Water quality management device, water quality management method, and water quality management program

Abstract

Provided is a water quality management device capable of accurately predicting residual chlorine concentration in a water channel located away from a chlorine injection point. A water quality management device (30) comprises: a change-amount data acquisition unit (32) that acquires data of a set of a change amount of an injected chlorine concentration and a change amount of a control-target residual chlorine concentration; a model generation unit (34) that generates a model indicating a relationship between the change amount of the injected chlorine concentration and the change amount of the control-target residual chlorine concentration on the basis of the acquired data; a change-amount prediction unit (43) that predicts the change amount of the control-target residual chlorine concentration using the generated model; and an optimal-chlorine-injection-concentration prediction unit (40) that calculates the optimal chlorine injection concentration on the basis of the predicted change amount of the control-target residual chlorine concentration.

Description

Water quality management apparatus, water quality management method, and water quality management program

[0001] The present invention relates to a water quality management apparatus, a water quality management method, and a water quality management program. More specifically, the present invention relates to a water quality management apparatus, a water quality management method, and a water quality management program for use in seawater utilizing plants such as power plants.

[0002] As a countermeasure against adherent organisms such as barnacles and mussels and biofilms that adhere to seawater systems of seawater utilizing plants such as thermal power plants and nuclear power plants, there is a technology of injecting chlorine generated by electrolyzing seawater or sodium hypochlorite (hereinafter referred to as "chlorine") into a water intake.

[0003] In relation to the above technology, Patent Document 1 discloses a water quality monitoring system for parameters such as chlorine concentration, which estimates parameters that cannot be measured by a sensor based on measurement parameters sequentially acquired from the sensor.

[0004] Japanese Unexamined Patent Publication No. 2020-121241

[0005] Conventional technologies including the technology disclosed in Patent Document 1 have a problem that it is difficult to accurately predict the residual chlorine concentration in a water channel distant from a chlorine injection point.

[0006] The present invention has been made in view of the above problem, and an object of the present invention is to provide a water quality management apparatus, a water quality management method, and a water quality management program that can accurately predict the residual chlorine concentration in a water channel distant from a chlorine injection point.

[0007] <1>A water quality management apparatus for a seawater utilizing plant, comprising: a change amount data acquisition unit that acquires data of a set of a change amount of injected chlorine concentration and a change amount of residual chlorine concentration to be controlled; a model generation unit that generates a model indicating a relationship between the change amount of injected chlorine concentration and the change amount of residual chlorine concentration to be controlled based on the acquired data; a change amount prediction unit that predicts the change amount of the residual chlorine concentration to be controlled using the generated model; and an optimum chlorine injection concentration calculation unit that calculates an optimum chlorine injection concentration based on the predicted change amount of the residual chlorine concentration to be controlled, the water quality management apparatus.

[0008] <2> The water quality management device of <1>, wherein the data of the pair of the amount of change in the injected chlorine concentration and the amount of change in the controlled residual chlorine concentration includes data acquired continuously for more than one year and data acquired within one month prior to calculating the optimal chlorine injection concentration.

[0009] <3> The water quality control device according to <1> or <2>, wherein the model is a linear model.

[0010] <4> The present invention relates to a water quality management device method for a seawater utilization plant, and includes: a change amount data acquisition step of acquiring data of a pair of change amounts of injected chlorine concentration and change amounts of controlled residual chlorine concentration; a model generation step of generating a model that shows the relationship between the change amount of injected chlorine concentration and the change amount of controlled residual chlorine concentration based on the acquired data; a change amount prediction step of predicting the change amount of residual chlorine concentration using the generated model; and an optimal chlorine injection rate prediction step of calculating the optimal chlorine injection concentration based on the predicted change amount of residual chlorine concentration.

[0011] <5> The water quality management program of the present invention is a water quality management program for a seawater utilization plant, and causes a computer to execute the following steps: a change amount data acquisition step of acquiring data of a pair of change amounts of injected chlorine concentration and change amounts of controlled residual chlorine concentration; a model generation step of generating a model that shows the relationship between the change amount of injected chlorine concentration and the change amount of controlled residual chlorine concentration based on the acquired data; a change amount prediction step of predicting the change amount of controlled residual chlorine concentration using the generated model; and an optimal chlorine injection rate prediction step of calculating the optimal chlorine injection concentration based on the predicted change amount of controlled residual chlorine concentration.

