Prediction device, prediction method, prediction program, facility control device, facility control method, and control program

Abstract

It is helpful to improve the operation of incineration facilities. A prediction device (1) is provided with: a movement data generation section (103) that generates movement data indicating a movement state of an incineration object in an incineration facility that feeds the incineration object in a hopper into an incinerator at a prescribed speed, from a plurality of images that are time-series images obtained by photographing the hopper from above; and a property prediction section (105) that predicts a property of the incineration object based on the generated movement data and the speed.

Description

Technical Field

[0001] This invention relates to a predictive device for predicting the properties of incinerated objects in an incinerator. Background Technology

[0002] In conventional waste incineration facilities, a funnel-shaped device called a hopper is installed. Waste stored in a storage device called a waste pit is grabbed by a crane or the like and fed into the hopper. Furthermore, the waste in the hopper is sequentially fed into the incinerator by a feeding device, starting with the first waste added. As a technology related to the hopper, Patent Document 1 is cited as an example. Patent Document 1 describes a method for detecting bridging based on periodically measured hopper levels. Existing technical documents Patent documents

[0003] Patent Document 1: Japanese Patent Application Publication No. 10-238735 Summary of the Invention (a) Technical problems to be solved

[0004] As described above, the waste fed into the hopper is sequentially fed into the incinerator by the feeding device for combustion. However, the combustion process varies depending on the characteristics of the waste. For example, when wet waste containing a high moisture content is fed in, the temperature inside the incinerator may decrease. On the other hand, when dry, easily combustible waste is fed in, the temperature inside the incinerator may increase. This phenomenon is not limited to waste and is common when incinerating any type of waste.

[0005] Therefore, it is considered that if the properties of the incinerator material can be understood before incineration, it would help improve the operation of the incineration facility. For example, the state inside the incinerator can be predicted based on the properties of the incinerator material, thereby preventing deterioration of the state. Furthermore, automatic control of the incinerator and automatic control of the feeding of the incinerator material into the hopper can also be implemented. However, existing technologies do not predict the properties of the incinerator material fed into the hopper. For example, Patent Document 1 mentioned above does not disclose any method for predicting the properties of the waste fed into the hopper.

[0006] One aspect of the present invention addresses the aforementioned problems, and its object is to provide a predictive device, etc., that helps improve the operation of incineration facilities. (II) Technical Solution

[0007] To address the aforementioned problems, a prediction apparatus according to one aspect of the present invention comprises: a movement data generation unit that generates movement data representing the movement state of the incineration object by feeding a hopper into an incineration facility of an incinerator at a specified speed, based on multiple time-series images obtained by taking pictures of the hopper from above; and a prediction unit that predicts the characteristics of the incineration object being incinerated based on the movement data generated by the movement data generation unit and the speed.

[0008] In addition, to solve the above problems, a facility control device according to one aspect of the present invention includes: a movement data generation unit that generates movement data representing the movement state of the incineration object by feeding the incineration object into the incineration facility of the incinerator at a specified speed, based on multiple time-series images obtained by taking pictures of the hopper from above; and an equipment control unit that performs at least one of the following controls based on the movement data generated by the movement data generation unit and the speed, namely: (1) control of feeding the incineration object into the hopper, (2) control of the speed, (3) control of the conveying speed of the incineration object in the incinerator, (4) combustion control of the incineration object in the incinerator, and (5) stirring control of the incineration object.

[0009] To address the aforementioned problems, one aspect of the present invention provides a prediction method executed by one or more information processing devices to predict the properties of an object to be incinerated, comprising the following steps: feeding the object to be incinerated into an incineration facility by feeding a hopper into an incinerator at a specified speed; generating movement data representing the movement state of the object to be incinerated based on multiple time-series images obtained by taking pictures of the hopper from above; and predicting the properties of the object to be incinerated based on the movement data generated in the above steps and the speed.

[0010] To address the aforementioned problems, a facility control method according to one aspect of the present invention is executed by one or more information processing devices and includes the following steps: in a combustion facility in which an incineration object is fed into an incinerator at a specified speed, motion data representing the movement state of the incineration object is generated based on multiple time-series images obtained by taking pictures of the hopper from above; and based on the motion data generated in the step and the speed, at least one of the following controls is performed: (1) control of feeding the incineration object into the hopper, (2) control of the speed, (3) control of the conveying speed of the incineration object in the incinerator, (4) combustion control of the incineration object in the incinerator, and (5) stirring control of the incineration object. (III) Beneficial Effects

[0011] According to one aspect of the present invention, it is possible to help improve the operation of incineration facilities. Attached Figure Description

[0012] Figure 1 This is a block diagram illustrating an example of the main structure of the prediction device according to Embodiment 1 of the present invention. Figure 2 This is a diagram illustrating a structural example of a facility control system that includes the aforementioned predictive device. Figure 3 This is a diagram illustrating an example of a method for determining whether or not garbage is present. Figure 4 This is a diagram illustrating an example of a method for generating mobile data. Figure 5 It is a graph showing the relationship between feed rate and movement. Figure 6 This is a flowchart illustrating an example of the processing performed by the aforementioned prediction device. Figure 7 This is a block diagram illustrating an example of the main structure of the prediction device according to Embodiment 2 of the present invention. Figure 8 This is a graph showing the relationship between predicted and actual values ​​for volume density. Figure 9 This is a diagram illustrating the method for predicting lower heating values. Figure 10 This is a flowchart illustrating an example of the processing performed by the aforementioned prediction device. Figure 11 This is a block diagram illustrating an example of the main structure of the facility control device according to Embodiment 3 of the present invention. Figure 12 This is a flowchart illustrating an example of the processing performed by the aforementioned facility control device. Detailed Implementation

[0013] (Implementation Method 1) (System Architecture) based on Figure 2 The structure of the facility control system 100 of this embodiment will be described. Figure 2 This is a diagram illustrating a structural example of a facility control system 100. In Figure 2 The diagram illustrates an example of applying the facility control system 100 to a waste incineration facility. Furthermore, the facility control system 100 is not limited to waste and can be applied to any incineration facility that incinerates any incineration object.

[0014] exist Figure 2The waste incineration facility shown includes a waste pit B for storing waste G and an incinerator C for incinerating the waste G. Waste G from the waste pit B is fed into a hopper A via a crane. An inclined surface A1, extending outwards, is formed at the top of the hopper A. Waste G slides down this inclined surface A1 and enters the interior of the hopper A, where it is temporarily contained. Furthermore, the waste G from the hopper A is sequentially fed into the incinerator C for incineration through an opening A2 at its lower part.

[0015] The facility control system 100 is a system for predicting the characteristics of waste incinerated in incinerator C, and includes a prediction device 1 and a photographic device 2. The photographic device 2 is positioned above hopper A and takes pictures of hopper A from above at predetermined intervals. Furthermore, the prediction device 1 acquires the images taken by the photographic device 2 and uses these images to predict the characteristics of the waste to be incinerated.

[0016] More specifically, the photographic device 2 captures multiple images of hopper A in a sequential manner. Furthermore, the prediction device 1 predicts the characteristics of the waste based on movement data generated from the multiple images captured by the photographic device 2, representing the movement state of the waste G during the capture of these multiple images.

