extracted from wipo:
formulas and tables are not copied:
FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
[See section 10, Rule 13]
OPERATION SUPPORT DEVICE AND OPERATION SUPPORT 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 100-8310, JAPAN
THE FOLLOWING SPECIFICATION PARTICULARLY DESCRIBES THE
INVENTION AND THE MANNER IN WHICH IT IS TO BE PERFORMED.
2
DESCRIPTION
5 Field
[0001] The present invention relates to an operation
support device and an operation support method for
supporting water treatment operation.
10 Background
[0002] One of the sewage treatment methods is a membrane
filtration method that removes a substance to be removed
from inflow water by causing the substance to adhere to the
membrane surface of a separation membrane. In the membrane
15 filtration method, the filtration capacity decreases and
the power consumption increases as the separation membrane
is clogged. Therefore, the separation membrane is
regularly cleaned until the separation membrane can no
longer be reused due to repeated cleaning, in which case
20 the separation membrane is replaced.
[0003] Because the membrane filtration method involves
the cleaning cost for separation membranes and the
replacement cost for separation membranes, there is a need
to know the proper cleaning schedule and replacement
25 schedule for separation membranes in order to achieve lowcost
filtration.
[0004] The operation support device described in Patent
Literature 1 calculates the separation membrane filtration
time and cleaning time for reducing the operation cost of
30 the membrane filtration treatment device based on the
operation information on the membrane filtration treatment
device, and predicts the replacement schedule for
separation membranes based on the calculation result.
3
Citation List
Patent Literature
[0005] Patent Literature 1: Japanese Patent Application
Laid-open 5 No. 2009-000580
Summary
Technical Problem
[0006] However, in the technique of Patent Literature 1,
10 a reference value for determining whether to clean a
separation membrane and a reference value for determining
whether to replace a separation membrane need to be input
by the user of the operation support device. Obtaining
these reference values requires specialized knowledge of
15 the membrane filtration method, so it is not possible to
easily calculate the cleaning schedule and replacement
schedule for separation membranes for reducing the
operation cost of separation membranes.
[0007] The present invention has been made in view of
20 the above, and an object thereof is to obtain an operation
support device capable of easily calculating the cleaning
schedule and replacement schedule for separation membranes
for reducing the operation cost of separation membranes.
25 Solution to Problem
[0008] In order to solve the above problem and achieve
an object, an operation support device according to an
aspect of the present invention includes a computer to:
calculate a cost of separation treatment that depends on a
30 deterioration of a separation membrane using a cleaning
efficiency model that expresses a deterioration of the
separation membrane relative to the number of cleaning
events, the separation membrane being configured to remove
4
a removal target from sewage; calculate an operation cost
using the cost of separation treatment, a cleaning cost for
one cleaning event for the separation membrane, and a
replacement cost for one replacement event for the
separation membrane, the operation cost 5 being a sum of the
cost of separation treatment, a cleaning cost for the
separation membrane, and a replacement cost for the
separation membrane in an operation period of the
separation membrane; determine a first reference value and
10 a second reference value based on the operation cost, the
first reference value being used for determining whether to
clean the separation membrane, the second reference value
being used for determining whether to replace the
separation membrane; and calculate a cleaning schedule and
15 a replacement schedule for the separation membrane using
the first reference value and the second reference value
determined.
Advantageous Effects of Invention
20 [0009] The present invention can achieve the effect of
easily calculating the cleaning schedule and replacement
schedule for separation membranes for reducing the
operation cost of separation membranes.
25 Brief Description of Drawings
[0010] FIG. 1 is a functional configuration diagram of
an operation support device according to a first embodiment.
FIG. 2 is a diagram illustrating an example of a
pattern set used in the operation support device according
30 to the first embodiment.
FIG. 3 is a diagram illustrating an example of a timeseries
change of the blockage rate calculated by the
operation support device according to the first embodiment.
5
FIG. 4 is a diagram illustrating the relationship
between calculation periods used by the operation support
device according to the first embodiment to calculate
pattern operation costs and calculation results of pattern
5 operation costs.
FIG. 5 is a diagram illustrating an example of a
presentation screen presented by the operation support
device according to the first embodiment.
FIG. 6 is a flowchart illustrating the procedure for
10 computation processing by a computation unit according to
the first embodiment.
FIG. 7 is a functional configuration diagram of an
operation support device according to a second embodiment.
FIG. 8 is a diagram for explaining the process of
15 changing the operating condition and the process of
recalculating the minimum reference value pattern by the
operation support device according to the second embodiment.
FIG. 9 is a diagram for explaining the replacement
cycle calculated by the operation support device according
20 to the second embodiment without changing the operating
condition.
FIG. 10 is a diagram for explaining the replacement
cycles calculated by the operation support device according
to the second embodiment by changing the operating
25 condition.
FIG. 11 is a functional configuration diagram of an
operation support device according to a third embodiment.
FIG. 12 is a diagram for explaining the process of
changing the operating condition and the process of
30 recalculating the minimum reference value pattern by the
operation support device according to the third embodiment.
FIG. 13 is a diagram for explaining the replacement
cycle calculated by the operation support device according
6
to the third embodiment without changing the operating
condition.
FIG. 14 is a diagram for explaining the replacement
cycles calculated by the operation support device according
to the third embodiment by changing the operating 5 condition.
FIG. 15 is a diagram illustrating a hardware
configuration for implementing the operation support device
according to any of the first to third embodiments.
FIG. 16 is a functional configuration diagram of an
10 operation support device according to a fourth embodiment.
FIG. 17 is a diagram illustrating an example of a
temporary reference value pattern set used in the operation
support device according to the fourth embodiment.
FIG. 18 is a diagram illustrating an example of a
15 time-series change of the blockage rate calculated by the
operation support device according to the fourth embodiment.
FIG. 19 is a diagram illustrating the relationship
between calculation periods used by the operation support
device according to the fourth embodiment to calculate
20 pattern operation costs and calculation results of pattern
operation costs.
FIG. 20 is a flowchart illustrating the procedure for
computation processing by a computation unit according to
the fourth embodiment.
25 FIG. 21 is a diagram illustrating another
configuration example of the operation support device
according to the fourth embodiment.
Description of Embodiments
30 [0011] Hereinafter, embodiments of an operation support
device and an operation support method according to the
present invention will be described in detail with
reference to the drawings. The present invention is not
7
limited to the embodiments.
[0012] First Embodiment.
FIG. 1 is a functional configuration diagram of an
operation support device according to the first embodiment.
The operation support device 100A, 5 which is a water
treatment operation support device, is a computer that
predicts the cleaning schedule and replacement schedule for
separation membranes for reducing the operation cost of
separation membranes used in water treatment such as sewage
10 treatment. A separation membrane is a membrane that
removes a removal target from sewage by filtration. In the
following description, separation membranes may be referred
to as membranes. The cleaning schedule and replacement
schedule for separation membranes may be referred to as the
15 maintenance schedule.
[0013] The operation support device 100A includes a
computation unit 1, a sewage information input unit 2, an
accumulation model input unit 3, an efficiency model input
unit 4, a cost input unit 5, a power consumption model
20 input unit 6, and a result presentation unit 7.
[0014] The sewage information input unit 2, the
accumulation model input unit 3, the efficiency model input
unit 4, the cost input unit 5, and the power consumption
model input unit 6 acquire data and input data to the
25 computation unit 1. Each of these input units may acquire
data by any means. Each input unit acquires data by manual
data input, data input from external software, input of
data acquired from a sensor, or the like. Each input unit
may input data to the computation unit 1 by any means such
30 as network communication or media transfer.
[0015] The sewage information input unit 2 acquires a
predicted water quality value D1 that is water quality
information on the sewage to be subjected to water
8
treatment, and inputs the predicted water quality value D1
to the computation unit 1. The predicted water quality
value D1 is data indicating a predicted value of the water
quality of sewage.
[0016] The accumulation model input 5 unit 3 acquires an
accumulation model D2 that expresses the amount of foulant
accumulation, and inputs the accumulation model D2 to the
computation unit 1. The accumulation model D2 is a model
that expresses the amount of foulant accumulation caused by
10 fouling. Fouling is a phenomenon in which a substance to
be separated contained in sewage physically or chemically
adheres to the membrane surface of a separation membrane.
Foulant is a substance adhered to the membrane surface of a
separation membrane.
15 [0017] The efficiency model input unit 4 acquires a
cleaning efficiency model D5 that is a model of cleaning
efficiency, and inputs the cleaning efficiency model D5 to
the computation unit 1. The cleaning efficiency model D5
is a model that expresses the deterioration of a separation
20 membrane relative to the number of cleaning events. In the
membrane filtration method, a separation membrane is
clogged through continuous use of the separation membrane,
and thus the efficiency of filtration treatment decreases
as the treatment time elapses. In the membrane filtration
25 method, it is necessary to clean and replace separation
membranes on a regular basis in order to regularly remove
foulant from membranes. In the cleaning of a separation
membrane, the separation membrane is cleaned with a
chemical solution or filtrate so that foulant is removed.
30 Generally, the cleaning cost for separation membranes is
lower than the replacement cost for separation membranes.
Therefore, the first step for dealing with clogging of a
separation membrane is to clean the separation membrane.
9
After the separation membrane is cleaned, the filtration
capacity of the separation membrane is recovered. However,
the filtration capacity of the cleaned separation membrane
is lower than that of a new separation membrane, and the
durability of the separation membrane is 5 also reduced. As
the number of times the separation membrane is cleaned
increases, the recovery rate of the filtration capacity and
the durability of the separation membrane decrease. When
the recovery rate of the filtration capacity or the
10 durability of the separation membrane becomes so low that
the separation membrane can no longer be reused, the
separation membrane is replaced. The cleaning efficiency
model D5 of the present embodiment expresses the recovery
rate of the filtration capacity and the durability of the
15 separation membrane relative to the number of times the
separation membrane is cleaned.
[0018] The cost input unit 5 acquires a maintenance cost
D6, and inputs the maintenance cost D6 to the computation
unit 1. The maintenance cost D6 is the cost required for
20 the maintenance of separation membranes, and includes the
cleaning cost required for one cleaning event for a
separation membrane and the replacement cost required for
one replacement event for a separation membrane.
[0019] The power consumption model input unit 6 acquires
25 a power consumption model D7, and inputs the power
consumption model D7 to the computation unit 1. The power
consumption model D7 is a model that expresses the power
consumption of the water treatment facility with respect to
the blockage rate of a separation membrane. In the
30 following description, one or more of the accumulation
model D2, the cleaning efficiency model D5, and the power
consumption model D7 may be referred to as a model.
[0020] The computation unit 1 computes a reference value
10
used for calculating the maintenance schedule for
separation membranes. The maintenance of a separation
membrane is exemplified by the cleaning of a separation
membrane and the replacement of a separation membrane. The
reference value is exemplified by a value 5 of the blockage
rate of a separation membrane. The following description
is based on the premise that the reference value is a
blockage rate.
[0021] The computation unit 1 uses the cost of
10 separation treatment that depends on the deterioration of a
separation membrane and the maintenance cost D6 to
determine the reference value such that the operation cost,
i.e. the sum of the cost of separation treatment, the
cleaning cost, and the replacement cost during the
15 operation of separation membranes, has the minimum value,
and calculates the cleaning schedule and replacement
schedule for separation membranes using the determined
reference value. The cost of separation treatment is a
cost such as the electricity expense per unit time required
20 for a separation membrane to perform separation treatment.
