求解多目标优化问题的新遗传算法  

Novel Genetic Algorithm for Multi-objective Optimization Problem

作　　者：韩丽霞[1] 

机构地区：[1]中国矿业大学计算机学院,徐州221116

出　　处：《计算机科学》2013年第06A期64-66,95,共4页Computer Science

基　　金：国家自然科学基金项目(60873099)资助

摘　　要：给出了求解多目标优化问题的一种新解法。定义了多目标优化问题的非劣方向,设计了方向杂交算子和简单的变异算子。标准算例的计算机仿真结果表明,新算法可以快速地找到一组范围广、分布均匀且数量充足的Pareto最优解。For multi-objective optimization problems（MOP）, a new algorithm was proposed. The nondominated direction was firstly defined. Then, a new crossover operator and a mutation operator were presented based on the definition.Finally, the computer simulations on benchmark problems demonstrate that the proposed algorithm is effective and can find high quality solution set in terms of convergence, diversity, and the distribution of the obtained Pareto optimal solutions.

关 键 词：多目标优化 遗传算法 PARETO最优解 全局收敛 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]