论文部分内容阅读
为了提高多目标优化算法的求解性能,提出一种启发式的基于种群的全局搜索与局部搜索相结合的多目标进化算法混合框架.该框架采用模块化、系统化的设计思想,不同模块可以采用不同策略构成不同的算法.采用经典的改进非支配排序遗传算法(NSGA-Ⅱ)和基于分解的多目标进化算法(MOEA/D)作为进化算法的模块算法来验证所提混合框架的有效性.数值实验表明,所提混合框架具有良好性能,可以兼顾算法求解的多样性和收敛性,有效提升现有多目标进化算法的求解性能.
In order to improve the performance of multi-objective optimization algorithm, a hybrid multi-objective evolutionary algorithm based on heuristic population-based global search and local search is proposed.This framework adopts modular and systematic design idea, and different modules can be used Different strategies constitute different algorithms.The validity of the proposed hybrid framework is verified by using the classical improved non-dominated ranking genetic algorithm (NSGA-II) and the decomposition-based multi-objective evolutionary algorithm (MOEA / D) as the modular algorithm of evolutionary algorithm. Numerical experiments show that the proposed hybrid framework has good performance, which can take into account the diversity and convergence of the algorithm, and effectively improve the performance of the existing multi-objective evolutionary algorithms.