VGB PowerTech

Areva´s Monitoring Concept
The Areva Integrated and Sustainable Concept of
Fatigue Design, Monitoring and Re-assessment
The paper was given at the “ANSYS Conference & 26th CADFEM Users’ Meeting 2008”,
Darmstadt/Germany, October 22 to 24, 2008
Jürgen Rudolph and Steffen Bergholz
Kurzfassung
VGB PowerTech · Autorenexemplar · © 2008
Areva Fatigue Concept (AFC) –
Ein geschlossenes Konzept zur
Verfolgung und Optimierung der auf
Ermüdung basierenden
Schädigungseffekte von thermisch
und mechanisch beanspruchten
Kraftwerksbauteilen
Die Vermeidung von Ermüdungsschäden an
Komponenten der nuklearen Kraftwerkstechnik ist ein wichtiges Thema in Hinblick auf die
Aufrechterhaltung der Anlagensicherheit sowie
auf die Erhöhung der Verfügbarkeit. Die Ermüdungsüberwachung ist integrierter Bestandteil
des Alterungs- und Lebensdauermanagements. Vor diesem Hintergrund und den sich
ändernden Rahmenbedingungen, wie z.B.
Neubau von KKW’s mit 60Jahren Laufzeit, der
Laufzeitverlängerung bestehender Anlagen
und der Berücksichtigung von medienbedingten Effekten, ist die Verbesserung der Nachweismethoden ein Hauptanliegen der F&EAktivitäten von Areva. Dabei wird das Hauptziel
verfolgt, den aktuellen Erschöpfungsgrad der
Kraftwerksbauteile realitätsnäher zu bestimmen. Der vorgeschlagene integrale und mehrstufige Ansatz trägt gleichzeitig zur Kostenoptimierung und Prüfzyklenminimierung bei.
Die Ermüdungsüberwachung sollte bereits in
der Inbetriebsetzungsphase einsetzen, da die
hohen auftretenden Lasten bereits einen signifikanten Beitrag zum Erschöpfungsgrad leisten. Areva empfiehlt den Einsatz des Systems
FAMOS (Fatigue Monitoring System) zur Bestimmung der realen transienten Lasten und
Weiterverarbeitung im Sinne einer ersten Ermüdungsschnellbewertung.
In einem mittelfristigen Zeitrahmen sollten die
aktuellen Erschöpfungsgrade mittels einer detaillierten regelwerkskonformen Ermüdungsanalyse bestimmt werden. Eingangsdaten dafür sind die realen gemessenen Lasten, wobei
die Spezifikation der Temperaturtransienten
aus systemtechnischer Expertensicht erfolgt.
Die Ermüdungsnachweise basieren auf FiniteElemente-Analysen unter Berücksichtigung
der transienten Temperaturfeldbelastungen am
Bauteil sowie der primären Belastungen. Potenzielle Problemzonen können auf diese Art
und Weise zuverlässig identifiziert werden, was
wiederum einen Beitrag zur Inspektionskostenreduktion leistet.
Authors
Dr.-Ing. Jürgen Rudolph
Dipl.-Ing. Steffen Bergholz
AREVA NP GmbH
Erlangen/Germany.
58
Introduction
F i g u r e 1 shows the typical arrangement of
components within the primary circuit of a
pressurised water reactor (PWR). The possible layout of measuring points for in-service
fatigue monitoring is marked with full circles.
Thus, the main role of the temperature measurement system can easily be identified.
The principal modules of the Areva Fatigue
Concept (AFC) are schematically shown and
logically combined in F i g u r e 2 .
It is obvious that for all plant life management
activities, the quality of all fatigue analyses
crucially depends on the determination of the
real operational loads in the plant.
However, the first analysis is carried out even
before the initial start-up. In this design stage,
the conservative approach may yield usage
factors above 1.0 requiring more refined analyses during operation. The real transient loads
are recorded and processed on site based on
FAMOS (Fatigue Monitoring System). A simplified fatigue estimation method (one branch
in Figure 2) is directly combined with the
measuring system. This simplified fatigue estimation is carried out after every operational
cycle. This procedure is highly automated and
allows for an estimation of the recent usage
factor. In this process, implausible measured
data have to be eliminated. A further division
into classes is carried out and usage factors
are estimated by application of the rainflow
algorithm.
As temperatures are measured on the
outer wall and fatigue relevant locations are
usually located on the inner wall the inverse
temperature field problem has to be solved
appropriately as one modular task for the
specification of realistic transients and subsequent thermomechanical analyses. Furthermore, temperatures are measured at thinwalled tubes within the piping system. So,
temperature and stress profiles due to the
measurements have to be transformed into
profiles at the fatigue relevant locations (e.g.
thick cross sections, nozzles, flanges and pipe
bends).