[0012] According to the present invention, it is possible to provide a water quality management device, a water quality management method, and a water quality management program that can accurately predict the chlorine concentration in a waterway located far from the chlorine injection point.

[0013] Figure 1 shows the overall configuration of the seawater utilization plant of this embodiment and the chlorine concentration at each point in the seawater utilization plant 1. Figure 2 shows the verification results of the residual chlorine concentration change prediction model. Figure 3 is a functional block diagram of the water quality management device according to the embodiment of the present invention. Figure 4 is a block diagram showing an example of the data flow when determining the optimal chlorine injection rate. Figure 5 shows the screen of the prediction software including the residual chlorine concentration change prediction model. Figure 6 is a flowchart showing the processing flow of the water quality management method according to the embodiment of the present invention.

[0014] (Seawater Utilization Plant) Embodiments of the present invention will be described with reference to the drawings. Figure 1 is a diagram showing the overall configuration of the seawater utilization plant 1 of this embodiment and the chlorine concentration at each point in the seawater utilization plant 1. The block diagram indicated by arrow 101 in Figure 1 shows the overall configuration of the seawater utilization plant 1. Hereinafter, this block diagram will be referred to as block diagram 101. The graph indicated by arrow 102 in Figure 1 shows the chlorine concentration at each point in the seawater utilization plant 1. Hereinafter, this graph will be referred to as graph 102. The horizontal axis of block diagram 101 and graph 102 shows the same point in the seawater utilization plant 1.

[0015] As shown in block diagram 101, the seawater utilization plant 1 includes a condenser 2, an intake channel 11, a discharge channel 12, and a water quality control device 30. The condenser 2 is a device that cools the steam that has passed through the steam turbine, etc., and returns it to water. The intake channel 11 is a channel that sends seawater from the sea 300 to the condenser 2. The discharge channel 12 is a channel that sends seawater from the condenser 2 to the sea 300.

[0016] The condenser 2 takes in seawater from the sea 300 via the intake channel 11 and uses it as cooling water. The condenser 2 cools the steam with the taken-in seawater. The condenser 2 then discharges the seawater cooling water, which has been used for cooling, back into the sea 300 via the discharge channel 12.

[0017] (Chlorine injection point) Chlorine is injected into the seawater that is taken in. The point where chlorine is injected into the seawater that is taken in is called the chlorine injection point 121. The chlorine injection point 121 is located near the sea 300 in the intake channel 11.

[0018] (Water quality management device) The water quality management device 30 is a device that manages the water quality in the seawater utilization plant 1, such as determining the amount of chlorine to be injected.

[0019] (Chlorine Concentration) The horizontal axis of Graph 102 shows the location in the seawater utilization plant 1. The vertical axis of Graph 102 shows the residual chlorine concentration. As shown in Graph 102, the residual chlorine concentration 141 in the waterway decreases as you move away from the chlorine injection point 121. The section indicated by arrow 161 in Figure 1 is the area 161 where measures against attached organisms are particularly necessary in the management of the waterway.

[0020] (Measures against attached organisms) In order to implement the desired level of measures against attached organisms in the area 161 where measures against attached organisms are necessary, it is preferable that the residual chlorine concentration 141 in the water channel at the condenser inlet point 122 is, for example, 0.04 mg / L or higher.

[0021] (Agreement Value) From the standpoint of preventing seawater pollution, etc., the residual chlorine concentration 141 in the waterway at the discharge point 123 is usually set at an agreement value of 142. When an agreement value of 142 is set, the residual chlorine concentration 141 in the waterway at the discharge point 123 must be less than or equal to the agreement value of 142.

[0022] With conventional technology, it is difficult to achieve the desired antimicrobial activity while adhering to agreed-upon standards without requiring significant effort. This is because it is difficult to accurately predict the residual chlorine concentration of 141 in the waterway.

[0023] (Prediction of Residual Chlorine Concentration) The water quality management device 30 of this embodiment can accurately predict the residual chlorine concentration 141 in the water channel at the condenser inlet point 122. In the following description, the residual chlorine concentration 141 in the water channel at the condenser inlet point 122 will be referred to as the controlled residual chlorine concentration. If a connecting well is provided at the condenser inlet point 122, the residual chlorine concentration in the connecting well will be referred to as the controlled residual chlorine concentration.

