FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
[See section 10, Rule 13]
AERATION VOLUME CONTROL DEVICE AND AERATION VOLUME 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:
AERATION VOLUME CONTROL DEVICE AND AERATION VOLUME CONTROL
5 METHOD
Field
[0001] The present disclosure relates to an aeration
volume control device and an aeration volume control method
10 for controlling the volume of aeration that is the volume
of air supplied to a reaction tank that performs biological
treatment.
Background
15 [0002] An example of a treatment system of
sewage/wastewater containing organic matter and ammonia
nitrogen is an activated sludge method. The activated
sludge method is a method that stores activated sludge,
which is a microbial community having a purification
20 function, in a reaction tank and performs aeration to
supply air while causing the activated sludge to be mixed
and in contact with wastewater, thereby oxidizing and
decomposing pollutants in the wastewater. The biological
reaction tank needs to be aerated with an appropriate
25 volume of air in response to a change in load of the inflow
pollutants, and the volume of aeration is feedforward
controlled by measuring ammonia concentration of the inflow
water. However, an ammonia concentration meter is
expensive. For this reason, Patent Literature 1 discloses
30 an aeration volume control method that measures a
conductivity value of inflow water, estimates an ammonia
concentration value from a correlation between a
conductivity value measured in advance and an ammonia
3
concentration value, and controls the volume of aeration on
the basis of the estimated ammonia concentration value.
Citation List
5 Patent Literature
[0003] Patent Literature 1: Japanese Patent Application
Laid-open No. 2009-119329
Summary
10 Technical Problem
[0004] In addition to ammonia, the inflow water contains
further coexisting substances such as other ions that
affect conductivity. An example of the coexisting
substance is a chloride ion. When the concentration of the
15 coexisting substance changes, the correlation between the
conductivity value of the inflow water and the ammonia
concentration value changes from a preset correlation.
Therefore, an error may occur between the ammonia
concentration value estimated on the basis of the measured
20 conductivity value and actual ammonia concentration of
treatment target water. However, the conventional
technique has not assumed that a change in the
concentration of the coexisting substance causes the
correlation between the conductivity value and the ammonia
25 concentration value to change from the preset correlation,
and an error occurs between the ammonia concentration value
estimated on the basis of the measured conductivity value
and the actual ammonia concentration. That is, the
conventional technique does not take into consideration the
30 change in the concentration of the coexisting substance in
the inflow water, and thus has had a problem that an
appropriate volume of air cannot be supplied to the
biological reaction tank in response to a change in load of
4
inflow ammonia.
[0005] The present disclosure has been made in view of
the above, and an object of the present disclosure to
provide an aeration volume control device capable of
5 supplying an appropriate volume of air to a biological
reaction tank in response to a change in load of inflow
ammonia even when concentration of a coexisting substance
in inflow water changes.
10 Solution to Problem
[0006] In order to solve the above problem and achieve
the object, the present disclosure is an aeration volume
control device that controls the volume of aeration which
is the volume of oxygen-containing gas supplied to a
15 biological reaction tank performing biological treatment on
treatment target water, the aeration volume control device
including an ammonia concentration sensor, a conductivity
sensor, a conductivity-concentration correlation
information storage unit, a first estimation unit, a target
20 aeration volume calculation unit, and a conductivityconcentration correlation information update unit. The
ammonia concentration sensor measures ammonia concentration
of treated water obtained when the treatment target water
in the biological reaction tank is subjected to the
25 biological treatment. The conductivity sensor measures
conductivity of the treatment target water flowing into the
biological reaction tank. The conductivity-concentration
correlation information storage unit stores conductivityconcentration correlation information indicating a
30 correlation between the conductivity of the treatment
target water and ammonia concentration of the treatment
target water. The first estimation unit estimates a first
estimated ammonia concentration value from the
5
conductivity-concentration correlation information on the
basis of a conductivity value measured by the conductivity
sensor, the first estimated ammonia concentration value
being an estimated value of the ammonia concentration of
5 the treatment target water. The target aeration volume
calculation unit calculates a target value of the volume of
aeration to the biological reaction tank on the basis of
the first estimated ammonia concentration value of the
treated water and the ammonia concentration value of the
10 treated water measured by the ammonia concentration sensor.
The conductivity-concentration correlation information
update unit updates the conductivity-concentration
correlation information. The conductivity-concentration
correlation information update unit includes a reception
15 unit and an update processing unit. The reception unit
receives a second estimated ammonia concentration value
that is a value of the ammonia concentration of the
treatment target water, the second estimated ammonia
concentration value being measured or estimated by a method
20 different from a method of estimating the first estimated
ammonia concentration value. The update processing unit
updates the conductivity-concentration correlation
information on the basis of the second estimated ammonia
concentration value received by the reception unit and the
25 conductivity value measured by the conductivity sensor.
Advantageous Effects of Invention
[0007] The aeration volume control device according to
the present disclosure has an effect of being able to
30 supply an appropriate volume of air to the biological
reaction tank in response to a change in load of the inflow
ammonia even when the concentration of the coexisting
substance in the inflow water changes.
6
Brief Description of Drawings
[0008] FIG. 1 is a diagram schematically illustrating an
example of a configuration of an aeration volume control
5 system including an aeration volume control device
according to a first embodiment.
FIG. 2 is a block diagram schematically illustrating
an example of a configuration of a database update unit of
the aeration volume control device according to the first
10 embodiment.
FIG. 3 is a flowchart illustrating an example of a
processing procedure of an aeration volume control method
according to the first embodiment.
FIG. 4 is a flowchart illustrating an example of a
15 procedure of a method of updating conductivityconcentration correlation information in the aeration
volume control device according to the first embodiment.
FIG. 5 is a block diagram illustrating an example of a
configuration of a database update unit of an aeration
20 volume control device according to a second embodiment.
FIG. 6 is a diagram illustrating an example of a
configuration of a learning device that generates a trained
model of the aeration volume control device according to
the second embodiment.
25 FIG. 7 is a diagram schematically illustrating an
example of a neural network used by a model generation unit
in FIG. 6.
FIG. 8 is a flowchart illustrating an example of a
procedure of learning processing by the learning device.
30 FIG. 9 is a diagram schematically illustrating an
example of a configuration of a second estimation unit of
the aeration volume control device according to the second
embodiment.
7
FIG. 10 is a flowchart illustrating an example of a
procedure of estimation processing by a second estimation
unit of the aeration volume control device according to the
second embodiment.
5 FIG. 11 is a flowchart illustrating an example of a
procedure of a method of updating conductivityconcentration correlation information in the aeration
volume control device according to the second embodiment.
FIG. 12 is a diagram illustrating an example of a
10 hardware configuration of a control circuit.
Description of Embodiments
[0009] Hereinafter, an aeration volume control device
and an aeration volume control method according to
15 embodiments of the present disclosure will be described in
detail with reference to the drawings.
[0010] First Embodiment.
FIG. 1 is a diagram schematically illustrating an
example of a configuration of an aeration volume control
20 system including an aeration volume control device
according to a first embodiment. An aeration volume
control system 1 includes a biological reaction tank 10, an
air diffuser plate 11, a blower 12, an air volume
regulation unit 13, a conductivity sensor 15, a treated
25 water ammonia concentration sensor 14, and an aeration
volume control device 30.
[0011] The biological reaction tank 10 is a water tank
for performing biological treatment on treatment target
water. Specifically, the biological reaction tank 10 is
30 the water tank that stores activated sludge therein and
uses the activated sludge to biologically treat the
treatment target water into treated water 101 having a
predetermined nitrogen concentration or less. The
8
biological reaction tank 10 is installed in a water
purification plant, a sewage treatment plant, a wastewater
treatment facility of a factory, or the like. An inflow
unit 102 and an outflow unit 103 are connected to the
5 biological reaction tank 10. The inflow unit 102 is a pipe
or a conduit into which the treatment target water as a
treatment target flows. The outflow unit 103 is a pipe or
a conduit through which the treated water 101 treated in
the biological reaction tank 10 flows out of the biological
10 reaction tank 10.