[0017] The inventors of this invention carefully examined images obtained from above of the hopper and discovered that differences in the properties of the incinerated material manifest as differences in its movement. For example, it was found that even when the incinerated material is fed into the incinerator at the same rate, materials with higher moisture content and greater difficulty in combustion tend to move a greater amount of material per unit time within the hopper compared to materials with lower moisture content and greater ease of combustion.

[0018] The prediction device 1 predicts the characteristics of the waste based on this insight. That is, the different characteristics of the waste are reflected in different movement states of the waste in the image, so according to the above structure, the characteristics of the waste can be predicted. Furthermore, understanding the characteristics of the waste helps in the proper operation of the incineration facility, so the prediction device 1 can help improve the operation of the incineration facility.

[0019] Specifically, the predictive device 1 controls the equipment within the waste incineration facility based on predictions of the waste's characteristics. This allows for appropriate control corresponding to the waste's characteristics, contributing to improved operation of the incineration facility. Furthermore, the control of the equipment within the waste incineration facility can be performed by operators of the facility; in this case, the predictive device 1 only needs to output the predicted waste characteristics to the operators. In this scenario, the predictive device 1 does not need to control the equipment.

[0020] The location for the prediction device 1 is not particularly limited; for example, it can be installed in a waste incineration facility or in a monitoring center that monitors multiple waste incineration facilities. Furthermore, Figure 2 An example is shown of using a prediction device 1 to predict the characteristics of waste in a single waste incineration facility, but it is also possible to use a prediction device 1 to predict the characteristics of waste in multiple waste incineration facilities. In this case, the prediction device 1 could be, for example, a cloud server.

[0021] (Structure of the prediction device) based on Figure 1 To illustrate the structure of prediction device 1. Figure 1 This is a block diagram illustrating an example of the main structural components of the prediction device 1. As shown, the prediction device 1 includes: a control unit 10, which performs comprehensive control over all parts of the prediction device 1; and a storage unit 11, which stores various data used by the prediction device 1. Furthermore, the prediction device 1 includes: a communication unit 12, which is used for communication between the prediction device 1 and other devices; an input unit 13, which receives various data inputs for the prediction device 1; and an output unit 14, which is used for the prediction device 1 to output various data.

[0022] Furthermore, the control unit 10 includes: a data acquisition unit 101, a determination unit 102, a motion data generation unit 103, a standard motion calculation unit 104, a characteristic prediction unit (prediction unit) 105, an influence prediction unit 106, and a device control unit 107. Additionally, the storage unit 11 stores an image DB 111 and a standard motion calculation formula 112.

[0023] The data acquisition unit 101 acquires multiple time-series images obtained by photographing hopper A from above. As described above, in the facility control system 100, the imaging device 2 captures multiple time-series images obtained by photographing hopper A from above, so the data acquisition unit 101 only needs to acquire the images captured by the imaging device 2. Furthermore, the data acquisition unit 101 can acquire images from the imaging device 2 via communication through the communication unit 12, or it can acquire images input via the input unit 13. The data acquisition unit 101 stores the acquired images in the image DB 111. Additionally, the imaging device 2 can also capture moving images. In this case, the data acquisition unit 101 only needs to acquire multiple time-series images (e.g., frame images at predetermined time intervals) from the moving images captured by the imaging device 2.

[0024] The determination unit 102 determines whether a garbage object region is present in the image obtained by the data acquisition unit 101, and this object region is the target of analysis for predicting garbage characteristics. The method for determining garbage in the object region will be discussed later. Figure 3 Please provide a detailed explanation.

[0025] The mobile data generation unit 103 generates mobile data representing the movement status of trash during the capture of multiple images acquired and recorded in the image DB111 by the data acquisition unit 101 in a time sequence. The method for generating mobile data will be discussed later. Figure 4 Please provide a detailed explanation.

[0026] The standard movement amount calculation unit 104 determines the speed at which waste is fed into the incinerator, i.e., the feeding speed. Furthermore, based on the determined feeding speed, the standard movement amount calculation unit 104 uses the standard movement amount calculation formula 112 to calculate the standard waste movement amount (hereinafter referred to as the standard movement amount) when waste is fed into the incinerator at the determined feeding speed.

[0027] Standard movement calculation formula 112 is a mathematical expression representing the relationship between the feed rate and the standard movement of the waste when it is fed into the incinerator at that feed rate. The standard movement calculation formula 112 and the method for calculating the standard movement using this formula 112 will be discussed later. Figure 5 Please provide a detailed explanation.

[0028] The property prediction unit 105 predicts the properties of the waste to be incinerated based on the movement data generated by the movement data generation unit 103 and the feeding speed. Specifically, the property prediction unit 105 calculates the difference between the movement amount represented by the movement data generated by the movement data generation unit 103 and the standard movement amount calculated by the standard movement amount calculation unit 104, and uses this difference as the prediction result of the waste properties.

[0029] The impact prediction unit 106 predicts the impact of incinerating the waste on the combustion state of the incinerator based on the characteristics of the waste predicted by the property prediction unit 105. The prediction result output by the impact prediction unit 106 can also represent the magnitude of the impact on the combustion state. For example, if the magnitude of the impact on the combustion state is represented by four stages—large, medium, small, and no impact—the impact prediction unit 106 only needs to output information representing one stage as the prediction result. Alternatively, the impact prediction unit 106 can also output a predicted value of the combustion state (e.g., a predicted value of the lower heating value) as the prediction result. The prediction of the impact on the combustion state will be discussed later based on... Figure 5 Please provide an explanation.

[0030] The equipment control unit 107 controls the equipment of the waste incineration facility based on the prediction results of the influence prediction unit 106. For example, the equipment control unit 107 can perform at least one of the following controls: (1) control of waste being fed into the hopper, (2) control of the feeding speed, (3) control of the waste conveying speed (grate speed) in the incinerator, and (4) control of the combustion of waste in the incinerator.

[0031] Furthermore, regarding the above (1), the following control can be implemented: when it is desired to promote combustion, the more easily combustible waste in the waste pit is fed into the hopper; when it is desired to suppress combustion, the less easily combustible waste is fed into the hopper. The distribution of the waste properties in the waste pit can also be managed separately, or it can be determined based on the prediction results of the property prediction unit 105 as explained later.

[0032] Furthermore, regarding (2) above, it is possible to increase the feed rate when it is desirable to promote combustion, and to decrease the feed rate when it is desirable to suppress combustion. The same applies to the grate speed in (3) above. In addition, depending on the situation, it may also be effective to use at least one method of reducing the feed rate and reducing the grate speed as a control for promoting combustion.

[0033] Furthermore, regarding the above (4), combustion control of waste can be achieved, for example, by heating the burner or controlling the supply of combustion air. That is, if it is desired to promote combustion (if it is desired to increase heat generation), it is sufficient to heat the burner or increase the supply of combustion air; conversely, if it is desired to suppress combustion (if it is desired to reduce heat generation), it is sufficient to reduce the supply of combustion air. In addition, depending on the situation, reducing the supply of combustion air is sometimes effective as a control for suppressing combustion.

