U.S. patent application number 10/278886 was filed with the patent office on 2003-05-01 for operation support system and method.
Invention is credited to Harada, Yasushi, Sugiyama, Shigeru, Tomita, Yasushi.
Application Number | 20030083788 10/278886 |
Document ID | / |
Family ID | 19147425 |
Filed Date | 2003-05-01 |
United States Patent
Application |
20030083788 |
Kind Code |
A1 |
Harada, Yasushi ; et
al. |
May 1, 2003 |
Operation support system and method
Abstract
An operation support system is arranged to effectively provide
an operator of an energy supply facility for supplying a plurality
of energies in different forms with guide information of an
operating method in which energy safety supply and operation cost
reduction are both realized. The operation support system takes the
steps of deriving demand prediction upper limit value and lower
value for each energy form based on recorded demand data; deriving
target operation pattern upper and lower limit values of each
energy supply device based on the demand prediction upper and lower
values; and displaying the target operation pattern upper and lower
limit values.
Inventors: |
Harada, Yasushi; (Hitachi,
JP) ; Tomita, Yasushi; (Mito, JP) ; Sugiyama,
Shigeru; (Hitachi, JP) |
Correspondence
Address: |
McDermott, Will & Emery
600, 13th Street, N.W.
Washington
DC
20005-3096
US
|
Family ID: |
19147425 |
Appl. No.: |
10/278886 |
Filed: |
October 24, 2002 |
Current U.S.
Class: |
700/291 |
Current CPC
Class: |
F02D 2041/1433 20130101;
Y04S 10/50 20130101; G06Q 10/04 20130101; Y02E 20/14 20130101; G05B
15/02 20130101; H02J 3/00 20130101; F02D 2041/141 20130101; H02J
3/003 20200101; F02D 2041/1437 20130101 |
Class at
Publication: |
700/291 |
International
Class: |
G05D 011/00 |
Foreign Application Data
Date |
Code |
Application Number |
Oct 30, 2001 |
JP |
2001-331928 |
Claims
What is claimed is:
1. An operation support method for an energy supply facility having
a plurality of energy forms, comprising: deriving demand prediction
upper and lower limits for each energy form; calculating target
operation pattern upper and lower limits corresponding to each
energy form based on the demand prediction upper and lower limits;
and displaying the target operation pattern upper and lower
limits.
2. An operation support method of an energy supply facility having
a plurality of energy forms, comprising: deriving a demand
prediction probability density distribution for each energy form;
calculating target operation pattern probability distributions
corresponding to each energy form based on the demand prediction
probability density distribution; and displaying the target
operation pattern probability density distributions.
3. An operation support method as claimed in claim 2, further
comprising setting upper and lower limits of the target operation
pattern so that an integrated value of the target operation pattern
probability density distribution between the upper and the lower
limits of the target operation pattern may become a specified
value.
4. An operation support method as claimed in claim 2, wherein the
displaying shows a time, a point of the target operation pattern
corresponding to a peak of the target operation pattern probability
density distribution, and the target operation pattern probability
density distribution in three-dimensional axes.
5. An operation support method as claimed in claim 1, further
comprising generating an alarm signal if the actual operation
pattern departs from a range between the upper and the lower limits
of the target operation pattern.
6. An operation support method as claimed in claim 2, further
comprising generating an alarm signal if the actual operation
pattern departs from a range between the upper and the lower limits
of the target operation pattern.
7. An operation support method according to claim 1, wherein the
deriving step includes retrieving past demand data having matched
conditions for an object of the demand prediction, and deriving
demand prediction upper and lower limits of each energy form based
on retrieved past demand data.
8. An operation support method according to claim 2, wherein the
deriving step includes retrieving past demand data having matched
conditions for an object of the demand prediction, and deriving
demand prediction upper and lower limits of each energy form based
on retrieved past demand data.
9. An operation support method according to claim 7, wherein the
calculating step includes extracting sample data from retrieved
past demand data between the demand prediction upper and lower
limits of each energy form, deriving optional operation pattern
corresponding to the each sample data, and assigning maximum value
and minimum value from the optional operation pattern to the target
operation.
10. An operation support method according to claim 8, wherein the
calculating step includes extracting sample data from retrieved
past demand data between the demand prediction upper and lower
limits of each energy form, deriving optional operation pattern
corresponding to the each sample data, and assigning maximum value
and minimum value from the optional operation pattern to the target
operation.
11. An operation support method of an energy supply facility having
a plurality of energy forms, comprising: deriving demand prediction
upper and lower limits of each energy form; deriving a demand
prediction probability density distribution of each energy form;
calculating target operation pattern probability distributions
corresponding with each energy form based on the upper and lower
limits and the demand prediction probability density distribution,
setting upper and lower limits of the target operation pattern
based on the target operation pattern probability distributions and
a specified probability value; and displaying the target operation
pattern probability density distributions including the upper and
lower limits of the target operation pattern probability density
distributions.
12. An operation support method according to claim 1, wherein the
plurality of energy forms includes a gas turbine generating
electric power and steam, a steam turbine generating electric power
by using steam, an absorption refrigerator supplying chilled water
by using steam, and a gas boiler generating steam.
13. An operation support method according to claim 2, wherein the
plurality of energy forms includes a gas turbine generating
electric power and steam, a steam turbine generating electric power
by using steam, an absorption refrigerator supplying chilled water
by using steam, and a gas boiler generating steam.
14. An operation support method according to claim 11, wherein the
plurality of energy forms includes a gas turbine generating
electric power and steam, a steam turbine generating electric power
by using steam, an absorption refrigerator supplying chilled water
by using steam, and a gas boiler generating steam.