[0012] The air diffuser plate 11 is disposed at a bottom
portion of the biological reaction tank 10 and supplies air
to the treatment target water in the biological reaction
tank 10. Note that, here, the case where air is supplied
15 is given as an example, but any gas containing oxygen like
air may be supplied.
[0013] The blower 12 is connected to the air diffuser
plate 11 via a pipe, and blows air to the air diffuser
plate 11. The air volume regulation unit 13 regulates the
20 volume of air that flows from the blower 12 to the air
diffuser plate 11. In one example, the air volume
regulation unit 13 is an air volume regulation valve
installed on the pipe connecting the blower 12 and the air
diffuser plate 11. In this case, the degree of opening of
25 the air volume regulation valve is adjusted to adjust the
volume of aeration provided to the air diffuser plate 11.
The air volume regulation unit 13 regulates the air volume
in accordance with a target value of the volume of
aeration, the target value being received from the aeration
30 volume control device 30.
[0014] The treated water ammonia concentration sensor 14
measures the ammonia concentration of the treated water 101
in the biological reaction tank 10. The treated water 101
9
is obtained when the treatment target water in the
biological reaction tank 10 is subjected to biological
treatment. In one example, when installed near the outflow
unit 103 or in the outflow unit 103 of the biological
5 reaction tank 10, the treated water ammonia concentration
sensor 14 can appropriately measure the ammonia
concentration of the treated water 101 treated with the
activated sludge in the biological reaction tank 10. The
treated water ammonia concentration sensor 14 is connected
10 to a target aeration volume calculation unit 33, which will
be described later, of the aeration volume control device
30 via a signal line, and transmits measurement data of the
measured ammonia concentration of the treated water 101 to
the target aeration volume calculation unit 33.
15 [0015] The conductivity sensor 15 measures conductivity,
that is, electrical conductivity of the treatment target
water. In one example, the conductivity sensor 15 is
installed in the inflow unit 102. Methods of measuring the
conductivity include an electrode method, an
20 electromagnetic induction method, and the like. The
conductivity sensor 15 is connected to a first estimation
unit 32 and a database update unit 34, which will be
described later, of the aeration volume control device 30
via a signal line and transmits a conductivity value, which
25 is measurement data of the measured conductivity of the
treatment target water, to the first estimation unit 32 and
the database update unit 34.
[0016] The aeration volume control device 30 controls
the volume of aeration, which is an amount of air supplied
30 to the biological reaction tank 10, on the basis of a
difference between an ammonia concentration value of the
treatment target water and an ammonia concentration value
of the treated water 101. The aeration volume control
10
device 30 includes a database 31, the first estimation unit
32, the target aeration volume calculation unit 33, and the
database update unit 34.
[0017] The database 31 stores conductivity-concentration
5 correlation information that is information indicating a
correlation between the conductivity of the treatment
target water and the ammonia concentration of the treatment
target water. The conductivity-concentration correlation
information is the information indicating the correlation
10 between the conductivity of the treatment target water and
the ammonia concentration thereof, but may be information
indicating a correlation between an amount of change in the
conductivity of the treatment target water and the ammonia
concentration thereof. Yet alternatively, the
15 conductivity-concentration correlation information may be
information indicating a correlation between the
conductivity of the treatment target water and also other
plant data related to the treatment target water, and the
ammonia concentration of the treatment target water. The
20 other plant data related to the treatment target water
includes a result of measurement of the concentration of
other ions such as chloride ions contained in the treatment
target water, a flow rate of the treatment target water,
measurement date and time, and the like. The database 31
25 is connected to the database update unit 34 and the first
estimation unit 32. The database 31 corresponds to a
conductivity-concentration correlation information storage
unit.
[0018] The first estimation unit 32 estimates, on the
30 basis of at least the conductivity value of the treatment
target water measured by the conductivity sensor 15, a
first estimated ammonia concentration value as an estimated
value of the ammonia concentration of the treatment target
11
water from the conductivity-concentration correlation
information in the database 31, and outputs the first
estimated ammonia concentration value of the treatment
target water to the target aeration volume calculation unit
5 33. The result of measurement of the conductivity used for
the estimation is not particularly limited and is an
instantaneous value, an average value over a predetermined
period, an amount of change, or the like. In a case where
the conductivity-concentration correlation information in
10 the database 31 is the information indicating the
correlation between the result of measurement of the
conductivity, the other plant data related to the treatment
target water, and the ammonia concentration of the
treatment target water, the first estimation unit 32 may
15 estimate the first estimated ammonia concentration value of
the treatment target water from the conductivityconcentration correlation information on the basis of the
result of measurement of the conductivity and the other
plant data related to the treatment target water. The
20 first estimation unit 32 is connected to the conductivity
sensor 15, the database 31, the target aeration volume
calculation unit 33, and the database update unit 34.
[0019] The target aeration volume calculation unit 33
calculates the target value of the volume of aeration from
25 the blower 12 to the biological reaction tank 10 at given
intervals, and transmits the target value of the volume of
aeration to the air volume regulation unit 13 via a signal
line. Specifically, the target aeration volume calculation
unit 33 calculates the target value of the volume of
30 aeration to the biological reaction tank 10 on the basis of
the first estimated ammonia concentration value of the
treatment target water transmitted from the first
estimation unit 32 and the ammonia concentration value of
12
the treated water 101 in the biological reaction tank 10
measured by the treated water ammonia concentration sensor
14. The target value of the volume of aeration is
desirably calculated at the intervals of one second or
5 longer and about five minutes or shorter, and can be set at
any value depending on the characteristics of the station.
The air volume regulation unit 13 regulates the volume of
aeration such that the volume of aeration provided to the
air diffuser plate 11 is equal to the target value of the
10 volume of aeration calculated by the target aeration volume
calculation unit 33. Note that depending on the scale of
the biological reaction tank 10 or the characteristics of
the station, the numbers of the air diffuser plates 11, the
air volume regulation units 13, and the target aeration
15 volume calculation units 33 can be changed to any values.
[0020] The database update unit 34 updates the
conductivity-concentration correlation information stored
in the database 31. As described above, it is known that
there is a correlation between the conductivity and the
20 ammonia concentration in the treatment target water. It is
also known that this correlation is affected by the
concentration of coexisting substances such as other ions
in the treatment target water. Therefore, in the first
embodiment, the database update unit 34 updates the
25 conductivity-concentration correlation information when
determining that the concentration of the coexisting
substance in the treatment target water has changed and the
conductivity-concentration correlation information needs to
be updated. The database update unit 34 corresponds to a
30 conductivity-concentration correlation information update
unit.
[0021] FIG. 2 is a block diagram schematically
illustrating an example of a configuration of the database
13
update unit of the aeration volume control device according
to the first embodiment. The database update unit 34
includes a reception unit 341 and an update processing unit
342.
5 [0022] The reception unit 341 externally receives a
second estimated ammonia concentration value that is a
value of the ammonia concentration of the treatment target
water. The second estimated ammonia concentration value is
the value of the ammonia concentration of the treatment
10 target water. The second estimated ammonia concentration
value is measured or estimated by a method different from
the method of estimating the first estimated ammonia
concentration value from the measured value of the
conductivity of the treatment target water. At this time,
15 the reception unit 341 may simultaneously receive the date
and time of the first estimated ammonia concentration value
of the treatment target water that has been received
externally. In one example, the reception unit 341 may
receive, as the second estimated ammonia concentration
20 value, a result of measurement of the ammonia concentration
of the treatment target water collected by an operator, the
measurement being performed by an analyzer such as an ion
chromatography analyzer. In another example, the reception
unit 341 may receive, as the second estimated ammonia
25 concentration value, an estimated value obtained by a
device that estimates the ammonia concentration of the
treatment target water. An example of the device that
estimates the ammonia concentration of the treatment target
water will be described in a second embodiment. The
30 reception unit 341 may also receive a data set in which the
second estimated ammonia concentration values for a
plurality of dates and times are input. The reception unit
341 transmits the received second estimated ammonia
14
concentration value of the treatment target water to the
update processing unit 342.