The computation unit 1 calculates the cost of separation
treatment using the cleaning efficiency model D5. The
computation unit 1 includes an accumulation amount
calculation unit 8 and a reference value calculation unit 9,
25 and outputs computation results to the result presentation
unit 7.
[0022] The accumulation amount calculation unit 8 uses
the predicted water quality value D1 and the accumulation
model D2 to calculate a time-series change D3 of the
30 fouling accumulation amount that occurs thereafter. The
time-series change D3 is data indicating a temporal change
in the fouling accumulation amount, which is the amount of
foulant accumulation. The accumulation amount calculation
11
unit 8 sends the time-series change D3 to the reference
value calculation unit 9.
[0023] The reference value calculation unit 9 includes a
plan calculation unit 13 and a pattern comparison unit 14.
The plan calculation unit 13 calculates 5 a plan set D4 for
separation membranes that shows plans for cleaning and
plans for replacement. Specifically, the plan calculation
unit 13 calculates the plan set D4 using the cleaning
efficiency model D5, the maintenance cost D6, and the time10
series change D3. At this time, the plan calculation unit
13 calculates the plan set D4 in accordance with a pattern
set P (wsh, cng). The pattern set P (wsh, cng) is a set of
combination patterns of a cleaning reference value which is
a first reference value for determining whether to clean a
15 separation membrane and a replacement reference value which
is a second reference value for determining whether to
replace a separation membrane. Here, wsh is a cleaning
reference value and cng is a replacement reference value.
The pattern set P (wsh, cng) is set before the maintenance
20 schedule for separation membranes is calculated. In the
following description, the pattern set P (wsh, cng) is
referred to as the pattern set P.
[0024] In the following description, a combination
pattern of a cleaning reference value and a replacement
25 reference value may be referred to as a reference value
pattern. That is, membrane maintenance reference values,
which are reference values for determining whether to
perform maintenance of separation membranes, may be
referred to as a reference value pattern.
30 [0025] The plan set D4 is maintenance plans that specify
the cleaning schedule and replacement schedule for
separation membranes. Therefore, the plan set D4 can also
be interpreted as separation membrane operation plans. The
12
cleaning schedule is represented by the date and time of a
separation membrane cleaning event or the elapsed time from
the start of filtration treatment to a cleaning event. The
replacement schedule is represented by the date and time of
a separation membrane replacement event or 5 the elapsed time
from the start of filtration treatment to a replacement
event. The plan set D4 is a list of the cleaning schedule
and replacement schedule specified by the time-series
change D3 for each reference value pattern.
10 [0026] In a case where filtration treatment with
separation membranes, cleaning, and replacement are
performed in accordance with each reference value pattern
included in the pattern set P, the cleaning schedule and
the replacement schedule vary between reference value
15 patterns. The plan calculation unit 13 calculates, as the
plan set D4, the cleaning schedule and the replacement
schedule that are based on a combination of a cleaning
reference value and a replacement reference value. The
plan calculation unit 13 sends the calculated plan set D4
20 to the pattern comparison unit 14.
[0027] The pattern comparison unit 14 uses the plan set
D4 and the power consumption model D7 to calculate the
operation cost required when a reference value pattern is
applied to each separation membrane in a certain period.
25 In the following description, the operation cost required
when a reference value pattern is applied to separation
membranes is referred to as a pattern operation cost. A
pattern operation cost includes the total cost of the
cleaning cost and the replacement cost and the cost of
30 separation treatment such as the electricity expense for
filtration. The sum of the cleaning cost and the
replacement cost is the cost required for the maintenance
of separation membranes, and the electricity expense for
13
filtration is the running cost of separation membranes.
The cleaning cost includes the cost of the cleaning liquid
and the like, and the replacement cost includes the price
of separation membranes.
[0028] Here, the reliability of a 5 calculation result
improves as the period used for calculating a pattern
operation cost becomes longer. In other words, if the
period for comparing pattern operation costs is too short,
a non-optimal reference value pattern may be adopted.
10 Therefore, the pattern comparison unit 14 compares
operation costs for reference value patterns in a period
including at least one separation membrane replacement
event after starting the use of a separation membrane or in
a period including at least two separation membrane
15 replacement events. In other words, the pattern comparison
unit 14 compares operation costs in a period longer than
the replacement cycle for separation membranes.
[0029] The pattern comparison unit 14 calculates a
pattern operation cost for each reference value pattern in
20 the pattern set P. The pattern comparison unit 14 compares
the calculated pattern operation costs and determines the
minimum pattern operation cost. The pattern comparison
unit 14 extracts, from the pattern set P, the reference
value pattern used for the calculation of the minimum
25 pattern operation cost. The pattern comparison unit 14
sends the minimum pattern operation cost and the extracted
reference value pattern to the result presentation unit 7.
[0030] The result presentation unit 7 presents the user
with the minimum pattern operation cost and the reference
30 value pattern that achieves the minimum pattern operation
cost. The presentation means by the result presentation
unit 7 may be any means such as displaying on a display or
printing on paper.
14
[0031] Next, each calculation performed by the
accumulation amount calculation unit 8, the plan
calculation unit 13, and the pattern comparison unit 14
will be described using a specific example. The fouling
accumulation amount is proportional to 5 the inflow load of
sewage. Therefore, in a case where the predicted water
quality value D1 of the input load is Xt and a proportional
expression is used as the accumulation model D2, the timeseries
change D3 of the fouling accumulation amount V(t)
10 that is calculated by the accumulation amount calculation
unit 8 is expressed as Formula (1) below. In Formula (1),
“a” is a proportional constant.
[0032]
[Math. 1]
... Formula (1)
[0033] Further, because the accumulation amount
15 calculation unit 8 can calculate the blockage rate of a
separation membrane using the fouling accumulation amount,
the blockage rate can be used as a cleaning reference value
and a replacement reference value for separation membranes.
In a case where a proportional expression with respect to
20 the fouling accumulation amount is used as a calculation
model for the blockage rate of a separation membrane, the
blockage rate R(t) of a separation membrane can be
expressed as Formula (2) below. In Formula (2), “b” is a
proportional constant.
25 [0034]
[Math. 2]
... Formula (2)
[0035] Instead of a proportional expression such as
Formula (2) that is used here for simplicity, Ruth's
equation represented by Formula (3) below or past data on
water quality changes may be used for the accumulation
t V(t)  a X
R(t)  b Vt
15
model D2. In Formula (3), Vf is the total amount of
filtrate, A is the membrane area of the separation membrane,
k is a resistance coefficient, Pf is the transmembrane
pressure difference during filtration, c is a cake ratio, μ
is the viscosity coefficient of 5 water, and V0 is a
filtration constant. The transmembrane pressure difference
is the pressure difference between the primary side, which
is the sewage side relative to the separation membrane, and
the secondary side, which is the filtrate side relative to
10 the separation membrane.
[0036]
[Math. 3]
... Formula (3)
[0037] The plan calculation unit 13 calculates the plan
set D4 for separation membranes using Formula (1), the
pattern set P set in advance, and the cleaning efficiency
15 model D5. The cleaning efficiency model D5 can express the
phenomenon that a separation membrane deteriorates due to
repeated cleaning by considering the efficiency of
separation membrane cleaning. If the efficiency of
separation membrane cleaning is not considered, that is, if
20 the efficiency of cleaning is always 100%, a separation
membrane will always recover by cleaning to the same level
as a new one regardless of the number of repetitions of
separation membrane cleaning. This means that the
separation membrane need not be replaced, which is not the
25 case. In a case where a model in which the efficiency of
separation membrane cleaning decreases exponentially with
respect to the number of cleaning events n for the
separation membrane is used as the separation membrane
cleaning efficiency model D5, the efficiency of separation
30 membrane cleaning Ref(n) can be expressed as Formula (4)
   0
2
c Vf V
A k Pf
dt
dVf
 
 

16
below. Here, “c” is the base of the exponential function.
[0038]
[Math. 4]
... Formula (4)
[0039] FIG. 2 is a diagram illustrating an example of a
pattern set used in the operation support device according
to the first embodiment. FIG. 2 depicts 5 the combination of
a blockage rate as a cleaning reference value and a
blockage rate as a replacement reference value in each
reference value pattern. In FIG. 2 and FIG. 4 described
later, examples of reference value patterns are denoted by
10 patterns A to I. Here, specific numerical examples of
cleaning reference values and numerical examples of
replacement reference values in patterns A to I are
illustrated.
[0040] FIG. 3 is a diagram illustrating an example of a
15 time-series change of the blockage rate calculated by the
operation support device according to the first embodiment.
FIG. 3 depicts a graph of the time-series change D3 of the
blockage rate for the case that cleaning and replacement
are performed using pattern A, pattern E, or pattern I of
20 the reference value patterns illustrated in FIG. 2. The
horizontal axis of the graph depicted in FIG. 3 represents
the time that elapses after the separation membrane starts
to filter sewage, and the vertical axis represents the
blockage rate of the separation membrane corresponding to
25 the transmembrane pressure difference.
[0041] The change transition of the blockage rate for
pattern A is denoted by change transition PA, the change
transition of the blockage rate for pattern E is denoted by
change transition PE, and the change transition of the
30 blockage rate for pattern I is denoted by change transition
PI. Focusing on the first cleaning event, it can be seen
  n
ef R n  c
17
that the cleaning in pattern A is performed at a stage
where the blockage rate is low, and the cleaning in pattern
I is performed at a stage where the blockage rate is high.
The plan calculation unit 13 can formulate the plan set D4
based on the change transition of the blockage 5 rate for the
case that each reference value pattern is adopted.
[0042] The pattern comparison unit 14 uses the
maintenance cost D6 and the power consumption model D7 to
calculate the pattern operation cost for each plan included
10 in the plan set D4. The pattern comparison unit 14
determines the minimum pattern operation cost of the
calculated pattern operation costs, and acquires the
reference value pattern corresponding to the minimum
pattern operation cost. The pattern comparison unit 14
15 creates a maintenance plan based on the acquired reference
value pattern. The maintenance plan created by the pattern
comparison unit 14 includes the cleaning schedule for
separation membranes and the replacement schedule for
separation membranes.
20 [0043] In a case where a proportional expression with
respect to the blockage rate is used as the power
consumption model D7, the power consumption W(t) can be
expressed as Formula (5) below. Here, “d” is a
proportional constant.
25 [0044]
[Math. 5]
... Formula (5)
[0045] For calculating a pattern operation cost, the
pattern comparison unit 14 determines the calculation
period used for calculating the pattern operation cost. In
this case, a calculation period sufficiently longer than
30 the replacement cycle for separation membranes is used as
described above. That is, the pattern comparison unit 14
Wt  d Rt
18
sets the calculation period based on the replacement cycle
for separation membranes. The replacement cycle for
separation membranes is the period from the replacement
event for a separation membrane to the next replacement
event. As the calculation period, the 5 pattern comparison
unit 14 may use a period designated by an instruction from
the user, or may use an initial value set in advance.
