一种基于“探测'与“开采'的多目标进化算法

A multi-objective evolutionary algorithm based on 'exploration' and 'exploitation'

针对实践中多目标优化问题(MOPs)的Pareto解集(PS)未知且比较复杂的特性,提出了一种基于'探测'(Exploration)与'开采'(Exploitation)的多目标进化算法(MOEA)——MOEA/2E。该算法在进化过程中采用'探测'与'开采'相结合的方法,用进化操作不断地探测新的搜索区域,用局部搜索充分开采优秀的解区域,并用隐最优个体保留机制保存每一代的最优个体。与目前最流行且有效的多目标进化算法NSGA-Ⅱ及SPEA-Ⅱ进行的比较实验结果表明,MOEA/2E获得的Pareto最优解集具有更好的收敛性与分布性。

In view of the fact that Pareto Set （PS） of multi-objective optimization problems （MOPs） is often unknown and complex in practice. This paper proposes a multi-objective evolutionary algorithm （MOEA） based on ＂Exploration＂ and ＂Exploitation＂, named MOEA/2E. This algorithm combines ＂Exploration＂ and ＂Exploitation＂ in the evolutionary process. It explores new searching areas with evolutionary operators, exploits promising areas effectively with local search and stores optimal individual of a population with elitism. Compared with two popular and efficient MOEAs--NSGA-Ⅱ and SPEA- Ⅱ, the experimental results demonstrate that MOEA/2E can obtain Pareto optimal solutions set with better convergence and diversity.