Hybrid CPN-neural dynamics model for discrete optimization of steel structures

Hojjat Adeli, Hyo Seon Park

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

In practical design of steel structures, the designer usually must choose from a limited number of commercially available shapes such as the widely used wide flange shapes. In this article, we present a hybrid counterpropagation-neural dynamics model and a new neural network topology for discrete optimization of large structures subjected to the AISCASD specifications. The constrained structural optimization problem is formulated in terms of a neural dynamics model with constraint and variable layers. The counterpropagation part of the model consists of the competition and interpolation layers. The CPN network is trained to learn the relationship between the cross-sectional area and the radius of gyration of the available sections. The robustness of the hybrid computational model is demonstrated by application to three examples representing the exterior envelope of high-rise and super-high-rise steel building structures, including a 147-story structure with 8904 members.

Original languageEnglish
Pages (from-to)355-366
Number of pages12
JournalComputer-Aided Civil and Infrastructure Engineering
Volume11
Issue number5
DOIs
Publication statusPublished - 1996 Jan 1

Fingerprint

Steel structures
Dynamic models
Structural optimization
Constrained optimization
Flanges
Interpolation
Topology
Neural networks
Specifications
Steel

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design
  • Computational Theory and Mathematics

Cite this

@article{823ee87698dd4da1ab25f7faab4a2856,
title = "Hybrid CPN-neural dynamics model for discrete optimization of steel structures",
abstract = "In practical design of steel structures, the designer usually must choose from a limited number of commercially available shapes such as the widely used wide flange shapes. In this article, we present a hybrid counterpropagation-neural dynamics model and a new neural network topology for discrete optimization of large structures subjected to the AISCASD specifications. The constrained structural optimization problem is formulated in terms of a neural dynamics model with constraint and variable layers. The counterpropagation part of the model consists of the competition and interpolation layers. The CPN network is trained to learn the relationship between the cross-sectional area and the radius of gyration of the available sections. The robustness of the hybrid computational model is demonstrated by application to three examples representing the exterior envelope of high-rise and super-high-rise steel building structures, including a 147-story structure with 8904 members.",
author = "Hojjat Adeli and Park, {Hyo Seon}",
year = "1996",
month = "1",
day = "1",
doi = "10.1111/j.1467-8667.1996.tb00449.x",
language = "English",
volume = "11",
pages = "355--366",
journal = "Computer-Aided Civil and Infrastructure Engineering",
issn = "1093-9687",
publisher = "Wiley-Blackwell",
number = "5",

}

Hybrid CPN-neural dynamics model for discrete optimization of steel structures. / Adeli, Hojjat; Park, Hyo Seon.

In: Computer-Aided Civil and Infrastructure Engineering, Vol. 11, No. 5, 01.01.1996, p. 355-366.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Hybrid CPN-neural dynamics model for discrete optimization of steel structures

AU - Adeli, Hojjat

AU - Park, Hyo Seon

PY - 1996/1/1

Y1 - 1996/1/1

N2 - In practical design of steel structures, the designer usually must choose from a limited number of commercially available shapes such as the widely used wide flange shapes. In this article, we present a hybrid counterpropagation-neural dynamics model and a new neural network topology for discrete optimization of large structures subjected to the AISCASD specifications. The constrained structural optimization problem is formulated in terms of a neural dynamics model with constraint and variable layers. The counterpropagation part of the model consists of the competition and interpolation layers. The CPN network is trained to learn the relationship between the cross-sectional area and the radius of gyration of the available sections. The robustness of the hybrid computational model is demonstrated by application to three examples representing the exterior envelope of high-rise and super-high-rise steel building structures, including a 147-story structure with 8904 members.

AB - In practical design of steel structures, the designer usually must choose from a limited number of commercially available shapes such as the widely used wide flange shapes. In this article, we present a hybrid counterpropagation-neural dynamics model and a new neural network topology for discrete optimization of large structures subjected to the AISCASD specifications. The constrained structural optimization problem is formulated in terms of a neural dynamics model with constraint and variable layers. The counterpropagation part of the model consists of the competition and interpolation layers. The CPN network is trained to learn the relationship between the cross-sectional area and the radius of gyration of the available sections. The robustness of the hybrid computational model is demonstrated by application to three examples representing the exterior envelope of high-rise and super-high-rise steel building structures, including a 147-story structure with 8904 members.

UR - http://www.scopus.com/inward/record.url?scp=19944362580&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=19944362580&partnerID=8YFLogxK

U2 - 10.1111/j.1467-8667.1996.tb00449.x

DO - 10.1111/j.1467-8667.1996.tb00449.x

M3 - Article

AN - SCOPUS:19944362580

VL - 11

SP - 355

EP - 366

JO - Computer-Aided Civil and Infrastructure Engineering

JF - Computer-Aided Civil and Infrastructure Engineering

SN - 1093-9687

IS - 5

ER -