Sludge treatment support system

Abstract

This sludge treatment support system continuously displays the appropriateness of the amount of coagulant injected, as well as the degree of coagulation of the coagulated sludge, thereby assisting the operation of the dewatering machine by the operator of the sludge dewatering system. [Solution] The sludge treatment support system of the embodiment is a sludge treatment support system that is added to a sludge dewatering system that adds a coagulant to sewage sludge, stirs and mixes it to form coagulated sludge, and dewaters the coagulated sludge in a dewatering machine to form dewatered sludge, and comprises an imaging unit and a coagulation degree determination device. The imaging unit is an imaging unit for the coagulated sludge. The coagulation degree determination device classifies the coagulated sludge image captured by the imaging unit and outputs the degree of appropriateness of the amount of coagulant added, the degree of size of the flocs in the coagulated sludge, and the degree of clarity of the filtrate other than the flocs in the coagulated sludge to an external terminal along with the coagulated sludge image.

Description

Technical Field

[0005] , ,

[0001] Embodiments of the present invention relate to a sludge treatment support system.

Background Art

[0002] In sewage treatment plants, excess sludge is generated during sewage treatment. This excess sludge is disposed of as waste outside the sewage treatment plant. At that time, for the purpose of reducing the disposal cost of the waste, the sludge is volume-reduced by dehydration treatment.

[0003] The sludge dewatering system of a sewage treatment plant inputs the excess sludge (hereinafter referred to as sewage sludge) of the sewage treatment plant into a coagulation tank, adds a coagulant, and stirs and mixes it with a stirrer to generate coagulated sludge. In the coagulated sludge, the suspended substances in the sewage sludge aggregate and coarsen to form flocs. Other than the flocs in the coagulated sludge, the suspended substances decrease and it becomes a liquid component with a reduced sludge concentration. The coagulated sludge is in a state where the coarsened flocs and the liquid with a reduced turbidity are mixed. The sludge dewatering system separates the coagulated sludge into dehydrated sludge with a moisture content of 50 to 90% and a liquid component removed from the coagulated sludge (hereinafter referred to as filtrate) by dehydration treatment with a dehydrator. Most of the flocs of the coagulated sludge migrate to the dehydrated sludge, and most of the liquid component of the coagulated sludge migrates to the filtrate.

[0004] The degree of coagulation of the coagulated sludge is affected by the injection amount of the coagulant with respect to the sewage sludge. The appropriate injection amount of the coagulant is determined by performing a coagulation and dehydration beaker test in which the coagulant is injected at a plurality of injection amounts with respect to the target sewage sludge in advance, and based on the degree of coagulation of the coagulated sludge. The criteria for judging the degree of coagulation include the size of the flocs of the coagulated sludge and the clarity of the filtrate (or something equivalent thereto). As an example of the criteria for judging the degree of coagulation, Non-Patent Document 1 is disclosed. The operation manager of the sludge dewatering system operates the dehydrator with the determined coagulant injection amount.

Prior Art Documents

Patent Documents

[0005] [Patent Document 1] Japanese Patent Publication No. 2024-51134 [Patent Document 2] Japanese Patent Publication No. 2023-163691 [Non-patent literature]

[0006] [Non-Patent Document 1] Tokyo Metropolitan Government Bureau of Sewerage Technical Survey Annual Report, Vol. 2006, pp. 185-199, 2007 [Overview of the Initiative] [Problems that the invention aims to solve]

[0007] However, the properties of sewage sludge vary depending on the properties of the sewage flowing into the sewage treatment plant and the operating conditions of the sewage treatment plant. Therefore, the appropriate amount of coagulant to inject is also expected to change. On the other hand, the coagulation and dewatering beaker test to determine the appropriate amount of coagulant to inject is time-consuming and labor-intensive, making it difficult to keep up with and respond to fluctuations in the properties of sewage sludge.

[0008] Furthermore, if a coagulation and dewatering beaker test is not performed, changes in the appropriate amount of coagulant injected can only be inferred from changes in the water content of the dewatered sludge and the turbidity of the filtrate after dewatering treatment. It is difficult to determine changes in the size of the flocs in the coagulated sludge or the clarity of the filtrate.