15. An operation support method according to claim 12, wherein, the
deriving step includes retrieving past demand data having matched
conditions for an object of the demand prediction, the past demand
data including past demand of electric power, steam, and chilled
water with at least one of day, weather, and temperature, and
calculating each average demand value and standard deviation of
electric power, steam, and chilled water based on retrieved past
demand data, and deriving demand prediction of upper and lower
limits of electric power, steam, and chilled water, the calculating
step includes calculating target operation pattern upper and lower
limits of the gas turbine, steam turbine, absorption refrigerator,
and gas boiler corresponding to the derived demand prediction of
upper and lower limits of electric power, steam, and chilled
water.
16. An operation support method according to claim 13, wherein, the
deriving step includes retrieving past demand data having matched
conditions for an object of the demand prediction, the past demand
data including past demand of electric power, steam, and chilled
water with at least one of day, weather, and temperature, and
calculating each average demand value and standard deviation of
electric power, steam, and chilled water based on retrieved past
demand data, and deriving demand prediction of upper and lower
limits of electric power, steam, and chilled water, the calculating
step includes calculating target operation pattern upper and lower
limits of the gas turbine, steam turbine, absorption refrigerator,
and gas boiler corresponding to the derived demand prediction of
upper and lower limits of electric power, steam, and chilled
water.
17. An operation support method according to claim 14, wherein, the
deriving step includes retrieving past demand data having matched
conditions for an object of the demand prediction, the past demand
data including past demand of electric power, steam, and chilled
water with at least one of day, weather, and temperature, and
calculating each average demand value and standard deviation of
electric power, steam, and chilled water based on retrieved past
demand data, and deriving demand prediction of upper and lower
limits of electric power, steam, and chilled water, the calculating
step includes calculating target operation pattern upper and lower
limits of the gas turbine, steam turbine, absorption refrigerator,
and gas boiler corresponding to the derived demand prediction of
upper and lower limits of electric power, steam, and chilled
water.
18. An operation support system for supporting operation of an
energy supply facility having a plurality of energy forms,
comprising: a data storage to store demand data of past demand for
each energy forms; a processor to perform demand prediction process
and target operation pattern calculation process; a display
displaying target operation pattern calculated by the processor; a
program executed by the processor, wherein execution of the program
by the processor causes the processor to implement a series of
steps, comprising: deriving demand prediction upper and lower
limits for each energy form based on the demand data stored in the
data storage; and calculating target operation pattern upper and
lower limits corresponding to each energy form based on the demand
prediction upper and lower limits.
19. An operation support system for supporting operation of an
energy supply facility having a plurality of energy forms,
comprising: a data storage to store demand data of past demand for
each energy forms; a processor to perform demand prediction process
and target operation pattern calculation process; a display
displaying target operation pattern probability density
distributions target operation pattern probability density
distributions calculated by the processor; a program executed by
the processor, wherein execution of the program by the processor
causes the processor to implement a series of steps, comprising:
deriving a demand prediction probability density distribution for
each energy form based on the demand data stored in the data
storage; and calculating target operation pattern probability
distributions corresponding to each energy form based on the demand
prediction probability density distribution.
20. A software product for operation support system having a data
storage, a processor, and a display, the product comprising: at
least one processor readable medium; programming code, carried by
the at least one medium, for execution by the processor, wherein
execution of the programming code by the processor causes the
system to implement a series of steps, comprising: deriving demand
prediction upper and lower limits for each energy form based on the
data stored in data storage; calculating target operation pattern
upper and lower limits corresponding to each energy form based on
the demand prediction upper and lower limits; and displaying the
target operation pattern upper and lower limits.
21. A software product for operation support system having a data
storage, a processor, and a display, the product comprising: at
least one processor readable medium; programming code, carried by
the at least one medium, for execution by the processor, wherein
execution of the programming code by the processor causes the
system to implement a series of steps, comprising: deriving a
demand prediction probability density distribution for each energy
form based on the data stored in data storage; calculating target
operation pattern probability distributions corresponding to each
energy form based on the demand prediction probability density
distribution; and displaying target operation pattern probability
density distributions.
22. A cogeneration energy supply system comprising: two or more
energy sources for generating energy; and, an operation support
system comprising: data storage to store demand data of past demand
for each energy sources; a processor to perform demand prediction
process and target operation pattern calculation process; and a
display displaying target operation pattern calculated by the
processor; wherein, the processor derives demand prediction upper
and lower limits for each energy sources based on the demand data
stored in the data storage and calculates target operation pattern
upper and lower limits corresponding to each energy sources based
on the demand prediction upper and lower limits.
23. A cogeneration energy supply system comprising: two or more
energy sources for generating energy; and, an operation support
system, comprising: data storage to store demand data of past
demand for each energy sources; a processor to perform demand
prediction process and target operation pattern calculation
process; and, a display displaying target operation pattern
probability density distributions target operation pattern
probability density distributions calculated by the processor;
wherein the processor derives a demand prediction probability
density distribution for each energy source based on the demand
data stored in the data storage; and calculates target operation
pattern probability distributions corresponding to each energy
source based on the demand prediction probability density
distribution.
24. A cogeneration energy supply system according to claim 22,
wherein each energy sources are selected from the group the group
consisting essentially of a gas turbine generating electric power
and steam, a steam turbine generating electric power by using
steam, an absorption refrigerator supplying chilled water by using
steam, and a gas boiler generating steam.
25. A cogeneration energy supply system according to claim 23,
wherein each energy sources are selected from the group the group
consisting essentially of a gas turbine generating electric power
and steam, a steam turbine generating electric power by using
steam, an absorption refrigerator supplying chilled water by using
steam, and a gas boiler generating steam.
Description
BACKGROUND OF THE INVENTION
[0001] The present invention relates to an operation support system
for a facility of supplying a plurality of energy forms such as
electric power and heat, and more particularly to the system which
properly supports the operation of the energy supply facility if
change of demands cannot be precisely predicted.
