TY - JOUR
T1 - Design of ultra-thin shell structures in the stochastic post-buckling range using Bayesian machine learning and optimization
AU - Bessa, M. A.
AU - Pellegrino, S.
PY - 2018
Y1 - 2018
N2 - A data-driven computational framework combining Bayesian regression for imperfection-sensitive quantities of interest, uncertainty quantification and multi-objective optimization is developed for the design of complex structures. The framework is used to design ultra-thin carbon fiber deployable shells subjected to two bending conditions. Significant increases in the ultimate buckling loads are shown to be possible, with potential gains on the order of 100% as compared to a previously proposed design. The key to this result is the existence of a large load reserve capability after the initial bifurcation point and well into the post-buckling range that can be effectively explored by the data-driven approach. The computational strategy here presented is general and can be applied to different problems in structural and materials design, with the potential of finding relevant designs within high-dimensional spaces.
AB - A data-driven computational framework combining Bayesian regression for imperfection-sensitive quantities of interest, uncertainty quantification and multi-objective optimization is developed for the design of complex structures. The framework is used to design ultra-thin carbon fiber deployable shells subjected to two bending conditions. Significant increases in the ultimate buckling loads are shown to be possible, with potential gains on the order of 100% as compared to a previously proposed design. The key to this result is the existence of a large load reserve capability after the initial bifurcation point and well into the post-buckling range that can be effectively explored by the data-driven approach. The computational strategy here presented is general and can be applied to different problems in structural and materials design, with the potential of finding relevant designs within high-dimensional spaces.
KW - Buckling
KW - Data mining
KW - Design charts
KW - Evolutionary optimization
KW - Heteroscedastic Gaussian process
KW - Post-buckling
KW - Ultra-thin composites
UR - http://www.scopus.com/inward/record.url?scp=85041917289&partnerID=8YFLogxK
U2 - 10.1016/j.ijsolstr.2018.01.035
DO - 10.1016/j.ijsolstr.2018.01.035
M3 - Article
AN - SCOPUS:85041917289
VL - 139-140
SP - 174
EP - 188
JO - International Journal of Solids and Structures
JF - International Journal of Solids and Structures
SN - 0020-7683
ER -