Proposal On A Unified Fatigue Design And Evaluation Procedure Incorporating Residual Stress Effect For Welded Steel Structures

In general, there are various recommendations for fatigue S-N curves of welded structures for design and evaluation. Most fatigue guidelines for welded structures depend on the joint type, loading mode, plate thickness, post-weld treatment, material, etc. Also, fatigue assessments of welded structures traditionally assume a high tensile residual stress level near the vulnerable weldment location. However, the level of residual stress in weldments is not always high in tension. Therefore, it is easy to be overly conservative in fatigue design and evaluation. This study proposed a pragmatic unified fatigue design and evaluation procedure incorporating the residual stress effect for welded steel structures, considering weld toe failure. The modified fatigue parameter was generated combined with the equivalent structural stress range parameter and the master S-N curve adopted by ASME BPVC VIII-2 (2007) and API 579-1/ASME FFS-1 (2007), and the residual stress effect expressed as an effective stress ratio using the fatigue strength concept. The proposed unified procedure was validated using existing fatigue test results with various residual stress levels, such as as-welded (high tensile), stress-relieved (low tensile), prestressed (compressive), and high-frequency mechanical impact treated (compressive). Therefore, the unified fatigue design and evaluation procedure for welded structures, which considers the effect of loading and residual stress level, can be effectively applied for welded steel structures using the master S-N curve by replacing the equivalent structural stress range with the modified equivalent structural stress range on the ordinate axis.