[0002] As a typical example of an electric power or heat energy
supply facility, there have been proposed a gas turbine (that
supplies electric power and exhaust heat vapor by gas), a steam
turbine (that supplies electric power by steam), a steam absorption
refrigerator (that supplies chilled water by steam), an electric
turbo refrigerator (that supplies chilled water by electric power),
and so forth. The energy supply facility composed of those devices
are difficult to be properly operated on account of the following
two reasons. First, the energy supply facility is arranged to keep
those devices closely related with one another in such a manner
that the gas turbine and the steam turbine are complemented with
each other in respect of the electric power supply and the steam
absorption refrigerator collects exhaust heat from the steam
turbine. Second, the energy demand may fluctuate depending on
variables such as a weather and a temperature or any obscure
cause.
[0003] In order to overcome these difficulties, the two background
arts have been proposed. The first background art is a method of
patterning an operation method (such as a start and stop time and a
number of operating devices) and operating the facility according
to the pattern. The concrete example of the target operation
pattern is described in the page 6 of "'99 Collection of Energy
Saving Cases (first volume) edited by the Energy Saving Center
(Incorporated Foundation), pages 786 to 787 of "'99 Collection of
Energy Saving Cases (second volume)" edited by the Energy Saving
Center (Incorporated Foundation), and page 53 of "Energy Saving
Diagnosis and Concrete Measures" edited by NTS (Limited). By
predetermining such an allowable target operation pattern as coping
with some demand fluctuations, the operation enables to operate the
energy supply facility with safety according to the predetermined
operation pattern, which makes the safety supply of energy
possible.
[0004] The second background art is described in JP-A-11-346438.
The method of the background art is arranged so that for supplying
an operator with a future predicted value of an electric power
demand, the electric power demand is predicted at each interval as
estimating a prediction error and are displayed. The application of
this method makes it possible for an operation to know a
prediction-deviating range or possibility and thereby to operate
the energy supply facility according to the range or possibility.
For example, if a future power demand may greatly exceed the
prediction value, the application of this background art makes it
possible for the operator to easily determine that one more
generator than usual needs to be started. Or, if the future power
demand may substantially match to the prediction value, the
application of this background art makes it possible for the
operator to easily determine that a reserve of energy may be
partially reduced.
[0005] The first background art has concerned with the method of
realizing the safety operation of the energy supply facility. This
art may, however, burdens itself with the additional operation
cost. For example, when both the electric power and the steam are
supplied by the gas turbine, assuming that the predetermined target
operation pattern of the gas turbine is at the "constant full
operation" and the gas turbine supplies more steam than needed
according to the target operation pattern, the surplus stream is
abandoned. It is certain that the energy may be supplied with
safety according to this target operation pattern, but the surplus
stream is abandoned. The operation cost thus may be increased more
than needed.
[0006] The second background art provides guide information of
keeping the electric power safety supply and the operation cost
reduction in prompting an electric power system in compatibility by
presenting the prediction value of the power demand with an error
range to an operator. However, unlike the promotion of the electric
power system, as to the electric power or heat energy supply
facility built in a factory or a building, since a plurality of
energy supply facilities are closely related with one another, even
if the demand prediction values of electric power, steam, heated
air and cool air may be presented to the operator together with
their error ranges, the operator cannot easily determine how to
cope with each prediction error with its error range. As one
example, consider the energy supply facility composed of a combined
cycle generator device and a heat recovery device, the combined
cycle generator device serving to drive a steam turbine by using
part of exhaust stream from the gas turbine and the heat recovery
device serving to utilize the remaining exhaust stream for a water
heater. Even if the gradual increase of the demand prediction
values of both the electric power and the water heater load is
quantitatively presented to the operator, the operator cannot
easily determine how to cope with the demand prediction value.
[0007] The reason of the difficulty in determining how to cope with
the prediction value is placed on a sophistication of an energy
supply facility configuration and function in using many kinds of
energies such as electric power, stream and chilled water. For
example, as one of the methods of coping with the increase of the
power demand, there may be proposed a method of raising a steam
turbine output. This method may bring about an adverse effect by a
shortage of a steam supply for a water heater because of increasing
the steam consumption of the steam turbine. Even if the steam
supply is not in short, at a certain unit price of the electricity
tariff, the cost may be lowered not by raising the output of the
steam turbine but by increasing the electricity bought from the
commercial electric power. As indicated by this example, the
operator cannot easily determine whether or not the manipulation of
raising the steam turbine output is in proper from both viewpoints
of safety supply and reduction of an operation cost.
SUMMARY OF THE INVENTION
[0008] This invention helps for an operator to keep two objects of
energy safety supply and operation cost reduction in compatibility
according to the demand continuously fluctuating on some variable
factors in a plurality of energy supply devices for supplying
different energy forms such as a cogeneration energy supply device
and in an energy supply system provided with a plurality of energy
demands.
[0009] To achieve above described object, an energy supply system
having a plurality of energy generating devices generating
different energy forms corresponding to a plurality of different
kind energy demands, displays a target operation pattern of each
energy generating device for each energy form with the upper and
the lower limits, the target operation pattern being based on the
upper and the lower limit values of the demand. For deriving the
upper and the lower limits of the target operation pattern, such
upper and lower limits of the target operation pattern as reducing
the operation cost as much as possible are derived in consideration
of the error range of the energy demand prediction caused by
variable factors. By displaying the upper and the lower limits of
the target operation pattern as guide information, the operator can
concretely obtain the operating policy of supplying energy with
safety and reducing an operation cost.
[0010] In particular, by displaying not the target operation
pattern but the upper and the lower limit of the target operation
pattern of each energy form, the operator clearly gets to know the
reliability of the target operation pattern. For example, if the
width between the upper and the lower limits of a certain target
operation pattern is narrow, the target operation pattern is less
likely to suffer from the adverse effect of the demand prediction
error. It means that the target operation pattern has a high
reliability. Conversely, if the width between the upper and the
lower limits of a certain target operation pattern is wide, the
target operation pattern is more likely to suffer from the adverse
effect of the demand prediction error. It means that the target
operation pattern has a low reliability.