[0034] Furthermore, the garbage shown when the hopper is photographed from above represents the garbage that has been present in the garbage pit up to this point, allowing us to determine the exact location within the garbage pit where the garbage was collected. Therefore, the distribution of the garbage's characteristics within the garbage pit can be predicted based on the prediction results of the characteristics prediction unit 105 or the prediction results of the influence prediction unit 106.

[0035] The equipment control unit 107 can control the mixing of waste in the landfill based on this property distribution. For example, the equipment control unit 107 can control the mixing to prioritize areas with non-combustible waste, or it can control the mixing / stirring of waste in areas with non-combustible waste and waste in areas with easily combustible waste to homogenize them. Furthermore, waste mixing refers to the processing carried out to improve the properties of the waste (especially its combustibility), such as by using a crane to grab the waste and drop it to disperse the grabbed waste.

[0036] As described above, the prediction device 1 includes a movement data generation unit 103 and a property prediction unit 105. The movement data generation unit 103 generates movement data representing the movement state of the waste by using multiple time-series images of the hopper taken from above, based on the waste being fed into the incinerator at a specified feeding rate. The property prediction unit 105 predicts the property of the waste based on the generated movement data and the feeding rate.

[0037] As mentioned above, the different properties of waste are reflected in different states of movement of waste in the image. Therefore, based on the above structure, the properties of waste can be predicted. Furthermore, understanding the properties of waste helps in the proper operation of incineration facilities; therefore, based on the above structure, it is possible to improve the operation of incineration facilities.

[0038] Furthermore, motion data can be generated using only two images taken at different times, as detailed later. Therefore, the operation of the prediction device 1 can be started immediately in the waste incineration facility without the need for machine learning or similar techniques. Additionally, even if the location or shooting angle of the camera device 2 changes slightly, the characteristics of the waste can be predicted without being affected. Moreover, it is conceivable that dirt may adhere to the lens of the camera device 2 installed in the waste incineration facility; however, since the dirt is immobile, its impact on the generation of motion data is limited. In other words, the facility control system 100 including the prediction device 1 has the advantages of easy implementation and stable operation.

[0039] (Regarding the determination of whether or not there is garbage) based on Figure 3 The determination of the presence or absence of garbage based on the determination unit 102 will be explained. Figure 3 This is a diagram illustrating an example of methods for determining the presence or absence of garbage. Figure 3 Image D shown is an image obtained by taking a picture of the hopper from above. Image D shows waste sliding down the inclined surface of the hopper, but not the entire inclined surface. Specifically, waste is shown in region d2 on the downstream side of the inclined surface of the hopper, but not in region d1 on the upstream side.

[0040] Here, if we assume that the movement data generation unit 103 generates movement data for areas such as region d1 where no garbage is reflected, the movement amount represented by the movement data may be zero or close to zero. Furthermore, when the movement amount is zero or close to zero, it is possible to output an incorrect prediction result about garbage that does not actually exist.

[0041] To avoid such erroneous detection, the prediction device 1 includes a determination unit 102. The determination unit 102 determines whether the object area in the image contains waste. Furthermore, a movement data generation unit 103 generates movement data, which represents the amount of movement between images determined by the determination unit 102 to contain waste in the object area. Therefore, images where the object area does not contain the object are not used for predicting the waste characteristics, thus preventing such erroneous predictions. Furthermore, it is preferable that the object area is set on the inclined surface of the hopper; more specifically, it is set on the downstream inclined surface, as in region d2, as will be described later.

[0042] The presence of litter can be determined using the captured image, or it can be easily determined by performing prescribed image processing on the captured image, followed by further processing. For example, the Canny method can be used for image edge detection. Figure 3 Image D1 shown is generated by edge detection of image D using the Canny method. The appearance of waste is often much more complex than the surface shape of the hopper. Therefore, in Figure 3 In the example, compared to region d1 where no trash was reflected, a large number of edges were detected in region d2 where trash was reflected.

[0043] Therefore, the determination unit 102 can use the Canny method to perform edge detection on the image obtained by photographing the hopper, and determine the image with the number of edges detected in the object area that is above a threshold as an image showing garbage, and determine the image with the number of edges detected in the object area that is less than a threshold as an image not showing garbage.

[0044] Of course, edge detection methods other than the Canny method can also be applied. Furthermore, the presence or absence of garbage can be determined using methods other than edge detection. For example, the determination unit 102 can determine the presence or absence of garbage by analyzing at least one of the brightness value and RGB value of each pixel constituting the image. Alternatively, the determination unit 102 can determine the presence or absence of garbage using a learned model (e.g., a neural network model) that has been learned in a way that can determine the presence or absence of garbage.

[0045] (Methods for generating mobile data) based on Figure 4 The method for generating mobile data based on the mobile data generation unit 103 will be described. Figure 4 This is a diagram illustrating an example of a method for generating mobile data. More specifically, Figure 4An example of generating motion data F using image correlation is shown, which represents the movement state of waste between image E1, obtained by photographing the inclined surface of the hopper at time t, and image E2, obtained by photographing the inclined surface of the hopper at time t+Δt. Furthermore, in Figure 4 In the diagram, the y-direction is the direction in which the waste descends, that is, from the inclined surface of the hopper (refer to...). Figure 2 The direction from the upstream side of the inclined surface A1 to the downstream side, the x direction is the direction perpendicular to the y direction on the inclined surface.

[0046] Image correlation is a method that, for two time-series images, calculates the movement of each particle reflected in each image through image processing and represents the calculated movement as a vector. For example, in image E1, the trash indicated by the dashed bounding box is reflected at position e1, but in image E2, it is reflected at position e2, which is lower than position e1. The movement state of this trash is represented in the movement data F as a vector f1 representing the displacement from position e1 to position e2. Vectors are calculated similarly for other positions.

[0047] Therefore, the movement data F represents the movement state of the waste at each location in images E1 and E2. Specifically, according to the movement data F, the waste at each location in images E1 and E2 moves in the y-direction (downstream of the hopper's inclined surface, i.e., the downward direction). Furthermore, in the movement data F, vectors with y-direction component values ​​above a threshold are represented by solid arrows, while vectors with values ​​below the threshold are represented by dashed arrows. This indicates that there are locations where the waste moves faster and slower.

[0048] As described above, the motion data generation unit 103 can generate motion data representing the amount of motion of an image element that is reflected in a plurality of predetermined positions in a downward direction between images E1 and E2. The plurality of predetermined positions are set on an inclined surface reflected in images E1 and E2.

[0049] Since the waste slides down the inclined surface of the hopper in roughly the same downward direction, the inclined surface is suitable for determining whether the waste is falling smoothly or has stopped falling. Therefore, based on the above structure, movement data that accurately indicates whether the waste is falling smoothly or has stopped falling can be generated. Furthermore, by using this movement data to predict the waste's characteristics, highly reliable predictions can be made. This is because, as mentioned above, different waste characteristics are manifested as different states of movement of the waste reflected in the image.

[0050] Furthermore, the moving data can also be used to generate image portions that reflect parts other than the inclined surface. In this case, the morphology prediction unit 105 only needs to use the moving data within the object area defined in the portion reflecting the inclined surface from the moving data generated by the moving data generation unit 103 to predict the morphology of the waste.

