摘要
分析了传统的人工免疫算法在寻优过程中易陷入局部极值点或过早收敛的原因,对算法进行了改进,提出了一种自适应克隆抑制免疫算法。改进的算法在克隆下一代抗体时,同时考虑了抗体亲和度和浓度两个因素,并给出了一种自适应调节两者关系的算子,兼顾了收敛速度和后代抗体种群多样性两个方面。对改进后的算法进行了分析,给出了数学描述,以便于工程应用。最后,通过典型的算例对提出算法的有效性进行了验证,结果证明,改进后的算法在收敛速度和寻优性能方面均优于传统的人工免疫算法和标准遗传算法。
Analyzed the reasons of the traditional artificial immune algorithm easily falling into local extreme point or premature convergence in the optimization process.This paper put forward a novel artificial immune algorithm,adaptive clone and suppression artificial immune algorithm（ACSAIA）.The algorithm took into account two factors of antibody affinity and concentration of antibody,and gave an adaptive operator to adjust them.Comparing with the corresponding evolutionary algorithm,ACSAIA could enhance the diversity of the population,avoid prematurity and solve deceptive problems to some extent.Meanwhile it had high convergence speed.The experiments show the proposed algorithm is superior to the traditional artificial immune algorithm and standard genetic algorithm in convergence speed and optimization performance.
出处
《计算机应用研究》
CSCD
北大核心
2011年第2期481-484,共4页
Application Research of Computers
基金
国家自然科学基金资助项目(60970105)
山东省自然科学基金资助项目(Y2007G22)
关键词
抗体亲和度
抗体浓度
人工免疫算法
antibody affinity
antibody concentration
artificial immune algorithm（AIA）
