FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
[See section 10, Rule 13]
WATER TREATMENT SYSTEM, AERATION AMOUNT CONTROL DEVICE, AND
AERATION AMOUNT CONTROL METHOD;
MITSUBISHI ELECTRIC CORPORATION, A CORPORATION ORGANISED AND
EXISTING UNDER THE LAWS OF JAPAN, WHOSE ADDRESS IS 7-3,
MARUNOUCHI 2-CHOME, CHIYODA-KU, TOKYO 1008310, JAPAN
THE FOLLOWING SPECIFICATION PARTICULARLY DESCRIBES THE
INVENTION AND THE MANNER IN WHICH IT IS TO BE PERFORMED.
2
DESCRIPTION
TITLE OF THE INVENTION: WATER TREATMENT SYSTEM, AERATION
AMOUNT CONTROL DEVICE, AND AERATION AMOUNT CONTROL METHOD
5 TECHNICAL FIELD
[0001] The present disclosure relates to a water treatment
system, an aeration amount control device, and an aeration
amount control method.
10 BACKGROUND ART
[0002] One of treatment methods for sewage containing
organic substances and nitrogen is an activated sludge
process. In the activated sludge process, microorganisms
(activated sludge) having a purification ability are stored
15 in a reaction tank and the microorganisms and waste water are
mixed and contact with each other while being aerated,
whereby contaminants in the waste water are oxidized and
decomposed. In order to sufficiently purify the
contaminants, it is necessary to supply (aerate) an
20 appropriate amount of air to the bioreactor tank. In
addition, the residence time in the reaction tank is as long
as about 10 hours, and therefore it is necessary to control
the aeration amount in accordance with variation in the
inflow water quality (contaminant concentration in inflow
25 water).
3
[0003] In this regard, it is known that aeration amount
control is performed using an inflow water quality acquired
through measurement by a sensor capable of continuously
measuring the inflow water quality or through estimation from
5 the past time-series data (see, for example, Patent Documents
1 and 2).
CITATION LIST
PATENT DOCUMENT
10 [0004] Patent Document 1: Japanese Patent No. 6764487
Patent Document 2: Japanese Laid-Open Patent
Publication No. 2005-125229
SUMMARY OF THE INVENTION
15 PROBLEM TO BE SOLVED BY THE INVENTION
[0005] In Patent Document 1, in order to continuously
measure the water quality of inflow water, a water quality
sensor such as an ammonia meter, a total nitrogen analyzer,
or a BOD meter is provided, whereby high-accuracy measurement
20 is achieved. However, these measurement instruments are
expensive and need to be calibrated and maintained highly
frequently, so that a lot of cost and effort are required for
sensor maintenance.
[0006] In Patent Document 2, in determining a target value
25 in water quality control, the inflow water quality is
4
predicted on the basis of the past time-series data.
However, in a case where the measured inflow water quality at
present has changed from the water quality at the time when
the past data was acquired, prediction accuracy might be
5 lowered.
[0007] The present disclosure has been made to solve the
above problem, and an object of the present disclosure is to
provide a water treatment system, an aeration amount control
device, and an aeration amount control method that can
10 estimate an inflow water quality with high accuracy without
using multiple measurement instruments for continuously
measuring the inflow water quality, and supply a necessary
amount of aeration without delay in accordance with variation
in the inflow water quality, thus suppressing variation in a
15 treated water quality.
MEANS TO SOLVE THE PROBLEM
[0008] A water treatment system according to the present
disclosure is a water treatment system for performing water
20 treatment through biological oxidation while performing
aeration from a blower to a reaction tank, the water
treatment system including: a first water quality measurement
value acquisition unit which acquires a water quality
measurement value at a first time point, of treatment target
25 water flowing into the reaction tank; a water quality
5
variation pattern acquisition unit which, from time-series
change in water quality information of the treatment target
water acquired in advance, acquires a water quality variation
pattern according to a condition at a time of acquisition of
5 the water quality information acquired in advance; an inflow
water quality estimation unit which selects a water quality
variation pattern that matches a condition at a time of
acquisition of the water quality measurement value at the
first time point, from the water quality variation patterns
10 included in the water quality variation pattern acquisition
unit, and estimates an inflow water quality value subsequent
to the first time point from the selected water quality
variation pattern; and a control unit which controls an
aeration amount of the blower subsequent to the first time
15 point, on the basis of the inflow water quality value
estimated by the inflow water quality estimation unit.
EFFECT OF THE INVENTION
[0009] The water treatment system according to the present
20 disclosure can estimate an inflow water quality value with
high accuracy without using a measurement instrument capable
of continuously measuring the inflow water quality and supply
a necessary amount of aeration in accordance with variation
in the inflow water quality value, thus suppressing variation
25 in a treated water quality value and reducing excessive
6
aeration.
BRIEF DESCRIPTION OF THE DRAWINGS
[0010] [FIG. 1] FIG. 1 is a block diagram showing the
5 configuration of a water treatment system according to
embodiment 1.
[FIG. 2] FIG. 2 is a block diagram showing the
aeration amount control device according to embodiment 1.
[FIG. 3] FIG. 3 illustrates an inflow water
10 quality variation pattern according to embodiment 1.
[FIG. 4] FIG. 4 shows an example of a water
quality variation pattern selected by the aeration amount
control device according to embodiment 1.
[FIG. 5] FIG. 5 shows an operation flow of the
15 aeration amount control device according to embodiment 1.
[FIG. 6] FIG. 6 is a block diagram showing an
aeration amount control device according to embodiment 2.
[FIG. 7] FIG. 7 shows an example of a water
quality variation pattern selected by the aeration amount
20 control device according to embodiment 2.
[FIG. 8] FIG. 8 is a block diagram showing the
configuration of a water treatment system according to
embodiment 3.
[FIG. 9] FIG. 9 is a hardware configuration
25 diagram of an inflow water quality estimation unit and the
7
aeration amount control device according to each of
embodiments 1 to 3.
[FIG. 10] FIG. 10 shows the configuration of an
inflow water quality inference device according to embodiment
5 4.
[FIG. 11] FIG. 11 shows the configuration of a
learning device of the inflow water quality inference device
according to embodiment 4.
[FIG. 12] FIG. 12 is a flowchart for performing
10 learning using the learning device shown in FIG. 11.
[FIG. 13] FIG. 13 shows the configuration of an
inference device according to embodiment 4.
[FIG. 14] FIG. 14 is a flowchart for inferring an
inflow water quality value using the inference device shown
15 in FIG. 13.
DESCRIPTION OF EMBODIMENTS
[0011] Hereinafter, embodiments will be described with
reference to the drawings. In the drawings, the same
20 reference characters denote the same or corresponding parts.
Hereinafter, with reference to the drawings,
embodiments of the present disclosure will be described in
detail. In the description and the drawings, components
having substantially the same function are denoted by the
25 same reference characters and will not repeatedly be
8
described. In the drawings, illustration of configurations
of devices and shapes of parts merely represent schematic
configurations and shapes of the devices and the parts.
Relative sizes and relative positions of parts shown in the
5 drawings do not necessarily represent accurate size
relationships and accurate positional relationships between
the actual parts.