[0023] The update processing unit 342 is connected to
the first estimation unit 32, the conductivity sensor 15,
5 and the database 31. On the basis of a difference between
the first estimated ammonia concentration value estimated
by the first estimation unit 32 and the second estimated
ammonia concentration value received by the reception unit
341, the update processing unit 342 determines whether or
10 not the conductivity-concentration correlation information
in the database 31 needs to be updated. In one example,
the update processing unit 342 determines that the
conductivity-concentration correlation information does not
need to be updated in a case where the difference between
15 the first estimated ammonia concentration value estimated
by the first estimation unit 32 and the second estimated
ammonia concentration value received by the reception unit
341 for the same time as the first estimated ammonia
concentration value is less than a preset determination
20 value. On the other hand, the update processing unit 342
determines that the conductivity-concentration correlation
information needs to be updated in a case where the
difference between the two is larger than the determination
value. In a case where the difference between the two is
25 equal to the determination value, the update processing
unit 342 may determine that the conductivity-concentration
correlation information needs or does not need to be
updated.
[0024] The determination value is, for example, a
30 threshold by which the first estimated ammonia
concentration value and the second estimated ammonia
concentration value can be determined to match within a
margin of error. The determination on whether or not the
15
conductivity-concentration correlation information needs to
be updated may be made on the basis of a preset ratio or
the like. The same time does not need to be the same to
the second, and need only be a period of time within a
5 predetermined range. The predetermined period of time can
be, for example, within one hour. However, in terms of
being able to increase the accuracy of the determination, a
short period of time is preferred as the predetermined
period of time.
10 [0025] If determining that the conductivityconcentration correlation information needs to be updated,
the update processing unit 342 newly constructs the
conductivity-concentration correlation information of the
treatment target water on the basis of the second estimated
15 ammonia concentration value of the treatment target water
received by the reception unit 341 and the conductivity
value measured by the conductivity sensor 15, and updates
the conductivity-concentration correlation information in
the database 31. At this time, the update processing unit
20 342 newly constructs the conductivity-concentration
correlation information of the treatment target water using
a plurality of pieces of data including a set that includes
the second estimated ammonia concentration value and the
conductivity of the treatment target water for the same
25 time.
[0026] Therefore, in the first embodiment, when a change
in the concentration of the coexisting substance affecting
the conductivity has caused a change in the correlation
between the conductivity and the ammonia concentration of
30 the treatment target water from the conductivityconcentration correlation information stored in the
database 31, the database update unit 34 updates the
conductivity-concentration correlation information at an
16
appropriate timing. Accordingly, the first estimation unit
32 estimates the first estimated ammonia concentration
value of the treatment target water with reference to the
conductivity-concentration correlation information updated
5 at the appropriate timing, and the target aeration volume
calculation unit 33 calculates the target value of the
volume of aeration using the first estimated ammonia
concentration value. As a result, an appropriate volume of
air can be supplied to the biological reaction tank 10 in
10 response to the change in load of the inflow ammonia.
[0027] Next, an aeration volume control method and a
method of updating the conductivity-concentration
correlation information in the aeration volume control
device 30 will be described one by one. Note that the
15 updating the conductivity-concentration correlation
information is performed when the aeration volume control
method is executed, and is a part of the aeration volume
control method. That is, the aeration volume control
method includes the method of updating the conductivity20 concentration correlation information.
[0028] FIG. 3 is a flowchart illustrating an example of
a processing procedure of the aeration volume control
method according to the first embodiment. When aeration
volume control is started, at certain time “t”, the
25 conductivity sensor 15 measures the conductivity of the
treatment target water (step S11). The conductivity sensor
15 outputs a result of measurement to the first estimation
unit 32.
[0029] Next, the first estimation unit 32 estimates, on
30 the basis of the conductivity value measured, the first
estimated ammonia concentration value of the treatment
target water from the conductivity-concentration
correlation information in the database 31 (step S12). The
17
first estimation unit 32 outputs the first estimated
ammonia concentration value of the treatment target water
to the target aeration volume calculation unit 33.
[0030] Also at time “t”, concurrently with steps S11 and
5 S12, the treated water ammonia concentration sensor 14
measures the ammonia concentration of the treated water 101
(step S13). The treated water ammonia concentration sensor
14 outputs a result of measurement to the target aeration
volume calculation unit 33.
10 [0031] After step S12 and step S13, the target aeration
volume calculation unit 33 calculates the target value of
the volume of aeration provided to the biological reaction
tank 10 on the basis of the first estimated ammonia
concentration value of the treatment target water that is
15 acquired from the first estimation unit 32 and the ammonia
concentration value of the treated water 101 in the
biological reaction tank 10 that is acquired from the
treated water ammonia concentration sensor 14 (step S14).
In one example, the target aeration volume calculation unit
20 33 calculates the target value of the volume of aeration
using a difference between the first estimated ammonia
concentration value and the ammonia concentration value of
the treated water 101 as an index. The target aeration
volume calculation unit 33 outputs the calculated target
25 value of the volume of aeration to the air volume
regulation unit 13.
[0032] After that, the air volume regulation unit 13
regulates the air volume such that the target value of the
volume of aeration is achieved, and supplies air to the
30 biological reaction tank 10 (step S15).
[0033] The above processing from step S11 to step S15 is
repeated at a certain time interval Δt1. Here, Δt1 is
desirably in a range of one second to about one hour.
18
However, a time longer than the time required to complete
the processing from step S11 to step S15 is set as Δt1.
[0034] FIG. 4 is a flowchart illustrating an example of
a procedure of the method of updating the conductivity5 concentration correlation information in the aeration
volume control device according to the first embodiment.
First, at certain time “t”, the reception unit 341 of the
database update unit 34 externally receives a data set of
the second estimated ammonia concentration value of the
10 treatment target water for one or more dates and times
(step S31). The second estimated ammonia concentration
value of the treatment target water is, as described above,
the value of the ammonia concentration of the treatment
target water that is measured or estimated by a method
15 different from the method of estimating the first estimated
ammonia concentration value from the conductivity of the
treatment target water. In one example, the second
estimated ammonia concentration value is the value obtained
by analyzing the ammonia concentration of the collected
20 treatment target water by the analyzer, or the value
estimated by the device that estimates the ammonia
concentration of the treatment target water. The reception
unit 341 outputs the received data set to the update
processing unit 342. The processing in step S31
25 corresponds to a second estimated ammonia concentration
value receiving step.
[0035] Also at time “t”, concurrently with step S31, the
conductivity sensor 15 measures the conductivity of the
treatment target water (step S32). The conductivity sensor
30 15 outputs a result of measurement to the first estimation
unit 32. The processing in step S32 corresponds to a
conductivity acquisition step. Moreover, the first
estimation unit 32 estimates, on the basis of the
19
conductivity value measured, the first estimated ammonia
concentration value of the treatment target water from the
conductivity-concentration correlation information in the
database 31 (step S33). The first estimation unit 32
5 outputs the first estimated ammonia concentration value as
a result of estimation to the update processing unit 342.
The processing in step S33 corresponds to a first estimated
ammonia concentration value estimating step.
[0036] After step S31 and step S33, on the basis of the
10 first estimated ammonia concentration value estimated by
the first estimation unit 32 and the second estimated
ammonia concentration value received by the reception unit
341, the update processing unit 342 determines whether the
conductivity-concentration correlation information in the
15 database 31 needs to be updated (step S34). In one
example, the update processing unit 342 determines whether
the difference between the first estimated ammonia
concentration value and the second estimated ammonia
concentration value, which have been estimated for the same
20 time, is less than the preset determination value. If the
difference between the two is less than the determination
value, the update processing unit 342 determines that the
conductivity-concentration correlation information does not
need to be updated. If the difference between the two is
25 larger than the determination value, the update processing
unit 342 determines that the conductivity-concentration
correlation information needs to be updated. The
processing in step S34 corresponds to a determination step.
