基于密集度的搅动多目标演化算法

Stir strategy based on multi-objective evolutionary algorithm

多目标演化算法的研究目标是使算法种群快速收敛并均匀分布于问题的非劣最优域。定义和使用密集度来保持群体中个体的均匀分布,将个体的Pareto强度值和密集度合并到个体的适应值定义中。提出搅动策略,以提高算法对解空间的遍历性,从而较大程度上避免算法的早熟,对每次搅动得到的部分非劣解个体进行邻域搜索以加快非劣解前沿的进化。最后,测试函数的实验结果表明了算法的可行性和有效性。

Capable of searching for multiple Pareto optimal solutions concurrently in a single simulation run,and the current research work focuses on the Pareto optimal-based MOO evolutionary approaches.The intensive degree is defined and used to maintain a good spread of solution in the population,and define the fitness of the individual through Pareto strength and intensive degree,and given the stir strategy,which results in a new population significantly indifferent from the old one while inheriting the evolutionary information from the history,by this way,the performance on global convergence is enhanced,and premature is avoided simultaneously.Test results show that the new approach is feasible and effective. Stir strategy based on multi-objective evolutionary algorithm LI Hong-mei1;2（1.College of Computer;Sun Yat-Sen University;Guangzhou 510275;China;2.Department of Computer;Guangdong Baiyun College;Guangzhou 510450;China） Capable of searching for multiple Pareto optimal solutions concurrently in a single simulation run,and the current research work focuses on the Pareto optimal-based MOO evolutionary approaches.The intensive degree is defined and used to maintain a good spread of solution in the population,and define the fitness of the individual through Pareto strength and intensive degree,and given the stir strategy,which results in a new population significantly indifferent from the old one while inheriting the evolutionary information from the history,by this way,the performance on global convergence is enhanced,and premature is avoided simultaneously.Test results show that the new approach is feasible and effective.