[0011] As mentioned above, it is possible for the operator to get
to know all the operation policies including all the factors up to
the reliability of the target operation pattern. The reliability of
the target operation pattern, termed herein, indicates how much the
operator may follow the target operation pattern with fidelity. It
means that if the reliability is lower, the operator is not
required to follow the target operation pattern so much, while if
it is high, the operator should follow the target pattern with
fidelity.
[0012] Furthermore, the operation support system can be arranged to
display a target operation pattern probability density
distribution. Like displaying the upper and the lower limits of the
target operation pattern (the first method), the operator can
concretely obtain guide information on the operating method in
keeping the energy safety supply and the operation cost reduction
in compatibility. For example, if the peak of the density
distribution is low and the skirts thereof are wide, it means that
the width between the upper and the lower limits of the target
operation pattern is wide in the first method. Conversely, if the
peak of the target operation pattern probability density
distribution is high and the skirts thereof are narrow, it means
that the width between the upper and the lower limits of the target
operation pattern is narrow in the first method.
[0013] By presenting the upper and the lower limits of the target
operation pattern derived in consideration of the demands
fluctuating on variable factors to the operator of the energy
facility, the operator is given an effective support in keeping the
energy safety supply and the operation cost reduction in
compatibility.
[0014] Additional objects, advantages and novel features of the
embodiments will be set forth in part in the description which
follows, and in part will become apparent to those skilled in the
art upon examination of the following and the accompanying drawings
or may be learned by production or operation of the embodiments.
The objects and advantages of the inventive concepts may be
realized and attained by means of the methodologies,
instrumentalities and combinations particularly pointed out in the
appended claims.
BRIEF DESCRIPTION OF THE DRAWINGS
[0015] The drawing figures depict preferred embodiments by way of
example, not by way of limitations. In the figures, like reference
numerals refer to the same or similar elements.
[0016] FIG. 1 is an exemplary function block diagram showing an
operation support system according to a first embodiment;
[0017] FIG. 2 is an exemplary system block diagram showing an
operation support system according to the first embodiment;
[0018] FIG. 3 is an exemplary flowchart showing a process of the
operation support system;
[0019] FIG. 4 is an exemplary table showing data on demand records
for describing a procedure of a demand predicting module;
[0020] FIG. 5 is an explanatory table showing an example of
generating sample data in a target operation pattern calculating
module;
[0021] FIG. 6 is an explanatory table showing an example of an
optimal operation pattern calculated in correspondence with each
sample data unit;
[0022] FIG. 7 is an explanatory table for describing an upper and a
lower limits of a target operation pattern;
[0023] FIG. 8 is an exemplary function block diagram showing an
operation support system provided with a target upper and lower
limits setting module according to a second embodiment;
[0024] FIG. 9 is an explanatory table for describing an example of
a target operation pattern probability density function of a gas
boiler included in the second embodiment;
[0025] FIG. 10 is an explanatory chart for describing an example of
a three-dimensional representation provided by an operation pattern
display;
[0026] FIG. 11 is an exemplary function block diagram showing an
operation support system provided with an alarming module according
to a third embodiment; and
[0027] FIG. 12 is an exemplary diagram for describing a
configuration of a cogeneration energy supply facility to which the
operation support system.
DETAILED DESCRIPTION OF THE EMBODIMENTS
[0028] Hereafter, the description will be oriented to the
configuration of the operation support system according to the
invention. As an example of applying an operation support system
according to the invention, a summary of the configuration of the
cogeneration energy supply facility will be described with
reference to FIG. 12. The cogeneration supply facility supplies
each of different energy forms such as electric power, stream and
chilled water in association with the commercial electric power.
The supply facility is composed of a gas turbine GT operated by
town gas as energy supplying means, a gas boiler GB operated by
town gas, a steam turbine ST operated by steam, and an absorption
refrigerator AC operated to supply chilled water by steam.
[0029] The gas turbine GT generates electric power by driving a
generator as well as to generate steam by using waste heat. The gas
boiler GB generates steam. The steam turbine ST generates electric
power by using steam. The absorption refrigerator AC supplies
chilled water by using steam. The electric power from the gas
turbine GT and the gas boiler GB is supplied to the electric demand
in association with the commercial electric power. The steam from
the gas turbine GT and the gas boiler GB is partially supplied to
the steam turbine ST in which it is converted into the electric
power and partially supplied to the absorption refrigerator AC from
which the chilled water is supplied. The remains of the steam are
supplied to the steam demand. The chilled water generated by the
absorption refrigerator AC is supplied to the chilled water
demand.
[0030] Hereafter, the description will be oriented to the operation
support system. FIG. 1 is an exemplary function block diagram for
realizing the operation support system according to the first
embodiment. The operation support system is arranged to have a
demand predicting module 11, a target operation pattern calculating
module 12, and a target operation pattern display screen 13. The
demand predicting module 11 derives demand prediction upper limit
value 14 and lower limit value 15. The target operation pattern
calculating module 12 derives target operation pattern upper and
lower limit values 16 and 17. On the target operation pattern
display screen 13, the target operation pattern upper and lower
limit values 16 and 17 is displayed. This makes it possible for the
operator to determine an operating policy, that is to say a
practical causes of action as to of how each of the energy supply
devices composing the energy supply facility should be operated on
the basis of the upper and the lower limits of the displayed target
operation pattern.
[0031] For example, if the width between the upper and the lower
limits of the target operation pattern in a certain energy supply
device is narrow, the operator tries to fit the actual operation
pattern in the narrow range, while if the width is wide, the
operator can easily fit the actual operation pattern in the wide
range.