[0037] If the update of the conductivity-concentration
30 correlation information is determined to be unnecessary (if
No in step S34), the processing returns to step S31 and
step S32. If determining that the conductivityconcentration correlation information needs to be updated
20
(if Yes in step S34), the update processing unit 342 newly
constructs the correlation between the conductivity and the
ammonia concentration of the treatment target water on the
basis of the second estimated ammonia concentration value
5 of the treatment target water received by the reception
unit 341 and the conductivity value measured by the
conductivity sensor 15, and updates the conductivityconcentration correlation information in the database 31
(step S35). Note that in a case where the data set
10 received in step S31 is the second estimated ammonia
concentration value of the treatment target water for one
date and time, the conductivity-concentration correlation
information is newly constructed using the second estimated
ammonia concentration value of the treatment target water
15 and the conductivity value for the same time of day in the
past. The processing in step S35 corresponds to an update
processing step.
[0038] The above processing from step S31 to step S35 is
repeated at a certain time interval Δt2.
20 [0039] As described above, in the first embodiment, when
a change in the coexisting substance affecting the
conductivity of the treatment target water has caused a
change in the correlation between the conductivity and the
ammonia concentration of the treatment target water from
25 the conductivity-concentration correlation information
stored in the database 31, the database update unit 34
determines whether the conductivity-concentration
correlation information needs to be updated on the basis of
the difference between the first estimated ammonia
30 concentration value and the second estimated ammonia
concentration value. In the case where the conductivityconcentration correlation information needs to be updated,
the database update unit 34 updates the conductivity-
21
concentration correlation information of the treatment
target water stored in the database 31 by using the second
estimated ammonia concentration value and the conductivity
value of the treatment target water. That is, the
5 conductivity-concentration correlation information is
updated at the appropriate timing. As a result, even when
the coexisting substance changes in the treatment target
water, the appropriate volume of air can be supplied to the
biological reaction tank 10 in response to the change in
10 load of ammonia in the inflow treatment target water.
[0040] Second Embodiment.
The aeration volume control device 30 according to a
second embodiment has the same configuration as that of
FIG. 1 of the first embodiment, but has a different
15 configuration of the database update unit 34. FIG. 5 is a
block diagram illustrating an example of the configuration
of the database update unit of the aeration volume control
device according to the second embodiment. Note that
components identical to those in the first embodiment are
20 denoted by the same reference numerals as those assigned to
such components in the first embodiment, so that the
description thereof will be omitted and differences from
the first embodiment will be described. The database
update unit 34 of the aeration volume control device 30
25 according to the second embodiment includes a second
estimation unit 343, the reception unit 341, and the update
processing unit 342.
[0041] The second estimation unit 343 estimates the
second estimated ammonia concentration value of the
30 treatment target water on the basis of plant data including
the inflow volume of the treatment target water into the
biological reaction tank 10, the volume of aeration by the
blower 12, and the ammonia concentration value of the
22
treated water 101 in the biological reaction tank 10. The
second estimation unit 343 outputs the second estimated
ammonia concentration value of the treatment target water
to the reception unit 341. In one example, the second
5 estimation unit 343 is equipped with a trained model using
an estimation algorithm that estimates an estimated value
of the ammonia concentration of the treatment target water.
That is, the second estimation unit 343 uses a result
obtained by inputting the plant data into the trained model
10 as the second estimated ammonia concentration value of the
treatment target water, the plant data including the inflow
volume of the treatment target water into the biological
reaction tank 10, the volume of aeration by the blower 12,
and the ammonia concentration value of the treated water
15 101 in the biological reaction tank 10. It is sufficient
that the trained model adopts a method capable of
estimating the ammonia concentration of the treatment
target water on the basis of the plant data that includes
the inflow volume of the treatment target water into the
20 biological reaction tank 10, the volume of aeration by the
blower 12, and the ammonia concentration value of the
treated water 101 in the biological reaction tank 10. For
the estimation algorithm used by the second estimation unit
343, an activated sludge model (ASM) that models a
25 biological reaction, a linear regression model, a nonlinear
regression model, machine learning, reinforcement learning,
deep reinforcement learning, deep learning, random forest,
a neural network, other prediction methods using artificial
intelligence, or the like can be used.
30 [0042] In a case where an ammonia load represented by a
product of the ammonia concentration of the treatment
target water and the inflow volume is high, the volume of
aeration that needs to be provided to the biological
23
reaction tank 10 for treating the ammonia load increases,
and when the volume of aeration is insufficient, the
ammonia concentration value of the treated water 101
detected by the treated water ammonia concentration sensor
5 14 increases. Conversely, in a case where the ammonia load
of the treatment target water is low, the volume of
aeration that needs to be provided to the biological
reaction tank 10 for treating the ammonia load decreases,
and when the volume of aeration is sufficient, the ammonia
10 concentration value of the treated water 101 detected by
the treated water ammonia concentration sensor 14
decreases. Thus, there is a correlation between the
ammonia load and the volume of aeration and ammonia
concentration value of the treated water 101, more
15 specifically, between the ammonia concentration of the
treatment target water and the inflow volume of the
treatment target water, the volume of aeration, and the
ammonia concentration value of the treated water 101. That
is, the correlation between the ammonia concentration of
20 the treatment target water and the plant data can be
expressed using the trained model. The plant data includes
the inflow volume of the treatment target water into the
biological reaction tank 10, the volume of aeration by the
blower 12, and the ammonia concentration value of the
25 treated water 101 in the biological reaction tank 10. The
trained model is generated using the estimation algorithm
described above. The second estimation unit 343 can then
use the trained model to estimate the ammonia concentration
of the treatment target water on the basis of the plant
30 data. The second estimation unit 343 is an example of the
device that estimates the ammonia concentration of the
treatment target water in the first embodiment.
[0043] An example will be described in which the trained
24
model used by the second estimation unit 343 is generated
by machine learning. FIG. 6 is a diagram illustrating an
example of a configuration of a learning device that
generates the trained model of the aeration volume control
5 device according to the second embodiment. A learning
device 50 includes a data acquisition unit 51, a model
generation unit 52, and a trained model storage unit 53.
[0044] The data acquisition unit 51 acquires, as
learning data, the plant data and the ammonia concentration
10 value of the treatment target water, the plant data
including the inflow volume of the treatment target water
into the biological reaction tank 10, the volume of
aeration by the blower 12, and the ammonia concentration
value of the treated water 101 in the biological reaction
15 tank 10. The ammonia concentration value of the treatment
target water can be obtained by, for example, measuring the
ammonia concentration of the treatment target water, which
is collected by an operator, using the analyzer such as the
ion chromatography analyzer. In the biological reaction
20 tank 10, the treatment target water flowing thereinto
through the inflow unit 102 gradually flows toward the
outflow unit 103. Therefore, when the ammonia
concentration of the treatment target water in the inflow
unit 102 changes, there is a time delay corresponding to
25 the downflow time until the ammonia concentration value of
the treated water 101 changes after change of the volume of
aeration required in the biological reaction tank 10.
Therefore, as data of the volume of aeration by the blower
12 used for estimating the ammonia concentration of the
30 treatment target water at time T and the ammonia
concentration value of the treated water 101 in the
biological reaction tank 10, it is preferable to use data
at time T+ΔT that is later than time T by the downflow time
25
ΔT from the inflow unit 102 to each measurement point.
Assuming that ΔT1 is a downflow time for the treatment
target water to flow from the inflow unit 102 to a position
where the aeration treatment is performed, and ΔT2 is a
5 downflow time for the treated water 101 to flow from the
inflow unit 102 to the treated water ammonia concentration
sensor 14, it is desirable to use, as the plant data, plant
data including the inflow volume of the treatment target
water to the biological reaction tank 10 at time T, the
10 volume of aeration by the blower 12 at time T+ΔT1, and the
ammonia concentration value of the treated water 101 in the
biological reaction tank 10 at time T+ΔT2.
