|
The primary goal of the proposed program is to advance research
in the area of life cycle management (LCM) of nuclear power
plant systems and renewal of energy infrastructure as a whole.
Life cycle management is
|
"an integrated approach to maximize plant value through a
scientific approach to condition assessment, inspection,
maintenance, and replacement activities that are systematically
scheduled throughout the service life of a system."
|
|
Since system parameters, operational conditions, costs, benefits
and other performance criteria tend to be uncertain, the concept
of “risk” is integrated with life cycle management philosophy.
The term “risk” combines probability of an event with its
consequence, and so it serves as a meaningful indicator of
system performance.
Program Overview
Life cycle management (LCM) is a process for timely detection
and mitigation of aging effects in systems, structures and
components (SSCs) important to plant safety, reliability and
economics. It is also considered a decision making process to
choose the best and balanced alternative for asset management.
In the nuclear plant, LCM defines in-service inspection and
maintenance programs, outage/generation plan and cost and
investment planning.
“Risk-Informed Decision Making” has become a cornerstone of the
regulatory framework world wide. Uncertainties about the
condition of existing assets and effects of aging are other
significant confounding factors in life cycle management.
Therefore, integration of techniques of risk and reliability
analysis with LCM is important.
As many plants in Canada are approaching the end of life,
cost-effective decisions are essential for the success of
refurbishment projects. The balance between “fitness for
service” and “asset preservation” is evolving over time. Fitness
for service is important for operation in the short term,
whereas asset preservation is important for refurbishment. This
suggests that LCM should be a dynamic model involving all
aspects of uncertainties.
The main research objective of the Chair program at the
University of Waterloo is the development and integration of
reliability models with LCM of NPP systems, structures and
components. The program is focused on
- Developing probabilistic models for risk analysis
- Benchmarking existing standards and FFS methodologies
- Solving a wide variety of practical problems related to
reliability of nuclear plant systems
- Supplying HQP to the industry
- Graduate Students (PhD, MASc), Post-Docs
- Education/Training of plant engineers (Undergraduates)
Fundamental research has focused on probabilistic risk and
reliability modeling, stochastic processes, statistical
estimation and extreme value analysis. Practical applications of
the basic research include risk-informed LCM of fuel channels,
steam generators, feeders and conventional systems, such as
electrical generators and transformers. These applications have
begun to yield considerable benefits to the operation and
maintenance of nuclear power plants in Canada.
Research Projects
The primary goal of the research program is to advance
research in the area of life cycle management (LCM) of nuclear
power plant systems and renewal of energy infrastructure as a
whole. The core research capabilities are in the areas of
reliability modeling, LCM and health monitoring techniques, as
shown in Figure 1.
Figure 1. Core research capabilities of the IRC
program.
A number of research projects have been initiated and
completed to address the effect of aging and improving life
cycle management of major elements of the primary heat transport
system (PHTS), such as fuel channels, feeders and steam generators.
Other conventional systems, such as main generators,
transformers and piping in the balance of the plant have also
been investigated.
These projects are not only relevant to the UNENE stakeholders
in the short term, but also generate knowledge to improve life
cycle management in the long term. Some of the key projects
initiated or completed in the past several years are summarized
below.
Back to Top
Feeder Projects
LCM Model
The wall thinning of feeder piping due to flow accelerated
corrosion (FAC) is a serious form of degradation affecting
nuclear piping systems. We have developed probabilistic methods
for estimating the corrosion rates taking into account probe
measurement errors and other uncertainties. The models are now
being applied to optimize the timing and sampling involved with
feeder inspection and replacement at operating stations. The
occurrence of leaks and ruptures of conventional piping in a
plant have also been modelled as a probabilistic process while
considering the impact of plant vintage and increase in failure
rate due to aging.
Figure 2. A schematic of a feeder section.
Figure 3. An application of feeder LCM model for
predicting the number of feeder replacements.
Feeder Cracking Susceptibility Analysis
We have developed a variogram approach for statistical
characterization of spatial distribution of Taylor factors in
feeder bends. The research objective is to develop indicators to
distinguish the susceptibility of feeder cracking in different
stations. Samples of the Taylor factor distributions are shown
in Figure 4.
