Water treatment system

Through data exchange and model export between representative series and subordinate series, it is estimated that the water quality of the sensor series is not set, which solves the problems of unstable control and high cost in the existing water treatment system, and achieves efficient and stable water quality control.

CN118946531APending Publication Date: 2024-11-12MITSUBISHI ELECTRIC CORP
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202280094354.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2022-04-15
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

Existing water treatment systems fail to fully consider individual differences between series when controlling water quality, resulting in instability in control of control and increased costs due to the demand for the number of sensors.

Method used

A water treatment system was designed to estimate the water quality of the sensor series without setting through data exchange and model derives between the representative series and subordinate series, and to use the water quality data model of the representative series and subordinate series to control the water quality to reduce the number of sensors. .

Benefits of technology

It realizes efficient water quality control on the basis of taking into account individual differences between series, reducing the cost of sensor setup and improving the stability of control.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118946531A_ABST
    Figure CN118946531A_ABST
Patent Text Reader

Abstract

A water treatment system according to the present invention comprises: a representative series for purifying water to be treated; a subordinate series for purifying the water to be treated; water quality control units (107, 207) for controlling the water quality of the water to be treated; a subordinate series water quality data acquisition unit (209) that acquires subordinate series water quality items; a representative series water quality data acquisition unit (110) for acquiring water quality items of the same type as the water quality items acquired by the subordinate series water quality data acquisition unit (209) and water quality items different from the water quality items acquired by the subordinate series water quality data acquisition unit (209); a representative series water quality variation model derivation unit (302) that derives a representative series water quality variation model; a subordinate-series water quality variation model derivation unit (303) that derives a subordinate-series water quality variation model; and an estimation unit (304) that estimates, on the basis of the representative series water quality variation model, the subordinate series water quality variation model, and the data stored in the data storage unit, a water quality item that is included in the representative series water quality data and that is different from the water quality item acquired by the subordinate series water quality data acquisition unit.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to a water treatment system for performing purification treatment by measuring the water quality of treated water such as sewage. Background Art

[0002] The activated sludge method is widely used in sewage treatment. The activated sludge method is a treatment method that activates aerobic microorganisms such as nitrifying bacteria in sewage by supplying air through a blower or the like, thereby treating ammonia nitrogen and other water pollutants in sewage.

[0003] In the activated sludge process, in order to fully remove water pollutants in sewage, it is necessary to measure water quality such as the concentration of water pollutants to be treated in sewage, and control the blower based on the measured value. However, in an existing water treatment system composed of multiple series, when sensors for measuring water quality are set in all series, sensors proportional to the number of series are required, which leads to a problem of requiring a large introduction cost.

[0004] In view of this problem, in an existing water treatment system composed of multiple series, when the water quality of a certain series of treated water is controlled, the water quality of the series as the control object is referenced by water quality data of other series instead of directly referring to the measured value of the series, in an effort to reduce the number of sensors that need to be installed. For example, in Patent Document 1, an ammonia nitrogen concentration meter and a dissolved oxygen concentration meter are configured in a certain series, while only a dissolved oxygen concentration meter is set in other series. In the water quality control of the series provided with the ammonia nitrogen concentration meter, the measured value of the ammonia nitrogen concentration meter is used for water quality control, and in the water quality control of the series provided with only the dissolved oxygen concentration meter, the water quality control is performed by referring not only to the measured value of the dissolved oxygen concentration meter of the series, but also to the value of the ammonia nitrogen concentration of the series provided with the ammonia nitrogen concentration meter, thereby trying to reduce the number of sensors that need to be installed. Prior art literature Patent Literature

[0005] Patent Document 1: Japanese Patent Application Publication No. 2005-199115 Summary of the invention Technical problem to be solved by the invention

[0006] However, in the above-mentioned conventional water treatment system, although the values ​​that can be measured in other series are used for control, the water quality of the series itself as the control object is not estimated. Therefore, since the individual differences of the equipment and sensors in each series are not taken into account, the control sometimes becomes unstable.

[0007] The present disclosure has been made in view of the above situation, and its object is to obtain a water treatment system that can perform water quality control by estimating the water quality of a series without a sensor based on the measurement value of a sensor set in other series while considering individual differences between series. Technical means for solving technical problems

[0008] In order to solve the above-mentioned problems and achieve the above-mentioned purposes, the present invention includes: a representative series for purifying the treated water; a subordinate series for purifying the treated water; a water quality control unit, which is arranged in the representative series and the subordinate series to control the water quality of the treated water; a subordinate series water quality data acquisition unit, which is arranged in the subordinate series to acquire at least one water quality item from the treated water; a representative series water quality data acquisition unit, which is arranged in the representative series to acquire at least one water quality item of the same type as the water quality items acquired by the subordinate series water quality data acquisition unit, and at least one water quality item different from the water quality items acquired by the subordinate series water quality data acquisition unit; a data storage unit, which stores control data representing the control content performed by the water quality control unit, the representative series water quality data acquired by the representative series water quality data acquisition unit, and the subordinate series water quality data. A representative series water quality change model derivation unit, which derives a representative series water quality change model based on the data stored in the data storage unit, and the representative series water quality change model is a mathematical model representing the changes of water quality items contained in the representative series water quality data; a subordinate series water quality change model derivation unit, which derives a subordinate series water quality change model based on the representative series water quality change model and the data stored in the data storage unit, and the subordinate series water quality change model is a mathematical model representing the changes of water quality items obtained by the representative series water quality data acquisition unit in the subordinate series; and an estimating unit, which estimates water quality items contained in the representative series water quality data that are different from the water quality items obtained by the subordinate series water quality data acquisition unit based on the representative series water quality change model, the subordinate series water quality change model and the data stored in the data storage unit. Effects of the Invention

[0009] According to the present disclosure, there is provided a water treatment system capable of performing water quality control by estimating water quality in a series where no sensor is provided based on sensor measurement values ​​provided in other series while taking individual differences between series into consideration. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 This is a diagram showing a schematic configuration of a water treatment system according to Embodiment 1. Figure 2 This is a diagram showing a schematic configuration of a water treatment system according to a second embodiment. Figure 3 This is a diagram showing an example of the hardware configuration of the data storage unit and the processing unit in the first and second embodiments. DETAILED DESCRIPTION

[0011] Hereinafter, a water treatment system according to an embodiment of the present disclosure will be described in detail based on the drawings.

[0012] Implementation method 1. Figure 1 This is a diagram showing a schematic configuration of a water treatment system according to Embodiment 1. Figure 1 The water treatment system of the present disclosure composed of more than two series is shown. To simplify the description, a water treatment system composed of two series, series 1 and series 2, is taken as an example for description.

[0013] Series 1 is a series including an anaerobic tank 102, an anaerobic tank 103, an aerobic tank 104, and a sedimentation tank 105. Sewage as treated water flows into series 1 from the outside through a pump 101, and then passes through the anaerobic tank 102, the anaerobic tank 103, the aerobic tank 104, and the sedimentation tank 105, so that the sewage is treated and discharged. In addition, in series 1, a water quality control unit 107 for controlling the water quality of the treated water in series 1 is provided. In this embodiment 1, the water quality control unit 107 is connected to the blower 106 provided in the aerobic tank 104, and the water quality control unit 107 aerates the treated water in the aerobic tank 104 by sending an aeration amount instruction value to the blower 106, and removes water quality pollutants such as ammonia nitrogen. In addition, a sensor for measuring the water quality items of the treated water contained in the aerobic tank 104 is provided in the aerobic tank 104. In this example, two sensors are provided: a dissolved oxygen concentration meter (sensor 108) for measuring the concentration of oxygen dissolved in the water to be treated and an ammonia nitrogen concentration meter (sensor 109) for measuring the concentration of ammonia nitrogen contained in the water to be treated.