[0045] The model generation unit 52 learns an estimated
value of the ammonia concentration of the treatment target
15 water on the basis of the learning data that is created on
the basis of a combination of the plant data output from
the data acquisition unit 51 and the ammonia concentration
value of the treatment target water. That is, the model
generation unit 52 generates the trained model that infers
20 the optimum estimated value of the ammonia concentration of
the treatment target water from the plant data of the
aeration volume control system 1 and the ammonia
concentration value of the treatment target water. Here,
the learning data is data in which the plant data and the
25 ammonia concentration value of the treatment target water
are associated with each other.
[0046] Note that the learning device 50 is used to learn
the estimated value of the ammonia concentration of the
treatment target water in the aeration volume control
30 system 1 and, for example, may be connected to the aeration
volume control system 1 via a network and be a device
separate from the aeration volume control system 1.
Alternatively, the learning device 50 may be built in the
26
aeration volume control system 1, particularly in the
aeration volume control device 30, or may exist on a cloud
server.
[0047] As the trained model used by the model generation
5 unit 52, a known algorithm such as supervised learning or
reinforcement learning can be used. As one example, a case
where a neural network is applied will be described.
[0048] The model generation unit 52 learns the estimated
value of the ammonia concentration of the treatment target
10 water by, for example, so-called supervised learning
according to a neural network model. Here, supervised
learning refers to a method that gives data sets of input
and label as result to the learning device 50, learns
features in these pieces of learning data, and infers the
15 result from the input.
[0049] The neural network includes an input layer
including a plurality of neurons, a middle layer including
a plurality of neurons, and an output layer including a
plurality of neurons. The middle layer is also called a
20 hidden layer, and may be one layer or two or more layers.
[0050] FIG. 7 is a diagram schematically illustrating an
example of the neural network used by the model generation
unit in FIG. 6. For example, in the case of a three-layer
neural network as illustrated in FIG. 7, when a plurality
25 of inputs is input to an input layer X1 to an input layer
X3, the input values are multiplied by weights represented
by w11 to w16 and are input to a middle layer Y1 and a
middle layer Y2. The weights w11 to w16 are referred to as
weights w1 when not individually distinguished. Moreover,
30 the results of the middle layer Y1 and the middle layer Y2
are further multiplied by weights represented by w21 to w26
and are output from an output layer Z1 to an output layer
Z3. The weights w21 to w26 are referred to as weights w2
27
when not individually distinguished. The output results of
the output layer Z1 to the output layer Z3 vary depending
on the values of the weights w1 and w2.
[0051] In the first embodiment, the neural network
5 learns the estimated value of the ammonia concentration of
the treatment target water by so-called supervised learning
according to the learning data created on the basis of a
combination of the plant data acquired by the data
acquisition unit 51 and the ammonia concentration value of
10 the treatment target water.
[0052] That is, the neural network performs learning by
adjusting the weights w1 and w2 such that the results
output from the output layers by inputting the plant data
to the input layers become close to the ammonia
15 concentration value of the treatment target water.
[0053] The model generation unit 52 generates and
outputs the trained model by performing learning as
described above.
[0054] The trained model storage unit 53 stores the
20 trained model output from the model generation unit 52.
[0055] Next, processing of learning by the learning
device 50 will be described with reference to FIG. 8. FIG.
8 is a flowchart illustrating an example of a procedure of
the learning processing by the learning device.
25 [0056] The data acquisition unit 51 acquires the plant
data and the ammonia concentration value of the treatment
target water (step S51). Note that although the plant data
and the ammonia concentration value of the treatment target
water are simultaneously acquired, it is sufficient that
30 the plant data and the ammonia concentration value of the
treatment target water can be input in association with
each other, and the plant data and the data of the ammonia
concentration value of the treatment target water may be
28
acquired at different timings.
[0057] Next, the model generation unit 52 learns the
estimated value of the ammonia concentration of the
treatment target water by so-called supervised learning
5 according to the learning data created on the basis of the
combination of the plant data acquired by the data
acquisition unit 51 and the ammonia concentration value of
the treatment target water. The model generation unit 52
generates the trained model using the estimation algorithm
10 indicating the correlation between the plant data and the
ammonia concentration value of the treatment target water.
The plant data includes the inflow volume of the treatment
target water into the biological reaction tank 10, the
volume of aeration by the blower 12, and the ammonia
15 concentration value of the treated water 101 in the
biological reaction tank 10 (step S52).
[0058] Then, the trained model storage unit 53 stores
the trained model generated by the model generation unit 52
(step S53). The processing is thus completed.
20 [0059] Next, details of the second estimation unit 343
will be described. FIG. 9 is a diagram schematically
illustrating an example of a configuration of the second
estimation unit of the aeration volume control device
according to the second embodiment. As described above,
25 the second estimation unit 343 is an inference device that
outputs a result obtained by inputting the plant data into
the trained model as the second estimated ammonia
concentration value, the plant data including the inflow
volume of the treatment target water into the biological
30 reaction tank 10, the volume of aeration by the blower 12,
and the ammonia concentration value of the treated water
101 in the biological reaction tank 10. The second
estimation unit 343 includes a data acquisition unit 3431
29
and an inference unit 3432.
[0060] The data acquisition unit 3431 acquires the plant
data including the inflow volume of the treatment target
water into the biological reaction tank 10, the volume of
5 aeration by the blower 12, and the ammonia concentration
value of the treated water 101 in the biological reaction
tank 10. As mentioned in describing the data acquisition
unit 51 of the learning device 50, in the biological
reaction tank 10, the treatment target water flowing
10 thereinto through the inflow unit 102 gradually flows
toward the outflow unit 103, so that when the ammonia
concentration of the treatment target water in the inflow
unit 102 changes, there is a time delay corresponding to
the downflow time until the ammonia concentration value of
15 the treated water 101 changes after changing of the volume
of aeration required in the biological reaction tank 10.
Accordingly, when ΔT1 is the downflow time for the
treatment target water to flow from the inflow unit 102 to
the position where the aeration treatment is performed, and
20 ΔT2 is the downflow time for the treated water 101 to flow
from the inflow unit 102 to the treated water ammonia
concentration sensor 14, it is desirable that the data
acquisition unit 3431 uses, as the plant data, the plant
data including the inflow volume of the treatment target
25 water to the biological reaction tank 10 at time T, the
volume of aeration by the blower 12 at time T+ΔT1, and the
ammonia concentration value of the treated water 101 in the
biological reaction tank 10 at time T+ΔT2.
[0061] The inference unit 3432 infers the second
30 estimated ammonia concentration value of the treatment
target water obtained using the trained model. That is,
when the plant data acquired by the data acquisition unit
3431 is input to the trained model, the second estimated
30
ammonia concentration value of the treatment target water
inferred from the plant data can be output.
[0062] Note that the second embodiment has described the
case of using the trained model learned from the plant data
5 of the aeration volume control system 1 and the ammonia
concentration value of the treatment target water.
However, a trained model learned from plant data of another
aeration volume control system 1 and an ammonia
concentration value of treatment target water may be
10 acquired from the outside, and the second estimated ammonia
concentration value of the treatment target water may be
output on the basis of the trained model.
[0063] Next, with reference to FIG. 10, processing for
obtaining the second estimated ammonia concentration value
15 of the treatment target water using the second estimation
unit 343 will be described. FIG. 10 is a flowchart
illustrating an example of a procedure of estimation
processing by the second estimation unit of the aeration
volume control device according to the second embodiment.
20 [0064] First, the data acquisition unit 3431 acquires
the plant data (step S71). The plant data is the data
including the inflow volume of the treatment target water
into the biological reaction tank 10, the volume of
aeration by the blower 12, and the ammonia concentration
25 value of the treated water 101 in the biological reaction
tank 10.
[0065] Next, the inference unit 3432 inputs the plant
data to the trained model stored in the trained model
storage unit 53, and obtains the second estimated ammonia
30 concentration value of the treatment target water (step
S72).
[0066] After that, the inference unit 3432 outputs the
second estimated ammonia concentration value of the
31
treatment target water obtained by the trained model to the
aeration volume control system 1, specifically, the
reception unit 341 of the database update unit 34 of the
aeration volume control device 30 (step S73).