[0046] Here, the difference in determination result
between the case of a short calculation period such as a
10 calculation period T1 and the case of a calculation period
T2 longer than the calculation period T1 will be described.
An example of the calculation period T1 is 240 days, and an
example of the calculation period T2 is 300 days. The
present embodiment is based on the premise that the unit of
15 periods is “day”, but any unit of periods may be used such
as “week” or “month”.
[0047] FIG. 4 is a diagram illustrating the relationship
between calculation periods used by the operation support
device according to the first embodiment to calculate
20 pattern operation costs and calculation results of pattern
operation costs. The calculation results of FIG. 4 contain
the calculation results for patterns A, E, and I in the
case that the pattern operation cost calculation period is
240 days and the calculation results for patterns A, E, and
25 I in the case that the pattern operation cost calculation
period is 300 days. The calculation results here are “the
number of cleaning events”, “the number of replacement
events”, “electricity expense”, and “pattern operation
cost”. “The number of cleaning events” is the number of
30 times the separation membrane is cleaned. “The number of
replacement events” is the number of times the separation
membrane is replaced. “Electricity expense” is the
electricity expense for the filtration treatment
19
corresponding to the calculation period.
[0048] Determination results obtained using a short
calculation period for pattern operation costs may differ
from those obtained using a long calculation period for
pattern operation costs. For example, 5 in FIG. 4, the
calculation period T1=240 days is shorter than the longest
replacement cycle in the compared patterns. Specifically,
calculation period T1<(replacement cycle for the case of
pattern I) is satisfied. Therefore, the determination
10 results differ between the case of the calculation period
T1=240 days and the case of the calculation period T2=300
days. That is, pattern E has the minimum pattern operation
cost in the case of the calculation period T1=240 days,
whereas pattern I has the minimum pattern operation cost in
15 the case of the calculation period T2=300 days.
[0049] Because the reliability of calculation results
increases as the calculation period for pattern operation
costs becomes longer, the reliability of the calculation
results in the case of the calculation period T2=300 days
20 is higher than that in the case of the calculation period
T1=240 days. That is, pattern I is considered to have the
minimum pattern operation cost. Therefore, the pattern
comparison unit 14 does not adopt a calculation period,
such as the calculation period T1=240 days, that makes the
25 determination results different from those with the
calculation period T2=300 days.
[0050] The pattern comparison unit 14 uses, as the
calculation period for pattern operation costs, a length
that is at least twice the longest replacement cycle in the
30 reference value patterns such as patterns A to I to be
compared. Note that it is desirable to extend the
calculation period for pattern operation costs and
recalculate pattern operation costs repeatedly until the
20
determination results converge. The pattern comparison
unit 14 compares the pattern operation costs calculated
using a sufficiently long calculation period, and
determines the reference value pattern that achieves the
minimum pattern operation cost. 5 Then, the result
presentation unit 7 presents the reference value pattern
that achieves the minimum pattern operation cost.
[0051] FIG. 5 is a diagram illustrating an example of a
presentation screen presented by the operation support
10 device according to the first embodiment. The result
presentation unit 7 displays a timing such as “when the
membrane is XX% blocked” as the cleaning timing for
separation membranes. The result presentation unit 7 also
displays a timing such as “when the transmembrane pressure
15 difference after cleaning reaches YY% or more” as the
replacement timing for separation membranes. The cleaning
timing is a timing defined by a condition for cleaning, and
the replacement timing is a timing defined by a condition
for replacement.
20 [0052] The result presentation unit 7 also displays the
reference value pattern. The result presentation unit 7
may display the calculation result in the form of converted
values representing information to be presented to the
operator of the water treatment facility. In this case,
25 the pattern comparison unit 14 converts the calculation
result into information to be presented to the operator of
the water treatment facility. Converted values are, for
example, the date of subsequent cleaning or replacement
events, any cleaning plan and replacement plan other than
30 the plan determined by the pattern comparison unit 14, and
the cost for the case of executing the cleaning plan and
replacement plan. For displaying the date of the next and
subsequent cleaning or replacement events, the result
21
presentation unit 7 displays the schedule of cleaning ahead
such as “Za days later, Zb days later, and Zc days later”,
and the schedule of replacement ahead such as “Zd days
later, Ze days later, and Zf days later”.
[0053] Next, the procedure for computation 5 processing by
the computation unit 1 will be described. FIG. 6 is a
flowchart illustrating the procedure for computation
processing by the computation unit according to the first
embodiment. This procedure is the same as the procedure
10 for the case that the computation unit 1 is executed as a
program.
[0054] The operation support device 100A prepares the
pattern set P of reference value patterns in advance before
the computation unit 1 starts computation. The pattern set
15 P includes various reference values. Further, the
operation support device 100A determines the initial value
of the calculation period in advance.
[0055] The computation unit 1 acquires the predicted
water quality value D1, which is information on sewage, and
20 the accumulation model D2 (steps S1 and S2). The
computation unit 1 calculates the future time-series change
D3 of the fouling accumulation amount based on the
predicted water quality value D1 and the accumulation model
D2 (step S3).
25 [0056] After that, the computation unit 1 acquires the
separation membrane cleaning efficiency model D5, the
maintenance cost D6, and the power consumption model D7 for
the water treatment facility with respect to the blockage
rate of a separation membrane (steps S4, S5, and S6). The
30 computation unit 1 acquires the values of the first
reference value pattern from the pattern set P, calculates
a cleaning plan and a replacement plan that are plans in
the calculation period (step S7), and calculates a pattern
22
operation cost based on the calculated cleaning plan and
replacement plan (step S8). Then, the computation unit 1
compares the value of the pattern operation cost calculated
in step S8 with the minimum value of the pattern operation
cost in the previous calculation results. 5 The computation
unit 1 determines whether the pattern operation cost
calculated in step S8 has the minimum value (step S9).
[0057] When the pattern operation cost calculated in
step S8 has the minimum value (step S9: Yes), the
10 computation unit 1 updates and records the minimum value of
the pattern operation cost and the reference value pattern
used for the calculation of the minimum pattern operation
cost (step S9-1). Note that in the case of the initial
calculation of a pattern operation cost, the value of the
15 pattern operation cost calculated in step S8 is necessarily
the minimum value. The computation unit 1 performs step
S10 after step S9-1.
[0058] When the pattern operation cost calculated in
step S8 does not have the minimum value (step S9: No), the
20 computation unit 1 determines whether there is a reference
value pattern in the pattern set P for which the
calculation of the pattern operation cost has not been
completed. That is, the computation unit 1 determines
whether the comparison with the minimum value has been
25 completed for all the reference value patterns (step S10).
When there is a reference value pattern for which the
calculation has not been completed (step S10: No), the
computation unit 1 acquires one reference value pattern for
which the calculation has not been completed, and updates
30 the reference value pattern for calculation (step S10-1).
Then, the computation unit 1 uses the new reference value
pattern to perform steps S7 to S10 again.
[0059] The computation unit 1 calculates the pattern
23
operation costs for all the reference value patterns, and
repeats steps S10-1 and S7 to S10 until the comparison
between the calculated pattern operation costs and the
minimum value is completed. After the pattern operation
costs for all the reference value patterns 5 are calculated
and compared with the minimum value (step S10: Yes), the
computation unit 1 determines whether the reference value
pattern that achieves the minimum pattern operation cost in
the calculation period for the calculation of the current
10 pattern operation cost is the same as the reference value
pattern having the minimum pattern operation cost in the
calculation of the previous pattern operation cost. The
minimum reference value pattern derived in the previous
calculation is the minimum reference value pattern
15 calculated in the previous loop of steps S7 to S10, and the
reference value pattern derived in the current calculation
is the minimum reference value pattern calculated in the
current loop of steps S7 to S10. Hereinafter, they are
referred to as minimum reference value patterns.
20 [0060] When the current minimum reference value pattern
is different from the previous minimum reference value
pattern, the computation unit 1 updates the latest minimum
reference value pattern to the current minimum reference
value pattern. The computation unit 1 determines whether
25 the minimum reference value pattern has been updated (step
S11).
[0061] In response to determining that the minimum
reference value pattern has been updated (step S11: Yes),
that is, the current minimum reference value pattern is
30 different from the previous minimum reference value pattern,
the computation unit 1 updates the current calculation
period T to a new calculation period T′=(T+ΔT) (step S11-1).
Then, the computation unit 1 executes steps S7 to S11 using
24
the new calculation period T'.
[0062] In response to determining that the minimum
reference value pattern has not been updated (step S11: No),
that is, the current minimum reference value pattern is the
same as the previous minimum reference 5 value pattern, the
computation unit 1 determines that the minimum reference
value patterns have converged. Upon determining that the
minimum reference value patterns have converged, the
computation unit 1 sends the current or previous minimum
10 reference value pattern to the result presentation unit 7.
Thus, the result presentation unit 7 displays the current
or previous minimum reference value pattern as a
calculation result (step S12).
[0063] As described above, in the first embodiment, the
15 reference value pattern is determined such that the pattern
operation cost, i.e. the sum of the cost of separation
treatment, the cleaning cost for separation membranes, and
the replacement cost for separation membranes during the
operation of separation membranes, has the minimum value,
20 and the cleaning schedule and replacement schedule for
separation membranes are calculated using the determined
reference value pattern. This makes it possible to easily
predict the cleaning schedule and replacement schedule for
separation membranes for reducing the pattern operation
25 cost of separation membranes without the need for reference
value input based on specialized knowledge.
[0064] Second Embodiment.
Next, the second embodiment of the present invention
will be described with reference to FIGS. 7 to 10. In the
30 first embodiment, the minimum reference value pattern is
calculated with no change in the current operating
situation of the water treatment facility. If this minimum
reference value pattern does not satisfy the membrane
25
maintenance condition required by the operator, it is
desirable to change the operating condition of the water
treatment facility such that it satisfies the membrane
maintenance condition. In the second embodiment, therefore,
the operating condition of the water treatment 5 facility is
changed so as to satisfy the membrane maintenance condition,
and the minimum reference value pattern satisfying the
membrane maintenance condition is calculated. The membrane
maintenance condition is a condition for maintenance of
10 separation membranes.
[0065] FIG. 7 is a functional configuration diagram of
an operation support device according to the second
embodiment. Components of the operation support device
100B illustrated in FIG. 7 that achieve the same functions
15 as those of the operation support device 100A of the first
embodiment illustrated in FIG. 1 are denoted by the same
reference signs, and duplicate descriptions are omitted.
[0066] The operation support device 100B includes a
membrane maintenance condition input unit 19, a reference
20 value adjustment unit 20B, and a facility control unit 21,
in addition to the components provided in the operation
support device 100A. The facility control unit 21 is
connected to a water treatment facility 23B including a
membrane maintenance facility 22.
25 [0067] The membrane maintenance condition input unit 19
acquires a membrane maintenance condition designated by the
operator, and inputs the membrane maintenance condition to
the reference value adjustment unit 20B. The membrane
maintenance condition is exemplified by a replacement cycle
30 representing the period to the replacement of a separation
membrane. The membrane maintenance condition input unit 19
acquires data by manual data input, data input from
external software, input of data acquired from a sensor, or
26
the like.