[0009] This invention was made to solve these problems and provides a sludge treatment support system that continuously indicates the appropriateness of the amount of coagulant injected, as well as the degree of coagulation of the coagulated sludge, thereby supporting the operation of the dewatering machine by the operator of the sludge dewatering system. [Means for solving the problem]

[0010] The sludge treatment support system of this embodiment is a sludge treatment support system that is added to a sludge dewatering system that adds a coagulant to sewage sludge, mixes it with a stirring to form coagulated sludge, and dewaters the coagulated sludge in a dewatering machine to form dewatered sludge, and comprises an imaging unit and a coagulation degree discrimination device. The imaging unit is an imaging unit for the coagulated sludge. The coagulation degree discrimination device classifies the coagulated sludge image captured by the imaging unit and outputs the degree of appropriateness of the amount of coagulant added, the degree of size of the flocs in the coagulated sludge, and the degree of clarity of the filtrate other than the flocs in the coagulated sludge to an external terminal along with the coagulated sludge image. [Brief explanation of the drawing]

[0011] [Figure 1] Figure 1 is a diagram showing an overview of an example of the configuration of the sludge treatment support system according to this embodiment. [Figure 2] Figure 2 shows an example of the configuration of a cohesiveness determination device in the sludge treatment support system according to this embodiment. [Figure 3] Figure 3 is a diagram illustrating an example of the function of the model creation unit of the cohesiveness determination device in the sludge treatment support system according to this embodiment. [Figure 4] Figure 4 is a diagram illustrating an example of the function of the cohesiveness determination unit of the cohesiveness determination device in the sludge treatment support system according to this embodiment. [Figure 5] Figure 5 shows an example of the relationship between the water content of dewatered sludge, filtrate turbidity, and the images of flocculated sludge and filtrate in relation to the injection rate of polymer flocculant. [Figure 6] Figure 6 shows an example of a coagulated sludge image and the label to which the coagulated sludge image is classified. [Modes for carrying out the invention]

[0012] An example of the sludge treatment support system according to this embodiment will be described below with reference to the attached drawings.

[0013] Figure 1 is a diagram showing an overview of an example configuration of the sludge treatment support system according to this embodiment. The sludge treatment support system 16 according to this embodiment includes a camera 9, a coagulation degree determination device 14, and an operation manager terminal 15. The sludge treatment support system 16 according to this embodiment is an example of a sludge treatment support system that is added to a sludge dewatering system, which adds a coagulant 3 to sewage sludge 1, stirs and mixes it to form coagulated sludge 5, and dewaters the coagulated sludge 5 in a dewatering machine 6 to form dewatered sludge 7.

[0014] Camera 9 is an example of a camera unit that photographs the coagulated sludge 5. Here, the coagulated sludge 5 is produced when excess sludge (sewage sludge 1) from a sewage treatment plant, which is introduced into the coagulation tank 2, is mixed with a coagulant 3 and stirred in a stirrer 4. In the coagulated sludge 5, the suspended solids in the sewage sludge 1 coagulate and become coarser, forming flocs. In addition, the parts of the coagulated sludge 5 other than the flocs become liquid with reduced sludge concentration due to the decrease in suspended solids. Therefore, the coagulated sludge 5 is a state in which coarsely enlarged flocs and liquid with reduced turbidity are mixed together.

[0015] The sludge dewatering system separates the coagulated sludge 5 into dewatered sludge 7 with a moisture content of 50-90% and filtrate 8, which is the liquid removed from the coagulated sludge 5, by dewatering it in a dewatering machine 6. Most of the flocs are transferred to the dewatered sludge 7, and most of the liquid from the coagulated sludge 5 is transferred to the filtrate 8.

[0016] The coagulation degree discrimination device 14 outputs a coagulated sludge image 10, which is an image of the coagulated sludge 5 captured by the camera 9, and the degree of appropriateness of the coagulant injection amount (an example of the degree of appropriateness of the amount of coagulant 3 added) 11, the degree of floc size in the coagulated sludge 5 12, and the degree of clarity of the filtrate 8 other than the flocs in the coagulated sludge 5 13, which are determined based on the coagulated sludge image 10. In other words, the coagulation degree discrimination device 14 is an example of a coagulation degree discrimination device that classifies the coagulated sludge image 10 captured by the camera 9 and outputs (for example, displays) the degree of appropriateness of the coagulant injection amount 11, the degree of floc size in the coagulated sludge 5 12, and the degree of clarity of the filtrate 8 other than the flocs in the coagulated sludge 5 13 together with the coagulated sludge image 10 to the operation manager terminal 15 (an example of an external terminal).

[0017] Here, the degree of appropriateness 11 of the flocculant injection amount is the degree of appropriateness of the injection amount (an example of the addition amount) of the flocculant 3 with respect to the flocculated sludge 5. The degree of appropriateness 11 of the flocculant injection amount is determined based on the water content of the dewatered sludge 7 obtained in the dewatering treatment of the flocculated sludge 5 and the turbidity of the filtrate 8 or the SS of the filtrate 8, and may include labels such as "appropriate", "excessive", and "insufficient" for the injection amount of the flocculant 3. Hereinafter, the degree of appropriateness 11 of the flocculant injection amount, the degree of size 12 of the flocs in the flocculated sludge 5, and the degree of clarity 13 of the filtrate 8 are also collectively referred to as labels.