[0014] Like series 1, series 2 is a series including an anaerobic tank 202, an anaerobic tank 203, an aerobic tank 204, and a sedimentation tank 205. Sewage as treated water flows into series 2 from the outside through a pump 201, and then passes through the anaerobic tank 202, the anaerobic tank 203, the aerobic tank 204, and the sedimentation tank 205, so that the sewage flows in and is treated and discharged. In addition, in series 2, a water quality control unit 207 for controlling the water quality of the treated water in series 2 is provided. In this embodiment 1, the water quality control unit 207 is connected to the blower 206 provided in the aerobic tank 204, and the water quality control unit 207 aerates the treated water in the aerobic tank 204 by sending an aeration amount instruction value to the blower 206, and removes water quality pollutants such as ammonia nitrogen. In addition, a sensor for measuring the water quality items of the treated water included in the aerobic tank 204 is provided in the aerobic tank 204. Different from the series 1, only one sensor is provided in the series 2, that is, only a dissolved oxygen concentration meter (sensor 208) for measuring the concentration of oxygen dissolved in the water to be treated.

[0015] In embodiment 1, a series capable of measuring two or more water quality items among the water quality items contained in the treated water is treated as a series representing all series and is referred to as a representative series. In addition, a series capable of measuring at least one or more water quality items among the water quality items measurable in the representative series and the water quality items contained in the treated water in each series is treated as a series subordinate to the representative series and is referred to as a subordinate series. Among the water quality items measurable in the representative series, at least one or more water quality items are different from the water quality items available in the subordinate series. That is, series 1 is treated as a representative series because a dissolved oxygen concentration meter (sensor 108) and an ammonia nitrogen concentration meter (sensor 109) are provided, and series 2 is treated as a subordinate series because only a dissolved oxygen concentration meter (sensor 208) is provided.

[0016] For the representative series, a representative series water quality data acquisition unit 110 is provided for acquiring water quality items that can be measured in the representative series, and for each subordinate series, a subordinate series water quality data acquisition unit 209 is provided for acquiring water quality items that can be measured in each subordinate series. The water quality data collected by the representative series water quality data acquisition unit 110 and the subordinate series water quality data acquisition unit 209 are sent to the data storage unit 301 and stored. In addition, the data storage unit 301 not only stores the water quality data of each series, but also stores control data indicating the control content applied to the treated water in each series. The control data is stored by sending the control data to the data storage unit 301 by the water quality control unit (water quality control unit 107 and water quality control unit 207) provided in each series. In the first embodiment, in order to control the blowers (blower 106 and blower 206) connected to the water quality control units 107 and 207, the aeration amount instruction values ​​sent to the blowers 106 and 206 are also sent to and stored in the data storage unit 301. In addition, the data storage unit 301 also stores the parameters included in the representative series water quality change model, the parameter correction amount included in the subordinate series water quality change model, and the data on the water quality estimated value of the subordinate series.

[0017] The water quality treatment system includes a processing unit 300. The processing unit 300 includes a representative series water quality change model deriving unit 302, a subordinate series water quality change model deriving unit 303, and an estimation unit 304. The representative series water quality model deriving unit 302 derives a representative series water quality change model based on various data stored in the data storage unit 301. The representative series water quality change model is a mathematical model that represents the change of water quality items that can be obtained in the representative series in the representative series. In the first embodiment, since the water quality items that can be obtained in the representative series are the dissolved oxygen concentration obtained by the sensor 108 and the ammonia nitrogen concentration obtained by the sensor 109, the representative series water quality change model deriving unit 302 derives a mathematical model that represents the change of these two water quality items in the representative series.

[0018] The subordinate series water quality change model derivation unit 303 derives the subordinate series water quality change model. The subordinate series water quality change model is a mathematical model that represents the change of water quality items that can be obtained in the representative series in each subordinate series based on the representative series water quality change model and various data stored in the data storage unit 301. Although the example of a subordinate series being one series is shown in the first embodiment, there may be multiple subordinate series, and the derived subordinate series water quality change model is not necessarily one.

[0019] In this embodiment 1, the water quality items that can be obtained in the representative series are the two types of dissolved oxygen concentration and ammonia nitrogen concentration as described above, and the subordinate series is only series 2, so a mathematical model representing the changes of these two water quality items in series 2 is derived from the subordinate series water quality change model derivation unit.

[0020] The estimation unit 304 estimates the parameters included in the representative series water quality change model, the water quality items available in the representative series in each subordinate series, and the parameter correction amount included in each subordinate series water quality change model based on the representative series water quality change model, each subordinate series water quality change model, and various data stored in the data storage unit 301. The estimation results are stored in the data storage unit 301.

[0021] The output unit 305 outputs various estimation results estimated by the estimation unit 304 and stored in the data storage unit 301. The output means may be in any form, such as output as a video on a display, output to paper on a printer, or output to other devices via a network.

[0022] The operation is described below. The representative series water quality change model is derived in the representative series water quality change model derivation unit 302. As described above, in the present embodiment 1, a mathematical model of the dissolved oxygen concentration and the ammonia nitrogen concentration in the series 1 as the representative series is derived. In the present embodiment 1, as an example of a mathematical model applicable to the representative series water quality change model, there is an activated sludge model (ASM). The activated sludge model is a series of mathematical models for the main water quality items contained in sewage advocated by the International Water Association (IWA). The mathematical models of versions such as ASM1, ASM2, ASM2D, and ASM3 are recorded in "Activated Sludge Model - ASM1, ASM2, ASM2D, ASM3 by Toshi Miyano, published by Environmental News Agency, January 31, 2005".

[0023] For example, a mathematical model regarding ammonia nitrogen concentration in ASM1 is proposed as follows.

[0024] [Mathematical formula 1]

[0025] The variables and parameters in mathematical formula (1) are as follows.

[0026]

Table 1

[0027] In addition, a simpler linear model as follows is also applicable to the first embodiment.

[0028] [Mathematical formula 2] x NH (t+1)=θ NH,NH x NH (t)+θ NH,o x o (t)+θ NH,u,b u b (t)+θ NH,B …(2)

[0029] The variables and parameters in mathematical formula (2) are as follows.

[0030]

Table 2

[0031] In addition, any mathematical model can be used as the representative series water quality change model in the present embodiment 1. In the embodiment 1, the description is based on the case where the linear model represented by the mathematical formula (2) is applied as the representative series water quality change model.

[0032] In the representative series, if the concentration of ammonia nitrogen at time t is x ( r ) NH (t), the dissolved oxygen concentration is set to x ( r ) 0(t), the aeration rate is set to u ( r ) NH (t), the representative series water quality change model in the first embodiment is expressed as follows.

[0033] [Mathematical formula 3]

[0034] [Formula 4]

[0035] In the representative series water quality change model derivation unit, the initial estimated values ​​of the parameters in the representative series water quality change model are derived using the data stored in the data storage unit. In this example, the parameters, θ, and θ in the representative series water quality change model represented by mathematical formula (3) and mathematical formula (4) are derived. (r) NH,NH ,θ (r) NH,0 ,θ (r) NH,u,b ,θ (r) NH,B ,θ (r)0,0 ,θ (r) 0,NH ,θ (r) 0,u,b and θ (r) 0,B Initial estimated value θ^ (r) NH,NH ,θ^ (r) NH,0 ,θ^ (r) NH,u,b ,θ^ (r) NH,B ,θ^ (r) 0,0 ,θ^ (r) 0,NH ,θ^ (r) 0,u,b and θ^ (r) 0,B In addition, when ^ is marked, it indicates an initial estimated value.

