文献类型：期刊文献

中文题名：Adaptive Immune Evolutionary Algorithms Based on Immune Network Regulatory Mechanism

英文题名：Adaptive immune evolutionary algorithms based on immune network regulatory mechanism

作者：He, Hong[1,2]; Qian, Feng[1]

机构：[1] State-Key Laboratory of Chemical Engineering, East China University of Science and Technology, Shanghai 200237, China; [2] College of Mechanical and Electronic Engineering, Shanghai Normal University, Shanghai 201418, China

年份：2007

卷号：24

期号：1

起止页码：141

中文期刊名：Journal of Donghua University(English Edition)

外文期刊名：Journal of Donghua University (English Edition)

收录：EI(收录号：20073810817358);Scopus

基金：National Science Funds for Distinguished Young Scholars ( No60625302);Major state Basic Research Program ofChina (973Program) (No2002CB312200) ;the 863 Hi-Tech Research and Development Programof China (No20060104Z1081);Science and Research Program of Shanghai Educational Committee (No06DZ030)

语种：英文

中文关键词：evolutionary algorithm; immune network;adaptation; stimulation level.

外文关键词：Antibodies - Immunology - Problem solving

摘要：Based on immune network regulatory mechanism, a new adaptive immune evolutionary algorithm (AIEA) is proposed to improve the performance of genetic algorithms (GA) in this paper. AIEA adopts novel selection operation according to the stimulation level of each antibody. A memory base for good antibodies is devised simultaneously to raise the convergent rapidity of the algorithm and adaptive adjusting strategy of antibody population is used for preventing the loss of the population adversity. The experiments show AIEA has better convergence performance than standard genetic algorithm and is capable of maintaining the adversity of the population and solving function optimization problems in an efficient and reliable way.
Based on immune network regulatory mechanism, a new adaptive immune evolutionary algorithm (AIEA) is proposed to improve the performance of genetic algorithms (GA) in this paper. AIEA adopts novel selection operation according to the stimulation level of each antibody. A memory base for good antibodies is devised simultaneously to raise the convergent rapidity of the algorithm and adaptive adjusting strategy of antibody population is used for preventing the loss of the population adversity. The experiments show ATEA has better convergence performance than standard genetic algorithm and is capable of maintaining the adversity of the population and solving function optimization problems in an efficient and reliable way.