In the framework of the periodic safety inspection (PSI) a detailed fatigue check conforming
to the code rules is carried out in order to determine the current state of the plant (see
workflow in Figure 2). This fatigue check is
based on monitored loads and updated specification of thermal transient loads and finite element (FE) analyses in connection with the local strain approach (strain life approach) to
design against fatigue. There is still potential
with respect to a less conservative specification of the thermal loads. Practicability still
demands for the definition of some covering
unit transients and the allocation of the real
number of occurrences (pressure, temperature
and mass flows). These unit transients are used
as load data in the FE stress and strain analyses and evaluated in terms of the applied design code (ASME, KTA, RCC-M [1, 2 and
3]). The intermediate step of load determination by computational fluid dynamics (CFD)
might be part of the workflow and will gain in
importance in the future. The determination of
realistic heat transfer coefficients as a function
of time is an important task. The FE analyses
always embrace transient thermal determination of the temperature field and subsequent
determination of (local) stresses and strains.
The latter analyses might be simplified elastic
plastic or fully elastic plastic. One code requirement is the additional check against progressive plastic deformation (ratcheting).
All conceptual activities are supported by respective R&D activities as well as the development of specialised methods and tools (see
lower workflow of Figure 2). Some of these
activities will be exemplarily addressed.
Figure 1. Primary circuit of a pressurised water
reactor.
VGB PowerTech 5/2009
Erection
Areva´s Monitoring Concept
Design data
• component design
• material data
• design & operating loads
Design analyses
stress/fatigue &
ratcheting
Approval documents
------------------------Basis for monitoring
and verification
Licensed status
Cycle counting
methods
Load data evaluation
Ageing and life time
management
Operation
Measured data
FAMOS
Transient
catalogue
(realistic transients)
Simplified
fatigue estimation
• Plant optimization
(operating modes, in-service
inspection, maintenance)
• Component availability
Detailed code-based
fatigue & ratcheting
analyses
Reliable verification
documents
• Early detection of failures
Supporting functions:
R&D
Further development
of codes
Methods
Tools
Figure 2. Principal modules of the AFC.
VGB PowerTech · Autorenexemplar · © 2008
First Design Analysis
before Operation
Before commissioning and operation of the
plant a catalogue of presumed thermal transients is compiled. Consequently, these specified thermal loads have to be considered as
design transients in contrast to the real transients based on temperature measurement during operation. In the past, the anticipated transients covered 40 years. Now, the period to be
considered is even 60 years. The specification
is done for level A and level B, e.g. normal and
upset conditions as well as partially levels C
and P (emergency and testing). The code based
fatigue analysis considers level A and level B
loads in the design phase. Design transients
before operation are specified according to
different plant models and experiences. As a
general rule, the specified transients are conservative with respect to frequency of occurrence, temperature range, rate of temperature
change and load type (thermal shock, thermal
stratification). The usage factors calculated in
the design fatigue analysis will considerably
differ from the results of the more detailed
consideration of the real operational (thermal)
loads. As a consequence, usage factors around
1.0 are still tolerable in the design phase. They
indicate the positions of further analysis which
is another aspect of the advantageous application of the first design analysis before operation. This reliable identification of points of
VGB PowerTech 5/2009
interest is valuable under different aspects. For
example, these locations are selected for future
instrumentation and non-destructive testing
(NDT). Thus, the pre-operational analysis is a
valuable aid for the specification of the testing
concept. It allows for an estimation of the
amount of testing required. Furthermore, it is
still possible to exert influence on the design
and construction of the components and the
operation mode of the plant. The consequences
of non-optimal operation modes can clearly be
shown. Plant staff is sensitised from the very
beginning to the influence of operation on the
lifetime of components. It contributes to the
development of a deepened consciousness of
responsibility and safety.
Thermal transients with low influence on the
fatigue behaviour are identified as well. The
design codes offer procedures for the exemption from fatigue analysis for non significant
loads. The so-called six points check should be
mentioned. This procedure is specified for primary circuit components in the ASME code
[1], section III, division 1, paragraph NB3222.4 and nearly identically in the German
KTA 3201.2 [2], paragraph 7.8.2. Note that all
six points have to be simultaneously fulfilled.
The points concerning pressure and temperature transients are of special interest in this
context. Insignificant pressure ranges ∆pNS are
calculated as a function of the design pressure
pD, the admissible stress amplitude at the endurance limit Sa(N = 106) and the Sm value of
the material at operational temperature SmT:
1
Sa(N = 106)
∆pNS < – · pD · –––––––––
3
SmT
(1)
Insignificant temperature ranges ∆TNS are
calculated as a function of the admissible
stress amplitude at the endurance limit
Sa (N = 106), Young’s modulus E and the linear thermal expansion coefficient α:
Sa(N = 106)
∆T < –––––––––
2·E·α
(2)
As far as non-primary circuit (class 2) components are concerned the German KTA rules
give further exemption criteria in KTA 3211.2,
paragraph 7.8.1.2.1.
It should be mentioned that section B3000
(Design) of the recent issues of the French
RCC-M code [3] does not contain a comparable procedure for a simplified fatigue check.
Hence, it depends on the code to be applied if
this kind of fatigue analysis is admissible for
nuclear power plant components.