[0024] In this embodiment, the water quality management device 30 predicts the amount of change in the target residual chlorine concentration. Then, after predicting the amount of change in the target residual chlorine concentration, it uses that amount of change to predict the amount of change in the chlorine injection concentration.

[0025] The change in the controlled residual chlorine concentration is determined as follows: Change in controlled residual chlorine concentration = f (change in injected chlorine concentration, water temperature, current injected chlorine concentration). f represents the residual chlorine concentration change prediction model f. The residual chlorine concentration change prediction model f is a model that predicts the change in the controlled residual chlorine concentration.

[0026] The data to be input into the residual chlorine concentration change prediction model f are: "the amount of change in the injected chlorine concentration during chlorine injection," "the water temperature when water is drawn from the sea 300," and "the current residual chlorine concentration at the condenser inlet point 122." Once this data is input, the residual chlorine concentration change prediction model f outputs a predicted value for the residual chlorine concentration. The water temperature when water is drawn from the sea 300 is called the intake temperature.

[0027] (Base Data) The residual chlorine concentration change prediction model f can be generated based on past data on changes in injected chlorine concentration and changes in residual chlorine concentration. This data is called the base data.

[0028] The residual chlorine concentration change prediction model f can be, for example, a linear model or a GAM (GAM: Generalized Additive Model). The linear model and GAM are generated based on the base data.

[0029] The base data can be data previously acquired at the seawater utilization plant 1 equipped with the water quality management device 30. Preferably, the base data is data acquired continuously for one year or more. By using data acquired continuously for one year or more as the base data, it is possible to generate a residual chlorine concentration change prediction model f that takes into account factors that repeatedly change throughout the seasons, such as temperature changes.

[0030] In the seawater utilization plant 1, the injected chlorine concentration at the chlorine injection point 121 and the residual chlorine concentration 141 in the water channel at the condenser inlet point 122 are normally continuously monitored. During continuous monitoring, for example, the residual chlorine concentration 141 in the water channel is acquired every 30 seconds. From the injected chlorine concentration at the chlorine injection point 121, the amount of change in the injected chlorine concentration at the chlorine injection point 121 can be determined. From the residual chlorine concentration 141 in the water channel at the condenser inlet point 122, the amount of change in the residual chlorine concentration 141 in the water channel at the condenser inlet point 122 can be determined. The data of the amount of change obtained in this way over a period of more than one year is used as base data.

[0031] (Recent Data) In addition, when generating the residual chlorine concentration change prediction model f, recent data may be considered in addition to the base data. Recent data can be, for example, data obtained within the last month, or more specifically, data from the last week or so. It is possible that the conditions of the seawater utilization plant 1 have changed since the conditions at which the base data was obtained. Therefore, the residual chlorine concentration change prediction model f is corrected based on the most recent intake water temperature, injected chlorine concentration, and residual chlorine concentration, etc. This makes the residual chlorine concentration change prediction model f a model that is in line with the current situation.

[0032] (Elapsed Time) When acquiring base data and recent data, the residual chlorine concentration can be the residual chlorine concentration after a predetermined time has elapsed since chlorine injection. This is because a predetermined time must elapse after chlorine injection before the effects of the injection become apparent. The predetermined time is preferably several hours, rather than just 10 minutes. More specifically, it can be, for example, 6 hours. This allows for obtaining a more accurate residual chlorine concentration change prediction model f.

[0033] (Exclusion of outliers) When acquiring base data and recent data, data where the change in injected chlorine concentration exceeds a predetermined value may be excluded. When the change in injected chlorine concentration is large, the behavior of the change in residual chlorine concentration may differ from that when the change in injected chlorine concentration is not large. Therefore, when the change in injected chlorine concentration is larger than the normal practical range, that data is excluded from the data used to generate the residual chlorine concentration change prediction model f. This makes it possible to generate a residual chlorine concentration change prediction model f that is more in line with actual conditions. A large change in injected chlorine concentration refers to cases where the change is, for example, ±0.25 mg / L or more.

[0034] (Verification of the Residual Chlorine Concentration Change Prediction Model) The verification results of the residual chlorine concentration change prediction model f will be explained with reference to Figure 2. Figure 2 is a graph showing the measured data of the change in residual chlorine concentration and the predicted value output by the residual chlorine concentration change prediction model f. The horizontal axis of the graph in Figure 2 shows the date. The vertical axis of the graph in Figure 2 shows the change in residual chlorine concentration (ppm).