[0018] The operation manager terminal 15 displays the flocculated sludge image 10, the degree of appropriateness 11 of the flocculant injection amount, the size 12 of the flocs of the flocculated sludge, and the degree of clarity 13 of the filtrate 8 output from the flocculability degree determination device 14.

[0019] FIG. 2 is a diagram showing an example of the configuration of the flocculability degree determination device included in the sludge treatment support system according to the present embodiment. FIG. 3 is a diagram for explaining an example of the function of the model creation unit of the flocculability degree determination device included in the sludge treatment support system according to the present embodiment. FIG. 4 is a diagram for explaining an example of the function of the flocculability degree determination unit of the flocculability degree determination device included in the sludge treatment support system according to the present embodiment. The flocculability degree determination device 14 includes a model creation unit 17 and a flocculability degree determination unit 18.

[0020] The coagulation degree discrimination unit 18 inputs the coagulated sludge image 10 into the trained coagulant injection amount appropriateness discrimination model 26, the floc size degree discrimination model 27, and the filtrate clarity degree discrimination model 28, each created by the model creation unit 17, and performs image classification, outputting the coagulant injection amount appropriateness degree 11, the floc size degree 12, and the filtrate clarity degree 13. Here, the coagulant injection amount appropriateness discrimination model 26 is a model that discriminates the coagulant injection amount appropriateness degree 11. The floc size degree discrimination model 27 is a model that discriminates the floc size 12 of the coagulated sludge 5. The filtrate clarity degree discrimination model 28 is a model that discriminates the filtrate clarity degree 13. Hereinafter, the coagulant injection amount appropriateness discrimination model 26, the floc size degree discrimination model 27, and the filtrate clarity degree discrimination model 28 are collectively referred to as the image classification model (an example of a model).

[0021] As shown in Figure 3, the model creation unit 17 creates an image classification model using the coagulated sludge image dataset 19. The coagulated sludge image dataset 19 is a list that includes pre-captured coagulated sludge images, the amount of coagulant 3 injected when the coagulated sludge image was taken, the water content of the dewatered sludge 7 of the dewatered coagulated sludge 5, and the turbidity of the filtrate 8, or SS (suspended solids) in the filtrate 8. Next, in the labeling process 20, the model creation unit 17 classifies the coagulated sludge images based on predetermined discrimination criteria for the appropriateness of the coagulant injection amount 11, the size of the flocs in the coagulated sludge 5 12, and the clarity of the filtrate 8 13, and adds labels to the coagulated sludge image dataset 19. The coagulated sludge image dataset 19 with added labels is called the coagulated sludge image dataset 19a.

[0022] Next, in the image augmentation 21, the model creation unit 17 performs rotation and inversion image processing on the coagulated sludge images of the coagulated sludge image dataset 19a to increase the number of images. The coagulated sludge images with increased image count are designated as the coagulated sludge image dataset 19b. The model creation unit 17 also uses the coagulated sludge image dataset 19a to perform coagulant injection amount appropriateness discrimination model training 22. The coagulant injection amount appropriateness discrimination model training 22 uses the coagulated sludge images and coagulant injection amount appropriateness 11 contained in the coagulated sludge image dataset 19b to perform machine learning and create a coagulant injection amount appropriateness discrimination model 26.

[0023] Furthermore, the model creation unit 17 performs floc size discrimination model training 23 using the floc size discrimination model dataset 19a. The floc size discrimination model training 23 uses the floc size discrimination model 27 to perform machine learning using the floc size discrimination model 27, which is contained in the floc size discrimination model dataset 19b.

[0024] Next, the model creation unit 17 performs image processing on the coagulated sludge images in the coagulated sludge image dataset 19b by removing flocs using a floc removal process 24, and replaces the original image (the coagulated sludge image in the coagulated sludge image dataset 19b) with the image from which the flocs have been removed to create the coagulated sludge image dataset 19c. The model creation unit 17 also uses the coagulated sludge image dataset 19c to train a model for discriminating the degree of clarity of the filtrate 8 25. The model for discriminating the degree of clarity of the filtrate 8 25 performs machine learning using the coagulated sludge images in the coagulated sludge image dataset 19c and the degree of clarity of the filtrate 8 13 to create a model for discriminating the degree of clarity of the filtrate 8 28.

