Finite element simulation-based fatigue assessment of welded high-strength steel joints
Pesonen, Tero (2026-06-02)
Väitöskirja
Pesonen, Tero
02.06.2026
Lappeenranta-Lahti University of Technology LUT
Acta Universitatis Lappeenrantaensis
School of Energy Systems
School of Energy Systems, Konetekniikka
Kaikki oikeudet pidätetään.
Julkaisun pysyvä osoite on
https://urn.fi/URN:ISBN:978-952-412-446-1
https://urn.fi/URN:ISBN:978-952-412-446-1
Kuvaus
ei tietoa saavutettavuudesta
Tiivistelmä
The fatigue performance of welded high-strength steels (HSSs) does not improve to the same extent as in quasi-static strength. To reduce this gap, HSSs require more detailed fatigue design methods, for which finite element (FE) simulations provide an efficient means. This thesis aims to incorporate a manufacturing and cyclic loading FE simulation to numerically account for their combined effect in simulation-based fatigue assessments in the high-cycle fatigue regime. The evaluations are performed at the notch-stress level using continuum mechanics and a local stress-based approach of the 4R method. The 4R method employs Smith-Watson-Topper mean stress correction in local elastoplastic material behavior for total fatigue life estimation. The local material behavior assessment incorporates a sequentially coupled welding simulation with material analysis, an optional high-frequency mechanical impact (HFMI) post-weld treatment simulation, and an external load analysis. The simulations were performed for an HSS fillet-welded longitudinal stiffener joint in as-welded, overloaded, and HFMI-treated joint conditions under constant- and variable-amplitude loading. Numerical fatigue-related manufacturing results were validated at different manufacturing stages, and computational fatigue strengths were validated through experimental fatigue tests.
The joint conditions due to the welding, HFMI, and tensile overload peaks were found to have a significant effect on the fatigue strength, which could be simulated with reasonable accuracy and included into the cyclic analysis. The 4R S-N master curve, obtained using simulated mean stress corrected stress ranges and experimental fatigue lives, showed relatively small scatter across different joint and loading conditions. The simulation-based 4R S-N master curve also showed consistency with previously published master curves based on the measurement and analytical approach. In the nominal stress system, the simulated capacity curves correlated well with the experimental results except at consistently high applied stress levels. The results demonstrate the suitability of the FE simulation approach presented in this thesis for incorporating various manufacturing and cyclic conditions using the master curve, thereby expanding the means for numerical fatigue assessment using the 4R method.
The joint conditions due to the welding, HFMI, and tensile overload peaks were found to have a significant effect on the fatigue strength, which could be simulated with reasonable accuracy and included into the cyclic analysis. The 4R S-N master curve, obtained using simulated mean stress corrected stress ranges and experimental fatigue lives, showed relatively small scatter across different joint and loading conditions. The simulation-based 4R S-N master curve also showed consistency with previously published master curves based on the measurement and analytical approach. In the nominal stress system, the simulated capacity curves correlated well with the experimental results except at consistently high applied stress levels. The results demonstrate the suitability of the FE simulation approach presented in this thesis for incorporating various manufacturing and cyclic conditions using the master curve, thereby expanding the means for numerical fatigue assessment using the 4R method.
Kokoelmat
- Väitöskirjat [1213]