5 [0067] Then, the reception unit 341 outputs the second
estimated ammonia concentration value of the treatment
target water to the update processing unit 342, and, as
described later, the update processing unit 342 uses the
second estimated ammonia concentration value of the
10 treatment target water and the first estimated ammonia
concentration value of the treatment target water output
from the first estimation unit 32 to determine a deviation
of the first estimated ammonia concentration value of the
treatment target water from the actual ammonia
15 concentration value (step S74). The second estimated
ammonia concentration value inferred by the inference unit
3432 is calculated while excluding the influence of a
change in the concentration of the coexisting substance,
and is thus closer to the actual ammonia concentration
20 value of the treatment target water than the first
estimated ammonia concentration value is. This makes it
possible to isolate whether the change in the conductivity
of the treatment target water is due to the ammonia
concentration or the influence of the coexisting substance
25 of the treatment target water. The processing is thus
completed.
[0068] Note that the second embodiment has described the
case where supervised learning is applied as the learning
algorithm used by the model generation unit 52, but the
30 present disclosure is not limited thereto. Besides
supervised learning, it is also possible to apply
reinforcement learning, semi-supervised learning, or the
like as the learning algorithm.
32
[0069] In addition, the model generation unit 52 may
learn the estimated value of the ammonia concentration of
the treatment target water according to learning data
created for a plurality of the aeration volume control
5 systems 1. Note that the model generation unit 52 may
acquire the learning data from a plurality of the aeration
volume control systems 1 used in the same area, or may use
the learning data collected from a plurality of the
aeration volume control systems 1 operating independently
10 in different areas to learn the estimated value of the
ammonia concentration of the treatment target water.
Moreover, the aeration volume control system 1 from which
the learning data is collected can be added or removed half
way through the learning. Furthermore, the learning device
15 50 that has learned the estimated value of the ammonia
concentration of the treatment target water for a certain
one of the aeration volume control system 1 may be applied
to another one aeration volume control system 1, and, for
the another one aeration volume control system 1, the
20 estimated value of the ammonia concentration of the
treatment target water may be relearned and updated.
[0070] Also, as the learning algorithm used in the model
generation unit 52, deep learning that learns extraction of
a feature value itself can be used, or machine learning may
25 be executed according to another known method such as
genetic programming, functional logic programming, or
support vector machine.
[0071] As described above, in order for the second
estimation unit 343 to estimate the second estimated
30 ammonia concentration value of the treatment target water,
it is necessary to use the plant data including the volume
of aeration by the blower 12 at time T+ΔT1, which is later
than time T by the downflow time ΔT1 required for the
33
treatment target water to flow from the inflow unit 102 to
the position where the aeration treatment is performed, and
the ammonia concentration value of the treated water 101 in
the biological reaction tank 10 at time T+ΔT2 which is
5 later than time T by the downflow time ΔT2 required for the
treated water 101 to flow from the inflow unit 102 to the
treated water ammonia concentration sensor 14. Therefore,
the ammonia concentration of the treatment target water
cannot be estimated in real time. However, it is
10 considered that the ammonia concentration of the treated
water 101 when a certain inflow volume of the treatment
target water is subjected to biological treatment with a
certain volume of aeration is mainly affected by the
ammonia concentration of the treatment target water. That
15 is, the ammonia concentration of the treated water 101 is
less likely to be affected by a change in the concentration
of the coexisting substance such as other ions, the effect
by the change in the concentration of the coexisting
substance having been the problem in estimating the ammonia
20 concentration using the conductivity of the treatment
target water. As a result, it can be considered that the
estimated value of the ammonia concentration of the
treatment target water that is estimated by the second
estimation unit 343 reflects the ammonia concentration
25 under a situation where the coexisting substance exists in
the treatment target water. Therefore, the estimated value
of the ammonia concentration of the treatment target water
estimated by the second estimation unit 343 can be used as
a reference when determining the deviation, from the actual
30 ammonia concentration of the treatment target water, of the
first estimated ammonia concentration value of the
treatment target water, the first estimated ammonia
concentration value being estimated by the first
34
estimation unit 32 from the conductivity-concentration
correlation information on the basis of the conductivity of
the treatment target water measured by the conductivity
sensor 15.
5 [0072] Returning to FIG. 5, the reception unit 341
receives the second estimated ammonia concentration value
of the treatment target water estimated by the second
estimation unit 343. The reception unit 341 may
simultaneously receive also the date and time when the
10 second estimated ammonia concentration value of the
treatment target water is estimated.
[0073] On the basis of a difference between the first
estimated ammonia concentration value estimated by the
first estimation unit 32 and the second estimated ammonia
15 concentration value estimated by the second estimation unit
343, the update processing unit 342 determines whether or
not the conductivity-concentration correlation information
in the database 31 needs to be updated. As described
above, the second estimated ammonia concentration value
20 estimated by the second estimation unit 343 can be
considered as the actual ammonia concentration value of the
treatment target water. Accordingly, the update processing
unit 342 determines whether or not the conductivityconcentration correlation information needs to be updated
25 on the basis of a degree of deviation, from the second
estimated ammonia concentration value, of the first
estimated ammonia concentration value estimated by the
first estimation unit 32.
[0074] In one example, if the difference between the
30 first estimated ammonia concentration value estimated by
the first estimation unit 32 and the second estimated
ammonia concentration value estimated by the second
estimation unit 343 for the same time as the first
35
estimated ammonia concentration value is less than a preset
determination value, the update processing unit 342
determines that the conductivity-concentration correlation
information does not need to be updated. If the difference
5 between the two is larger than the determination value, the
update processing unit 342 determines that the
conductivity-concentration correlation information needs to
be updated. The determination on whether or not the
conductivity-concentration correlation information needs to
10 be updated may be based on a preset ratio or the like. In
a case where the difference between the two is equal, the
update processing unit 342 may determine that the
conductivity-concentration correlation information needs or
does not need to be updated.
15 [0075] If determining that the database 31 needs to be
updated, the update processing unit 342 newly constructs
the correlation between the conductivity of the treatment
target water and the ammonia concentration of the treatment
target water on the basis of the second estimated ammonia
20 concentration value of the treatment target water estimated
by the second estimation unit 343 and the conductivity
value measured by the conductivity sensor 15, and updates
the conductivity-concentration correlation information in
the database 31. At this time, it is desirable to use a
25 plurality of sets of data of the second estimated ammonia
concentration value and the conductivity value acquired at
a plurality of different times.
[0076] Therefore, in the second embodiment, even when a
change in the coexisting substance affecting the
30 conductivity has caused a change in the correlation between
the conductivity and the ammonia concentration of the
treatment target water from the conductivity-concentration
correlation information stored in the database 31, it is
36
possible to estimate the second estimated ammonia
concentration value of the treatment target water in the
presence of the coexisting substance on the basis of the
plant data. The second estimated ammonia concentration
5 value of the treatment target water is less affected by the
concentration of the coexisting substance, and is thus
closer to the actual ammonia concentration of the treatment
target water than the first estimated ammonia concentration
value obtained from the conductivity of the treatment
10 target water. By using the second estimated ammonia
concentration value of the treatment target water, the
database update unit 34 can update the conductivityconcentration correlation information of the treatment
target water stored in the database 31 at an appropriate
15 timing. As a result, in response to a change in load of
the inflow ammonia, an appropriate volume of air can be
supplied to the biological reaction tank 10 while
eliminating the influence of the coexisting substance.
[0077] Next, a method of updating the conductivity20 concentration correlation information in the database
update unit 34 of the aeration volume control device 30 of
the second embodiment will be described. FIG. 11 is a
flowchart illustrating an example of a procedure of the
method of updating the conductivity-concentration
25 correlation information in the aeration volume control
device according to the second embodiment.
[0078] First, at certain time “t”, the second estimation
unit 343 acquires the plant data including the inflow
volume of the treatment target water into the biological
30 reaction tank 10, the volume of aeration by the blower 12,
and the ammonia concentration value of the treated water
101 in the biological reaction tank 10 (step S91). The
processing of step S91 corresponds to a plant data
37
acquisition step.
