Application of the Improved Immune Algorithm to Structural Design Support System
Publication: Journal of Structural Engineering
Volume 130, Issue 1
Abstract
Genetic algorithms (GAs), based on a multipoint search method and a crossover operation, are useful search procedures for combinatorial optimization problems. They also are applied to many kinds of practical optimization problem. However, in general, the GAs have a tendency to decrease rapidly of population diversities in processes of searching. In order to address this drawback, some researchers have proposed new algorithms for maintaining the population diversity. On the other hand, immune algorithms (IAs) are optimization techniques that imitate immune systems in an organism. The IAs are able to obtain a multiple quasi-optimum solution while maintaining the population diversity compared with GAs. In this paper, in order to consider the application of the IAs to optimum structural design problems, we developed a useful design support system for reinforced concrete (RC) slabs by combining a three-dimensional (3D) nonlinear dynamic finite element method (FEM) and an improved immune algorithm (IA). For the analysis of impact failure behavior of RC slabs, layered nonlinear FEM models with thin plate bending elements, and 3D elastoplastic FEM models with eight-node isoparametric hexahedral elements were used. First, an existing IA was modified to make it suitable for RC slab design, and both genetic and immune algorithms were applied to optimization problems with multiple optimum solutions. This modified IA was also applied to quasi-optimization problems with multiple maximum values. By comparing the above two approaches, the usefulness of the IA for design problems with discrete parameters was verified. Next, indices for evaluating impact resistance were discussed, and design simulation was conducted, based on these indices, by combining the layered nonlinear FEM and the IA. From results obtained, the accuracy and usefulness of the IA-based design support system were discussed. For obtaining a more practical design support system, shear reinforcement will be taken into account by combining the IA with the 3D elastoplastic FEM. Furthermore, the effectiveness and future potential of the design support system were confirmed by investigating an effect of introducing a database of design plans.
Get full access to this article
View all available purchase options and get full access to this article.
References
Adeli, H., and Hung, S. L. (1997). Machine learning: Neural networks, genetic algorithms and fuzzy systems, Wiley, New York.
Chiji-iwa, H., Mihara, T., and Ohta, T. (1996). “An optimal location design method of a steel pile foundation using a genetic algorithm,” Proc., Japan Society of Civil Engineers, No. 549/I-37, JSCE, Tokyo, 97–105.
Emoto, H., Nakamura, H., and Miyamoto, A.(1999). “Development of impact resistance design support system for RC slabs by genetic algorithms.” J. Struct. Eng., 45A, 453–464.
Farmer, J. D., Packard, N. H., and Perelson, A. S.(1986). “The immune system, adaptation, and machine learning.” Physica D, 22, 187–204.
Goldberg, D. E. (1989). Genetic algorithms in search, optimization, and machine learning, Addison-Wesley.
Hagiwara, M. (1994). Neuro/fuzzy/genetic algorithms, Sangyo-Tosho CO. LTD, Tokyo.
Ichikawa, A., and Ishii, J.(1994). “Maintenance of a diversity of genetic algorithms based on the evaluation of gene distributions.” Proc. Soc. Instrument and Control Eng., 30(10), 1242–1250.
Ishida, R., Murase, H., and Koyama, S. (1997). Basics and applications of genetic algorithms, Morikita-Shuppan CO. LTD, Tokyo.
King, W. M., Miyamoto, A., and Nishimura, A. (1991). “Failure criteria and failure modes for concrete slabs under impulsive loads.” Mem. Grad. School Sci. Technol., Kobe Univ., 9-A.
Miyamoto, A., King, M. W., and Fujii, M.(1991). “Nonlinear dynamic analysis of reinforced concrete slabs under impulsive loads.” ACI Struct. J., 88(4), 411–419.
Mori, K., Tsukiyama, M., and Fukuda, T.(1997). “Application of an immune algorithm to multi-optimization problems.” J. Electr. Eng., C117(5), 593–597.
Mori, K., Yoshida, I., Kita, A., and Nishikawa, K.(1996). “Proposal of thermo-dynamic selection rules in genetic algorithms.” Proc. Inst. Syst., Control Information Eng., 9(2), 82–90.
Sakawa, M., and Tanaka, M. (1995). Genetic algorithms, Asakura-Shoten CO, LTD, Tokyo.
Wada, K., Tanaka, S., and Wada, K.(1992). “Evolution systems–Basics of genetic algorithms. Part 5.” Comput. Today, 51(9), 82–87.
Information & Authors
Information
Published In
Journal of Structural Engineering
Volume 130 • Issue 1 • January 2004
Pages: 108 - 119
Copyright
Copyright © 2004 American Society of Civil Engineers.
History
Received: Feb 13, 2002
Accepted: Mar 25, 2003
Published online: Dec 15, 2003
Published in print: Jan 2004
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.