[0035] Figure 2 shows the measured change in the controlled residual chlorine concentration and the predicted value output by the residual chlorine concentration change prediction model f for one month. The base data used to create the residual chlorine concentration change prediction model f was the data from the most recent year. The plotted values ​​in the graph are the daily averages. The controlled residual chlorine concentration refers to the residual chlorine concentration in the junction well.

[0036] As shown in Figure 2, the predicted values ​​output by the residual chlorine concentration change prediction model f closely match the measured values. It was confirmed that the predicted values ​​output by the residual chlorine concentration change prediction model f almost perfectly reproduce the measured values. Furthermore, the linear model showed higher reproducibility than the GAM model.

[0037] (Functional Block of Water Quality Management Device) The water quality management device 30 of this embodiment will be described with reference to Figure 3. Figure 3 is a functional block diagram of the water quality management device 30 of this embodiment. As shown in Figure 3, the water quality management device 30 includes a change amount data acquisition unit 32, a residual chlorine concentration change amount prediction model generation unit 34, an optimal chlorine injection rate prediction unit 40, and an input / output unit 36. The optimal chlorine injection rate prediction unit 40 includes a control target residual chlorine concentration change amount prediction unit 42.

[0038] The change amount data acquisition unit 32 is responsible for acquiring base data and the most recent data. Specifically, the change amount data acquisition unit 32 acquires the injected chlorine concentration at the chlorine injection point 121 and the residual chlorine concentration 141 in the waterway at the condenser inlet point 122. At that time, it acquires data in the form of a pair of the amount of change in the injected chlorine concentration and the amount of change in the residual chlorine concentration in the waterway when that change is made. The change amount data acquisition unit 32 includes, for example, a sensor provided at the chlorine injection point 121. This sensor measures the concentration of the injected chlorine. The change amount data acquisition unit 32 also includes, for example, a sensor provided at the junction well. This sensor measures the residual chlorine concentration in the waterway.

[0039] The residual chlorine concentration change prediction model generation unit 34 (model generation unit) is the part that generates the residual chlorine concentration change prediction model f. The residual chlorine concentration change prediction model generation unit 34 acquires base data and recent data, etc. The residual chlorine concentration change prediction model generation unit 34 generates the residual chlorine concentration change prediction model f by analyzing or learning from the acquired data. The generated residual chlorine concentration change prediction model f can be a linear model or a GAM, etc.

[0040] The optimal chlorine injection rate prediction unit 40 is a part that determines the chlorine injection rate to be injected to achieve a target residual chlorine concentration. This chlorine injection rate is referred to as the optimal chlorine injection rate. The optimal chlorine injection rate prediction unit 40 includes a residual chlorine concentration change amount prediction unit 42. The optimal chlorine injection rate prediction unit 40 predicts the optimal chlorine injection rate based on the residual chlorine concentration change amount predicted by the residual chlorine concentration change amount prediction unit 42 and the corresponding change amount of injected chlorine concentration. Note that the optimal chlorine injection rate prediction unit 40 may predict the optimal chlorine injection rate using a model such as, for example, an optimal chlorine injection rate prediction model. The optimal chlorine injection rate prediction model will be described later with reference to FIG. 4.

[0041] The residual chlorine concentration change amount prediction unit 42 (change amount prediction unit) is a part that predicts the residual chlorine concentration change amount required to achieve the target residual chlorine concentration and the change amount of injected chlorine concentration required to generate said residual chlorine concentration change amount. This prediction is performed using a residual chlorine concentration change amount prediction model f.

[0042] The input / output unit 36 is a part that inputs data necessary for determining the optimal chlorine injection rate, and outputs the determined optimal chlorine injection rate.

[0043] (Example of Input and Output to the Model) With reference to FIG. 4, an example of data input and output when determining the optimal chlorine injection rate will be described. FIG. 4 is a block diagram showing an example of a data flow when determining the optimal chlorine injection rate. FIG. 4 illustrates a case where the residual chlorine concentration in the junction well is set as a target value for the residual chlorine concentration.

[0044] Determination of the optimal chlorine injection rate is performed using an optimal chlorine injection rate prediction model. The optimal chlorine injection rate prediction model is a prediction model that takes, as input items, the target value of residual chlorine concentration and the current residual chlorine concentration, etc., and outputs the optimal chlorine injection rate.

