Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Computational Technology Reviews
ISSN 2044-8430
Computational Technology Reviews
Volume 5, 2012
Structural Seismic Optimisation using Meta-Heuristics and Neural Networks: A Review
E. Salajegheh1 and S. Gholizadeh2

1Department of Civil Engineering, University of Kerman, Iran
2Department of Civil Engineering, Urmia University, Iran

Full Bibliographic Reference for this paper
E. Salajegheh, S. Gholizadeh, "Structural Seismic Optimisation using Meta-Heuristics and Neural Networks: A Review", Computational Technology Reviews, vol. 5, pp. 109-137, 2012. doi:10.4203/ctr.5.4
Keywords: optimisation, earthquake, meta-heuristic, neural network.

Summary
Structural optimisation for earthquake induced loads is one of the major concerns in the field of structural engineering. As the dynamic analysis of structures is a computationally intensive task, achieving the structural optimisation for transient time history loading is an expensive process in terms of computational costs. Therefore, utilising an efficient optimisation algorithm possessing a global search ability is an important task for the seismic design optimisation of structures.

Employing meta-heuristic algorithms allows a larger fraction of the design space to be explored and increases the probability of finding a global or near global optimum compared with gradient-based optimisation methods. A comprehensive review of the meta-heuristics and their applications in the field of structural optimisation for static loads has been carried out by Saka [1] and Lamberti and Pappalettere [2].

Structural design optimisation for earthquake loads is a computationally extensive process. In order to reduce the computational burden, neural network techniques can be effectively utilised. In recent years, neural networks were widely used to solve complex problems in the fields of civil and structural engineering. A review of applications of neural networks in engineering has been undertaken in [3].

The main aim of this paper is to provide a state-of-the-art review of seismic design optimisation of structures using meta-heuristics assisted by neural networks. In this case, the hybrid computational strategies which are currently being developed in literature to improve convergence behaviour of meta-heuristic algorithms are considered and their efficiency is discussed. As the amount of research is very limited in this area, in the literature survey part of the paper only a few papers are reviewed. In this paper, a numerical example is presented and the computational performance of the genetic algorithm (GA), particle swarm optimisation (PSO), ant colony optimisation (ACO), and harmony search (HS) are investigated for the optimisation of a steel frame structure subjected to earthquake loading. The required structural responses during the optimisation process are predicted using an advanced meta-model.

References
[1]
M.P. Saka, "Optimum Design of Steel Frames using Stochastic Search Techniques Based on Natural Phenomena: A Review", in B.H.V. Topping, (Editor), "Civil Engineering Computations: Tools and Techniques", Saxe-Coburg Publications, Stirlingshire, UK, Chapter 6, 105-147, 2007. doi:10.4203/csets.16.6
[2]
L. Lamberti, C. Pappalettere, "Metaheuristic Design Optimization of Skeletal Structures: A Review", Computational Technology Reviews, 4, 1-32, 2011. doi:10.4203/ctr.4.1
[3]
I. Flood, "Current and Future Trends in the Use of Artificial Neural Networks in Engineering", Computational Technology Reviews, 4, 93-114, 2011. doi:10.4203/ctr.4.4

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to Computational Technology Reviews
purchase this volume (price £80 +P&P)