Predicting the fatigue damage evolution in welded steel joints
The present thesis is investigating the fatigue damage evolution in welded steel joints subjected to dynamic repetitive loading. The work embraces various joint geometries and loading modes. The focus is on the fatigue damage occurring at the weld toe in fillet welded attachments. The work is mainly...
| Main Author: | |
|---|---|
| Format: | Doctoral or Postdoctoral Thesis |
| Language: | English |
| Published: |
University of Agder
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/11250/3146086 |
| Summary: | The present thesis is investigating the fatigue damage evolution in welded steel joints subjected to dynamic repetitive loading. The work embraces various joint geometries and loading modes. The focus is on the fatigue damage occurring at the weld toe in fillet welded attachments. The work is mainly focusing on joints subjected to constant amplitude (CA) loading but some considerations for variable amplitude (VA) loading are also included. The damage mechanisms involved are discussed and modelled both for the initiation phase and the subsequent crack growth phase. The basic theories for these phases are examined and discussed and new models and prediction methods are proposed. The focus is on the probabilistic modelling such that the large scatter found in the fatigue damage evolution can be accounted for. An advanced probabilistic model for life predictions designated the Random Fatigue-Limit Model (RFLM) is fitted to the experimental data. The model is suggested as a support and even as an alternative to the conventional SN curves currently recommended for fatigue life predictions in the building codes. The proposed RFLM is enhanced such that it can explicitly account for the mean stress effect for a given applied stress range. A distinction is made between the initiation phase and the crack growth phase regarding the impact from the mean stress. Data from experimental investigations are collected and used to corroborate the proposed models and applied calculation methodologies. Additional experimental work with test series for fillet welded longitudinal attachments is carried out as a supplement to the collected life data. For this test series both the initiation phase and the crack growth phase are monitored by an Alternating Current Potential Drop (ACPD) method. Possible improvements of the models are suggested based on this empirical background. The present work provides new knowledge regarding: • How to describe and model the entire damage evolution in welded joints. • The importance and the modelling of ... |
|---|