated in the feeder 104 is captured. The second collapse determination unit 76 binarizes the luminance contained in the fifth target region 46 based, for example, on a predetermined luminance threshold. The luminance threshold is stored in advance in the storage unit 90. The second collapse determination unit 76 binarizes the fifth target region 46 by referring to the luminance threshold stored in the storage unit 90 in a timely manner. Hereinafter, the fifth target region 46 binarized by the second collapse determination unit 76 will be referred to as "binarized data 46'".

[0077] Figure 16 is a diagram conceptually illustrating the method for calculating the similarity score calculated by the second collapse determination unit 76. As shown in Figure 16, the second collapse determination unit 76 calculates the difference between the binarized data 46' based on the first infrared image at time t and the binarized data 46' based on each of the multiple infrared images acquired in the past prior to the first infrared image (X1, X2, ..., X shown in Figure 16). y-1 , X y The following are obtained: the difference is calculated and the average value of the multiple obtained differences (ΣX / y shown in Figure 16) is used as the similarity. The binarized data 46' based on each of the multiple infrared images referred to here means, for example, the binarized data 46' based on the infrared image obtained at time t-1, the binarized data 46' based on the infrared image obtained at time t-2, ..., the binarized data 46' based on the infrared image obtained at time t-(y-1), and the binarized data 46' based on the infrared image obtained at time ty. y is, for example, an integer, and a value of 2 or greater (such as 5) is adopted.

[0078] The second collapse determination unit 76 determines whether the calculated similarity satisfies predetermined conditions. In this embodiment, the second collapse determination unit 76 determines whether the similarity is equal to or greater than a predetermined similarity threshold. In this case, the second collapse determination unit 76 determines that the predetermined conditions are met if the similarity is equal to or greater than the similarity threshold, and that the predetermined conditions are not met if the similarity is less than the similarity threshold. The similarity threshold is stored in the storage unit 90 in advance. The second collapse determination unit 76 determines whether the calculated similarity satisfies predetermined conditions by referring to the similarity threshold stored in the storage unit 90 in a timely manner. If the similarity does not satisfy the predetermined conditions, the second collapse determination unit 76 determines that there is a collapse of the first scale. That is, the second collapse determination unit 76 detects a collapse of the first scale. When the second collapse determination unit 76 detects a collapse of the first scale, it outputs a first-scale collapse detection flag indicating the presence of a collapse. On the other hand, the second collapse determination unit 76 determines that there is no collapse if the similarity satisfies predetermined conditions. In other words, the second collapse determination unit 76 detects that there is no collapse. When the second collapse determination unit 76 detects a collapse of the first magnitude, it outputs a collapse detection flag indicating that there is no collapse.

[0079] Next, an example of the operation of the information processing device 4 will be explained with reference to Figure 17. However, the order of the processes described below is not limited to the example below and may be rearranged as appropriate. Also, explanations of parts that overlap with the operation of the information processing device 4 explained using Figure 13 will be omitted.

[0080] If the second collapse determination unit 76 determines that there is a change in the representative value of brightness (step S8: YES), it calculates the similarity and determines whether the calculated similarity satisfies predetermined conditions (step S20). On the other hand, if the second collapse determination unit 76 determines that there is no change in the representative value of brightness (step S8: NO), it detects that there is no collapse (step S10). If the similarity does not satisfy predetermined conditions (step S20: NO), the second collapse determination unit 76 detects a collapse of the first scale (step S9). On the other hand, if the similarity satisfies predetermined conditions (step S20: YES), the second collapse determination unit 76 executes the process in step S10.

[0081] Figure 18 shows an example of the time-series results of the collapse detection flags output by the first collapse determination unit 75 and the second collapse determination unit 76 described above. In other words, Figure 18 shows the final determination result (detection result) by the information processing device 4. The graph shown in Figure 18 is the result obtained by the inventors' analysis. In Figure 18, the time when the operator of the incineration equipment 100 determined that there was a collapse of first-scale waste Fg by visually inspecting the combustion chamber R is indicated by a triangle (△), and the time when the operator determined that there was a collapse of second-scale waste Fg is indicated by a circle (〇). Also, the time corresponding to the interval between the time scale lines (dotted lines) extending vertically in the graph shown in Figure 18 is the same for all intervals. In the results shown in Figure 18, compared with the results shown in Figure 14, there is a time (time T in Figure 14) when the information processing device 4 may have excessively detected the presence of a waste Fg collapse. -18 and time T -17 There is no interval between these two events. Furthermore, it is confirmed that the time when the operator of incineration equipment 100 determined that a collapse had occurred coincides with the time when the collapse determination flag was raised.

[0082] <Third embodiment of the incineration facility> Next, a third embodiment of the incineration equipment 100 according to this disclosure will be described. In the third embodiment described below, components common to the first embodiment described above are denoted by the same reference numerals in the figures and their descriptions are omitted. In the third embodiment, the configuration of the collapse detection unit 41 and the storage unit 90 differs from that of the first embodiment described above.

[0083] (Configuration of the collapse detection unit) As shown in Figure 19, the collapse detection unit 41 in this embodiment includes a first calculation unit 50, a second calculation unit 55, a fifth calculation unit 66, a sixth calculation unit 67, and a determination unit 70. The determination unit 70 includes a disturbance determination unit 74, a fourth determination unit 77, a fifth determination unit 78, a third collapse determination unit 79, and a fourth collapse determination unit 81.

[0084] (First Calculation Unit) As shown in Figure 20, in this embodiment, the first target region 42 that the first calculation unit 50 calculates is divided into 24 sub-regions of equal area, arranged in a 6x4 matrix. However, the first target region 42 is not limited to being divided equally into 24 sub-regions in a matrix.