Areva’s Advanced Fatigue
Monitoring System (FAMOS)
The acquisition of realistic operational data in
the power plant is one essential pillar of the
AFC. Its role as the basis for a fast fatigue estimation and detailed data processing (e.g. for
realistic load determination) is pointed up in
Figure 2. Areva disposes of a fatigue monitor-
59
Areva´s Monitoring Concept
Thermocouples
Mounting strap
Cover shell
T1
Hot
Cold
T7
T2
T3
T4
y [mm]
T5
T6
T [K]
Figure 3. FAMOS measurement section.
VGB PowerTech · Autorenexemplar · © 2008
ing system called FAMOS since the early
1980ies [4]. At that time, German licensing
authorities demanded for the realisation of a
comprehensive programme of measurement
in the nuclear power plant of Grafenrheinfeld/
Germany in order to get detailed information
on the real component loadings during plant
operation. It was shown that the real operating
conditions partially differed significantly
from the design data. It should be pointed out
that all measured data fell into admissible
limits. At this occasion, the advantages of
monitoring the real operating conditions and
using the acquired data as an input for fatigue
analyses based on realistic loading data became obvious. This gave a strong development impetus towards a sophisticated fatigue
monitoring system. As a consequence, the
German nuclear power plants as well as many
plants in other countries were equipped with
FAMOS (more than 20 nuclear power plants
since 1988; 20 to 50 measurement sections
and up to 150 thermocouples per plant).
The objectives of FAMOS [4] can be summarised as follows:
–– determining the fatigue status of the most
highly stressed components,
–– identifying operating modes which are unfavourable to fatigue,
–– establishing a basis for fatigue analysis
based on realistic operating loads and
–– using the results for lifetime management
and lifetime extension.
In fact, the practical application resulted in
various improvements of operating procedures, transient catalogues and fatigue results.
FAMOS proved to be a reliable basis for the
integrity concept, inspection plans and lifetime extension considerations.
In the following the technical bases of
FAMOS are shortly described. F i g u r e 3
shows the application of thermocouples at the
outer surface of a piping section. The standard
configuration consists of a mounting strap
with the thermocouples and a second clamp-
60
ing strap used to press the thermocouples to
the surface of the pipe. In 1996, a new measurement section which significantly improved
the thermal sensitivity was successfully introduced. The FAMOS special instrumentation
features consist of seven temperature transducers around one half of the pipe circumference in order to detect thermal stratification.
This arrangement of thermocouples resolves
the course of temperature as a function of
time and the vertical coordinate y. In case of
anticipated plug flow the application of only
two thermocouples (T1 and T7 according to
Figure 3) is sufficient.
The practical installation of FAMOS in a nuclear power plant starts with the generation of
a so-called FAMOS manual. This process
contains a weak point analysis identifying
components in the primary circuit and its
safety systems as well as the secondary circuit
that are expected to be particularly prone to
fatigue damage. All design documents and
operating experience are considered. A measuring point plan is elaborated and all activities
are co-ordinated with the plant operator and if
required with independent experts. As an example, points of interest within the primary
circuit of a nuclear power plant are marked
with full circles in Figure 1. In this case the
following components are fatigue monitoring
locations: the surgeline, the pressuriser spray
lines, the nozzle of the residual heat removal
system at the cold loop, the volume control
system, the nozzle at the primary loop, the recuperative heat exchanger and the feedwater
nozzle at the steam generator.
The measured temperatures are to be transferred to the fatigue most severe locations.
TInside, 2
TOutside, 2
TMedium, 2
TMedium, 1
TInside, 1
ṁ
TOutside, 1
FAMOS
TMedium
Figure 4. Determination of realistic inner wall
transients.
Design
&
construction
Operation modes
Thermodynamic
loading data
Stress &
fatiqueanalysis
Lifetimemonitoring
Figure 5. Interaction between operating mode,
loading data and stress & fatigue
analysis.
Nowadays, this is done by FE analyses based
on the measured data as input parameters. In
the future, this task should be fulfilled automatically considering the capacitive corrections between thin and thick cross sections as
it is shown F i g u r e 4 . This requires the simultaneous solution of the inverse temperature field problem and the automatic correction of the measuring tolerance deviation.
Thus, the required transient temperatures will
be directly available at the fatigue critical locations.
Of course, the transient temperatures Tinsid e,2
in Figure 4 can be used as input data for a
simplified as well as a detailed code based fatigue analysis. However, at least the detailed
code based fatigue analysis still requires the
specification of thermal transients as an intermediate step.
Detailed Data Processing and
Specification of Thermal Transients
An important connecting link between the recording of measurements and the stress and
fatigue analysis is the specification of thermal
loads. For this task it is essential to use the
specific knowledge of system and process engineers to identify the relevant load mechanism. Consequently, it is possible to develop
smarter operational modes or more favourable
hardware designs to achieve lower fatigue values which are necessary for an extended lifetime ( F i g u r e 5 ). Additionally, this procedure is a plausibility check of the available
data. As a result, implausible data are detected
and adjusted. The staff is sensitised to the
consequences of certain operational actions,
preferably beginning with the commissioning
and initial start-up phase. Note that the transients can be significantly different in one
power plant compared to another one due to
the numerous influences. In spite of the strict
regulations of the operational manuals the
operator still takes an active influence on the
operation of the power plant.
