基于差分变异与扰动变异的多目标克隆选择算法

Multi-objective Colonial Selection Algorithm Based on Mutation of Differential and Disturbance

为了更加有效解决多目标优化问题，提出了一种基于差分变异和扰动变异相结合的多目标克隆选择算法（MCSA-MDD）。该算法对非支配抗体进行克隆操作，从而有利于向着理想Pareto前端搜索；采用差分变异与扰动变异相结合的方式来进行免疫基因操作，有利于解的多样性。数值实验分为两组，一组选取4个常用测试函数并与其它五个多目标算法进行比较，数值实验结果表明了MCSA-MDD算法的有效性。另一组仅用差分变异和扰动变异的多目标免疫克隆算法进行比较，数值实验结果验证了采用差分变异与扰动变异相结合的免疫操作提高了算法的性能。

In order to more effectively solve the multi-objective optimization problem, a mult-objective clonal selection algorithm based on mutation of differential and disturbance （MCSA-MDD） is presented. Cloning of the non-dominated antibodies is available for searching towards the true Pareto-front. tial mutation implements to the immune Adopting the combination methods of differen-genetic manipulation and disturbance mutation and assures the diversity of solutions. Numerical experiments are divided into two groups. One compared with five other multi-objective algorithms implementation on four benchmark problems, numerical results show that the MCSA-MDD is effective. The other compared with the muir-objective clonal selection algorithm based on disturbance mutation and the mult-objective clonal selection algorithm based on differential mutation, numerical experiments show that the MCSA-MDD improves the performance of the algorithm.