An Immune Complement Optimization Algorithm
Guang-zhu CHEN,Zhi-shu LI,Zhen-cai ZHU,Dun-wei GONG
2009-01-01
Abstract:Some immune optimization algorithms inspired by the biological immune system are lack of fast convergence speed,high robustness and are difficult for attaining the optimal solution of optimization problems.The complement system,which represents a chief component of innate immunity,not only participates in inflammation but also acts to enhance the adaptive immune response.In order to improve the capability of immune system optimization,a novel immune algorithm based on the complement activation theory-an immune complement optimization algorithm(ICOA) was presented.In ICOA,two complement operators,cleave operator and bind operator,were presented firstly.Cleave operator cleaved a complement individual into two sub-individuals,while bind operator binded two complement individuals together to form a big complement individual.Then,the optimal problem was optimized continuously through the complement operators according to the complement activation process,and the overall optimal solution was obtained.Finally,the convergence,robustness of ICOA were analyzed theoretically,which proved that ICOA could converge to the optimal solution and had high robustness.The comparison of ICOA with the classical clonal selection algorithm(CSA) showed that the optimal solution,convergence rate,robustness of ICOA were better than of CSA.