JP2020065964A - Wastewater treatment method and wastewater treatment system - Google Patents

Abstract

To provide a wastewater treatment method capable of performing, during wastewater treatment, a real-time determination of whether a coagulant addition rate in wastewater treatment is excessive or insufficient.SOLUTION: The wastewater treatment method includes: forming coagulation flocs by adding a coagulant to wastewater; generating image data of coagulation flocs by an image acquisition device 12; inputting at least the image data of coagulation flocs into a coagulant determination model constructed by a machine learning algorithm; and outputting, from the coagulant determination model, a determination result indicating whether the coagulant addition rate is excessive or insufficient.SELECTED DRAWING: Figure 1

Description

  TECHNICAL FIELD The present invention relates to a wastewater treatment method and a wastewater treatment system for determining an excess or deficiency of an addition rate of a flocculant from a model using a machine learning algorithm and an image of floc flocs.

  In recent years, due to the dramatic improvement in computer computing ability, the improvement of communication infrastructure environment, and the evolution of analysis algorithms, it has become the third artificial intelligence boom. Due to the development of the communication infrastructure environment, it is possible to easily acquire a large amount of data, and unlike the conventional artificial intelligence boom, expectations for its use in the real world are increasing.

  Machine learning / artificial intelligence is expected to be applied to various fields, and even in the field of water treatment, exploration of its utilization method is being made. One of the methods of utilizing machine learning / artificial intelligence in the water treatment field is to predict an objective variable in an unknown environment, for example, temporally in the future. An analytical approach is one of the methods for predicting the objective variable in an unknown environment. This is to analyze a physical phenomenon theoretically and to make it an equation. On the other hand, machine learning / artificial intelligence predicts objective variables in an unknown environment, and statistically analyzes the correlation between input and output without analyzing the theoretical relationship between input and output according to the natural law. Analysis is performed and a regression equation is created as an example of the relational expression.

  In the water treatment field, it is sometimes difficult to theoretically analyze what is happening, and machine learning / artificial intelligence that can predict output from input without clarifying the theoretical background is effective. It is expected that there are many aspects that can be utilized.

  As mentioned above, the artificial intelligence boom itself has reached its third boom this time. Neural networks were the main algorithms in the second artificial intelligence boom that occurred in the late 1980s. During the first artificial intelligence boom, by adding a hidden layer to the perceptron, which was composed of only input and output layers, and by establishing a learning method called "error backpropagation", It became an artificial intelligence boom. After all, the second artificial intelligence boom will be abolished, but it is said that one of the factors was that we could not obtain satisfactory prediction accuracy depending on the problem. In the third artificial intelligence boom of this time, deep learning (deep learning) in which the hidden layers of the neural network are multi-layered is the main role, and due to the progress of learning algorithms, it will be possible to demonstrate high accuracy depending on the problem to be applied. This is one of the reasons for the boom. In this specification, machine learning using a neural network composed of an input layer, two or more hidden layers, and an output layer is called deep learning. In addition, artificial intelligence is a broader concept than machine learning, and machine learning is one of the means for creating artificial intelligence.

  Now, returning to the water treatment field, especially wastewater treatment, in wastewater treatment, suspended components such as SS (suspended substance) in raw water and dissolution of COD (chemical oxygen demand), chromaticity, etc. Flocculant is coarsened with a polymer flocculant after aggregating suspension and soluble components with a flocculant containing polyvalent cations such as aluminum and iron in order to efficiently remove volatile components. In many cases, aggregation treatment is performed. The flocs are precipitated in a settling tank provided after the flocculation tank or are separated from the treated water by floating separation using bubbles.

  At this time, the state of agglomerated flocs greatly affects the processing performance of each processing. For example, in the precipitation process, the size and density of the floc flocs affect the sedimentation rate of the flocs. Also in the floating process, the size and density of the floc of flocs affect the floating speed. As a result, the quality of the treated water that has passed through the sedimentation or flotation tank is greatly affected. Therefore, monitoring the state of flocculation flocs is extremely important in wastewater treatment.

  Further, in the coagulation treatment process in wastewater treatment, when biological treatment is performed in the previous stage, the ratio of activated sludge flocs and biofilm fragments mixed in the raw water of the coagulation treatment process varies depending on the operating state of the biological treatment, In the case of factory effluent, the chromaticity of raw water may change due to the operation of the factory. Fluctuations in raw water have a great influence on the coagulation treatment process, so it is important to capture such fluctuations in order to stabilize the treatment.

JP-A-3-175339 JP, 2005-192960, A JP, 2017-72404, A

  It is difficult to judge the state of this floc in real time only by the quality of treated water. That is, after the addition of a flocculant, flocculation flocs were formed, and finally treated water was obtained through precipitation or flotation, and the flocculation operation was appropriate until the quality of the treated water was measured. It is difficult to determine if was good.

  At times, the raw water quality changes depending on the fluctuation of the drainage property due to the operation of the factory and the operating conditions in the aerobic treatment process when there is aerobic treatment in the previous stage. In such a situation, after confirming the water quality of the treated water as described above, even if the addition rate of the coagulant is revised, it is usually several tens of minutes to several hours until the coagulation tank-treated water is obtained. Therefore, it will not be in time.

  Therefore, for fluctuations in raw water, veteran operators decide the coagulant addition rate based on the raw water quality based on past knowledge, etc., and make further decisions while visually checking the floc state of the flocculation tank. Whether or not the addition rate was appropriate was determined, and if necessary, the coagulant addition rate was reviewed again. Alternatively, the raw water is actually collected, divided into equal amounts in a beaker, and several types of coagulant addition rates are set to perform a coagulation test, a so-called jar test, to determine an appropriate coagulant addition rate for the current raw water. There is.

  However, in the former case, it depends largely on the experience and skill of the operator, and it is hard to say that it is a wise method in the water treatment industry, which has the problem of inheriting technology. In the latter case, the unevenness caused by the operator is reduced, but it takes time and it takes about 15 minutes before the result of the jar test is obtained. In any case, it is difficult to improve efficiency such as unmanned processing because it requires manpower.

Therefore, there is a demand for a technique for grasping raw water, flocculation floc state, and treated water in real time and more appropriately performing the flocculation treatment.
Patent Document 1 describes a floc monitoring device that images a floc under the water surface, performs image processing to quantify the particle size of each floc, and obtains the particle size distribution. In addition, there is a description regarding operation control using the same device.

  In Patent Document 2, in a floc forming tank, or in a mixing tank and a floc forming tank, a floc particle size measuring device is used to measure the floc particle diameter at a plurality of points in the floc forming process, and the difference in the obtained floc particle diameters is measured. Alternatively, there is a description regarding a water treatment system that detects a floc formation abnormality from the ratio.

  In either case, the agglomeration state is determined based on the particle size of the floc, but in reality, it is difficult to judge the quality of the agglomerated floc only by the particle size of the floc. For example, when the flocculant addition rate is set to 80, 100, 120, 140, 160, 180 mg / L for raw water for which the optimum coagulant addition rate is determined to be 120 mg / L, and the respective floc particle sizes are measured When the coagulant addition rate exceeds 120 mg / L, the coagulant becomes excessive and floc growth cannot be seen. Then, the floc particle size at the coagulant addition rate of 120 mg / L and the floc particle size at the coagulant addition rate of 180 mg / L may be almost the same.

