U.S. patent application number 09/683004 was filed with the patent office on 2003-05-08 for system and method for continuous optimization of control-variables during operation of a nuclear reactor.
Invention is credited to Kropaczek, David Joseph, Russell, William Earl II, Watford, Glen Alan.
Application Number | 20030086520 09/683004 |
Document ID | / |
Family ID | 24742160 |
Filed Date | 2003-05-08 |
United States Patent
Application |
20030086520 |
Kind Code |
A1 |
Russell, William Earl II ;
et al. |
May 8, 2003 |
System and method for continuous optimization of control-variables
during operation of a nuclear reactor
Abstract
A system and method is provided for a continual updating of the
optimization of multiple operational control-variables during the
operation of a nuclear reactor over a plurality of fuel cycles. A
networked computer system includes one or more hosts programmed to
execute an optimization process to identify and make changes in
quantitative values of operational control-variables that result in
improved efficiency and operational flexibility. Optimization and
updating of operational control-variables may proceed selectively
under manual control for inputting specific optimization
constraints and reactor state-point information or may proceed
autonomously through a repetitive performing of the optimization
process based upon a predetermined user-defined strategy stored on
the network. Communications between users and networked processors
is facilitated by use of a TCP/IP server connected to the Internet
so that portions of the optimization process may be conducted
contemporaneously at remote locations and/or the results made
accessible to users via conventional browser enabled computers.
Inventors: |
Russell, William Earl II;
(Wilmington, NC) ; Kropaczek, David Joseph; (Kure
Beach, NC) ; Watford, Glen Alan; (Wilmington,
NC) |
Correspondence
Address: |
Nixon & Vanderhye, P. C.
8th Floor
1100 North Glebe RD.,
Arlington,
VA
22201-4714
US
|
Family ID: |
24742160 |
Appl. No.: |
09/683004 |
Filed: |
November 7, 2001 |
Current U.S.
Class: |
376/259 |
Current CPC
Class: |
G21D 3/08 20130101; G21D
3/001 20130101; Y02E 30/30 20130101; Y02E 30/00 20130101 |
Class at
Publication: |
376/259 |
International
Class: |
G21C 017/00 |
Claims
1. A method for determining and updating a projected strategy for
maintaining optimal operations of a nuclear reactor based on one or
more optimization input parameters and/or reactor state-point
information stored in a database, comprising the steps performed by
a computer or computer network of: a) accepting input data for
initializing or modifying one or more optimization input parameter
values which collectively define a particular reactor operation
strategy; b) computing optimized quantitative values for a
plurality of reactor core operational control-variables based on
current reactor plant state-point data, wherein the computed
optimized quantitative values meet predetermined design constraints
for the reactor; and c) displaying, on a display device coupled to
said computer or computer network, at least one or more of said
computed optimized quantitative control-variable values, said
information being indicative of a projected strategy for optimal
reactor performance.
2. The method of claim 1 further including the step of computing
optimal reactor simulation results based on optimized quantitative
values computed in step (b).
3. The method of claim 1 wherein the step of computing optimized
quantitative values for a plurality of reactor core operational
control-variables further comprises computing polynomial response
surface data based on a plurality of reactor core operation
simulations.
4. The method of claim 1 wherein the step of computing optimized
quantitative values for a plurality of reactor core operational
control-variables further comprises using predetermined polynomial
response surface data to predict optimal changes in independent
control-variable values, said polynomial response surface data
based on a plurality of reactor core operation simulations.
5. The method of claim 1 further including the step of periodically
computing updated values for one or more optimization inputs in
accordance with a current time and/or a time remaining in a current
fuel cycle.
6. The method of claim 1 wherein optimization input data of step
(a) comprises predetermined stored optimization input parameter
values which collectively define a preferred operational
strategy.
7. The method of claim 6 wherein a stored reactor operational
strategy incorporates reactor plant licensing requirements.
8. The method of claim 1 wherein optimization input data of step
(a) is automatically retrieved from a database of optimization
input data.
9. The method of claim 1 wherein optimization input data of step
(a) is manually input by a user.
10. The method of claim 1 further including the step of selectively
displaying historical and/or current reactor plant performance
data.
11. The method of claim 1 wherein computation of optimized
control-variables is performed in response to changes made to an
existing or preferred reactor operational strategy.
12. The method of claim 1 further including the step of converting
thermal and/or reactivity margin data into fuel cycle energy
data.
13. The method of claim 1 further including the step of converting
reactor fuel cycle energy data into thermal and/or reactivity
margin data for subsequent display via graphical user
interface.
14. The method of claim 1 wherein computing optimized quantitative
values for one or more reactor core operational control-variables
is performed in response to poor or inadequate process computer
and/or design simulator predictions.
15. The method of claim 1 wherein computing optimized quantitative
values for one or more reactor core operational control-variables
is performed for a reactor during a given fuel cycle, N, so as to
improve efficiency of reactor operations for a subsequent fuel
cycle, N+1.
