Computational & Technology Resources
an online resource for computational,
engineering & technology publications 

CivilComp Proceedings
ISSN 17593433 CCP: 80
PROCEEDINGS OF THE FOURTH INTERNATIONAL CONFERENCE ON ENGINEERING COMPUTATIONAL TECHNOLOGY Edited by: B.H.V. Topping and C.A. Mota Soares
Paper 94
Optimal Design of Scissorlink Foldable Structures using Genetic Algorithms A. Kaveh and S. Shojaee
Department of Civil Engineering, Iran University of Science and Technology, Narmak, Tehran, Iran A. Kaveh, S. Shojaee, "Optimal Design of Scissorlink Foldable Structures using Genetic Algorithms", in B.H.V. Topping, C.A. Mota Soares, (Editors), "Proceedings of the Fourth International Conference on Engineering Computational Technology", CivilComp Press, Stirlingshire, UK, Paper 94, 2004. doi:10.4203/ccp.80.94
Keywords: optimisation, scissorlink foldable structures, design, uniplet, genetic algorithm.
Summary
The need for mobile, reusable structures that are
characterized by fast and easy erection existed for along time. Such
structures found many applications in temporary constructions. The first
such structure has been designed and constructed by Pinero. Substantial
contribution to the general understanding of geometric and kinematic
behavior of scissorlink structures is due to Escrig [1], Gantes et al [2],
Shan [3], and Kaveh and Davaran [4], among many others.
In structural engineering, Goldberg and Samtani [5], Rajeev and
Krishnamoorthy [6], Jenkins [7], Lin and Hajela [8], Saka [9],
Kaveh and Kalatjari [10,11,12] used genetic algorithm for optimisation.
In this article, the genetic algorithm is employed to optimize scissorlink
foldable structures. The advantage of using GA lies in the fact that the
discrete spaces can be optimized without any complexity. Here,
displacement method is used for analysis, employing uniplet elements.
A genetic algorithm is employed to optimize scissorlink foldable structures. For justification of the developed algorithm, first a classical example of a 25bar space truss is studied. Then, two foldable structures are optimised using the present algorithm. The first example is a 32uniplet foldable barrel vault. Application of GA resulted in a structure with the wieght 817.38N (183.7lb). The second example is an 80uniplet foldable dome as shown in Figure 1. In this example, the number of generation is taken as 50, the population size is chosen as 100, the mutation rate is 0.15, and the constant for penalty function is taken 3.6. The optimaization process leads to a dome with 2310.6 N (519.2 lb). The main emphasis of this article is on the suitability of the genetic algorithm for the opimal design of foldable structures. The present algorithm, achieves optimal designs with a good convergence. The selection method prevents omitting the best individuals, and the considered high rate of mutation, reduces the chance of local optima. Finally, the use of uniplets simplifies the analysis of flodable structures and hence increases the efficiency of the optimisation process.
References
purchase the fulltext of this paper (price £20)
go to the previous paper 