[0051] Of course, the target area can be any area where the movement of waste is in roughly the same direction, or it can be any area outside the inclined surface. For example, the target area can be the area on the inner surface of the hopper that is connected to the inclined surface on the downstream side (usually extending in the vertical direction), or it can be the area that includes both the inclined surface and other parts.

[0052] Furthermore, motion data can be any data representing the movement status of the garbage, and is not limited to the examples above. For example, the speed at which the garbage moves can be used as motion data. Alternatively, a difference image representing the difference between two images taken at different times can also be used as motion data.

[0053] (Calculation of standard movement quantity ~ Prediction of combustion state) based on Figure 5 To explain: the standard movement quantity calculation unit 104 calculates the standard movement quantity, the property prediction unit 105 predicts the property of the waste, and the influence prediction unit 106 predicts the combustion state. Figure 5 It is a graph showing the relationship between feed rate and movement.

[0054] Figure 5 The graph H shown depicts the movement of various types of waste measured at different feed rates in a waste incineration facility. Furthermore, a histogram h1 representing the distribution of feed rates is shown above graph H, and a histogram h2 representing the distribution of movement is shown to the right of graph H. Here, movement refers to the amount of movement in the downward direction of the waste between two images taken in sequence.

[0055] As shown in the figure, there is a correlation between the feeding rate and the amount of waste moved; the higher the feeding rate, the greater the amount of waste moved. This correlation can be formalized through methods such as regression analysis. Figure 5 In the example, h3 represents the straight line y = F(x) obtained through regression analysis, which is a regression equation representing the relationship between the feed rate and the amount of waste movement. That is, h3 is the regression line. The prediction device 1 stores such a regression equation as the standard movement amount calculation equation 112 in the storage unit 11.

[0056] The standard movement calculation unit 104 calculates the standard movement at the current feed rate by determining the current feed rate and substituting it into the standard movement calculation formula 112. This allows for the calculation of an appropriate standard movement value. For example, when the current feed rate is x1, the standard movement is y = F(x1).

[0057] In this case, the property prediction unit 105 can calculate the difference between the movement amount represented by the movement data generated by the movement data generation unit 103 and the standard movement amount of the waste as a prediction result. As described above, according to the inventors' research results according to the present invention, waste with higher moisture content and harder to burn tends to have a larger movement amount per unit time within the hopper compared to waste with lower moisture content and easier to burn. Therefore, it can be said that the difference between the movement amount represented by the movement data and the standard movement amount represents the ease of combustion of the waste. Therefore, based on the above structure, a value representing the ease of combustion of the waste can be calculated as a prediction result.

[0058] For example, such as Figure 5 As shown in the example, assuming the current feed rate is x1, the movement amount represented by the movement data generated by the movement data generation unit 103 is y1. Figure 5 The combination of feed rate and movement is represented by point h4 on the curve plane of curve H. As mentioned above, since the standard movement when the feed rate is x1 is y = F(x1), in this example, the property prediction unit 105 calculates y' = y1 - F(x1) as the prediction result.

[0059] The larger the calculated y' value, the more the waste's properties deviate from the standard waste characteristics. More specifically, when y' is negative, the larger the absolute value of y', the less the waste moves. Generally speaking, lighter waste moves less and has less moisture compared to standard-weight waste, making it easier to burn. Therefore, waste with negative y' values ​​and large absolute values ​​of y' is more easily burned.

[0060] On the other hand, when y' is a positive value, it means that the larger the absolute value of y', the greater the movement of the waste. Generally speaking, heavier waste moves faster than standard-weight waste, has more moisture, and is more difficult to burn. Therefore, waste with a positive y' and a large absolute value of y' is difficult to burn.

[0061] Therefore, for example, in Figure 5 In the case where the movement and feeding speed of a piece of waste are plotted within the region h5 above the regression line h3, the waste is considered difficult to burn. On the other hand, in Figure 5In the context of a scenario where the movement and feeding speed of a piece of waste are depicted within a region h6 below the regression line h3, this waste is considered easily combustible. Point h4 is located within the region h6 below the regression line h3; therefore, the waste at point h4 is also easily combustible.

[0062] Based on the above, the impact prediction unit 106 can classify the waste according to the value of y' and output the classification result as a prediction result. For example, the impact prediction unit 106 can also predict that there is an impact on the combustion state if the value of y' calculated by the property prediction unit 105 is drawn within region h5 or h6. Regarding the magnitude of the impact, the impact prediction unit 106 only needs to determine it according to the magnitude of the value of y'. In addition, the impact prediction unit 106 can predict that there is no impact on the combustion state if the area drawn is between region h5 and h6.

[0063] Furthermore, the method for predicting the combustion state can be any method that appropriately corresponds to the combustion state of the anticipated object, and is not limited to the examples described above. For example, the influence prediction unit 106 can predict the lower heating value. In this case, it is sufficient to standardize the correlation between the value of y' and the lower heating value. Moreover, the influence prediction unit 106 can simply use this formula to predict the lower heating value.

[0064] Furthermore, the method for determining the standard movement amount is not limited to the examples described above. For instance, similar to the examples above, after measuring the movement amount of waste of various types at various feed rates in a waste incineration facility, the feed rates can be divided into multiple numerical ranges, and the standard movement amount for each range can be determined. The standard movement amount can be, for example, the average value or median value of the movement amounts in each range. In this case, the standard movement amount calculation unit 104 only needs to determine the standard movement amount based on which range the current feed rate belongs to.

[0065] Alternatively, instead of using a linear model such as regression analysis to calculate the standard movement amount, a nonlinear model such as a neural network can be used to calculate the standard movement amount. In this case, a standard movement amount prediction model that pre-learns the relationship between the feed rate and the standard movement amount is prepared using machine learning. Furthermore, the standard movement amount calculation unit 104 only needs to input the feed rate into the standard movement amount prediction model to calculate the standard movement amount.

[0066] (Processing flow) based on Figure 6 This will explain the processing flow (prediction method) performed by prediction device 1. Figure 6 This is a flowchart illustrating an example of the processing performed by the prediction device 1. Furthermore, for example, this is performed whenever the imaging device 2 captures a new image. Figure 6 The processing.

[0067] In S1, the data acquisition unit 101 acquires a timing image obtained from the overhead camera hopper and records it in image DB111. As described above, due to the camera device 2 (refer to...) Figure 2 These images are captured, so the data acquisition unit 101 only needs to acquire the images captured by the photography device 2.

[0068] In S2, the determination unit 102 determines whether the object region in the image obtained through S1 contains existing garbage. The method for determining whether the object region contains existing garbage is similar to that based on... Figure 3 As previously explained, this will not be repeated here. If the condition is yes in S2, proceed to S3; if the condition is no in S2, the process ends. Figure 6 The processing.