16. The method of claim 1 wherein the step of computing optimized
quantitative values for one or more reactor core operational
control-variables is performed and/or an optimal projected reactor
operational strategy is displayed in direct response to one or more
specific commands input via graphical user interface.
17. The method of claim 1 wherein the step of computing optimized
quantitative values includes displaying all independent control
variables via graphical user interface.
18. The method of claim 1 wherein the step of computing optimized
quantitative values includes utilizing exposure dependent
constraints.
19. The method of claim 1 wherein the step of computing optimized
quantitative values includes providing multiple solutions from
varying constraint criteria.
20. The method of claim 1 wherein the step of computing optimized
quantitative values includes performing computations based upon a
user selected extended reactor fuel cycle duration value.
21. The method of claim 1 further including the step of computing a
maximum fuel cycle energy corresponding to a particular reduced
coolant flow capability.
22. The method of claim 1 further including the step of performing
a coastdown reduction computation.
23. The method of claim 1 further including the step of computing a
maximum undisturbed sequence length.
24. A nuclear reactor operations optimization system for
determining and updating one or more strategies for optimal
operations of a nuclear reactor plant based on one or more
optimization input parameters and/or reactor state-point
information stored in a database, comprising: a plurality of
optimization system host processors coupled via a digital
communications network, at least one or more of said host
processors operative as a control-variable optimization engine and
at least one or more of said host processors providing a graphic
user interface for selecting optimization inputs and inputting
associated parameter values and/or displaying information
indicative of an optimized operational strategy based on
optimization output results provided by the optimization engine;
and a database storage device accessible by one or more of said
host processors via the communications network for storage of
optimization inputs and parameter values.
25. The reactor operations optimization system of claim 24 wherein
at least one or more of said host processors process one or more
reactor core operation simulation cases corresponding to a
predetermined set of independent control-variable values.
26. The reactor operations optimization system of claim 24 wherein
reactor state-point information is produced by a reactor simulation
process performed by one or more of processors connected to the
network and is stored in a database accessible via the
communications network.
27. The system of claim 24 wherein the communications network
comprises a LAN.
28. The system of claim 24 wherein the communications network
comprises a WAN.
29. The system of claim 24 wherein the communications network
comprises, at least in part, existing Internet communications
infrastructure.
30. The system of claim 24 wherein the display device includes a
graphical user interface that enables retrieval and selective
display of historical and/or current reactor plant performance data
stored on one or more storage devices accessible via the
communications network.
Description
BACKGROUND OF INVENTION
[0001] This invention generally concerns nuclear reactor operations
optimization and management. More particularly, the present
invention is directed toward identifying optimal reactor plant
operational settings and an ongoing management strategy that
incorporates a consideration of plant-specific constraints for a
multiplicity of operational control-variables such as, for example,
control blade positioning, cycle flow strategy, location of
sequence exchanges, and other critical-to-quality control variables
relevant to operation of a nuclear reactor plant throughout one or
more reactor core refueling cycles.
[0002] A nuclear reactor plant includes many different individual
components that have dynamic characteristics which may affect any
given strategy devised for eliciting a more efficient operation of
the reactor core. For example, a nuclear reactor core has many,
e.g., several hundred, control blades which each require position
and location identification throughout the direction of one or more
cycles of fuel depletion. In addition, many other controllable
elements and factors that affect the reactivity and overall
efficiency of a reactor core must also be taken into consideration
if one is to design or develop an effective control strategy for
optimizing the performance of a reactor core at a particular
reactor plant. Such variable "operational controls" (also referred
to herein as "independent control-variables") include various
physical component and controllable operating conditions
configurations within the reactor that can be individually adjusted
or set before or during operation of the reactor.
[0003] For example, the locations of the various control blades
within the reactor core are one of the many independent
controllable variables that significantly affect the generated
power output and overall efficiency of operation of a reactor.
Other operational controls include such controllable variables as
"core flow" (rate of water flow through the core) and the timing of
the sequence exchange or exposure interval at which groups of
control blades are changed. Each of these so called variable
operational controls may be considered as an independent
"control-variable" which has a measurable effect on the overall
performance of the reactor core. Due to vast number of possible
different operational values and combinations of values that these
independent control-variables can assume, it is both a formidable
challenge and a very time consuming task using conventional
computer-aided methodologies to attempt to analyze and optimize
most if not all of the individual influences that may have an
impact core reactivity and performance.
[0004] In order to furnish and maintain a required energy output,
the reactor core is periodically refueled with fresh fuel bundles.
The duration between one refueling and the next is commonly
referred to as a "fuel cycle", "core cycle", or "cycle" of
operations and, depending on the particular reactor plant, is on
the order of twelve to twenty-four months. During the course of a
cycle, the excess energy capability of the core, defined as the
excess reactivity or "hot excess", is controlled by core coolant
(water) flow and the control blades. Typically, a reactor core
contains many such control blades which are fit between selected
fuel bundles and are axially positioned within the core.