[0012] Embodiment 1
Hereinafter, a water treatment system according to
10 embodiment 1 will be described with reference to FIG. 1 to
FIG. 5.
FIG. 1 shows the configuration of the water
treatment system according to embodiment 1, and FIG. 2 shows
the configuration of an aeration amount control device. In
15 FIG. 1, a water treatment system 100 includes: a bioreactor
tank 10 provided with an inflow section 15 for introducing
treatment target water, an outflow section 16 for discharging
treated water, and diffuser plates 11, 12, 13 disposed
inside; a blower 20 which sends air to the diffuser plates
20 11, 12, 13; air volume adjustment valves 71, 72, 73 for
adjusting the volume of air from the blower 20; target
aeration amount calculation units 61, 62, 63 which calculate
aeration amounts for controlling the air volume adjustment
valves 71, 72, 73; an inflow water quality estimation unit 30
25 which estimates an inflow water quality value for the target
9
aeration amount calculation units 61, 62, 63 to calculate the
aeration amounts; an inflow water quality measurement value
acquisition unit 50 which acquires a measurement value of a
water quality of the treatment target water flowing into the
5 bioreactor tank 10 and transmits the measurement value to the
inflow water quality estimation unit 30; and an inflow water
quality variation pattern acquisition unit 40 which acquires
a variation pattern of an inflow water quality and transmits
the variation pattern to the inflow water quality estimation
10 unit 30.
[0013] The aeration amount control device is composed of
air volume adjustment valves 70 (71, 72, 73), target aeration
amount calculation units 60 (61, 62, 63), the inflow water
quality estimation unit 30, the inflow water quality
15 measurement value acquisition unit 50, and the inflow water
quality variation pattern acquisition unit 40. When the air
volume adjustment valves and the target aeration amount
calculation units are collectively mentioned, they are
respectively referred to as air volume adjustment valves 70
20 and target aeration amount calculation units 60, and when
they are mentioned individually, they are respectively
referred to as air volume adjustment valves 71, 72, 73 and
target aeration amount calculation units 61, 62, 63.
Hereinafter, the details of each component will be
25 described.
10
[0014] The bioreactor tank 10 is a reaction tank storing
activated sludge therein. Air supplied from the blower 20
passes through a pipe 20a and is supplied into the bioreactor
tank 10 from the diffuser plates 11, 12, 13.
5 [0015] The air volume adjustment valves 71, 72, 73 are
provided one by one to pipes branched from the pipe 20a to
the diffuser plates 11, 12, 13, respectively. Through
adjustment of the opening degrees of the air volume
adjustment valves 71, 72, 73, the amounts of aeration to be
10 supplied to the diffuser plates 11, 12, 13 can be
individually adjusted.
[0016] The target aeration amount calculation units 61,
62, 63 calculate target values for the amounts of aeration to
be supplied from the diffuser plates 11, 12, 13, in every
15 arbitrary cycle, and transmit the target values for the
aeration amounts to the air volume adjustment valves 71, 72,
73 via signal lines 61a, 62a, 63a. The cycle of calculation
of the target values for the aeration amounts is desirably
about 1 second to 5 minutes, but may be arbitrarily set in
20 accordance with the characteristic of the plant site. The
opening degrees of the air volume adjustment valves 71, 72,
73 are adjusted so that the amounts of aeration supplied from
the diffuser plates 11, 12, 13 become equal to the target
values for the aeration amounts calculated by the target
25 aeration amount calculation units 61, 62, 63. In FIG. 1, the
11
example in which three diffuser plates, three air volume
adjustment valves, and three target aeration amount
calculation units are provided is shown, but the numbers
thereof are not limited thereto. The numbers of the diffuser
5 plates, the air volume adjustment valves, and the target
aeration amount calculation units can be arbitrarily changed
in accordance with the scale of the bioreactor tank 10 and
the characteristic of the plant site.
[0017] The inflow section 15 is a pipe or a water channel
10 through which treatment target water flows into the
bioreactor tank 10. The outflow section 16 is a pipe or a
water channel through which treated water treated in the
bioreactor tank 10 flows to the outside of the bioreactor
tank 10. The treatment target water flowing from the inflow
15 section 15 comes into contact with activated sludge and air
supplied through aeration in the bioreactor tank 10, whereby
oxidation and decomposition of contaminants in the treatment
target water are promoted, and the resultant water flows out
as the treated water.
20 [0018] Next, the inflow water quality variation pattern
acquisition unit 40 will be described.
A pattern of contaminant concentration variation
over time in which contaminant concentrations in treatment
target water acquired in advance are arranged in time series
25 in accordance with the acquisition times, is inputted to and
12
stored in the inflow water quality variation pattern
acquisition unit 40. As the contaminants, one or more kinds
are selected from treatment targets in the bioreactor tank
10. Examples include biochemical oxygen demand (BOD),
5 chemical oxygen demand (COD), ammonia nitrogen, total
nitrogen, Kjeldahl nitrogen, total phosphorus, and phosphate
phosphorus. A plurality of inflow water quality variation
patterns can be inputted in accordance with the kinds of
contaminants or a feature of date and time.
10 [0019] FIG. 3 illustrates an inflow water quality
variation pattern acquired by the inflow water quality
variation pattern acquisition unit 40. In a general sewage
treatment plant, a certain pattern appears in variation in
the inflow water quality through one day. The variation
15 pattern differs among plant sites. For example, in a
variation pattern characteristic to a residential district,
peaks appear in the early morning and the evening, and in a
variation pattern characteristic to an office district, a
peak appears only during daytime. For example, through an
20 all-day examination, inflow water on one day is sampled at
intervals of one or two hours and the water quality thereof
is measured, whereby variation in the inflow water quality
value through the day is acquired a plurality of times in
advance, and using an inflow water quality value at a given
25 time, inflow water quality values at other times are
13
normalized.
[0020] In FIG. 3, the vertical axis indicates the
contaminant concentration in given treatment target water,
and an example of temporal change, i.e., a variation pattern
5 of the water quality, under normalization in which the inflow
water quality value at 9:00 in the morning on a day is
defined as 1. As described above, the variation pattern of
the inflow water quality depends on an activity pattern of
people in the treatment district, and thus, the variation
10 pattern of the inflow water quality might differ between a
weekday and a holiday, for example. Therefore, variation
patterns of the inflow water quality are acquired separately
for a weekday and a holiday in advance, and an inflow water
quality variation pattern on the weekday indicated by a solid
15 line and an inflow water quality variation pattern on the
holiday indicated by a broken line in the graph are inputted
to the inflow water quality variation pattern acquisition
unit 40.
[0021] The above method in which the inflow water quality
20 variation patterns are separated between a weekday and a
holiday is merely an example. In accordance with the
characteristic of the treatment district, inflow water
quality variation patterns may be separated using a plurality
of indices such as month, day, and the operation rates of
25 factories in the treatment district. In addition, such an
14
inflow water quality variation pattern does not necessarily
need to be a one-day-basis pattern, and a pattern period such
as one minute, one hour to one week, or one month may be
selected in accordance with the variation cycle of the inflow
5 water quality on each plant site. An inflow water quality
variation pattern is inputted to the inflow water quality
variation pattern acquisition unit 40 for every set pattern
period. That is, in a case of an inflow water quality
variation pattern on a one-day basis, an inflow water quality
10 variation pattern is inputted to the inflow water quality
variation pattern acquisition unit 40 on a one-day basis.
