Reliability Mechanical Components under Fatigue

Today`s highly competitive environment requires continued improvement of the quality of products. This applies particularly to components and structures as used in aerospace, mechanical-, civil- and others types of engineering. The safety and reliability of these structures has to be assured not only against ultimate loading, e.g. caused by extreme events, but also, even more importantly against fracture and fatigue failure, e.g. caused by cyclic loading. Structures and components subjected to those types of loading may be rockets, satellites, etc in space engineering, cars, cranes, etc, in mechanical engineering, bridges, tall masts, etc in civil engineering, fuselage skin, etc, in aeronautical engineering, etc. In the context of damage tolerance of e.g. ageing aircrafts, the residual strength of e.g. the fuselage skin damaged by multiple fatigue cracks received considerable attention. In the stress field of the fuselage skin, the rivet holes of the skin splice joints constitute a number of stress concentrations, where Multi Site Damage may develop: several small cracks may coalesce to form a large crack, which may then progress fast and lead to ultimate failure. The small cracks may also lead to Widespread Fatigue Damage, where unstable crack growth of a lead crack becomes possible due to strength reduction. The determination of the crack initiation points as well as the crack propagation is a difficult task, due to the fact that the two quantities are associated with a certain amount of uncertainty. A statistical perspective should be adopted to make possible to combine all uncertainties and variations to predict fatigue life of components and consequently to perform a reliability analysis. The results of reliability analysis are needed in aerospace engineering as well as in other types of industries (e.g. car industry, off-shore engineering, mechanical and civil engineering, gas transmission lines, fossil and nuclear power plants, etc.). In fact the unavoidable existence of cracks in some components may lead to increased safety concerns about the loss of structural strength an possibly failure of structural systems. This research project provides an opportunity for the IfM to continue the collaboration with Prof. Zabaras` group of Material Process Design and Control Laboratory at Cornell University in Ithaca, New York. The aim of this collaboration is to improve the accuracy over the currently applied engineering approaches, e.g. cohesive elements, adopted to model the fracture initiation. This bridges the gap in the fracture initiation model between the engineering approach, based on the stochastic analysis on macroscale, and the scientific approach by modelling the material uncertainty effects on the macroscale by the microstructure. With this scientific approach a more realistic prediction of the initial crack distribution and consequently the fatigue life will be feasible.