[0005] The total number of control blades utilized in a reactor
varies with core size and geometry, and is typically between fifty
(50) and one-hundred and fifty (150). The axial position of control
blades (e.g., fully inserted, fully withdrawn, or somewhere in
between) is based on the need to control excess reactivity and to
meet other operational constraints, such as thermal or reactivity
margins. For each control blade, there may be, for example,
twenty-five or more possible axial positions and twenty-five or
more "exposure" (duration of use) steps. Considering symmetry and
other requirements that may reduce the number of control blades
that are available for application at any one time, there are more
than several million existing combinations of control blade
positions possible for even the simplest arrangement. Larger
arrangements may have more than a googol (1.times.10.sup.100)
possible configurations. However, only a small fraction of these
configurations will satisfy all the applicable design and safety
constraints, and of those, only a smaller fraction prove
economical. Moreover, the axial positioning of control blades also
influences core cycle energy and potential thermal limits. Since it
is desirable to maximize the core-cycle energy in order to minimize
nuclear fuel cycle costs, developing an optimum control blade
positioning strategy is yet another type of independent
control-variable optimization problem that should be taken into
consideration when attempting to optimize operational management
strategies.
[0006] Historically, cycle operations and core management,
including control blade positioning, sequence exchange lengths, and
core flow selection, are determined on a "trial-and-error" basis
based primarily on the past experiences of the reactor engineers.
Due to circumstances that require a rapid response to changing
plant operating conditions, a reactor engineer may be faced with
the formidable challenge of specifying values for over one-hundred
or more independent control-variables within a very short time
frame. If a particular design constraint is not satisfied by an
identified arrangement, then the arrangement is modified and a
computer simulation is run. Because of the relatively long computer
simulation time required for assessing the impact of a change in
the value of even a single given independent control-variable,
man-weeks of human and computer resources are typically required
before an appropriate operational strategy is identified using this
procedure. Moreover, using this trial-and-error approach, once a
operation strategy that satisfies all identifiable design and
safety constraints is determined, it may turn out that the
identified arrangement does not provide the best cycle-energy
economics. In that case, the trial-and-error selection process must
continue until the engineers believe that an optimum operational
strategy has been identified. In practice, however, it is very
possible that a particular core arrangement that is inconsistent
with past experience may actually be the optimum operational
strategy to use.
[0007] Numerous systems have attempted to address various aspects
of the above problem through the implementation of various
improvements in display interfaces to the reactor engineer (e.g.,
see U.S. Pat. Nos. 5,859,885, 4,853,1 75 and 5,812,622),
improvements in data management of information (e.g., see U.S. Pat.
Nos. 5,793,636 and 4,459,259), improvements towards reactor
operation alarms (e.g., see U.S. Pat. Nos. 5,311,562 and
5,023,045), and improvements in the instantaneous monitoring of the
reactor (e.g., see U.S. Pat. Nos. 4,997,617, 5,309,485; 5,392,320
and 5,091,139). Although such efforts have somewhat improved the
real-time monitoring and display of information required for
operation of a nuclear reactor, none provide the tools needed for
determining the appropriate settings of the independent control
variables for an entire full cycle or longer. Moreover, the above
prior art systems all rely significantly on a manual input/data
selection process in the development of any operational
strategy.
[0008] There have been a few attempts to provide automated
predictive capabilities for one or more aspects of the above
problem through the use of so called "decision tree" or "neural
net" technologies. For example, U.S. Pat. No. 4,552,718 to Impink,
Jr. et al. discloses a system for monitoring the operational status
of a nuclear reactor that provides indication of "off-normal"
conditions, and a path of operation by which the reactor can be
restored to normal conditions by way of "decision tree" logic. Such
"decision tree" technologies are capable of providing correct logic
to a limited number of independent variables when adequate
supporting data is provided such that the cause and effect
relationships are well defined, such as global reactor problems and
consequent human responses. However, optimization of all current
and future independent variables as described above for operation
of a boiling water reactor (BWR) for entire full cycle of
operations essentially requires an infinite number of cause/effect
relations and is not particularly feasible. Similarly, U.S. Pat.
No. 5,009,833 to Takeuchi et al. describes an expert system rule
based optimization approach. Much like "decision tree"
technologies, "expert system" rule based technologies are only as
good as the rules provided to the system. Consequently, while these
technologies are capable of identifying global reactor issues and
the subsequent necessary human response, their application is often
not practical and does not include an application for a continuous
operations optimization of a working nuclear reactor.
[0009] In another example, U.S. Pat. No. 5,790,616 to R. O. Jackson
et al., issued Aug. 4, 1998, describes an early attempt to perform
optimization on control blade locations for a nuclear reactor. In
this example, optimizations are performed using a genetics based
algorithm at a single time sequence. Once the preferred rod
patterns at a given time sequence are determined, the rods are
established and the following time step is studied. A heuristic
assumption is integrated into the system by assuming that the
"best" set of rod patterns for the cycle are the rod patterns that
provide for the lowest axial peak in the core. Although this
heuristic assumption enabled the Jackson et al. system to optimize
on a subset of the total number of independent variables
(approximately 6-12), the assumption precludes the obtainment of a
true optimal solution. Moreover, on many BWR reactors, extremely
hard bottom burns at the beginning of cycle (BOC) can lead to
conditions at the end of cycle (BOC)--where thermal limits are
excessive and require the reactor to lower power levels to maintain
safe operation. Consequently, the system disclosed by Jackson et
al. provides neither optimal nor potentially usable solutions.