If there is no existing compilation of the load
cases which must be considered, e.g. from the
permission of the plant, such a compilation
must be created. This can be done specific for
the system or for the whole pressure retaining
boundary. In some cases such an overall load
VGB PowerTech 5/2009
Areva´s Monitoring Concept
case compilation was carried out for German
power plants for the pressure retaining boundary together with the secondary side systems
and the most important neighbouring systems
such as the chemical and volume control system (CVCS) and the residual heat removal
system (RHRS).
All the information is used for the specification of the plant specific thermodynamic
loads in official reports and transient handbooks.
For thermal loads three types of specification
are state of the art:
–– Design specification for new plants or
backfitting projects. This type of specification is based on theoretical analysis as well
as measurements in comparable plants. It is
necessary to reflect all possible load cases
in the lifetime of the plant.
–– Determination of the actual fatigue value
(note that it is only possible to examine an
upper limit of this number). Here all the
events in the present lifetime must be considered which are relevant for loading. Normally this exercise will be repeated every
ten years.
–– Overall specification for all possible and
hypothetical load cases which must be considered for the lifetime.
The first step to specify thermal loads is the
identification of the operational processes lead-
ing to relevant transients of temperature and/or
pressure. For these events an appropriate
number of model transients are selected. Usually, it is necessary to split up these model transients into subclasses which are different, e.g.
in the temperature difference of the transient.
In a next step all the (relevant) events in the
considered lifetime period are allocated to the
specified model transients (or the corresponding sub-classes) and all the factors defining
the model transient or the corresponding subclasses must be specified conservatively. All
allocated events must be covered. Occasionally, an iterative process is required in order to
achieve suitable model transients with an
acceptable measure of conservatism.
As the location of the existing measurements
is not identical to the locations of maximum
stresses it is necessary to translate the information of the outside transients to the inner
wall and the medium including the appropriate heat transfer coefficients (see Figure 4).
Special tools are used for the determination of
these parameters.
The thermodynamic loads are harmonised
with the system specification, the transient
handbook, the instructions of the operating
manual, isometric diagrams, data of the
FAMOS system and other inputs such as test
instructions, logic diagrams and other work
reports.
If hypothetical load cases must be considered
as well, the experience of normal events must
be transferred to these theoretical boundary
conditions (e.g. thermal stratification can be
expected between pressuriser temperature and
main coolant temperature). Knowledge of the
processes and the operation of the systems is
essential. As an example, a spray event under
emergency conditions can be derived from a
spray event of the start-up of the plant.
Normally, only part of the required information is available. Obviously, the exchange of
experiences and information between similar
plants is synergetic and is a valuable aid in the
permanent improvement of the entire process
of data evaluation (Figure 2). This holds especially true for the German KONVOI plants
which have a very similar basic design. If the
conditions are not identical it is still possible
to acquire additional information by transfer
and re-scaling to hypothetical boundary conditions. Again, missing information has to be
complemented based on an excellent knowledge of systems and processes of the plant.
Finally, the specified thermodynamic loading
data are the basis for the code based fatigue
and ratcheting analyses. It should be pointed
out that the quality and accuracy of the thermomechanical analyses highly depends on the
plausibility of the specified transient loads.
Together with the information about e.g. fluid
dynamical or seismic loads the following
ACHEMA 2009
Halle 10.1, Stand B22-C23
VGB PowerTech · Autorenexemplar · © 2008
Probenahme- und
Analysensysteme
• Thiedig-Probenahmeeinrichtungen
für alle Messaufgaben
• Engineering kompletter Systemlösungen
• Hohe Funktionalität und Lebensdauer
• 70 Jahre Know-how und Innovation
Digox 601 silica
• Medientransport erfolgt ohne Pumpen
• Geringer Chemikalienverbrauch
• Automatische Zweipunktkalibrierung
• Individuell wählbare Messhäufigkeit
für jeden Kanal
Dr. Thiedig – Engineering Solutions
info@thiedig.com • www.thiedig.com
Areva´s Monitoring Concept
tion and the detection of fatigue critical locations. Furthermore, the investigation of the
results allows for the detection of anomalies.
Begin
D ≥ 1.0
Component geometry
and (thermal) loading
Load-time-functions
(transients)
D < 1.0
End
Fatigue check
Usage factor
yes
Simplified check
no
Sn,m<3Sm
(steel)
Sn,m<4Sm
(cast iron)
yes
Elastic fatigue analysis
Counting method
according to ASMEcode + linear damage
accumulation rule
no
Sn,m<3Sm
(steel)
Sn,m<4Sm
(cast iron)
no
yes
Simplified elasticplastic fatigue analysis
Ke-approach
Fatigue curves (strain
lite curves plotted as
pseudostress
amplitudes)
Elastic-plastic fatigue
analysis
non-linear FE
yes (optional)
Ratcheting check
It is important to distinguish this cycle based
method for simplified stress based fatigue
(SSBF) from the frequently applied event
based fatigue method for the approximate assessment of fatigue damage increments based
on the recording and evaluation of so-called
counting lists of load cases. In the design
phase the calculated fatigue effects are related
to certain presumed events. Such an event, to
which a defined set of fatigue load cycles is
assigned, may be for instance the unit startup, the unit shutdown, power increase or decrease, exchange of water, emergency shutdown etc. The fatigue calculation assigns a
fraction of fatigue damage to each of these
events. The fatigue fraction is then multiplied
by the presumed number of occurrences of
the event during the component’s planned
service life. The cumulated fatigue is obtained
by linear summation of all fatigue fractions of
all the events considered (Miner’s rule).