[0069] In step S3 (mobility data generation step), the mobility data generation unit 103 generates mobility data representing the movement status of trash during the period when these images were captured, based on the images obtained in step S1 and images captured before those images. For example, if an image captured at time t+Δt is obtained in step S1, the mobility data generation unit 103 reads the image captured at time t from image DB111. Furthermore, the mobility data generation unit 103 generates mobility data representing the movement status of trash during the period from time t to time t+Δt. Regarding the method for generating mobility data, it is similar to that based on... Figure 4 As already explained, I will not repeat it here.

[0070] In S4, the standard movement calculation unit 104 determines the feeding speed and calculates the standard movement amount using the standard movement amount calculation formula 112 based on the determined feeding speed. Furthermore, the feeding speed can be set to a value that is input by the user of the prediction device 1 via the input unit 13, or it can be determined by communicating with the control device of the waste treatment facility via the communication unit 12.

[0071] In step S5 (predicting characteristics), the characteristics prediction unit 105 predicts the characteristics of the waste to be incinerated. More specifically, the characteristics prediction unit 105 calculates the difference between the movement amount represented by the movement data generated by the movement data generation unit 103 in step S3 and the standard movement amount calculated by the standard movement amount calculation unit 104 using the feed rate in step S4 as the prediction result. As described above, since the standard movement amount is calculated using the feed rate, in step S5, the characteristics prediction unit 105 predicts the characteristics of the waste based on the movement data and the feed rate.

[0072] In S6, the impact prediction unit 106 predicts the impact of incinerating the waste on the combustion state of the incinerator based on the characteristics of the waste predicted by the property prediction unit 105 in S5. Furthermore, the prediction of the combustion state is not mandatory; subsequent processing steps (S7 and beyond) can also be based on the prediction results from S5.

[0073] In step S7, the equipment control unit 107 determines, based on the prediction result of step S6, whether to perform control to maintain a good combustion state in the incinerator. If the determination is yes in step S7, the process proceeds to step S8, where the equipment control unit 107 performs control to maintain a good combustion state in the incinerator. Conversely, if the determination is no in step S7, step S8 is not performed, and the process ends. Figure 6 The processing.

[0074] Furthermore, the judgment criteria for S7 only need to be predetermined, and the method for determining the control content for S8 only needs to be predetermined. For example, it could be that when the magnitude of the influence on the combustion state is predicted using four stages—large, medium, small, and no influence—control is performed if the predicted result is anything other than no influence (determined as yes in S7). Moreover, it could be that the greater the predicted influence on the combustion state, the more the equipment control unit 107 can increase the control quantity used to promote or suppress combustion.

[0075] Here, the equipment control unit 107 can determine whether to perform combustion-promoting control or combustion-suppressing control as follows: First, the equipment control unit 107 determines whether the movement amount represented by the movement data generated by the movement data generation unit 103 in S3 is greater than the standard movement amount calculated by the standard movement amount calculation unit 104 using the feed rate in S4. In this determination, if it is determined that the movement amount represented by the movement data generated by the movement data generation unit 103 is greater, it is considered that the waste has more moisture than standard waste and is difficult to burn. Therefore, the equipment control unit 107 only needs to determine to perform combustion-promoting control (e.g., heating with the burner, increasing the amount of combustion air). On the other hand, if it is determined that the standard movement amount is greater, it is considered that the waste has less moisture than standard waste and is easy to burn. Therefore, the equipment control unit 107 only needs to determine to perform combustion-suppressing control (e.g., reducing the amount of combustion air). As a result, the combustion state in the incinerator can be maintained in a good state.

[0076] Additionally, in S7, control can also be implemented that takes into account the delay in the waste fed into the hopper being sent to the incinerator. For example, the equipment control unit 107 can use the feeding speed, etc., to calculate the time it takes for the waste fed into the hopper to be sent to the incinerator, and when that time has elapsed, perform control on the waste corresponding to its impact on the predicted combustion state.

[0077] Furthermore, depending on the control content, there may be a delay in the reflection of the control result on the combustion state. Therefore, the equipment control unit 107 can also perform control so that the time when the control result is reflected on the combustion state is the same as or earlier than the time when the waste is fed into the incinerator.

[0078] Alternatively, in S7, the equipment control unit 107 can determine the control content based on the prediction result of the property prediction unit 105. In this case, the processing in S6 is omitted, and it is not necessary to affect the prediction unit 106.

[0079] As described above, the prediction method of this embodiment includes steps (S3) and (S5). In step (S3), movement data representing the movement state of the waste is generated based on multiple time-series images obtained by taking pictures of the hopper from above in an incineration facility where waste is fed into the hopper at a specified feeding rate. In step (S5), the characteristics of the waste being incinerated are predicted based on the movement data generated in step (S3) and the feeding rate. This prediction method can help improve the operation of the incineration facility.

[0080] Furthermore, as described at the beginning of Embodiment 1, the facility control system 100 is not limited to a waste incineration facility, but can be applied to an incineration facility that incinerates any incineration object, and the incineration object stored in the hopper is not limited to waste. That is, the "waste" described in the above embodiments can be replaced with any "incineration object". This is also true in Embodiment 2 and thereafter.

[0081] (Implementation Method 2) Another embodiment of the present invention will be described below. Furthermore, for ease of explanation, reference numerals that have the same function as those used for components described in the above embodiments will not be repeated. This is also true in Embodiment 3, which will be described later.

[0082] (Structure of the prediction device) Figure 7 This is a block diagram illustrating an example of the main structural components of the prediction device 1A according to this embodiment. The prediction device 1A differs from the other embodiments in that it lacks a standard movement calculation unit 104 and instead includes a volume density prediction unit (prediction unit) 105A and a calorific value prediction unit 106A, replacing the property prediction unit 105 and the influence prediction unit 106. Figure 1 The prediction device 1 shown is different. In addition, the prediction device 1A does not store the standard movement amount calculation formula 112 in the storage section 11, but instead stores the volume density prediction formula 112A and the calorific value prediction formula 113A.

[0083] The volumetric density prediction unit 105A uses the movement amount, feed rate, and volumetric density prediction formula 112A represented by the movement data generated by the movement data generation unit 103 to calculate the volumetric density. The volumetric density prediction formula 112A is a mathematical expression representing the relationship between the movement amount and feed rate of the waste and the volumetric density of that waste. The following is based on... Figure 8 The method for predicting bulk density is explained in detail.

[0084] The calorific value prediction unit 106A uses calorific value prediction formula 113A, which expresses the relationship between the bulk density of the waste and the lower heating value (calorific value) when the waste is incinerated, to calculate the predicted value of the lower heating value (calorific value) when the waste is incinerated. The following is based on... Figure 9 The method for predicting lower heating value (heat value) is explained in detail.

[0085] (Methods for predicting bulk density) The inventors of this invention investigated whether the volumetric density could be represented by the amount of waste moving and the feed rate. The results of this study found that the volumetric density of waste of various volumes could be predicted using the following mathematical formula (1), wherein the mathematical formula (1) was obtained by regression analysis using the results of measuring the amount of waste moving and the feed rate. y=ax1+bx2+cx1*x2+d (1) Furthermore, in the above mathematical formula (1), y is the volume density [t / m³]. 3 (t is the unit of mass: tons), x1 is the feed rate [m / h], X2 is the amount of waste moved (a dimensionless quantity calculated using the image correlation method), and a to d are coefficients obtained through regression analysis.