[0022] The instantaneous value of a contaminant
concentration measured for inflow water sampled at a spot in
the inflow section 15 or near the inflow section 15 in the
15 bioreactor tank 10 is inputted to the inflow water quality
measurement value acquisition unit 50 together with the
sampling date and time. It is desirable that the kinds of
contaminants for input to the inflow water quality
measurement value acquisition unit 50 are the same as the
20 kinds of contaminants for input to the inflow water quality
variation pattern acquisition unit 40. Means for inputting
the inflow water quality measurement value to the inflow
water quality measurement value acquisition unit 50 may be
any means, e.g., a tablet, mouse operation, keyboard
25 operation, or input on a screen of a central monitoring
15
system.
[0023] The frequency at which the inflow water quality
measurement value is inputted to the inflow water quality
measurement value acquisition unit 50 may be arbitrary, but
5 in order to perform inflow water quality estimation with
higher accuracy, it is desirable that the inflow water
quality measurement value is inputted to the inflow water
quality measurement value acquisition unit 50 in a cycle not
longer than the cycle in which the inflow water quality
10 variation pattern is inputted to the inflow water quality
variation pattern acquisition unit 40. For example, in a
case where the inflow water quality variation pattern is
inputted to the inflow water quality variation pattern
acquisition unit 40 on a one-day (24-hours) basis, it is
15 desirable that a measurement value is inputted to the inflow
water quality measurement value acquisition unit 50 at a
frequency of one or more times a day.
[0024] The inflow water quality estimation unit 30
estimates an inflow water quality value subsequent to the
20 sampling date and time of inflow water quality measurement,
on the basis of the inflow water quality variation pattern
transmitted from the inflow water quality variation pattern
acquisition unit 40 via a signal line 40a and the measurement
value of the inflow water quality transmitted from the inflow
25 water quality measurement value acquisition unit 50 via a
16
signal line 50a. The inflow water quality value estimated by
the inflow water quality estimation unit 30 is transmitted to
the target aeration amount calculation unit 60 via the signal
line 30a. The target aeration amount calculation unit 60
5 calculates the target value for the aeration amount on the
basis of the inflow water quality value estimated by the
inflow water quality estimation unit 30.
[0025] Next, operation of the inflow water quality
estimation unit 30 will be described. The inflow water
10 quality estimation unit 30 includes a sampling date-and-time
extraction unit 31, an inflow water quality variation pattern
selection unit 32, and an inflow water quality calculation
unit 33. The sampling date-and-time extraction unit 31
extracts a feature of the sampling date and time of the
15 inflow water quality measurement value inputted to the inflow
water quality measurement value acquisition unit 50, and
associates the extracted feature with a feature of the date
and time of the inflow water quality variation pattern stored
in the inflow water quality variation pattern acquisition
20 unit 40. For example, in a case where inflow water quality
variation patterns for a weekday and a holiday have been
inputted to the inflow water quality variation pattern
acquisition unit 40 and an inflow water quality measurement
value for a weekday is inputted to the inflow water quality
25 measurement value acquisition unit 50, the sampling date-and-
17
time extraction unit 31 extracts "weekday" as the feature of
the sampling date and time.
[0026] The inflow water quality variation pattern
selection unit 32 selects the variation pattern of the inflow
5 water quality corresponding to the feature of the sampling
date and time extracted by the sampling date-and-time
extraction unit 31, from the inflow water quality variation
patterns inputted to the inflow water quality variation
pattern acquisition unit 40. For example, in a case where
10 "weekday" is extracted as the feature of the sampling date
and time as described above, the inflow water quality
variation pattern for "weekday" is selected.
[0027] The inflow water quality calculation unit 33
calculates an inflow water quality value subsequent to the
15 sampling date and time of inflow water quality measurement,
on the basis of the variation pattern of the inflow water
quality transmitted from the inflow water quality variation
pattern selection unit 32 and the measurement value of the
inflow water quality inputted to the inflow water quality
20 measurement value acquisition unit 50. That is, with the
sampling date and time of inflow water quality measurement
defined as a first time, an inflow water quality value
subsequent to the first time is calculated.
[0028] With reference to FIG. 4, an estimation value of
25 the inflow water quality calculated by the inflow water
18
quality calculation unit 33 will be described. For example,
it is assumed that a variation pattern of an inflow total
nitrogen concentration on a one-day basis is inputted to the
inflow water quality variation pattern acquisition unit 40,
5 and 25 mg/L is inputted as an inflow total nitrogen
concentration at 9:00 on October 1, to the inflow water
quality measurement value acquisition unit 50. In this case,
the inflow water quality calculation unit 33 outputs the
value of the inflow total nitrogen concentration for each
10 time until 9:00 on October 2 which is one day later, starting
from the measurement value of 25 mg/L at 9:00 on October 1.
That is, a variation pattern indicated by a solid line in
FIG. 4 is calculated.
[0029] For example, it is assumed that the sampling cycle
15 is 8 hours and an inflow water quality measurement value is
next inputted from the inflow water quality measurement value
acquisition unit 50 at 17:00 on October 1. Then, since the
day is still a weekday, the pattern is not changed in the
inflow water quality variation pattern selection unit 32, but
20 the value of the inflow total nitrogen concentration for each
time until 9:00 on October 2 which is one day later is
estimated again in accordance with the inputted inflow water
quality measurement value. Similarly, also when an inflow
water quality measurement value is inputted at 1:00 on
25 October 2 which is further 8 hours later, estimation is
19
performed again.
[0030] Here, an estimation algorithm implemented in the
inflow water quality calculation unit 33 may be any algorithm
that can estimate temporal variation in the inflow water
5 quality, using the inflow water quality variation pattern and
the inflow water quality measurement value as inputs, and for
example, a linear/nonlinear regression model, machine
learning, reinforcement learning, deep reinforcement
learning, deep learning, random forest, a neural network, and
10 another prediction method using artificial intelligence, may
be used. The inflow water quality variation pattern obtained
as a target of estimation can be used as a learned inflow
water quality variation pattern to the inflow water quality
variation pattern acquisition unit 40.
15 [0031] In a case where a variation pattern of the inflow
total nitrogen concentration on a one-week basis is inputted
to the inflow water quality variation pattern acquisition
unit 40, inflow total nitrogen concentrations for respective
times from 9:00 on October 1 to 9:00 on October 8 are
20 outputted.
[0032] As described above, since variation patterns of the
inflow water quality have been acquired in advance, it is not
necessary to use a sensor capable of continuously measuring
the inflow water quality. Further, an inflow water quality
25 subsequent to the sampling date and time of inflow water
20
quality measurement is estimated by combination with the
inflow water quality measurement value, whereby accuracy of
inflow water quality estimation can be improved in each
sampling.