[0085] The first calculation unit 50 calculates, for example, the average value of the brightness of each sub-region in the first target region 42 in the first infrared image. Note that the representative value calculated by the first calculation unit 50 is not limited to the average value of the brightness of each sub-region, but may be a statistical quantity such as the median.

[0086] (Fifth Calculation Department) If the disturbance determination unit 74 determines that there is no disturbance, the fifth calculation unit 66 calculates a fifth feature based on the representative brightness value (first feature) received from the first calculation unit 50 and the representative brightness value (second feature) received from the second calculation unit 55. In this embodiment, the fifth feature is the sum of the brightness changes obtained by time-difference from the average brightness of each of the 24 sub-regions in the first target region 42 in the first infrared image. Note that the sum calculated by the fifth calculation unit 66 is not limited to the sum of the brightness changes of each sub-region, but may be the sum of statistical quantities such as the median. The fifth calculation unit 66 inputs the sum of the brightness changes of each sub-region in the first target region 42 in the first infrared image to the third collapse determination unit 79.

[0087] (6th Calculation Department) The sixth calculation unit 67 performs unsupervised learning using the autoencoder algorithm and calculates the sixth feature using the dimensionality-reduced information placed in the hidden layer of the trained autoencoder. In this embodiment, the autoencoder 96 is pre-stored in the memory unit 90 (see Figure 19). The sixth feature is infrared image information in which a series of infrared images acquired by the imaging device 2 are encoded by the autoencoder 96, thereby reducing the number of dimensions. The imaging device 2 acquires one frame of infrared image every 0.1 seconds. The sixth calculation unit 67 inputs a series of infrared images acquired by the imaging device 2 at 0.1-second intervals to the autoencoder 96. In the autoencoder 96, the series of infrared images are compressed (encoded) as they flow from the input layer to the hidden layer, reducing the number of dimensions to 512. Subsequently, the dimensionality-reduced infrared images are restored (decoded) to the original information as they flow from the hidden layer to the output layer. If the infrared image placed in the input layer can be reconstructed from the infrared image flowing from the hidden layer to the output layer, then the training data for that infrared image does not contain abnormal data such as disturbance images, and it is considered that correct training has been performed. When correct training is performed, the infrared image information with reduced dimensionality has noise in the image reduced. Note that the interval for acquiring infrared images by the imaging device 2 is not limited to 0.1 seconds. Also, the dimensionality reduction performed in the autoencoder 96 is not limited to 512 dimensions.

[0088] As described later, when the third collapse determination unit 79 determines that there is a collapse inside the furnace, the sixth calculation unit 67 packages 40 consecutive frames of noise-removed infrared image information, taken from multiple consecutive infrared image information at 0.1-second intervals, which have had their dimensionality reduced and noise removed from the images, and which are acquired over a total of 4 seconds (2 seconds before and 2 seconds after the point in time when a human judged that the garbage had collapsed), and inputs this packaged infrared image information into the long short-term memory (LSTM) network included in the fourth collapse determination unit 81. The long short-term memory network is a type of RNN that performs learning and prediction (regression / classification) of time series data, and when the packaged infrared image information is input into the long short-term memory network, the fourth collapse determination unit 81 performs a determination using a trained model that has been trained to output a determination result regarding the possibility that a collapse has occurred. Hereinafter, the trained model used for determination by the fourth collapse determination unit 81 will be referred to as the "trained model for collapse determination 97". The trained model 97 for collapse detection is pre-stored in the memory unit 90 (see Figure 19). Note that the number of information frames packaged is not limited to 40. The number of frames can be increased or decreased as appropriate by observing how long it takes for a series of events to progress, from when the waste peels off the waste surface inside the furnace, when the waste settles on the hearth, when the waste is scattered inside the furnace, and when the scattered waste falls back down, taking into account the characteristics of the incinerator. Furthermore, since there is uncertainty in the human judgment that a waste collapse has occurred, the infrared image information at the time when a human judges that a waste collapse has occurred, and the following 5 frames, may be used as training data for collapse occurrence.

[0089] (4th judgment part) The fourth determination unit 77 makes a determination regarding collapse based on one or more infrared images received from the acquisition unit 40. In this embodiment, the fourth determination unit 77 makes a determination regarding collapse using one first infrared image (the most recent infrared image) from the images received from the acquisition unit 40 that is the target of calculation by the first calculation unit 50. The fourth determination unit 77 may also use one image different from the first infrared image from the images received from the acquisition unit 40 when making a collapse determination. Furthermore, the fourth determination unit 77 may use multiple images when making a collapse determination, and the first infrared image may be included among the multiple images.

[0090] The fourth determination unit 77, upon receiving the first infrared image, performs a determination using a trained model that has been trained to output a determination result regarding the possibility of collapse occurring. Hereinafter, the determination made by the fourth determination unit 77 will be referred to as the "fourth determination," and the trained model used by the fourth determination unit 77 for the fourth determination will be referred to as the "fourth determination trained model 94." The fourth determination trained model 94 is pre-stored in the memory unit 90 (see Figure 19). The fourth determination unit 77 inputs the first infrared image received from the acquisition unit 40 into the fourth determination trained model 94 stored in the memory unit 90, and obtains the output determination result as the fourth determination. The fourth determination trained model 94 is, for example, a deep learning model (supervised learning model) such as a convolutional neural network (CNN). The fourth pre-trained model 94 for judgment is generated (trained) by repeatedly performing a learning step in which an infrared image captured by the imaging device 2 is input, and the presence or absence of collapse in the infrared image (ground truth data judged as correct by a human) is taught. Note that the fourth pre-trained model 94 for judgment may use a recurrent neural network (RNN) or the like instead of a CNN.