[0086] Figure 8 This is a graph that uses the predicted volume density calculated using the above mathematical formula (1) as the vertical axis and the actual volume density as the horizontal axis, showing the relationship between the predicted and actual values. Figure 8 As shown in graph J, the volume density can be predicted with high accuracy using the above mathematical formula (1). In the example shown in graph J, the coefficient of determination is 0.8565.

[0087] In addition, the inventors of this invention also attempted to predict the volume density using the following mathematical formula (2) for comparison. y = ax1 + b (2) Although the predicted volumetric density using mathematical formula (2) is consistent with the actual volumetric density to some extent, its coefficient of determination is lower than that using mathematical formula (1). Therefore, the amount of waste movement helps improve the accuracy of volumetric density estimation.

[0088] Of course, in addition to using mathematical formula (1), various methods can be used to calculate the predicted value of bulk density based on the feed rate and the amount of waste movement. That is, the predicted value of bulk density can be calculated using any prediction model that models the relationship between the feed rate, the amount of waste movement, and bulk density.

[0089] As described above, the inventors of this invention have found that there is a correlation between the amount of waste moving within the hopper, the feeding speed, and the bulk density of the waste. Therefore, the prediction device 1A can calculate an appropriate value of bulk density as a prediction result. This prediction device 1A includes a bulk density prediction unit 105A, which calculates the bulk density using a bulk density prediction formula 112A that expresses the relationship between the amount of waste moving, the feeding speed, and the bulk density of the waste.

[0090] Furthermore, the feed rate changes every second, providing high real-time accuracy. Additionally, the amount of waste moving is also highly real-time information, obtainable, for example, in minutes. Moreover, as explained below, the amount of waste moving depends on its bulk density.

[0091] First, it is known that there is a correlation between the feed rate and the amount of waste moving. However, even at the same feed rate, the amount of waste moving varies. One reason for this variation is the difference in the weight of the waste. That is, as waste settles along the hopper, heavier waste is compressed more by its own weight than lighter waste, thus increasing its movement per unit time, i.e., its speed. Therefore, the speed at which waste settles along the hopper is affected by both the feed rate and the compression caused by the waste's own weight. Consequently, the weight per unit volume, i.e., the bulk density, varies depending on the feed rate and the degree of compression based on the waste's own weight. Therefore, the amount of waste moving depends not only on the feed rate but also on its bulk density.

[0092] Based on the aforementioned structure, the bulk density of the waste within the hopper can be estimated using highly real-time information. Specifically, by taking multiple photos of the hopper from above and determining the feeding speed during these photos, a predicted value for the bulk density can be calculated. Consequently, the lower heating value of the waste can also be estimated in real time based on the bulk density, and the estimation results can be effectively used in the control of the incinerator.

[0093] Furthermore, the bulk density can also be calculated using the mathematical formula (3) described later. However, in this method, in order to make the amount (weight) of waste processed an appropriate value, the above-mentioned specified period needs to be set to a relatively long period, such as 1 day, and the bulk density cannot be calculated at a time that can be used for the control of the incinerator.

[0094] (Methods for predicting lower heating values) based on Figure 9 The method for predicting lower heating value using the heating value prediction unit 106A is explained. Figure 9 This is a diagram illustrating the method for predicting lower heating values. Additionally, in Figure 9 Also shown Figure 2 A schematic cross-sectional view of the area near the lower opening A2 of hopper A shown.

[0095] In waste incineration facilities, the waste fed into the hopper remains in the hopper for a period of time before being sent to the incinerator for combustion. Therefore, it is difficult to measure the low calorific value of the waste during combustion as reflected in the image obtained by taking a picture of the hopper from above.

[0096] Therefore, the inventors of this invention first determined the relationship between bulk density and lower heating value based on the average calorific value of the incinerator over a day and the average bulk density of that day. The calorific value prediction unit 106A uses this determined relationship to calculate the lower heating value based on the bulk density.

[0097] like Figure 9 As shown, the waste collected inside the hopper A is fed into the incinerator C through the opening A2. Therefore, the processing capacity of the incinerator C after one day (24 hours) is expressed as follows, with the cross-sectional area of ​​the opening A2 set to So, the average feed rate set to V, and the incinerator C operating for one day (24 hours).

[0098] (Daily processing capacity) [tons / day] = {V[m / h] × 24[h / day]} × So[m 2 × (bulk density) [tons / m³] 3 ] Therefore, the bulk density is expressed as follows.

[0099] (Bulk density) = (Daily processing capacity) / (V × 24 × So) (3) Figure 9 K is a graph representing the relationship between the bulk density, calculated as described above, and the net calorific value (NCV) Hu in incinerator C for that day. As shown in the figure, although there are deviations, there is a correlation between bulk density and NCV as a whole. More specifically, there is a proportional relationship between bulk density and NCV; therefore, the relationship between bulk density and NCV can be represented by a straight line k1.

[0100] Therefore, if the mathematical expression of the straight line k1 is stored in the storage unit 11 as the calorific value prediction expression 113A, the calorific value prediction unit 106A can use the mathematical expression and calculate the lower heating value based on the volume density predicted by the volume density prediction unit 105A.

[0101] As described above, it is known that the bulk density of waste is related to the lower heating value when the waste is incinerated. Therefore, the prediction device 1A can calculate an appropriate value of lower heating value as a prediction result, wherein the prediction device 1A includes a heating value prediction unit 106A, which uses a heating value prediction formula 113A that expresses the relationship between bulk density and lower heating value to predict the lower heating value.

[0102] Furthermore, as described above, the bulk density can be calculated based on the amount of waste moving and the feeding rate, and the lower heating value can be calculated based on the bulk density. Therefore, the lower heating value can be calculated based on the amount of waste moving and the feeding rate even without calculating the bulk density. That is, the heating value prediction unit 106A can also use a mathematical formula that represents the relationship between the amount of waste moving, the feeding rate, and the lower heating value when incinerating waste fed into the incinerator at that feeding rate, as expressed in the moving data generated by the moving data generation unit 103. Even with this structure, an appropriate value of lower heating value can be calculated as a prediction result representing the waste characteristics.

[0103] As a mathematical formula for calculating the lower heating value of waste, a regression formula such as the one described above can be applied, but it is not limited to this example. The calorific value prediction unit 106A can use any prediction model that models the relationship between the feeding speed and the amount of waste movement and the lower heating value to calculate the predicted value of the lower heating value.

[0104] (Processing flow) based on Figure 10 This section explains the processing flow (prediction method) performed by the prediction device 1A. Figure 10 This is a flowchart illustrating an example of the processing performed by the prediction device 1A. Furthermore, Figure 10 The processing of S11 to S13 and Figure 6 S1 to S3 are the same, so the explanation is omitted.

[0105] In S14 (the step of predicting characteristics), the bulk density prediction unit 105A substitutes the amount of movement represented by the movement data generated in S13 and the feeding speed when the waste moves at that amount of movement into the bulk density prediction formula 112A to calculate the predicted value of the bulk density.