5 [0033] The target aeration amount calculation units 60
(61, 62, 63) calculate target values for the amounts of
aeration to be supplied from the diffuser plates 11, 12, 13,
on the basis of the inflow water quality value estimated by
the inflow water quality estimation unit 30. The target
10 aeration amount calculation units 61, 62, 63 calculate the
target aeration amounts in association with the inflow water
quality estimation value corresponding to the time at which
the target aeration amounts are calculated. For example,
calculation is performed on the basis of the following
15 Formulae (1) to (3).
Qair1 = K1 × TN ... (1)
Qair2 = K2 × TN ... (2)
Qair3 = K3 × TN ... (3)
Here, Qair1, Qair2, and Qair3 are the target
20 aeration amounts calculated by the target aeration amount
calculation units 61, 62, 63, TN is the inflow water quality
estimated by the inflow water quality estimation unit 30, and
K1, K2, and K3 are proportionality constants.
[0034] By calculating the target aeration amounts on the
25 basis of Formulae (1) to (3), it becomes possible to control
21
the aeration amounts so as to follow variation in the inflow
water quality. Thus, control delay is suppressed, whereby
variation in the water quality after treatment is also
suppressed and a favorable treated water quality is obtained,
5 while an excessive aeration amount can be reduced. The
values of K1, K2, and K3 do not necessarily need to be all
equal to each other, and they may be set arbitrarily in
accordance with the positions of the diffuser plates 11, 12,
13. In particular, in a case where the influence of
10 variation in the inflow water quality is desired to be made
as small as possible, the proportionality constants may be
set so as to satisfy K1 > K2 > K3, whereby variation in the
inflow water quality can be suppressed earlier on the
preceding-stage side in the bioreactor tank 10. Formulae (1)
15 to (3) are merely an example, and in a case of desiring to
control the aeration amounts so as to follow an inflow load
(inflow contaminant concentration × inflow water amount), a
term of the inflow water amount may be added.
[0035] Next, an operation flow of the aeration amount
20 control device in embodiment 1 will be described with
reference to FIG. 5.
Variation patterns of the inflow water quality have
been inputted to the inflow water quality variation pattern
acquisition unit 40 in advance. From this state, control by
25 the aeration amount control device is started.
22
The operation flow of the aeration amount control
device includes six steps ST1 to ST6 shown in FIG. 5. After
the aeration amount control is started, the flow through
steps ST1 to ST6 is repeated at certain time intervals Δt.
5 Here, Δt is set at a value not greater than an update cycle T
of the target aeration amount of the target aeration amount
calculation unit 60, and it is desirable that Δt is about one
second to one minute. However, a period longer than a period
required for going through all the steps ST1 to ST6 needs to
10 be set as Δt.
[0036] First, in step ST1, whether or not time t
corresponds to the update cycle of the target aeration amount
of the target aeration amount calculation unit 60, is
determined. In a case where the update cycle T of the target
15 aeration amount is 5 minutes, step ST1 gives determination as
Yes at 5-minute intervals. In a case of Yes in step ST1, the
process proceeds to the next step ST2, and in a case of No in
step ST1, the process proceeds to step ST5.
[0037] In step ST2, whether or not a measurement value of
20 an inflow water quality has been newly inputted to the inflow
water quality measurement value acquisition unit 50 during a
period to time t from time t-Δt which is a time preceding by
one cycle of the repetition interval Δt, is determined. In a
case where the input cycle of the inflow water quality to the
25 inflow water quality measurement value acquisition unit 50 is
23
one day, step ST2 gives determination as Yes at one-day
intervals. In a case of Yes in step ST2, the process
proceeds to the next step ST3, and in a case of No in step
ST2, the process proceeds to step ST5.
5 [0038] Here, after water for inflow water quality
measurement is sampled, water quality measurement is
performed, and a result thereof is inputted to the inflow
water quality measurement value acquisition unit 50.
Therefore, the time at which the sampling for inflow water
10 quality measurement was performed is before the time t. For
example, it is assumed that the sampling for inflow water
quality measurement is performed at 9:00 on October 1, and
then, after one hour is taken for water quality analysis, the
measurement value at 9:00 on October 1 is inputted to the
15 inflow water quality measurement value acquisition unit 50.
In this case, the inflow water quality measurement value at
the time of 9:00 on October 1 is inputted when the time t in
the operation flow of the aeration amount control device is
10:00 on October 1.
20 [0039] Next, in step ST3, the inflow water quality
estimation unit 30 estimates an inflow water quality value
subsequent to the sampling date and time of inflow water
quality measurement. In the present embodiment, as shown in
FIG. 4, the inflow water quality variation pattern on a one25 day basis has been input to the inflow water quality
24
variation pattern acquisition unit 40. Therefore, variation
in the inflow water quality value from 9:00 on October 1 to
9:00 on October 2 is estimated.
[0040] Next, in step ST4, the inflow water quality
5 estimation value recorded at time t-Δt is updated to the
inflow water quality estimation value estimated in step ST3.
[0041] In step ST5, the target aeration amount calculation
units 61, 62, 63 refer to the inflow water quality estimation
value at the time t. In the present embodiment, in a case
10 where the time t is 10:00 on October 1, the inflow water
quality estimation value at 10:00 on October 1 is referred
to.
[0042] In step ST6, using the inflow water quality
estimation value referred to in step ST5 and Formulae (1) to
15 (3), each target aeration amount calculation unit 61, 62, 63
calculates a target aeration amount at time t. The
calculated target aeration amounts are transmitted to the air
volume adjustment valves 71, 72, 73 via the signal lines 61a,
62a, 63a, and the opening degrees of the air volume
20 adjustment valves 71, 72, 73 are adjusted so as to achieve
the target aeration amounts.
[0043] In the operation flow of the aeration amount
control device shown in FIG. 5, the water quality of the
treatment target water flowing into the bioreactor tank 10 at
25 9:00 on October 1 which is the first time is sampled, to
25
perform water quality measurement. Thereafter, at 10:00 on
October 1 which is a second time, the inflow water quality
measurement value acquisition unit 50 acquires an inflow
water quality measurement value at the first time, and the
5 inflow water quality estimation unit 30 estimates an inflow
water quality subsequent to the sampling date and time of
inflow water quality measurement. Here, since the
acquisition cycle for the water quality measurement value by
the inflow water quality measurement value acquisition unit
10 50 is one day, the water quality of the treatment target
water flowing into the bioreactor tank 10 at 9:00 on October
2 which is a third time is sampled, to perform water quality
measurement. Thereafter, at 10:00 on October 2 which is a
fourth time, the inflow water quality measurement value
15 acquisition unit 50 acquires an inflow water quality
measurement value at the third time, and the inflow water
quality estimation unit 30 estimates an inflow water quality
subsequent to the sampling date and time (third time) of
inflow water quality measurement. In this way, sampling of
20 the inflow water quality, acquisition of the measurement
value, and estimation of the inflow water quality are
repeated. The third time may be before 9:00 on October 2, as
long as sampling and determination of the measurement value
are performed so that an inflow water quality measurement
25 value subsequent to the first time can be acquired at the
26
fourth time. As a matter of course, a newer inflow water
quality measurement value can be acquired if the third time
is closer to 9:00 on October 2.