[0091] Figure 21 shows examples of infrared images captured by the imaging device 2, including images (a,d) showing debris Fg being scattered during a collapse (collapse present), images (b,e) showing ash being scattered without the effects of a collapse (no collapse), and images (c,f, including water vapor accumulation) showing other conditions (no collapse). In this embodiment, when the infrared image of a target region, which is a part of the first infrared image, is input to the fourth determination unit 77, it classifies whether or not a collapse has occurred in the infrared image captured by the imaging device 2 based on the three types of image examples described above. The fourth determination unit 77 is pre-generated by being taught whether or not a collapse has occurred in the infrared image (ground truth data). When a new infrared image is input to the fourth determination unit 94 after training, it outputs a numerical value related to the possibility of a collapse occurring in the infrared image as a "determination score".

[0092] The fourth determination unit 77 determines whether the determination score obtained from the fourth determination trained model 94 is equal to or greater than a predetermined fourth threshold (fourth determination). The fourth determination unit 77 determines that there is a possibility of collapse if the obtained determination score is equal to or greater than the fourth threshold, and determines that there is no possibility of collapse if the determination score is less than the fourth threshold. In this embodiment, the fourth determination unit 77 outputs a flag (i=1) indicating that there is a collapse if the determination score is equal to or greater than the fourth threshold. On the other hand, the fourth determination unit 77 outputs a flag (i=0) indicating that there is no collapse if the determination score is less than the fourth threshold. Hereinafter, the flag (i=1 or 0) related to the collapse determination output by the fourth determination unit 77 will be referred to as the "fourth collapse determination flag". In other words, the fourth collapse determination flag indicates the result of the fourth determination and shows a value of 1 or 0. The fourth threshold is stored in advance in the storage unit 90. The fourth determination unit 77 performs a fourth determination by referring to the fourth threshold value stored in the memory unit 90 in a timely manner. The fourth determination unit 77 inputs the fourth collapse determination flag, which is the result of the fourth determination, to the third collapse determination unit 79.

[0093] (5th judgment part) The fifth determination unit 78 makes a determination regarding collapse based on the visible light image received from the acquisition unit 40. In this embodiment, the fifth determination unit 78 makes a determination regarding collapse using the first visible light image (the most recent visible light image) from the images received from the acquisition unit 40. The fifth determination unit 78 may also use one image different from the first visible light image from the images received from the acquisition unit 40 when making a collapse determination. Furthermore, the fifth determination unit 78 may use multiple images when making a collapse determination, and the first visible light image may be included among the multiple images.

[0094] The fifth determination unit 78, upon receiving the first visible light image, performs a determination using a trained model that has been trained to output a determination result regarding the possibility of collapse occurring. Hereinafter, the determination made by the fifth determination unit 78 will be referred to as the "fifth determination," and the trained model used by the fifth determination unit 78 for the fifth determination will be referred to as the "fifth determination trained model 95." The fifth determination trained model 95 is pre-stored in the memory unit 90 (see Figure 19). The fifth determination unit 78 inputs the first visible light image received from the acquisition unit 40 into the fifth determination trained model 95 stored in the memory unit 90, and obtains the output determination result as the fifth determination. The fifth determination trained model 95 is, for example, a deep learning model (supervised learning model) such as a convolutional neural network (CNN). The fifth pre-trained model 95 for judgment is generated (trained) by repeatedly performing a learning step in which a visible light image captured by the imaging device 2 is input, and the presence or absence of collapse in the visible light image (ground truth data judged as correct by a human) is taught. Note that instead of a CNN, a recurrent neural network (RNN) or the like may be used for the fifth pre-trained model 95 for judgment.

[0095] The fifth determination unit 78 classifies the first visible light image based on an example image showing debris covering flames during a large-scale collapse, and an example image showing a different state (no collapse). Specifically, a visible light image of a target region, which is a part of the first visible light image, is input to the fifth determination trained model 95, and the presence or absence of a collapse in the first visible light image is classified based on the two example images mentioned above. The fifth determination trained model 95 is pre-generated by being taught the presence or absence of a collapse (ground truth data) in the visible light image. Once the fifth determination trained model 95 has finished training, when a new visible light image is input, it outputs a numerical value related to the possibility of a collapse occurring in the visible light image as a "determination score".

[0096] The fifth determination unit 78 determines whether the determination score obtained from the fifth determination trained model 95 is equal to or greater than a predetermined fifth threshold (fifth determination). The fifth determination unit 78 determines that there is a possibility of collapse if the obtained determination score is equal to or greater than the fifth threshold, and determines that there is no possibility of collapse if the determination score is less than the fifth threshold. In this embodiment, the fifth determination unit 78 outputs a flag (i=1) indicating that there is a collapse if the determination score is equal to or greater than the fifth threshold. On the other hand, the fifth determination unit 78 outputs a flag (i=0) indicating that there is no collapse if the determination score is less than the fifth threshold. Hereinafter, the flag (i=1 or 0) related to the collapse determination output by the fifth determination unit 78 will be referred to as the "fifth collapse determination flag". The fifth threshold is stored in advance in the storage unit 90. The fifth determination unit 78 performs the fifth determination by referring to the fifth threshold stored in the storage unit 90 in a timely manner. The fifth determination unit 78 inputs the fifth collapse determination flag, which is the result of the fifth determination, to the third collapse determination unit 79.

