Multiobjective Optimization Using Fitness Function and Sequence Approximate Technology

Abstract:

In order to obtain well-distributed Pareto front with less number of high-fidelity analysis during simulation optimization, a Pareto multiobjective optimization method is put forward by combining the fitness function and sequence approximate technology. Research shows that better Pareto points can be identified for the multiobjective optimization problems with convex, nonconvex or discontinue Pareto front, and the number of high-fidelity analysis can be largely reduced.

• Periodical: Advanced Materials Research (Volumes 139-141)
• Pages: 1192-1195
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.139-141.1192
• Online since: October 2010
• Authors: Zhan Si Jiang, Hui Jiang
• Keywords: Fitness Function, Multi-Objective Optimization (MOP), Sequence Approximate Technology

References

[1] I. Das and J. E. Dennis : Structural Optimization Vol. 14 No. 1(1997), pp.63-69.

[2] B. Wilson, D. J. Cappelleri, T. W. Simpson and M. Frecker: Opt. and Eng. Vol. 2(2001), pp.31-50.

[3] R. Balling, in: Proceedings of EMO2003, (2003), pp.1-15.

[4] G.G. Wang and S. Shan: Journal of Mechanical Design, Vol. 129 No. 4 (2007), pp.370-380. Fig. 6 Symmetric two-bar truss.