[0010] To more adequately address the above problems, it would be
most desirable to have a very efficient computer system arrangement
that would be capable of performing a comprehensive nuclear reactor
plant operations optimization process that would identify most, if
not all, of the appropriate changes/modifications in operational
control-variables that are needed to result in an improved fuel
cycle efficiency, better global reactor economics and enhanced
operational flexibility.
SUMMARY OF INVENTION
[0011] In one aspect, the present invention is a system and method
for identifying the best possible quantitative values for one or
more operational control-variables that are collectively associated
with the functioning and control of a nuclear reactor core and for
determining an optimal operational strategy for one or more
refueling cycles of the core. In another aspect, the present
invention is a system and method for updating and maintaining an
optimal operational strategy for a nuclear reactor continuously
during the operation of the reactor over a duration of multiple
refueling cycles. In this aspect, once a set of preferred
optimization constraints are identified, the system effectuates a
continued optimization of reactor operations throughout the
duration of a multiplicity of fuel cycles by effectively
continuously providing updated reactor operations parameters
(optimized control-variables) which may be directly implemented in
controlling the reactor to result in a more flexible, economical
and safe manner of operation.
[0012] In an example embodiment of the present invention, a system
and method for optimizing multiple operational control-variables
during operation of a nuclear reactor comprises a networked
computer system that includes one or more host processors or
computers programmed to execute an optimization process that
identifies and/or implements changes to one or more of the many
operational control-variables of a reactor plant in order to
improve reactor fuel cycle efficiency, global reactor economics,
and enhance operational flexibility. In a further aspect, the
present invention encompasses a computer network system with
communications enhanced by connection to the Internet so as to
distribute processing loads and to facilitate access and control of
the optimization process from a wide variety of locations.
[0013] In the example embodiment, optimization and updating of
operational control-variables may proceed selectively under manual
control for inputting specific optimization constraints and reactor
state-point information or may proceed autonomously through a
repetitive performing of the optimization process based upon a
predetermined user-defined strategy stored on the network. Using a
graphics input and display user-interface (GUI), a reactor design
specialist/engineer may selectively input and review various
independent variable selections and their resulting dependent
variable responses, or change various optimization constraints and
controls in pursuit of alternative design strategies. As will
become evident by the following description, once a set of
preferred optimization constraints are identified, the present
invention may also be used to effectuate automatic repetitive
adjustments to reactor controls to effectively provide continuous
optimizations in reactor operation over the duration of one or more
reactor fuel cycles.
BRIEF DESCRIPTION OF DRAWINGS
[0014] FIG. 1A is a block diagram illustrating an example reactor
control-variable optimization system;
[0015] FIG. 1B is a flowchart illustrating the two fundamental
operational processing loops and a general data processing overview
of an example system for optimization of multiple operational
control-variables of a nuclear reactor in accordance with the
present invention;
[0016] FIG. 2 is a block diagram illustrating exemplary groups of
information and parameter types stored on a general database
storage device of the present invention;
[0017] FIG. 3 is a block diagram illustrating some example
operational strategy change issues;
[0018] FIG. 4 is a flowchart illustrating an example process for
communicating a notice of change in reactor state point and an
intent to automatically re-optimize an initial strategy in
accordance with the present invention;
[0019] FIG. 5 is a block diagram illustrating exemplary information
stored in an Optimization Inputs database of the present
invention;
[0020] FIG. 6A is a flowchart illustrating a general overview of
example software processing steps performed by the optimization
process in accordance with the present invention;
[0021] FIG. 6B is a flowchart illustrating processing steps of an
example software optimization engine for performing optimization
computations to determine preferred values for control-variables in
accordance with the present invention; and
[0022] FIG. 7 is a block diagram illustrating exemplary information
stored in an Optimization Outputs database of the present
invention.
DETAILED DESCRIPTION
[0023] The following description is directed toward a presently
preferred embodiment of the present invention, which may be
implemented as a remote user application running, for example,
under the Microsoft Windows 2000 operating system on Intel Pentium
hardware with WEB for communications, Oracle for database
structure, and UNIX for a database operating system. The present
invention, however, is not limited to any particular computer
system or any particular environment. Instead, those skilled in the
art will find that the system and methods of the present invention
may be advantageously applied to any environment requiring
continuous optimization of multiple control-variable critical
processes or systems, including industrial, chemical, mechanical,
nuclear, and biotech. Moreover, the invention may be embodied using
a variety of computer operation system platforms, including UNIX,
LINUX, Mac OS, Open VMS, Solaris, SCO UNIX, Digital UNIX, HP UNIX,
AIX, OSF, DOS, OS/2, BSD, Plan-9, and the like. Similarly, the
invention may be implemented using a variety of different hardware
environments, including X86, Power PC, Strongarm, Alpha, Sparc,
RISC, Cray, and the like. Therefore, the particular description of
example embodiments of the invention provided herein is for
purposes of illustration and not limitation.
