Information processing device, driving assistance method, driving assistance program, and control system
The information processing device addresses the challenge of analyzing overlapping flocs by calculating index values from images to support plant operations, ensuring efficient and automated chemical adjustments for optimal moisture content in the dehydrated cake.
Patent Information
- Application Number
- JP2021195506
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2021-12-01
- Publication Date
- 2024-11-11
- Estimated Expiration
- 2041-12-01
AI Technical Summary
Existing systems face difficulties in accurately analyzing the condition of flocs in images where multiple flocs are overlapped, leading to challenges in efficiently supporting the operation of plants performing solid-liquid separation, particularly due to issues with imaging devices becoming dirty and requiring manual cleaning.
An information processing device calculates an index value from overlapping floc images to support plant operations, using a system that includes a flocculator, dehydrator, and imaging device to control the addition of chemicals and adjust processing conditions based on the calculated index values, reflecting the formation state and properties of flocs.
Enables stable and automated operation support for plants processing liquids with overlapping flocs, allowing for accurate adjustment of chemical amounts and processing conditions to maintain optimal moisture content in the dehydrated cake.
Smart Images

Figure 0007580365000001 
Figure 0007580365000002 
Figure 0007580365000003
Abstract
Description
[Technical field]
[0001] The present invention relates to an apparatus for supporting the operation of a plant that adds an agent for coagulating suspended solid matter to a liquid to be treated to form flocs, and performs solid-liquid separation of the liquid to be treated in which the flocs have formed. [Background technology]
[0002] A conventional technique involves adding chemicals such as a flocculant to a liquid to be treated such as sewage sludge, stirring the liquid to flocculate suspended solids to form flocs, and then subjecting the flocs, which are an aggregate of flocs, to solid-liquid separation to obtain dewatered suspended solids. In a plant that performs the above-mentioned treatment, in order to perform solid-liquid separation stably, it is necessary to constantly check the state of floc formation and to appropriately adjust the treatment conditions, such as the amount of chemicals added, in response to changes in the state of floc formation, but such checking and adjustments impose a large labor burden.
[0003] For this reason, the development of technologies for supporting the operation of such plants has been underway.For example, the following Patent Document 1 discloses a technology for supporting the operation of a plant by extracting individual floc parts from an image captured inside an aeration tank where flocs are formed, and obtaining a floc aggregation state index from the average brightness value of the extracted individual floc parts. [Prior art documents] [Patent documents]
[0004] [Patent Document 1] Patent Publication No. 06-031292 Summary of the Invention [Problem to be solved by the invention]
[0005] When analyzing images of flocs to understand their state, there is a problem that it is difficult to accurately understand their state if multiple flocs are overlapping in the image. For example, a floc coagulation tank is provided with a viewing window for observing the inside. When the inside of the coagulation tank during floc formation is photographed through this viewing window, the photographed image shows many overlapping flocs, making it difficult to extract individual flocs from the image.
[0006] In this regard, the technology of Patent Document 1 involves immersing an imaging device in the liquid of an aeration tank to capture images of flocs. With this configuration, it is believed possible to capture images that allow individual flocs to be identified by appropriately adjusting the position of the imaging device. However, when the imaging device is immersed, it becomes dirty due to the adhesion of sludge, flocs, etc., and this requires the trouble of cleaning the imaging device.
[0007] An object of one aspect of the present invention is to provide an information processing device or the like that can provide operational support for a plant that treats a liquid to be treated by using an image in which a plurality of overlapping flocs are captured. [Means for solving the problem]
[0008] In order to solve the above problems, an information processing device according to one embodiment of the present invention includes an index value calculation unit that calculates an index value indicating a color tone of the entire image from an image of a plurality of overlapping flocs before solid-liquid separation taken in a plant that adds an agent that coagulates suspended solids to the liquid to be treated in a coagulation tank to form flocs and performs solid-liquid separation of the liquid to be treated in which the flocs have formed, and an operation support unit that provides operation support for the plant based on the index value.
[0009] In order to solve the above problems, an operation support method according to one embodiment of the present invention is a plant operation support method executed by one or more information processing devices, which adds an agent that coagulates suspended solids to a liquid to be treated in a coagulation tank to form flocs, and performs solid-liquid separation of the liquid to be treated in which the flocs have been formed, and includes an index value calculation step of calculating an index value indicating a color tone of an entire image taken at the plant, from an image in which a plurality of the flocs are overlapped and before solid-liquid separation, and an operation support step of providing operation support for the plant based on the index value.