[0068] The reference value adjustment unit 20B is
connected to the result presentation unit 7, the membrane
maintenance condition input unit 19, and the facility
control unit 21. The reference value adjustment 5 unit 20B
is also connected to the accumulation model input unit 3,
the efficiency model input unit 4, the cost input unit 5,
and the power consumption model input unit 6.
[0069] The reference value adjustment unit 20B acquires
10 the calculation result by the computation unit 1 from the
result presentation unit 7. Specifically, the reference
value adjustment unit 20B acquires the minimum reference
value pattern and the maintenance schedule from the result
presentation unit 7. The reference value adjustment unit
15 20B changes the operating condition of the water treatment
facility 23B such that the minimum reference value pattern
computed by the computation unit 1 satisfies the membrane
maintenance condition. The operating condition is a
condition that affects at least one of the accumulation
20 model D2, the cleaning efficiency model D5, the maintenance
cost D6, and the power consumption model D7. The reference
value adjustment unit 20B here sets an operating condition
that makes the maintenance schedule satisfy the membrane
maintenance condition, and sends the set operating
25 condition to the accumulation model input unit 3, the
efficiency model input unit 4, the cost input unit 5, or
the power consumption model input unit 6. For example, the
reference value adjustment unit 20B sets a new accumulation
model D2 and sends the set accumulation model D2 to the
30 accumulation model input unit 3. The computation unit 1
recalculates the minimum reference value pattern using the
changed operating condition.
[0070] The reference value adjustment unit 20B repeats
27
the process of changing the operating condition of the
water treatment facility 23B and causing the computation
unit 1 to calculate the minimum reference value pattern
until the calculation result by the computation unit 1
satisfies the membrane maintenance 5 condition.
[0071] Upon determining the operating condition that
makes the maintenance schedule satisfy the membrane
maintenance condition, the reference value adjustment unit
20B inputs the determined operating condition to the
10 facility control unit 21.
[0072] The facility control unit 21 controls the water
treatment facility 23B. The facility control unit 21 of
the present embodiment changes the operating condition of
the membrane maintenance facility 22 in accordance with the
15 operating condition from the reference value adjustment
unit 20B. The membrane maintenance facility 22 is a
facility or device for maintaining the state of separation
membranes. An example of the membrane maintenance facility
22 is a blower that regulates the volume of aeration air on
20 the membrane surface or a pump that regulates the
concentration of the cleaning liquid used for membrane
cleaning.
[0073] The operation support device 100B calculates the
minimum reference value pattern and the maintenance
25 schedule using a method similar to that of the operation
support device 100A. Hereinafter, processing that is
performed by the operation support device 100B after the
operation support device 100B calculates the minimum
reference value pattern and the maintenance schedule will
30 be described. After calculating the minimum reference
value pattern and the maintenance schedule, the operation
support device 100B changes the operating condition such
that it satisfies the membrane maintenance condition, and
28
recalculates the minimum reference value pattern and the
maintenance schedule.
[0074] FIG. 8 is a diagram for explaining the process of
changing the operating condition and the process of
recalculating the minimum reference value 5 pattern by the
operation support device according to the second embodiment.
Here, consider the case in which the membrane maintenance
facility 22 is a blower 22X that performs membrane surface
aeration, and the operation support device 100B changes the
10 operating condition of the blower 22X such that the
maintenance schedule satisfies the membrane maintenance
condition.
[0075] The water treatment facility 23B includes a
pretreatment tank 30A, a membrane filtration tank 30B, a
15 separation membrane 31, and the blower 22X. The
pretreatment tank 30A precipitates suspended matter
contained in sewage W1 and then sends the sewage W1 to the
membrane filtration tank 30B. The membrane filtration tank
30B includes the separation membrane 31, and the sewage W1
20 is filtered by the separation membrane 31. The blower 22X
performs membrane surface aeration on the separation
membrane 31. The membrane filtration tank 30B sends the
filtered sewage W1 to the outside.
[0076] Generally, as the air volume of membrane surface
25 aeration for the separation membrane 31, that is, the
output of the blower 22X, increases, the increase of the
amount of foulant accumulation becomes small. In addition,
generally, as the output of the blower 22X increases, the
amount of power consumption per unit time increases.
30 Therefore, the operating condition that can be changed by
changing the output of the blower 22X is the accumulation
model D2 and the power consumption model D7. The following
description is based on the premise that the accumulation
29
model D2 and the power consumption model D7 that are the
same as the models described in the first embodiment are
adopted. That is, the operation support device 100B here
uses Formulas (1) and (2) as the accumulation model D2 and
uses Formula (5) as the power consumption 5 model D7. The
reference value adjustment unit 20B inputs the accumulation
model D2 adjusted to satisfy the membrane maintenance
condition to the accumulation model input unit 3, and
inputs the power consumption model D7 adjusted to satisfy
10 the membrane maintenance condition to the power consumption
model input unit 6. It is assumed that the maintenance
condition input to the membrane maintenance condition input
unit 19 is that the replacement cycle for the separation
membrane 31 is 51 days or more. In a case where the
15 concentration of the cleaning liquid used for membrane
cleaning is changed as the operating condition, the
cleaning cost for the separation membrane 31 varies
according to the concentration of the cleaning liquid, and
thus the maintenance cost D6 varies.
20 [0077] Hereinafter, specific procedures for the process
of changing the operating condition and the process of
recalculating the minimum reference value pattern by the
operation support device 100B will be described. The
computation unit 1 of the operation support device 100B, in
25 a similar manner to that of the operation support device
100A of the first embodiment, calculates the minimum
reference value pattern under the current operating
condition, and calculates the replacement cycle for the
separation membrane 31 based on the calculated minimum
30 reference value pattern. The computation unit 1 inputs the
replacement cycle, which is the calculation result, to the
result presentation unit 7, and the result presentation
unit 7 inputs the replacement cycle to the reference value
30
adjustment unit 20B. The reference value adjustment unit
20B compares the membrane maintenance condition input from
the membrane maintenance condition input unit 19 with the
replacement cycle as the calculation result.
[0078] FIG. 9 is a diagram 5 for explaining the
replacement cycle calculated by the operation support
device according to the second embodiment without changing
the operating condition. FIG. 9 depicts the correspondence
relationship between the initial value of the blower output
10 that is the output of the blower 22X, the initial value of
“a” in Formula (1), the initial value of “b” in Formula (2),
the initial value of “d” in Formula (5), the initial
cleaning reference value, the initial replacement reference
value, the replacement cycle, and the pattern operation
15 cost. Note that in FIG. 9 and FIGS. 10, 13, and 14
described later, the cleaning reference values and
replacement reference values are expressed in percentage.
[0079] As illustrated in FIG. 9, if the water treatment
facility 23B is operated with the initial value of the
20 blower output, the replacement cycle is 44 days, which is
shorter than 51 days specified by the maintenance condition.
That is, because the replacement cycle calculated with the
initial value of the blower output is shorter than the
replacement condition specified by the maintenance
25 condition, the initial value of the blower output does not
satisfy the maintenance condition. In this case, the
reference value adjustment unit 20B creates the
accumulation model D2 and the power consumption model D7
for the case of increasing the blower output. The
30 reference value adjustment unit 20B creates a new
accumulation model D2 by updating “b” in Formula (2) and
creates a new power consumption model D7 by updating “d” in
Formula (5).
31
[0080] The reference value adjustment unit 20B inputs
the accumulation model D2 for the case of increasing the
blower output to the accumulation model input unit 3, and
inputs the power consumption model D7 for the case of
increasing the blower output to the power 5 consumption model
input unit 6. Then, the computation unit 1 recalculates
the minimum reference value pattern.
[0081] In this way, the operation support device 100B
repeatedly recalculates the minimum reference value pattern
10 while updating models, and determines the models or the
maintenance cost D6 that can satisfy the maintenance
condition. That is, the computation unit 1 changes the
operating condition of the membrane maintenance facility 22
such that it satisfies the maintenance condition, updates
15 at least one of the models and the maintenance cost D6
based on the changed operating condition, and recalculates
the cleaning schedule and replacement schedule for the
separation membrane 31.
[0082] FIG. 10 is a diagram for explaining the
20 replacement cycles calculated by the operation support
device according to the second embodiment by changing the
operating condition. FIG. 10 depicts the correspondence
relationship between the number of replacement cycle
calculations, blower output values, “a” in Formula (1), “b”
25 in Formula (2), “d” in Formula (5), cleaning reference
values, replacement reference values, replacement cycles,
and pattern operation costs.
[0083] The number of replacement cycle calculations is
the number of times the reference value adjustment unit 20B
30 calculates the replacement cycle. The number of
calculations “1” indicates that the replacement cycle is
calculated using the initial value of the blower output.
The number of calculations “2” indicates that the
32
replacement cycle is recalculated for the first time after
the blower output is changed, and the number of
calculations “3” indicates that the replacement cycle is
further recalculated after the blower output is further
5 changed.
[0084] The reference value adjustment unit 20B increases
the blower output each time the blower output is calculated.
As the blower output increases, “b” in Formula (2)
decreases and “d” in Formula (5) increases. The
10 replacement cycle becomes longer as the blower output
increases. As illustrated in FIG. 10, the replacement
cycle reaches 54 days, which is more than 51 days specified
by the maintenance condition, by the fourth calculation of
the replacement cycle. The reference value adjustment unit
15 20B changes the accumulation model D2 using “b” in Formula
(2) that has changed, and changes the power consumption
model D7 using “d” in the Formula (5).
[0085] Upon determining the models that can satisfy the
maintenance condition, the reference value adjustment unit
20 20B inputs the blower output that can satisfy the
maintenance condition to the facility control unit 21.
[0086] In response to receiving the changed operation
condition such as the blower output, the facility control
unit 21 calculates the amount of operation for the water
25 treatment facility 23B such that it matches the received
operation condition. Because the facility control unit 21
can satisfy the maintenance condition by increasing the
blower output for membrane surface aeration by 1.6 times,
the facility control unit 21 sends a signal for increasing
30 the amount of operation by 1.6 times to the blower 22X.
Thus, the blower 22X increases the amount of membrane
surface aeration by 1.6 times. As a result, the water
treatment facility 23B can filter the sewage W1 while
33
satisfying the maintenance condition. Note that the reason
why the cleaning reference values vary depending on the
number of calculations is that replacement cycles can be
calculated using different calculation periods, and
different calculation periods can 5 lead to different
cleaning reference values.
[0087] As described above, the second embodiment makes
it possible to easily predict the cleaning schedule and
replacement schedule for the separation membrane 31 for
10 reducing the pattern operation cost of the separation
membrane 31 without the need for reference value input
based on specialized knowledge, while satisfying the
maintenance condition designated by the operator by
controlling the membrane maintenance facility 22.
15 [0088] Third Embodiment.
Next, the third embodiment of the present invention
will be described with reference to FIGS. 11 to 14. In the
second embodiment, the membrane maintenance facility 22 is
controlled in order to satisfy the maintenance condition.
20 Alternatively, the water quality of the sewage W1 may be
controlled by the pretreatment tank 30A, which is the
facility at the pre-stage of membrane separation. In the
third embodiment, the operating condition in the
pretreatment tank 30A is changed so as to satisfy the
25 maintenance condition designated by the operator, and the
minimum reference value pattern satisfying the maintenance
condition is calculated.