[0044] As described above, in the water treatment system
5 according to the present embodiment 1, the inflow water
quality estimation unit 30 estimates an inflow water quality
value on the basis of the inflow water quality variation
pattern inputted to the inflow water quality variation
pattern acquisition unit 40 and the inflow water quality
10 measurement value inputted to the inflow water quality
measurement value acquisition unit 50, whereby it becomes
possible to estimate the inflow water quality value with high
accuracy while reflecting both of the most recent water
quality measurement value and difference in the inflow water
15 quality variation pattern depending on the sampling date and
time. Further, the target aeration amount calculation units
61, 62, 63 calculate the target aeration amounts, using the
estimated inflow water quality value, and adjust the air
volume adjustment valves 71, 72, 73 in accordance with the
20 calculated target aeration amounts. Thus, it is possible to
perform treatment according to the water quality value of
inflow water while suppressing control delay, whereby it
becomes possible to reduce an excessive aeration amount while
obtaining a favorable treated water quality.
25 [0045] Conventionally, an operator who measures the water
27
quality and an operator who performs operation control for
the water treatment system are different, and thus it takes
time to reflect a water quality result in control for the
water treatment system. However, as in the present
5 embodiment, control for the water treatment system is
performed using the estimated inflow water quality estimation
value, whereby it becomes possible to perform control in
accordance with the water quality.
[0046] Embodiment 2
10 Hereinafter, a water treatment system according to
embodiment 2 will be described with reference to FIG. 6.
FIG. 6 is a block diagram showing the configuration
of an aeration amount control device of the water treatment
system according to embodiment 2. Difference from embodiment
15 1 is that the inflow water quality estimation unit 30
includes a rain influence determination unit 34. The other
configurations are the same as those in embodiment 1 and
therefore the description thereof is omitted.
[0047] In FIG. 6, the rain influence determination unit 34
20 determines whether or not the inflow water quality
measurement value inputted to the inflow water quality
measurement value acquisition unit 50 has been influenced by
rain, and transmits the inflow water quality measurement
value inputted to the inflow water quality measurement value
25 acquisition unit 50 and a rain influence degree to the inflow
28
water quality calculation unit 33.
[0048] In rainy weather, the treatment target water is
diluted by rainwater, so that the contaminant concentration
in the inflow water becomes smaller than in fine weather, and
5 thus estimation accuracy for the inflow water quality might
be deteriorated. Therefore, in a case where the inflow water
quality measurement value is smaller than a determination
threshold preset in the rain influence determination unit 34,
the rain influence determination unit 34 calculates a rain
10 influence degree. In calculation of the rain influence
degree, weather information from a weather radar in a target
treatment district of the plant site may be acquired to
determine the influence of rain, and a result thereof may be
used.
15 [0049] The rain influence degree is a ratio of dilution of
inflow water due to rain entry water, and is calculated by
Formula (4).
Dilution ratio = 1 - Ssp_r/Ssp_ave ... (4)
Here, Ssp_r is an inflow water quality measurement
20 value sampled at the date and time when there is an influence
of rain, and Ssp_ave is an average value of inflow water
quality measurement values in fine weather. As the average
value of the inflow water quality measurement values, an
inflow water quality measurement value at the same sampling
25 date and time is acquired a plurality of times, and the
29
average value thereof is inputted to the rain influence
determination unit 34.
[0050] Next, an estimation value of an inflow water
quality calculated by the inflow water quality calculation
5 unit 33 in a case where the rain influence determination unit
34 determines that there is an influence of rain, will be
described with reference to FIG. 7.
For example, it is assumed that a variation pattern
of an inflow total nitrogen concentration on a one-day basis
10 is inputted to the inflow water quality variation pattern
acquisition unit 40, 5 mg/L is inputted as an inflow total
nitrogen concentration at 9:00 on October 8, to the inflow
water quality measurement value acquisition unit 50, and the
rain influence determination unit 34 determines that "there
15 is an influence of rain". In this case, the inflow water
quality calculation unit 33 outputs the value of the inflow
total nitrogen concentration for each time until 9:00 on
October 9 which is one day later, starting from the inflow
water quality measurement value of 5 mg/L at 9:00 on October
20 8. Here, in accordance with the dilution ratio due to rain,
estimation is performed so that the estimation value of the
inflow total nitrogen concentration becomes smaller and the
variation width of the inflow water quality value also
becomes smaller, than in a case where there is no influence
25 of rain.
30
[0051] A broken line in FIG. 7 indicates a variation
pattern of an estimated inflow total nitrogen concentration
on a weekday in a case where the weather is fine, i.e., there
is no influence of rain. A future variation is estimated
5 from the dilution ratio and the inflow water quality
measurement value transmitted from the rain influence
determination unit 34, whereby a pattern indicated by a solid
line can be obtained. In this way, it is possible to
estimate the inflow water quality value accurately even on a
10 day when there is an influence of rain. An estimation
algorithm implemented in the inflow water quality calculation
unit 33 may be any algorithm that can estimate temporal
variation in the inflow water quality value, using the inflow
water quality variation pattern, the inflow water quality
15 measurement value, and the dilution ratio due to rain, as
inputs, and for example, a linear/nonlinear regression model,
machine learning, reinforcement learning, deep reinforcement
learning, deep learning, random forest, a neural network, and
another prediction method using artificial intelligence, may
20 be used as in embodiment 1.
[0052] The target aeration amount calculation units 61,
62, 63 calculate target values for the amounts of aeration to
be supplied from the diffuser plates 11, 12, 13, on the basis
of the inflow water quality value estimated by the inflow
25 water quality estimation unit 30. The target aeration amount
31
calculation units 61, 62, 63 calculate the target aeration
amounts in association with the inflow water quality
estimation value corresponding to the time at which the
target aeration amounts are calculated. In general, in a
5 case where there is an influence of rain, the contaminant
concentration in inflow water is extremely small, and
therefore, if the same aeration amounts as in fine weather
are used, excessive aeration would be supplied. In the
present embodiment, the target aeration amount calculation
10 units 61, 62, 63 calculate the target aeration amounts on the
basis of the estimation value of the inflow water quality for
which the influence of rain has been considered. Thus, it is
possible to not only suppress variation in the inflow water
quality and stabilize the treated water quality but also
15 further reduce excessive aeration as compared to a case of
fine weather.
[0053] As described above, according to embodiment 2, the
inflow water quality estimation unit 30 estimates the inflow
water quality, using the rain influence degree determined by
20 the rain influence determination unit 34. Thus, it becomes
possible to estimate high-accuracy inflow water quality value
even in a case where an inflow water quality variation
pattern different from a normal one in fine weather is
predicted from a measured inflow water quality value.
25 Further, the target aeration amount calculation units 61, 62,
32
63 calculate target aeration amounts, using the inflow water
quality estimation value for which the influence of rain has
been considered, and adjust the air volume adjustment valves
71, 72, 73 in accordance with the calculated target aeration
5 amounts. Thus, it is possible to perform treatment according
to the water quality of inflow water even in rainy weather,
whereby it becomes possible to reduce an excessive aeration
amount while obtaining a favorable treated water quality.
[0054] Embodiment 3
10 Hereinafter, a water treatment system according to
embodiment 3 will be described with reference to FIG. 8.
FIG. 8 is a block diagram showing the configuration
of the water treatment system according to embodiment 3.
Difference from embodiment 1 is that a contaminant
15 concentration measurement unit 80 having a concentration
measurement instrument is provided for measuring a
contaminant concentration at a time point when treatment in
the bioreactor tank 10 for the treatment target water flowing
into the bioreactor tank 10 is finished. The other
20 configurations are the same as those in embodiment 1 or 2 and
the description thereof is omitted.