[0036] When deriving the parameters in the linear model represented by equations (3) and (4), in the first embodiment, the parameters are derived by the least squares method. First, when the initial estimated value θ^ of the parameter in equation (3) is derived by the least squares method, (r) NH,NH ,θ^ (r) NH,0 ,θ^ (r) NH,u,b ,θ^ (r) NH,B When using the data stored in the data storage department, from a certain time T earlier than the current time to T+N T -1 represents a series of ammonia nitrogen concentration, dissolved oxygen concentration, and aeration volume, and the matrix A (T:T+N-1) is obtained:

[0037] [Formula 5]

[0038] Then, the data from a certain time in the past earlier than the current time T+1 to T+N which is also stored in the data storage unit is used. T The ammonia nitrogen concentration, dissolved oxygen concentration and aeration volume between the two represent a series, and the vector b(T+1:T+N T ),

[0039] [Mathematical formula 6]

[0040] Then, a vector formed by arranging the initial estimated values ​​of the parameters to be derived is derived as follows.

[0041] [Mathematical formula 7]

[0042] In the mathematical formula, a matrix with -1 on the upper right side thereof represents an inverse matrix of the target matrix, and a matrix with T on the upper right side thereof represents a transposed matrix of the target matrix.

[0043] [Mathematical formula 8] θ^ (r) (0)=(A T (T:T+N T -1)A(T:T+N T -1) -1 A T (T:T+N T -1)b(T+1:T+N T )…(8)

[0044] Similarly, the initial estimated values ​​of the parameters in equation (4), θ^ (r) 0,0 ,θ^ (r) 0,NH ,θ^ (r) 0,u,b and θ^ (r) 0,B . In addition, in the present embodiment 1, the case where the variables processed in the representative series water quality change model are only two variables, namely, the ammonia nitrogen concentration and the dissolved oxygen concentration, is described. However, even when the number of variables increases, the initial estimated values ​​of the parameters in the representative series water quality change model can be similarly derived by naturally extending the above method. In addition, even when the representative series water quality change model is not a linear model such as mathematical formula (2) but a nonlinear model such as mathematical formula (1), the initial estimated values ​​of the parameters can be derived by a nonlinear least squares method such as the Gauss-Newton method. Therefore, the initial estimated values ​​of the parameters included in the representative series water quality change model can be derived without relying on the content of the representative series water quality change model and using the data stored in the data storage unit 301 without compromising generality.

[0045] Next, the operation of the subordinate series water quality change model derivation unit 303 will be described. In the subordinate series water quality change model derivation unit 303, each subordinate series water quality change model is derived. As described above, in this example, the mathematical model of the dissolved oxygen concentration and the ammonia nitrogen concentration in the series 2 as the subordinate series is derived. When deriving each subordinate series water quality change model, if the difference between the series is taken into account, each series is a series that constitutes the same water treatment system, so it can be considered that sewage of the same water quality flows into each series, and the sewage treatment environment such as water temperature is also basically the same between the series. Therefore, between the series, although each device and each sensor will produce individual differences, when a certain series is used as a benchmark, it can be considered that the difference between the series used as the benchmark and other series is small. In implementation mode 1, the series used as the benchmark is set as a representative series, considering that the water quality changes in each subordinate series show changes similar to the water quality changes in the representative series, it is considered that each subordinate series water quality change model can be derived by adding a sufficiently small parameter correction amount to the parameters in the representative series water quality change model.

[0046] Taking the representative series water quality change model as the mathematical formula (3) and the mathematical formula (4) as an example, the process of the subordinate series water quality change model derivation unit 303 deriving the water quality change model of the series 2 as the subordinate series is described. As described above, the parameter θ included in the representative series water quality change model is used. (r) NH,NH ,θ (r) NH,0 ,θ (r) NH,u,b ,θ (r) NH,B ,θ (r) 0,0 ,θ (r) 0,NH ,θ (r) 0,u,b and θ (r) 0,B Here, in series 2, the concentration of ammonia nitrogen at time t is set to x (s,2) NH (t), the dissolved oxygen concentration is set to x (s,2) 0(t), the aeration rate is set to u (s,2) 0(t), the parameter correction amount for each parameter in the water quality change model is set to Δθ (s,2) NH,NH , Δθ (s,2) NH,0 , Δθ (s,2) NH,u,b , Δθ (s,2) NH,B , Δθ (s,2)0,0 , Δθ (s,2) 0,NH , Δθ (s,2) 0,u,b and Δθ (s,2) 0,B , the subordinate series water quality change model of series 2 can be expressed as follows.

[0047] [Mathematical formula 9]

[0048] [Formula 10]

[0049] In the subordinate water quality change model derivation unit, an initial estimated value of a parameter correction amount included in the subordinate water quality change model is derived. In this example, the parameter correction amount Δθ in the subordinate water quality change model represented by equations (9) and (10) is derived. (s,2) NH,NH , Δθ (s,2) NH,0 , Δθ (s,2) NH,u,b , Δθ (s,2) NH,B , Δθ (s,2) 0,0 , Δθ (s,2) 0,NH , Δθ (s,2) 0,u,b and Δθ (s,2) 0,B The initial estimated value of Δθ^ (s,2) NH,NH , Δθ^ (s,2) NH,0 , Δθ^ (s,2) NH,u,b , Δθ^ (s,2) NH,B , Δθ^ (s,2) 0,0 , Δθ^ (s,2) 0,NH , Δθ^ (s,2) 0,u,b and Δθ^ (s,2) 0,B However, unlike the representative series, there is no data for uniquely determining the initial estimated values ​​of the parameter correction amounts. Therefore, these initial estimated values ​​are all set to 0, and in the estimation unit 304, while estimating the water quality items in each subordinate series that can be measured in the representative series, the estimated values ​​of these parameter correction amounts are updated and processed to be close to the true values.

[0050] Next, the operation of the estimation unit 304 will be described. As described above, the estimation unit 304 estimates the water quality items that can be measured in the representative series in each subordinate series, and estimates the parameters in the representative series water quality change model and the parameter correction amount included in the water quality change model of each subordinate series. In the first embodiment, the ammonia nitrogen concentration of the series 2 as the subordinate series, the parameters in the various models, and the parameter correction amount are estimated.

[0051] In the first embodiment, the estimation process is performed by a particle filter. A particle filter is an algorithm that performs estimation by assuming that there are multiple virtual state variables called particles and using these particles to approximate the probability distribution of the state variables to be estimated. In the estimation of the particle filter, it is first necessary to define the state equation and observation equation of the estimated system. Regarding the state equation, in this system, various parameters and parameter correction amounts required to represent the water quality items as measurement objects in the representative series and each subordinate series and their water quality changes constitute the state equation of the system. In addition, regarding the observation equation, in this system, the measurable water quality items in the representative series and each subordinate series constitute the observation equation of the system.