[0079] Next, the second estimation unit 343 estimates
the second estimated ammonia concentration value of the
treatment target water on the basis of the plant data
5 including the inflow volume of the treatment target water
into the biological reaction tank 10, the volume of
aeration by the blower 12, and the ammonia concentration
value of the treated water 101 in the biological reaction
tank 10 (step S92). Specifically, the second estimation
10 unit 343 determines, as the second estimated ammonia
concentration value of the treatment target water, a value
output by inputting the plant data as input data to the
trained model. The second estimation unit 343 outputs the
second estimated ammonia concentration value of the
15 treatment target water to the reception unit 341. The
processing in step S92 corresponds to a second estimated
ammonia concentration value estimating step.
[0080] The reception unit 341 receives the second
estimated ammonia concentration value of the treatment
20 target water estimated by the second estimation unit 343
(step S93). The reception unit 341 outputs, to the update
processing unit 342, the data set of the received second
estimated ammonia concentration value of the treatment
target water. The processing in step S93 corresponds to
25 the second estimated ammonia concentration value receiving
step.
[0081] Also at time “t”, concurrently with the
processing from step S91 to step S93, the conductivity
sensor 15 measures the conductivity of the treatment target
30 water (step S94). The conductivity sensor 15 outputs a
result of measurement to the first estimation unit 32. The
processing in step S94 corresponds to the conductivity
acquisition step.
38
[0082] Next, the first estimation unit 32 estimates, on
the basis of the conductivity value measured, the first
estimated ammonia concentration value of the treatment
target water from the conductivity-concentration
5 correlation information in the database 31 (step S95). The
first estimation unit 32 outputs the first estimated
ammonia concentration value of the treatment target water
as a result of estimation to the update processing unit
342. The processing in step S95 corresponds to the first
10 estimated ammonia concentration value estimating step.
[0083] After step S93 and step S95, on the basis of a
difference between the first estimated ammonia
concentration value of the treatment target water estimated
by the first estimation unit 32 and the second estimated
15 ammonia concentration value estimated by the second
estimation unit 343, the update processing unit 342
determines whether the conductivity-concentration
correlation information in the database 31 needs to be
updated (step S96). In one example, if the difference
20 between the first estimated ammonia concentration value
estimated by the first estimation unit 32 and the second
estimated ammonia concentration value estimated by the
second estimation unit 343 for the same time as the first
estimated ammonia concentration value is less than a preset
25 determination value, the update processing unit 342
determines that the conductivity-concentration correlation
information does not need to be updated. If the difference
between the two is larger than the determination value, the
update processing unit 342 determines that the
30 conductivity-concentration correlation information needs to
be updated. The processing in step S96 corresponds to the
determination step.
[0084] If the update of the conductivity-concentration
39
correlation information is determined to be unnecessary (if
No in step S96), the processing returns to step S91 and
step S94. If determining that the conductivityconcentration correlation information needs to be updated
5 (if Yes in step S96), the database update unit 34 newly
constructs the correlation between conductivity and the
ammonia concentration of the treatment target water on the
basis of the second estimated ammonia concentration value
estimated by the second estimation unit 343 and the
10 conductivity value measured by the conductivity sensor 15,
and updates the conductivity-concentration correlation
information in the database 31 (step S97). This update
processing of the conductivity-concentration correlation
information uses a plurality of sets of data of the second
15 estimated ammonia concentration value estimated by the
second estimation unit 343 and the conductivity value
measured by the conductivity sensor 15. The processing in
step S97 corresponds to the update processing step.
[0085] The above processing from step S91 to step S97 is
20 repeated at a certain time interval Δt3.
[0086] As described above, in the second embodiment,
when a change in the coexisting substance affecting the
conductivity has caused a change in the correlation between
the conductivity and the ammonia concentration of the
25 treatment target water from the conductivity-concentration
correlation information stored in the database 31, the
second estimated ammonia concentration value eliminating
the influence of the coexisting substance of the treatment
target water is estimated on the basis of the plant data
30 including the inflow volume of the treatment target water
into the biological reaction tank 10, the volume of
aeration by the blower 12, and the ammonia concentration
value of the treated water 101 in the biological reaction
40
tank 10. The database update unit 34 can appropriately
determine the update time of the conductivity-concentration
correlation information of the treatment target water by
comparing the first estimated ammonia concentration value
5 with the second estimated ammonia concentration value.
Also, in the case of updating the conductivityconcentration correlation information, the database update
unit 34 updates the conductivity-concentration correlation
information with the correlation between the conductivity
10 and the ammonia concentration of the treatment target
water, the correlation being newly constructed on the basis
of the second estimated ammonia concentration value and the
conductivity value measured by the conductivity sensor 15.
As a result, an appropriate volume of air can be supplied
15 to the biological reaction tank 10 in response to the
change in load of the inflow ammonia.
[0087] Note that the above description has described the
case where the target aeration volume calculation unit 33
outputs the target value of the volume of aeration to the
20 air volume regulation unit 13, and the air volume
regulation unit 13 regulates the degree of opening of the
air volume regulation valve. However, it is sufficient
that the volume of aeration can be regulated in the end,
and thus the air volume of the blower 12 may be regulated
25 instead of the air volume regulation unit 13. That is, the
target aeration volume calculation unit 33 may output the
target value of the volume of aeration to the blower 12,
and the blower 12 may regulate the air volume such that the
target value of the volume of aeration is achieved.
30 [0088] Here, the aeration volume control device 30
illustrated in FIG. 1 may be configured as a single circuit
or device, or each of the database 31, the first estimation
unit 32, the target aeration volume calculation unit 33,
41
and the database update unit 34 may be configured as one
circuit or device. Moreover, each unit may be implemented
by a control circuit including a memory and a processor
executing a program stored in the memory, or may be
5 implemented by dedicated hardware. Here, a case where the
aeration volume control device 30 is implemented by the
control circuit will be described as an example.
[0089] FIG. 12 is a diagram illustrating an example of a
hardware configuration of the control circuit. A control
10 circuit 400 illustrated in FIG. 12 includes an input unit
401, a processor 402, a memory 403, and an output unit 404.
The components of the control circuit 400 are connected to
one another via a bus 411.
[0090] The input unit 401 receives a signal from the
15 outside. The output unit 404 outputs a signal generated in
the control circuit 400 to the outside. The processor 402
is, for example, a central processing unit (CPU), a central
processor, a processing unit, an arithmetic unit, a
microprocessor, a microcomputer, a digital signal processor
20 (DSP), or the like. The processor 402 executes various
processings.
[0091] The memory 403 is, for example, a non-volatile or
volatile semiconductor memory such as a random access
memory (RAM), a read only memory (ROM), a flash memory, an
25 erasable programmable read only memory (EPROM), or an
electrically erasable programmable read only memory (EEPROM
(registered trademark)), a magnetic disk, a flexible disk,
an optical disk, a compact disc, a mini disc, or a digital
versatile disk (DVD). The memory 403 stores a program for
30 operating the aeration volume control device 30, the
conductivity-concentration correlation information, and the
like.
[0092] The processor 402 reads and executes the program
42
stored in the memory 403 via the bus 411, and is
responsible for the processing and control of the entire
aeration volume control device 30. The functions of the
target aeration volume calculation unit 33, the first
5 estimation unit 32, and the database update unit 34 of the
aeration volume control device 30 illustrated in FIG. 1 are
implemented using the processor 402.
[0093] The memory 403 is used as a work area of the
processor 402. The memory 403 also stores programs such as
10 a boot program and an aeration volume control program for
executing the aeration volume control method and the method
of updating the conductivity-concentration correlation
information. When the aeration volume control method
described in the first and second embodiments is executed,
15 the processor 402 loads the aeration volume control program
into the memory 403 and executes various processings.