[0010] In order to solve the above problems, a control system according to one embodiment of the present invention includes an apparatus for adding an agent that coagulates suspended solids to the liquid to be treated in a coagulation tank to form flocs, an apparatus for performing solid-liquid separation of the liquid to be treated in which the flocs have been formed, an imaging device for photographing the flocs formed in the coagulation tank, and an information processing device for calculating an index value indicating the color tone of the entire image from an image captured by the imaging device showing a plurality of overlapping flocs before solid-liquid separation, and for controlling at least any of the apparatus based on the calculated index value. Effect of the Invention
[0011] According to one aspect of the present invention, it is possible to provide operational support for a plant that treats a liquid to be treated by using an image in which a plurality of overlapping flocs are captured. [Brief description of the drawings]
[0012] [Figure 1] 1 is a block diagram showing an example of a configuration of a main part of an information processing device according to a first embodiment of the present invention. [Diagram 2] FIG. 2 is a diagram illustrating an example of the configuration of a control system including the information processing device. [Diagram 3] FIG. 13 is a diagram illustrating the relationship between the average brightness value of a floc image and the moisture content of a dehydrated cake. [Figure 4] 4 is a flowchart illustrating an example of a process executed by the information processing device. [Diagram 5]FIG. 11 is a block diagram showing an example of a configuration of a main part of an information processing device according to a second embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0013] [Embodiment 1] (System Configuration) The configuration of a control system according to one embodiment of the present invention will be described with reference to Fig. 2. Fig. 2 is a diagram showing an example of the configuration of a control system 100. The control system 100 is a system used in a plant where a chemical agent for flocculating suspended solid matter is added to a liquid to be treated in a coagulation tank to form flocs, and solid-liquid separation of the liquid to be treated in which the flocs have been formed is performed. In the following, an example in which the liquid to be treated is sludge will be described, but the control system 100 can also be applied to a plant that treats liquid to be treated other than sludge. Incidentally, sludge is a liquid containing fine solid matter generated in wastewater treatment and the like, and can also be called a slurry.
[0014] As will be described in detail below, the control system 100 performs each process of the sludge treatment process from a process of converting the sludge to be treated into flocculated sludge by flocculating solid suspended matter in the sludge to be treated to form flocs, to a process of dewatering the flocculated sludge to obtain dehydrated sludge (also called dehydrated cake) and dehydrated filtrate. As shown in Fig. 2, the control system 100 includes an information processing device 1, a control device 3, a flocculator 5, and a dehydrator 9.
[0015] The flocculator 5 is a device that adds a chemical agent for flocculating suspended solids to the liquid to be treated in a coagulation tank and moderately stirs the liquid to form flocs. Specifically, the flocculator 5 uses sludge as the liquid to be treated, coagulates suspended solids in the sludge to form flocs, and produces coagulated sludge. The flocculator 5 in FIG. 2 includes a coagulation tank 51, a stirring blade 52, a motor 53, and an inspection window 54. The flocculator 5 also includes a sludge inlet 55, a chemical inlet 56, and a discharge outlet 57.
[0016] Furthermore, a photographing device 72 and a lighting device 71 for photographing are attached to the inspection window 54. The photographing device 72 may be any device capable of at least taking still images. It is preferable that the coagulation tank 51 is opaque so that the way light hits the flocs does not change while the control system 100 is in operation. Moreover, it is preferable that the photographing device 72 and the lighting device 71 are housed in a light-shielding dark box with an opening on the inspection window 54 side, as in the illustrated example.
[0017] The dehydrator 9 is a device for performing solid-liquid separation of the liquid to be treated in which flocs are formed. Specifically, the dehydrator 9 is disposed after the flocculator 5, and performs solid-liquid separation by dehydrating the coagulated sludge discharged from the flocculator 5. The dehydrator 9 in FIG. 2 is a screw press type dehydrator including an outer body screen 91 and a screw 92. The dehydrator 9 is also provided with a sludge inlet 93, a filtrate outlet 94, and a dehydrated cake outlet 95. Although not shown, the dehydrator 9 also includes a motor for rotating the screw 92. Of course, the dehydrator 9 is not limited to the screw press type as long as it can dehydrate the coagulated sludge. For example, a centrifugal dehydrator, a filter press type dehydrator, a belt press dehydrator, or the like can be applied.
[0018] In the control system 100, the sludge to be treated is continuously or intermittently supplied from a sludge inlet 55 into the coagulation tank 51 of the flocculator 5 by a supply device (not shown). The supply rate of the sludge may be automatically controlled by the supply device or its control device 3 according to the treatment rate of the sludge by the flocculator 5 and the dehydrator 9.
[0019] Then, chemicals (including at least a flocculant) for flocculating the sludge are fed into the sludge in the flocculation tank 51 through chemical inlet 56. In this state, motor 53 is driven to rotate agitator blade 52, which agitates the sludge and chemicals and forms flocs. The flocculated sludge, which is a mixture of the formed flocs and the water contained in the sludge, is discharged from outlet 57.
[0020] Next, this flocculated sludge is supplied into the outer body screen 91 from the sludge inlet 93 of the dewatering machine 9. In the dewatering machine 9, the flocculated sludge is dewatered under pressure by the screw 92, and the filtrate is discharged from the filtrate outlet 94, while the dewatered cake, which is a mass of the dewatered flocculated sludge, is discharged from the dewatered cake outlet 95.