[0052] In the first embodiment, the variables and parameters constituting the state equation are the ammonia nitrogen concentration and the dissolved oxygen concentration in the series 1 as the representative series and the series 2 as the subordinate series, and the parameter θ required to express them. (r) NH,NH ,θ (r) NH,0 ,θ (r) NH,u,b ,θ (r) NH,B ,θ (r) 0,0 ,θ (r) 0,NH ,θ (r) 0,N,b ,θ (r) 0,B , parameter correction value Δθ (s,2) NH,NH , Δθ (S,2) NH,0 , Δθ (s,2) NH,u,b , Δθ (s,2) NH,B , Δθ (s,2) 0,0 , Δθ (s,2) 0,N H、Δθ (s,2) 0,u,b and Δθ (s,2) 0,BIn addition, the variables constituting the observation equation are the dissolved oxygen concentration and the ammonia nitrogen concentration of the series 1 which can be measured by the sensor 108 and the sensor 109, and the dissolved oxygen concentration in the series 2 which can be measured by the sensor 208. If these state equations and observation equations use the above-mentioned representative series water quality change model and subordinate series water quality change model, these state equations and observation equations can be expressed as follows.

[0053] [Mathematical formula 11] x(t+1)=f(x(t),u(t))+w x (t)…(11)

[0054] [Mathematical formula 12] y(t)=h(x(t))+w y (t)…(12)

[0055] Mathematical formula (11) represents the state equation, and mathematical formula (12) represents the observation equation of the water treatment system. In mathematical formula (11), x(t) is a vector composed of various parameters and parameter correction amounts required to represent the water quality items to be measured in the representative series and each subordinate series at time t and their water quality changes, and is composed of the following vectors.

[0056] [Mathematical formula 13]

[0057] [Formula 14]

[0058] [Mathematical formula 15]

[0059] [Formula 16]

[0060] [Mathematical formula 17]

[0061] In addition, the vector u(t) in the mathematical formula (11) is a vector composed of control command values ​​for the system. In the present embodiment 1, the control command values ​​are aeration amount command values ​​for the blower 106 or the blower 206 output by the water quality control unit 107 of the series 1 and the water quality control unit 207 of the series 2, so the vector u(t) can be expressed as follows.

[0062] [Mathematical formula 18]

[0063] In addition, f(x(t), u(t)) in mathematical formula (11) is a vector function corresponding to each element constituting x(t+1) on the left side, and in this example, is constructed as follows.

[0064] [Mathematical formula 19]

[0065] The vector function f that constitutes mathematical formula (19) (x(r)) (x(t), u(t)) corresponds to the water quality change model representing the series, the first dimension is equal to the right side of mathematical formula (3), and the second dimension is equal to the right side of mathematical formula (4). The vector function f (x(s,2)) (x(t), u(t)) corresponds to the subordinate series water quality change model of series 2, the first dimension is equal to the right side of mathematical formula (9), and the second dimension is equal to the right side of mathematical formula (10). The vector function f (θ(r)) (x(t), u(t)) corresponds to the parameters included in the model representing the series of water quality changes, and the vector function f (Δθ(s,2)) (x(t), u(t)) corresponds to the parameter correction amount included in the subordinate series water quality change model of series 2. Since these values ​​are constants, each vector function can be expressed as follows.

[0066] [Mathematical formula 20] f (θr)) (x(t),u(t))=θ (r) (t)…(20)

[0067] [Mathematical formula 21]

[0068] In addition, the vector w included in mathematical formula (11) x (t) is the system noise added to the system state at time t. In this example, assume that w x Each element of (t) is normally distributed with a mean value of 0, and the system user can set any value for the variance of the normal distribution of each element.

[0069] The vector y(t) in the mathematical formula (12) represents an observation value vector, which is a vector composed of elements that can be measured at time t. In addition, the vector function h(x(t)) is a vector function that represents a mathematical model of the observation vector y(t) obtained from the state x(t). In this example, since the ammonia nitrogen concentration and the dissolved oxygen concentration in series 1 and the dissolved oxygen concentration in series 2 can be measured from the state x(t), the observation vector y(t) and the vector function h(x(t)) can be expressed as follows.

[0070] [Mathematical formula 22]

[0071] [Mathematical formula 23]

[0072] [Mathematical formula 24]

[0073] [Mathematical formula 25]

[0074] In addition, the vector w included in equation (12) y (t) is the observation noise added to the observation at time t. In this example, we assume that w x Each element of (t) is normally distributed with a mean value of 0, and the system user can set any value for the variance of the normal distribution of each element.

[0075] Based on the state equation shown in equation (11) and the observation equation shown in equation (12), the particle filter estimates the state of the system at each moment. First, before the particle filter performs the estimation process, it is necessary to define the state of the particles and initialize them. Set the number of assumed particles to N. p , the particle i(1≤i≤N p ) The state at time t is set as vector x (i) (t). In the particle filter, each particle is regarded as a sample of the system’s state vector x(t), so the state vector x of each particle is (i) (t) has the same elements as the state vector x(t) of the system. In addition, since state transition and observation can be performed according to the same state equation and observation equation as the system shown in mathematical formulas (11) and (12), the following mathematical formula holds. In the following formula, the state vector x of particle i is (i) (t), observation vector y (i) (t), and system noise w x (i) (t), observation noise w y (i) (t) has the same structure as the actual system described above, so the description is omitted.

[0076] [Mathematical formula 26]

[0077] [Mathematical formula 27]

[0078] The initial state of each particle is determined by sampling from a normal distribution with each measured value or initial estimated value as the average value. (r) NH (0), x (r) 0(0), x (s,2) 0(0) and the initial estimated value θ^ (r) 0,0 (0),θ^ (r) 0,NH (0),θ^ (r) 0,u,b (0),θ^ (r) 0,B (0),θ^ (r) NH,NH (0),θ^ (r) NH,0 (0),θ^ (r) NH,u,b (0),θ^ (r) NH,B (0),θ^ (r) 0,0 (0),θ^ (r) 0,NH (0),θ^ (r) 0,u,b (0),θ^ (r) 0,B (0) As the average value of the normal distribution, the sampling values ​​obtained from these normal distributions are defined as the initial state of each particle. The variance of these normal distributions can be set by the system user at will, and for x for which the measured value and initial estimated value cannot be obtained, (s,2) NH The mean and variance of (0) can also be set by the system user.

[0079] The first step of the estimation process in the particle filter is to transfer the state of each particle according to the state equation. For the transfer, the state x of each particle is generated by sampling (i) (t) and the input u(t) applied to the system and the system noise w for each particle (i) x (t), and calculate the state x of the transferred particle based on mathematical formula (26): (i) (t+1).

[0080] The second step of the inference process is to predict the state x of the particle after the transfer. (i) The observation vector y of each particle at (t+1) (i) (t+1). Regarding the prediction of the observation vector, the state x of the particle after the transfer is generated by sampling (i) (t+1) and the observation noise w for each particle(i) y (t), and calculate the predicted value y of the observation vector based on mathematical formula (24) (i) (t+1).

[0081] The third step of the estimation process is performed after the actual observation value y(t+1) is obtained. (i) The likelihood function of particle i at time t+1 is calculated by using the actual observed value y(t+1) and the likelihood of each particle. (i) (t+1) is calculated as follows.

[0082] [Mathematical formula 28]

[0083] In equation (28), σ is a parameter representing the variance of the likelihood function, and its value is set arbitrarily by the system user. (i) (t+1)-y(t+1)|is the returned vector y (i) The function of the Euclidean norm of (t+1)-y(t+1) is calculated as follows in this example.

[0084] [Mathematical formula 29]

[0085] The fourth step of the estimation process is to normalize the calculated likelihood, weight the states of each particle according to the normalized likelihood, and find the average value. The vector obtained by this calculation is the estimated value x^(t+1) of the system x(t+1) at time t+1. The calculation is expressed as follows.

