系统工程与电子技术（英文版）
繫統工程與電子技術（英文版）
계통공정여전자기술（영문판）
JOURNAL OF SYSTEMS ENGINEERING AND ELECTRONICS
2004年
4期
702-709
,共8页
artificial immune system%immune algorithm%evolutionary algorithm%genetic algorithm%optimization computation%adaptive
Adaptive immune evolutionary algorithm is proposed based on the principle of adaptive immune response. Two new algorithm parameters of expansion radius and mutation radius are defined to construct a small neighborhood and a large neighborhood, then expansion and mutation operations are designed to search the local and global regions of solution space simultaneously by using the two neighborhoods, thus, two-level neighborhood search mechanism is realized. The degree of the diversity in the population is described with the average Euclideandistance among all individuals, and it is used to adjust algorithm parameters adaptively to accelerate convergence and avoid getting stuck at local optima. The algorithm is proved to be convergent and its optimization principle is analyzed. The experiment results of multi-modal function optimization show that the algorithm is effective.