[0021] The information processing device 1 calculates an index value indicating the color tone of the entire image from an image (hereinafter referred to as a floc image) of multiple overlapping flocs before solid-liquid separation captured by the image capture device 72. Then, based on the calculated index value, the information processing device 1 controls at least one of the flocculator 5 and the dewatering machine 9, which are devices that add a chemical agent for flocculating suspended solids to the liquid to be treated in the coagulation tank to form flocs. Note that the control range of these devices includes the supply amount from a sludge / chemical supply device (not shown).
[0022] The information processing device 1 may directly perform the above control on the flocculator 5 and the dehydrator 9, but in this embodiment, an example will be described in which the operation of these devices is controlled via the control device 3. The control device 3 is a device that controls the operation of at least one of the flocculator 5 and the dehydrator 9. The control device 3 may be, for example, a PLC (Programmable Logic Controller).
[0023] As described above, the control system 100 includes the flocculator 5, which is an apparatus for forming flocs by adding an agent that flocculates suspended solids to the liquid being treated in a coagulation tank, the dehydrator 9, which is an apparatus for performing solid-liquid separation of the liquid being treated in which flocs have been formed, the photographing device 72 that photographs the flocs formed in the coagulation tank 51, and the information processing device 1 that calculates an index value that indicates the color tone of the entire floc image from the floc image photographed by the photographing device 72, in which multiple flocs are overlapping each other before solid-liquid separation, and controls at least one of the flocculator 5 and the dehydrator 9 based on the calculated index value.
[0024] The color of the flocks reflects the state of formation and properties of the flocks. In addition, in a flock image in which a plurality of flocks are overlapped, the color of the entire flock image reflects the state of formation and properties of the flocks. Therefore, according to the above configuration, it is possible to provide driving assistance according to the state of formation and properties of the flocks using a flock image in which a plurality of flocks are overlapped.
[0025] As described above, in this embodiment, an example will be described in which the liquid to be treated is sludge. This sludge may be, for example, mixed raw sludge, which is a combination of primary settling sludge and final settling sludge (excess sludge) generated in a sewage treatment plant. It has been found through experiments by the inventors of the present invention that when a chemical is added to mixed raw sludge generated in a sewage treatment plant to form flocs, the color of the flocs reflects the state of formation and properties of the flocs. Therefore, if the liquid to be treated is mixed raw sludge, it is possible to appropriately treat the mixed raw sludge by providing operation support according to the state of formation and properties of the flocs generated from the mixed raw sludge.
[0026] (Device configuration) The configuration of the information processing device 1 will be described with reference to Fig. 1. Fig. 1 is a block diagram showing an example of a main configuration of the information processing device 1. As shown in the figure, the information processing device 1 includes a control unit 10 that controls each unit of the information processing device 1, and a storage unit 11 that stores various data used by the information processing device 1. The information processing device 1 also includes a communication unit 12 that allows the information processing device 1 to communicate with other devices, an input unit 13 that accepts input of various data to the information processing device 1, and an output unit 14 that allows the information processing device 1 to output various data.
[0027] The control unit 10 also includes a data acquisition unit 101, an index value calculation unit 102, a driving support unit 103, and a learning unit 104. The memory unit 11 stores teacher data 111 and a trained model 112. Details of the learning unit 104, teacher data 111, and trained model 112 will be described later in "Method for estimating moisture content."
[0028] The data acquiring unit 101 acquires an image of a plurality of overlapping flocs before solid-liquid separation, which is taken in a plant where a chemical agent for flocculating suspended solids is added to the liquid to be treated in a coagulation tank to form flocs, and where solid-liquid separation of the liquid to be treated in which the flocs have formed is performed. Specifically, the data acquiring unit 101 acquires the above-mentioned floc image (an image of the sludge in the coagulation tank 51 taken by the photographing device 72). The floc image may be input via the input unit 13, or may be received via the communication unit 12. The data acquiring unit 101 may also transmit a photographing trigger to the photographing device 72 directly or via another device such as the control device 3 to have the floc image photographed.
[0029] The index value calculation unit 102 calculates an index value indicating the color tone of the entire flock image from the flock image acquired by the data acquisition unit 101. Here, color tone is a concept that encompasses hue, saturation, and brightness. In other words, the index value calculated by the index value calculation unit 102 may indicate the hue, saturation, or brightness of the entire flock image, or may indicate a combination of these. Details of the index value calculation method will be explained later in "Regarding Index Values."
[0030] The operation support unit 103 supports the operation of various devices in the control system 100, particularly at least one of the flocculator 5 and the dehydrator 9, based on the index value calculated by the index value calculation unit 102. As an example, the operation support unit 103 estimates the moisture content of the dehydrated cake based on the index value. Then, the operation support unit 103 adjusts the amount of chemicals to be added to the flocculator 5 based on the estimated moisture content of the dehydrated cake.