[0055] The contaminant concentration measurement unit 80
according to the present embodiment 3 corresponds to a second
water quality measurement value acquisition unit different
25 from the inflow water quality measurement value acquisition
33
unit 50. A main role of the contaminant concentration
measurement unit 80 is to measure a contaminant concentration
at a time point when treatment is finished, and therefore it
is desirable that the concentration measurement instrument
5 thereof is provided at a position closer to the outflow
section 16 in the bioreactor tank 10. Alternatively, the
concentration measurement instrument of the contaminant
concentration measurement unit 80 may be provided at the
outflow section 16.
10 [0056] For the concentration measurement instrument, one
or more kinds are selected from treatment targets in the
bioreactor tank 10. Examples include BOD, COD, ammonia
nitrogen, total nitrogen, Kjeldahl nitrogen, total
phosphorus, and phosphate phosphorus. The contaminant
15 concentration measurement unit 80 does not necessarily need
to have a concentration measurement instrument for the same
kind as the water quality inputted to the inflow water
quality variation pattern acquisition unit 40 and the inflow
water quality measurement value acquisition unit 50, and may
20 have a concentration measurement instrument for a desired
kind of treated water quality to be monitored.
The contaminant concentration measured by the
contaminant concentration measurement unit 80 is transmitted
to the target aeration amount calculation units 61, 62, 63
25 via a signal line 80a. The target aeration amount
34
calculation units 61, 62, 63 calculate target values for
aeration amounts on the basis of the estimation value of the
inflow water quality estimated by the inflow water quality
estimation unit 30 and the contaminant concentration measured
5 by the contaminant concentration measurement unit 80.
[0057] Next, a calculation method for target values for
aeration amounts by the target aeration amount calculation
units 61, 62, 63 will be described. Hereinafter, it is
assumed that a variation pattern of an inflow water total
10 nitrogen concentration and an inflow water quality
measurement value are respectively inputted to the inflow
water quality variation pattern acquisition unit 40 and the
inflow water quality measurement value acquisition unit 50,
and the contaminant concentration measurement unit 80 has an
15 ammonia nitrogen concentration meter as a contaminant
concentration meter.
[0058] The target values for the aeration amounts are
determined through control for following variation in the
inflow water quality value estimated by the inflow water
20 quality estimation unit 30 and control (proportional integral
(PI) control) for performing operation so that the
contaminant concentration measured by the contaminant
concentration measurement unit 80 becomes equal to a certain
target water quality. As a specific example, the target
25 values of the aeration amounts are determined on the basis of
35
Formulae (5) to (7) shown below. Among components of a
control unit in the present embodiment 3, a target water
quality setting unit for setting a target water quality is
not shown, but the target aeration amount calculation units
5 61, 62, 63 have a target water quality setting unit therein,
and an operator can set a target water quality from the
outside.
Qair1 = K1 × TN + Kp1 × (NH4 - NH4*)
+ Ki1 × Σ(NH4 - NH4*) ... (5)
10 Qair2 = K2 × TN + Kp2 × (NH4 - NH4*)
+ Ki2 × Σ(NH4 - NH4*) ... (6)
Qair3 = K3 × TN + Kp3 × (NH4 - NH4*)
+ Ki3 × Σ(NH4 - NH4*) ... (7)
[0059] Here, Qair1, Qair2, and Qair3 are target aeration
15 amounts calculated by the target aeration amount calculation
units 61, 62, 63, TN is the estimation value of the inflow
water total nitrogen concentration estimated by the inflow
water quality estimation unit 30, K1, K2, and K3 are
proportionality constants, Kp1, Kp2, and Kp3 are proportional
20 gains (constants), Ki1, Ki2, and Ki3 are integral gains
(constants), NH4 is an ammonia nitrogen concentration
measured by the contaminant concentration measurement unit
80, and NH4* is a target value for the ammonia nitrogen
concentration. In addition, Σ represents the sum of
25 measurement values of (NH4 - NH4*) after calculation of
36
aeration amounts by Formulae (5) to (7) is started. For
example, in a case where calculation of aeration amounts
based on Formulae (5) to (7) is performed at one-minute
intervals, the value of Σ(NH4 - NH4*) after one hour is the
5 sum of 60 measurement values of (NH4 - NH4*) obtained per one
minute from a time just after calculation of aeration amounts
by Formulae (5) to (7) is started.
[0060] The first terms in Formulae (5) to (7) represent
proportional control for the estimation value of the inflow
10 water total nitrogen concentration estimated by the inflow
water quality estimation unit 30, whereby it becomes possible
to control the aeration amounts so as to follow variation in
the inflow water quality.
[0061] The second terms and the third terms in Formulae
15 (5) to (7) represent PI control based on a difference between
the ammonia nitrogen concentration measured by the
contaminant concentration measurement unit 80 and the target
value for ammonia nitrogen, whereby the aeration amounts are
controlled so that the treated water quality becomes constant
20 with respect to the target value, and thus necessary amounts
of aeration can be supplied to the bioreactor tank 10 without
excess or deficiency. The proportionality constants (K1, K2,
K3), the proportional gains (Kp1, Kp2, Kp3), and the integral
gains (Ki1, Ki2, Ki3) for the estimation value of the inflow
25 water quality need not be all equal among the target aeration
37
amount calculation units 61, 62, 63, and may be set at
arbitrary values in accordance with the positions of the
diffuser plates 11, 12, 13. In particular, in a case of
desiring to make the influence of variation in the inflow
5 water quality value as small as possible, the proportionality
constants are set so as to satisfy K1 > K2 > K3, whereby
variation in the inflow water quality value can be suppressed
early on the preceding-stage side in the bioreactor tank 10.
[0062] In addition, in a case of desiring to make
10 variation in the treated water quality value as small as
possible, setting is made so as to satisfy Kp1 < Kp2 < Kp3
and Ki1 < Ki2 < Ki3, whereby the influence of PI control can
be made greater for the air volume adjustment valve that is
closer to the location where the contaminant concentration
15 measurement unit 80 is provided. Formulae (5) to (7) are
merely an example, and in a case of desiring to control the
aeration amounts so as to follow an inflow load (inflow
contaminant concentration × inflow water amount), a term of
an inflow water amount may be added at the first term.
20 [0063] As described above, according to embodiment 3, the
target aeration amount calculation units 61, 62, 63 calculate
target aeration amounts, using the inflow water quality
estimation value estimated by the inflow water quality
estimation unit 30 and the contaminant concentration measured
25 by the contaminant concentration measurement unit 80, whereby
38
the treated water quality can be controlled to be constant
with improved response of the aeration amounts with respect
to variation in the inflow water quality, without using a
sensor capable of continuously measuring the inflow water
5 quality. Thus, it is possible to reduce an excessive
aeration amount by supplying aeration without excess or
deficiency, while suppressing variation in the treated water
quality value and obtaining a favorable treated water
quality.