[0086] [Math 30]

[0087] The fifth step of the estimation process is to resample the particles. The particle filter is an algorithm that estimates by approximating the probability distribution of the state vector to be estimated with a large number of samples (particles). The more particles there are, the more accurately the desired probability distribution can be approximated. However, if the first to fourth steps are repeated, degradation will occur in which the likelihood of only a few particles becomes high. If degradation occurs, the target probability distribution is approximated using only a small number of particles, and the approximation accuracy is extremely deteriorated. Resampling is to prevent this degradation, and particles are regenerated so that the likelihood of each particle is close to 1 / N. p In this process, based on the likelihood of each particle, the new N pParticles to regenerate particles.

[0088] As mentioned above, by sequentially executing the first to fifth steps, the estimated values ​​of the system at each moment, i.e., the water quality items as the estimation object in each subordinate series, the parameters included in the water quality change model of the representative series, and the correction parameters included in the water quality change model of each subordinate series can be estimated. In the present embodiment 1, the estimation method of the system is described by taking the situation that the number of series is 2, an ammonia nitrogen concentration meter and a dissolved oxygen concentration meter are set for the representative series, and only a dissolved oxygen concentration meter is set for the subordinate series, and the control input is only the aeration volume instruction value of each series as an example. The estimation method is an implementable method that is independent of the number of series or the type of water quality items measured, and the content of the control input. In addition, in the present embodiment 1, the processing content is described by taking a simple linear model as an example, but even a complex nonlinear model such as ASM1 can be implemented without compromising generality. In addition, although a particle filter is used as an estimation algorithm, other state estimation algorithms such as an extended Kalman filter can also be applied, and the present disclosure is not limited by these factors.

[0089] As a water quality control algorithm using an estimated value, it can be performed by classical control such as PID (proportional-integral-differential) control using one estimated value and two-degree-of-freedom PID control using two estimated values, or it can be performed by a modern control method such as optimal control based on multiple estimated values ​​and a derived water quality change model. In addition, when performing the above control, in addition to the estimated value, the actual measurement value can also be used.

[0090] Implementation method 2. Figure 2 1 is a diagram showing a simplified structure of a water treatment system according to Embodiment 2. In Embodiment 1, a method is described in which, among the series constituting the water treatment system, a series capable of measuring two or more water quality items is defined as a representative series, and a series other than the representative series, a series capable of obtaining at least one or more water quality items among the water quality items measured in the representative series, is defined as a subordinate series, and by expressing the difference between each subordinate series and the representative series with a small number of small correction parameters, a water quality item that is a measurement object of the representative series and is not a measurement object of each subordinate series is estimated.

[0091] On the other hand, in embodiment 2, if Figure 2As shown, the same estimation method is applied to the reaction tanks included in a certain series, not to the differences between the series. That is, among the reaction tanks included in the same series, the reaction tank that can measure two or more water quality items is defined as a representative reaction tank, and the reaction tank other than the representative reaction tank and that can obtain at least one of the water quality items measured in the representative reaction tank is defined as a subordinate reaction tank, and the difference between each subordinate reaction tank and the representative reaction tank is represented by a small amount of small correction parameters, thereby being able to estimate the water quality items that are the measurement objects of the representative reaction tank and are not the measurement objects of each subordinate reaction tank.

[0092] like Figure 2 As shown, the water treatment system of this embodiment 2 includes more than one series, and one of the series is composed of more than two reaction tanks. In this embodiment 2, in order to simplify the description, a water treatment system with one series and five reaction tanks in the series is used as an example for description.

[0093] Series 1 includes an anaerobic tank 402, an anaerobic tank 403, an aerobic tank 404, an aerobic tank 405, and a sedimentation tank 406. Sewage as treated water flows into series 1 from the outside through a pump 401, and then passes through the anaerobic tank 402, the anaerobic tank 403, the aerobic tank 404, the aerobic tank 405, and the sedimentation tank 406, so that the inflowing sewage is treated and discharged. In addition, in series 1, a water quality control unit 409 for controlling the water quality of the treated water in series 1 is provided. In this embodiment 2, the water quality control unit 409 is connected to a blower 407 and a blower 408, and the blower 407 aerates the aerobic tank 404, and the blower 408 aerates the aerobic tank 405. The water quality control unit 409 aerates the treated water in the aerobic tank 404 and the aerobic tank 405 by sending an aeration volume instruction value to the blower 407 and the blower 408, thereby removing water quality pollutants such as ammonia nitrogen. In addition, a sensor for measuring the water quality items of the treated water contained in the aerobic tank 404 is provided in the aerobic tank 404. In the present embodiment 2, two sensors are provided in the aerobic tank 404, namely, a dissolved oxygen concentration meter (sensor 410) for measuring the oxygen concentration dissolved in the treated water and an ammonia nitrogen concentration (sensor 411) for measuring the ammonia nitrogen concentration contained in the treated water. In addition, a sensor for measuring the water quality items of the treated water contained in the aerobic tank 405 is provided in the aerobic tank 405. Unlike the aerobic tank 404, in the present embodiment 2, a sensor is provided in the aerobic tank 405, and a dissolved oxygen concentration meter (sensor 413) for measuring the oxygen concentration dissolved in the treated water is provided.

[0094] In the second embodiment, the reaction tank capable of measuring two or more water quality items contained in the treated water in each reaction tank in the system is treated as a reaction tank representing all the reaction tanks, and is referred to as a representative reaction tank. In addition, the reaction tank capable of measuring at least one or more water quality items in the water quality items measurable in the representative reaction tank and in the treated water in each reaction tank is treated as a reaction tank subordinate to the representative reaction tank, and is referred to as a subordinate reaction tank. At least one or more of the water quality items measurable in the representative reaction tank is a water quality item different from the water quality items available in the subordinate reaction tank. In the second embodiment, since a dissolved oxygen concentration meter (sensor 410) and an ammonia nitrogen concentration meter (sensor 411) are provided in the aerobic tank 404, the aerobic tank 404 is treated as a representative reaction tank, and only a dissolved oxygen concentration meter (sensor 413) is provided in the aerobic tank 405, so the aerobic tank 405 is treated as a subordinate reaction tank.

[0095] A representative reaction tank water quality data acquisition unit 412 is provided for the representative reaction tank to obtain water quality items that can be measured in the representative reaction tank, and a subordinate reaction tank water quality data acquisition unit 414 is provided for each subordinate reaction tank to obtain water quality items that can be measured. The water quality data collected by the representative reaction tank water quality data acquisition unit 412 and the subordinate reaction tank water quality data acquisition unit 414 are sent to the data storage unit 501 and stored. In addition, the data storage unit 301 not only stores the water quality data of the representative reaction tank and each subordinate reaction tank, but also stores control data indicating the control content applied to the treated water in each reaction tank. The control data is stored by sending the control content for each reaction tank to the data storage unit by the water quality control unit 409 to send the control data to the data storage unit. In this example, since the water quality control unit controls the blower 407 and the blower 408, the aeration amount instruction value sent to each blower is also sent to the data storage unit and stored. The data storage unit also stores parameters included in a representative reaction tank water quality change model described later, parameter correction amounts included in each slave reaction tank water quality change model, and data on water quality estimated values ​​for each slave reaction tank.