  Further, sometimes, when the coagulant is excessive, charge neutralization does not occur, defective floc formation occurs, and the particle size of floc may be reduced. In that case, when the optimum addition rate is 120 mg / L, the particle diameters of the floc of 100 mg / L and 140 mg / L may be almost the same. In such a case, if the control based on the floc particle size is performed, the addition rate of the coagulant may be erroneously reviewed.

  Patent Document 3 describes an apparatus and a method for monitoring the aggregation state in the aggregating tank by scattered light in the aggregating treatment of water to be treated having a high SS concentration. In this document, the crest value of scattered light obtained from agglomerate flocs, and the number of appearance of scattered light whose crest value exceeds a set threshold value is to determine the excess or deficiency of the aggregating agent, but the actual determination result is , It is considered that it depends only on the particle size of the floc. In that respect, the above-mentioned problems exist as in Patent Documents 2 and 3.

  The present invention has been made in view of the above points, and an object thereof is to learn the state of agglomerated flocs, including various factors including particle size, by a machine learning algorithm such as deep learning, and to operate a veteran. By constructing a model that can determine coagulation flocs that are equal to or more than the number of coagulants, a wastewater treatment method that can determine the excess or deficiency of the coagulant addition rate in wastewater treatment in real time during wastewater treatment To provide. Moreover, the objective of this invention is providing the wastewater treatment system which can implement such a wastewater treatment method.

  In one aspect, a flocculant is added to wastewater to form flocculants, image data of the flocculation flocs is generated by an image acquisition device, and at least image data of the flocculation flocs is a flocculant constructed by a machine learning algorithm. A wastewater treatment method is provided, which is input to a determination model and outputs a determination result indicating excess or deficiency of a coagulant addition rate from the coagulant determination model.

  In the present specification, wastewater treatment refers to water treatment intended for sewage and industrial wastewater, and in particular, polyaluminum chloride (PACl), sulfate sulfate, ferric chloride, polyiron, polysilicairon ( PSI), organic coagulants, coagulation operations that add inorganic coagulants such as polymers and organic coagulants, and water treatment that involves the operation of precipitating or floating separating coagulated flocs and discharging them as sludge to the outside of the system. .

  As a machine learning algorithm, a deep learning method (deep learning method) is suitable. The deep learning method is a learning method based on a neural network in which hidden layers are multilayered. In this specification, machine learning using a neural network composed of an input layer, two or more hidden layers, and an output layer is called deep learning. By using the deep learning method, it is possible to determine the state of agglomerate flocs, which was previously determined to be good or bad based on human eyes and experience, by a computer based on the image data of the agglomerate flocs.

  For example, the floc particle size when the coagulant addition rate is 100 mg / L and the floc particle size when the coagulant addition rate is 140 mg / L may be almost the same. However, comparing the actual floc states at 100 mg / L and 140 mg / L, there are slight differences in the floc density, so-called tightness, color shade, shape, and the like. According to machine learning / deep learning, it is possible to discriminate an image at a higher accuracy than a human by sometimes causing a computer to repeatedly learn a large number of combinations of image data and correct data corresponding to the image data.

  The image data of the flocculant floc is input to the coagulant determination model constructed by the machine learning algorithm to determine whether the coagulant addition rate is appropriate, that is, whether the wastewater treatment is properly performed. To judge in real time. Machine learning algorithms learn a variety of factors including the particle size of agglomerated flocs. Therefore, the coagulant determination model (that is, the learned model) constructed by the machine learning algorithm enables more accurate determination of the coagulant addition rate than the conventional agglomeration determination technology based on particle size. is there.

  In the construction of the coagulant determination model, the explanatory variable of the learning data is the image data of the coagulation floc, and the objective variable of the learning data is a numerical value representing the excess or deficiency of the coagulant addition rate, for example, 1: small, 2: slightly small, 3: It can be given as a five-level evaluation of optimum, 4: slightly high, 5: high. However, the present invention is not limited to the 5-step evaluation, and may be, for example, 3-step evaluation or 10-step evaluation.

  Whether the actual coagulant addition rate corresponds to the above five-stage evaluation is determined by the turbidity and chromaticity of the treated water in the precipitation treatment after the coagulation treatment, the appearance of the flocculation flocs of the experienced operator, Further, it is possible to make a determination from the water quality of the sand filtration water or the membrane filtration water in the latter stage, the rate of increase in filtration resistance of sand filtration or membrane filtration, and the like. A data set consisting of a combination of numerical values (1 to 5 in the above example) representing the excess or deficiency of the actual coagulant addition rate and the corresponding coagulation floc image data is used to build and update the coagulant determination model. . The numerical value that represents the excess or deficiency of the actual coagulant addition rate is correct data and is used for machine learning such as deep learning.

  FIG. 1 is a schematic diagram showing an embodiment of a coagulant determination model for determining whether the coagulant addition rate is excessive or insufficient. The coagulant determination model is a neural network having an input layer 301, a plurality of hidden layers (also referred to as intermediate layers) 302, and an output layer 303. The coagulant determination model shown in FIG. 1 has four hidden layers 302, but the configuration of the coagulant determination model is not limited to the embodiment shown in FIG. Image data of floc flocs is input to the input layer 301 of the flocculant determination model. More specifically, the numerical value of each pixel forming the image data of the aggregated floc is input to the input layer 301. In one example, when the image data of the floc floc is grayscale image data, the gray level of each pixel (usually a value within 0 to 255) is input to each node (neuron) of the input layer 301 of the flocculant determination model. To be done. When the image data of the flocculation floc is color image data, the numerical values representing red, green, and blue of each pixel are input to the corresponding nodes (neurons) of the input layer 301 of the flocculant determination model.

  In the example shown in FIG. 1, the output layer 303 of the coagulant determination model outputs a numerical value indicating the determination result of the excess or deficiency of the coagulant addition rate. For example, in the case where the excess or deficiency of the coagulant addition rate is represented by a five-level evaluation of 1: small, 2: slightly small, 3: optimal, 4: slightly large, 5: large, the output layer of the coagulant determination model 303 outputs any numerical value from 1 to 5. However, the configuration of the coagulant determination model shown in FIG. 1 is an example, and the present invention is not limited to the example shown in FIG.

  When the image data of the current flocculant floc is input to the coagulant determination model constructed as described above, the computer executes an operation according to the algorithm of the multilayer perceptron that constitutes the neural network, and the output layer 303 displays the coagulant addition. Outputs a numerical value that represents the result of determination of excess or deficiency of the rate. The operator can aim to optimize the operation of the wastewater treatment system based on the determination result.

  In one aspect, the wastewater treatment method further includes a step of generating image data of the wastewater before the coagulant is added by an image acquisition device, and in addition to the image data of the floc flocs, an image of the wastewater. Data is entered into the Flocculant Judgment Model.

