一种求解多目标优化问题的协同演化算法  

Coevolutionary Algorithm Solving Multiobjective Optimization Problems

作　　者：董红斌 黄厚宽[2] 邓大勇[2] 

机构地区：[1]哈尔滨师范大学计算机科学系,黑龙江哈尔滨150025 [2]北京交通大学计算机与信息技术学院,北京100044

出　　处：《北京交通大学学报》2007年第5期67-71,共5页JOURNAL OF BEIJING JIAOTONG UNIVERSITY

基　　金：国家自然科学基金资助项目(6044300340771154);黑龙江自然科学基金资助项目(F200605)

摘　　要：提出了一种求解多目标优化问题的协同演化算法.新算法改进了Kwee-Bo的协同演化的思想,将混合策略演化规划用于协同演化过程中,混合策略指导算法有效搜索过程,两个种群协同优化目标函数.标准测试函数的数值实验验证了新算法的有效性.This paper presents a new evolutionary problems--A Coevolutionary Algorithm to Solve algorithm approach to multiobjective optimization the Multiobjective Optimization Problems （MO- CEP）. Through the evolutionary game, populations try to optimize their own objective function and all individuals of the population are regenerated after populations have been rewarded. Based on the performance of mutation strategies mixed strategy distribution is dynamically adjusted, The new approach is compared with other evolutionary optimization techniques in several benchmark functions. The benchmark problems numerical experiment results demonstrate that the proposed method can rapidly converge to the Pareto optimal front and spread widely along the front.

关 键 词：多目标优化 协同演化 PARETO最优前沿 混合策略 演化规划 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]