应用档案精英学习和反向学习的多目标进化算法

发布时间：2018-06-08 14:42

本文选题：档案精英学习 + 动态一般反向学习　；参考：《计算机学报》2017年03期

【摘要】：现实中的多目标优化问题日益复杂,对多目标优化算法提出了新的挑战.受混合多目标优化算法的启发,该文提出了一种应用档案精英学习和反向学习的多目标进化算法(Multi-objective Evolutionary Algorithm Based on Archive-Elite Learning and Opposition-based Learning,AOL-MOEA)以解决困难的多目标优化问题.AOLMOEA算法利用档案精英学习算子增强算法全局搜索能力,促进算法较快收敛;运用动态一般反向学习机制代替变异算子以增加种群逃逸局部极值的机会;使用3-点最短路径方法维持解群的多样性.AOL-MOEA算法与另外5种代表性多目标优化算法在12个基准多目标测试函数上进行性能比较,实验结果表明:AOL-MOEA算法在收敛性、多样性和稳定性等方面均优于或部分优于其他的对比算法.
[Abstract]:The multi-objective optimization problem is becoming more and more complex in reality, which brings a new challenge to the multi-objective optimization algorithm. Inspired by the hybrid multi-objective optimization algorithm, In this paper, a multi-objective evolutionary algorithm named Multi-Objective Evolutionary Algorithm based on Archive-Elite Learning and based on AOL-MOEAA is proposed to solve the difficult multi-objective optimization problem. AOLMOEA uses the archival elite learning operator to enhance the global search ability of the algorithm. It promotes the fast convergence of the algorithm and uses the dynamic general reverse learning mechanism to replace the mutation operator to increase the opportunity of escaping the local extremum of the population. Using the 3-point shortest path method to maintain the diversity of solution groups. AOL-MOEA algorithm is compared with other five representative multi-objective optimization algorithms on 12 benchmark multi-objective test functions. The experimental results show that the proposed algorithm is convergent. Diversity and stability are superior to or partly superior to other comparison algorithms.
【作者单位】：华东交通大学软件学院;
【基金】：国家自然科学基金(61165004) 江西省自然科学基金(20114BAB201025) 江西省教育厅科技项目(GJJ12307,GJJ14373)资助
【分类号】：TP18

【相似文献】