[0094] Also, in a case where each processing unit
included in the aeration volume control device 30, each
processing unit included in the database update unit 34,
20 each processing unit included in the second estimation unit
343, or each processing unit included in the learning
device 50 is implemented by dedicated hardware, the
dedicated hardware is, for example, a single circuit, a
complex circuit, a programmed processor, a parallel25 programmed processor, an application specific integrated
circuit (ASIC), a field programmable gate array (FPGA), or
a combination of these. In the case where each processing
unit is implemented by the dedicated hardware, the
processing units are connected via a signal line. Data is
30 then communicated among the processing units via the signal
line.
[0095] Furthermore, when the aeration volume control
program described above is executed by a computer, the
43
computer has functions similar to those of the aeration
volume control device 30.
[0096] In addition, although the aeration volume control
program described above is stored in the memory 403 in
5 advance, the present disclosure is not limited thereto.
The aeration volume control program described above may be
written on a recording medium such as a compact disc (CD)-
ROM or a digital versatile disc (DVD)-ROM to be supplied to
a user, and may be installed in the memory 403 by the user.
10 Alternatively, the aeration volume control program
described above may be provided to a user via a network
such as the Internet.
[0097] The configurations illustrated in the above
embodiments each illustrate an example so that another
15 known technique can be combined, the embodiments can be
combined together, or the configurations can be partially
omitted and/or modified without departing from the scope of
the present disclosure.
20 Reference Signs List
[0098] 1 aeration volume control system; 10 biological
reaction tank; 11 air diffuser plate; 12 blower; 13 air
volume regulation unit; 14 treated water ammonia
concentration sensor; 15 conductivity sensor; 30 aeration
25 volume control device; 31 database; 32 first estimation
unit; 33 target aeration volume calculation unit; 34
database update unit; 50 learning device; 51, 3431 data
acquisition unit; 52 model generation unit; 53 trained
model storage unit; 101 treated water; 102 inflow unit;
30 103 outflow unit; 341 reception unit; 342 update
processing unit; 343 second estimation unit; 400 control
circuit; 401 input unit; 402 processor; 403 memory; 404
output unit; 411 bus; 3432 inference unit.

We Claim :
1. An aeration volume control device that controls volume
of aeration as volume of oxygen-containing gas supplied to
a biological reaction tank that performs biological
5 treatment on treatment target water, the aeration volume
control device comprising:
an ammonia concentration sensor to measure ammonia
concentration of treated water that is obtained when the
treatment target water in the biological reaction tank is
10 subjected to the biological treatment;
a conductivity sensor to measure conductivity of the
treatment target water that flows into the biological
reaction tank;
a conductivity-concentration correlation information
15 storage unit to store conductivity-concentration
correlation information that indicates a correlation
between the conductivity of the treatment target water and
ammonia concentration of the treatment target water;
a first estimation unit to estimate a first estimated
20 ammonia concentration value from the conductivityconcentration correlation information on the basis of a
conductivity value measured by the conductivity sensor, the
first estimated ammonia concentration value being an
estimated value of the ammonia concentration of the
25 treatment target water;
a target aeration volume calculation unit to calculate
a target value of the volume of aeration to the biological
reaction tank on the basis of the first estimated ammonia
concentration value of the treated water and the ammonia
30 concentration value of the treated water measured by the
ammonia concentration sensor; and
a conductivity-concentration correlation information
update unit to update the conductivity-concentration
45
correlation information, wherein
the conductivity-concentration correlation information
update unit includes:
a reception unit to receive a second estimated ammonia
5 concentration value that is a value of the ammonia
concentration of the treatment target water, the second
estimated ammonia concentration value being measured or
estimated by a method different from a method of estimating
the first estimated ammonia concentration value; and
10 an update processing unit to update the conductivityconcentration correlation information on the basis of the
second estimated ammonia concentration value received by
the reception unit and the conductivity value measured by
the conductivity sensor.
15
2. The aeration volume control device according to claim
1, wherein
the conductivity-concentration correlation information
update unit further includes a second estimation unit to
20 output the second estimated ammonia concentration value
from plant data by using a trained model that infers the
second estimated ammonia concentration value of the
treatment target water from the plant data, the plant data
including an inflow volume of the treatment target water
25 into the biological reaction tank, the volume of aeration,
and the ammonia concentration value of the treated water in
the biological reaction tank, and
the reception unit receives the second estimated
ammonia concentration value from the second estimation
30 unit.
3. The aeration volume control device according to claim
1 or 2, wherein the update processing unit determines
46
whether the conductivity-concentration correlation
information needs to be updated from the first estimated
ammonia concentration value and the second estimated
ammonia concentration value, and updates the conductivity5 concentration correlation information when the
conductivity-concentration correlation information needs to
be updated.
4. The aeration volume control device according to claim
10 3, wherein, when a difference between the first estimated
ammonia concentration value and the second estimated
ammonia concentration value is larger than a predetermined
determination value, the update processing unit determines
that the conductivity-concentration correlation information
15 needs to be updated, newly constructs a correlation between
the conductivity of the treatment target water and the
ammonia concentration of the treatment target water using
the second estimated ammonia concentration value and the
conductivity value for the same time, and updates the
20 conductivity-concentration correlation information in the
conductivity-concentration correlation information storage
unit.
5. The aeration volume control device according to claim
25 2, wherein, when ΔT1 represents a time during which the
treatment target water flows down to a position where
aeration is performed after flowing into the biological
reaction tank at time T, and ΔT2 represents a time during
which the treatment target water flows down to a position
30 of the ammonia concentration sensor after flowing into the
biological reaction tank at time T,
the inflow volume of the treatment target water into
the biological reaction tank is data at time T, the volume
47
of aeration is data at time T+ΔT1, and the ammonia
concentration value of the treated water in the biological
reaction tank is data at time T+ΔT2.
5 6. The aeration volume control device according to claim
2 or 5, further comprising
a learning device to generate the trained model,
wherein
the learning device includes:
10 a data acquisition unit to acquire learning data
including the plant data and the ammonia concentration
value of the treatment target water, the plant data
including the inflow volume of the treatment target water
into the biological reaction tank, the volume of aeration,
15 and the ammonia concentration value of the treated water in
the biological reaction tank; and
a model generation unit to use the learning data to
generate the trained model for inferring the second
estimated ammonia concentration value from the plant data
20 that includes the inflow volume of the treatment target
water into the biological reaction tank, the volume of
aeration, and the ammonia concentration value of the
treated water in the biological reaction tank.
25 7. An aeration volume control method in an aeration
volume control device that estimates a first estimated
ammonia concentration value that is an estimated value of
ammonia concentration of treatment target water flowing
into a biological reaction tank in which biological
30 treatment is performed on the treatment target water, by
using conductivity-concentration correlation information
indicating a correlation between conductivity and the
ammonia concentration of the treatment target water, and
48
controls volume of aeration that is volume of oxygencontaining gas supplied to the biological reaction tank, by
using the first estimated ammonia concentration value, the
aeration volume control method comprising:
5 a conductivity acquisition step in which the aeration
volume control device acquires a conductivity value of the
treatment target water flowing into the biological reaction
tank;
a second estimated ammonia concentration value
10 receiving step in which the aeration volume control device
externally receives a second estimated ammonia
concentration value that is a value of the ammonia
concentration of the treatment target water, the second
estimated ammonia concentration value being measured or
15 estimated by a method different from a method of estimating
the first estimated ammonia concentration value; and
an update processing step in which the aeration volume
control device updates the conductivity-concentration
correlation information of the treatment target water on
20 the basis of the second estimated ammonia concentration
value and the conductivity value.
8. The aeration volume control method according to claim
7, further comprising:
25 a first estimated ammonia concentration value
estimating step in which the aeration volume control device
estimates, on the basis of the conductivity value acquired,
the first estimated ammonia concentration value of the
treatment target water from the conductivity-concentration
30 correlation information; and
a determination step in which the aeration volume
control device determines whether the conductivityconcentration correlation information needs to be updated
49
from the first estimated ammonia concentration value and
the second estimated ammonia concentration value, wherein
when update of the conductivity-concentration
correlation information is determined to be necessary, the
5 update processing step is executed.