[0089] FIG. 11 is a functional configuration diagram of
an operation support device according to the third
30 embodiment. Components of the operation support device
100C illustrated in FIG. 11 that achieve the same functions
as those of the operation support device 100A or the
operation support device 100B are denoted by the same
34
reference signs, and duplicate descriptions are omitted.
[0090] The operation support device 100C includes a
reference value adjustment unit 20C, instead of the
reference value adjustment unit 20B of the operation
support device 100B. The reference value 5 adjustment unit
20C is connected to the result presentation unit 7, the
membrane maintenance condition input unit 19, and the
facility control unit 21. The reference value adjustment
unit 20C is also connected to the sewage information input
10 unit 2, the accumulation model input unit 3, and the power
consumption model input unit 6. The facility control unit
21 is connected to a water treatment facility 23C.
[0091] The reference value adjustment unit 20C acquires
the minimum reference value pattern and the maintenance
15 schedule from the result presentation unit 7. The
reference value adjustment unit 20C changes the operating
condition of the water treatment facility 23C such that the
minimum reference value pattern and the maintenance
schedule computed by the computation unit 1 satisfy the
20 membrane maintenance condition. The operating condition is
a condition that affects at least one of the predicted
water quality value D1, the accumulation model D2, and the
power consumption model D7. The reference value adjustment
unit 20C here sets an operating condition that makes the
25 minimum reference value pattern and the maintenance
schedule satisfy the membrane maintenance condition, and
sends the set operating condition to the sewage information
input unit 2, the accumulation model input unit 3, or the
power consumption model input unit 6. For example, the
30 reference value adjustment unit 20C sets a new power
consumption model D7 and sends the set power consumption
model D7 to the power consumption model input unit 6. The
computation unit 1 recalculates the minimum reference value
35
pattern using the changed operating condition.
[0092] The reference value adjustment unit 20C repeats
the process of changing the operating condition of the
water treatment facility 23C and causing the computation
unit 1 to calculate the minimum reference 5 value pattern
until the calculation result by the computation unit 1
satisfies the membrane maintenance condition.
[0093] Upon determining the operating condition that
makes the minimum reference value pattern and the
10 maintenance schedule satisfy the membrane maintenance
condition, the reference value adjustment unit 20C inputs
the determined operating condition to the facility control
unit 21.
[0094] The facility control unit 21 controls the water
15 treatment facility 23C. The facility control unit 21 of
the present embodiment changes the operating condition of
the pretreatment tank 30A in accordance with the operating
condition from the reference value adjustment unit 20C. An
example of a device for which the operating condition is
20 changed is a facility or device for maintaining the state
of the pretreatment tank 30A, such as a blower of an
aerobic tank for the activated sludge process or a chemical
injection pump for freshwater treatment. The following
description is based on the premise that the device for
25 which the operating condition is changed is a blower of an
aerobic tank.
[0095] The operation support device 100C calculates the
minimum reference value pattern and the maintenance
schedule using a method similar to that of the operation
30 support device 100A. Processing that is performed by the
operation support device 100C after the operation support
device 100C calculates the minimum reference value pattern
and the maintenance schedule will be described. After
36
calculating the minimum reference value pattern and the
maintenance schedule, the operation support device 100C
changes the operating condition, and recalculates the
minimum reference value pattern and the maintenance
5 schedule.
[0096] FIG. 12 is a diagram for explaining the process
of changing the operating condition and the process of
recalculating the minimum reference value pattern by the
operation support device according to the third embodiment.
10 Here, consider the case in which the device for which the
operating condition is changed is a blower 25 that sends
air to an aerobic tank for activated sludge treatment, and
the operation support device 100C changes the operating
condition of the blower 25 such that the maintenance
15 schedule satisfies the membrane maintenance condition.
[0097] The water treatment facility 23C includes the
pretreatment tank 30A, the membrane filtration tank 30B,
the separation membrane 31, and the blower 25. The
pretreatment tank 30A includes an aerobic tank, and the
20 blower 25 sends air to the aerobic tank.
[0098] Generally, as the blower output of the aerobic
tank increases, the water quality of the sewage W1 improves.
In addition, generally, as the blower output increases, the
amount of power consumption per unit time increases.
25 Therefore, the operating condition that can be changed by
changing the output of the blower 25 is the predicted water
quality value D1 and the power consumption model D7. The
operation support device 100C uses the models same as those
used in the first embodiment by regarding the improvement
30 of the water quality of the sewage W1 flowing into the
separation membrane 31 as the reduction of the amount of
foulant accumulation. That is, the operation support
device 100C uses Formula (1) as the predicted water quality
37
value D1 and uses Formula (5) as the power consumption
model D7. The reference value adjustment unit 20C inputs
the predicted water quality value D1 adjusted to satisfy
the membrane maintenance condition to the sewage
information input unit 2, and inputs the 5 power consumption
model D7 adjusted to satisfy the membrane maintenance
condition to the power consumption model input unit 6. It
is assumed that the maintenance condition input to the
membrane maintenance condition input unit 19 is that the
10 replacement cycle for the separation membrane 31 is 51 days
or more.
[0099] Hereinafter, specific procedures for the process
of changing the operating condition and the process of
recalculating the minimum reference value pattern by the
15 operation support device 100C will be described. The
computation unit 1 of the operation support device 100C, in
a similar manner to that of the operation support device
100A of the first embodiment, calculates the minimum
reference value pattern under the current operating
20 condition, and calculates the replacement cycle for the
separation membrane 31 based on the calculated minimum
reference value pattern. The computation unit 1 inputs the
replacement cycle, which is the calculation result, to the
result presentation unit 7, and the result presentation
25 unit 7 inputs the replacement cycle to the reference value
adjustment unit 20C. The reference value adjustment unit
20C compares the membrane maintenance condition input from
the membrane maintenance condition input unit 19 with the
replacement cycle as the calculation result.
30 [0100] FIG. 13 is a diagram for explaining the
replacement cycle calculated by the operation support
device according to the third embodiment without changing
the operating condition. FIG. 13 depicts the
38
correspondence relationship between the initial value of
the blower output that is the output of the blower 25, the
initial value of “a” in Formula (1), the initial value of
“d” in Formula (5), the initial cleaning reference value,
the initial replacement reference value, 5 the replacement
cycle, and the pattern operation cost.
[0101] As illustrated in FIG. 13, if the water treatment
facility 23C is operated with the initial value of the
blower output, the replacement cycle is 40 days, which is
10 shorter than 51 days specified by the maintenance condition.
That is, because the replacement cycle calculated with the
initial value of the blower output is shorter than the
replacement condition specified by the maintenance
condition, the initial value of the blower output does not
15 satisfy the maintenance condition. In this case, the
reference value adjustment unit 20C creates the predicted
water quality value D1 and the power consumption model D7
for the case of increasing the blower output. The
reference value adjustment unit 20C creates a new predicted
20 water quality value D1 by updating “a” in Formula (1) and
creates a new power consumption model D7 by updating “d” in
Formula (5).
[0102] The reference value adjustment unit 20C inputs
the predicted water quality value D1 for the case of
25 increasing the blower output to the sewage information
input unit 2, and inputs the power consumption model D7 for
the case of increasing the blower output to the power
consumption model input unit 6. Then, the computation unit
1 recalculates the minimum reference value pattern.
30 [0103] In this way, the operation support device 100C
repeatedly recalculates the minimum reference value pattern
while updating models, and determines the predicted water
quality value D1 or the power consumption model D7 that can
39
satisfy the maintenance condition. That is, the
computation unit 1 changes the operating condition of the
blower 25 such that it satisfies the maintenance condition,
updates at least one of the predicted water quality value
D1 and the power consumption model D7 based 5 on the changed
operating condition, and recalculates the cleaning schedule
and replacement schedule for the separation membrane 31.
[0104] FIG. 14 is a diagram for explaining the
replacement cycles calculated by the operation support
10 device according to the third embodiment by changing the
operating condition.
[0105] FIG. 14 depicts the correspondence relationship
between the number of replacement cycle calculations,
blower output values, “a” in Formula (1), “d” in Formula
15 (5), cleaning reference values, replacement reference
values, replacement cycles, and pattern operation costs.
[0106] The reference value adjustment unit 20C increases
the blower output each time the blower output is calculated,
in a similar manner to the reference value adjustment unit
20 20B. As the blower output increases, “a” in Formula (1)
decreases and “d” in Formula (5) increases. The
replacement cycle becomes longer as the blower output
increases. As illustrated in FIG. 14, the replacement
cycle reaches 59 days, which is more than 51 days specified
25 by the maintenance condition, by the fourth calculation of
the replacement cycle. The reference value adjustment unit
20C changes the predicted water quality value D1 using “a”
in Formula (1) that has changed, and changes the power
consumption model D7 using “d” in the Formula (5).
30 [0107] Upon determining the predicted water quality
value D1 and the power consumption model D7 that can
satisfy the maintenance condition, the reference value
adjustment unit 20C inputs the blower output that can
40
satisfy the maintenance condition to the facility control
unit 21.
[0108] In response to receiving the changed operation
condition such as the blower output, the facility control
unit 21 calculates the amount of operation 5 for the water
treatment facility 23C such that it matches the received
operation condition. Because the facility control unit 21
can satisfy the maintenance condition by increasing the
blower output of the blower 25 that sends air to the
10 aerobic tank by 1.38 times, the facility control unit 21
sends a signal for increasing the amount of operation by
1.38 times to the blower 25. Thus, the blower 25 increases
the amount of air that is sent to the aerobic tank by 1.38
times. As a result, the water treatment facility 23C can
15 filter the sewage W1 while satisfying the maintenance
condition.
[0109] As described above, the third embodiment makes it
possible to easily predict the cleaning schedule and
replacement schedule for the separation membrane 31 for
20 reducing the pattern operation cost of the separation
membrane 31 without the need for reference value input
based on specialized knowledge, while satisfying the
maintenance condition designated by the operator by
controlling the device for maintaining the state of the
25 pretreatment tank 30A.
[0110] Here, a hardware configuration for implementing
the functions of the operation support devices 100A to 100C
according to the first to third embodiments will be
described. FIG. 15 is a diagram illustrating a hardware
30 configuration for implementing the operation support device
according to any of the first to third embodiments.
Because the operation support devices 100A to 100C have a
similar hardware configuration, the hardware configuration
41
of the operation support device 100C will be described here.
[0111] The operation support device 100C can be
implemented by a control circuit 300 illustrated in FIG. 15,
that is, a processor 301, a memory 302, an input unit 303,
and a display unit 304. The processor 5 301 is a central
processing unit (CPU, also referred to as a central
processing device, a processing device, a computation
device, a microprocessor, a microcomputer, or a digital
signal processor (DSP)), a system large scale integration
10 (LSI), or the like. The memory 302 is a random access
memory (RAM), a read only memory (ROM), or the like.
[0112] The memory 302 stores a program for executing the
function of the computation unit 1, a program for executing
the function of the reference value adjustment unit 20C,
15 and a program for executing the function of the facility
control unit 21.
