Abstract:
Curved steel buildings are frequently designed to supply the users of the structure with
ordinary light with a sense of capaciousness as well as grandness in public facilities such as
stations, buying malls, leisure centres and airports.
This paper presents a method for analysis and optimum design of 2D and 3D curved roof trusses
subjected to static loading and specified set of constraints. Here the optimization refers to
minimization of total weight of curved roof structures such that they can resist applied forces
(stress constraint) and don’t exceed certain deformations (displacement constraints). The finite
element formulations is developed and implemented for the static analysis of curved roof trusses
to determine the stresses and displacements.
The use of a reliable and competitive procedure for finding the optimum solutions for problems
involving continuous design variables based on genetic algorithms is demonstrated and used in
this study .The performance of genetic algorithms is affected by various factors such as
coefficients and constants, genetic operators, parameters and some strategies. Member grouping
and initial population strategies are also important factors.
Optimization is an automated design procedure in which the computers are utilized to obtain the
best results. The numerical methods of structural optimization with applications of computers
automatically generate a near optimal design (converge to solve) in interactive manner. A
program was modified and used to automate analysis and optimization of the structure written in
FORTRAN language based Finite Element analysis and Genetic Algorithm optimization
technique. The developed method is tested on several examples and compared with previous researches or SAP2000 results. It is concluded that this method can serve as a useful tool in
engineering design and optimization of curved roofs.