10 [0064] In the above embodiments 1 to 3, the inflow water
quality estimation unit 30 and the target aeration amount
calculation unit 60 of the water treatment system 100 are
composed of a processor 301 and a storage device 302, as
shown in a hardware example in FIG. 9. Although not shown,
15 the storage device includes a volatile storage device such as
a random access memory, and a nonvolatile auxiliary storage
device such as a flash memory. The storage device may
include an auxiliary storage device of a hard disk, instead
of a flash memory. The processor 301 executes a program
20 inputted from the storage device 302. In this case, the
program is inputted from the auxiliary storage device to the
processor 301 via the volatile storage device. The processor
301 may output data such as a calculation result to the
volatile storage device of the storage device 302, or may
25 store such data into the auxiliary storage device via the
39
volatile storage device.
[0065] Embodiment 4
Hereinafter, a water treatment system according to
embodiment 4 will be described with reference to FIG. 10 to
5 FIG. 14. In the present embodiment, an inflow water quality
inference device in which the inflow water quality estimation
unit 30 shown in each of embodiments 1 to 3 is implemented
with a machine learning device provided, will be described.
FIG. 10 shows the configuration of an inflow water
10 quality inference device 400 of the water treatment system
according to the present embodiment 4, and the inflow water
quality inference device 400 includes a learning device 410
and an inference device 420. The inflow water quality
inference device 400 may be the inflow water quality
15 estimation unit 30. Hereinafter, a procedure for inferring
an inflow water quality value will be described, with the
procedure separated into a "learning phase" and a
"utilization phase" for actually performing inference.
[0066]
20 FIG. 11 shows the configuration of a learning
device 410. The learning device 410 includes a data
acquisition unit 411, a model generation unit 412, and a
trained model storage unit 413.
FIG. 12 is a flowchart showing a processing
25 procedure for executing the learning phase using the learning
40
device 410.
[0067] The data acquisition unit 411 acquires data b1 of
the kind and the concentration of a contaminant and data b2
of the measurement date and time, as input data, from the
5 inflow water quality measurement value acquisition unit 50,
and combines both data to obtain time-series training data
(step ST101).
[0068] The model generation unit 412 learns variation in
the inflow water quality on the basis of the training data
10 outputted from the data acquisition unit 411 (step ST102).
That is, from a plurality of time-series data of the
concentrations of contaminants over a certain period, for
example, water quality variation for each kind of
contaminants on each day (weekday or holiday) is learned to
15 generate a trained model 414 including water quality
variation patterns. In learning for generating the trained
model 414, for example, a linear/nonlinear regression model,
machine learning, reinforcement learning, deep reinforcement
learning, deep learning, random forest, a neural network, and
20 another prediction method using artificial intelligence, may
be used.
[0069] By executing learning as described above, the model
generation unit 412 generates and outputs the trained model
414, and the trained model storage unit 413 stores the
25 trained model 414 outputted from the model generation unit
41
412 (step ST103).
[0070]
FIG. 13 shows the configuration of the inference
device 420. The inference device 420 includes a data
5 acquisition unit 421 and an inference unit 422.
FIG. 14 is a flowchart showing a processing
procedure for executing the utilization phase for inferring a
water quality value of inflow water using the inference
device 420.
10 [0071] The data acquisition unit 421 acquires water
quality measurement data b11 of a contaminant from the inflow
water quality measurement value acquisition unit 50 and an
inflow water quality variation pattern b12 selected by the
inflow water quality variation pattern selection unit 32, as
15 input data (step ST111).
[0072] The inference unit 422 infers variation in the
inflow water quality, using the trained model 414. That is,
new water quality measurement data b11 of a contaminant at a
given time point acquired by the data acquisition unit and a
20 selected inflow water quality variation pattern b12, are
inputted to the trained model 414 (step ST112), whereby
subsequent water quality variation values can be inferred
with high accuracy and can be outputted from the inference
device 420 (step ST113).
25 A result of inference of inflow water quality
42
values is outputted to the target aeration amount calculation
unit 60 (step ST114).
[0073] The water quality measurement data b11 of a
contaminant from the inflow water quality measurement value
5 acquisition unit 50 and the inflow water quality variation
pattern b12 selected by the inflow water quality variation
pattern selection unit 32 are inputted from the data
acquisition unit 421 to the inference unit 422. However, the
inflow water quality variation pattern b12 selected by the
10 inflow water quality variation pattern selection unit 32
merely serves as reference data, and may be omitted. This is
because the water quality measurement data b11 of a
contaminant acquired from the inflow water quality
measurement value acquisition unit 50 is imparted with water
15 quality information such as the date and time of measurement
and the kind of the contaminant and thus can be applied to
the trained data.
[0074] In the above description, the example corresponding
to embodiment 1 has been described. However, as a matter of
20 course, the inflow water quality inference device 400 for
rainy weather as in embodiment 2 can be constructed by
including rain information in data to be inputted to the
inflow water quality inference device 400.
In addition, as a matter of course, the inflow
25 water quality inference device 400 is also applicable to
43
embodiment 3.
[0075] The inference device 420 may be the inflow water
quality estimation unit 30, and the learning device 410 and
the trained model storage unit 413 may be externally provided
5 to the inflow water quality estimation unit 30 of each of
embodiments 1 to 3, or these may be integrated as the inflow
water quality inference device 400 as shown in FIG. 10.
[0076] An example of hardware of the inflow water quality
inference device 400 is composed of a processor and a storage
10 device as in the configuration shown in FIG. 9, and therefore
the description thereof is omitted. The processor executes a
program to implement the functions of the learning device 410
and the inference device 420 of the inflow water quality
inference device 400.
15 [0077] As the patterns to be inputted to the inflow water
quality variation pattern acquisition unit 40 in each of
embodiments 1 to 4, inflow water quality variation patterns
based on the trained model may be used.
[0078] In the present embodiment 4, it has been described
20 that an inference result of the inflow water quality is
outputted using the trained model 414 trained in the model
generation unit 412. However, a trained model may be
acquired from the outside and an inference result of the
inflow water quality may be outputted on the basis of the
25 acquired trained model.
44
In the present embodiment 4, regarding the learning
phase and the utilization phase described separately from
each other, the learning phase may be carried out first, and
then the utilization phase may be carried out, or both phases
5 may be carried out in parallel. In a case of carrying out
both phases in parallel, a threshold for completion of
learning, e.g., the number of acquired data, may be set.
Then, only learning may be carried out during a period in
which learning has not been completed yet, and after learning
10 is completed, both phases may be carried out in parallel.
[0079] As described above, according to embodiment 4,
effects of embodiment 1 are provided, and in addition, data
b1 of the kind and the concentration of a contaminant and
data b2 of the measurement date and time are acquired from
15 the inflow water quality measurement value acquisition unit
50, and both data are combined to obtain time-series training
data, thus generating a trained model, and new present water
quality measurement data b11 of a contaminant at a given time
point and a selected inflow water quality variation pattern
20 b12 are inputted to the trained model, whereby subsequent
water quality variation values can be inferred with high
accuracy. Thus, it becomes possible to perform aeration
amount control with higher accuracy on the basis of a result
of the above inference.
25 [0080] Although the disclosure is described above in terms
45
of various exemplary embodiments and implementations, it
should be understood that the various features, aspects, and
functionality described in one or more of the individual
embodiments are not limited in their applicability to the
5 particular embodiment with which they are described, but
instead can be applied, alone or in various combinations to
one or more of the embodiments of the disclosure.
