Computational & Technology Resources
an online resource for computational,
engineering & technology publications |
|
Civil-Comp Proceedings
ISSN 1759-3433 CCP: 102
PROCEEDINGS OF THE FOURTEENTH INTERNATIONAL CONFERENCE ON CIVIL, STRUCTURAL AND ENVIRONMENTAL ENGINEERING COMPUTING Edited by:
Paper 226
Optimal Shape Design of Membrane Structures C.W. Lim1 and V.V. Toropov1,2
1School of Civil Engineering, University of Leeds, United Kingdom
C.W. Lim, V.V. Toropov, "Optimal Shape Design of Membrane Structures", in , (Editors), "Proceedings of the Fourteenth International Conference on Civil, Structural and Environmental Engineering Computing", Civil-Comp Press, Stirlingshire, UK, Paper 226, 2013. doi:10.4203/ccp.102.226
Keywords: membranes, wrinkling, shape optimization, aggregated constraints, design of experiments, metamodel, multi-objective optimization.
Summary
This paper deals with stress-based optimization of membrane structures, where constraints on compressive stresses are used to avoid the formation of wrinkles on the membrane surface. The difficulty arises because a large number of constraints must be considered since stress is a local quantity and hence the problem becomes computationally intensive. The strategy of replacing the local compressive stresses by a global measure using the aggregated constraints approach is hence adopted. A shape optimization problem of a prestressed membrane roof is formulated such that the structural stiffness is increased by minimizing the strain energy subject to a single aggregated stress constraint. Further, an unconstrained multi-objective optimization problem is formulated and solved using a multi-objective genetic algorithm to find a representative set of Pareto-optimal solutions. For that purpose, a moving least squares metamodel has been built using sampling from an optimum latin hypercube design of experiments. Comparisons are then made between the single- and multi-objective optimization problem formulations.
purchase the full-text of this paper (price £20)
go to the previous paper |
|