[0031] As described above, the information processing device 1 includes the index value calculation unit 102 that calculates an index value indicating the color tone of the entire floc image, and the operation support unit 103 that supports the operation of the plant based on the index value. Therefore, it is possible to support the operation of a plant that treats sludge by using a floc image in which multiple flocs are overlapped.
[0032] (Regarding index values) The index value calculation unit 102 may calculate, for example, an average brightness value of the entire flock image as an index value indicating the color tone of the entire flock image. This will be described with reference to Fig. 3. Fig. 3 is a diagram illustrating the relationship between the average brightness value of the flock image and the moisture content of the dehydrated cake.
[0033] FIG. 3 shows three floc images a1 to a3 obtained by photographing flocs with different moisture contents. The average brightness value of the entire floc image a1 is 76, the average brightness value of the entire floc image a2 is 97, and the average brightness value of the entire floc image a3 is 114. Thus, the difference in moisture content of the photographed flocs appears as a difference in the average brightness value. Note that these average brightness values are calculated by adding up all the brightness values of the pixels constituting the floc image and dividing by the number of pixels in the floc image. The floc image may be a color image or a grayscale image.
[0034] Graph b1 in FIG. 3 is a graph showing the relationship between the average brightness value of the floc image and the moisture content of the dehydrated cake when the operating conditions of the flocculator 5 and the dehydrator 9 are approximately the same. As shown in graph b1, as the average brightness value of the floc image increases, the moisture content of the dehydrated cake increases. In this way, the average brightness value of the floc image is correlated with the moisture content of the dehydrated cake. In addition, since the moisture content of the dehydrated cake is determined by the formation state and properties of the flocs, it can be said that the average brightness value of the floc image reflects the formation state and properties of the flocs. Therefore, the average brightness value of the floc image can be used as an index value for predicting the moisture content of the dehydrated cake, and can also be used as an index value for predicting the formation state and properties of the flocs.
[0035] Graph b2 in Fig. 3 is a graph showing the relationship between the average area of the background small regions in the floc image and the moisture content of the dehydrated cake when the operating conditions of the flocculator 5 and the dehydrator 9 are approximately the same. The background small regions are multiple continuous regions that constitute the background region in the floc image where the light from the lighting device 7 does not reach the flocs. In the floc images a1 to a3 shown in Fig. 3, the black areas between the overlapping flocs are the background small regions, in contrast to the flocs that appear whitish as they are illuminated by the lighting device 71.
[0036] As shown in graph b2, as the average area of the background small regions increases, the moisture content of the dehydrated cake increases. In this way, the average area of the background small regions is also correlated with the moisture content of the dehydrated cake. In addition, since the moisture content of the dehydrated cake is determined by the state of floc formation and properties, it can be said that the average area of the background small regions also reflects the state of floc formation and properties. Therefore, the average area of the background small regions can be used as an index value for predicting the moisture content of the dehydrated cake, as well as an index value for predicting the state of floc formation and properties.
[0037] In this way, it was found that in a floc image in which multiple flocs overlap each other, the formation state and properties of the flocs are reflected in the average brightness value of the entire floc image, including the flocs and their background. Therefore, the index value calculation unit 102 can calculate an appropriate index value that reflects the formation state and properties of the flocs through simple calculations by calculating the average brightness value of the entire floc image as the index value.
[0038] It can also be said that the block images a1 to a3 in Fig. 3 have different overall color hues. Since the above-mentioned average luminance value can be said to indicate the brightness of the entire image, the configuration for calculating the average luminance value as an index value can be said to calculate an index value indicating brightness by regarding differences in color as differences in brightness. Of course, various index values indicating brightness can be applied other than the average luminance value. For example, the index value calculation unit 102 may calculate a median value of the luminance values instead of the average luminance value.
[0039] Furthermore, when the flock image is a color image, the index value calculation unit 102 may calculate an index value indicating the hue of the flock image. For example, the index value calculation unit 102 may calculate an average value of the RGB (Red, Green, Blue) values of each pixel constituting the flock image as an index value indicating the hue of the flock image. Similarly, the index value calculation unit 102 may calculate an index value indicating the saturation of the flock image.
[0040] (About driving assistance) The operation support unit 103 may support the operation of the control system 100 by estimating the moisture content of the liquid to be treated after solid-liquid separation based on the index value calculated by the index value calculation unit 102. When the liquid to be treated is sludge, the moisture content of the liquid to be treated after solid-liquid separation is the moisture content of the dehydrated cake, so that the "moisture content of the dehydrated cake" in the following description can be read as the "moisture content of the liquid to be treated after solid-liquid separation".