[0032] FIG. 2 shows an exemplary system block diagram of the
hardware for realizing the operation support system. The operation
support system is composed of a data storage 21, a processor 22,
and a display 23. These items may comprise types of devices
commonly used in various types of computers or other similar types
of programmable systems.
[0033] The data storage 21 stores the actual demand data 24 that
corresponds to the past demand records for each energy form. The
actual demand data 24 is used to the demand predicting module 11.
The data storage unit 21 may take the form of hard disk drive,
optical disk drive, a tape drive or other form of storage drive or
memory device, of a type typically formed in computers and the
like.
[0034] The processor 22 performs the operations required for both
of demand predicting module 11 and the target operation pattern
calculating module 12. In this embodiment, both demand predicting
module 11 and the target operation pattern calculating module 12
are implemented by a software module. For example, the processor 22
includes, CPU, RAM, ROM, etc. The software to implement the
processor functions may be stored in ROM, or other memory. For
example, the CPU performs these functions by executing the software
stored in memory. In this case, program memory is a processor
readable medium, but the term "processor readable medium" as used
herein refers to any medium that participates in providing
instructions and/or data to programmable processor (CPU) for
execution or other processing. Such a medium may take many form,
including but not limited to carrier waves and physical media for
transporting such waves, non-volatile storage media, and volatile
storage media. Non-volatile media include, for example, optical or
magnetic disks, such as ROM, or Hard disc. However, the functions
of the modules can be implemented by hardware modules with
proprietary circuits.
[0035] The demand predicting module 11 calculates the demand
prediction upper and lower limit values of each energy form through
the use of the actual demand prediction data 24. The target
operation pattern calculating module 12 calculates the target
operation pattern and its upper and lower limit values of each
energy supply device based on the demand prediction upper and lower
limit values of each energy form.
[0036] The display 23 has the target operation pattern display
screen 13 and displays the target operation pattern calculated by
the target operation pattern calculating module 12 on the target
operation pattern display screen 13.
[0037] The process of realizing the operation support system
according to the this embodiment will be described with reference
to the flowchart of FIG. 3. The operation support system executes
the processes of the demand prediction 31, the target operation
pattern calculation 32, and the target operation pattern display 33
in the describing sequence.
[0038] The demand prediction process 31 to be executed by the
demand predicting module 11 is executed to retrieve the recorded
demand data 24 for extracting the past recorded data that have the
same conditions (a weather, a day of the week and so forth) at the
target time point when the consideration is started, for preparing
the basic data required for predicting the demand (31a).
[0039] If the number of extracted data units is not enough, the
accuracy of the demand precision is made lower. Hence, it is
checked if the number of the extracted data units is enough (31b).
As the determining criterion of it, some threshold values may be
provided.
[0040] If the number of the extracted data units is not enough, the
retrieval condition is relaxed. Then, the actual demand data 24 is
again retrieved for extracting the past recorded data (31c). As the
method of relaxing the retrieval condition, for example, when the
predicted temperature is 25.degree. C., the data including
25.degree. C. as well as 24.degree. C. and 26.degree. C. are
extracted from the recorded temperatures. Or, when the demand on
Wednesday is predicted, the data including the recorded data on
Wednesday as well as the other days of the week are extracted from
the recorded data.
[0041] If a sufficient number of the recorded data units are
extracted, the demand prediction upper and lower limits of each
energy form are derived through the use of the statistical
technology based on the extracted data units (31d). As a method of
deriving the demand prediction upper and lower limits, it is
possible to utilize the method of assigning an average
value.+-.standard deviation to the upper and the lower limits of
the demand prediction. The foregoing process corresponds to the
content of the demand prediction process 31.
[0042] Then, the target operation pattern calculation process 31 to
be executed by the target operation pattern calculating module 12
prepares some sample data units about the demand value for each
energy form as the basic data for deriving the optimal operation
pattern (32a). In this embodiment, the demand prediction value has
some width (that is, the upper and the lower limits of the demand
prediction are set). In general, however, the optimal operation
pattern is different depending on a certain demand prediction value
between the upper and the lower limits. For deriving the optimal
operation pattern, it is necessary to extract some sample data
units between the upper and the lower limits of the demand
prediction. In addition, for a plurality of energy forms whose
demands are to be predicted, such as electric power, steam and
chilled water, the combination of the demand values for each energy
form is treated as sample data.
[0043] In turn, the optimal operation pattern for each energy
supply device is derived by using each sample data unit (32b). The
optimal operation pattern, termed herein, means the operation to be
executed at a minimum cost. However, the concept of this embodiment
may be also applied to the operation at a minimum C02 exhaust or
the operation at a minimum energy consumption converted into oil.
The method of deriving the optimal operation pattern may be
obtained by applying the technology described in ITO Kouichi and
YOKOYAMA Ryouhei "Optimal Planning of Cogeneration", pages 57 to
63, Sangyou-Tosho (1990). The application of this method makes it
possible to obtain the operation pattern at a minimum cost in the
constraint condition that energy can be positively supplied in
response to the demand. The method of deriving the optimal
operation pattern is not limited to the foregoing method. In place,
the experimental operating method or the operating rules based on
the experimental knowledge of an operator may be applied to the
method.
[0044] The maximum value and the minimum value of the optimal
operation pattern for each energy supply device is assigned to the
upper limit value and the lower limit value of the target operation
pattern, respectively (32c). The foregoing process is the content
of the target operation pattern calculating process 32.
[0045] Lastly, the target operation pattern displaying process to
be executed by the view display 23 displays the upper and the lower
limits of the target operation pattern on the target operation
pattern displaying screen 13 (33a).
[0046] The foregoing series of processes make it possible for the
operator to view the target operation pattern with the upper and
the lower limits of each energy supply device, thereby allowing the
operator to obtain guide information how to operate the energy
supply device as keeping the energy safety supply and the operation
cost reduction in compatibility.