  In wastewater treatment, the type of wastewater varies depending on the factory operations. For example, the wastewater A when the factory is producing the product A and the wastewater B when producing the product B have different water qualities such as chromaticity and COD (chemical oxygen demand). Further, there may be a plurality of products other than the products A and B, or the manufacturing ratio of the products may change. Therefore, in the wastewater treatment process, it is necessary to adjust the coagulant addition rate in consideration of the operating state of the factory.

  In addition, if there is a biological treatment such as an activated sludge method or a biofilm method before the coagulation treatment, the activated sludge flocs and biofilm may be mixed in the wastewater depending on the operating state of the biological treatment, and The chromaticity and COD of the inflowing wastewater will change. Therefore, it is necessary to adjust the coagulant addition rate according to the operating state of the biological treatment.

  If these changes in drainage properties can be automatically captured, it will be useful information for the operation of the coagulation process. In order to obtain the information on the drainage property, installation of a COD meter, SS meter, chromaticity meter, etc. is an option, but installation may be difficult from the viewpoint of water quality and cost. Therefore, by obtaining the drainage state as image information, appearance information such as the color of the drainage and the state of the activated sludge flocs can be obtained.

  In one embodiment, the learning data including the image data of the wastewater before the coagulant is added, the image of the coagulation floc, and the numerical value indicating the excess or deficiency of the actual coagulant addition rate (that is, the correct answer data) is used, Build a coagulant determination model with a machine learning algorithm. Machine learning such as deep learning can find the relationship between the image data of wastewater before the addition of the coagulant and the coagulant addition rate at which appropriate coagulation flocs can be formed. The coagulant determination model constructed by the machine learning algorithm can accurately determine the excess or deficiency of the coagulant addition rate with respect to changes in the properties of wastewater.

  In one aspect, the step of adding the aggregating agent to the wastewater to form the aggregating flocs comprises adding an inorganic aggregating agent to the drainage to form an aggregating floc, and then adding an organic aggregating agent to the wastewater to form the agglomerates. A step of growing flocs, wherein the step of generating the image data of the floc flocs is performed by the first image acquisition device using the first image data of the floc flocs formed by adding the inorganic flocculant to the wastewater. It is a step of generating the second image data of the aggregated flocs grown by adding the organic coagulant to the wastewater by the second image acquisition device, and the first image data and the second image data are generated. Input to the coagulant determination model.

  In the coagulation process of wastewater treatment, an inorganic flocculant is added to the wastewater in the first-stage stirring tank to generate coagulation flocs, and an organic flocculant such as a polymer coagulant is added to the wastewater in the second-stage stirring tank. Then, the aggregated flocs may be aggregated and coarsened. In one embodiment, the first image data of the floc flocs in the first-stage agitation tank, the second image data of the floc flocs in the second-stage agitation tank, and the actual excess / deficiency of the coagulant addition rate are displayed. A coagulant determination model is constructed by a machine learning algorithm using learning data including numerical values (that is, correct answer data).

  In one aspect, the wastewater treatment method further comprises the step of separating the aggregated flocs from the wastewater by precipitating the aggregated flocs to generate treated water, and generating image data of the treated water by an image acquisition device. In addition to the image data of the floc, the image data of the treated water is input to the flocculant determination model.

  The image data of the treated water generated by the coagulation sedimentation treatment (that is, the supernatant water in the settling tank) includes the appearance information of the treated water such as the color tone and the suspension state of the obtained treated water. By using the learning data including the image data of the treated water in addition to the image data of the floc, it is possible to construct a more accurate coagulant determination model by the machine learning algorithm. In addition, the coagulant determination model constructed using such learning data can output a determination result with higher accuracy.

  In one aspect, the wastewater treatment method produces treated water by levitating the flocculation flocs and removing the flocculated flocs from the wastewater, and an image of sludge composed of the flocculation flocs removed from the wastewater. Data, and further comprising a step of generating at least one of the image data of the treated water by an image acquisition device, in addition to the image data of the floc flocs, one of the image data of the sludge and the image data of the treated water or Both are entered into the Flocculant Judgment Model.

  The image data of the treated water generated by the coagulation flotation method (that is, the water from which coagulated flocs are removed) includes the appearance information of the treated water such as the color tone and the suspension state of the obtained treated water. By using the learning data including the image data of the treated water in addition to the image data of the floc, it is possible to construct a more accurate coagulant determination model by the machine learning algorithm. In addition, the coagulant determination model constructed using such learning data can output a determination result with higher accuracy.

  In one aspect, the wastewater treatment method separates the flocculation flocs from the wastewater to generate treated water, generates image data of the treated water by an image acquisition device, and machine-learns the image data of the treated water. The method further includes the step of inputting to the water quality judgment model constructed by an algorithm and outputting a predicted value of the water quality of the treated water from the water quality judgment model.

  In the construction of the water quality judgment model, the image data of the treated water is used as the explanatory variable of the learning data, and the actual measured value of the water quality of the treated water is used as the objective variable of the learning data. Examples of the measured water quality of the treated water include turbidity of the treated water and COD (chemical oxygen demand) of the treated water. The water quality judgment model constructed using such learning data can output the predicted value of the water quality of the treated water obtained by the coagulation treatment.

  An alarm may be issued when the predicted water quality value output from the water quality determination model exceeds a predetermined threshold value. For example, when the predicted value of the turbidity of the treated water exceeds 1 as the predicted value of the water quality, or when the COD of the treated water exceeds 10 mg / L as the predicted value of the water quality, an alarm may be issued. By setting such a threshold value, safer and more reliable operation can be performed even in equipment with reduced labor.

  In one aspect, the wastewater treatment method further includes a step of separating the aggregate flocs from the wastewater to generate treated water, and forming the aggregate flocs in addition to image data of the aggregate flocs, The time difference from the step of separating coagulated flocs from the waste water is input to the coagulant determination model.

  The wastewater treatment process takes minutes to hours. For example, there is a time delay of several hours from the addition of the flocculant to form floc and the completion of the precipitation treatment. Therefore, there is a time difference between the floc formation step and the floc separation step. It is possible to calculate the time lag from when the coagulant addition rate is changed to when the result of precipitation treatment changes, but it is possible to calculate it with a hydraulic simulation model, but in order to match with the reality, it is a highly accurate model. And requires high computing power. Therefore, there is a cost problem in applying the hydraulic simulation model to many sites.

  In the present invention, when constructing the flocculant determination model, an explanatory variable indicating the time difference between the floc floc generation step and the floc separation step, for example, by adding the elapsed time from a certain reference time to the learning data, It is possible to construct a coagulant determination model that includes the temporal relationship between the parameter change operation that actually occurs and the change in the processing result caused by the change. The coagulant determination model including such a temporal relationship is a highly accurate coagulant even when the operating parameters such as a change in the water quality of wastewater and a change operation of the rotation speed of the stirrer are accompanied by a time delay. It is possible to output the determination result of excess or deficiency of the addition rate.

  In one aspect, in addition to the image data of the floc flocs, information data related to wastewater treatment is input to the flocculant determination model.