[0045] Specifically, the input items are the target value Y of the residual chlorine concentration in the junction well and the current value. The current value refers to the current residual chlorine concentration in the junction well and the current intake water temperature. For example, an operator of the water quality management apparatus 30 may input the input items to the optimal chlorine injection rate prediction model.

[0046] In addition, as data to be prepared in advance, data of the residual chlorine concentration in the junction well, intake water temperature and injected chlorine concentration for approximately the most recent one week shall be prepared in advance. These data may be input in advance into the optimal chlorine injection rate prediction model or the residual chlorine concentration change amount prediction model. Alternatively, they may be input together with the current values when obtaining the optimal chlorine injection rate.

[0047] Note that the residual chlorine concentration in the junction well may be measured with a continuous residual chlorine concentration analyzer. In addition, the intake water temperature and the injected chlorine concentration can be acquired from the plant data management system. For example, these data measured every 30 seconds can be input in csv file format to the optimal chlorine injection rate prediction model or the residual chlorine concentration change amount prediction model.

[0048] The optimal chlorine injection rate prediction model predicts the optimal chlorine injection concentration (mg / L) based on the residual chlorine concentration change amount predicted by the residual chlorine concentration change amount prediction model and the corresponding change amount of the injected chlorine concentration.

[0049] (Prediction Software) With reference to Fig. 5, the prediction software including the residual chlorine concentration change amount prediction model will be described. Fig. 5 is a diagram showing a screen of the prediction software. In the example shown in Fig. 5, the prediction software constitutes a part of the chlorine injection prediction system. The prediction software shown in Fig. 5 uses a linear model as the residual chlorine concentration change amount prediction model f. The residual chlorine concentration change amount prediction model f used by the prediction software is not limited to a linear model, and for example, GAM can also be used.

[0050] The operation of obtaining the optimal chlorine injection concentration using the prediction software can be performed in the following three steps from (1) to (3): (1) inputting parameters required for prediction, (2) clicking a "Run Prediction (Linear Model)" button, (3) displaying the optimal chlorine injection concentration for adjusting the residual chlorine concentration to a target value. In Fig. 5, display positions corresponding to each step are indicated by (1) to (3).

[0051] The prediction software defines the optimal injection concentration as the point where the input target value of the residual chlorine concentration in the junction well (B) intersects with the prediction formula (A) derived from the linear model. In the example shown in Figure 5, the current optimal injection concentration is 0.42 [mg / L]. In addition to the optimal injection concentration, the prediction software may also display the estimated error range (C) of the predicted value. The estimated error range of the predicted value is the expected daily variation in residual chlorine concentration when chlorine is injected at the optimal injection concentration. The estimated error range of the predicted value can be calculated using statistical methods based on the training data used when generating the prediction model.

[0052] Thus, this prediction software allows you to determine the optimal chlorine injection concentration with simple operation. Furthermore, the aforementioned prediction software can be operated standalone. Operating the prediction software standalone helps to suppress information leakage.

[0053] (Water Quality Management Method) The processing flow of the water quality management method of this embodiment will be explained with reference to Figure 6. Figure 6 is a flowchart showing the processing flow of the water quality management method of this embodiment. In the following description and drawings, S1 means step 1. The same applies to S2 and subsequent steps.

[0054] (S1) S1 is the change amount data acquisition step. In S1, the change amount data acquisition unit 32 acquires the base data and the most recent data.

[0055] (S2) S2 is the residual chlorine concentration change prediction model generation step (model generation step). In S2, the residual chlorine concentration change prediction model generation unit 34 generates a residual chlorine concentration change prediction model f based on the data acquired in S1.

[0056] (S3) S3 is the most recent data input step. In S3, as explained with reference to Figure 4, data such as residual chlorine concentration data for the past week or so is input. This improves the prediction accuracy of the residual chlorine concentration change prediction model f.

[0057] (S4) S4 is the step for inputting the target value and current value of the residual chlorine concentration. In S4, for example, the target value and current value of the residual chlorine concentration are input into the prediction software. The current value is the current residual chlorine concentration value and the current intake water temperature value, as explained with reference to Figure 4.

[0058] (S5) S5 is the residual chlorine concentration change prediction step. In S5, the residual chlorine concentration change is predicted using the residual chlorine concentration change prediction model f.

