Journal of Steel & Structure

Journal of Steel & Structure

Optimum Design (Minimum Weight) of Sigle-layer-lattice Domes with Different Topologies using Genetic Algorithm and Artificial Neural Networks

Document Type : Original Article

Authors
Mehdi Bandegi
Farzad Shahabian
10.22034/jss.2010.238448
Abstract
In this paper, the two methods of genetic algorithm and neural networks have been combined to achieve the optimum design (minimum weight) of single-layer-lattice domes subjected to uniform gravitational loading. Herein, 378 models with six topologies, spans between 10 and 50 m and different span-to-height ratio from 1/8 to 1/2 have been analyzed and designed for the optimum weight and configuration using genetic algorithm. Furthermore, the results were being used to train the artificial neural networks to predict the optimum weight of the space domes. Thus, for a circular site plan with a specific diameter, an optimum single-layer-lattice dome with specific configuration could be proposed by using the neuro-genetic system.
Keywords
Single-layer-lattice Dome
Optimization
Genetic Algorithm
Artificial Neural Networks
Journal of Steel & Structure
Volume 4, Issue 8 - Serial Number 9
September 2010
Pages 33-47