  Information data includes turbidity of wastewater, color of wastewater, COD of wastewater, pH of wastewater, SS concentration of wastewater, ammonia concentration of wastewater, total nitrogen concentration of wastewater, total phosphorus concentration of wastewater, alkalinity of wastewater, wastewater Temperature, flow rate of waste water, rotation speed of agitator equipped in waste water treatment equipment, amount and frequency of sludge withdrawal from sedimentation tank, and data expressed by numerical values related to operation. By using the learning data including the information data related to the wastewater treatment in addition to the image data of the floc, it is possible to construct the coagulant determination model with higher accuracy by the machine learning algorithm. In addition, the coagulant determination model constructed using such learning data can output a determination result with higher accuracy.

In one aspect, the determination result is used to control the coagulant addition rate.
The operation of the wastewater treatment system can be automatically optimized based on the determination result of the excess or deficiency of the coagulant addition rate. For example, the determination result is output in a predetermined cycle, for example, every minute. At a certain time T, if the judgment result is 1: “small”, the current coagulant addition rate is increased by + 50% in 0.5 minutes, and if 2: “slightly small”, the current coagulant addition is performed. The rate is increased by + 25% in 0.5 minutes, and if 3: "optimal", the current coagulant addition rate is unchanged, and if 4: "moderately high", the current coagulant addition rate is 0. Decrease by -25% in 5 minutes, 5: If "high", decrease current coagulant addition rate by -50% in 0.5 minutes. Subsequently, at time T + 1, the determination result is output again, and the coagulant addition rate is adjusted in the same manner. This step is repeated until the optimum determination result is output.

  In the above example, the numerical value indicating the determination result of the excess or deficiency of the coagulant addition rate is any of 1 to 5, but the present invention is not limited to the above example. In one embodiment, the numerical value representing the determination result of the excess or deficiency of the coagulant addition rate is + 50% (corresponding to "small"), + 25% (corresponding to "slightly small"), 0% (corresponding to "optimal"). , -25% (corresponding to "slightly high"), -50% (corresponding to "high").

  Here, at time 0, that is, when switching from the existing control in the wastewater treatment system to the control based on the above model, or when the operation of the wastewater treatment system is newly started, the coagulant addition rate at the time when the control is started is The value may be determined based on the past knowledge of the wastewater treatment system, judgment of experts, batch tests such as jar tests, and the like. Control of the wastewater treatment system is started based on the coagulant addition rate determined here. After that, after the coagulant is added, considering the water arrival time to the point where the image is captured by the image acquisition device, the control for optimizing the coagulant addition rate using the above model is started. good.

  In one aspect, the wastewater treatment method is irrelevant to the state of the floc floc, the reference value of the coagulant addition rate is determined by a reference value calculation formula, and the correction value of the reference value is the coagulant addition rate. The method further includes the step of making a determination based on the determination result indicating excess or deficiency and correcting the coagulant addition rate from the reference value and the correction value.

  When the quality of the wastewater flowing into the wastewater treatment system changes rapidly in a short time, control may be delayed and the quality of the treated water may deteriorate. Therefore, the computer has a reference value calculation formula for determining the reference value of the coagulant addition rate irrespective of the state of floc flocs, and using this reference value calculation formula, the reference value of the coagulant addition rate To decide. Further, the computer determines the correction value of the reference value from the determination result indicating the excess or deficiency of the coagulant addition rate.

For the standard value calculation formula, for example, the standard value (D ₀ ) of the coagulant addition rate is determined from the values of the effluent quality (turbidity, chromaticity, pH, COD, etc.) that is constantly monitored by the effluent quality meter. Is a formula for. This reference value calculation formula is, for example, a dataset of past operation results in the wastewater treatment system, that is, the value of the wastewater quality and the corresponding coagulant addition rate, from the value of the wastewater quality by a statistical analysis method, It is possible to use a regression equation to calculate the corresponding coagulant addition rate. The coagulant is added based on the standard value of the coagulant addition rate thus obtained, and coagulated flocs are formed.

This coagulation floc is imaged by the method described above, and the image data of the coagulation floc is input to the coagulant determination model described above to determine whether the coagulant addition rate is excessive or insufficient. Here, when D ₀ is a reference value (mg / L) of the coagulant addition rate, and d is a correction value represented by a positive or negative percentage (%) indicating the determination result of the coagulant addition rate, The corrected coagulant addition rate (D, mg / L) is represented by the following correction formula (1).
D = D ₀ × (1 + d / 100) (1)

The correction value d represents the determination result of the excess or deficiency of the coagulant addition rate, + 50% (when "small"), + 25% (when "slightly small"), 0% (when "optimal"), -25% It is either (when “somewhat”) or −50% (when “some”). For example, when D ₀ = 20 mg / L and d = −25%, the corrected coagulant addition rate D is 15 mg / L. According to the embodiment described above, it is possible to more quickly control the coagulant addition rate based on the image data of the floc.

  In one aspect, the wastewater treatment method updates the coagulant determination model using a data set consisting of a combination of image data of the floc floc and a numerical value indicating an excess or deficiency of the actual coagulant addition rate. Further includes.

  The coagulant determination model constructed by the machine learning algorithm is continuously updated on the basis of newly obtained data in the wastewater treatment system that is operating using the coagulant determination model. This allows the model to be updated according to the actual situation even if the properties of the wastewater in the wastewater treatment system have changed significantly after the coagulant determination model was first constructed, or if the operating method was changed. Can continuously output an appropriate determination result.

  In one aspect, it has an aggregating floc formation tank that forms an aggregating floc by adding an aggregating agent to wastewater, an image acquisition device that generates image data of the aggregating floc, and an aggregating agent determination model constructed by a machine learning algorithm. A computer, wherein the computer stores a storage device in which the aggregating agent determination model is stored and at least the image data is input to the aggregating agent determination model, and the aggregating agent indicates a determination result indicating an excess or deficiency of the aggregating agent addition rate. A wastewater treatment system is provided that includes a treatment device that performs an operation for outputting from a determination model.

  In one aspect, a step of inputting at least image data of agglomerated flocs to the input layer of a model including a neural network having an input layer, a plurality of hidden layers, and an output layer, and a multilayer perceptron configuring the neural network. A computer-readable recording medium having a program recorded thereon for executing the step of outputting a numerical value indicating the excess or deficiency of the coagulant addition rate from the output layer by executing the calculation according to the algorithm is provided.

  According to the wastewater treatment method according to the present invention, the coagulant determination model constructed by the machine learning algorithm, and the image data of the floc floc obtained by the image acquisition device such as a small camera, the excess or deficiency of the coagulant addition rate. , It is possible to automatically judge with high accuracy in real time, and it becomes possible to operate a more optimal wastewater treatment system.

It is a schematic diagram which shows one Embodiment of the coagulant | flocculant determination model which determines the excess / deficiency of a coagulant | flocculant addition rate. It is a figure showing one embodiment of a wastewater treatment system. It is a figure which shows other embodiment of the wastewater treatment system. It is a figure which shows other embodiment of the wastewater treatment system. It is a figure which shows other embodiment of the wastewater treatment system. It is a schematic diagram which shows one Embodiment of the computer which comprises at least one part of a computing system.

