论文部分内容阅读
进化算法可并行处理多个解的特性使得它特别适合解决多目标优化问题。针对高维决策空间,将基因表达式编程引入多目标优化,设计了新的个体结构和操作,提出了一个进化多目标优化算法EMOGEP。实验结果表明,新算法在低维决策空间是可行和有效的;在高维决策空间中,表现出了比传统进化多目标优化算法更好的性能;多模态情况下,新算法能很好的逼近理论Pareto前沿。
Evolutionary algorithms that can handle multiple solutions in parallel make it particularly suitable for solving multi-objective optimization problems. For high dimensional decision space, gene expression programming is introduced into multi-objective optimization, a new individual structure and operation is designed, and an evolutionary multi-objective optimization algorithm EMOGEP is proposed. The experimental results show that the new algorithm is feasible and effective in low-dimensional decision space, and shows better performance than the traditional evolutionary multi-objective optimization algorithm in high-dimensional decision space. The new algorithm can be very good in multi-modal case Approximation theory Pareto frontier.