应用精英档案和反向学习的多目标差分进化算法

A Multi-objective Differential Evolution Algorithm with Elite-archive and Opposition-based Learning

针对多目标优化问题日渐复杂的情况,受集成算法思想的启发,提出一种应用精英档案和反向学习的多目标差分进化算法。该算法通过建立一个外部档案来保存种群进化过程中的非支配解,提高算法收敛速度。在进化过程中根据反向学习代跳跃概率,使用反向学习生成反向解,扩大搜索范围,提高种群多样性。利用网格系统确定解的坐标,并根据一定的约束生成交叉池,在交叉池中选择父代个体,利用差分进化算法产生新个体,通过网格约束分解排序算法选择下一代种群。将此算法与其他算法在UF测试函数上进行实验,结果表明:所提出的算法在解决无约束多目标优化问题上得到Pareto前沿形状有较强的鲁棒性。

The multi-objective optimization problem became more and more complex.Inspired by the ensemble algorithm,a multi-objective differential evolution algorithm with elite-archive and opposition-based learning is proposed in this paper.In this algorithm,an external archive was created to save the nondominated solutions in the evolutionary process of the population.It used the preset opposition-based generation jumping andopposition-based learning to generate the different solutions of the individual and stored in the elite archive,to expand the search scope and improve population diversity.The grid was used to determines the coordinates of the solutions,and the restricted mating pool was generated according to certain constraints.The parent solutions were selected in the restricted mating pool to produce the new individual by using differential evolution algorithm,then generated the next iteration population by constrained decomposition with grids sorting.The experimental results showed that the proposed algorithm had strong robustness with the shapes of PFs in solving unconstrained multi-objective optimization problems was superior to some state-of-the-art multi-objective algorithms in diversity and convergence on UF test problems.