Options
Quasi-static sizing optimization for stability of stiffened shell structures under manufacturing uncertainty
Publikationstyp
Conference Presentation
Date Issued
2022-06
Sprache
English
Author(s)
Daussault Systèmes Deutschland GmbH
Citation
8th European Congress on Computational Methods in Applied Sciences and Engineering (ECCOMAS 2022)
Contribution to Conference
Frequently, deterministic optimization approaches provide designs being sensitive to deviations from the nominal configuration. This is especially relevant for shell structures prone to buckling under in-plane compression. The buckling load of curved shells is known to be depending on geometric imperfections. The optimization of thin-walled shell structures is numerically challenging when the determination of the buckling load is part of the optimization problem. This holds even more for stiffened shells for which the first buckling (local buckling) is acceptable and therefore, global buckling is considered as objective for the optimization by performing postbuckling analyses in each optimization iteration.
The present studies apply the backward-Euler integration method for analyzing the postbuckling behavior of a stiffened shell structure subject to geometric imperfections. This enables the formulation of time dependent transient design responses for sizing optimization. The buckling capacity of the shell structure is maximized without applying an explicit determination of the buckling load for the optimization formulation by maximizing the sum of discrete reaction forces in time subject to a mass constraint.
The geometrical imperfections are parametrized using Fourier-series approximations and transformed to a reduced set of random variables. Subsequently, the parametrized geometrical imperfections are considered in a computationally efficient Taylor-based robust design optimization formulation. Combining robust design optimization with the quasi-static sizing optimization yields optimized shell structures having significantly increased buckling performance and reduced sensitivity to the geometrical imperfections in comparison to the deterministically optimized designs.
The present studies apply the backward-Euler integration method for analyzing the postbuckling behavior of a stiffened shell structure subject to geometric imperfections. This enables the formulation of time dependent transient design responses for sizing optimization. The buckling capacity of the shell structure is maximized without applying an explicit determination of the buckling load for the optimization formulation by maximizing the sum of discrete reaction forces in time subject to a mass constraint.
The geometrical imperfections are parametrized using Fourier-series approximations and transformed to a reduced set of random variables. Subsequently, the parametrized geometrical imperfections are considered in a computationally efficient Taylor-based robust design optimization formulation. Combining robust design optimization with the quasi-static sizing optimization yields optimized shell structures having significantly increased buckling performance and reduced sensitivity to the geometrical imperfections in comparison to the deterministically optimized designs.
Subjects
imperfection sensitivity
nonlinear analysis
postbuckling
quasi-static solution
Robust design optimization
thickness sizing optimization
DDC Class
624: Civil Engineering, Environmental Engineering
Funding(s)
Calls for Transfer C4T532
Funding Organisations