一种保持群体多样性的多目标遗传算法

Multi-objective optimization genetic algorithm keeping diversity of population

提出一种保持群体多样性的多目标遗传算法.该算法采用一种基于信息熵的测度,以衡量群体在多目标空间下的多样性,并利用该测度将群体当前的进化状态(多样性)与算法的运行机制相关联,设计了若干种增强算法探索力度的策略,有效地开拓了算法的搜索范围,提高了进化过程中群体的多样性,防止了算法早熟收敛.对所提出算法的计算复杂度进行了理论分析.仿真实验表明,所提出的算法具有较好的收敛性能和分布特性.

A multi-objective genetic algorithm keeping diversity of the population is proposed. The algorithm uses a metric based on entropy to measure the diversity of the population in the case of multi-objective space. The evolving state of the current population is associated with the running mechanism of the algorithm by the diversity metric, and several strategies are designed to enhance the extent of exploration of the algorithm, which widens the searching range of the algorithm, and increases the diversity of the evolving population and prevents premature convergence. The computational complexity of the algorithm is analyzed theoretically. Simulation results indicate that the proposed algorithm has good performance of convergence and distribution.