[0113] The processor 301 receives necessary information
via the input unit 303, and reads and executes the programs
stored in the memory 302, thereby executing processing by
20 the computation unit 1, the reference value adjustment unit
20C, and the facility control unit 21. It can be said that
the programs stored in the memory 302 cause a computer to
execute a plurality of instructions corresponding to the
procedures or methods of the computation unit 1, the
25 reference value adjustment unit 20C, and the facility
control unit 21. The memory 302 is also used as a
temporary memory when the processor 301 performs various
processes.
[0114] The sewage information input unit 2, the
30 accumulation model input unit 3, the efficiency model input
unit 4, the cost input unit 5, the power consumption model
input unit 6, and the membrane maintenance condition input
unit 19 are implemented using the input unit 303. The
42
result presentation unit 7 is implemented using the display
unit 304.
[0115] The programs executed by the processor 301 may be
a computer program product having a computer-readable nontransitory
recording medium including 5 a plurality of
computer-executable instructions for performing data
processing.
[0116] Note that the processor 301 and the memory 302
illustrated in FIG. 15 may be replaced with processing
10 circuitry. For example, processing circuitry is a single
circuit, a composite circuit, a programmed processor, a
parallel programmed processor, an application specific
integrated circuit (ASIC), a field-programmable gate array
(FPGA), or a combination thereof. Note that the functions
15 of the computation unit 1, the reference value adjustment
unit 20C, and the facility control unit 21 may be partially
implemented by dedicated hardware and partially implemented
by software or firmware.
[0117] Fourth Embodiment.
20 Next, the fourth embodiment of the present invention
will be described with reference to FIGS. 16 to 21. In the
fourth embodiment, reference values are determined before
the start of the water treatment operation (design stage)
or during the operation of the water treatment facility.
25 That is, the fourth embodiment relates to the case in which
the reference value for determining the cleaning schedule
(hereinafter referred to as the reference value for
cleaning) and the reference value for determining the
replacement schedule (hereinafter referred to as the
30 reference value for replacement) for separation membranes
for a newly constructed water treatment facility are
determined in the design stage before the start of the
water treatment operation, or the case in which the
43
reference value for cleaning and the reference value for
replacement are redetermined for the water treatment
facility in operation. For a newly constructed water
treatment facility, the reference value for cleaning and
the reference value for replacement are 5 newly determined.
[0118] Furthermore, regarding the water treatment
facility in operation, conditions for water treatment in
the initial operation of the water treatment facility may
change due to changes in the water quality of the sewage to
10 be treated or changes in the environment of the water
treatment facility. Therefore, for the water treatment
facility in operation, the reference value for cleaning and
the reference value for replacement are also updated
appropriately.
15 [0119] Note that the timing of updating may be any
timing such as when each condition such as the condition of
sewage changes or when a certain period of time elapses
after the previous calculation. The procedure of the
present embodiment is repeatedly applied every time the
20 reference value for cleaning and the reference value for
replacement need to be recalculated during operation, so
that the reference value for cleaning and the reference
value for replacement suitable for the state of the water
treatment facility can be continuously adopted.
25 [0120] FIG. 16 is a functional configuration diagram of
an operation support device according to the fourth
embodiment. The operation support device 100D is a
computer that calculates a first reference value (reference
value for cleaning) used for determining whether to clean a
30 separation membrane for water treatment and a second
reference value (reference value for replacement) used for
determining whether to replace a separation membrane.
[0121] The operation support device 100D includes the
44
computation unit 1, the sewage information input unit 2,
the accumulation model input unit 3, the efficiency model
input unit 4, the cost input unit 5, the power consumption
model input unit 6, and the result presentation unit 7.
[0122] The sewage information 5 input unit 2, the
accumulation model input unit 3, the efficiency model input
unit 4, the cost input unit 5, and the power consumption
model input unit 6 acquire data and input data to the
computation unit 1. Each of these input units may acquire
10 data by any means. Each input unit acquires data by manual
data input, data input from external software, input of
data acquired from a sensor when the water treatment
facility is in operation, or the like. Each input unit may
input data to the computation unit 1 by any means such as
15 network communication or media transfer.
[0123] The sewage information input unit 2 acquires the
predicted water quality value D1 that is water quality
information on the sewage to be subjected to water
treatment, and inputs the predicted water quality value D1
20 to the computation unit 1. The predicted water quality
value D1 of the present embodiment is data indicating a
time-series change of the water quality of sewage or data
indicating a predicted value of the water quality of sewage.
[0124] The accumulation model input unit 3 acquires the
25 accumulation model D2 that expresses the amount of foulant
accumulation with respect to the water quality, and inputs
the accumulation model D2 to the computation unit 1. The
accumulation model D2 is a model that expresses the amount
of foulant accumulation caused by fouling.
30 [0125] The efficiency model input unit 4 acquires the
cleaning efficiency model D5 that is a model of cleaning
efficiency, and inputs the cleaning efficiency model D5 to
the computation unit 1. The cleaning efficiency model D5
45
is a model that expresses a change of the deterioration of
a separation membrane relative to the number of cleaning
events. In the membrane filtration method, a separation
membrane is clogged with foulant through continuous use of
the separation membrane, and thus 5 the efficiency of
filtration treatment (the amount of water that can be
filtered in a unit time) decreases as the treatment time
elapses. In the membrane filtration method, it is
necessary to clean and replace separation membranes on a
10 regular basis in order to remove foulant from membranes.
[0126] Generally, the cleaning cost for separation
membranes is lower than the replacement cost for separation
membranes. Therefore, the first step for dealing with
clogging of a separation membrane is to clean the
15 separation membrane. In the cleaning of a separation
membrane, the separation membrane is cleaned with a
chemical solution or filtrate so that foulant is removed.
After the separation membrane is cleaned, the filtration
capacity of the separation membrane (the amount of water
20 that the separation membrane can treat) is recovered to
some extent.
[0127] However, compared with the filtration capacity of
a new separation membrane, the filtration capacity and
durability of the cleaned separation membrane are low. As
25 the number of times the separation membrane is cleaned
increases, the recovery rate of the filtration capacity
decreases. When the recovery rate of the filtration
capacity becomes so low that the separation membrane can no
longer be reused, the separation membrane is replaced. The
30 cleaning efficiency model D5 of the present embodiment
expresses the recovery rate of the filtration capacity
relative to the number of times the separation membrane is
cleaned.
46
[0128] The cost input unit 5 acquires the maintenance
cost D6, and inputs the maintenance cost D6 to the
computation unit 1. The maintenance cost D6 is the cost
required for the maintenance of separation membranes, and
includes the cleaning cost, namely the 5 cost required for
one cleaning event for a separation membrane, and the
replacement cost, namely the cost required for one
replacement event for a separation membrane. In the
present embodiment, as described above, the cleaning cost
10 includes the price of the chemical solution used for
cleaning, and the replacement cost includes the price of
separation membranes.
[0129] The power consumption model input unit 6 acquires
the power consumption model D7, and inputs the power
15 consumption model D7 to the computation unit 1. The power
consumption model D7 is a model that expresses the power
consumption of the water treatment facility with respect to
the blockage rate of a separation membrane. In the present
embodiment, one or more of the accumulation model D2, the
20 cleaning efficiency model D5, and the power consumption
model D7 may also be referred to as a model.
[0130] The computation unit 1 computes a reference value
used for calculating the maintenance schedule for
separation membranes (cleaning schedule and replacement
25 schedule for separation membranes). The reference value
can be, for example, a value of the blockage rate of a
separation membrane and the period until the blockage rate
reaches that value. Specific examples of the use of the
reference value are to clean a separation membrane when the
30 blockage rate thereof reaches A% and to replace a
separation membrane when the time until the blockage rate
thereof reaches A% is less than B months. The following
description is based on the premise that the reference
47
value is a value of the blockage rate and the period until
the blockage rate reaches that value.
[0131] The computation unit 1 determines the reference
value for cleaning and the reference value for replacement
such that the operation cost, i.5 e. the sum of the
maintenance cost D6 and the power cost for water treatment
calculated using the predicted water quality value D1, the
accumulation model D2, the cleaning efficiency model D5,
and the power consumption model D7, has the minimum value,
10 and calculates the cleaning schedule and replacement
schedule for separation membranes using the determined
reference values. The computation unit 1 includes the
accumulation amount calculation unit 8 and the reference
value calculation unit 9, and outputs computation results
15 to the result presentation unit 7.
[0132] The accumulation amount calculation unit 8 uses
the predicted water quality value D1 and the accumulation
model D2 to calculate the time-series change D3 of the
fouling accumulation amount that occurs thereafter. The
20 time-series change D3 of the fouling accumulation amount is
data indicating a temporal change in the fouling
accumulation amount, which is the amount of foulant
accumulation. The fouling accumulation amount corresponds
to the blockage rate of a separation membrane. Therefore,
25 the time-series change D3 of the fouling accumulation
amount corresponds to the time-series change of the
blockage rate of a separation membrane. The accumulation
amount calculation unit 8 sends the time-series change D3
of the fouling accumulation amount to the reference value
30 calculation unit 9.
[0133] The reference value calculation unit 9 includes
the plan calculation unit 13 and the pattern comparison
unit 14. The plan calculation unit 13 calculates the plan
48
set D4 that expresses the schedule for cleaning and
replacement. The plan calculation unit 13 first creates a
plurality of temporary reference values for cleaning as
candidates for the reference value for cleaning. Temporary
reference values for cleaning are temporary 5 values of the
reference value for cleaning. The reference value
calculation unit 9 further creates a plurality of temporary
reference values for replacement as candidates for the
reference value for replacement. Temporary reference
10 values for replacement are temporary values of the
reference value for replacement. Then, the reference value
calculation unit 9 creates a set of temporary reference
value pairs including all these combinations. That is, the
reference value calculation unit 9 creates a set of
15 temporary reference value pairs that are all combinations
of temporary reference values for cleaning and temporary
reference values for replacement.
[0134] The reference value calculation unit 9 creates a
set of temporary reference value pairs as a temporary
20 reference value pattern set Px (wash, change). Hereinafter,
the set of temporary reference value pairs created by the
reference value calculation unit 9 may be referred to as
the temporary reference value pattern set Px. For example,
suppose that the criterion for cleaning is the blockage
25 rate of a separation membrane, temporary reference values
for cleaning are three types of A%, B%, and C%, the
criterion for replacement is the blockage rate of a
separation membrane after cleaning, and reference values
for replacement are three types of a%, b%, and c%. In this
30 case, the temporary reference value pattern set Px includes
nine elements of (A%, a%), (A%, b%), (A%, c%), (B%, a%),
(B%, b%), (B%, c%), (C%, a%), (C%, b%), and (C%, c%).
[0135] Then, the plan calculation unit 13 calculates a
49
separation membrane operation plan for each element of the
temporary reference value pattern set Px using the timeseries
change D3 of the fouling accumulation amount and the
cleaning efficiency model D5. In the present embodiment, a
set of operation plans for temporary 5 reference value
patterns is referred to as the plan set D4. The plan set
D4 is maintenance plans that contain the cleaning schedule
and replacement schedule for separation membranes. The
plan calculation unit 13 sends the calculated plan set D4
10 to the pattern comparison unit 14. Hereinafter, elements
of temporary reference value patterns may be referred to as
patterns.