[0096] The representative reaction tank water quality change model deriving unit 502 derives a representative reaction tank water quality change model based on various data stored in the data storage unit. The representative reaction tank water quality change model is a mathematical model that represents the change of water quality items that can be obtained in the representative reaction tank. In this example, since the water quality items that can be obtained in the representative reaction tank are the dissolved oxygen concentration obtained by the sensor 410 and the ammonia nitrogen concentration obtained by the sensor 411, the representative reaction tank water quality change model deriving unit derives a mathematical model that represents the change of these two water quality items in the representative reaction tank.

[0097] The slave reaction tank water quality model derivation unit 502 derives the slave reaction tank water quality change model based on the representative reaction tank water quality change model and various data stored in the data storage unit. The slave reaction tank water quality change model is a mathematical model representing the change of water quality items available in each slave reaction tank and in the representative reaction tank. Since there is only one representative reaction tank and there can be multiple slave reaction tanks, the derived slave reaction tank water quality reaction model is not necessarily one. In this example, the water quality items available in the representative reaction tank are the two types of dissolved oxygen concentration and ammonia nitrogen concentration as described above, and the slave reaction tank is only the aerobic tank 405, so the slave reaction tank water quality model derivation unit derives a mathematical model representing the change of these two water quality items in the aerobic tank 405.

[0098] The estimation unit 504 estimates the parameters included in the representative reaction tank water quality change model, the water quality change models of each slave reaction tank, and the parameter correction amount included in the water quality change model of each slave reaction tank based on the representative reaction tank water quality change model, the water quality change models of each slave reaction tank, and various data stored in the data storage unit, and stores the estimation results in the data storage unit.

[0099] The output unit 505 outputs various estimation results stored in the data storage unit and estimated by the estimation unit. The output unit may be in any form, such as output as a video on a display, output to paper on a printer, or output to other devices via a network.

[0100] The operation is described below. The water quality treatment system includes a processing unit 500. The processing unit 500 includes a representative series water quality change model derivation unit 502, a subordinate series water quality change model derivation unit 503 and an estimation unit 504. In the representative reaction tank water quality change model derivation unit 502, the representative reaction tank water quality change model is derived. As described above, in this example, a mathematical model of the dissolved oxygen concentration and the ammonia nitrogen concentration in the aerobic tank 404 representing the reaction tank is derived. Regarding this mathematical model, as in embodiment 1, it can be any mathematical model such as ASM1 represented by mathematical formula (1) or a linear model represented by mathematical formula (2). In this example, if the ammonia nitrogen concentration at time t in the representative reaction tank is set to x (r) NH (t), dissolved oxygen concentration is set to x0 (r) NH (t), the aeration volume is set to u (r) NH (t), then the water quality change model of the reaction tank is the same as that of equations (3) and (4).

[0101] In the representative reaction tank water quality change model derivation unit 502, the initial estimated values ​​of the parameters in the representative reaction tank water quality change model are derived using the data stored in the data storage unit. In this example, the parameter θ in the representative reaction tank water quality change model that is the same as the mathematical formula (3) and the mathematical formula (4) is derived. (r) NH,NH ,θ (r) NH,0 ,θ (r) NH,u,b ,θ (r) NH,B ,θ (r) 0,0 ,θ (r) 0,NH ,θ (r) 0,u,b and θ (r) 0,B The initial estimated value θ^ (r) NH,NH ,θ^ (r) NH,0 ,θ^ (r) NH,u,b ,θ^ (r) NH,B ,θ^ (r) 0,0 ,θ^ (r) 0,NH ,θ^ (r) 0,u,b and θ^ (r) 0,B . In addition, regarding the derivation method, if the linear model of mathematical formula (3) and mathematical formula (4) is used, it can be derived by the least square method as shown in Implementation Example 1, and if a nonlinear model such as ASM1 is used, the initial estimated value of the parameter can be derived by a nonlinear least square method headed by the Gauss-Newton method. Therefore, it is possible to derive the initial estimated value of the parameter included in the representative reaction tank water quality change model without relying on the content of the representative reaction tank water quality change model and using the data stored in the data storage unit 501 without compromising generality.

[0102] Next, the operation of the slave reaction tank water quality model derivation unit 503 is described. The slave reaction tank water quality change model derivation unit 503 derives the water quality change model of each slave reaction tank. As described above, in the present embodiment 2, the mathematical model of the dissolved oxygen concentration and the ammonia nitrogen concentration in the reaction tank 305 as the slave reaction tank is derived. When deriving the water quality change model of each slave reaction tank, if the difference between the reaction tanks is taken into account, although there may be individual differences between each device and sensor according to the same discussion as in the embodiment 1, it can be considered that when a certain reaction tank is used as a reference, the difference between the reaction tank used as the reference and the other reaction tanks is small. In the present embodiment, if the reaction tank used as the reference is considered to be a representative reaction tank, and the water quality change in each slave reaction tank shows a change similar to the water quality change in the representative reaction tank, it is considered that the water quality change model of each slave reaction tank can be derived by adding a sufficiently small parameter correction amount to the parameters in the water quality change model of the representative reaction tank.

[0103] Taking the case where the representative reaction tank water quality change model is the same as the mathematical formula (3) and the mathematical formula (4) as an example, the process of the slave reaction tank water quality change model derivation unit deriving the water quality change model of the reaction tank 405 as the slave reaction tank is described. As described above, the parameter θ included in the representative reaction tank water quality change model is used. (r) NH,NH ,θ (r) NH,O ,θ (r) NH,u,b ,θ (r) NH,B ,θ (r) O,O ,θ (r) O,NH ,θ (r) O,u,b and θ (r) O,B Here, the concentration of ammonia nitrogen in the reaction tank 405 at time t is set to x (s,405) NH (t), the dissolved oxygen concentration is set to x (s,405) O (t), the aeration volume is set to u (s,405) O(t), and the parameter correction amount for each parameter in the representative series water quality change model is Δθ (s,405) NH,NH , Δθ (s,405) NH,O , Δθ (s,405) NH,u,b , Δθ (s,405) NH,B , Δθ (s,405) O,O , Δθ(s,405) O,NH , Δθ (s,405) O,u,b and Δθ (s,405) O,B In the case of , the subordinate series water quality change model of the reaction tank 405 can be expressed in the same form as the mathematical formula (9) and the mathematical formula (10).

[0104] In the slave reaction tank water quality change model derivation unit 503, an initial estimated value of the parameter correction amount included in the slave reaction tank water quality change model is derived. (s,405) NH,NH , Δθ (s,405) NH,O , Δθ (s,405) NH,u,b , Δθ (s,405) NH,B , Δθ (s,405) O,O , Δθ (s,405) O,NH , Δθ (s,405) O,u,b and Δθ (s,405) O,B The initial estimated value of Δθ^ (s,405) NH,NH , Δθ^ (s,405) NH,O , Δθ^ (s,405) NH,u,b , Δθ^ (s,405) NH,B , Δθ^ (s,405) O,O , Δθ^ (s,405) O,NH , Δθ^ (s,405) O,u,b and Δθ^ (s,405) O,B However, unlike the representative reaction tank, there is no data for uniquely determining the initial estimated values ​​of the parameter correction amounts. Therefore, these initial estimated values ​​are all set to 0, and the estimating unit 504 estimates the water quality items that can be measured in the representative reaction tank of each slave reaction tank, and at the same time, updates the estimated values ​​of these parameter correction amounts and performs processing to make them close to the true value.

[0105] Next, the operation of the estimation unit 504 will be described. As described above, the estimation unit 504 estimates the water quality items that can be measured in the representative reaction tank of each slave reaction tank, and at the same time estimates the parameters in the water quality change model of the representative reaction tank and the parameter correction amount included in each slave series water quality change model. In the second embodiment, the ammonia nitrogen in the reaction tank 405 as the slave reaction tank, the parameters in various models and the parameter correction amount are estimated.