[0041] The moisture content of the dehydrated cake is an important index for proper sludge treatment. Therefore, according to the above configuration in which the moisture content of the dehydrated cake is estimated based on the index value, it becomes possible to provide proper operational support for the plant. Note that the estimation result may be used in any manner for operational support. For example, the estimation result may be output to the output unit 14 to present it to a user for operational support, or various devices in the plant may be controlled in accordance with the estimation result for operational support.
[0042] Regarding the above control, for example, the operation support unit 103 may at least adjust the amount of chemicals added to the sludge and / or adjust the operating conditions of the equipment for treating the sludge based on the estimated moisture content so that the moisture content of the dehydrated cake falls within a predetermined range. According to this configuration, the control for keeping the moisture content of the dehydrated cake within the predetermined range is automatically performed, so that the plant can be automatically operated stably. The predetermined range can also be rephrased as a normal range.
[0043] Regarding adjustment of the operating conditions of the equipment treating the sludge, for example, the operation support unit 103 may control at least any of the amount of sludge supplied to the sludge inlet 55, the rotation speed of the screw 92 of the dehydrator 9, the feeding pressure of the sludge to the sludge inlet, the amount of chemicals supplied to the flocculator 5, and the rotation speed of the agitator blades 52 of the flocculator 5. Since all of these affect the moisture content of the dehydrated cake, it is possible to keep the moisture content within a predetermined range by adjusting these. The operation support unit 103 may determine what kind of control to apply to which equipment depending on the degree of deviation between the estimated moisture content and the range of the target moisture content, etc.
[0044] (Method of estimating moisture content) The operation support unit 103 may estimate the moisture content of the dehydrated cake by a statistical method based on the index value calculated by the index value calculation unit 102. As an example, the operation support unit 103 may estimate the moisture content of the dehydrated cake based on an output value obtained by inputting the index value calculated by the index value calculation unit 102 into the trained model 112.
[0045] Trained model 112 is a model for estimating the moisture content of a dehydrated cake, constructed by machine learning the relationship between a plurality of explanatory variables including at least the index value calculated by index value calculation unit 102 and a target variable indicating the moisture content of the dehydrated cake. By using such trained model 112, it becomes possible to estimate the moisture content with high accuracy.
[0046] The algorithm of the trained model 112 is not particularly limited, and for example, a trained model 112 generated by multiple regression analysis may be used, or the trained model 112 may be generated by random forest, neural network, gradient boosting, etc.
[0047] Trained model 112 may be generated by learning unit 104 through machine learning using training data 111. Training data 111 is data indicating a relationship between a plurality of explanatory variables including at least an index value calculated by index value calculation unit 102 and a target variable indicating the moisture content of a dehydrated cake. For example, training data 111 may be data in which the moisture content of a dehydrated cake obtained during past operation of a plant to which control system 100 is applied is associated with an index value calculated from a floc image of a sludge floc that is the source of the dehydrated cake. In addition to the index value, various explanatory variables correlated with the moisture content of the dehydrated cake may be associated with the moisture content of the dehydrated cake.
[0048] For example, the explanatory variables may include at least one of (1) the average unit area of a plurality of small regions constituting the background of the flocs in the floc image, (2) the amount of sludge supplied to the coagulation tank 51, (3) the amount of chemical added, (4) the concentration of sludge supplied to the coagulation tank 51, and (5) the mixing speed at which the sludge and the chemical are mixed. In addition, in the case of a plant using a screw press type dehydrator such as the dehydrator 9, the explanatory variables may further include at least one of (6) the rotation speed of the screw and (7) the input pressure at the time of squeezing by the screw. Note that the above (1) is synonymous with the "average area of the background small regions" explained based on FIG. 3.
[0049] As a result of experiments conducted by the inventors of the present invention, it has been found that the state of formation and properties of flocs are reflected in the values of (1) to (7) above. Therefore, by using trained model 112 that uses the above (1) to (7) as explanatory variables together with the index value calculated by index value calculation unit 102, it becomes possible to estimate the moisture content of the dehydrated cake with high accuracy.
[0050] Note that the learning unit 104 is not a required component of the information processing device 1, and for example, the information processing device 1 may obtain a trained model 112 generated by another device having the function of the learning unit 104 and perform processing using the trained model 112. An advantage of providing the information processing device 1 with the learning unit 104 is that the information processing device 1 can re-learn the trained model 112 by itself.
[0051] For example, the user of the information processing device 1 may measure the actual moisture content after the driving assistance unit 103 inputs various explanatory variables into the trained model 112 to estimate the moisture content. Then, the user may add training data in which the measured moisture content is associated with the input explanatory variables to the training data 111, and cause the learning unit 104 to perform re-learning. This makes it possible to maintain or improve the estimation accuracy of the trained model 112.
[0052] (Processing flow) The flow of the process (driving support method) executed by the information processing device 1 will be described with reference to Fig. 4. Fig. 4 is a flowchart showing an example of the process executed by the information processing device 1.