[0097] (3rd collapse judgment section) The third collapse determination unit 79 determines whether or not the waste inside the furnace has collapsed based on the sum of the brightness changes of each of the 24 sub-regions in the first target region 42 in the first infrared image calculated by the fifth calculation unit 66, the fourth collapse determination flag which is the result of the fourth determination performed by the fourth determination unit 77, and the fifth collapse determination flag which is the result of the fifth determination performed by the fifth determination unit 78. Specifically, there are three patterns that combine the sum of the brightness changes in the first target region 42, the fourth collapse determination flag, and the fifth collapse determination flag. As shown in Figure 22, the first pattern is when the sum of the brightness changes in each of the first target region 42 is 22 or more or -26 or less, and the fourth collapse determination flag input from the fourth determination unit 77 is the flag indicating collapse (i=1). The second pattern is when the sum of the brightness changes of each sub-region in the first target region 42 is 20 or more or -13 or less, and the fifth collapse determination flag input from the fifth determination unit 78 is the flag indicating collapse (i=1). The third pattern is when the sum of the brightness changes of each sub-region in the first target region 42 is 10 or more or -13 or less, and both the fourth collapse determination flag input from the fourth determination unit 77 and the fifth collapse determination flag input from the fifth determination unit 78 are the flags indicating collapse (i=1).

[0098] The third collapse determination unit 79 determines that there is a collapse in the furnace if the sum of the brightness changes of each of the 24 sub-regions in the first target region 42 in the first infrared image calculated by the fifth calculation unit 66, the fourth collapse determination flag input from the fourth determination unit 77, and the fifth collapse determination flag input from the fifth determination unit 78 all fall into one of the first to third patterns, and determines that there is no collapse in the furnace if none of the first to third patterns fall into one of the first to third patterns. The threshold value for the sum of the brightness changes of the infrared images included in the first to third patterns may be changed depending on the type of image selected. That is, if the deep learning determination result of the visible light image indicates that there is a collapse, it can be assumed that a large-scale collapse of waste has occurred in the furnace, so the range of the threshold value for the sum of the brightness changes of the infrared images included in the second pattern may be smaller than the range of the threshold value for the sum of the brightness changes of the infrared images included in the first pattern. Furthermore, if both the deep learning-based determination results for infrared images and the deep learning-based determination results for visible light images indicate the presence of collapse, it can be considered that there is a high probability that waste collapse is occurring inside the furnace. Therefore, the threshold range for the sum of the brightness changes in infrared images included in the third pattern may be smaller than the threshold range for the sum of the brightness changes in infrared images included in the first and second patterns.

[0099] (4th collapse judgment section) The fourth collapse determination unit 81 includes a long-short-term memory network, which is a type of RNN that performs learning and prediction (regression / classification) of time-series data. When the fourth collapse determination unit 81 receives a package of infrared image information for 40 consecutive frames from which noise has been removed from the image from the sixth calculation unit 67 into the long-short-term memory network, it uses a trained model 97 for collapse determination, which is trained to output a determination result regarding the possibility of a collapse occurring, to determine whether or not the waste inside the furnace has collapsed. The fourth collapse determination unit 81 inputs the package of infrared image information for 40 consecutive frames from which noise has been removed from the image, which has been input from the sixth calculation unit 67, into the trained model 97 for collapse determination stored in the memory unit 90, thereby obtaining the determination result from the long-short-term memory network.

[0100] When the fourth collapse determination unit 81 receives a package of 40 consecutive frames of infrared image information from the sixth calculation unit 67 into its long- and short-term memory network, it trains the long- and short-term memory network with the package and determines the state inside the furnace from the state transitions of the 40 frames of infrared images. In other words, when 40 consecutive frames of infrared image information are input into the collapse determination trained model 97, the presence or absence of collapse (ground truth data) for the infrared image information is taught based on the state inside the furnace over time, thereby generating the collapse determination trained model 97 in advance. Once the training is complete, when the collapse determination trained model 97 receives a package of infrared image information, it outputs a numerical value as a "determination score" that indicates the possibility of collapse occurring for the 40 frames of infrared image information contained in the package.

[0101] The fourth collapse determination unit 81 determines whether the determination score obtained from the trained model 97 for collapse determination is equal to or greater than a predetermined sixth threshold (sixth determination). The fourth collapse determination unit 81 determines that there is a possibility of collapse if the obtained determination score is equal to or greater than the sixth threshold, and determines that there is no possibility of collapse if the determination score is less than the sixth threshold. In this embodiment, the fourth collapse determination unit 81 outputs a flag (i=1) indicating the presence of a collapse when the determination score is equal to or greater than the sixth threshold. On the other hand, the fourth collapse determination unit 81 outputs a flag (i=0) indicating the absence of a collapse when the determination score is less than the sixth threshold. The determination unit 70 ultimately determines that there is a collapse in the furnace when the fourth collapse determination unit 81 outputs the flag (i=1), and ultimately determines that there is no collapse in the furnace when the fourth collapse determination unit 81 outputs the flag (i=0).

[0102] (Configuration of the control unit) The control unit 80 controls multiple controlled devices S based on the collapse detection flag received from the fourth collapse determination unit 81 (see Figure 19). In this embodiment, when the control unit 80 receives a collapse detection flag indicating a collapse from the fourth collapse determination unit 81, it controls one or more of the multiple controlled devices S so that they operate at their rated capacity. For example, the control unit 80 transmits signals to each controlled device S indicating increases or decreases in the movement speed of the extrusion arm 124, the movement speed of the grate 126, the rotation speed of the blower 138, the valve opening of the first flow control valve 140, and the valve opening of the second flow control valve 142. The control unit 80 may also control one or more of the multiple controlled devices S so as to reduce the concentration of unburned gas in the combustion chamber R.

[0103] (Operation of information processing device) Next, an example of the operation of the information processing device 4 in this embodiment will be described with reference to Figure 23. However, the order of the processes described below is not limited to the following example and may be rearranged as appropriate.

