Weight minimization of tubular dome structures by a particle swarm methodology
The weight of tubular steel dome structures is minimized by utilizing a traditional particle swarm optimization (PSO) approach. Since an extensive welding process is utilized to connect the members of dome structures, joint strengths-related requirements are included into design constraints. The des...
Main Author: | |
---|---|
Format: | Article |
Language: | English |
Published: |
Kuwait University
2013-06-01
|
Series: | Maǧallaẗ al-abḥāṯ al-handasiyyaẗ |
Subjects: | |
Online Access: | http://pubcouncil.kuniv.edu.kw/jer/files/18Jun20130337137_Weight.pdf |
id |
doaj-78d687f691a6464dbc066f5c151455ff |
---|---|
record_format |
Article |
spelling |
doaj-78d687f691a6464dbc066f5c151455ff2020-11-25T00:11:35ZengKuwait UniversityMaǧallaẗ al-abḥāṯ al-handasiyyaẗ2307-18772307-18852013-06-0111145180Weight minimization of tubular dome structures by a particle swarm methodologyTalaslioglu TugrulThe weight of tubular steel dome structures is minimized by utilizing a traditional particle swarm optimization (PSO) approach. Since an extensive welding process is utilized to connect the members of dome structures, joint strengths-related requirements are included into design constraints. The design constrains proposed to check both member and joint-related strengths are taken from the provisions of American Petroleum Institute (API) specification. In order to improve the search capacity of PSO, the exploiting ability of PSO is enhanced thereby hybridizing PSO with a neural network. The hybridized PSO (HPSO) is applied to optimize the designs of two benchmark domes with 354 and 756 members along with a dome structure, shape, topology and size of which are generated by an automatic dome generating tool. The computational efficiencies of HPSO and PSO are evaluated considering the convergence degrees of optimal designations obtained. Thus, it is demonstrated that i) although HPSO achieves to obtain the optimal designations, the fixed geometrical configurations of domes with 354 and 756 members prevents it to explore an optimal designation that satisfies the joint strength-related design constraints, ii) the automatic dome generation tool, which has a capability of using mixed (continuous and integer) design variables, plays an important role in generating the appropriate geometrical dome configurations without any violation of the member and joint strength-related design constraints. Consequently, the inclusion of the joint strength-related design constraints into conceptual design stage leads to both a new looks at the optimization problems of tubular steel structures and an increase in the design reliability of dome structures.http://pubcouncil.kuniv.edu.kw/jer/files/18Jun20130337137_Weight.pdfTubular domeAPIparticle swarmoptimization. |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Talaslioglu Tugrul |
spellingShingle |
Talaslioglu Tugrul Weight minimization of tubular dome structures by a particle swarm methodology Maǧallaẗ al-abḥāṯ al-handasiyyaẗ Tubular dome API particle swarm optimization. |
author_facet |
Talaslioglu Tugrul |
author_sort |
Talaslioglu Tugrul |
title |
Weight minimization of tubular dome structures by a particle swarm methodology |
title_short |
Weight minimization of tubular dome structures by a particle swarm methodology |
title_full |
Weight minimization of tubular dome structures by a particle swarm methodology |
title_fullStr |
Weight minimization of tubular dome structures by a particle swarm methodology |
title_full_unstemmed |
Weight minimization of tubular dome structures by a particle swarm methodology |
title_sort |
weight minimization of tubular dome structures by a particle swarm methodology |
publisher |
Kuwait University |
series |
Maǧallaẗ al-abḥāṯ al-handasiyyaẗ |
issn |
2307-1877 2307-1885 |
publishDate |
2013-06-01 |
description |
The weight of tubular steel dome structures is minimized by utilizing a traditional particle swarm optimization (PSO) approach. Since an extensive welding process is utilized to connect the members of dome structures, joint strengths-related requirements are included into design constraints. The design constrains proposed to check both member and joint-related strengths are taken from the provisions of American Petroleum Institute (API) specification. In order to improve the search capacity of PSO, the exploiting ability of PSO is enhanced thereby hybridizing PSO with a neural network. The hybridized PSO (HPSO) is applied to optimize the designs of two benchmark domes with 354 and 756 members along with a dome structure, shape, topology and size of which are generated by an automatic dome generating tool. The computational efficiencies of HPSO and PSO are evaluated considering the convergence degrees of optimal designations obtained. Thus, it is demonstrated that i) although HPSO achieves to obtain the optimal designations, the fixed geometrical configurations of domes with 354 and 756 members prevents it to explore an optimal designation that satisfies the joint strength-related design constraints, ii) the automatic dome generation tool, which has a capability of using mixed (continuous and integer) design variables, plays an important role in generating the appropriate geometrical dome configurations without any violation of the member and joint strength-related design constraints. Consequently, the inclusion of the joint strength-related design constraints into conceptual design stage leads to both a new looks at the optimization problems of tubular steel structures and an increase in the design reliability of dome structures. |
topic |
Tubular dome API particle swarm optimization. |
url |
http://pubcouncil.kuniv.edu.kw/jer/files/18Jun20130337137_Weight.pdf |
work_keys_str_mv |
AT talaslioglutugrul weightminimizationoftubulardomestructuresbyaparticleswarmmethodology |
_version_ |
1725403249850187776 |