Event-based fatigue methods compare the
numbers of occurrences of operational events
with the presumed events from the design
phase. The evaluation does not consider occurred loads. Thus, it is not investigated
whether the course of events during operation
complies with those from the design phase.
As an example, different start-up events may
cause different fatigue usage contributions
due to different numbers of hot standby hours
and the derived different thermal loads.
Figure 6. Code-based fatigue analysis.
steps in the chain of certification can be
done.
VGB PowerTech · Autorenexemplar · © 2008
It is pointed out that the FAMOS system
should be installed in the power plant from
the very beginning of operation, i.e. in the
commissioning and initial start-up phase. This
phase is often characterised by the highest
loads of the entire lifetime of the power plant.
If the loads of this phase are unknown the initial pre-damage state cannot be specified. The
specification of the initial starting-up phase
has to be done in a very conservative way in
that case.
Simplified Stress-based
Fatigue Estimation
The results of the temperature measurement
are to be processed quickly in order to get a
first estimation of the fatigue state. One important task before the simplified and automated evaluation is the verification of the acquired data. Detection and adjustment of implausible data are parts of this process. These
plausibility and quality checks of the measured data have to be done by experienced spe-
62
cialists. The result is a pre-processed database
for (automated) data evaluation and fatigue
assessment.
Detailed Code-based Fatigue
and Ratcheting Analysis
C o d e - b a s e d Fa t i g u e A n a ly s i s
In the very first step of the simplified fatigue
assessment procedure the changes of temperatures are subject to a rain flow cycle counting
algorithm [5]. In this process the temperature
ranges at the locations of measurement are
identified, counted and classified. These thermal load cycles are input data for a stress and
fatigue assessment of the monitored components based on conservative analytical computation formulae. This rough real time fatigue estimation is done after every operational cycle and allows for a direct comparison of
thermal loads and fatigue damage increments
as well as an evaluation of the actual fatigue
usage.
The detailed code-based fatigue analysis is
usually carried out after a certain time period
of plant operation (e.g. ten years) in the framework of the periodic safety inspection. Loading data of the operational period as well as
anticipated loads of future operation are used
as essential input parameters. Hence, usage
factors are calculated for the current state of
the plant and until the end of life (e.g. 40 or
60 years). The fatigue analysis for primary
circuit components is based on design codes
such as the ASME code [1], the German KTA
rules [2] or the French RCC-M code [3]. The
code based procedure [1, 2] is exemplified in
Figure 6.
The result of this simplified fatigue assessment is usually conservative (idealised thermal shock assumption, thermal stratification
with sharp separation). Although the correlation of the real temperature ranges is fairly
simple it is suitable for a comparison of different real sequences of loads and allows for a
qualitative evaluation of the mode of opera-
The simplified elastic-plastic fatigue analysis
(not to be confused with the simplified stressbased fatigue estimation explained above)
based on elastic FE analysis and plasticity
correction (fatigue penalty or strain concentration factors Ke) e.g. according to paragraph
7.8.4 of [2] or equally NB 3228.5 of [1] is
known to yield conservative results [6,7]. In
VGB PowerTech 5/2009
Areva´s Monitoring Concept
space. Classical kinematic material models
(Prager, Besseling etc.) often do not describe
this translation [11] in a sufficiently accurate
way. As a consequence, they fail in the simulation of experimentally proved load paths.
Both considerable over- and underestimations
are possible.
Allowable limit
1.0
gin
Mar
Fatigue
usage U
Commissioning and
initial start-up
Consideration of more
realistic operational data
Implementation
of improvements
Operating without
fatigue monitoring
0
Start of fatigue
monitoring
40 years
60 years
Figure 7. Development of fatigue usage factor considering local monitoring and improvements.
VGB PowerTech · Autorenexemplar · © 2008
the practical application this will often yield
high calculated usage factors. As a consequence, the branch of elastic-plastic fatigue
analysis methodology based on non-linear FE
analysis of Figure 6 will often be used for
fatigue design. This is associated with an increased calculation effort. Computing times
for complex 3D geometries and numerous
transients (coupling of transient thermal and
mechanical analyses) are significant. Under
these circumstances the specified transients
have to be rearranged in a small set (often 7
to 10) of covering transients for calculation
purposes. There is still potential for improvement at the interface between “transient catalogue” and “detailed code based fatigue &
ratcheting analyses” in Figure 2.