[0053] In S11, the data acquiring unit 101 acquires a floc image obtained by photographing the inside of the coagulation tank 51 of the flocculator 5 with the photographing device 72. The data acquiring unit 101 outputs the acquired floc image to the index value calculation unit 102. As described above, the data acquiring unit 101 may transmit a photographing trigger to the photographing device 72 to photograph the floc image. For example, the data acquiring unit 101 may transmit the photographing trigger at the timing of the impeller angle at which the flow of flocs by the agitating impeller 52 is the gentlest.
[0054] In S12 (index value calculation step), the index value calculation unit 102 calculates an index value indicating the color tone of the entire flock image from the flock image acquired by the data acquisition unit 101. As an example, the index value calculation unit 102 calculates an average brightness value of the entire image as the index value. The index value calculation unit 102 outputs the calculated index value to the driving support unit 103.
[0055] In S13, the driving support unit 103 estimates the moisture content of the dehydrated cake based on the index value calculated in S12. For example, the driving support unit 103 may estimate the moisture content of the dehydrated cake using the trained model 112. In this case, the driving support unit 103 also acquires other explanatory variables to be input to the trained model 112, and inputs them to the trained model 112 together with the index value calculated in S12.
[0056] In S14, the driving support unit 103 determines whether driving support is necessary or not based on the estimation result of S13. As an example, when the estimated moisture content of the dehydrated cake is not within a predetermined range, the driving support unit 103 determines that driving support is necessary (Yes in S14) and executes the process of S15. On the other hand, when the estimated moisture content of the dehydrated cake is within the predetermined range, the driving support unit 103 determines that driving support is not necessary (No in S14) and ends the process.
[0057] In S15 (operation support step), the operation support unit 103 performs operation support for the plant based on the index value calculated in S12. For example, the operation support unit 103 may perform operation support so that the moisture content of the dehydrated cake estimated using the index value calculated in S12 falls within a predetermined range. As an example, the operation support unit 103 may adjust the amount of chemical added to the sludge by controlling a device that supplies chemicals to the coagulation tank 51 so that the estimated moisture content of the dehydrated cake falls within the predetermined range. This control may be performed directly on the device or via the control device 3. When the process of S15 ends, the process of FIG. 4 ends.
[0058] As described above, the operation support method according to this embodiment is a plant operation support method executed by the information processing device 1, in which an agent for coagulating suspended solids is added to sludge in the coagulation tank 51 to form flocs, and the sludge in which the flocs have been formed is subjected to solid-liquid separation. This operation support method includes an index value calculation step (S12) of calculating an index value indicating the overall color tone of a floc image from a floc image taken in the plant, and an operation support step (S15) of supporting the operation of the plant based on the index value calculated in S12. According to this operation support method, it is possible to support the operation of the plant using an image in which a plurality of flocs are overlapping each other.
[0059] [Embodiment 2] Other embodiments of the present invention will be described below. For ease of explanation, the same reference numerals are given to members having the same functions as those described in the above embodiment, and the description thereof will not be repeated.
[0060] (Device configuration) The configuration of the information processing device 1A according to this embodiment will be described with reference to Fig. 5. Fig. 5 is a block diagram showing the configuration of the information processing device 1A. The information processing device 1A differs from the information processing device 1 shown in Fig. 1 in that it includes a control unit 10A including a driving support unit 103A and a learning unit 104A instead of the control unit 10 including the driving support unit 103 and the learning unit 104. The information processing device 1A also differs from the information processing device 1 in that it includes a storage unit 11A that stores teacher data 111A and a learned model 112A instead of the storage unit 11 that stores teacher data 111 and a learned model 112.
[0061] Similar to the operation support unit 103, the operation support unit 103A supports the operation of the plant based on the index value calculated by the index value calculation unit 102. More specifically, the operation support unit 103A supports the operation of the control system 100 by adjusting the amount of chemical agent added to the coagulation tank 51 based on the index value calculated by the index value calculation unit 102. As an example, the operation support unit 103A adjusts the amount of chemical agent added to the flocculator 5 of the control system 100 based on the average brightness value of the entire floc image calculated by the index value calculation unit 102.
[0062] The state of formation and properties of flocs are reflected in the index value calculated by the index value calculation unit 102. Therefore, the operation support unit 103A can automatically adjust the amount of chemical agent added according to the state of formation and properties of flocs by adjusting the amount of chemical agent added based on the index value.
[0063] For example, the driving support unit 103A may adjust the amount of medicine to be added based on an output value obtained by inputting the index value calculated by the index value calculation unit 102 to the trained model 112A. The trained model 112A is a model for estimating an appropriate amount of medicine to be added, which is constructed by machine learning the relationship between a plurality of explanatory variables including at least the index value calculated by the index value calculation unit 102 and a target variable indicating an appropriate amount of medicine to be added. By using such trained model 112A, it becomes possible to appropriately adjust the amount of medicine to be added. Note that, like the trained model 112, the algorithm of the trained model 112A is not particularly limited.
