Page 34 - Nuclear Plant Journal - March April 2013
Basic HTML Version
34
Nuclear Plant Journal, March-April 2013
On-line Monitoring...
and asset management system designed
to capture and incorporate knowledge,
experience, and intelligence from
diversified operating systems and
monitoring environments. Advanced
pattern recognition modeling is a
fundamentally different approach from
data trending analysis.With data trending,
the primary knowledge source is trained
plant personnel, who can view trends and
understand when data has changed based
on their experience of the system. With
advanced pattern recognition modeling,
the primary knowledge source is a trained
numerical model, and in many cases,
the numerical models can automatically
detect much smaller changes than is
possible with data trending. Further,
numerical models are amenable to higher
levels of automation and integration
into advanced information management
systems, thereby supporting modern
business management programs.
Digital measurement and signal
processing
- Monitoring technologies
enable more effective analysis of systems,
structures, and components (SSCs).
Nuclear plants were originally equipped
with a minimum set of measurements and
signals to support plant operation and
protection systems. These measurements
are primarily based on the original
analog technologies available during the
plant design and construction phases.
To more fully assess SSC conditions,
additional measurements are needed as
inputs to monitoring programs. Vibration,
temperature and flow monitoring of
specific equipment are examples of
measurements required for modeling and
analysis of SSCs. Data preprocessing
also is important because large amounts
of data do not provide useful actionable
information. Preprocessing technologies
are used to condense large amounts
of data into organized sets of data for
efficient use in analytical programs.
Modeling of systemand component
behaviors
-The application of monitoring
technologies
requires
a
greater
understanding of system and component
behaviors as precursors to model
development. Monitoring programs do
not have inherent intelligence about the
behavior of monitored equipment or
systems. Application of these programs
generated a need for greater intelligence
and research in support of component
behavior and the modeling of monitored
SSCs.
Analytical methods to optimize
measurement and data requirements
-
Performance and condition analysis
identified the need for advanced analytical
methods to define the measurement and
data requirements for classes of SSCs.
The integration of data processing, analog
systems, digital systems, and variations in
measurement quantity and quality results
in low-value data inputs to the analytical
engines. High-performance results for
effectiveness required development
of analytical methods to optimize
data requirements and supporting
measurements.
Information Management and
industrial business applications
- A key
functionalrequirementofawell-developed
monitoring program is its information
interface with the operating plant and
associated staff. The integration of
advanced information detailing operating
conditions of a wide range of plant assets
into individual plants is an important step
in implementing monitoring programs.
This step must be analyzed, designed,
planned, published, and managed for
success and maximize the return on the
OLM program investment.
4.
Will the monitoring system help
extend the life of the plants? If so, please
explain how
.
OLM systems are effective at
identifying equipment operating in
conditions that may shorten their
remaining life. Monitoring programs
inform life extension decisions by
identifying conditions that contribute
to asset life limiting or reduction
behaviors. They also provide the ability
to gather substantially more data through
automated means and to analyze and trend
performance using new methods to make
more informed decisions about nuclear
plant operation and safety management.
OLM programs process sensor
signals to derive parameter estimates of
specific aging and performance features
required to characterize the state and
condition of material properties. This
information can then be used to refine
“diagnostic” assessments about material
aging and degradation, enabling more
accurate understanding of a system’s
ability to achieve its design function
and the types and timing of needed
interventions. Of particular importance
will be the capability to determine the
“remaining useful life” of a component
to justify its continued operation over
an extended plant life or until the next
scheduled maintenance opportunity.
Exelon is applying EPRI-developed
fleetwide online monitoring technology
at its 17 nuclear units to improve plant
productivity, equipment reliability, and
optimize long-term plant management.
Contact: Richard Rusaw, Electric
Power Research Institute, email:
.