The possible modification of design codes
with respect to more severe fatigue curves [8]
and the consideration of environmental,
surface, load sequence and other effects [9]
will significantly influence the code based
fatigue design. Of course, these developments
are attentively followed and actively accompanied (see “supporting functions” in Figure 2). As an example, surface and load
sequence effects can be described on a mechanistic basis by application of short crack
fracture mechanics. For the practical fatigue
calculation process, conservative assumptions will have to be scrutinised in order to
compensate the more severe fatigue assessment approach.
Fi n i t e E l e m e n t M o d e l l i n g
Requirements
FE analyses as one essential pillar of the codebased fatigue approach have to meet strict
quality requirements. Results obtained by different commercial FE codes and by different
engineers must be comparable.
The appropriate idealisation and discretisation
strategy has to be formulated in terms of mandatory rules. Where applicable such rules
VGB PowerTech 5/2009
should be verified by checking calculations
and convergence tests. The elaboration and
formulation of modelling guidelines is part of
the quality management process.
Note that special rules apply for thermal transient analyses in terms of discretisation requirements and time step sizes.
The compliance with such rules ensures an
accurate determination of notch stresses, thermal stresses and stress gradients.
Ratcheting: Incremental
E l a s t i c - p l a s t i c A n a ly s i s
Ratcheting as a possible damage mechanism
is a design code relevant issue. The ratcheting
phenomenon is defined in paragraph NB3213.33 of the ASME code [1] as follows:
“Ratcheting is a progressive incremental inelastic deformation or strain which can occur
in a component that is subjected to variations
of mechanical stress, thermal stress, or both”.
In contrast, the term Shakedown is defined in
paragraph NB-3213.34 of [1] as follows:
“Shakedown of a structure occurs if, after a
few cycles of load application, ratcheting
ceases. The subsequent structural response is
elastic, or elastic-plastic, and progressive incremental inelastic deformation is absent.
Elastic shakedown is the case in which the
subsequent response is elastic”. In fact, ratcheting is a complex phenomenon of cyclic
plasticity. The realistic simulation of the process of strain accumulation requires the application of the incremental theory of plasticity.
In the case of the elastic-plastic approach,
care has to be taken with respect to the application of an appropriate material law. The development of appropriate material laws for
ratcheting simulation is the subject of numerous international research activities [10]. The
key for any successful ratcheting simulation is
the description of the kinematic hardening,
i.e. the translation of the yield surface in stress
It is important to point out that available experimental data and research results mostly
refer to mechanical loading. By contrast,
components of nuclear power plants are mainly subject to thermal transient loading conditions. This requires special emphasis on thermal ratcheting issues. It does not only demand
for the determination of the material parameters as a function of temperature but for the
additional implementation of a differential
temperature term. An expanded version of the
Ohno & Wang material law is actually being
implemented as a user routine within the commercial FE code ANSYS® [12].
Consideration of Fatigue Damage
beyond Recent Code Requirements
D e s c r i p t i o n o f Fa t i g u e D a m a g e
B a s e d o n S h o r t C r a c k Fr a c t u r e
Mechanics
Instead of using the abstract notion of a usage
factor, fatigue damage can be interpreted as
the initiation and propagation of microscopically, physically and mechanically short
cracks to an engineering crack size of about
0.5 to 3.0 mm. Fatigue curves of un-notched
(polished) standard specimens resulting from
strain controlled tests usually describe this
limiting damage criterion. This is one characteristic feature of the local strain approach.
The design fatigue curves of the applicable
design codes [1 to 3] are based on these types
of tests. The phase of technical crack initiation to an engineering crack size as part of the
real fatigue damage process can alternatively
be described by means of short crack fracture
mechanics [13,14]. The main advantages of
this approach are listed below:
–– realistic description of the fatigue damage
process,
–– determination of more accurate usage factors,
–– additional consideration of the influence of
surface roughness by a fracture mechanical
interpretation as an initial damage (crack),
–– consideration of size and gradient effects,
–– implicit consideration of mean stress
effects by simulation of the crack closure
behaviour,
–– realistic interpretation of load sequence
effects by consideration of the transient
crack closure behaviour under variable amplitude loading and
–– evaluation of NDT findings on the basis of
63
Areva´s Monitoring Concept
experimentally approved crack propagation
laws.
The concept has already been extended to
multiaxial loading conditions [15] and is successfully applied for the simulation of fatigue
damage of engineering components with emphasis on mechanical loadings [16]. The
strengthening of the approach for thermomechanical loading conditions is part of current research activities. The calculated local
elastic-plastic stresses and strains and the derived fracture mechanical parameters serve as
input data for the fracture mechanics based
laws of crack initiation and crack propagation.
The short crack model should be combined
with replica measurement of the crack growth.
Hence, measured and modelled crack lengths
can be compared in a very early stage of fatigue damage. As a result, NDT findings of
short fatigue cracks can be directly allocated
to the real usage of the material.
The short crack approach is based on parameters of elastic-plastic fracture mechanics.
