基于相关性选择的高维多目标优化算法

Correlative Selection Mechanism-based Evolutionary Algorithm for Many-objective Optimization

在科学研究领域普遍存在高维多目标优化问题且难有较好的解决方法,论文提出了基于相关性选择的高维多目标优化算法,首先,论文算法借鉴差分算法中的变异引导算子和交叉算子来提高搜索能力和搜索精度;其次,该算法没有采用传统的非支配排序的选择机制,而是使用基于相关性情况选择个体来维持种群的多样性。为检验算法的性能,将所提出的算法应用于多个基准测试问题,与同类算法相比,所提的方法在收敛性和分布性方面效果较好。

In the field of scientific research are widespread high-dimensional multi-objective optimization problem and difficult to have a better solution.Correlative selection mechanism-based evolutionary algorithm for many-objective optimization is proposed.First,the mutation algorithm is using finite difference algorithm to guide operator and it is using crossover operator to improve the search ability and search accuracy;Then,the algorithm does not adopt the traditional non dominated sorting selection mechanism,but selecting individuals based on the correlation is used to maintain the diversity of population.Some standard benchmark problems are tested to demonstrate the effectiveness of the algorithm.Experimental results show that the algorithm performes better than other algorithms in convergence and diversity.