[0104] The acquisition unit 40 acquires an infrared image from the imaging device 2 (step S1). The acquisition unit 40 also acquires a visible light image from the imaging device 2 (step S11). Following the processing in step S1, the first calculation unit 50 calculates a representative value of brightness based on the first infrared image acquired in step S1. The second calculation unit 55 also calculates a representative value of brightness based on the second infrared image acquired in step S1 (step S2). Next, the disturbance determination unit 74 performs disturbance determination based on the first infrared image acquired in step S1 (step S3). If the disturbance determination unit 74 determines that there is a disturbance (step S3: YES), the process returns to step S1. On the other hand, if the disturbance detection unit 74 determines that there is no disturbance (step S3: NO), the fifth calculation unit 66 calculates a fifth feature based on the representative brightness value (first feature) received from the first calculation unit 50 and the representative brightness value (second feature) received from the second calculation unit 55 (step S21). The fourth determination unit 77 makes a determination regarding collapse based on one or more infrared images received from the acquisition unit 40 (step S22). The fifth determination unit 78 also makes a determination regarding collapse based on the visible light image received from the acquisition unit 40 (step S23).

[0105] Following the processing in steps S21, S22, and S23 above, the third collapse determination unit 79 determines whether or not the waste in the furnace has collapsed based on the sum of the brightness changes of each sub-region in the first target region 42 in the first infrared image calculated in step S21, the result of the determination in step S22, and the result of the determination in step S23 (step S24). If the third collapse determination unit 79 determines that a collapse has occurred (step S24: i=1), the sixth calculation unit 67 packages 40 frames of noise-removed infrared image information, which are continuous at 0.1-second intervals and have had their dimensionality reduced and noise removed from the image, for a total of 4 seconds, starting from the point in time when a human judged that the waste had collapsed, by 2 seconds before and after, and inputs this packaged infrared image information into the long- and short-term memory network included in the fourth collapse determination unit 81 (step S25). If the third collapse determination unit 79 determines that there is no collapse (step S24: i=0), it is detected that there is no collapse of waste inside the furnace (step 28). When the process in step S28 is completed, the process returns to step S1.

[0106] Following the processing in step S25, the fourth collapse determination unit 81 receives a package of infrared image information for 40 consecutive frames from which noise has been removed from the image into the long- and short-term memory network from the sixth calculation unit 67, and uses the trained model 97 for collapse determination to determine whether or not the waste inside the furnace has collapsed (step S26). If the fourth collapse determination unit 81 determines that there is a collapse (step S26: i=1), it is detected that there is a collapse of waste inside the furnace (step S27). When the processing in step S27 is completed, the process returns to step S1. If the fourth collapse determination unit 81 determines that there is no collapse (step S26: i=0), it is detected that there is no collapse (step 28). When the processing in step S28 is completed, the process returns to step S1.

[0107] The operation of the information processing device 4 described above is repeatedly performed during the operation phase of the incineration equipment 100.

[0108] (Effects / Actions) Even if an image captures a change in the furnace environment that does not necessarily indicate a collapse, such as black ash moving from bottom to top, if only a single frame is used for judgment without considering the passage of time, the movement of ash from bottom to top cannot be grasped, and it may be recognized as one element indicating the occurrence of a collapse, potentially leading to false positives. According to this embodiment, the element of time progression is considered when grasping the state inside the furnace. Information from 40 consecutive infrared images acquired by the acquisition unit 40 is input into a long-short-term memory network, which is one of the trained models with a recurrent structure, for training, and the state inside the furnace is determined from the state progression of 40 consecutive infrared images. As a result, changes in the furnace environment that do not necessarily indicate a collapse can be correctly grasped, and false positives can be suppressed.

[0109] (Other embodiments) Although embodiments of this disclosure have been described in detail above with reference to the drawings, the specific configurations are not limited to those of each embodiment, and additions, omissions, substitutions, and other modifications to the configurations are possible without departing from the gist of this disclosure.

[0110] Furthermore, the first collapse determination unit 75 described above may determine whether or not there is a second-scale collapse, which is larger in scale than the first-scale collapse, based on the results of both the first and second determinations. In this case, the first collapse determination unit 75 only needs to determine that there is a second-scale collapse if the sum of the values ​​of the first collapse determination flag and the second collapse determination flag is equal to or greater than the determination threshold.

[0111] Furthermore, the collapse detection unit 41 may determine whether a collapse has occurred based on a first input element, which is a first infrared image acquired by the acquisition unit 40, and a second input element, which is two or more second infrared images captured within the time required for one collapse of waste Fg, among multiple time-series images acquired by the acquisition unit 40 before the first infrared image. In this case, the collapse detection unit 41 makes the determination using a trained model 93 (see Figure 2; hereinafter referred to as the "collapse detection trained model") that has been trained to output a determination result regarding the possibility of a collapse occurring when the first and second input elements are input. The collapse detection trained model 93 is pre-stored in the memory unit 90. The collapse detection unit 41 obtains the determination result output by inputting the first and second input elements received from the acquisition unit 40 to the collapse detection trained model 93 stored in the memory unit 90. The collapse detection trained model 93 is, for example, a deep learning model such as a convolutional neural network. The trained model 93 for collapse detection is generated by repeatedly performing a training step in which infrared images, which are the first and second input elements captured by the imaging device 2, are input, and the presence or absence of a collapse in the infrared images is taught.

[0112] Furthermore, the images received by the first calculation unit 50 and the second calculation unit 55 from the acquisition unit 40 are not limited to infrared images. Also, the images received by the third calculation unit 60 and the fourth calculation unit 65 from the acquisition unit 40 are not limited to visible light images.

[0113] Furthermore, although the incineration equipment 100 is a stoker-type waste incinerator in this embodiment, it is not limited to a stoker-type waste incinerator. The incineration equipment 100 may be, for example, a kiln-stoker furnace, a biomass fluidized bed boiler, or a sludge incinerator. Therefore, the collapse detection system 1 described above may be a system that can be applied to these incineration equipment such as kiln-stoker furnaces, biomass fluidized bed boilers, and sludge incinerators.