[0047] The exemplary composition of the recorded demand data 24 and
the exemplary method of obtaining the upper and the lower values of
the demand prediction will be described with reference to FIG. 4.
In FIG. 4, a numeral 4a denotes an example of the recorded demand
data 24 to be used when the demand predicting module 11 derives the
demand prediction upper limit value 14 and lower limit value 15. In
the example of FIG. 4, the recorded demand data includes "year",
"month", "day", "day of the week", "time", "weather",
"temperature", "electric power demand", "stream demand", and
"chilled water demand" as data items. Hereafter, the description
will be oriented to the method of deriving the upper and the lower
limit values of the demand prediction with reference to the
data.
[0048] First, a process is executed to extract the recorded data
matching to the demand-prediction condition from all the data
items. For example, when the demands for an electric power, steam
and chilled water amounts are to be predicted at 1500 hours,
Tuesday, Jun. 26, 2001, in a cloudy weather, and at a temperature
of 25.degree. C., the data is extracted from all the data items
with the items of June, Tuesday, three o'clock in the afternoon,
and cloudy as a key. In the example 4a, the row 4a06, that is, the
data on June 19 corresponds to the concerned data. If the number of
the extracted data items is not enough, the weather item may be
ignored, the temperature of 25.degree. C. may be included as a key
item, or the temperature of 24.degree. C. to 26.degree. C. may be
included as a key item. For example, in the example 4a, assuming
that the weather is ignored and the temperature range is 24.degree.
C. to 26.degree. C., the rows 4a01 and 4a02, that is, the data on
June 5 and June 12 are extracted.
[0049] Second, based on the extracted recorded demand data of each
energy form, the basic statistics (specifically, average values and
standard deviations) 4b of electric power, steam and chilled water
are derived. The basic statistics 4b in FIG. 4 is derived on the
extracted data 4a01, 4a02 and 4a06. The method of calculating an
average value and a standard deviation may utilize the method
described in ISHIMURA Sadao, "Easily Understandable Statistics
Analysis", pages 14 to 27, Tokyo-Tosho (1993). The basic statistics
4b is obtained by the following expressions. 1 Average Value of
Electric Power = 351 + 354 + 354 3 = 353 Average Value of Steam =
238 + 265 + 244 3 = 249 Average Value of Chilled Water = 150 + 151
+ 149 3 = 150 Standard Deviation of Electric Power = ( 351 - 353 )
2 + ( 354 - 353 ) 2 + ( 354 + 353 ) 2 3 - 1 / 3 = 1.0 Standard
Deviation of Steam = ( 238 - 249 ) 2 + ( 265 - 249 ) 2 + ( 244 -
249 ) 2 3 - 1 / 3 = 8.2 Standard Deviation of Chilled Water = ( 150
- 150 ) 2 + ( 151 - 150 ) 2 + ( 149 - 150 ) 2 3 - 1 / 3 = 0.6
[0050] Last, based on the basic statistical amount, the upper and
the lower limit values of the demand prediction are derived.
Concretely, assume that the demand prediction upper limit
value=average value+constant times of standard deviation and demand
prediction lower limit value=average value-constant times of
standard deviation. Herein, assuming that spreads of the prediction
errors follow the Gaussian distribution, if the constant is 2.0,
the demand prediction upper and lower limit values are obtained in
the range of the corresponding probability 95%. If the constant is
3.0, it is obtained in the range of the corresponding probability
99%.
[0051] The demand prediction upper and lower limit values 4c of
FIG. 4 are calculated on the basic statistics 4b. The foregoing
description has concerned with the method of applying a standard
deviation for deriving spreads of prediction errors. As another
method, the table for holding the prediction error width for each
condition may be predefined. The reference values of the demand
prediction upper and lower limits may utilize not only the
above-mentioned average values but also the recorded values in the
near past. The use of the latter data allows the operator to obtain
a policy of how to operate the supply device on the basis of the
displayed target operation pattern upper and lower limits.
[0052] The generation of the sample data will be described with
reference to FIG. 5. The sample data 51 shown in FIG. 5 is
generated by using the demand prediction upper and lower limit
values 4c shown in FIG. 4. The sample data 51 includes an
occurrence probability of each data item, which is derived on the
assumption that the occurrence probabilities of an electric power
demand value, a steam demand value and a chilled water demand value
correspond to the values as described by 5a1, 5a2 and 5a3,
respectively. In the sample data 51, the possible demand value is
divided into three stages, a lower limit value, a center value and
an upper limit value. In actual, it may be divided into more
stages.
[0053] As a result, the highest occurrence probability 0.216 takes
place in the case that the demand predictions of the electric
power, the stream and the chilled water are 353 kJ/s, 249 kJ/s and
150 kJ/s, respectively.
[0054] The obtention of the optimal operation pattern will be
described with reference to FIG. 6. Then, the optimal operation
pattern of each combination of demand values will be obtained along
the flow of FIG. 6. FIG. 6 exemplarily shows the optimal operation
pattern 61 of each energy supply device generated on the sample
data. As shown in FIG. 5, the number of the sample data units is
27, so 27 optimal operation patterns are prepared. The left-end
numbers of FIG. 6 correspond with the left-end ones of FIG. 5. It
means that the occurrence probabilities of FIG. 5 are common to
those of FIG. 6. The method of obtaining the optimal operation
pattern is executed by applying the technology described in ITO
Kouichi and YOKOYAMA Ryouhei, "Optimal Planning of Cogeneration",
pages 57 to 63, Sangyou-Tosho (1990). The illustrated data is based
on the following configuration of the energy supply system. The
electric power is generated by the gas turbine GT and the steam
turbine ST. The steam is supplied from the gas turbine GT and the
gas boiler GB and is consumed by the steam turbine ST and the
absorption refrigerator AC. The remaining steam is supplied to the
steam demand. The chilled water is merely supplied by the
absorption refrigerator AC. The cogeneration energy supply facility
shown in FIG. 12 consumes the energy as described below. That is,
the gas turbine GT generates a steam of 600 kJ/s (double) as a
byproduct when it produces an electric power of 300 kJ/s. The
stream turbine ST consumes steam of 150 kJ/s (triple) when it
produced an electric power of 50 kJ/s. The absorption refrigerator
AC consumes steam of 750 kJ/s (quintuple) when it produces chilled
water of 150 kJ/s.