[0106] In S15, the calorific value prediction unit 106A substitutes the predicted value of the bulk density calculated in S14 into the calorific value prediction formula 113A to calculate the predicted value of the lower heating value. Furthermore, as described above, the lower heating value can be calculated based on the amount of waste movement and the feeding rate without calculating the bulk density. In this case, S14 is omitted, and S15 becomes the step of predicting the properties of the waste (specifically, the lower heating value).

[0107] In S16, the equipment control unit 107 determines whether to perform control to maintain a good combustion state in the incinerator based on the prediction result of S15. If the determination is yes in S16, the process proceeds to S17, where the equipment control unit 107 performs control to maintain a good combustion state in the incinerator. Conversely, if the determination is no in S16, the process in S17 is not performed, and the process ends. Figure 10 The processing.

[0108] Furthermore, the judgment criteria for S16 only need to be predetermined, and the method for determining the control content in S17 also only needs to be predetermined. For example, control can be performed when the predicted value of the lower heating value calculated in S15 is outside the predetermined normal range (determined as yes in S16). Moreover, if the predicted value of the generated heat is less than the lower limit of the normal range, the equipment control unit 107 only needs to perform control to increase the generated heat (e.g., heating using the burner, or controlling the increase of the combustion air volume). On the other hand, if the predicted value of the generated heat exceeds the upper limit of the normal range, the equipment control unit 107 only needs to perform control to reduce the generated heat (e.g., reducing the combustion air volume). Therefore, the combustion state inside the incinerator can be maintained in a good state.

[0109] (Implementation Method 3) (Structure of the facility control device) Figure 11 This is a block diagram illustrating an example of the main structure of the facility control device 1B according to this embodiment. The facility control device 1B controls the equipment within the waste incineration facility based on movement data and feed rate generated from multiple time-series images of the hopper obtained from above.

[0110] Compared to the aforementioned prediction devices 1 and 1A, facility control device 1B is similar in generating movement data based on multiple time-series images obtained from above photographs of the hopper, but differs in controlling the equipment without predicting the waste characteristics. Therefore, in Figure 2 In addition, it can also form a facility control system 100 that replaces the prediction device 1 with the facility control device 1B.

[0111] Facility control device 1B and Figure 1 Compared to the prediction device 1 shown, the device lacks the standard movement calculation unit 104, the property prediction unit 105, and the influence prediction unit 106, and instead has a control content determination unit 105B. Furthermore, the storage unit 11 of the facility control device 1B does not store the standard movement calculation formula 112, but instead stores the control content determination model 112B.

[0112] The control content determination unit 105B determines the control content for the equipment within the waste incineration facility based on the movement data generated by the movement data generation unit 103 and the feeding speed. For example, the control content determination unit 105B may also determine at least one of the following control contents: (1) control of feeding waste into the hopper, (2) control of the feeding speed, (3) control of the conveying speed (grate speed) of waste in the incinerator, (4) control of the combustion of waste in the incinerator, and (5) control of the stirring of waste in the pit. The control content may also include, for example, the controlled object and the controlled quantity.

[0113] Furthermore, these controls are necessary to maintain a proper combustion state. Therefore, the control content determination unit 105B determines not to perform controls when there are no controls required to maintain a proper combustion state.

[0114] More specifically, the control content determination unit 105B uses a control content determination model 112B to determine the control content. The control content determination model 112B is a model constructed in a way that allows the control content to be determined based on movement data and feed rate. For example, when determining the control quantity of grate speed, the control content determination model 112B can be used, which uses the movement quantity and feed rate as explanatory variables and the optimal grate speed for maintaining proper combustion as the target variable. Such a control content determination model 112B can be constructed using regression analysis or other algorithms such as neural networks.

[0115] Furthermore, the equipment control unit 107 performs control on the equipment of the waste incineration facility as determined by the control content determination unit 105B. As described above, since the control content determination unit 105B determines the control content based on movement data and feeding speed, the equipment control unit 107 controls the equipment of the waste incineration facility based on movement data and feeding speed.

[0116] As described above, the facility control device 1B includes a movement data generation unit 103 and an equipment control unit 107. The movement data generation unit 103 generates movement data representing the movement status of the waste based on multiple images obtained by taking pictures of the hopper from above in an incineration facility where waste is fed into the hopper at a specified feeding speed. The equipment control unit 107 performs at least one of the following based on the movement data generated by the movement data generation unit 103 and the feeding speed: (1) control of waste feeding into the hopper, (2) control of the feeding speed, (3) control of the waste conveying speed in the incinerator, (4) control of the combustion of waste in the incinerator, and (5) control of the stirring of waste in the pit.

[0117] As described above, the different properties of incineration materials such as waste manifest as different movement states of the incineration materials reflected in the image. Furthermore, these different properties of the incineration materials affect the content of the controls required to maintain an appropriate combustion state. Therefore, the controls required to maintain an appropriate combustion state of the waste can be determined based on the movement state of the waste reflected in the image. Thus, by implementing appropriate controls corresponding to the properties of the waste according to the above structure, the operation of the incineration facility can be improved.

[0118] (Processing flow) based on Figure 12 To explain the process flow (facility control method) performed by facility control device 1B. Figure 12 This is a flowchart illustrating an example of the processing performed by facility control device 1B. Furthermore, Figure 12 The processing of S21 to S23 and Figure 6 S1 to S3 are the same, so the explanation is omitted.

[0119] In S24, the control content determination unit 105B uses the control content determination model 112B to determine the control content. For example, if the movement data generated in S23 is set to represent the amount of movement, the control content determination model 112B is a model that uses the amount of movement and the feed rate as explanatory variables, and the change in the optimal feed rate for maintaining a proper combustion state as the target variable. In this case, the control content determination unit 105B determines the change in the optimal feed rate for maintaining a proper combustion state based on the value obtained by inputting the amount of movement represented by the movement data generated in S23 and the feed rate into the control content determination model 112B.

[0120] In step S25, the equipment control unit 107 determines whether to perform control to maintain a good combustion state in the incinerator. If the determination is yes in S25, the process proceeds to step S26, where the equipment control unit 107 performs the control determined in step S24. Conversely, if the determination is no in S25, step S26 is not performed, and the process ends. Figure 12 The processing.

[0121] Furthermore, the decision criteria for S25 can be predetermined. For example, if it is uncertain whether control will be performed in S24, or if the control quantity is zero, then control can be determined to be performed (yes in S25). Alternatively, for example, if model 112B is determined using control content that is predicted to be optimal and outputs a value representing the reliability of that prediction, then control can be determined to be performed (yes in S25) if the reliability is above a threshold. Alternatively, for example, the operator can be prompted with the control content determined in S24 and asked to input whether to execute the control.

[0122] As described above, the facility control method of this embodiment includes steps (S23) and (S26). In step (S23), movement data representing the movement state of the waste is generated based on multiple images obtained by taking pictures of the hopper from above in an incineration facility where waste is fed into the hopper at a specified feeding rate. In step (S26), at least one of the following is performed based on the movement data generated in step (S23) and the feeding rate: (1) control of waste feeding into the hopper, (2) control of the feeding rate, (3) control of the waste conveying rate in the incinerator, (4) control of the combustion of waste in the incinerator, and (5) control of the agitation of waste in the pit. According to this facility control method, it is possible to improve the operation of the incineration facility.