[0114] Figure 24 is a hardware configuration diagram showing the configuration of the computer 1100 according to this embodiment. The computer 1100 includes a processor 1110, main memory 1120, storage 1130, and interface 1140.

[0115] The information processing device 4 described above is implemented in one or more computers 1100. The operation of each processing unit described above is stored in storage 1130 in the form of a program. The processor 1110 reads the program from storage 1130, loads it into main memory 1120, and executes the above processing according to the program. The processor 1110 also reserves a storage area in main memory 1120 corresponding to the storage unit 90 described above, according to the program. The program may be for realizing a part of the functions to be performed by the computer 1100. For example, the program may perform its functions in combination with other programs already stored in storage 1130, or in combination with other programs implemented in other devices. In addition to the above configuration, or in place of the above configuration, the computer 1100 may be equipped with a custom LSI (Large Scale Integrated Circuit) such as a PLD (Programmable Logic Device). 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, some or all of the functions implemented by processor 1110 may be implemented by the integrated circuit.

[0116] Examples of storage 1130 include magnetic disks, magneto-optical disks, and semiconductor memory. Storage 1130 may be an internal medium directly connected to the bus of computer 1100, or an external medium connected to computer 1100 via interface 1140 or a communication line. Furthermore, if this program is distributed to computer 1100 via a communication line, computer 1100 that receives the program may load it into main memory 1120 and execute the above processing. In the above embodiment, storage 1130 is a tangible storage medium that is not temporary. Furthermore, the program may be for the purpose of realizing some of the functions described above. Moreover, the program may be a so-called differential file (differential program) that realizes the above functions in combination with other programs already stored in storage 1130.

[0117] <Note> The collapse detection system and collapse detection method described in each embodiment can be understood, for example, as follows.

[0118] (1) The collapse detection system 1 according to the first embodiment includes: an acquisition unit 40 that acquires images of the waste to be incinerated (garbage Fg) that is accumulated in the feeder 104 of the incineration equipment 100 and pushed toward the combustion chamber R at a first predetermined period; a first calculation unit 50 that calculates a representative value of brightness based on the first image (first infrared image) acquired by the acquisition unit 40; a second calculation unit 55 that calculates a representative value of brightness based on two or more images (second infrared images) taken within the time required for one collapse of the waste to be incinerated, from among a plurality of time-series images acquired by the acquisition unit 40 before the first image; and a determination unit 70 that makes a determination regarding the collapse based on the representative value of brightness calculated by the first calculation unit 50 and the representative value of brightness calculated by the second calculation unit 55.

[0119] This allows for more accurate determination of whether or not waste Fg has collapsed compared to, for example, determining the presence or absence of collapse based on the brightness of each image acquired at two different timings. As a result, the combustion state of waste Fg within the combustion chamber R can be accurately understood.

[0120] (2) The collapse detection system 1 according to the second embodiment is the collapse detection system 1 of (1), wherein the second calculation unit 55 may calculate the average or median value of the brightness for the two or more images as a representative value of brightness based on the two or more images.

[0121] This allows us to represent the state of the first infrared image and the state of the second infrared image using specific statistical quantities.

[0122] (3) The collapse detection system 1 according to the third embodiment is the collapse detection system 1 of (1) or (2), wherein the two or more images are images taken at intervals of a first time from each other, and the two or more images may be images taken before the first image over a second time period which is at least twice the first time.

[0123] This allows the above effects to be realized with more specific settings.

[0124] (4) The collapse detection system 1 according to the fourth embodiment is any one of the collapse detection systems 1 from (1) to (3), wherein the determination unit 70 may include a first determination unit 71 that makes a first determination regarding the collapse based on a representative value of brightness calculated by the first calculation unit 50 and a representative value of brightness calculated by the second calculation unit 55, a second determination unit 72 that makes a second determination regarding the collapse based on one or more images acquired by the acquisition unit 40, and a first collapse determination unit 75 that determines whether or not there is a second-scale collapse which is larger in scale than a first-scale collapse, based on the result of the first determination and the result of the second determination.

[0125] This allows for the classification of the scale of collapse of the incinerated material. Therefore, the combustion state of the incinerated material within the combustion chamber R can be understood more accurately.

[0126] (5) The collapse detection system 1 according to the fifth embodiment is the collapse detection system 1 of (4), wherein the second determination unit 72 may perform the second determination using a trained model (second determination trained model 91) that has been trained to output a determination result regarding the possibility of the collapse occurring when an image acquired by the acquisition unit 40 is input.

[0127] This allows for a more accurate classification of the scale of collapse of materials being incinerated.

[0128] (6) The collapse detection system 1 according to the sixth embodiment is the collapse detection system 1 of (4) or (5), wherein the acquisition unit 40 acquires an infrared image which is the image at the first predetermined period and acquires a visible light image taken inside the combustion chamber R at the second predetermined period, the second determination unit 72 makes the second determination based on one or more infrared images acquired by the acquisition unit 40, the determination unit 70 further includes a third determination unit 73 which makes the third determination regarding the collapse based on one or more visible light images acquired by the acquisition unit 40, and the first collapse determination unit 75 may determine whether or not there is a collapse of the second scale based on the result of the first determination, the result of the second determination, and the result of the third determination.

[0129] This allows for a more accurate classification of the scale of collapse of materials being incinerated.

[0130] (7) The collapse detection system 1 according to the seventh embodiment is any one of the collapse detection systems 1 from (4) to (6), wherein the determination unit 70 includes a disturbance determination unit 74 that determines the presence or absence of disturbances based on the brightness feature quantities of the image acquired by the acquisition unit 40, and the determination by the first collapse determination unit 75 may be performed when the disturbance determination unit 74 determines that there are no disturbances.