[0059] (S6) S6 is the optimal chlorine injection rate prediction step. In S6, the optimal chlorine injection rate is predicted by taking into account the target value and current value of residual chlorine concentration entered in S4, as well as the amount of residual chlorine concentration change predicted in S5.

[0060] (S7) S7 is the output value setting step. In S7, the optimal chlorine injection rate predicted in S6 is set to, for example, the water quality management device 30. Specifically, the optimal chlorine injection rate is set to, for example, an electrolytic device that generates chlorine. The step ends in S7. Note that the order of the steps described above is an example. The order of the steps can be changed as appropriate.

[0061] (Water Quality Management Program) The water quality management method using the water quality management device is implemented by software. When implemented by software, the program constituting this software is installed on a computer that serves as the water quality management device. The program may also be distributed to users by being recorded on removable media, or by being downloaded to the user's computer via a network. Furthermore, these programs may be provided to the user's computer as a web service via a network without being downloaded.

[0062] Embodiments of the present invention have been described above. The present invention is not limited to the embodiments described above, and various modifications, variations, and combinations are possible. For example, the change amount data acquisition unit 32 and the residual chlorine concentration change amount prediction model generation unit 34 may be provided separately from the water quality management device 30.

[0063] Conventional methods of predicting injection concentration from the controlled residual chlorine concentration may result in an overestimation of the injection concentration. This is likely due to the small contribution of changes in injection concentration to the residual chlorine concentration. In contrast, this disclosure predicts residual chlorine concentration using the relationship between the change in injection chlorine concentration and the change in residual chlorine concentration. Therefore, it is possible to accurately predict residual chlorine concentration in waterways far from the chlorine injection point.

[0064] 1 Seawater utilization plant 2 Condenser 11 Intake channel 12 Discharge channel 30 Water quality management device 32 Change amount data acquisition unit 34 Residual chlorine concentration change amount prediction model generation unit 36 ​​Input / output unit 40 Optimal chlorine injection rate prediction unit 42 Control target residual chlorine concentration change amount prediction unit 42 Residual chlorine concentration change amount prediction unit 121 Chlorine injection point 122 Condenser inlet point 123 Discharge outlet point 141 Residual chlorine concentration in the water channel 142 Agreed value 300 Sea

Claims

1. A water quality management device for a seawater utilization plant, comprising: a change amount data acquisition unit that acquires data of a pair of changes in the injected chlorine concentration and changes in the controlled residual chlorine concentration; a model generation unit that generates a model showing the relationship between the change in the injected chlorine concentration and the change in the controlled residual chlorine concentration based on the acquired data; a change amount prediction unit that predicts the change in the controlled residual chlorine concentration using the generated model; and an optimal chlorine injection rate prediction unit that calculates the optimal chlorine injection concentration based on the predicted change in the controlled residual chlorine concentration.

2. The water quality management device according to claim 1, wherein the data set of the amount of change in the injected chlorine concentration and the amount of change in the controlled residual chlorine concentration includes data acquired continuously for more than one year and data acquired within one month prior to calculating the optimal chlorine injection concentration.

3. The water quality control device according to claim 1 or 2, wherein the model is a linear model.

4. A water quality management device method for a seawater utilization plant, comprising: a change amount data acquisition step of acquiring data of a pair of change amounts in injected chlorine concentration and change amounts in the controlled residual chlorine concentration; a model generation step of generating a model showing the relationship between the change amount in injected chlorine concentration and the change amount in the controlled residual chlorine concentration based on the acquired data; a change amount prediction step of predicting the change amount in residual chlorine concentration using the generated model; and an optimal chlorine injection rate prediction step of calculating the optimal chlorine injection concentration based on the predicted change amount in residual chlorine concentration.

5. A water quality management program for a seawater utilization plant, which causes a computer to perform the following steps: a change amount data acquisition step of acquiring data of a pair of changes in injected chlorine concentration and changes in the controlled residual chlorine concentration; a model generation step of generating a model showing the relationship between the change in injected chlorine concentration and the change in the controlled residual chlorine concentration based on the acquired data; a change amount prediction step of predicting the change in the controlled residual chlorine concentration using the generated model; and an optimal chlorine injection rate prediction step of calculating the optimal chlorine injection concentration based on the predicted change in the controlled residual chlorine concentration.

Images