[0136] The pattern comparison unit 14 uses the plan set
D4, the maintenance cost D6, and the power consumption
15 model D7 to calculate the operation cost for each plan.
Because the plans in the plan set D4 differ in the number
of cleaning events, the number of replacement events, or
the change of the blockage rate of a separation membrane in
a certain operation period, the operation costs differ
20 between the plans. Here, a certain operation period means
the period for calculating an operation cost, and the
reliability of a calculation result improves as the
operation period becomes longer.
[0137] In other words, if the period for comparing
25 operation costs is too short, a non-optimal reference value
pattern may be adopted. Therefore, the pattern comparison
unit 14 compares operation costs for reference value
patterns in a period including at least one separation
membrane replacement event after starting the use of a
30 separation membrane or in a period including at least two
separation membrane replacement events.
[0138] In other words, the pattern comparison unit 14
compares operation costs in a period longer than the
50
replacement cycle for separation membranes. Then, the
pattern comparison unit 14 compares the calculated
operation costs for the plans and determines the minimum
operation cost. The pattern comparison unit 14 extracts,
from the operation cost calculation results, 5 the pair of
temporary reference values used for the calculation of the
minimum operation cost plan, and sends the pair of
temporary reference values to the result presentation unit
7 as the combination of the reference value for cleaning
10 and the reference value for replacement.
[0139] The result presentation unit 7 presents the user
with the combination of the reference value for cleaning
and the reference value for replacement that achieves the
minimum operation cost. The result presentation unit 7 may
15 present the user with the operation cost itself. Further,
the result presentation unit 7 may present the user with
the operation plan that uses the reference value for
cleaning and the reference value for replacement. The
presentation means by the result presentation unit 7 may be
20 any means such as displaying on a display or printing on
paper.
[0140] In the fourth embodiment, the result presentation
unit 7 displays information similar to that in the first
embodiment. Specifically, the result presentation unit 7
25 displays a reference value for cleaning such as “when the
membrane is XX% blocked” as the cleaning schedule for
separation membranes. The result presentation unit 7 also
displays a reference value for replacement such as “when
the transmembrane pressure difference after cleaning
30 reaches YY% or more” as the replacement schedule for
separation membranes.
[0141] The result presentation unit 7 may display the
calculation result in the form of converted values
51
representing information to be presented to the operator of
the water treatment facility. In this case, the pattern
comparison unit 14 converts the calculation result into
information to be presented to the operator of the water
treatment facility. Information to be 5 presented to the
operator of the water treatment facility is, for example,
the date of the next cleaning or replacement event and the
operation cost for the case of executing the separation
membrane operation plan.
10 [0142] For displaying the date of the next and
subsequent cleaning or replacement events, the result
presentation unit 7 displays the schedule of cleaning ahead
such as “Za days later, Zb days later, and Zc days later”,
and the schedule of replacement ahead such as “Zd days
15 later, Ze days later, and Zf days later”.
[0143] Next, each calculation performed by the
accumulation amount calculation unit 8, the plan
calculation unit 13, and the pattern comparison unit 14
will be described using a specific example. The
20 accumulation amount calculation unit 8 calculates the timeseries
change D3 of the fouling accumulation amount. The
fouling accumulation amount is almost proportional to the
inflow load of sewage.
[0144] Therefore, the accumulation amount calculation
25 unit 8 can use a proportional expression as the
accumulation model D2, for example. Therefore, in a case
where the predicted water quality value D1 of the input
load is Xt and a proportional expression is used as the
accumulation model D2, the time-series change D3 of the
30 fouling accumulation amount V(t) that is calculated by the
accumulation amount calculation unit 8 is expressed as
Formula (1) described in the first embodiment.
[0145] Instead of a proportional expression such as
52
Formula (1) that is used here for simplicity, Ruth's
equation represented by Formula (3) in the first embodiment
or the like may be used for the accumulation model D2. In
addition, prediction data for water quality changes may be
used as the predicted water 5 quality value D1.
[0146] The plan calculation unit 13 calculates the plan
set D4 for separation membranes using Formula (1), which is
the time-series change D3 of the accumulation amount, the
temporary reference value pattern set Px, and the cleaning
10 efficiency model D5.
[0147] An example of the temporary reference value
pattern set Px is illustrated in FIG. 17. FIG. 17 is a
diagram illustrating an example of a temporary reference
value pattern set used in the operation support device
15 according to the fourth embodiment. FIG. 17 depicts the
temporary reference value pattern set Px for the case that
the blockage rate of a separation membrane is adopted as
the criterion for cleaning and the blockage rate of a
separation membrane after cleaning is adopted as the
20 criterion for replacement. As illustrated in FIG. 17, the
temporary reference value pattern set Px includes nine
temporary reference value patterns Ax to Ix.
[0148] The cleaning efficiency model D5 is a model
similar to the cleaning efficiency model D5 described in
25 the first embodiment. That is, the cleaning efficiency
model D5 is a model that expresses the phenomenon that a
separation membrane deteriorates due to repeated cleaning.
For example, in a case where a model in which the
efficiency of separation membrane cleaning decreases
30 exponentially with respect to the number of cleaning events
n for the separation membrane is used as the separation
membrane cleaning efficiency model D5, the efficiency of
separation membrane cleaning Ref(n) can be expressed as
53
Formula (4) described in the first embodiment. In Formula
(4), “c” is the base of the exponential function.
[0149] An example of the plan set D4 calculated by the
plan calculation unit 13 is illustrated in FIG. 18. FIG.
18 is a diagram illustrating an example 5 of a time-series
change of the blockage rate calculated by the operation
support device according to the fourth embodiment. In
practice, operation plans, i.e. a total of nine operation
plans, are calculated for all elements of the temporary
10 reference value pattern set Px illustrated in FIG. 17.
However, FIG. 18 depicts only three patterns Ax, Ex, and Ix
for simplicity of the drawing. The horizontal axis of the
graph depicted in FIG. 18 represents the time that elapses
after the separation membrane starts to filter sewage, and
15 the vertical axis represents the blockage rate of the
separation membrane.
[0150] In FIG. 18, the change transition of the blockage
rate for pattern Ax is denoted by change transition PAx,
the change transition of the blockage rate for pattern Ex
20 is denoted by change transition PEx, and the change
transition of the blockage rate for pattern Ix is denoted
by change transition PIx. Focusing on the first cleaning
event, it can be seen that the cleaning in pattern Ax is
performed at a stage where the blockage rate is low, and
25 the cleaning in pattern Ix is performed at a stage where
the blockage rate is high.
[0151] The pattern comparison unit 14 uses the plan set
D4, the maintenance cost D6, and the power consumption
model D7 to calculate the operation cost in a certain
30 operation period for each plan included in the plan set D4.
The pattern comparison unit 14 determines the minimum
operation cost of the calculated operation costs, and
acquires the temporary reference value pattern
54
corresponding to the minimum operation cost.
[0152] The maintenance cost D6 is, for example, a
cleaning cost of Ay yen for one cleaning event and a
replacement cost of By yen for one replacement event. In a
case where cleaning is performed a times 5 and replacement is
performed b times within a certain operation period, the
maintenance cost Costwash,change(t) can be expressed as
Formula (6) below.
[0153]
[Math. 6]
... Formula (6)
10 [0154] In a case where a proportional expression with
respect to the blockage rate is used as the power
consumption model D7, the power cost Costw(t) for water
treatment can be expressed as Formula (7) below. Here, “d”
is a proportional constant.
15 [0155]
[Math. 7]
... Formula (7)
[0156] From these Formulas (6) and (7), the operation
cost for a temporary reference value pattern in a certain
operation period can be expressed as Formula (8) below.
[0157]
[Math. 8]
... Formula (8)
20 [0158] Here, for calculating an operation cost, the
pattern comparison unit 14 sets a certain operation period
as the calculation period used for calculating the
operation cost. In this case, a period sufficiently longer
than the replacement cycle for separation membranes is used
25 as described above.
[0159] In other words, the pattern comparison unit 14
       t
wash,change 0 Cost t Ay a By b
Cost t d Rt w  
Cost t  Cost t  Cost t  total wash ,change w  
55
sets the calculation period based on the replacement cycle
for separation membranes. The replacement cycle for
separation membranes is the period from the replacement
event for a separation membrane to the next replacement
event. As the calculation period, the 5 pattern comparison
unit 14 may use a period designated by an instruction from
the user, or may use an initial value set in advance.
[0160] Here, the difference in determination result
between the case of a short calculation period such as a
10 calculation period T1x (FIG. 18) and the case of a
calculation period T2x (FIG. 18) longer than the
calculation period T1x will be described. An example of
the calculation period T1x is 240 days, and an example of
the calculation period T2x is 300 days. The present
15 embodiment is based on the premise that the unit of periods
is “day”, but any unit of periods may be used such as “week”
or “month”.
[0161] FIG. 19 is a diagram illustrating the
relationship between calculation periods used by the
20 operation support device according to the fourth embodiment
to calculate pattern operation costs and calculation
results of pattern operation costs. The calculation
results of FIG. 19 contain the calculation results for
patterns Ax, Ex, and Ix in the case that the operation cost
25 calculation period is 240 days and the calculation results
for patterns Ax, Ex, and Ix in the case that the operation
cost calculation period is 300 days.
[0162] The calculation results here are “the number of
cleaning events”, “the number of replacement events”,
30 “electricity expense”, and “operation cost”. “The number
of cleaning events” is the number of times the separation
membrane is cleaned. “The number of replacement events” is
the number of times the separation membrane is replaced.
56
“Electricity expense” is the electricity expense for the
filtration treatment corresponding to the calculation
period. The calculation period T1x=240 days is shorter
than the longest replacement cycle in the compared patterns.
[0163] Specifically, “5 calculation period
T1x<(replacement cycle for the case of pattern Ix)” is
satisfied. Therefore, the determination results differ
between the case of the calculation period T1x=240 days and
the case of the calculation period T2x=300 days. Pattern
10 Ex has the minimum operation cost in the case of the
calculation period T1x=240 days, whereas pattern Ix has the
minimum operation cost in the case of the calculation
period T2x=300 days.
[0164] Because the reliability of calculation results
15 increases as the calculation period for operation costs
becomes longer, the reliability of the calculation results
in the case of the calculation period T2x=300 days is
higher than that in the case of the calculation period
T1x=240 days. That is, pattern Ix is considered to have
20 the minimum operation cost. Therefore, the pattern
comparison unit 14 does not adopt a calculation period,
such as the calculation period T1x=240 days, that makes the
determination results different from those with the
calculation period T2x=300 days.
25 [0165] The pattern comparison unit 14 uses, as the
calculation period for operation costs, a length that is at
least twice the longest replacement cycle in the patterns
Ax to Ix to be compared. Note that it is desirable to
extend the calculation period for operation costs and
30 recalculate operation costs repeatedly until the
determination results converge.
[0166] The pattern comparison unit 14 compares the
operation costs calculated using a sufficiently long
57
calculation period, and determines the temporary reference
value pattern that achieves the minimum operation cost.
[0167] Next, the procedure for computation processing by
the computation unit 1 will be described. FIG. 20 is a
flowchart illustrating the procedure 5 for computation
processing by the computation unit according to the fourth
embodiment. This procedure is the same as the procedure
for the case that the computation unit 1 is executed as a
program.
