Probabilistic shape design optimization of structural components under fatigue. (1st April 2017)

• Record Type: Journal Article
• Authors: Georgioudakis, Manolis; Lagaros, Nikos D.; Papadrakakis, Manolis
• Is Part Of: Computers & structures. Volume 182(2017)
• Page Start: 252
• Page End: 266
• Publication Date: 2017-04-01
• Subjects: Crack propagation -- Extended finite element method -- Fatigue -- Fracture mechanics -- Metaheuristic algorithms -- Shape optimization; Structural engineering -- Data processing -- Periodicals; Electronic data processing -- Structures, Theory of -- Periodicals; 624.171
• DOI: 10.1016/j.compstruc.2016.12.008
• Languages: English
• ISSNs: 0045-7949

Abstract

Highlights: A structural reliability analysis based on shape design optimization is proposed. Shape design optimization problems formulation within the context of the XFEM. Nature inspired optimization techniques used for solving the optimization problems. Significant effect of the initial imperfection on the optimal shapes. Abstract: Many failures of structural components are attributed to fatigue due to repeated loading and unloading conditions. The crack growth due to fatigue, represents a critical issue for the integrity and capacity of structural components. Apart from the loading conditions, the shape of the structural components plays an important role in their service life. In this study, extended finite element and level set methods are integrated into a probabilistic shape design optimization framework aiming to improve the service life of structural components under fatigue. In this context, the relation between the geometry of the structural components and their service life is investigated. The effect of uncertain material properties as well as the crack tip initialization, described by random variables is also examined. Comparisons between optimized shapes obtained for various targeted fatigue life values are addressed, while the location of the initial imperfection along with its orientation are found to have a significant effect on the optimal shapes for the components studied.