[0131] This makes it possible to suppress the detection of collapse even when, for example, the material being incinerated has not collapsed.

[0132] (8) The eighth collapse detection system 1 is any one of the collapse detection systems 1 from (4) to (7), wherein the determination unit 70 further includes a second collapse determination unit 76 that determines whether or not there is a collapse of the second scale based on changes in brightness of a plurality of images acquired by the acquisition unit 40 when the first collapse determination unit 75 determines that there is no collapse of the second scale.

[0133] This allows for a more accurate classification of the scale of collapse of materials being incinerated.

[0134] (9) The collapse detection system 1 according to the ninth aspect is the collapse detection system 1 of (8), wherein the second collapse determination unit 76 calculates the similarity between the image acquired by the acquisition unit 40 and one or more images previously acquired by the acquisition unit 40, and if the similarity does not satisfy predetermined conditions, it may determine that there is no collapse.

[0135] This makes it possible to further suppress false positives, such as detecting a collapse when the material being incinerated has not actually collapsed.

[0136] (10) The collapse detection system 1 according to the tenth embodiment is the collapse detection system 1 of (1), wherein the determination unit 70 may determine the presence or absence of a collapse by performing a process that includes inputting feature quantities extracted based on one or more images acquired by the acquisition unit 40 into a trained model 97 having a recurrent structure.

[0137] This helps to suppress the occurrence of false positives.

[0138] (11) The collapse detection system 1 according to the eleventh embodiment is the collapse detection system 1 of (10), wherein the determination unit 70 may include a fourth determination unit 77 that performs a fourth determination by inputting feature quantities extracted based on one or more infrared images acquired by the acquisition unit 40 into a deep learning model, a fifth determination unit 78 that performs a fifth determination by inputting feature quantities extracted based on visible images acquired by the acquisition unit 40 into a deep learning model, and a third collapse determination unit 79 that performs a determination regarding the collapse based on feature quantities calculated based on representative values ​​of brightness calculated by the first calculation unit 50 and representative values ​​of brightness calculated by the second calculation unit 55, the result of the fourth determination, and the result of the fifth determination.

[0139] (12) The 12th embodiment of the collapse detection system 1 is the collapse detection system 1 of (10) or (11), wherein the determination unit 70 packages the feature quantities of a plurality of images acquired continuously at predetermined time intervals by the acquisition unit 40, and inputs the feature quantities of the packaged plurality of images into the trained model 97 having a recurrent structure.

[0140] (13) The collapse detection system 1 according to the 13th embodiment is the collapse detection system 1 of (12), wherein the dimensions of a plurality of images acquired continuously at predetermined time intervals by the acquisition unit 40 are reduced using an autoencoder 96, and the feature quantities of the plurality of images with reduced dimensions are packaged.

[0141] (14) The collapse detection method according to the 14th embodiment includes one or more computers 1100 acquiring images of the material to be incinerated that is accumulated in the feeder 104 of the incineration equipment 100 and pushed toward the combustion chamber R at a first predetermined period, calculating a representative value of brightness based on the acquired first image, calculating a representative value of brightness based on two or more images taken within the time required for one collapse of the material to be incinerated from a plurality of time-series images acquired before the first image, and making a judgment regarding the collapse based on the representative value of brightness based on the first image and the representative values ​​of brightness based on the two or more images.

[0142] (15) The collapse detection system 1 according to the 15th embodiment includes an acquisition unit 40 that acquires images of the material to be incinerated that is accumulated in the feeder 104 of the incineration equipment 100 and pushed toward the combustion chamber R at a first predetermined period, and a collapse detection unit 41 that makes a determination regarding the collapse based on a first input element which is a first image acquired by the acquisition unit 40, and a second input element which is two or more images taken within the time required for one collapse of the material to be incinerated, from among a plurality of time-series images acquired by the acquisition unit 40 before the first image. [Industrial applicability]

[0143] This disclosure relates to a system for detecting the collapse of materials to be incinerated in the combustion chamber of an incinerator. The collapse detection system of this disclosure can improve the accuracy of detecting the collapse of materials to be incinerated. [Explanation of Symbols]

[0144] 1… Collapse detection system 2…Imaging device 4…Information Processing Devices 5…Infrared camera 6…Visible light camera 8…Filter device 40…Acquisition part 41... Collapse detection unit 42…First Target Area 43…Second Target Area 44…Third Target Area 45, 45a, 45b, 45c… Fourth target area 46…Fifth Target Area 46'...Binarized data 42a…First area 42b…Second area 42c...Third area 42d…Fourth area 42e...5th area 42f…6th area 42g…7th area 42h…8th area 42i…9th area 43c…Central area 43l…left area 43r…Right area 50...First Calculation Unit 55...Second Calculation Unit 60... Third Calculation Department 65...Fourth Calculation Section 66... ​​Fifth Calculation Unit 67…6th Calculation Department 70…Judgment section 71...1st judgment section 72…Second judgment section 73...Third judgment section 74…Disturbance detection unit 75...1st collapse judgment part 76…Second collapse judgment part 77…Fourth judgment part 78…5th judgment part 79…Third collapse judgment part 80... Control Unit 81…4th collapse judgment part 90...Storage section 91... Pre-trained model for second decision 92... Pre-trained models for disturbance detection 93... Pre-trained model for collapse detection 94... Pre-trained model for the 4th decision 95...Pre-trained model for the 5th decision 96... Autoencoder 97... Pre-trained model for collapse detection 100... Incineration equipment 102... Hoppa 104... Feeder 108... Furnace body 110... Extruder 112... Air supply device 114… Heat recovery boiler 116... Defrosting tower 118... Dust collection device 120... Chimney 121...Downstream end 122...Intake port 124... Extrusion arm 126... Fire Gates 128…Dry area 130... Combustion range 131... Flame 132...Post-combustion region 135...ash 136...Air supply pipe 138... Blower 140...First flow control valve 142... Second flow control valve 143... Exhaust gas 144...Flute 145...Harashiri 146... Gray Shot 1100... Computer 1110… Processor 1120... Main memory 1130...Storage 1140… Interface Fg... Garbage (material to be incinerated) Fr...Front R...combustion chamber S...Controlled device W1... Conveying direction