  FIG. 2 is a diagram showing an embodiment of the wastewater treatment system. In this wastewater treatment system, a model that has been learned in a wastewater treatment system that uses a filtration method is used to determine whether the coagulant addition rate is excessive or deficient (coagulation floc state) and to control the coagulant addition rate. All the descriptions including the model construction, the model configuration, and the like described in “Means for Solving the Problems” are applied to each embodiment of the wastewater treatment system described below.

  The wastewater treatment system includes a rapid stirring tank 1 and a slow stirring tank 2 as stirring tanks, and a coagulant addition pump 5 for adding a coagulant to the wastewater. The rapid agitation tank 1 and the slow agitation tank 2 function as an aggregation floc formation tank that adds an aggregating agent to waste water to form aggregation flocs. The coagulant addition pump 5 is connected to the rapid stirring tank 1 and is arranged to add the coagulant to the waste water in the rapid stirring tank 1. The wastewater treatment system further includes a coagulant addition pump 10 that adds the coagulant to the wastewater in the slow stirring tank 2. The coagulant added to the rapid stirring tank 1 by the coagulant addition pump 5 is an inorganic coagulant, and the coagulant added to the slow stirring tank 2 by the coagulant addition pump 10 is an organic coagulant. In one embodiment, the slow stirring tank 2 and the flocculant addition pump 10 may not be provided.

  The wastewater introduction line 3 is connected to the rapid stirring tank 1, and the wastewater to be treated is supplied to the rapid stirring tank 1 through the wastewater introduction line 3. The slow stirring tank 2 is connected to the rapid stirring tank 1. The rapid agitation tank 1 is an agitation tank that agitates wastewater to which a flocculant is added and agglomerates suspended substances in the wastewater to form floc flocs. The slow-speed stirring tank 2 is a stirring tank that stirs wastewater containing agglomerated flocs to coarsen the agglomerated flocs.

  The wastewater treatment system further includes a settling tank 7 as a separation tank for separating the floc of flocs from the wastewater. The precipitation tank 7 is connected to the slow stirring tank 2. The waste water containing the flocs growing in the slow stirring tank 2 is sent to the precipitation tank 7. The flocculated flocs are settled in the settling tank 7, whereby the flocculated flocs are separated from the waste water. The flocculated flocs that have settled are discharged from the bottom of the settling tank 7 as sludge. The supernatant water in the settling tank 7 is treated water, and this treated water is discharged from the gutter 7a of the settling tank 7.

  The wastewater treatment system inputs an image acquisition device 12 for generating image data of floc flocs formed in the rapid agitation tank 1 and at least the image data into a flocculant determination model constructed by a machine learning algorithm, The calculation system 100 is further provided that outputs a determination result indicating excess or deficiency of the addition rate from the coagulant determination model. The computing system 100 includes a program for building and updating a flocculant determination model, a storage device 110 in which the flocculant determination model is stored, and a processing device 120 (such as a GPU or a CPU) that executes a computation according to the program. ing. The processing device 120 inputs at least image data of a flocculant floc into the flocculant determination model, and executes a calculation for outputting a determination result indicating excess or deficiency of the coagulant addition rate from the flocculant determination model.

  The computing system 100 is composed of at least one computer. The at least one computer may be a server or a plurality of servers. The coagulant determination model for determining whether the coagulant addition rate is excessive or deficient (that is, whether or not the coagulant floc state is good or bad) based on at least the image data of the coagulant floc is stored in the storage device 110 of the arithmetic system 100. Further, the above-described reference value calculation formula is also stored in the storage device 110 of the arithmetic system 100.

  The computing system 100 may be an edge server connected to the image acquisition device 12 by a communication line, a cloud server connected to the image acquisition device 12 via a network such as the Internet, or an image acquisition. It may be a fog computing device (gateway, fog server, router, etc.) installed in the network connected to the device 12. The computing system 100 may be a plurality of servers connected by a network such as the Internet. For example, the computing system 100 may be a combination of an edge server and a cloud server.

  In the present embodiment, the image acquisition device 12 photographs the floc of flocs in the rapid stirring tank 1 after the addition of the flocculant. The image acquisition device 12 generates image data of agglomerated flocs and sends the image data to the arithmetic system 100. The computing system 100 acquires image data from the image acquisition device 12 and stores it in the storage device 110. The image acquisition device 12 is installed above the rapid stirring tank 1 (above the water surface). An illuminator such as an LED light is used to illuminate the floc in the rapid stirring tank 1 so that the image acquisition device 12 can clearly capture the floc. The image acquisition device 12 may be installed in the rapid stirring tank 1 (below the water surface) or in the pipe connecting the rapid stirring tank 1 and the slow stirring tank 2 as long as the floc can be clearly imaged. Absent.

  The image acquisition device 12 is a digital camera equipped with an image sensor (for example, a CCD image sensor or a CMOS image sensor) capable of generating a still image or a continuous image. As a specification of the digital camera, a camera having a manual focus function is preferable in order to always keep the focal length constant, and the number of pixels is preferably 5 million pixels or more, preferably 10 million pixels or more.

  In the present embodiment, the aggregation flocs to be imaged are the aggregation flocs in the rapid stirring tank 1. This is because the control delay can be shortened and it is preferable from the viewpoint of controlling the coagulant addition rate. In a treated water field where fluctuations in wastewater are scarce, a coarse flocculation floc in the slow stirring tank 2 may be a target for imaging.

  The coagulant addition pump 5 is electrically connected to the arithmetic system 100. The arithmetic system 100 adjusts the coagulant addition rate by issuing a command to the coagulant addition pump 5 based on the determination result of the excess or deficiency of the coagulant addition rate output from the coagulant determination model. Specifically, when the determination result shows that the current coagulant addition rate is low, the arithmetic system 100 issues a command to the coagulant addition pump 5 to increase the coagulant addition rate. When the determination result indicates that the current coagulant addition rate is high, the arithmetic system 100 issues a command to the coagulant addition pump 5 to reduce the coagulant addition rate. In this way, the arithmetic system 100 can optimize the coagulant addition rate by controlling the operation of the coagulant addition pump 5 based on the determination result output from the coagulant determination model. At the start of the optimization of the coagulant addition rate using the coagulant determination model, the coagulant addition rate is determined from knowledge based on past operation of the wastewater treatment system and jar test results.

  In one embodiment, the turbidity sensor 21, the pH sensor 22, and the COD measuring device 23 for acquiring information data related to wastewater treatment may be connected to the wastewater introduction line 3. The turbidity sensor 21, the pH sensor 22, and the COD measuring device 23 are electrically connected to the calculation system 100, and the measured values of the turbidity, pH, and COD of the wastewater are input to the calculation system 100. Further, as the information data, among the chromaticity of the wastewater, the treatment flow rate of the wastewater, the stirring rotation speed of the rapid stirring tank 1 and / or the slow stirring tank 2, the sludge withdrawal amount and frequency of the settling tank 7, and the coagulant addition rate, At least one may be input to the computing system 100. By using the learning data including the above information data in addition to the image data of the floc floc, a more accurate flocculant determination model can be constructed by the machine learning algorithm.

