Send link
Dissertation: Engineering tools for robust creep modeling
04.02.2010
Stefan Holmström, Senior Research Scientist from VTT, presents his doctoral dissertation “Engineering tools for robust creep modeling” at the Aalto University School of Science and Technology on the 5th of February, 2010, at 12 noon (Otakaari 4, Espoo, Finland).
High temperature creep is often dealt with simplified models to assess and
predict the future behavior of materials and components. Also, for most
applications the creep properties of interest require costly long-term testing
that limits the available data to support design and life assessment. Such
test data sets are even smaller for welded joints that are often the weakest
links of structures. It is of considerable interest to be able to reliably
predict and extrapolate long term creep behavior from relatively small sets of
supporting creep data.
For creep strain, the current tools
for model verification and quality assurance are very limited. The ECCC PATs
can be adapted to some degree but the uncertainty and applicability of many
models are still questionable outside the range of data. In this thesis tools
for improving the model robustness have been developed.
The
toolkit includes creep rupture, weld strength and creep strain modeling
improvements for uniaxial prediction. The applicability is shown on data set
consisting of a selection of common high temperature steels and the
oxygen-free phosphorous doped (OFP) copper. The steels assessed are 10CrMo9-10
(P22), 7CrWVMoNb9-6 (P23), 7CrMoVTiB10-10 (P24), 14MoV6-3 (0.5CMV),
X20CrMoV11-1 (X20), X10CrMoVNb9-1 (P91) and X11CrMoWVNb9-1-1(E911).
The
work described in this thesis has provided simple yet well performing tools to
predict creep strain and life for material evaluation, component design and
life assessment purposes. The uncertainty related to selecting the type of
material model or determining weld strength factors has been reduced by the
selection procedures and by linking the weld behavior to the base material
master equation. Much of the resulting improvements and benefits are related
to the reduced requirements for supporting creep data. The simplicity and
robustness of the new tools also makes them easy to implement for both
analytical and numerical solutions.
Dissertation