[0064] The trained model 112A may be generated by the learning unit 104A through machine learning using the teacher data 111A. The teacher data 111A is data showing a relationship between a plurality of explanatory variables including at least an index value calculated by the index value calculation unit 102 and a target variable showing an appropriate amount of the chemical to be added. For example, the teacher data 111A may be data that associates the amount of the chemical to be added during a period in which the moisture content of the dehydrated cake was within an appropriate range during past operation of the plant to which the control system 100 is applied with an index value calculated from a floc image of the sludge floc that was the source of the dehydrated cake. In addition to the index value, various explanatory variables that are correlated with the appropriate amount of the chemical to be added may be associated with the amount of the chemical to be added.
[0065] For example, the explanatory variables may include at least one of (1) the average unit area of multiple small regions constituting the background of the floc in the floc image, (2) the amount of sludge supplied to the coagulation tank 51, (3) the amount of chemical added, (4) the concentration of sludge supplied to the coagulation tank 51, and (5) the mixing speed at which the sludge and the chemical are mixed, as in the trained model 112. In addition, in the case of a plant using a screw press type dehydrator such as the dehydrator 9, the explanatory variables may further include at least one of (6) the rotation speed of the screw and (7) the input pressure at the time of squeezing by the screw. Furthermore, the explanatory variables may include the moisture content of the dehydrated cake.
[0066] The objective variable for the explanatory variables as described above is not limited to the amount of chemicals added. For example, at least one of an appropriate amount of sludge supplied to the sludge inlet 55, an appropriate rotation speed of the screw 92 of the dehydrator 9, an appropriate pressure for feeding sludge to the sludge inlet, and an appropriate rotation speed of the agitator blades 52 of the flocculator 5 may be used as the objective variable. In this way, the configuration of this embodiment can be applied to determining any control content for any device that affects the normality of flocs.
[0067] (Processing flow) The flow of the process (driving support method) executed by the information processing device 1A is the same as that of the driving support method of the first embodiment (see FIG. 4). The main difference is that in the driving support method of the present embodiment, instead of estimating the moisture content in S13, an appropriate amount of the chemical to be added is estimated. In addition, in the driving support method of the present embodiment, in S14, the necessity of driving support is determined based on the estimation result in S13. For example, the driving support unit 103A may determine that driving support is necessary when the difference between the current amount of addition and the amount of addition estimated in S13 is equal to or greater than a threshold, and may determine that driving support is unnecessary when the difference is less than the threshold. Then, in S15, the driving support unit 103A controls a device that supplies chemicals to the flocculation tank 51 so as to add the amount of chemicals estimated in S13.
[0068] [Modifications] The operation support in each of the above-mentioned embodiments can be combined with other processes for supporting plant operation. For example, the information processing device 1 or 1A may perform a process of calculating the particle size of the flocs by analyzing a floc image and rotating the agitator blades 52 of the flocculator 5 at a speed corresponding to the particle size, and perform the above-mentioned operation support. This makes it possible to obtain a dehydrated cake with an appropriate moisture content by the operation support unit 103 or 103A, even in cases where a dehydrated cake with an appropriate moisture content cannot be obtained by only adjusting the rotation speed of the agitator blades 52.
[0069] It is assumed that it may be difficult to calculate the particle size of each floc from a floc image in which multiple flocs overlap each other. In this case, the average area of the background small regions described above may be used as an index value indicating the particle size of each floc. The average area of the background small regions correlates with the particle size of each floc, and therefore can be used as an index value indicating the particle size of each floc.
[0070] In addition, in each of the above embodiments, the information processing device 1 and the control device 3 are described as independent devices, but they may be integrated into one or two devices. Also, a part of the processing executed by each of these devices may be executed by another information processing device other than these devices. In this way, the executing entity of each process described in each of the above embodiments can be changed as appropriate, and the control system 100 according to the present invention can be realized in various system configurations. For example, each process of the driving support method shown in FIG. 4 can be executed in a form in which it is shared among a plurality of information processing devices.
[0071] [Software implementation example] The functions of the information processing devices 1 and 1A (hereinafter referred to as the "devices") can be realized by a program for causing a computer to function as the device, and a program (driving assistance program) for causing a computer to function as each control block of the device (particularly each part included in the control unit 10 or 10A).
[0072] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the program. The control device and storage device execute the program to realize each function described in each of the above embodiments.
[0073] The program may be non-transitory and may be recorded in one or more computer-readable recording media. The recording media may or may not be included in the device. In the latter case, the program may be provided to the device via any wired or wireless transmission medium.
[0074] In addition, some or all of the functions of each of the control blocks can be realized by a logic circuit. For example, an integrated circuit in which a logic circuit that functions as each of the control blocks is formed is also included in the scope of the present invention. In addition, the functions of each of the control blocks can be realized by, for example, a quantum computer.