  In one embodiment, the above-mentioned information data is input to the learned coagulant determination model together with the image data of the floc, and the coagulant determination model outputs the determination result of the excess or deficiency of the coagulant addition rate. Since the water quality of the wastewater that has reached the image acquisition device 12 at the time of coagulation floc shooting may differ from the water quality of the wastewater in the wastewater introduction line 3 at that time, the wastewater corresponding to the acquired coagulation floc image data As the water quality, it is desirable to adopt a value that takes into account the time required for the water to reach the image acquisition device 12 from the wastewater introduction line 3.

  The coagulant determination model is constructed by using a deep learning method or the like and acquiring various data by the wastewater treatment system shown in FIG. 2 before the operation using the coagulant determination model. Here, in the coagulant determination model using the machine learning algorithm, the prediction accuracy for input data that has never been experienced is low, so the data in the construction of the coagulant determination model is Those that include changes are desirable. Therefore, it is desirable to construct a coagulant determination model using the data acquired for preferably one year or more. However, the data acquisition period can be shortened if the drainage water quality is stable. In addition, when a coagulant determination model is already constructed in a similar wastewater treatment system that treats similar wastewater, the coagulant determination model can be used.

  The flocculation floc information output from the flocculant determination model relates to the quality of the flocculation floc, that is, whether the flocculant addition rate is excessive or insufficient. Although the operator of the wastewater treatment system may review the coagulant addition rate with reference to this information, if this information is incorporated into the control of the coagulant as shown in Fig. 2, the coagulant addition rate can be continuously calculated. It is possible to optimize. In this way, the wastewater treatment system can be continuously and stably operated.

  However, if the operation of the wastewater treatment system is prolonged, the quality of the wastewater may change or a part of the operation flow may change, and the coagulant determination model may not be able to output an appropriate determination result. Therefore, it is necessary to update the coagulant determination model appropriately. Whether the determination result of the excess or deficiency of the coagulant addition rate is correct or not can be determined based on the turbidity and chromaticity of the treated water (supernatant water) at the outlet of the precipitation tank 7. If the flocculation treatment is not performed properly, the sedimentation of flocculation flocs becomes poor and the amount of suspended solids remaining in the treated water without agglomeration increases, resulting in high turbidity of the treated water.

  Therefore, in the present embodiment, during the operation of the coagulant determination model, the value of the turbidity of the treated water is continuously acquired by the turbidity sensor 30, and the delivery time is taken into consideration, and the explanatory variable and the objective variable are included. Add as supplementary information to learning data. This supplementary information is compared with the output result of the coagulant determination model to periodically verify the accuracy of the coagulant determination model. When the accuracy of the coagulant determination model is lower than that at the beginning of operation, correct data is re-assigned to the objective variable of the learning data based on supplemental information, and the coagulant determination model is updated by a deep learning method or the like. To do.

  When the computing system 100 includes an edge server at the edge (site) of the wastewater treatment system and a cloud server connected to the edge server by a network (for example, the Internet), the update work of the coagulant determination model is performed. It is desirable to perform it on the cloud server and then send the updated flocculant determination model to the edge server and store it in the edge server.

  FIG. 3 is a diagram showing another embodiment of the wastewater treatment system. The configuration of this embodiment that is not particularly described is the same as that of the embodiment shown in FIG. 2, and thus the duplicate description thereof will be omitted. In the present embodiment, similarly to the above-described embodiment, a model learned in a processing system using a filtration method is used to determine whether the coagulant addition rate is excessive or insufficient, and to control the coagulant addition rate. Is an example.

  The wastewater treatment system includes an image acquisition device 32 that generates image data of wastewater before being introduced into the rapid stirring tank 1, and an image acquisition device 33 that generates image data of a floc floc formed in the slow stirring tank 2. An image acquisition device 34 is further provided for generating image data of treated water (supernatant water) from which floc flocs have been removed in the settling tank 7. The image acquisition device 32 is installed facing the drainage introduction line 3, the image acquisition device 33 is installed facing the slow stirring tank 2, and the image acquisition device 34 is installed facing the outlet of the precipitation tank 7.

  The image acquisition device 32 generates image data of the wastewater flowing through the wastewater introduction line 3, that is, the wastewater before the coagulant is added, and sends the image data to the calculation system 100. The image acquisition device 33 generates image data of the floc flocs grown in the slow stirring tank 2 and sends the image data to the arithmetic system 100. The image acquisition device 34 generates image data of treated water from which floc flocs are removed by the sedimentation treatment in the sedimentation tank 7, and sends this image data to the arithmetic system 100. The arithmetic system 100 stores these image data in the storage device 110.

  The calculation system 100 includes image data of wastewater before addition of a flocculant, image data of floc flocs in the rapid stirring tank 1, image data of floc flocs in the slow stirring tank 2, and an image of treated water. A coagulant determination model is constructed by a machine learning algorithm using data and learning data including numerical values (ie, correct answer data) representing the excess or deficiency of the actual coagulant addition rate.

  During the operation of the wastewater treatment system, the arithmetic system 100 causes the image data of the wastewater before the addition of the flocculant, the image data of the floc in the rapid stirring tank 1, the image data of the floc in the slow stirring tank 2. , And the image data of the treated water are input to the constructed coagulant determination model, and the coagulant determination model outputs a determination result indicating excess or deficiency of the coagulant addition rate. In one embodiment, either one or both of the image data of the wastewater before the coagulant is added and the image data of the treated water may be omitted.

  The coagulant addition pumps 5 and 10 are electrically connected to the arithmetic system 100. The computing system 100 adjusts the coagulant addition rate by issuing a command to the coagulant addition pumps 5 and 10 based on the determination result of the excess or deficiency of the coagulant addition rate output from the coagulant determination model.

  FIG. 4 is a diagram showing another embodiment of the wastewater treatment system. The configuration of this embodiment that is not particularly described is the same as that of the embodiment shown in FIG. 3, and thus the duplicate description thereof will be omitted. In the present embodiment, similarly to the above-described embodiment, a model learned in a processing system using a filtration method is used to determine whether the coagulant addition rate is excessive or insufficient, and to control the coagulant addition rate. Is an example.

  The wastewater treatment system of the present embodiment includes a flotation tank 40 instead of the precipitation tank 7 as a separation tank for separating coagulated flocs from wastewater. The floating tank 40 is connected to the slow stirring tank 2. A pressurized water supply line 41 that supplies pressurized water containing pressurized air is connected to the levitation tank 40. The pressurized water is injected into the levitation tank 40 through the pressurized water supply line 41. The air contained in the pressurized water forms minute bubbles in the wastewater held in the floating tank 40. The air bubbles adhere to the floc and cause the flocs to float. The flocculated flocs that have floated are scraped by the scum skimmer 44, collected as sludge in the gutter 45, and further discharged from the flotation tank 40 through the gutter 45. The treated water from which the flocs of flocculation have been removed is discharged from the floating tank 40 through the treated water discharge line 47.

  The wastewater treatment system further includes an image acquisition device 48 that generates image data of sludge composed of coagulated flocs removed from the wastewater, and an image acquisition device 49 that generates image data of the treated water discharged from the flotation tank 40. There is. The image acquisition device 48 is arranged facing the gutter 45 that collects the flocs, and the image acquisition device 49 is installed facing the treated water discharge line 47.

