A comparative study of different fatigue failure assessments of welded bridge details

Abstract

Five different welded joints frequently used in steel bridges have been selected to investigate the accuracy and applicability of three fatigue assessment methods. The first method, also categorised as the global method, is the nominal stress method, while the more advanced methods are the hot spot and the effective notch stress methods. Solid element based finite element models for welded bridge details were created by following the modelling requirements of each fatigue assessment method. A statistical evaluation based on the results of the finite element analyses and the fatigue test data collected from the literature was performed to determine the mean and characteristic fatigue strength. In addition, the standard deviation for each data series was also determined to conclude how well each method describes the fatigue strength of each welded detail. A method with a lower standard deviation is regarded as more accurate. Moreover, the evaluated results from each method were compared with the recommended fatigue strength values in the Eurocode 3 (EN 1993-1-9:2005) and IIW codes. In the light of the test results in this study, it appears that the codes are in reasonable agreement with the test data, even though a few examples of the opposite occurred. The conclusion based on the revised results in this article indicates that the nominal stress method yields satisfactory results, despite its simplicity. When considering the effort involved in creating FE models for numerical analysis, it seems clear that the choice of the nominal method is fairly acceptable.