Figure 4. Second bend samples from outlet feeders of
two different reactors.
Back to Top
Fuel Channel Projects
The performance of fuel channels in CANDU reactors are
affected by pressure tube degradation mechanisms such as
irradiation enhanced deformation (e.g., elongation, diametral
expansion, sag, and wall thinning), delayed hydride cracking and
changes in material properties. Due to large uncertainties
associated with the degradation mechanisms, limited data, and
high cost of inspection, the industry has emphasized the need
for a prudent risk-informed framework for managing the reactor
core in existing plants. Various sub-projects related to the LCM
of fuel channels are briefly described below.
Figure 5. A schematic of a CANDU fuel channel.
Probabilistic Leak-Before-Break (LBB) Assessment
The LBB assessment of pressure tubes is intended to
demonstrate that in the event of through wall cracking of the
tube, there will be sufficient time following leak detection for
a controlled shutdown of the reactor prior to the rupture of the
pressure tube. CSA Standard N285.8 (2005) has specified
deterministic and probabilistic methods for LBB assessment.
Although the deterministic method is simple, the associated
degree of conservatism is not quantified and it does not provide
a risk-informed basis for the fitness for service assessment. On
the other hand, full probabilistic methods based on simulation
require excessive amount of information and computation time,
making them impractical for routine LBB assessment work. We have
therefore developed a risk-informed approach by calibrating the
existing criterion to specified target probabilities of rupture
based on the concept of partial factors.
Delayed Hydride Cracking Assessment
CSA-Standard N285.8 recommends deterministic and
probabilistic procedures for the assessment of potential for DHC
initiation in planar flaws. We have proposed an innovative
method in which the deterministic assessment criterion of CSA
N285.8 Standard is calibrated to specified target probabilities
of DHC initiation using the concept of partial factors. The main
advantage of this approach is that it provides a practical,
risk-informed basis for DHC initiation assessment while
retaining the simplicity of the deterministic method.
Probabilistic End of Life (EOL) Assessment
Advanced random process models developed in the program have
been applied to the analysis of inspection data regarding
dimensional changes in pressure tubes and for predicting the end
of life distribution. We have analyzed reactor specific data for
diametral strain to estimate the distribution of remaining life
of pressure tubes and the number of tubes unfit for service in a
prescribed operating interval.
Material Variability Analysis
The nuclear industry has undertaken comprehensive
experimental programs to determine the material properties, such
as fracture toughness (KC), DHC threshold toughness (KIH) and
DHC rate using the samples taken from the in-service pressure
tubes removed from the reactor core. We have developed in
collaboration with AECL researchers multivariate statistical
models to predict the distribution and bounds of material
properties that are useful in the fitness for service assessment
of pressure tubes.
Risk-Informed Inspection of Pressure Tubes: Optimum
Sample Size
A key question in deciding the scope of pressure tube
inspection is the sample size, as it affects the outage time as
cost of inspection program. A case in point is the inspection
for flaws that could become potential sites of DHC. We have
developed a risk informed model to estimate the sample size. For
example, if we want to demonstrate through inspection that
probability of DHC initiation is less than 0.01, then the
optimal sample size as a function of statistical efficiency can
be estimated using our model, as shown in Figure 6.
Figure 6. Risk informed selection of sample size for
inspection of pressure tubes.
Back to Top
Steam Generator Projects
Alloy 800 Lifetime Assessment
Alloy 800 tubing has demonstrated excellent resistance to
in-service degradation modes, in particular stress corrosion
cracking (SCC). In planning refurbishment of a CANDU station, a
key concern is the longevity of existing steam generators up to
60 year lifetime of the refurbished plant. We have developed
probabilistic methods for estimating the lifetime distribution
of Alloy 800 and 600 SG tubing using in-service experience and
test data. We are also participating with a COG working group
for experimental evaluation of Alloy 800 material.
Impact of Inspection Uncertainties in the Assessment of
SG Tubing
In this project, we analyzed the influence of inspection
uncertainties on the maintenance and replacement of SG tubing.
Our analysis has shown that ignoring inspection uncertainties
(such as POD and sizing error) will lead to highly biased
decisions.
