基于物种选择的遗传算法求解约束非线性规划问题

来源期刊:中南大学学报(自然科学版)2009年第1期

论文作者:梁昔明 朱灿 颜东煌

文章页码:185 - 189

关键词:遗传算法;种群划分;物种选择;交叉算子;非线性规划

Key words:genetic algorithms; population decomposition; species selection; cross operator; nonlinear constrained optimization

摘    要:将信赖域思想和基于稳定进化策略思想相结合,提出一种基于物种选择的遗传算法。根据当前代最优点,采用稳定最优种群数目和收缩最优种群边界的方法将种群划分为最优种群和全局种群,并提出基于构造优化方向的一种新的交叉算子。研究结果表明:对这2种群按不同的策略协调进化,较好地平衡了种群的多样性和选择压力,兼顾了局部搜索和全局搜索;缺少合适的搜索方向是进化后阶段收敛速度慢的重要原因之一;本算法能有效地提高遗传算法的收敛速度,并具有比较好的鲁棒性。

Abstract: A novel genetic algorithm based on species selection was proposed with combination of trust reign and evolutionary stable strategy. The population was divided into two parts based on the distances between individuals and the current optimal individual. One was the optimal population of current generation, and the other aggregated hypo-opt individuals. A new cross operator based on optimal vector was presented and some numerical tests were made. The results show that this new approach can enhance local searching by bounding constrained optimal population and could rise the population diversities by introducing self-adaptive mutation probability in hypo-opt population. Rational searching vector unreachable is one of the reasons for slow convergent speed. The algorithm is effective and robust.

基金信息:国家重点基础研究发展规划(973)项目
高等学校博士学科点专项科研基金资助项目

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号