摘要
针对免疫算法在全局优化过程中多样性不足的问题,提出一种新型的免疫进化算法.随机克隆扩张和多受体随机编辑算子是该算法的主要特色,同时引入改进的超变异算子加强个体的学习能力;提出一种新的算法性能评价准则,以比较不同算法在全局优化中的表现.实验环节中,首先确定了克隆扩张比;然后将免疫进化算法与快速克隆算法和Opt-IMMALG算法进行比较.结果表明,免疫进化算法明显优于另外两种算法.
In order to increase the diversity of immune algorithm when solving global optimization problems, a novel immune evolutionary algorithm(IEA) is proposed. The main characteristics of IEA are clonal expansion and multiple-parent random receptor editor operators. In addition, a modified hypermutation operator is introduced to improve the learning ability of individuals. Particularly, a novel performance evaluation criterion is constructed, by which the performance of different algorithms can be compared easily. In the experimental study, the ratio of clonal expansion is determined, and the IEA is compared with fast clonal algorithm(FCA) and Opt-IMMALG. The results show that IEA is significantly better than FCA and Opt-IMMALG.
出处
《控制与决策》
EI
CSCD
北大核心
2011年第1期59-64,共6页
Control and Decision
基金
国家自然科学基金项目(60373000
90820302
60805027)
国家博士点基金项目(200805330005)
湖南省院士基金项目(02152)
关键词
人工免疫系统
全局优化问题
性能评价准则
artificial immune systems
global optimization
performance evaluation criterion