Evolutionary Elementary Cooperative Strategy for Global Optimization
Nonlinear functions optimization is still a challenging problem of great importance. This paper proposes a novel optimization technique called Evolutionary Elementary Cooperative Strategy (EECS) that integrates ideas form interval division in an evolutionary scheme. We compare the performances of the proposed algorithm with the performances of three well established global optimization techniques namely Interval Branch and Bound with Local Sampling (IVL), Advanced Scatter Search (ASS) and Simplex Coding Genetic Algorithm (SCGA). We also present the results obtained by EECS for higher dimension functions. Empirical results for the functions considered reveal that the proposed method is promising.
KeywordsGlobal Optimization Scatter Search Dimension Number Global Optimization Technique Space Division
Unable to display preview. Download preview PDF.
- 3.Floudas, C.A., Pardalos, P.M.: A collection of test problems for constraint global optimization algorithms. Springer, Heidelberg (1990)Google Scholar
- 4.Grosan, C., Abraham, A.: A simple strategy for nonlinear optimization. In: Proceedings of the Third International Conference on Neural, Parallel and Scientific Computation, Atlanta, USA (in press, 2006)Google Scholar
- 5.Hedar, A.R., Fukushima, M.: Simplex coding genetic algorithm for the global optimization of nonlinear functions. In: Tanino, T., Tanaka, T., Inuiguchi, M. (eds.) Multi-Objective Programming and Goal Programming, pp. 135–140. Springer, Heidelberg (2003)Google Scholar
- 6.Laguna, M., Marti, R.: Scatter search: methodology and implementations. Kluwer Academic Publishers, Dordrecht (2003)Google Scholar
- 12.http://www.cyberiad.net/realbench.htm (accessed on May 20, 2006)
- 13.http://www.solon.cma.univ.ie.ac.at/~neum/glopt/my_problems.html (accessed on May 20, 2006)