[0025] As shown in Figure 4, the coagulation degree discrimination unit 18 inputs the captured coagulated sludge image 10 into the created image classification models (coagulant injection amount appropriateness discrimination model 26, floc size degree discrimination model 27, filtrate 8 clarity degree discrimination model 28), performs image classification, and outputs labels (for example, coagulant injection amount appropriateness 11, floc size degree 12, filtrate 8 clarity degree 13) along with the coagulated sludge image 10. In other words, the image classification that outputs the coagulant injection amount appropriateness 11 and floc size degree 12 is performed by image classification models created by machine learning of the coagulated sludge image and the coagulant injection amount appropriateness 11 and floc size degree 12, which have been previously determined and labeled based on the said coagulated sludge image.

[0026] Furthermore, the image classification that outputs the clarity degree 13 of the filtrate 8 is performed by an image classification model created through machine learning, which uses an image obtained by removing flocs from a previously captured image of coagulated sludge and the clarity degree 13 of the filtrate 8 that has been previously determined and labeled based on that image. Here, the method for removing flocs from the coagulated sludge image may be by masking the image of the flocs. Alternatively, the method for removing flocs from the coagulated sludge image may be by generating an image of only the filtrate 8 with the flocs removed using a generating AI.

[0027] (Example 1) In this example, a coagulation and dewatering test was conducted on digested sludge (TS 2.11%) collected from a sewage treatment plant, using different injection rates (1%, 1.4%, 1.8%, and 2.2%) of a polymer flocculant (an example of flocculant 3). In this example, a predetermined amount of polymer flocculant prepared to a concentration of 0.2% was added to the digested sludge and stirred to obtain coagulated sludge 5, which was then photographed. Subsequently, in this example, the coagulated sludge 5 was dewatered to separate it into dewatered sludge 7 and filtrate 8. The water content of the dewatered sludge 7 (dewatered sludge water content) and the turbidity of the filtrate 8 (filtrate turbidity) were measured, and the filtrate 8 was also photographed.

[0028] Figure 5 shows an example of the relationship between the injection rate of polymer flocculant, the water content of dewatered sludge, the filtrate turbidity, and the images of flocculated sludge and filtrate. At an injection rate of polymer flocculant of 1%, almost no flocs were formed in the flocculated sludge image 10, and a large amount of suspended solids (SS) was observed in the filtrate 8. Furthermore, at an injection rate of polymer flocculant of 1%, the water content of dewatered sludge and the filtrate turbidity were the highest, with the filtrate turbidity being particularly high. Therefore, it is suggested that the injection rate of polymer flocculant is very insufficient.

[0029] Furthermore, at a polymer flocculant injection rate of 1.4%, floc formation progressed compared to an injection rate of 1%. However, filtrate 8 showed a significant amount of suspended solids (SS), suggesting insufficient floc formation. The water content of the dewatered sludge and the filtrate turbidity decreased compared to an injection rate of 1% with polymer flocculant, but were higher than at an injection rate of 1.8%, suggesting that the injection rate of polymer flocculant was insufficient.

[0030] Furthermore, at a polymer flocculant injection rate of 1.8%, the flocs became coarser, filtrate 8 was clearer, and suspended solids (SS) were significantly reduced. The water content of the dewatered sludge was the lowest, and the filtrate turbidity was also low and nearly saturated, suggesting that the polymer flocculant injection rate was appropriate.

[0031] Furthermore, at a polymer flocculant injection rate of 2.2%, the flocs became coarser and reached their maximum size. Filtrate 8 was as transparent as at a polymer flocculant injection rate of 1.8%. However, the volume of filtrate 8 appeared smaller, suggesting that liquid was retained in the coarsened flocs. A slimy feeling was also observed due to the residual polymer flocculant. The water content of the dewatered sludge worsened compared to the polymer flocculant injection rate of 1.8%. This suggests that the removal of liquid retained in the coarsened flocs was impaired, and the viscosity increased due to the residual polymer. Based on the above, the flocculant injection rate is considered excessive.

[0032] Furthermore, as shown in Figure 5, it was suggested that the flocculated sludge image 10 can be classified as "very insufficient," "insufficient," "appropriate," or "excessive" as a degree of appropriateness of flocculant injection amount 11. This label for the degree of appropriateness of flocculant injection amount 11 is highly useful in the operation of the dewatering machine 6 because it includes information on countermeasures, such as reducing the amount of flocculant injected if it is determined to be "excessive," and increasing the amount of flocculant injected if it is determined to be "insufficient."

