Weight Optimization of Steel Frames Using Genetic Algorithm

Abstract

Genetic Algorithm (GA) is a new technique in optimization procedure that works best in design problems with discrete variables. It employs the survival of the fittest philosophy in determining the optimum combination. GA optimization procedure is applied to weight optimization of steel plane frames subjected to different load cases. Database of steel beam sizes is provided as the discrete variables. Both elitist and non-elitist search procedures are used to optimize the total weight of steel frames. Crossover types used are 20- and 50-percent uniform. Optimization result using population sizes 10, 20, and 40 are compared. Elitist search procedure showed superior results when compared to non-elitist for higher population sizes search because of its faster convergence rate. Performance of non-elitist is superior when using lesser population sizes. To examine the performance of genetic algorithms, case studies are conducted by varying material groups and the results are compared with the results from other optimization techniques. Genetic optimization showed superior results when compared to other techniques especially to problems with few material groupings.

Keywords

genetic algorithm structural optimization minimum weight steel frames

Get full access to this article

View all access options for this article.

References

Adeli

and Cheng

(1994). “Augmented lagrangian genetic algorithm for structural optimization”, ASCE Journal of Aerospace Engineering, Vol. 7, No. 1, 104–118.

ASTM (1993). American Society for Testing and Materials: Annual Book of ASTM Standard Vol. 01.04.

Bagley

J.D.

(1967). The behaviour of adaptive systems which employ genetic and correlation algorithms, Doctoral dissertation, University of Michigan.

Chen

S.Y.

and Rajan

S.D.

(1998). “Improving the efficiency of genetic algorithms for frame designs”, Engineering Optimization, Vol. 30, 281–307.

Goldberg

D.E.

(1989). Genetic algorithms in search, optimization and machine learning, Addison-Wesley

Khan

M.R.

(1984). “Optimality criterion techniques applied to frames having general cross-sectional relationships”, IAAA 22, 669–676.

Q.S.

Liu

D.K.

and Fang

J.Q.

(1999). “Optimum Design of Actively Controlled Structures Using Genetic Algorithms”, Advances in Structural Engineering, Vol. 2, No. 2, 109–118.

Liu

Ding

Y.L.

, and Xiao

S.D.

(1996). “An Improved Strategy for GAs in Structural Optimization”, Computers and Structures, Vol. 61, No. 6, 1185–1191.

Rajeev

and Khrisnamoorthy

(1992). “Discrete optimization of structures using genetic algorithm”, ASCE Journal of Structural Engineering, Vol. 118, No. 5, 1233–1250.

10.

Sadek

E.A.

(1992). “Optimization of structures having general cross-sectional relationships using an optimality criterion method”, Computer and Structures, Vol. 43, No. 5, 959–969.

11.

Thevendran

, (1992). “Minimum weight design of multi-bay multi-storey steel frames”, Computer and Structures, Vol. 43, No. 3, 495–503.

12.

Torregosa

R.F.

(1998). Optimization of steel frames using genetic algorithms, M. Eng'g. thesis, Asian Institute of Technology, Bangkok, Thailand.

13.

Yamakawa

(1999). “Some examinations of optimization for topology, shape of structural systems and design of control systems using genetic algorithms”, Proceeding to the 3rd World Congress of Structural and Multidisciplinary Optimization, New York.

14.

Whitley

(1994). “A Genetic Algorithm Tutorial”, Statistics and Computing, Vol. 4, 65–85.