  The image acquisition device 48 generates image data of sludge made of coagulated flocs removed from the wastewater, and sends this image data to the arithmetic system 100. The image acquisition device 49 generates image data of the treated water flowing through the treated water discharge line 47, that is, treated water from which flocculated flocs have been removed by the floating separation process in the floating tank 40, and sends the image data to the arithmetic system 100. . The arithmetic system 100 stores these image data in the storage device 110.

  The calculation system 100 includes image data of wastewater before addition of a flocculant, image data of floc flocs in the rapid stirring tank 1, image data of floc flocs in the slow stirring tank 2, and image data of sludge. Then, a coagulant determination model is constructed by a machine learning algorithm by using the learning data including the image data of the treated water and the numerical value (that is, the correct answer data) representing the excess or deficiency of the actual coagulant addition rate.

  During the operation of the wastewater treatment system, the arithmetic system 100 causes the image data of the wastewater before the addition of the flocculant, the image data of the floc in the rapid stirring tank 1, the image data of the floc in the slow stirring tank 2. The image data of the sludge and the image data of the treated water are input to the constructed coagulant determination model, and the coagulant determination model outputs a determination result indicating excess or deficiency of the coagulant addition rate. In one embodiment, either or both of the sludge image data and the treated water image data may be omitted.

  The computing system 100 adjusts the coagulant addition rate by issuing a command to the coagulant addition pumps 5 and 10 based on the determination result of the excess or deficiency of the coagulant addition rate output from the coagulant determination model.

  FIG. 5 is a diagram showing another embodiment of the wastewater treatment system. The configuration of this embodiment that is not particularly described is the same as that of the embodiment shown in FIG. 1, and thus the duplicate description thereof will be omitted. The wastewater treatment system of this embodiment includes a biological reaction tank 51 and a precipitation tank 52. The biological reaction tank 51 functions as a floc formation tank for adding flocculant to waste water to form floc. The precipitation tank 52 functions as a separation tank for separating the floc of flocs from the waste water. The biological reaction tank 51 and the precipitation tank 52 are connected by a communication channel 55.

  The biological reaction tank 51 is a processing tank for performing biological treatment on wastewater according to a standard activated sludge method, a biofilm method, a biological nitrogen removal method, a biological phosphorus removal method, and the like. The wastewater to be treated flows into the biological reaction tank 51 through the wastewater introduction line 3, and then flows into the settling tank 52 through the communication channel 55. The biological reaction tank 51 has an overflow section 51A, and the wastewater subjected to biological treatment is transferred to the precipitation tank 52 after passing through the overflow section 51A. The settling tank 52 also has an overflow section 52A. The treated water (supernatant water) from which coagulated flocs have been removed is discharged from the settling tank 52 through the overflow section 52A.

  The wastewater treatment system includes a flocculant addition pump 57 that adds a flocculant to the wastewater in the biological reaction tank 51, an image acquisition device 58 that generates image data of floc flocs formed in the biological reaction tank 51, and a precipitation tank. An image acquisition device 59 for generating image data of treated water (supernatant water) in 52 is provided. The coagulant addition pump 57, the image acquisition device 58, and the image acquisition device 59 are electrically connected to the arithmetic system 100.

  The coagulant addition pump 57 is arranged to add the coagulant to the overflow portion 51A of the biological reaction tank 51. An inorganic coagulant is used as the coagulant. The image acquisition device 58 is arranged facing the communication channel 55, and generates image data of agglomerated flocs in the wastewater flowing through the communication channel 55. The image acquisition device 59 is installed above the overflow section 52A of the settling tank 52, and generates image data of treated water (supernatant water) from which floc flocs have been removed by the sedimentation processing in the settling tank 52. The image acquisition device 58 and the image acquisition device 59 send the image data of the floc and the treated water to the arithmetic system 100. The arithmetic system 100 stores these image data in the storage device 110.

  The computing system 100 uses a coagulant agent by a machine learning algorithm by using coagulant floc image data, treated water image data, and learning data including numerical values (i.e., correct answer data) indicating an excess or deficiency of the actual coagulant addition rate. Build a decision model. During the operation of the wastewater treatment system, the arithmetic system 100 inputs the image data of the floc flocs and the image data of the treated water into the constructed flocculant determination model, and the flocculant determination model determines whether the coagulant addition rate is excessive or insufficient. Is output.

  The calculation system 100 adjusts the coagulant addition rate by issuing a command to the coagulant addition pump 57 based on the determination result of the coagulant addition rate which is output from the coagulant determination model.

  The calculation system 100 further includes a water quality determination model that outputs a predicted value of the water quality of the treated water, in addition to the above-described flocculant determination model. The water quality determination model is a neural network having an input layer, a plurality of hidden layers (also referred to as intermediate layers), and an output layer, like the coagulant determination model. The water quality determination model is stored in the storage device 110 of the arithmetic system 100. In the construction of the water quality judgment model, the image data of the treated water is used as the explanatory variable of the learning data, and the actual measured value of the water quality of the treated water is used as the objective variable of the learning data. Examples of the measured water quality of the treated water include turbidity of the treated water and COD (chemical oxygen demand) of the treated water. The water quality judgment model constructed using such learning data can output the predicted value of the water quality of the treated water obtained by the coagulation treatment.

  During the operation of the wastewater treatment system, the arithmetic system 100 inputs the image data of the treated water generated by the image acquisition device 59 into the constructed water quality determination model, and the water quality determination model uses the predicted value of the water quality of the treated water. Is output. The computing system 100 may issue an alarm when the predicted water quality value output from the water quality determination model exceeds a predetermined threshold value. For example, when the predicted value of the turbidity of the treated water exceeds 1 as the predicted value of the water quality, or when the COD of the treated water exceeds 10 mg / L as the predicted value of the water quality, the arithmetic system 100 issues an alarm. Good. By setting such a threshold value, safer and more reliable operation can be performed even in equipment with reduced labor.

  FIG. 6 is a schematic diagram showing an embodiment of a computer forming at least a part of the arithmetic system 100. The computer includes a storage device 110 in which programs and data are stored, a processing device 120 such as a CPU (central processing unit) or GPU (graphic processing unit) that performs an operation according to a program stored in the storage device 110, and data. , A program, and various kinds of information in the storage device 110, an output device 140 for outputting a processing result and processed data, and a communication device 150 for connecting to a network such as the Internet. I have it.

  The storage device 110 includes a main storage device 111 accessible by the processing device 120 and an auxiliary storage device 112 that stores data and programs. The main storage device 111 is, for example, a random access memory (RAM), and the auxiliary storage device 112 is a storage device such as a hard disk drive (HDD) or a solid state drive (SSD).

  The input device 130 includes a keyboard and a mouse, and further includes a recording medium reading device 132 for reading data from the recording medium, and a recording medium port 134 to which the recording medium is connected. The recording medium is a non-transitory tangible computer-readable recording medium, and is, for example, an optical disk (eg, CD-ROM, DVD-ROM) or a semiconductor memory (eg, USB flash drive, memory card). is there. Examples of the recording medium reading device 132 include an optical drive such as a CD-ROM drive and a DVD-ROM drive, and a card reader. An example of the recording medium port 134 is a USB port. The program and / or data stored in the recording medium is introduced into the computer via the input device 130 and stored in the auxiliary storage device 112 of the storage device 110. The output device 140 includes a display device 141 and a printing device 142.