10 [0168] The operation support device 100D prepares the
pattern set P, which is a set of reference value patterns,
in advance before the computation unit 1 starts computation.
The pattern set P includes various reference values.
Further, the operation support device 100D determines the
15 initial value of the calculation period in advance.
[0169] The computation unit 1 acquires the predicted
water quality value D1, which is information on sewage, and
the accumulation model D2 (steps S21 and S22). The
computation unit 1 calculates the future time-series change
20 D3 of the fouling accumulation amount based on the
predicted water quality value D1 and the accumulation model
D2 (step S23).
[0170] After that, the computation unit 1 acquires the
separation membrane cleaning efficiency model D5, the
25 maintenance cost D6, and the power consumption model D7 for
the water treatment facility with respect to the blockage
rate of a separation membrane (steps S24, S25, and S26).
[0171] The computation unit 1 acquires the values of the
first reference value pattern from the pattern set P,
30 calculates a cleaning plan and a replacement plan that are
plans in the calculation period (step S27), and calculates
an operation cost based on the calculated cleaning plan and
replacement plan (step S28). Then, the computation unit 1
58
compares the value of the operation cost calculated in step
S28 with the minimum value of the operation cost in the
previous calculation results. The computation unit 1
determines whether the operation cost calculated in step
S28 has the minimum 5 value (step S29).
[0172] When the operation cost calculated in step S28
has the minimum value (step S29: Yes), the computation unit
1 updates and records the minimum value of the operation
cost and the reference value pattern used for the
10 calculation of the minimum operation cost (step S29-1).
Note that in the case of the initial calculation of an
operation cost, the value of the operation cost calculated
in step S28 is necessarily the minimum value. The
computation unit 1 performs step S30 after step S29-1.
15 [0173] When the operation cost calculated in step S28
does not have the minimum value (step S29: No), the
computation unit 1 determines whether there is a reference
value pattern in the pattern set P for which the
calculation of the operation cost has not been completed.
20 That is, the computation unit 1 determines whether the
comparison with the minimum value has been completed for
all the reference value patterns (step S30).
[0174] When there is a reference value pattern for which
the calculation has not been completed (step S30: No), the
25 computation unit 1 acquires one reference value pattern for
which the calculation has not been completed, and updates
the reference value pattern for calculation (step S30-1).
Then, the computation unit 1 uses the new reference value
pattern to perform steps S27 to S30 again.
30 [0175] The computation unit 1 calculates the operation
costs for all the reference value patterns, and repeats
steps S30-1 and S27 to S30 until the comparison between the
calculated operation costs and the minimum value is
59
completed. After the operation costs for all the reference
value patterns are calculated and compared with the minimum
value (step S30: Yes), the computation unit 1 determines
whether the reference value pattern that achieves the
minimum operation cost in the calculation 5 period for the
calculation of the current operation cost is the same as
the reference value pattern having the minimum operation
cost in the calculation of the previous operation cost.
[0176] The minimum reference value pattern derived in
10 the previous calculation is the minimum reference value
pattern calculated in the previous loop of steps S27 to S30,
and the reference value pattern derived in the current
calculation is the minimum reference value pattern
calculated in the current loop of steps S27 to S30.
15 [0177] When the current minimum reference value pattern
is different from the previous minimum reference value
pattern, the computation unit 1 updates the latest minimum
reference value pattern to the current minimum reference
value pattern. The computation unit 1 determines whether
20 the minimum reference value pattern has been updated (step
S31).
[0178] In response to determining that the minimum
reference value pattern has been updated (step S31: Yes),
that is, the current minimum reference value pattern is
25 different from the previous minimum reference value pattern,
the computation unit 1 updates the current calculation
period T to a new calculation period T′=(T+ΔT) (step S31-1).
Then, the computation unit 1 executes steps S27 to S31
using the new calculation period T'.
30 [0179] In response to determining that the minimum
reference value pattern has not been updated (step S31: No),
that is, the current minimum reference value pattern is the
same as the previous minimum reference value pattern, the
60
computation unit 1 determines that the minimum reference
value patterns have converged. Upon determining that the
minimum reference value patterns have converged, the
computation unit 1 sends the current or previous minimum
reference value pattern to the result presentation 5 unit 7.
Thus, the result presentation unit 7 displays the current
or previous minimum reference value pattern as a
calculation result (step S32).
[0180] FIG. 21 is a diagram illustrating another
10 configuration example of the operation support device
according to the fourth embodiment. FIG. 21 depicts the
functional configuration of an operation support device
100E. The operation support device 100E is applied, for
example, when redetermining the reference value for
15 cleaning and the reference value for replacement for the
water treatment facility in operation. The operation
support device 100E includes a maintenance plan comparison
unit 32 in addition to the components provided in the
operation support device 100D.
20 [0181] The operation support device 100E may or may not
include a maintenance plan input unit 33. The maintenance
plan input unit 33 receives the current maintenance plan
input by an external device or the user, and sends the
current maintenance plan to the maintenance plan comparison
25 unit 32.
[0182] The maintenance plan comparison unit 32 compares
the maintenance plan currently adopted by the water
treatment facility with a new maintenance plan calculated
under new conditions, and sends the maintenance plan with a
30 lower cost to the result presentation unit 7.
[0183] In a case where the maintenance plan currently
adopted by the water treatment facility is the maintenance
plan calculated by the operation support device 100E, the
61
maintenance plan comparison unit 32 receives information on
the current maintenance plan from the computation unit 1
(reference value calculation unit 9). In a case where the
maintenance plan currently adopted by the water treatment
facility is not the maintenance plan 5 calculated by the
operation support device 100E, the operation support device
100E uses the maintenance plan input unit 33. In this case,
the maintenance plan input unit 33 receives the currently
adopted maintenance plan from an external device or the
10 user, and inputs the currently adopted maintenance plan to
the maintenance plan comparison unit 32.
[0184] As described above, in the fourth embodiment, the
reference value pattern is determined such that the
operation cost, i.e. the sum of the power cost for
15 separation treatment, the cleaning cost for separation
membranes, and the replacement cost for separation
membranes during the operation of separation membranes, has
the minimum value. This makes it possible to easily
predict the cleaning schedule and replacement schedule for
20 separation membranes for reducing the operation cost of
separation membranes without the need for reference value
input based on specialized knowledge.
[0185] The configurations described in the abovementioned
embodiments indicate examples of the contents of
25 the present invention. The configurations can be combined
with another well-known technique, and some of the
configurations can be omitted or changed in a range not
departing from the gist of the present invention.
30 Reference Signs List
[0186] 1 computation unit; 2 sewage information input
unit; 3 accumulation model input unit; 4 efficiency model
input unit; 5 cost input unit; 6 power consumption model
62
input unit; 7 result presentation unit; 8 accumulation
amount calculation unit; 9 reference value calculation
unit; 13 plan calculation unit; 14 pattern comparison
unit; 19 membrane maintenance condition input unit; 20B,
20C reference value adjustment unit; 21 5 facility control
unit; 22 membrane maintenance facility; 22X, 25 blower;
23B, 23C water treatment facility; 30A pretreatment tank;
30B membrane filtration tank; 31 separation membrane;
100A to 100E operation support device; D1 predicted water
10 quality value; D2 accumulation model; D3 time-series
change; D4 plan set; D5 cleaning efficiency model; D6
maintenance cost; D7 power consumption model; W1 sewage.
63
We Claim :
1. An operation support device comprising
a computation unit to: calculate a cost of separation
treatment that depends on a deterioration of a separation
membrane using a cleaning efficiency model 5 that expresses a
deterioration of the separation membrane relative to the
number of cleaning events, the separation membrane being
configured to remove a removal target from sewage;
calculate an operation cost using the cost of separation
10 treatment, a cleaning cost for one cleaning event for the
separation membrane, and a replacement cost for one
replacement event for the separation membrane, the
operation cost being a sum of the cost of separation
treatment, a cleaning cost for the separation membrane, and
15 a replacement cost for the separation membrane in an
operation period of the separation membrane; determine a
first reference value and a second reference value based on
the operation cost, the first reference value being used
for determining whether to clean the separation membrane,
20 the second reference value being used for determining
whether to replace the separation membrane; and calculate a
cleaning schedule and a replacement schedule for the
separation membrane using the first reference value and the
second reference value determined.
25
2. The operation support device according to claim 1,
wherein
the computation unit
calculates the operation cost using:
30 water quality information on the sewage including a
concentration of the removal target and a flow rate;
an accumulation model that expresses an amount of
foulant accumulation on a membrane surface of the
64
separation membrane;
the cleaning efficiency model;
a maintenance cost including the cleaning cost for one
cleaning event and the replacement cost for one replacement
5 event; and
a power consumption model that expresses a power
consumption that depends on a blockage rate of the
separation membrane in a water treatment facility that
performs the separation treatment.
10
3. The operation support device according to claim 2,
wherein
the computation unit
changes an operating condition of a facility or device
15 for maintaining a state of the separation membrane such
that the operating condition satisfies a maintenance
condition that is a condition for maintenance of the
separation membrane, updates at least one of the
accumulation model, the cleaning efficiency model, the
20 maintenance cost, and the power consumption model based on
the operating condition changed, and calculates the
cleaning schedule and the replacement schedule.
4. The operation support device according to claim 3,
25 wherein
the computation unit
changes an operating condition of a facility or device
that is used for treatment at a pre-stage of the separation
treatment such that the operating condition satisfies a
30 maintenance condition that is a condition for maintenance
of the separation membrane, updates at least one of the
water quality information, the accumulation model, and the
power consumption model based on the operating condition
65
changed, and calculates the cleaning schedule and the
replacement schedule.
5. The operation support device according to any one of
claims 5 1 to 4, wherein
the computation unit
calculates the operation cost in the operation period
by calculating a time-series change of the blockage rate of
a membrane using the first reference value and the second
10 reference value and calculating the cost of separation
treatment based on the time-series change of the blockage
rate.
6. The operation support device according to any one of
15 claims 1 to 5, wherein
the computation unit
determines the first reference value and the second
reference value by calculating a plurality of the operation
costs in the operation period for combinations of a
20 plurality of first temporary reference values that are
temporary values of the first reference value and a
plurality of second temporary reference values that are
temporary values of the second reference value and
comparing the plurality of operation costs.
25
7. An operation support method comprising:
a calculation step of calculating a cost of separation
treatment that depends on a deterioration of a separation
membrane using a cleaning efficiency model that expresses a
30 deterioration of the separation membrane relative to the
number of cleaning events, the separation membrane being
configured to remove a removal target from sewage;
a reference value determination step of calculating an
operation cost using the cost of separation
cleaning cost for one cleaning event for the separation
membrane, and a replacement cost for one replacement event
for the separation membrane, the operation cost being a sum
of the cost of separation treatment, a 5 cleaning cost for
the separation membrane, and a replacement cost for the
separation membrane in an operation period of the
separation membrane, and determining a first reference
value and a second reference value based on the operation
10 cost, the first reference value being used for d
whether to clean the separation membrane, the second
reference value being used for determining whether to
replace the separation membrane; and
a schedule calculation step of calculating a cleaning
15 schedule and a replacement schedule for the se
membrane using the first reference value and the second
reference value determined.