Obálka

Construction of Integral Objective Function/Fitness Function of Multi-Objective/Multi-Disciplinary Optimization

Uloženo v:
Vydáno v:Computer Modeling in Engineering & Sciences vol. 6, no. 6 (2004), p. 567
Hlavní autor: Zhu, Z Q
Další autoři: Liu, Z, Wang, X L, Yu, R X
Vydáno:
Tech Science Press
Témata:
Maxwell's equations
Hierarchies
Euler-Lagrange equation
Multiple objective analysis
Genetic algorithms
Fitness
Optimization
Integrals
Electromagnetic fields
On-line přístup:Citation/Abstract
Full Text - PDF
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstrakt:To extend an available mono-objective optimization method to multi-objective/multi-disciplinary optimization, the construction of a suitable integral objective function (in gradient based deterministic method-DM) or fitness function (in genetic algorithm-GA) is important. An auto-adjusting weighted object optimization (AWO) method in DM is suggested to improve the available weighted sum method (linear combined weighted object optimizationLWO method). Two formulae of fitness function in GA are suggested for two kinds of design problems. Flow field solution is obtained by solving Euler equations. Electromagnetic field solution is obtained by solving Maxwell equations. Bi-disciplinary optimization computation is carried out by coupling these two solutions with a nonlinear optimization method. Numerical results show that the needed Pareto solutions can be effectively obtained by using these suggested methods to meet the original design requirements.
ISSN:1526-1492
1526-1506
DOI:10.3970/cmes.2004.006.567
Zdroj:Advanced Technologies & Aerospace Database