[0024] An example embodiment of the system of the present invention
utilizes a network of independent processors for contemporaneously
conducting multiple simulations of a reactor core operating under
different conditions and constraints. Each simulation is
representative of a different virtual operational case and
comprises different sets of values for various reactor core
operational parameters (i.e., the independent control-variables).
The reactor core simulations provide output data that is indicative
of selected performance parameters which reflect the operational
state of the reactor throughout the duration of a reactor core fuel
cycle. Once all reactor core simulations are completed, the
simulation output data for each control-variable case is
accumulated, normalized and mapped by a host processor to
corresponding high-order polynomials that fit the reactor core
simulation results data for each control-variable case.
Coefficients that uniquely describe each polynomial are collected
in an associated memory device as a multidimensional data array
that serves as a type of virtual "response surface".
[0025] In this manner, the virtual response surface acts as a
cyber-workspace and repository for storing resultant output data
from many control-variable case simulations. The polynomials are
then used to predict quantitative values (i.e., dependent
variables) for the reactor performance parameters over a limited
range of independent control-variable values. The predicted
performance parameter values from each polynomial predictor are
compared using an objective function to determine which particular
associated independent control-variable(s) is likely to produce the
greatest improvement. Another core simulation using the identified
values is then performed to provide calibration of the polynomial
predictors and to calibrate the polynomial coefficient data in the
response surface with the simulation process.
[0026] Upon completion of the above global optimization process,
optimized parameter values for the independent variables associated
with different reactor operations (e.g., rod patterns, core flow,
and sequence exchange times) may be digitally communicated via, for
example, LAN/WAN, the Internet or other network facilities for
using and displaying at various locations. For example, a
Web-browser enabled computer connected to the Internet may be
utilized as an output display device/terminal and may also serve as
one of the optimization system processors as well as one of the
reactor core simulation machines (core simulator). Through this
distributed computing and display arrangement, system users at
various locations may view preferred independent variable
selections and resulting dependent variable responses, and/or
change various optimization constraints and controls for studying
and implementing alternative operation strategies.
[0027] In one aspect of the present invention, quantitative values
for several thousand operational control-variables produced by an
optimization process host processor and one or more networked
reactor core simulators are stored in an network accessible storage
device for maintaining a database of both past and projected
reactor operation information. A reactor design specialist may then
study such information to determine practical ranges and limits
that constitute allowable and safe operation of the reactor. A
collection of permissible ranges for the operational
control-variables is then identified and stored in the general
database as a preferred operational strategy for a given reactor.
Although, a given operational strategy may remain as the preferred
strategy for the duration of an entire fuel cycle of a particular
reactor, the present invention also enables the design
specialist/engineer to make on-the-fly modifications to the stored
strategy should the need occur.
[0028] For example, operational strategy changes may occur due to a
desire for a more economically efficient reactor performance,
changes in the NRC licensing requirements, or bad predictions by
the design simulators. Once a design strategy is identified, an
optimization process that provides recommendations for operation
control variable selection. After the optimization process has
completed execution, the predicted reactor performance data is
stored in a general database accessible as part of the system
network. Optimization output results including recommended values
for the operational control-variables, resulting projected
dependent variable values, comparisons of the projected values to
limit values and the like are computed and made selectively
available as an output.
[0029] Once a set of initial strategy definitions have been stored
in an accessible database, the present invention may also be used
to schedule periodic re-optimizations on a regular or continual
basis. Such automatically performed re-optimizations allow
differences between expected simulator biases to be constantly
re-calibrated to actual simulator results. Frequent such
re-optimizations performed based on the most current reactor
state-points, results in more accurate projections of future
operations. The duration of the automatic re-optimizations is only
limited by the speed of the individual host processors used in
performing the optimization computations and the rate at which the
general database is updated with current reactor operational
data.
[0030] FIG. 1A shows an example hardware arrangement of components
for providing a reactor control-variable optimization system. In
this example, one or more host processors 10 are coupled via a
local area network (LAN) 11, a wide area network (WAN) 17 or the
Internet (TCP/IP network). Each processor 10 may host reactor
simulation software and/or client software for accessing and
displaying information provided via a graphic user interface (GUI)
on a display device (12) coupled to the processor. The optimization
system components may include one or more database storage devices
14 accessed via, for example, one or more database servers 13. In
addition, the optimization system may include remotely located host
processors and/or database storage devices in communication with
local LAN 11 via connection to a remote LAN/WAN 17 or over the
Internet, for example, via TCP/IP servers 15 and 16.
[0031] A beneficial aspect of the present invention comes from the
implementation of a system configuration that uncouples the
graphical user interface (GUI) from the location of the
computational environment where the optimization calculations are
actually performed. For example, in an example embodiment of the
present invention, a TCP/IP network, LAN, WAN or a combination
these and other of digital communications infrastructures may be
used to connect a computer or terminal having a graphical user
interface to one or more database storage devices and
processors/servers that perform the optimization process
computations.
