PSI - Issue 22

Carlos D.S. Souto et al. / Procedia Structural Integrity 22 (2019) 376–385 Author name / Structural Integrity Procedia 00 (2018) 000–000

377

2

the design S-N curves are organized into di ff erent categories, where each detail category is represented by their stress range at 2 million cycles. These fatigue curves are also represented graphically by double slope and a cut-o ff limit for stress range called by endurance limit. The fatigue life analysis methodologies used in structural design are based on nominal and hot-spot stresses ap proaches applied to the bolted and welded steel connections, respectively. Besides, other fatigue approaches are based on local fatigue damage parameters such as stress, strain and energy. Kuz˙awa et al. (2018) proposed a practical way to evaluate the fatigue lifetime of existing riveted bridges based on di ff erent fatigue safety levels and assessment phases. This methodology is presented using the fatigue assumptions of the EN1993-1-9 standard and fatigue loads of the Eurocode 1 EN1991-2 (2004) standard. Additionally, these authors have based their studies on recommendations from the European Recommendations for Estimation of Remaining Fatigue Life of Existing Steel Structures. Alencar et al. (2018) and Viana et al. (2019) applied the hot-spot stress approach to estimate the fatigue lifetime of critical welded steel details of a railway composite bridge. Liu et al. (2017, 2019a) presented a fatigue life study of critical connections of a historic eyebar suspension bridge based on the EN1993-1-9 and AASHTO standards. These authors have suggested a global-local fatigue methodology of an ancient riveted metallic bridge based on the submodeling of the critical detail and using the strain damage parameter (Liu et al., 2019b). In this way, several local fatigue damage parameters have been proposed to be used in global-local fatigue analyses considering full-range fatigue curves or fatigue curves based on low- to high-cycle fatigue regimes (Barbosa et al., 2019a,b; Correia et al., 2018; De Jesus et al., 2010). Moreover, Imam and Salter (2018) presented a study taking into account the e ff ects of historical loads on the fatigue life of metallic railway bridges. Rakoczy et al. (2019) suggested a new methodology to estimate the fatigue life of a railroad bridge using a probabilistic method combined with a new fatigue curve for riveted details. In this paper, a software named Fatigue Damage Tool (FDT) is presented. This tool assesses fatigue damage ac cording to the EN1993-1-9 standard. Also in this paper, FDT is applied to evaluate the fatigue accumulated damage of the critical detail at the mid-span of the Va´rzeas railway bridge (Boavida-Barroso, 2019), considering the fatigue load defined by a type 5 locomotive-hauled freight train (adapted from the EN1991-2 standard). The time-history analyses of the nodal stresses, from the dynamic behaviour, were collected (Boavida-Barroso, 2019) and used to calculate the fatigue accumulated damage using FDT. Fatigue damage analysis of the critical detail under consideration was made according to the accumulated damage method proposed in the EN1993-1-9 standard. The cycle counting was done by the rainflow counting algorithm implemented in FDT based on the ASTM E1049-85 (2017) standard. Additionally, FDT can also be applied to other types of engineering structures such as marine structures, pressure vessels, machine details, etc. Global S-N approaches based on the Wo¨hler’s S-N curves, have been proposed for fatigue life prediction taking into account both fatigue crack initiation and propagation phases (Pedrosa et al., 2019). These fatigue curves were based on experimental fatigue data for fatigue lifetime prediction of metallic structures and components. Fatigue curves, which relate the stress range, ∆ σ , and fatigue life, N f , are expressed in logarithmic scale. Another important aspect of the design S-N curves is the fatigue limit, ∆ σ D , which is also known as the endurance limit or fatigue strength, and it relates to the stress level below which an infinite number of loading cycles can be applied to a structural component without causing fatigue failure. In the design codes, the fatigue limit is considered the stress level for the numbers of cycles of 5 × 10 6 and 1 × 10 8 corresponding to the constant and variable amplitude loading, respectively. The statistical treatments of the fatigue data for structural details are described by design codes (Eurocode 3 EN1993-1-9, 2014). Usually, the ASTM E739-10 (2015) standard is used to generate the design S-N curves from the experimental data. According to the EN1993-1-9 standard, the design fatigue curves for constant amplitude loading are given by the following expressions: For m = 3 and N ≤ 5 × 10 6 → ∆ σ m R · N R = ∆ σ m C · 2 × 10 6 (1) 2. Structural fatigue damage assessment 2.1. S-N curves