  The trained model constructed by the machine learning algorithm is stored in the storage device 110. As shown in FIG. 1, this learned model is a neural network having an input layer 301, a plurality of hidden layers (also referred to as intermediate layers) 302, and an output layer 303. The computer operates according to a program electrically stored in the storage device 110. In one embodiment, the computer acquires the image data of the floc from the image acquisition device 12, inputs at least the image data into a flocculant determination model constructed by a machine learning algorithm, and configures a neural network. By executing an operation according to the algorithm of the multi-layer perceptron, the output layer 303 outputs the numerical value indicating the determination result of the excess / deficiency of the coagulant. Further, the computer executes a step of constructing the coagulant determination model and a step of updating the coagulant determination model. A program for causing a computer to execute these steps is recorded in a computer-readable recording medium that is a non-transitory tangible material, and is provided to the computer via the recording medium. Alternatively, the program and the aggregating agent determination model may be input to the computer from the communication device 150 via a communication network such as the Internet.

  The above-described embodiment is described for the purpose of enabling a person having ordinary knowledge in the technical field to which the present invention belongs to implement the present invention. Various modifications of the above-described embodiment can be naturally made by those skilled in the art, and the technical idea of the present invention can be applied to other embodiments. Therefore, the present invention is not limited to the described embodiments, but is to be construed in its broadest scope according to the technical idea defined by the claims.

1 Rapid Stirring Tank 2 Slow Stirring Tank 3 Wastewater Introducing Line 5 Coagulant Addition Pump 7 Precipitator 10 Coagulant Addition Pump 12 Image Acquisition Device 21 Turbidity Sensor 22 pH Sensor 23 COD Measuring Device 30 Turbidity Sensor 32 Image Acquisition Device 33 Image acquisition device 34 Image acquisition device 40 Floating tank 41 Pressurized water supply line 44 Scum skimmer 45 Gutter 47 Treated water discharge line 48 Image acquisition device 49 Image acquisition device 51 Biological reaction tank 51A Overflow part 52 Precipitation tank 52A Overflow part 55 Communication channel 57 Flocculant addition pump 58 Image acquisition device 59 Image acquisition device 100 Computing system 110 Storage device 111 Main storage device 112 Auxiliary storage device 120 Processing device 130 Input device 132 Recording medium reading device 134 Recording medium port 140 Output device 141 Display device 142 Printing Device 150 Communication device 301 On Force layer 302 Hidden layer 303 Output layer

Claims (13)

Add flocculant to wastewater to form floc,
Image data of the flocculation floc is generated by an image acquisition device,
At least the image data of the floc floc is input to a flocculant determination model constructed by a machine learning algorithm,
A wastewater treatment method, which outputs a determination result indicating excess or deficiency of a coagulant addition rate from the coagulant determination model. The wastewater treatment method according to claim 1,
Further comprising the step of generating image data of the wastewater before the coagulant is added by an image acquisition device,
A wastewater treatment method, wherein image data of the wastewater is input to the coagulant determination model in addition to the image data of the floc. The wastewater treatment method according to claim 1 or 2, wherein
The step of adding the aggregating agent to the wastewater to form the aggregating floc includes adding an inorganic aggregating agent to the drainage to form an aggregating floc, and then adding an organic aggregating agent to the wastewater to grow the aggregating floc. Is a process,
In the step of generating the image data of the floc, the first image data of the floc formed by adding the inorganic flocculant to the drainage is generated by a first image acquisition device, and the organic flocculant is generated. A step of generating second image data of the aggregated flocs grown by adding to the wastewater by a second image acquisition device,
A wastewater treatment method, wherein the first image data and the second image data are input to the coagulant determination model. The wastewater treatment method according to any one of claims 1 to 3,
Separating the agglomerated flocs from the wastewater by precipitating the agglomerated flocs to produce treated water,
Further comprising the step of generating image data of the treated water by an image acquisition device,
A wastewater treatment method, wherein image data of the treated water is input to the flocculant determination model in addition to the image data of the floc. The wastewater treatment method according to any one of claims 1 to 3,
The aggregated flocs are floated, and the flocculated flocs are removed from the wastewater to produce treated water,
Image data of the sludge consisting of the flocculation flocs removed from the wastewater, and further comprising a step of generating at least one of the image data of the treated water by an image acquisition device,
A wastewater treatment method, in which, in addition to the image data of the floc, the one or both of the image data of the sludge and the image data of the treated water is input to the coagulant determination model. The wastewater treatment method according to any one of claims 1 to 3,
Separating the floc from the wastewater to produce treated water,
Image data of the treated water is generated by an image acquisition device,
Input the image data of the treated water to a water quality judgment model constructed by a machine learning algorithm,
The wastewater treatment method further comprising the step of outputting a predicted water quality value of the treated water from the water quality determination model. The wastewater treatment method according to any one of claims 1 to 3,
Further comprising separating the aggregated flocs from the wastewater to produce treated water,
A wastewater treatment method, in which, in addition to the image data of the floc, the time difference between the step of forming the floc and the step of separating the floc from the wastewater is input to the flocculant determination model. The wastewater treatment method according to any one of claims 1 to 7,
A wastewater treatment method, wherein information data relating to wastewater treatment is input to the coagulant determination model in addition to the image data of the floc. The wastewater treatment method according to any one of claims 1 to 8,
The determination result is a wastewater treatment method used for controlling the coagulant addition rate. The wastewater treatment method according to claim 9,
Irrespective of the state of the floc flocs, the reference value of the coagulant addition rate is determined by a reference value calculation formula,
The correction value of the reference value is determined based on the determination result indicating excess or deficiency of the coagulant addition rate,
The wastewater treatment method further comprising a step of correcting the coagulant addition rate based on the reference value and the correction value. The wastewater treatment method according to any one of claims 1 to 10,
The wastewater treatment method further comprising a step of updating the coagulant determination model using a data set consisting of a combination of image data of the floc flocs and a numerical value representing an excess or deficiency of the actual coagulant addition rate. A floc formation tank for adding flocculant to waste water to form floc.
An image acquisition device for generating image data of the floc,
A computer having a flocculant judgment model constructed by a machine learning algorithm,
The computer inputs the at least the image data to the coagulant determination model and a storage device storing the coagulant determination model, and outputs a determination result indicating excess or deficiency of the coagulant addition rate from the coagulant determination model. A wastewater treatment system, which comprises a treatment device that executes a calculation for Inputting at least image data of agglomerated flocs to the input layer of a model consisting of a neural network having an input layer, a plurality of hidden layers, and an output layer,
A computer readable program recording a program that causes a computer to execute a step of outputting a numerical value indicating an excess or deficiency of a coagulant addition rate from the output layer by executing an operation according to an algorithm of a multilayer perceptron forming the neural network. recoding media.

Images