[0032] Referring now to FIG. 1B, a data processing flow diagram
illustrates an example system for continuous optimization of
multiple operational control-variables of a nuclear reactor in
accordance with the present invention. The flowchart shown provides
a general processing overview of an example system and illustrates
two fundamental operational processing modes: a manual input
constraint definition process (manual loop 10) and an automated
optimization updating process (automated loop 100). By way of the
manual process, updated results from a general database 101 may be
viewed (102) by using a conventional display device (12) driven,
for example, by a graphical user interface (GUI). General database
101 may consist of a central data base (as shown in the Figures)
located on a single storage device or it may be a distributed data
base located on multiple storage devices distributed throughout the
optimization system network. Should a user (e.g., a design
engineer) desire to modify or test an alternative operating
strategy (103), such modifications may also be initiated and input
(104) through the GUI. Although though various features of the GUI
are described herein, details of the GUI driving software and other
conventional GUI features that may be appropriate for use with the
optimization system of the present invention may be readily
developed without undue experimentation by a programmer of ordinary
skill and, as such, are not discussed in detail herein.
[0033] Once selected optimization inputs have been modified, the
various inputs are stored, for example, within optimization inputs
database 106, which may be distinct from, or form a portion of,
general database 101. Next, using the appropriate inputs stored in
optimization inputs database 106, an optimization program 107
determines appropriate values for the independent control-variables
and provides resulting values for all dependent variables. This
optimization output 108 is stored to general database 101 for
subsequent access and viewing as described above. Optimized values
for operational control variables (e.g., rod pattern, flow
strategy, sequence exchange times, sequence lengths, etc.) are
provided as displayable outputs for use in the operational
management of the nuclear reactor core.
[0034] As previously mentioned, one aspect of the present invention
provides automatic updates to control-variables and automatically
updates the status of a currently operating reactor based on a
predefined preferred operating strategy. To implement this
automated aspect of the invention, an updated nuclear reactor
state-point is first obtained from a general database 101 (loop
100). The updated state-point data may be produced, for example,
from actual monitoring devices and sensors on the reactor or as a
result of simulating reactor operations by a conventional reactor
simulator process or program provided on one or more host
processors 10 connected via networks 11, 17 or 18 of FIG. 1A. The
updated reactor state-point information is then used to make
modifications to various optimization input parameters stored in
Optimization Inputs database 106 based on an operational strategy
set up during the manual input loop process(10).
[0035] Referring now to FIG. 2, a block diagram illustrates some
example information (content) of a general (central) database (201)
used for of the present invention. In this example, the information
in general database 201 may be stored, for example, using one or
more mass data storage devices interconnected via the digital
communications network of the system or it may reside entirely on a
single centralized storage device. In this example embodiment, the
general database also includes at least four basic types of data
items: 1) user profile data 204; 2) process computer data 205; 3)
offline model data 206; and 4) optimization data 207. User Profile
Data 204, for example, enables the system to control access of each
user to various information, while preventing access to certain
predetermined restricted data. Because a centralized general
database may include information for a multitude of nuclear
reactors owned by a plurality of different companies, controlled
access to such data may be implemented, for example, by requiring a
password for access or some other conventional security access
arrangement. Security may be enhanced, for example, by requiring
identification of the user IP address and allowing only specific
user and specific machine access privileges. Other profile data
information may include, for example, cumulative user performance
measures of the optimization performance status. Such information
may provide insight towards additional user training requirements
of specific users and may also be used to monitor optimization
value for a specified user and reactor.
[0036] The second type of data stored in general database 201 is
Process Computer Data 205. This process computer data is the
results of actual reactor plant monitoring of operational
parameters such as: LHGR results, CPR results, cycle exposure,
bundle exposures, core average exposures, blade depletions, core
inlet enthalpy, blade positions, core flow rate, LPRM data, hot
reactivity bias, cold reactivity bias, thermal power, electric
power, etc. Such data is highly related to individual reactor plant
instrumentation and, therefore, is considered as the "official"
operation status of a particular nuclear reactor.
[0037] Somewhat analogous to Process Computer Data 205 is the
current and historical "Off-line" Model Data 206. This data is
similar, although not exactly the same, to Process Computer Data
205. Although design inputs may be identical, differences may occur
in reactor performance outputs due to various reactor simulator
biases and other uncertainties. Moreover, different reactor
simulators may have been used in producing the Process Computer
Data and the Off-line Model Data. Such different reactor simulators
may implement the simulation of reactor operations by using
substantially different computational methodologies. Consequently,
output from these reactor simulators (even though provided with
identical input information) will often result in substantially
different reactor output data. Because the optimization execution
begins from the off-line simulator model, incorporation of the
differences between Process Computer Data 205 and Off-line Model
Data 206 is utilized to provide calibration to accommodate
differences between different simulation models. Using such prudent
calibration between different types of data allows optimization
predictions for future operation to be more accurate and
useful.
[0038] The third type of data stored in general database 201 is
Optimization Data 207. This type of data includes both current and
historic information used for optimization input specification,
strategy definition inputs, and optimization output results.
