基于Pareto的多目标优化免疫算法

免疫算法具有搜索效率高、避免过早收敛、群体优化、保持个体多样性等优点。将其应用于多目标优化问题,建立了一种新型的基于Pareto的多目标优化免疫算法(MOIA)。算法中,将优化问题的可行解对应抗体,优化问题的目标函数对应抗原,Pareto最优解被保存在记忆细胞集中,并利用有别于聚类的邻近排挤算法对其进行不断更新,进而获得分布均匀的Pareto最优解。文章最后,对MOIA算法与文献3]中SPEA算法进行仿真,通过比较两者的收敛性和分布性,得到了MOIA优于SPEA的结论。

Multi-objective Optimization Immune Algorithm Based on Pareto

Immune algorithm has many merits,such as high searching efficiency,avoiding immature convergence,colony optimization,keeping individual varieties and so on.In this article,immune algorithm is used to Pareto multi-objective optimization problems,a new Pareto Multi-objective Optimization Immune Algorithm(MOIA) is established.In the algorithm,the feasible solutions are regarded as antibodies,the multi-objective functions are regarded as antigens,Pareto optimal solutions are preserved in memory cells population updated by a vicinity crowding algorithm different from the cluster algorithm.Finally,simulation is carried on the MOIA and the SPEA3.By comparing the convergence and thedistribution of these two algorithms,the article obtains that MOIA is better than SPEA.