[0055] In FIG. 5, the sample data of No. 1 includes an electric
power demand of 353 kJ/s, a steam demand of 249 kJ/s and a chilled
water demand of 150 kJ/s. In FIG. 6, the operation pattern of No. 1
generates steam of 600 kJ/s when the gas turbine GT produces an
electric power of 300 kJ/s. The steam turbine ST consumes steam of
150 kJ/s and thereby produces an electric power of 50 kJ/s. Since
the electric power demand is 353 kJ/s, the shortage of 3 kJ/s is
fed from the commercial electric power. The absorption refrigerator
AC consumes steam of 750 kJ/s and thereby produces chilled water of
150 kJ/s. The steam generated by the gas turbine GT is 600 kJ/s and
the steam consumed by the steam turbine ST and the absorption
refrigerator AC is 150+750=900 kJ/s. Hence, the shortage is 300
kJ/s. The gas boiler GB is thus required to generate steam of 549
kJ/s in response to the steam demand of 249 kJ/s and the short
steam demand of 300 kJ/s.
[0056] The steam turbine ST is adjusted to generate an electric
power of 50 kJ/s if the steam demand is 232.6 kJ/s or 249 kJ/s or
an electric power of 40 kJ/s when the steam demand is 265.4 kJ/s.
If the electric power generated by the gas turbine GT and the steam
turbine ST is not enough to the power demand, the shortage is fed
from the commercial power.
[0057] Lastly, based on the 27 operation patterns shown in FIG. 6,
the process is executed to derive the target operation pattern
upper and lower limits 71 shown in FIG. 7. FIG. 7 exemplarily shows
the upper and lower limits of the target operation pattern. In this
example, the maximum value and the minimum value of the operation
pattern correspond to the upper limit and the lower limit of the
target operation pattern, respectively. In this example, the upper
limit and the lower limit of the target operation pattern in the
gas turbine are both 300 kJ/s. Hence, it is understood that the gas
turbine should be operated at an output of 300 kJ/s. The absorption
refrigerator should be operated in a narrow range of 149 to 151.
The steam turbine should be operated in a relatively wide range of
40 to 50 kJ/s and the gas boiler should be also operated in a
relatively wide range of 528 to 544 kJ/s.
[0058] In general, the width between the upper and the lower limits
of the target operation pattern depends on the following two
factors. One factor is a width between the upper and the lower
limits of the demand prediction and the other factor is a
sensitivity to a demand change of the optimal operation pattern. As
will be understood from the comparison of the standard deviations
of the basic statistics 4b, the steam demand is so variable that
the width between the upper and the lower limits of the steam
demand prediction is made relatively wide as indicated by the upper
and the lower limits 4c of the demand prediction. The gas turbine
and the gas boiler are assumed as the steam supply source. If each
optimal operation pattern of the gas turbine and the gas boiler has
the same sensitivity to the demand change of the steam, the width
between the upper and the lower limits of the target operation
pattern of the gas turbine is substantially equal to that of the
gas boiler. However, the resulting target operation pattern upper
and lower limits 71 indicate that the width between the upper and
the lower limits of the target operation pattern of the gas turbine
is zero, while the upper limit of the gas boiler is 528 kJ/s and
the lower limit thereof is 544 kJ/s. The difference of the width
between the pattern upper and lower limits takes place between the
gas turbine and the gas boiler results from the fact that the gas
turbine has a small (zero) sensitivity of the optimal operation
pattern to the steam demand change, while the gas boiler has a
large sensitivity.
[0059] This sort of information is quite significant to the safely
and economical operation of the energy supply facility by the
operator. The sort of information can be obtained only if the
system displays the upper and lower limits 71 of the target
operation pattern shown in FIG. 7. It cannot be obtained if the
background art merely displays the upper and the lower limits of
the demand prediction.
[0060] Then, the description will be oriented to the function
configuration of the operation support system according to the
second embodiment with reference to FIG. 8. This operation support
system includes a target upper and lower limit setting module 81,
which determines the upper and the lower limits of the target
operation pattern so that the integrated value of the probability
density distribution is made equal to a specified probability value
84. The action of the operation support system shown in FIG. 8 will
be described with reference to the examples shown in FIGS. 4 to 7.
As similar as the first embodiment, these functional blocks are
implemented by hardware system in FIG. 2. The target upper and
lower limit setting module 81 is included to the processor 22. The
demand predicting module 11 derives the demand prediction upper and
lower limit values 14 and 15 as well as a demand prediction
probability density function 82, which correspond to the occurrence
probabilities 5a1, 5a2 and 5a3, respectively.
[0061] Then, the target operation pattern calculating module 12
derives a target operation pattern probability density function 83
through the use of the demand prediction upper and lower limit
values 14 and 15 and the demand prediction probability density
function 82. The coordination of the operation pattern and the
occurrence probability of the optimal operation pattern 61 with
respect to the gas boiler results in producing the demand
prediction probability density function 91 of the target operation
pattern of the gas boiler shown in FIG. 9, which corresponds to the
target operation pattern probability density function 83 of the gas
boiler.
[0062] Next, the target upper and lower limits setting module 81
derives the target operation pattern upper and lower limits 16 and
17 through the use of the target operation pattern probability
density function and the specified probability value 84. For this
purpose, the setting module 81 gradually wide the width between the
upper and the lower limits with the target operation pattern as a
center, the target operation pattern corresponding with the peak of
the target operation pattern probability density distribution, for
searching the width between the upper and the lower limits in which
the integrated value coincides with the specified value.