(Optimization Inputs and Optimization Outputs are discussed in
greater detail below with respect to FIG. 5, FIG. 6A, FIG. 6B, and
FIG. 7). Database 201 may be accessed and/or updated both manually
by a user through GUI 203 and automatically by the system during
automated processing 202.
[0039] FIG. 3 provides an example list (304) of typical operational
strategy issues that are likely to be considered, for example, by
an engineer/operator when deciding whether a change in operating
strategy or a new operation strategy should be implemented. For
example, after viewing core performance from the process computer
data, core performance from a simulation model, and/or optimization
inputs and outputs for projected future optimization through the
graphical user interface (102), one may desire to implement changes
(103) to the predefined operational inputs strategy definition
using the GUI (104). The GUI may also provide a selection menu
listing for editing or selecting predetermined strategy profiles
and/or optimization inputs for affecting a change in operational
strategy and the equivalent constraints. Strategy changes affecting
predetermined values of parameters may be implemented, for example,
via GUI 104. Listed immediately below are a few example reasons why
implementing one or more strategy constraint change may be
desirable:
[0040] to increase reactor full-power cycle length;
[0041] to provide additional reactor margin if needed for thermal
limits or reactivity margin limits;
[0042] to recover from erroneous predictions produced by the
reactor core simulator/simulation runs;
[0043] to perform sensitivity studies between reactor parameter
control-variables;
[0044] to recover energy during flow window modifications;
[0045] to accommodate a desire for longer or shorter control blade
sequence lengths;
[0046] to reduce impact due to a leaking bundle of radioactive
material;
[0047] to minimize release of radioactive material from a leaking
fuel bundle;
[0048] to improve reactor parameter robustness;
[0049] to change control blade group selection;
[0050] to recover from changes in design due to licensing
modifications;
[0051] to perform alternate blade management;
[0052] to store additional reactivity in following cycles for
improved long term cycle economic efficiency; and
[0053] to enhance overall operational flexibility.
[0054] FIG. 4 illustrates a flowchart of an example procedure for
computing and communicating an automated modification of
optimization inputs of a predefined operational strategy as a
consequence of a change in the last state-point of an operating
reactor plant. A new updated state-point is determined and, using
data from the general database 401, a comparison 402 is performed
to determine if the most recent updated state-point is different
than the state-point obtained from a previously run simulation. If
the latest state-point has not changed (403), state-point
comparisons 402 are continued. If the state-point has changed
(404), a copy of the new state-point is copied to Optimization
Inputs Database 409. In addition, a small change is made to the
operational strategy (405) to reflect the change in the starting
exposure. With the strategy starting point update and the small
modification made to reflect the new starting point time, a
optimization request flag is set (406) to identify the system for
an optimization request. Since this aspect of the present invention
is automated and requires no human intervention, notification of
the automated implementation is provided via e-mail 408 to a
predetermined specified distribution of recipients 407.
[0055] FIG. 5 is a block diagram illustrating example contents of
an Optimization Input database (503) as a result of updated
computed reactor state-point information (501) or manually input
modifications. As described above Optimization Input database
modifications may originate from the automated process loop
operations 501 or the manual process loop operations via graphical
user interface 502. For example, Optimization Inputs may include,
but are not limited to data and parameters such as: state-point
location, state-point filename, simulator model, allowable
independent variables to optimize, optimization controls, preferred
cycle exposure to be modeled, allowable blade movement ranges,
starting depletion example, design constraints, constraint weights,
licensing limits, target hot reactivity bias, target cold
reactivity bias, search breadth for independent variables, response
surface breadth, and optimization request flag. Each of these
optimization inputs are stored in Optimization Input database 503
for use during Optimization process execution 504.
[0056] FIG. 6A illustrates example steps performed by optimization
software running on one or more of the host processors (10) for
performing the control-variable optimization process of the present
invention. When Optimization Inputs are initially placed in
Optimization Inputs database, an optimization request flag is set.
During an optimization process run, the values of the Optimization
Inputs are obtained from the database (601) only after the request
flag is tested (602) to determine if any changes have been made to
the Optimization Inputs. If no changes to the Optimization Inputs
have been made (i.e., request flag not set), the process reverts to
the beginning entry point of the current manual or automatic loop
(FIG. 1B). If changes have been made, the new values are obtained
from the database (601) and an optimization computation process is
queued to run on a host processor (603). Once the optimization
process computations are completed (604), the results are stored to
the general database (605).
[0057] FIG. 6B illustrates a flowchart of an example optimization
process computational engine. At the outset, as indicated at block
612, the most recent simulation state-point (501) information and
user-specified optimization constraints (505) are obtained from
Optimization Inputs database 611. Next, at block 613, the
processing of two reactor simulator cases is initiated for each
independent variable in order to determine the functional
relationship of dependent variables to a change in a specified
independent variable. Next, at block 614, the generation of a
polynomial response surface is determined by solving for the
coefficients of the polynomial. (The response surface transfer
functions being normalized about the center-point to prolong
usefulness during the optimization phase). Since there may be as
many as several hundred independent variables, and a couple hundred
thousand dependent variables for each independent variable, the
above processing may potentially result in producing millions of
polynomial response surface transfer functions.