[0123] (Regarding the bridge construction) In a hopper, bridging can occur, causing blockages in the waste. If bridging occurs, the amount of waste moving in the upper part of the hopper (the area where bridging occurs and above it) decreases significantly or becomes zero, regardless of the waste's characteristics. Therefore, it is difficult to predict the waste's characteristics based on images taken when bridging occurs.

[0124] Therefore, it is also possible that the prediction devices 1 and 1A described in the above embodiments include a bridging detection unit that detects the occurrence of bridging, and when the bridging detection unit detects the occurrence of bridging, it does not perform the prediction of waste characteristics. Alternatively, it is also possible that the prediction devices 1 and 1A maintain the prediction results of waste characteristics when bridging occurs, and restart the prediction of characteristics after the bridging is eliminated.

[0125] There are no particular limitations on the method for detecting and eliminating bridging. For example, the occurrence and elimination of bridging can also be detected based on the amount of waste movement represented by the movement data generated by the movement data generation unit 103. Similarly, the facility control device 1B can detect the occurrence of bridging, maintain control when bridging occurs, and restart control after the bridging is eliminated.

[0126] Conversely, bridging can also be detected based on the predicted waste properties, taking into account the impact of bridging on the prediction results. For example, prediction device 1A can determine that bridging has occurred when there is a discrepancy between the predicted and measured bulk density. Furthermore, if bridging persists for an extended period, it also affects the lower heating value; therefore, prediction device 1A can detect bridging based on the predicted lower heating value. Figure 9 If the straight line k1 deviates from the line shown, it is determined that a bridge has been built.

[0127] (Modified Example) The execution entities for each process described in the above embodiments are arbitrary and not limited to the examples described above. That is, multiple interconnectable devices can be used to construct a system with the same functions as the prediction device 1, 1A, and the facility control device 1B. For example, an information processing device can be used to perform the process. Figure 6 The process involves steps S1 to S3, and the generated mobile data is sent to another information processing device, where steps S4 to S8 are performed. Figure 9 and Figure 12 The processes shown are the same.

[0128] (An example implemented using software) The functions of the prediction device 1, 1A and facility control device 1B (hereinafter referred to as "the device") can be realized by making the computer a program for the device to function, that is, by making the computer a program (prediction program / control program) for the various control blocks of the device (especially the various parts included in the control unit 10) to function.

[0129] In this case, the aforementioned device is equipped with a computer as hardware for executing the aforementioned program, the computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory). By utilizing the control device and the storage device to execute the aforementioned program, the functions described in the above embodiments are realized.

[0130] The aforementioned program can be recorded on one or more non-transitory recording media that are computer-readable. The aforementioned device may or may not have such a recording medium. In the latter case, the aforementioned program can be provided to the aforementioned device via any wired or wireless transmission medium.

[0131] Furthermore, some or all of the functions of the aforementioned control blocks can also be implemented using logic circuits. For example, integrated circuits that form the logic circuits that enable the functions of the aforementioned control blocks are also included within the scope of this invention. Additionally, the functions of the aforementioned control blocks can also be implemented using a quantum computer, for example.

[0132] This invention is not limited to the embodiments described above. Various modifications can be made within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of this invention. Explanation of reference numerals in the attached figures

[0133] 1-Prediction device 103-Mobile Data Generation Department 104-Standard Movement Calculation Department 105 - Trait Prediction Department (Prediction Section) 1A-Prediction Device 103-Mobile Data Generation Department 105A - Bulk Density Prediction Unit (Prediction Unit) 1B - Facility Control Device 103-Mobile Data Generation Department 107 - Equipment Control Department.

Claims

1. A prediction device comprising: A motion data generation unit generates motion data representing the movement state of an object to be incinerated by feeding a hopper into the incinerator at a specified speed, based on multiple time-series images of the hopper taken from above; and The prediction unit predicts the properties of the object to be incinerated based on the movement data generated by the movement data generation unit and the speed.

2. The prediction device according to claim 1, characterized in that, The movement data represents the amount of movement of the object to be incinerated. The prediction unit calculates the difference in movement as the prediction result, namely: the movement represented by the movement data generated by the movement data generation unit; and the standard movement of the incineration object when it is fed into the incinerator at the speed.

3. The prediction device according to claim 2, characterized in that, The device includes a standard movement calculation unit that uses a mathematical formula to calculate the standard movement amount, the formula representing the relationship between the speed at which the object to be incinerated is fed into the incinerator and the standard movement amount of the object to be incinerated when it is fed into the incinerator at that speed.

4. The prediction device according to claim 1, characterized in that, The movement data represents the amount of movement of the object to be incinerated. The prediction unit uses the following mathematical formula to calculate the volume density as the prediction result, the mathematical formula representing the relationship between the amount of movement represented by the movement data generated by the movement data generation unit, the speed at which the incineration object is fed into the incinerator, and the volume density of the incineration object.

5. The prediction device according to claim 1, characterized in that, The movement data represents the amount of movement of the object to be incinerated. The prediction unit uses the following mathematical formula to calculate the lower heating value as the prediction result, the mathematical formula representing the relationship between the amount of movement represented by the movement data generated by the movement data generation unit, the speed at which the object to be incinerated is fed into the incinerator, and the lower heating value when the object to be incinerated is burned.

6. A facility control device, comprising: A motion data generation unit generates motion data representing the movement state of an object to be incinerated by feeding a hopper into the incinerator at a specified speed, based on multiple time-series images of the hopper taken from above; and The equipment control unit performs at least one of the following controls based on the movement data generated by the movement data generation unit and the speed: (1) control of feeding the incineration object into the hopper, (2) control of the speed, (3) control of the conveying speed of the incineration object in the incinerator, (4) combustion control of the incineration object in the incinerator, and (5) stirring control of the incineration object.

7. A prediction method, executed by one or more information processing devices, for predicting the properties of an object to be incinerated, comprising the following steps: In an incineration facility that feeds incinerated material into an incinerator at a specified speed via a hopper, movement data representing the movement state of the incinerated material is generated based on multiple time-series images of the hopper taken from above; and Based on the movement data and speed generated in the above steps, the properties of the incinerated object are predicted.

8. A facility control method, executed by one or more information processing devices, comprising the following steps: In an incineration facility that feeds incinerated material into an incinerator at a specified speed via a hopper, movement data representing the movement state of the incinerated material is generated based on multiple time-series images of the hopper taken from above; and Based on the movement data and speed generated in the above steps, at least one of the following controls are performed: (1) control of feeding the incinerator into the hopper, (2) control of the speed, (3) control of the conveying speed of the incinerator in the incinerator, (4) combustion control of the incinerator in the incinerator, and (5) stirring control of the incinerator.

9. A prediction program for enabling a computer to function as the prediction device of claim 1, and for enabling the computer to function as the aforementioned mobile data generation unit and the aforementioned prediction unit.

10. A control program for enabling a computer to function as a facility control device as claimed in claim 6, and for enabling the computer to function as the mobile data generation unit and the equipment control unit.

Images