期刊
APPLIED MATHEMATICS AND COMPUTATION
卷 215, 期 12, 页码 4172-4184出版社
ELSEVIER SCIENCE INC
DOI: 10.1016/j.amc.2009.12.038
关键词
Constrained optimization; Decision maker (DM); Evolutionary algorithm (EA); Multiobjective optimization; Penalty function
资金
- National Natural Science Foundation of China [70921001, 60574058]
- Natural Science Foundation of Hunan Province, China [07JJ3126]
- Hunan Provincial Science and Technology Department [2009WK2009]
An adaptive decision maker (ADM) is proposed for constrained evolutionary optimization. This decision maker, which is designed in the form of an adaptive penalty function, is used to decide which solution candidate prevails in the Pareto optimal set and to choose the individuals to be replaced. By integrating the ADM with a model of a population-based algorithm-generator, a novel generic constrained optimization evolutionary algorithm is derived. The performance of the new method is evaluated by 13 well-known benchmark test functions. It is shown that the ADM has powerful ability to balance the objective function and the constraint violations, and the results obtained are very competitive to other state-of-the-art techniques referred to in this paper in terms of the quality of the resulting solutions. (C) 2009 Elsevier Inc. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据