[0075] The present invention is not limited to the above-described embodiments, and various modifications are possible 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 the present invention. [Explanation of symbols]
[0076] 1, 1A Information processing device 5 Flocculator (equipment) 9 Dehydrator (equipment) 72 Imaging Equipment 100 Control System 102 Index value calculation unit 103, 103A Driving Support Department 112, 112A Trained Model
Claims
1. an index value calculation unit that calculates an index value indicating the overall color tone of an image in which a plurality of overlapping flocs are captured before solid-liquid separation in a plant that adds an agent for flocculating suspended solids to a liquid to be treated in a flocculation tank to form flocs and performs solid-liquid separation of the liquid to be treated in which the flocs have formed; an operation support unit that supports the operation of the plant by at least one of adjusting the amount of the chemical to be added based on the index value, adjusting the operating conditions of an apparatus that treats the liquid to be treated based on the index value, and outputting a water content of the liquid to be treated after solid-liquid separation that is estimated from the index value; The index value calculation unit calculates an average luminance value of the entire image or a median luminance value of the entire image as the index value indicating a color tone of the entire image.
2. The information processing device described in claim 1, wherein the operation support unit supports the operation of the plant by at least one of estimating the moisture content of the treated liquid after solid-liquid separation based on the index value, adjusting the amount of the chemical to be added based on the estimated moisture content, adjusting the operating conditions of equipment that treats the treated liquid based on the estimated moisture content, and outputting the estimated moisture content.
3. The information processing device described in claim 2, wherein the operation support unit adjusts at least one of the amount of chemical added and the operating conditions of the equipment treating the treated liquid so that the moisture content of the treated liquid after solid-liquid separation is within a predetermined range based on the estimated moisture content.
4. The information processing device according to claim 1, wherein the driving assistance unit adjusts the amount of the drug to be added based on an output value obtained by inputting the index value calculated by the index value calculation unit into a trained model constructed by machine learning the relationship between a plurality of explanatory variables including at least the index value and a target variable indicating an appropriate amount of the drug to be added.
5. The information processing device described in claim 2 or 3, wherein the operation support unit estimates the moisture content of the treated liquid after solid-liquid separation based on the output value obtained by inputting the index value calculated by the index value calculation unit into a learned model constructed by machine learning the relationship between a plurality of explanatory variables including at least the index value and a target variable indicating the moisture content of the treated liquid after solid-liquid separation.
6. 6. The information processing device according to claim 4 or 5, wherein explanatory variables of the trained model include at least any of: (1) an average unit area of a plurality of small regions constituting a background of flocs in the image; (2) an amount of the liquid to be treated supplied to the coagulation tank; (3) an amount of the chemical to be added; (4) a concentration of the liquid to be treated supplied to the coagulation tank; and (5) a stirring speed at which the liquid to be treated and the chemical are stirred.
7. The information processing apparatus according to claim 1 , wherein the liquid to be treated is mixed raw sludge generated in a sewage treatment plant.
8. 1. A plant operation support method executed by one or more information processing devices, comprising: adding an agent for coagulating suspended solids to a liquid to be treated in a coagulation tank to form flocs, and performing solid-liquid separation of the liquid to be treated in which the flocs have formed, the method comprising: an index value calculation step of calculating an index value indicating a color tone of an entire image taken in the plant, the image showing a plurality of overlapping flocs before solid-liquid separation; and an operation support step of supporting the operation of the plant by at least one of adjusting the amount of the chemical to be added based on the index value, adjusting the operating conditions of an apparatus for treating the liquid to be treated based on the index value, and outputting a water content of the liquid to be treated after solid-liquid separation estimated from the index value, In the index value calculation step, an average luminance value of the entire image or a median luminance value of the entire image is calculated as the index value indicating a color tone of the entire image.
9. A driving assistance program for causing a computer to function as the information processing device according to claim 1 , the driving assistance program causing a computer to function as the index value calculation unit and the driving assistance unit.
10. An apparatus for adding a chemical agent for flocculating suspended solids to the liquid to be treated in the flocculation tank to form flocs; an apparatus for performing solid-liquid separation of the liquid to be treated in which flocs have been formed; an imaging device for imaging the flocs formed in the coagulation tank; an information processing device that calculates an index value indicating a color tone of an entire image from an image of the plurality of overlapping flocks before solid-liquid separation captured by the photographing device, and controls at least any of the devices based on the calculated index value; The information processing device calculates an average luminance value of the entire image or a median luminance value of the entire image as the index value indicating a color tone of the entire image.
Citation Information
Patent Citations
Device for controlling sewage treatment process
JP1989034497A
Flock flocculation state monitor
JP1994031292A
Sludge flocculating device
JP1996197100A
Sludge treatment apparatus and sludge treatment method
JP2011189321A
Dewatering system
JP2021037508A