Hence, it is appropriate for the simulation of
low cycle fatigue conditions typical for nuclear power plant components.
D i r e c t S c a n n i n g o f Fa t i g u e D a m a g e
VGB PowerTech · Autorenexemplar · © 2008
A long-term R&D project has been launched
with the aim of being able to detect fatigue
damage directly on the component based on
NDT methods. Basic research results of various German institutes give evidence that the
different stages of fatigue damage in power
plant steels can be successfully related to
NDT data [17]. A new practical approach to
the evaluation of fatigue damage based on the
fractal dimension of deformation structures is
promising [18] in this context. It was found
that the fractal dimension of the deformation
structure derived from the surface topography
increases during fatigue load. Simultaneously,
the fractal dimension of the changing magnetic domain structure shows the same behaviour. Thus, magnetic noise measurements collect information on small sample regions. The
scaling parameter “fractal dimension” is derived and correlated to fatigue life [18]. Areva
aims at the application and verification of this
method.
Conclusions
The Areva integrated and sustainable concept
of fatigue design, monitoring and re-assessment is an expression of the significance of
design against fatigue of nuclear power plant
components. New plants with scheduled operating periods of 60 years, lifetime extension,
modification of the code-based approaches
and the improvement of operational availability are driving forces in this process. This is
an expression of the sense of responsibility as
well as an economic requirement.
64
It is pointed out that the fatigue concept is
widely supported by measured data. Although
the FAMOS system does not interfere the operation of the plant the results of the fatigue
monitoring can be the basis for decisions of
optimised operating modes and thus influence
the fatigue usage U. This is shown schematically in F i g u r e 7 .
The application of fatigue monitoring in an
early phase of plant operation permits an early
implementation of improvements and consideration of more realistic operational data in
the detailed fatigue analysis. As a rule of
thumb the actual operating conditions result
in lower fatigue usage than those applied in
the design calculations. Of course, additional
loadings may occur as a result of the mode of
operation or faulting conditions within specific systems.
The main modules are the first design analysis before operation, the advanced load monitoring system, the simplified fatigue assessment, the detailed data processing combined
with the specification of occurring thermodynamic loads and the code based fatigue and
ratcheting analysis. The fatigue monitoring
process should be considered from the very
beginning (initial start-up) until the end of life
(e.g. 60 years). As all modules are closely
connected, it is reasonable to apply the approach as a whole with an additional cost reduction effect compared to separate solutions.
Thus, the integrated fatigue approach makes a
significant contribution to the operational
availability and the protection of investment.
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
Acknowledgements
[14]
Of course, the implementation of the Areva
integrated fatigue concept depends on the coordination of numerous specialists of various
disciplines in a cooperative effort. This paper
gives a descriptive overview of the ongoing
process. The authors wish to express special
thanks to the following colleagues for their
valuable contributions, helpful discussions
and amendments to the manuscript: K. Degmayr, G. Desfontaines, M. Franz,
F. Guena, P. Gilles, N. Gillet, R. Gurdal,
S. Krüger, W. Kleinöder, H. Lang, C. Pöckl,
P. Quinot, G. Richter, G. Schön, K. Schramm,
P. Ulmer, A. Willuweit and K. Wirtz.
[15]
[16]
[17]
References
[1]
[2]
2007 ASME Boiler & Pressure Vessel Code,
Section III, Division 1 – Subsection NB:
Class 1 Components, Rules for Construction
of Nuclear Power Plant Components.
KTA 3201.2 (06/96): Components of the Reactor Coolant Pressure Boundary of Light
Water Reactors. Part 2: Design and Analysis.
[18]
RCC-M Edition 2000, Section I, Subsection
B: Class 1 Components. Design and Construction Rules for Mechanical Components
of PWR Nuclear Islands.
Kleinöder, W., and Pöckl, C.: Developing and
implementation of a fatigue monitoring system for the new European pressurized water
reactor EPR. Proceedings of the International
Conference on Nuclear Energy for New Europe 2007, Portoroz, September, 10 to 13,
2007.
Matsuishi, M., and Endo, T.: Fatigue of metals subjected to varying stresses. Proceedings
of the Kyushu branch of the Japanese Society
of Mechanical Engineers, March 1968.
Adams, S.A.: An Alternate Simplified ElasticPlastic Analysis Method. ASME Code Correspondence. Technical Support Document,
January 19, 2005.
Hoffmann, M., Seeger, T., and Führing, H.:
Proposal for an improvement of the Ke-procedure of the ASME-Code. Technical University of Darmstadt, FF-2/1985, 1985.
O’Donnell, B.: Proposed Section III Fatigue
Design Curves for Air Environments.ASME
Code Correspondence, November 7, 2007.
Effect of LWR Coolant Environments on the
Fatigue Life of Reactor Materials. Final Report. NUREG/CR-6909 ANL-06/08. Argonne
National Laboratory, February 2007.
Bari, S., and Hassan, T.: Anatomy of coupled
constitutive models for ratcheting simulation.
International Journal of Plasticity 16 (2000),
pp. 381/409.