Probabilistic Modelling of Pitting Corrosion
We have developed a methodology to analyze all historically
reported inspection data at a nuclear station and prepared a life
cycle management approach. We have also correlated the extent of
pitting with the distribution of sludge in the boilers (Figure
7).
Figure 7. Correlation of sludge distribution with
pitting in a steam generator.
Back to Top
Generation Risk Assessment (GRA)
The objective of the project is to integrate the results of
the advanced degradation models of CANDU systems developed by
the Chair into a system-wide assessment of generation risk. Key
aging mechanisms affecting the availability of HTS systems are
shown in Figure 8.
Figure 8. Key HTS aging mechanisms.
The project includes a pilot study of the primary heat
transport system at an existing nuclear station. We have
developed degradation models using the station specific data and
then quantified the forced loss rate for the station.
The GRA model has been applied to study a number of scenarios
of plant maintenance and refurbishment, such as HTS pump
vibration monitoring, extending testing interval of liquid
relief valves (LRV) from 3 years to 6 years, aging of HTS pump
seals, and the FAC of feeders. Figure 9 shows results of
optimization of the time of refurbishment that would minimize
the generation risk. The unit cost in this figure refers to the
unit cost of feeder replacement.
Figure 9. HTS system generation risk versus
refurbishment considering feeder degradation.
The developed GRA model to more accurately assess the risk
and likelihood of failure over time will not only support the
assessment of the timing and scope of refurbishment and
replacement decisions, but also increase the flexibility,
stability, and reliability of a plant’s capital investment
planning and decision-making processes.
Risk-Informed Asset Management of Main Generator and
Transformers
The operation of conventional systems, such as the
turbine/generator and large transformers is often second in
criticality only to the operation of the reactor itself. Because
these components require long repair times, their failures can
be significant contributors to lost power generation and plant
trips. The developed life cycle cost analysis model incorporates
time-dependent failure rates and uncertainty in costs and
unavailability parameters. This model has been used to support
the life cycle management of large transformers at a CANDU
station.
Back to Top
Seismic Risk Analysis
Seismic design loads for systems, structures and components (SSCs)
in old nuclear stations were evaluated using the Design Basis
Earthquake (DBE) response spectrum defined by prevailing
practices and standards of the nuclear industry in the 1970s,
which did not have consistent probabilistic basis. As old CANDU
stations are planning for refurbishment, a seismic gap analysis
between an existing structural design and new Code requirements
based on probabilistic methods is necessary. To support a case
for refurbishment of a plant, we have investigated the
differences between the original DBE and response spectra
calculated on the basis of modern standards and regulations.
Back to Top
Research Significance
The refurbishment and renewal of nuclear reactors is a
complicated process with several critical issues that require
prudent planning and cost-effective management. The research
program on risk-based management has significant potential for
improving the performance and extending the service life of
safety critical structures, and improving the effectiveness of
operating costs in power generation plants. An additional
benefit will be the improvement in overall fleet management
capability. The general impacts of life cycle
management are summarized in Figure 10 below.
Figure 10. The impacts of implementing a life cycle management program in a power plant.
The direct benefits of the many projects undertaken by the NSERC-UNENE
industrial research chair include
- Effective fitness-for-service assessment of the reactor PHTS
- Improved communication with the regulator about managing the
risk associated with degradation
- Minimize cost penalties associated with increased inspection
and outage duration
- Overall improvement in operational efficiency
Back to Top
Interaction with Industry
A key objective of the IRC program is to facilitate the
transfer of knowledge to industry through applications of the
results of the fundamental research to practical industry
problems. Active collaboration and consultation with UNENE’s
member industries has continued within conferences, workshops,
and industry working groups, as well as through the publication
of numerous technical papers and memoranda.
The Chair has maintained an active and on-going role in knowledge transfer
within the broader nuclear industry. Most recently, the
Chair has acted as a member of the organizing committee for the CNS
8th International Conf. on CANDU Maintenance, held in November
2008, in Toronto, Ontario. The Chair also offered well attended
workshops in “Engineering Reliability and Life Cycle Management”
as part of the International Conference on Nuclear Engineering
organized by the American Society for Mechanical Engineers (ASME)
in May 2008 (ICONE-16), in Orlando, Florida and June 2009
(ICONE-17) in Brussels, Belgium.
Back to Top
|
