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 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.
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)