Rahman, S.M. and Hassan, T.: Advanced Cyclic Plasticity Models in Simulating Ratcheting Responses of Straight and Elbow Piping
Components, and Notched Plates. Proceedings of PVP-2005, Denver, Colorado, USA.
ANSYS Rev. 11 - Multi Purpose Finite Elemente Code.
Vormwald, M.: The Consequences of Short
Crack Closure on Fatigue Crack Growth Under Variable Amplitude Loading. Fatigue
Fract. Engng. Mater. Struct. 14 (1991), No.
2/3, pp. 205/225.
Vormwald, M., Heuler, P., and Krae, C.: Spectrum Fatigue Life Assessment of Notched
Specimens Using a Fracture Mechanics
Based Approach. ASTM STP 1231, 1994,
pp. 221/240.
Savaidis, G., and Seeger, T.: Consideration
of Multiaxiality in Fatigue Life Prediction
Using the Closure Concept. Fatigue Fract.
Engng. Mater. Struct. 20 (1997), No. 7,
pp. 985/1004.
Savaidis, G., Savaidis, A., and Seeger, T.:
Engineering Components under Multiaxial
Loading. Fatigue Analysis and Lifetime Evaluation. MP Materialprüfung 43 (2001),
No. 3, pp. 76/86.
Palm, S., Gertkemper, H., Knoch, P., and
Maile, K.: A study of the early stages of fatigue damage in power plant steels. Proceedings of the 31st MPA-Seminar in conjunction
with the Symposium on Materials & Components Behaviour in Energy & Plant Technology, Stuttgart, Germany, October 13-14,
2005.
Schreiber, J., and Kröning, M.: Evaluation of
fatigue damage of steel for nuclear power stations by the fractal features of deformation
structures and their noise behaviour (in the
German language). Proceedings of the 31st
MPA-Seminar in conjunction with the Symposium on Materials & Components.
h
VGB PowerTech 5/2009
DVD-VGB_2007-e
04.07.2008
15:15 Uhr
Seite 1
VGB PowerTech-DVD
More than 10,000 digitalised pages with data and expertise
(incl. search function for all documents)
Please fill in and return by mail or fax
I would like to order the VGB PowerTech-DVD
1990 to 2007 (single user license).
Name, First Name
Euro 950.–* (Subscriber of VGB PowerTech Journal 1)
Euro 1950.–* (Non-subscriber of VGB PowerTech Journal 2)
Plus postage, Germany Euro 7.50 and VAT
Street
Network license (corporate license), VGB members’
edition (InfoExpert) and education license on request
(phone: +49 201 8128-200).
Postal Code
City
Country
* Plus VAT.
Annual update 1) Euro 150.–; 2) Euro 350.–
The update has to be ordered annually.
Phone/Fax
Return by fax or in business envelope with window to
Cancellation: This order may be cancelled within 14 days. A notice must be
sent to to VGB PowerTech Service GmbH within this period. The deadline
will be observed by due mailing. I agree to the terms with my 2nd signature.
VGB PowerTech Service GmbH
Fax No. +49 201 8128-329
InfoExpert
Date
Date
1st Signature
2nd Signature
DVD-VGB_2007-e
04.07.2008
15:15 Uhr
Seite 2
VGB PowerTech – www.vgb.org
The generation of electricity and the disposal of heat is in all parts of the world a central
topic of technology, economy, politics and daily live. Experts are responsible for the
construction and operation of power plants, their development and monitoring as well
as for various tasks in connection with service and management.
The technical journal VGB PowerTech is a competent and internationally accepted
publication for power plant engineering. It appears with 11 bilingual issues
(German/English) annually. VGB PowerTech informs with technical/scientific papers and
up-to-date news on all important questions of electricity and heat generation.
VGB PowerTech appears with VGB PowerTech Service GmbH, publishing house of
technical-scientific publications.
VGB PowerTech e.V., the German and European technical association, is the publisher.
VGB PowerTech DVD 1990 to 2007:
Digitalised technical papers of VGB Kraftwerkstechnik and VGB PowerTech.
You find the competent technical know-how from 18 years on more than 10,000 pages
VGB Kraftwerkstechnik (German issues until 2000) and the international technical
journal VGB PowerTech (as of 2001) with:
– More than 2300 technical papers,
– All documents in PDF-format (up to the year 2000 for technical reasons as b/w scan),
– Convenient search function in all papers as full-text search and/or deliberate search
for authors and documents titles,
– Navigate quickly to the desired papers with a few mouse clicks.
The VGB PowerTech-DVD is available as single license or multi-user license for
companies, research institutions and authorities.
The single license can be ordered by form and by post/fax or use our online shop
under www.vgb.org.
A quotation for a multi-user license is made on demand.
You can bring up to date your DVD annually with the VGB PowerTech update.
The update has to be ordered annually.
Your contact at VGB PowerTech Service GmbH,
Jürgen Zimander, Phone: +49 201 8128-200, E-Mail: mark@vgb.org
VGB PowerTech Service GmbH
P.O. Box 10 39 32
45039 Essen
ALLEMAGNE
InfoExpert