[0106] In the second embodiment, the estimation process is performed by a particle filter. As in the first embodiment, in the estimation using the particle filter, it is necessary to define the state equation and observation equation of the system to be estimated. Regarding the state equation, in this system, various parameters and parameter correction amounts required to represent the water quality items as the measurement objects in the representative reaction tank and each slave reaction tank and their water quality changes constitute the state equation of the system. In addition, regarding the observation equation, in this system, the water quality items that can be measured in the representative reaction tank and each slave reaction tank constitute the observation equation of the system.

[0107] In the second embodiment, the variables and parameters constituting the state equation are the concentration of ammonia nitrogen and the concentration of dissolved oxygen in the reaction tank 404 as the representative reaction tank and the reaction tank 405 as the slave reaction tank, and the parameter θ required to express them. (r) NH,NH ,θ (r) NH,O ,θ (r) NH,u,b ,θ (r) NH,B ,θ (r) O,O ,θ (r) O,NH ,θ (r) O,u,b ,θ (r) O,B , parameter correction value Δθ (s,405) NH,NH , Δθ (s,405) NH,O , Δθ (s,405) NH,u,b , Δθ (s,405) NH,B , Δθ (s,405) O,O , Δθ (s,405) O,NH , Δθ (s,405) O,u,b and Δθ (s,405) O,B , and the variables constituting the observation equation are the dissolved oxygen concentration and the ammonia nitrogen concentration in the reaction tank 404 which can be measured by the sensors 410 and 411, and the dissolved oxygen concentration in the reaction tank 405 which can be measured by the sensor 413. If the above-mentioned representative reaction tank water quality change model and the subordinate reaction tank water quality change model are used, these state equations and observation equations can be expressed as follows.

[0108] [Mathematical formula 31] x(t+1)=f(x(t),u(t)))+w x (t)…(31)

[0109] [Math 32] y(t)=h(x(t))+w y (t)…(32)

[0110] Mathematical formula (31) represents the state equation, and mathematical formula (32) represents the observation equation of this system. In mathematical formula (31), x(t) is a vector composed of various parameters and parameter correction amounts required to represent the water quality items to be measured in the representative reaction tank and each subordinate reaction tank at time t and their water quality changes, and is composed of the following vectors.

[0111] [Mathematical formula 33]

[0112] [Math 34]

[0113] [Mathematical formula 35]

[0114] [Mathematical formula 36]

[0115] [Mathematical formula 37]

[0116] In addition, the vector u(t) in the mathematical formula (31) is a vector composed of control command values ​​for the system. In the second embodiment, the control command value is an aeration amount command value for the blower 407 or the blower 408 output by the water quality control unit 409, so the vector u(t) can be expressed as follows.

[0117] [Math 38]

[0118] In addition, f(x(t), u(t)) in mathematical formula (31) is a vector function corresponding to each element constituting the left side x(t+1), and in this embodiment 2, it is constructed as follows.

[0119] [Math 39]

[0120] The vector function f(x(t), u(t)) constituting the mathematical formula (39) corresponds to the water quality change model representing the reaction tank. The vector function f (x(s,405)) (x(t), u(t)) corresponds to the dependent series water quality change model of the reaction tank 405. The content of the function is the same as that of the first embodiment and is therefore omitted. The vector function f (θ(r))(x(t), u(t)) corresponds to the parameters included in the model representing the series of water quality changes, and the vector function f (Δθ(s,405)) (x(t), u(t)) corresponds to the parameter correction amount included in the dependent series water quality change model of the reaction tank 405. Since these values ​​are constants, each vector function can be expressed as follows.

[0121] [Math 40] f (θr)) (x(t),u(t))=θ (r) (t)…(40)

[0122] [Mathematical formula 41]

[0123] In addition, the vector w included in mathematical formula (31) x (t) is the system noise added to the system state at time t. In this example, assume that w x Each element of (t) is normally distributed with a mean value of 0, and the system user can set any value for the variance of the normal distribution of each element.

[0124] The vector y(t) in the mathematical formula (32) represents an observation value vector, which is a vector composed of elements that can be measured at time t. In addition, the vector function h(x(t)) is a vector function that represents a mathematical model of the observation vector y(t) obtained according to the state x(t). In this example, since the ammonia nitrogen concentration and the dissolved oxygen concentration in the reaction tank 404 and the dissolved oxygen concentration in the reaction tank 405 can be measured according to the state x(t), the observation vector y(t) and the vector function h(x(t)) can be expressed as follows.

[0125] [Mathematical formula 42]

[0126] [Mathematical formula 43]

[0127] [Mathematical formula 44]

[0128] [Mathematical formula 45]

[0129] In addition, the vector w contained in equation (32) y (t) is the observation noise added to the observation at time t. In this example, we assume that w y Each element of (t) is normally distributed with a mean value of 0, and the system user can set any value for the variance of the normal distribution of each element.

[0130] As described above, in the second embodiment, the state equation is defined as equation (31), and the observation equation is defined as equation (32). Since equations (31) and (32) are the same as equations (11) and (12) of the first embodiment, the ammonia nitrogen concentration of the reaction tank 405 included in the state vector x(t), the parameters included in the representative reaction tank water quality change model, and the parameter correction amount included in the water quality change model of each slave reaction tank can be estimated in sequence by the same method as in the first embodiment. In this example, the number of reaction tanks as the object of the system is 2, and the ammonia nitrogen concentration and the dissolved oxygen concentration are set for the representative reaction tank, and only the dissolved oxygen concentration is set for the slave reaction tank, and the control input is only the aeration amount instruction value of the representative reaction tank and the slave reaction tank. This situation is used to illustrate the estimation method of the system. This method is a method that can be implemented regardless of the number of reaction tanks, the type of water quality item to be measured, or the content of the control input. In addition, in this embodiment 2, a simple linear model is used as an example to illustrate the processing content, but even a complex nonlinear model such as ASM1 can be implemented without compromising generality. In addition, although a particle filter is used as an estimation algorithm, other state estimation algorithms such as an extended Kalman filter can also be applied, and the present disclosure is not limited by these factors.

[0131] As a water quality control algorithm using an estimated value, it can be performed by classical control such as PID (proportional-integral-differential) control using one estimated value and two-degree-of-freedom PID control using two estimated values, or it can be performed by a modern control method such as optimal control based on multiple estimated values ​​and a derived water quality change model. In addition, when performing the above control, in addition to the estimated value, the actual measurement value can also be used.

[0132] Figure 3 1 is a diagram showing an example of the hardware configuration of the data storage unit and the processing unit in Embodiment 1 and Embodiment 2. The data storage unit 301, 501 and the processing unit 300, 500 include a processor 11 that performs various processes and a memory 12 that stores information. The processor 11 and the memory 12 can send and receive information to and from each other via a bus 13.

[0133] The data storage unit 301 is implemented by the memory 12. The processor 11 reads and executes the program stored in the memory 12, thereby functioning as a representative system water quality change model derivation unit 302, 502, a dependent system water quality change model derivation unit 303, 503, and an estimation unit 304, 504.

[0134] The processor 11 is an example of a processing circuit, and includes one or more of a CPU (Central Processing Unit), a DSP (Digital Signal Processor), and a system LSI (Large Scale Integration).