[0058] Once the transfer function polynomial response surface is
generated, it can be used to "predict" the response of the
dependent variables for a given change in value of an independent
variable 615. Consequently, computing simulated value changes for
each of the independent variables provides an estimate of an
optimum modification (i.e., change in quantitative value) which may
be made to each independent variable. When such predictions
indicate that an improvement exists relative to a previous
iteration, the scenario is simulated using a reactor core operation
simulator which may, for example, be a conventional core simulation
program or process performed by one or more other host processors
coupled to the network. A looping (619) of computing polynomial
response surface predictions (615) and performing simulator
calibrations/corrections is repeated until either: 1) the response
surface becomes inaccurate, 2) a predetermined number of iterations
is reached, or 3) until no further significant improvements to the
computed solution are realized. Once loop 619 is exited, the range
of the independent variable selection is reduced (616) and a new
response surface is regenerated via processing "loop" 620. This
larger response surface computation loop (620) is pursued until
changes to an independent variable no longer improve the computed
solution by a predetermined margin--which may be specified by
optimization the user--input constraints. Once the optimization is
complete, computed optimization output values (617) are stored in
an Optimization Output Results database (618), which may be part of
the general (central) database.
[0059] FIG. 7 is a block diagram illustrating example contents of
information stored in an Optimization Output database 702, provide
on a storage device in the system network. Three primary categories
of optimization database contents are illustrated which include: 1)
optimization status data 704, 2) optimization independent
control-variables 705, and 3) resulting optimization dependent
variable output predictions 706. The Optimization Status data 704
may include, but is not limited to, comparison results to design
values, cycle length improvement, optimization results,
optimization path, optimization status, and strategy comparisons.
The Optimization independent Control-Variables 705 may include, for
example, the location of the preferable control blades and
equivalent blade groupings at all future requested exposures, the
preferable core average flow at all future requested exposures, and
the preferable sequence exchange exposure intervals. The
Optimization Dependent variable output predictions, 706, may
include (but are not limited to), for example, LHGR results, CPR
results, cycle exposure, bundle exposure, core average exposure,
blade depletions, core inlet enthalpy, LPRM data, hot reactivity
bias, cold reactivity bias, thermal power, and electric power.
[0060] The system and method of the present invention as described
above may significantly improve the economic efficiency of
operating nuclear reactors by providing suggested specifications of
the operational control-variables that maximize energy and cycle
length while providing the same or greater design margins needed to
perform safe and flexible operation. Other practical uses of the
present invention may include, but are not limited to:
[0061] Continuous optimization based on continuously updated
reactor state-points;
[0062] Optimization based on operating plant history;
[0063] Optimization based on response to changing operational
requirements;
[0064] Continuous maintenance optimal projection;
[0065] Conversion of operating margin into energy or cycle
length;
[0066] Conversion of energy or cycle length into operating
margin;
[0067] Recovery from poor 3D simulation predictions;
[0068] Automatic updating of projections for use in next cycle
design and licensing;
[0069] Energy storage during current cycle for following cycle fuel
cycle efficiency improvements;
[0070] Reactor operator feedback on optimal operations
recommendations;
[0071] Reactor operator feedback on sensitivity of all independent
variables;
[0072] Ability to receive and optimize operational
control-variables using exposure and/or location dependent
constraints;
[0073] Remote access to centralized database;
[0074] Customer access to vendor hardware;
[0075] Secure access by customer user to customer specific
data;
[0076] Ability to perform multiple solutions from varying
constraints;
[0077] Determining economic cost of margin;
[0078] Extending reactor cycle length;
[0079] Improving reactor limit margin;
[0080] Recovering energy due to loss in reactor flow
capability;
[0081] Ability to reduce number of required sequence exchanges;
[0082] Increasing length of a given sequence exchange interval;
[0083] Ability to respond to emergency conditions such as pump
loss, leakage control, scram recovery, etc.;
[0084] Ability to respond to changes in licensing requirements of a
nuclear plant; and
[0085] Comparing various optimization strategies.
[0086] The method of the invention presented herein and described
above may be practiced using most any type of computer network or
interconnected system of processors having sufficient processing
speed and associated data storage capacity and is not necessarily
intended to be limited to any particular type of data processor or
network. Moreover, software elements of the present invention may
be operative as one or more modules and may be embodied on a
computer-readable medium for ease of transport between and/or
installation on one or more host processors/computers in a
networked environment. In addition, the method and system presented
herein are applicable toward optimizing the operations of many
different types of reactor plants, including but not limited to
boiling water reactors (BWRs) and pressurized water reactors
(PWRs).
[0087] While the invention has been described in connection with
what is presently considered to be the most practical and preferred
embodiment, it is to be understood that the invention is not to be
limited to the disclosed embodiment, but on the contrary, is
intended to cover various modifications and equivalent arrangements
included within the spirit and scope of the appended claims.
* * * * *