[0063] In turn, the description will be oriented to the operation
of the target upper and lower limits setting module 81 on the
assumption that the specified probability value 84 is 0.8, that is,
80% with an example of the target operation pattern probability
density function 91 shown in FIG. 9. The target upper and lower
limit setting module 81 has an original function of setting the
target operation pattern upper and lower limits as to the gas
turbine, the steam turbine, the absorption refrigerator and the gas
boiler. Herein, as shown in FIG. 9, the gas boiler is used as an
example. The same principle is applied to the other devices rather
than the gas boiler. At first, the target upper and lower limits
setting module 81 determines the reference operation point. The
method of determining the point may be a method of assuming the
reference operation point as an average value, a center value or a
most frequent value. In the graph shown in FIG. 9, the average
value=528.times.0.04+530-
.times.0.04+533.times.0.12+535.times.0.12+538.times.0.04+540.times.0.04+54-
4.times.0.12+549.times.0.36+544.times.0.12=543, the center
value=(528+554)/2=541, and the most frequent value=549. Herein, the
following description will be expanded as assuming the reference
operation point as the most frequent value 549 kJ/s.
[0064] Hereafter, the process is executed to symmetrically move the
upper limit and the lower limit from the reference operation point
in the plus and the minus directions. When the total of the
probabilities located between the upper and the lower limits
reaches 80%, the upper and the lower limits are searched. At first,
assuming that the reference operation point of 549 kJ/s is located
on the upper and the lower limits, the probability is 0.36, which
does not reach 0.8. If the width between the upper and the lower
limits is widened like .+-.1, .+-.2, .+-.3, .+-.4, the probability
is still 0.36. If the width is .+-.5, the lower limit is 544 kJ/s
and the upper limit is 554 kJ/s. The probability of the area
between the upper and the lower limits is 0.12+0.36+0.12=0.5, which
does not still reach 0.8. If the width is widened into .+-.14, the
lower limit is 535 kJ/s and the upper limit is 563 kJ/s, in which
the probability of the area between the upper and the lower limits
is 0.12+0.12+0.04+0.04+0.12+0.36+0.12=0.80, which finally reaches
the specified value 0.8. Hence, the target upper and lower limits
setting module 81 determines that the target operation pattern
lower limit is 535 kJ/s and the upper limit is 563 kJ/s. Further,
the upper limit is delimited by the maximum value of 554 kJ/s of
the operation pattern having nonzero probability so that the range
between the upper and the lower limits of the target operation
pattern is 535 to 554 kJ/s.
[0065] After the upper and the lower limits of the target operation
pattern are determined, view display 23 displays the values to the
operator on the target operation pattern display screen 13.
[0066] The application of the foregoing method makes it possible to
adjust the width between the upper and the lower limits of the
target operation pattern by adjusting the specified probability
value. In general, as a smaller specified probability is taken, the
width between the upper and the lower limits of the target
operation pattern is made smaller, which brings about an advantage
that the target is so narrowed that it may be more understandable.
On the other hand, it brings about a disadvantage that the
difference of the width between the upper and the lower limits in
each device is obscure. On the other hand, as a larger specified
probability value is taken, the width between the upper and the
lower limits of the target operation pattern is likely to be
larger, which brings about a disadvantage that the target is so
obscure that it cannot be easily grasped. On the other hand, it
bring about an advantage that the difference of the width between
the upper and the lower limits in each device is so clear. For
example, in the example shown in FIG. 6, the probability in 300
kJ/s of the gas turbine reaches 100%, so that how largely the
specified probability value may be assumed, the upper and the lower
limits are both kept in 300 kJ/s. On the other hand, in the gas
boiler, the width between the upper and the lower limits is made
wider according to a larger specified probability value.
[0067] As will be understood from this example, a larger specified
probability value requires the gas turbine to be operated in 300
kJ/s, while the gas boiler does not need to strictly consider the
specified probability value. Roughly speaking, it is requested that
a beginner operator assume a smaller specified probability value
for making the target more obvious, while a skilled operator assume
a larger value for grasping a target reliability.
[0068] The description will be oriented to the three-dimensional
representation on the target operation pattern displaying screen 13
in which a time, a target operation pattern and a target operation
pattern probability density distribution are assigned to three axes
with reference to FIG. 10. As shown, view display 23 outputs a
three-dimensional representation in which the X axis corresponds
with the time, the Y axis corresponds with the target operation
pattern for the peak of the target operation pattern probability
density distribution, and the Z axis corresponds with the target
operation pattern probability density function. This type of
three-dimensional representation makes it possible to more clearly
present the time shift of the target operation pattern probability
density distribution to the operator.
[0069] In turn, the description will be oriented to the function
configuration of the operation support system according to a third
embodiment. This operation support system is characterized to have
an alarming module 111 in addition to the operation support system
shown in FIG. 1. The alarming module 111 is included to the
processor 22. The alarming module 111 compares the target operation
pattern upper limit 16, the target operation pattern lower limit 17
and the current operation pattern 112 with one another and outputs
an alarm if the current operation pattern departs from the target
operation pattern upper and lower limits. The constantly checking
of the proper fitting of the actual operation pattern in the range
between the target operation pattern upper and lower limits is a
dull work for a skilled operator having a deep knowledge of the
optimal operation. The use of the alarming module 111 allows the
skilled operator to be released from the dull constant checking
work.
[0070] While the foregoing has described what are considered to be
the best mode and/or other preferred embodiments, it is understood
that various modifications may be made therein and that the
invention or inventions disclosed herein may be implemented in
various forms and embodiments, and that they may be applied in
numerous applications, only some of which have been described
herein. It is intended by the following claims to claim any and all
modifications and variations that fall within the true scope of the
inventive concepts.
* * * * *