[0135] The memory 12 includes one or more of RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory) and EEPROM (registered trademark) (Electrically Erasable Programmable Read Only Memory). In addition, the memory 12 includes a recording medium on which a program readable by a computer is recorded. The above-mentioned recording medium includes one or more of a non-volatile or volatile semiconductor memory, a magnetic disk, a floppy disk, an optical disk, a compressed disk and a DVD (Digital Versatile Disc). In addition, the processing unit 12, 43 may include an integrated circuit such as an ASIC (Application Specific Integrated Circuit) and an FPGA (Field Programmable Gate Array).

[0136] In addition, the water quality control unit 107, 207, 409 is not limited to a device that controls the blower 106, 206, 406, as long as it is a device that controls the water quality by controlling at least one of the pumps 101, 201, 401, the valve (not shown), and the chemical addition device (not shown). The valve is, for example, a valve that is set in the pipe through which the treated water of the inflow series passes and adjusts the inflow amount of the treated water. The chemical addition device is, for example, a device that puts a chemical for purifying the treated water into the treated water.

[0137] In addition, it is preferred that the water quality items acquired by each water quality data acquisition unit 110, 209, 412, 414 include at least one of dissolved oxygen concentration, ammonia nitrogen concentration, nitrite nitrogen concentration, nitrate nitrogen concentration, total nitrogen concentration, total phosphorus concentration, pH (hydrogen ion concentration), alkalinity, water temperature, MLSS (Mixed Liquor Suspended Solids, activated sludge suspended solids), ORP (oxidation-reduction potential), turbidity, chromaticity, conductivity, BOD (biochemical oxygen demand), and COD (chemical oxygen demand).

[0138] The configuration described in the above embodiment is an example, and can be combined with other known technologies, the embodiments can be combined with each other, and part of the configuration can be omitted or changed within the scope of the gist. Description of symbols

[0139] 101, 201, 401 pump, 102, 202, 402 anaerobic tank, 103, 203, 403 anaerobic tank, 104, 204, 404, 405 aerobic tank, 105, 205 sedimentation tank, 106, 206, 406, 407, 408 blower, 107, 207, 409 water quality control unit, 108, 109, 208, 410, 411, 413 sensor, 110 represents Series water quality data acquisition unit, 209 is subordinate to the series water quality data acquisition unit, 300, 500 processing unit, 301, 501 data storage unit, 302, 502 represents the series water quality change model derivation unit, 303, 503 is subordinate to the series water quality change model derivation unit, 304, 504 estimation unit, 305, 505 output unit, 412 represents the reaction tank water quality data acquisition unit, 414 is subordinate to the reaction tank water quality data acquisition unit.

Claims

1. A water treatment system, characterized in that: include: A representative series, which performs purification treatment on the treated water; A subordinate series, which performs purification treatment on the treated water; A water quality control unit, which is provided in the representative series and the subordinate series and controls the water quality of the treated water; A subordinate series water quality data acquisition unit, which is arranged in the subordinate series and acquires at least one water quality item from the treated water; a representative series water quality data acquisition unit, which is arranged in the representative series and acquires from the treated water at least one water quality item of the same type as the water quality items acquired by the subordinate series water quality data acquisition unit and at least one water quality item different from the water quality items acquired by the subordinate series water quality data acquisition unit; a data storage unit that stores control data indicating the content of control performed by the water quality control unit, the representative series water quality data acquired by the representative series water quality data acquisition unit, and the subordinate series water quality data acquired by the subordinate series water quality data acquisition unit; a representative series water quality change model deriving unit, which derives a representative series water quality change model based on the data stored in the data storage unit, the representative series water quality change model being a mathematical model representing changes in water quality items included in the representative series water quality data; a subordinate series water quality change model deriving unit, which derives a subordinate series water quality change model based on the representative series water quality change model and the data stored in the data storage unit, the subordinate series water quality change model being a mathematical model representing the change of the water quality item acquired by the representative series water quality data acquisition unit in the subordinate series; as well as An estimating unit estimates a water quality item included in the representative system water quality data that is different from the water quality item acquired by the subordinate system water quality data acquiring unit based on the representative system water quality change model, the subordinate system water quality change model, and the data stored in the data storage unit.

2. The water treatment system according to claim 1, characterized in that: The method further includes an output unit that outputs an estimation result estimated by the estimation unit.

3. The water treatment system according to claim 1 or 2, characterized in that: In the subordinate series water quality change model derivation unit, The subordinate series water quality variation model is derived by deriving a parameter correction amount for each parameter included in the representative series water quality variation model and adding the derived parameter correction amount to the corresponding parameter included in the representative series water quality variation model.

4. A water treatment system, characterized in that: include: A representative reaction tank, which performs purification treatment on the treated water; A slave reaction tank, which performs purification treatment on the treated water; A water quality control unit, which is disposed in the representative reaction tank and the slave reaction tank and controls the water quality of the treated water; A slave reaction tank water quality data acquisition unit, which is disposed in the slave reaction tank and acquires at least one water quality item from the treated water; a representative reaction tank water quality data acquisition unit, the representative reaction tank water quality data acquisition unit being arranged in the representative reaction tank, and acquiring from the treated water at least one water quality item of the same type as the water quality items acquired by the slave reaction tank water quality data acquisition unit and at least one water quality item different from the water quality items acquired by the slave reaction tank water quality data acquisition unit; a data storage unit storing control data indicating the content of control performed by the water quality control unit, representative reaction tank water quality data acquired by the representative reaction tank water quality data acquisition unit, and subordinate reaction tank water quality data acquired by the subordinate reaction tank water quality data acquisition unit; a representative reaction tank water quality change model deriving unit, the representative reaction tank water quality change model deriving unit deriving a representative reaction tank water quality change model based on the data stored in the data storage unit, the representative reaction tank water quality change model being a mathematical model representing changes in water quality items included in the representative reaction tank water quality data; a slave reaction tank water quality change model deriving unit, which derives the slave reaction tank water quality change model based on the representative reaction tank water quality change model and the data stored in the data storage unit, wherein the slave reaction tank water quality change model is a mathematical model representing the change of the water quality item acquired by the representative reaction tank water quality data acquisition unit in the slave reaction tank; as well as An estimating unit estimates water quality items included in the representative reaction tank water quality data that are different from the water quality items acquired by the slave reaction tank water quality data acquiring unit based on the representative reaction tank water quality change model, the slave reaction tank water quality change model and the data stored in the data storage unit.

5. The water treatment system according to claim 4, characterized in that: The method further includes an output unit that outputs an estimation result estimated by the estimation unit.

6. The water treatment system according to claim 4 or 5, characterized in that: In the slave reaction tank water quality change model derivation unit, The slave reaction tank water quality change model is derived by deriving a parameter correction amount for each parameter included in the representative reaction tank water quality change model and adding the derived parameter correction amount to the corresponding parameter included in the representative reaction tank water quality change model.

7. The water treatment system according to any one of claims 1 to 6, characterized in that: The water quality control unit controls at least one of a blower, a pump, a valve, and a chemical adding device.

8. The water treatment system according to any one of claims 1 to 6, characterized in that: The water quality items include at least one of dissolved oxygen concentration, ammonia nitrogen concentration, nitrite nitrogen concentration, nitrate nitrogen concentration, total nitrogen concentration, total phosphorus concentration, hydrogen ion concentration, alkalinity, water temperature, activated sludge suspended matter, redox potential, turbidity, chromaticity, conductivity, biochemical oxygen demand and chemical oxygen demand.

Citation Information

Patent Citations

  • Aeration air quantity controller for sewage treatment plant

    JP2005199115A