Fatigue life assessment with random loadings: spectral methods, dynamic simulations, testing Denis Benasciutti
Adam Nieslony
Filippo Cianetti
Numerical and Physical endurance validation of a mountain bike from in-situ strain gage measurements   HBM-nCode
Multiaxial fatigue under variable amplitude loading Jaime T. P. de Castro
Marco Antonio Meggiolaro
Validation of fatigue prediction models and of fatigue solvers Jan Papuga

Fatigue life assessment with random loadings: spectral methods, dynamic simulations, testing

Spectral methods are useful tools to analyse structures subjected to random fatigue loadings, as they provide analytical expressions to estimate fatigue life directly from Power Spectral Density. Spectral methods are very efficient when combined, for example, with dynamic analysis by Finite Element Method and Multi-Body Simulation.

Spectral methods have been formulated for uniaxial and multiaxial random processes with narrow-band or wide-band PSD, as well as validated by experimental tests. Despite some underlying principles are beginning to emerge, further improvements of theoretical models and experimental testing are still possible and they can be stimulated by joint cooperation between academic research and industrial applications.

This mini-symposium will promote open discussions among participants with academic and industrial background, with the aim to share experience and to identify accepted methods in random fatigue analysis. Some topics suggested for discussion are, but are not limited to:

  • Structural durability using spectral method
  • Fatigue life testing with random loadings
  • The influence of mean stress on fatigue in spectral methods
  • Dynamic simulation vs. spectral methods


Numerical and Physical endurance validation of a mountain bike from in-situ strain gage measurements

The HBM-nCode Mini symposium is aimed at Engineers who are looking to use in-service loads to develop realistic in-house testing methods to greater understand how their products are used in the field.

Usage data and Strain measurements collected on the component will be shown and then analysed for fatigue life. We will share a real-world example where it was necessary to replicate failure modes on test rigs and in finite element analysis, and then show how we reconstructed the external loads required to be applied to the structure from the measured responses using the live example as a transfer function. The results of this example established that it was not practical to simulate days upon days of data collection. Learn how we were able to reduce the testing time whilst maintaining equivalent damage content, and then how subsequently test and CAE simulations were used to correlate the stresses and the fatigue life.

The attendees will be taken through a process which involves data acquisition, fatigue analysis, and physical and numerical simulations. Our live example will show how to adapt to using real usage data. The aim of course is to fine-tune the testing and analysis processes to change component design accordingly and avoid over or under engineering.

Multiaxial fatigue under variable amplitude loading

The aim of this minisymposium is to be a meeting place to promote discussion and exchange of ideas in Multiaxial Fatigue. Participants are invited to present short talks (optional), which can be based on current or past research, without the need for peer review or an associated paper. The format should be a set of presentations followed by discussion, but it is not intended to be a work-shop. It is important for the presenting participants to send presentation material before the confer-ence, not only to publish on the conference website but also to plan the talk and discussion sched-ules. To narrow the scope, the main topics include:

  • High-cycle multiaxial fatigue damage models, including multiaxial fatigue limit calculation.
  • Low-cycle multiaxial fatigue damage models, including strain and energy-based models.
  • Comparison among multiaxial damage approaches based on critical planes, stress or strain invar-iants, mesoscopic models, and/or integral-averaging methods, and their applicability to metallic and non-metallic materials.
  • Proportional and non-proportional variable-amplitude loading, including calculation of non-proportionality factors of load histories.
  • Multiaxial cycle counting (e.g. rainflow) and filtering (e.g. racetrack), including methods to iden-tify path-equivalent stress and strain amplitudes.
  • Multiaxial plasticity models for proportional and non-proportional loading, including kinematic, isotropic and non-proportional hardening, and elastoplastic notch corrections.
  • Short and long crack growth under mixed-mode behavior.

Detailed biography is available here.


Validation of fatigue prediction models and of fatigue solvers

Whichever fatigue prediction model is designed, it stays to be a theory only until it is checked on real experimental data. If fatigue prediction in engineering conditions should get more reliable, the models should be validated on extensive experimental inputs. Fatigue experiments are lengthy and expensive, so the second-hand use of their results is a common practice, but only if possible. Many experiments (e.g. in thermo-mechanical fatigue) are so expensive that they do not get published publicly and remain for a proprietary use only. The minisymposium will be focused on current "optimum" validation sets and will discuss the possible future, allowing us to reach more reliable fatigue prediction."



Conference Secretariat: C-IN, 5. kvetna 65, 140 21 Prague 4, CZE | tel.: +420 261 174 301 | fax: +420 261 174 307
Home | Sitemap |
Copyright © 2014 | Powered and created by E-WORKS - web studio | XHTML 1.0 | CSS 2