Claims

1. An acquisition unit that acquires images of the material to be incinerated, which is accumulated in the feeder of the incineration equipment and pushed towards the combustion chamber, at a first predetermined cycle, A first calculation unit calculates a representative value of brightness based on the first image acquired by the acquisition unit, A second calculation unit calculates a representative brightness value based on two or more images captured within the time required for one collapse of the incinerated material, among a plurality of time-series images acquired by the acquisition unit before the first image, A collapse detection system comprising: a determination unit that performs a determination regarding the collapse based on a representative value of luminance calculated by the first calculation unit and a representative value of luminance calculated by the second calculation unit.

2. The collapse detection system according to claim 1, wherein the second calculation unit calculates the average or median value of the brightness for the two or more images as a representative value of brightness based on the two or more images.

3. The two or more images mentioned above are images taken at intervals of one time from each other. The collapse detection system according to claim 1 or 2, wherein the two or more images are images taken before the first image over a second period of time that is at least twice the length of the first period.

4. The determination unit, A first determination unit performs a first determination regarding the collapse based on a representative value of luminance calculated by the first calculation unit and a representative value of luminance calculated by the second calculation unit, A second determination unit performs a second determination regarding the collapse based on one or more images acquired by the acquisition unit, A collapse detection system according to claim 1 or 2, comprising: a first collapse determination unit that determines, based on the result of the first determination and the result of the second determination, whether or not there is a second-scale collapse which is larger in scale than a first-scale collapse.

5. The collapse detection system according to claim 4, wherein the second determination unit performs the second determination using a trained model that has been trained to output a determination result regarding the possibility of the collapse occurring when an image acquired by the acquisition unit is input.

6. The acquisition unit acquires the infrared image, which is the image, at the first predetermined period, and acquires the visible light image of the inside of the combustion chamber at the second predetermined period. The second determination unit performs the second determination based on one or more infrared images acquired by the acquisition unit. The determination unit further includes a third determination unit that performs a third determination regarding the collapse based on one or more visible light images acquired by the acquisition unit, The collapse detection system according to claim 4, wherein the first collapse determination unit determines whether or not a collapse of the second magnitude has occurred based on the result of the first determination, the result of the second determination, and the result of the third determination.

7. The determination unit includes a disturbance determination unit that determines the presence or absence of disturbances based on the brightness feature quantities of the image acquired by the acquisition unit. The collapse detection system according to claim 4, wherein the determination by the first collapse determination unit is performed when the disturbance determination unit determines that there is no disturbance.

8. The collapse detection system according to claim 4, wherein the determination unit further includes a second collapse determination unit that determines whether or not there is a collapse of the second scale based on changes in brightness of a plurality of images acquired by the acquisition unit when the first collapse determination unit determines that there is no collapse of the second scale.

9. The collapse detection system according to claim 8, wherein the second collapse determination unit calculates the similarity between the image acquired by the acquisition unit and one or more images previously acquired by the acquisition unit, and determines that there is no collapse if the similarity does not meet predetermined conditions.

10. The collapse detection system according to claim 1, wherein the determination unit determines whether or not a collapse has occurred by performing a process that includes inputting feature quantities extracted based on one or more images acquired by the acquisition unit into a trained model having a recurrent structure.

11. The determination unit, A fourth determination unit performs a fourth determination by inputting the feature quantities extracted based on one or more infrared images acquired by the acquisition unit into a deep learning model. A fifth determination unit performs a fifth determination by inputting the features extracted based on the visible image acquired by the acquisition unit into a deep learning model, The collapse detection system according to claim 10, comprising: a third collapse determination unit that performs a determination regarding the collapse based on a characteristic quantity calculated based on a representative value of luminance calculated by the first calculation unit and a representative value of luminance calculated by the second calculation unit, the result of the fourth determination, and the result of the fifth determination.

12. The collapse detection system according to claim 10 or 11, wherein the determination unit packages the feature quantities of a plurality of images acquired continuously at predetermined time intervals by the acquisition unit, and inputs the packaged feature quantities of the plurality of images into the trained model having a recurrent structure.

13. The collapse detection system according to claim 12, comprising reducing the dimensions of a plurality of images acquired continuously at predetermined time intervals by the acquisition unit using an autoencoder, and packaging the feature quantities of the plurality of images with reduced dimensions.

14. One or more computers, Images of the material to be incinerated, which is accumulated in the feeder of the incineration equipment and pushed towards the combustion chamber, are acquired at a first predetermined cycle. A representative value of brightness is calculated based on the first image acquired. From among multiple time-series images acquired prior to the first image, a representative brightness value is calculated based on two or more images captured within the time required for one collapse of the material to be incinerated. A collapse detection method that includes making a determination regarding the collapse based on a representative value of brightness based on the first image and representative values ​​of brightness based on the two or more images.

15. An acquisition unit that acquires images of the material to be incinerated, which is accumulated in the feeder of the incineration equipment and pushed towards the combustion chamber, at a first predetermined cycle, A collapse detection system comprising: a first input element which is a first image acquired by the acquisition unit; and a second input element which is two or more images taken within the time required for one collapse of the incinerated material, from among a plurality of time-series images acquired by the acquisition unit before the first image; and a collapse detection unit which makes a determination regarding the collapse.

Images