PSI - Issue 33

Aprianur Fajri et al. / Procedia Structural Integrity 33 (2021) 19–26 Fajri et. al/ Structural Integrity Procedia 00 (2019) 000–000

20 2

1. Introduction The development of technology, especially in the field of engineering structures today, is increasing rapidly. The challenges and obstacles faced are also increasingly diverse. One of the problems that often occur is related to material failure due to operating load that often causes losses. The operating load can be distinguished into two types, namely static load and dynamic load. Failures resulting from static loads are infrequent, as engineers always apply high safety factor calculations when designing a structural component. Problems that often occur are always related to the dynamic load that works, for example, accidental load/impact load that is sourced from various possibilities (Li et al., 2013; Prabowo et al., 2017; 2018; 2019; 2020). Furthermore, dynamic loads, which are relatively small in size, can still cause failure if they occur continuously. This kind of phenomenon is called fatigue failure. It is estimated that 80 90% of engineering material failures are caused by the fatigue phenomenon (Campbell, 2012). Fatigue is permanent damage that occurs to objects due to the influence of conditions that produce stress-strain fluctuations (Bishara et al., 2018). According to Hansen and Winterstein (1995), fatigue failure occurs in four phases, namely: (1) crack nucleation; (2) structurally dependent crack propagation; (3) crack propagation; (4) failure. Several factors can affect fatigue resistance, including the type of applied loading, the kind of material used, mechanical properties, manufacturing procedures, surface roughness, operation temperature, environment, microstructure state, residual stress, corrosion, and crack initiation (Boardman, 1987; Gaidai et al., 2020; Ślęzak, 2020; Vaara et al., 2020). Marine structures, engine components, critical infrastructure, reservoirs, turbines, nuclear reactors, and features that work in extreme conditions are engineering structures with a high risk of fatigue failure. The collapse of the Alexander L. Kielland rig/platform in 1980 that killed 123 people (France, 2019) is a terrible accident caused by the fatigue phenomenon. The accident illustrates that catastrophic failure is very detrimental and dangerous. Therefore, estimating fatigue life must be done carefully by considering all the variables to prevent disasters (Kamal and Rahman, 2018). Before use, a material needs to undergo a series of tests to look for its characteristics, including fatigue resistance. Fatigue data from these test results can be used as a consideration in the engineering structure design process. In the next step, the designed structure's fatigue characteristics need to be predicted first using the numerical simulation method. Then, the simulation results must be validated using experimental methods before they are operated directly in the field. In this study, some fatigue failure phenomena and research with different methods and approaches will be presented. The goal is to demonstrate the state of development and achievement and the gaps that can still be researched. 2. Timeline of fatigue research history and notable fatigue failures In 1837, Wilhelm Albert became the first person to publish an article on the fatigue phenomenon. Later, William John Macquorn Rankine realized the effect of stress concentration on the axle train's failure in 1842. The systematic fatigue test was only invented in 1860 by Sir William Fairbairn and August Wöhler, which was later refined in 1870, followed by introducing the S-N curve concept and endurance limit. In 1945, A.M. Miner popularized Palmgren's theory of linear damage hypothesis. This theory is still used as a reference when investigating the fatigue phenomenon using the stress life approach. Paul Croce Paris, in 1961 introduced a method to predict the rate of growth of cracks caused by cyclic loading. Furthermore, this method is now known as the fatigue crack growth/ fatigue crack propagation approach (Schütz, 1996). The de Havilland Comet plane crash in 1954 caused by fatigue failures forced manufacturers to redesign all aircraft components, which changed the window shape from square to oval (Schijve, 2003; Wanhill et al., 2015). The sinking of the MV Kurdistan in 1979 and the collapse of Alexander L. Kielland (see Fig. 1a) in 1980 due to fatigue crack growth from a weld defect (see Fig. 1b) were two significant accidents in the field of marine structures (Vukelic, 2018). As a result of this accident, the investigation showed that material properties, weld quality, and geometry design must be considered to avoid structural failure. The importance of using fatigue estimation with the elastic-plastic fracture mechanics approach was introduced to prevent similar accidents. It should be underlined that how fatigue life is predicted is not an exact science, and the method developed so far is still an approach that must be developed continuously.