It is therefore understood that numerous
modifications which have not been exemplified can be devised
10 without departing from the scope of the present disclosure.
For example, at least one of the constituent components may
be modified, added, or eliminated. At least one of the
constituent components mentioned in at least one of the
preferred embodiments may be selected and combined with the
15 constituent components mentioned in another preferred
embodiment.
DESCRIPTION OF THE REFERENCE CHARACTERS
[0081] 10 bioreactor tank
20 11, 12, 13 diffuser plate
15 inflow section
16 outflow section
20 blower
20a pipe
25 30 inflow water quality estimation unit
46
30a signal line
31 sampling date-and-time extraction unit
32 inflow water quality variation pattern
selection unit
5 33 inflow water quality calculation unit
34 rain influence determination unit
40 inflow water quality variation pattern
acquisition unit
40a signal line
10 50 inflow water quality measurement value
acquisition unit
50a signal line
60, 61, 62, 63 target aeration amount calculation
unit
15 61a, 62a, 63a signal line
70, 71, 72, 73 air volume adjustment valve
80 contaminant concentration measurement unit
80a signal line
100 water treatment system
20 301 processor
302 storage device
400 inflow water quality inference device
410 learning device
411 data acquisition unit
25 412 model generation unit
47
413 trained model storage unit
414 trained model
420 inference device
421 data acquisition unit
5 422 inference unit

We Claim :
[Claim 1] A water treatment system for performing water
treatment through biological oxidation while performing
aeration from a blower to a reaction tank, the water
5 treatment system comprising:
a first water quality measurement value acquisition
unit which acquires a water quality measurement value at a
first time point, of treatment target water flowing into the
reaction tank;
10 a water quality variation pattern acquisition unit
which, from time-series change in water quality information
of the treatment target water acquired in advance, acquires a
water quality variation pattern according to a condition at a
time of acquisition of the water quality information acquired
15 in advance;
an inflow water quality estimation unit which
selects a water quality variation pattern that matches a
condition at a time of acquisition of the water quality
measurement value at the first time point, from the water
20 quality variation patterns included in the water quality
variation pattern acquisition unit, and estimates an inflow
water quality value subsequent to the first time point from
the selected water quality variation pattern; and
a control unit which controls an aeration amount of
25 the blower subsequent to the first time point, on the basis
49
of the inflow water quality value estimated by the inflow
water quality estimation unit.
[Claim 2] The water treatment system according to claim 1,
5 further comprising a second water quality measurement value
acquisition unit, wherein
the second water quality measurement value
acquisition unit acquires a water quality measurement value
of treated water that has been treated in the reaction tank,
10 and
the control unit further controls the aeration
amount of the blower subsequent to the first time point, on
the basis of the water quality measurement value of the
treated water acquired by the second water quality
15 measurement value acquisition unit.
[Claim 3] The water treatment system according to claim 1
or 2, wherein
the inflow water quality estimation unit includes
20 a learning device including a data acquisition
unit which acquires time-series change in water quality
information of the treatment target water, as training data,
in advance, and a model generation unit which generates a
trained model for inferring a plurality of the water quality
25 variation patterns each of which is time-series change
50
according to a condition at a time of acquisition of the
water quality information, using the training data, and
an inference device which infers the inflow
water quality value subsequent to the first time point, from
5 the water quality measurement value at the first time point
of the treatment target water acquired by the first water
quality measurement value acquisition unit, using the trained
model.
10 [Claim 4] An aeration amount control device used for a water
treatment system for performing water treatment through
biological oxidation while performing aeration from a blower
to a reaction tank, the aeration amount control device
comprising:
15 a first water quality measurement value acquisition
unit which acquires a water quality measurement value at a
first time point, of treatment target water flowing into the
reaction tank;
a water quality variation pattern acquisition unit
20 which, from time-series change in water quality information
of the treatment target water acquired in advance, acquires a
water quality variation pattern according to a condition at a
time of acquisition of the water quality information acquired
in advance;
25 an inflow water quality estimation unit which
51
selects a water quality variation pattern that matches a
condition at a time of acquisition of the water quality
measurement value at the first time point, from the water
quality variation patterns included in the water quality
5 variation pattern acquisition unit, and estimates an inflow
water quality value subsequent to the first time point from
the selected water quality variation pattern; and
a target aeration amount calculation unit which
calculates a target aeration amount of aeration to be
10 supplied from the blower to the reaction tank subsequent to
the first time point, on the basis of the inflow water
quality value estimated by the inflow water quality
estimation unit.
15 [Claim 5] The aeration amount control device according to
claim 4, wherein
the first water quality measurement value
acquisition unit acquires a water quality measurement value
of treatment target water flowing into the reaction tank at a
20 second time point subsequent to the first time point, and
the inflow water quality estimation unit selects
the water quality variation pattern from the water quality
variation pattern acquisition unit, correspondingly to
acquisition date and time of the water quality measurement
25 value at the second time point acquired by the first water
52
quality measurement value acquisition unit, and estimates an
inflow water quality value subsequent to the second time
point.
5 [Claim 6] The aeration amount control device according to
claim 4 or 5, wherein
the inflow water quality estimation unit includes a
rain influence determination unit which determines whether
the water quality measurement value acquired by the first
10 water quality measurement value acquisition unit has been
influenced by rain, and
in a case where the rain influence determination
unit determines that the water quality measurement value has
been influenced, the inflow water quality value subsequent to
15 the first time point is estimated on the basis of a rain
influence degree and the water quality variation pattern
selected from the water quality variation pattern acquisition
unit.
20 [Claim 7] The aeration amount control device according to any
one of claims 4 to 6, further comprising a second water
quality measurement value acquisition unit, wherein
the second water quality measurement value
acquisition unit acquires a water quality measurement value
25 of treated water that has been treated in the reaction tank,
53
and
the target aeration amount calculation unit
calculates the target aeration amount of aeration to be
supplied from the blower to the reaction tank, on the basis
5 of the inflow water quality value estimated by the inflow
water quality estimation unit and the water quality
measurement value of the treated water acquired by the second
water quality measurement value acquisition unit.
10 [Claim 8] An aeration amount control method used for a water
treatment system for performing water treatment through
biological oxidation while performing aeration from a blower
to a reaction tank, the aeration amount control method
comprising:
15 a water quality variation pattern acquisition step
of, from time-series change in water quality information of
treatment target water flowing into the reaction tank in
advance, acquiring a water quality variation pattern
according to a condition at a time of acquisition of the
20 water quality information acquired in advance;
a water quality measurement value acquisition step
of acquiring a water quality measurement value at a first
time point, of the treatment target water flowing into the
reaction tank;
25 an inflow water quality estimation step of
54
selecting a water quality variation pattern that matches an
acquisition condition for the water quality measurement value
at the first time point, from a plurality of the water
quality variation patterns acquired in the water quality
5 variation pattern acquisition step, and estimating an inflow
water quality value subsequent to the first time point from
the selected water quality variation pattern; and
a target aeration amount calculation step of
calculating a target aeration amount of aeration to be
10 supplied from the blower to the reaction tank subsequent to
the first time point, on the basis of the inflow water
quality value estimated in the inflow water quality
estimation step.