[0033] Furthermore, the flocculated sludge image 10 can be classified by the general size of the flocs, and by removing the flocs, it can also be classified by the clarity 13 of the filtrate 8. Flocs in the flocculated sludge image 10 were removed by either detecting (segmenting) the flocs and then masking the floc regions, or by using a generation AI to generate an image of only the filtrate 8 with the flocs removed from the flocculated sludge image 10.

[0034] Figure 6 shows an example of a coagulated sludge image and the labels to which the coagulated sludge image is classified. As shown in Figure 6, all coagulated sludge images 10 are assigned three types of labels: the degree of appropriateness of coagulant injection amount 11, the degree of floc size 12, and the degree of clarity of filtrate 8 13. An image classification model is created by machine learning using the coagulated sludge images 10 and the respective labels assigned to them. The accuracy of the created image classification model increases with a large number of images, and augmenting the coagulated sludge images 10 through image processing such as rotation and inversion is useful for improving the accuracy of the image classification model.

[0035] By displaying the degree of appropriateness of the coagulant injection amount 11, the degree of floc size 12, and the degree of clarity 13 of the filtrate 8, as determined by the image classification model, along with the coagulated sludge image 10, the operator of the dewatering machine 6 can maintain the appropriate amount of coagulant injection 3 and also obtain information on the floc size and clarity of the filtrate 8, thereby facilitating the operation and management of the dewatering machine 6. Furthermore, information useful for the operation and management of multiple dewatering machines 6 can be obtained from images captured by only one camera 9, enabling low-cost and effective operation and management.

[0036] Thus, the sludge treatment support system of this embodiment continuously displays the degree of appropriateness of the coagulant injection amount 11, as well as the degree of coagulation of the coagulated sludge 5, thereby supporting the operation of the dewatering machine by the operator of the sludge dewatering system.

[0037] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims of the invention and its equivalents. [Explanation of Symbols]

[0038] 1. Sewage sludge 2 Coagulation tank 3. Flocculant 4. Stirrer 5. Coagulated sludge 6 Dehydrator 7. Dewatered sludge 8. filtrate 9 cameras 10. Images of aggregated sludge 11. Appropriateness of the amount of coagulant injected 12. Size and degree of floc 13. Degree of clarity of the filtrate 14. Apparatus for determining the degree of cohesion 15. Operation Manager Terminal 16. Sludge Treatment Support System 17 Model Creation Department 18. Aggregation degree discrimination unit 19. Aggregated Sludge Image Dataset 20. Labeling process 21. Image padding 22. Learning a model for determining the appropriateness of coagulant injection amount. 23. Learning a model for discriminating the size and degree of flocs. 24. Flocculation treatment 25. Learning a model for discriminating the degree of clarity of filtrate. 26. Model for determining the appropriateness of coagulant injection amount 27. Model for determining the size and degree of flocs 28. Model for determining the degree of clarity of filtrate

Claims

1. A sludge treatment support system to be added to a sludge dewatering system that adds a coagulant to sewage sludge, mixes it with the added coagulant to form coagulated sludge, and dewaters the coagulated sludge in a dewatering machine to form dewatered sludge, The aforementioned coagulated sludge imaging unit, A coagulation degree determination device that classifies the coagulated sludge image captured by the aforementioned imaging unit and outputs the degree of appropriateness of the amount of coagulant added, the degree of floc size in the coagulated sludge, and the degree of clarity of the filtrate other than the floc in the coagulated sludge to an external terminal along with the coagulated sludge image, A sludge treatment support system equipped with the following features.

2. The image classification that outputs the degree of appropriateness of the amount of coagulant added and the degree of floc size is performed by a machine learning model created from a pre-captured image of coagulated sludge and the degree of appropriateness of the amount of coagulant added and the degree of floc size, which have been pre-determined and labeled based on the said coagulated sludge image. The sludge treatment support system according to claim 1, wherein the image classification that outputs the degree of clarity of the filtrate is performed by a model created by machine learning using an image obtained by removing the flocs from a previously captured image of coagulated sludge and the degree of clarity of the filtrate which has been previously determined and labeled based on the said image.

3. The sludge treatment support system according to claim 2, wherein the degree of appropriateness of the amount of coagulant added is determined based on the water content of the dewatered sludge obtained in the dewatering treatment of the coagulated sludge and the turbidity of the filtrate or the SS of the filtrate, and the amount of coagulant added is a label including appropriate, excessive, and insufficient.

4. The sludge treatment support system according to claim 2, wherein the method for removing the flocs from the aggregated sludge image is performed by masking the image of the flocs.

5. The sludge treatment support system according to claim 2, wherein the method for removing the flocs from the aggregated sludge image is to generate an image of only the filtrate from which the flocs have been removed by generating AI.

Images