一种免疫补体优化算法

针对目前提出的免疫优化算法在求解优化问题时还存在收敛速度慢，往往不能求得最优解，鲁棒性低的问题，基于生物免疫补体激活原理，提出了一种免疫补体优化算法。在算法中，依据补体激活理论，设计了主要的补体算子：分裂算子和结合算子，并根据补体激活过程，通过补体算子的作用对问题解不断优化，求得全局最优解。最后对算法的收敛性和鲁棒性进行了理论分析，并将免疫补体优化算法与典型的克隆选择算法进行了对比实验。理论与实验结果表明了免疫补体优化算法是收敛的，并且收敛速度更快,求得的最优解更好,鲁棒性更高。

An Immune Complement Optimization Algorithm

Some immune optimization algorithms inspired by the biological immune system, for example the clonal selection theory, the immune network algorithm have been presented, but these algorithms 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 resolving optimization problems, 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 are 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 convergence, robustness of ICOA were analyzed theoretically,which proved that ICOA could converge to the optimal solution and had high robustness.Finally, the experiments results of ICOA compared with the standard clonal selection algorithm (CSA) showed that the optimal solution,convergence rate, robustness of ICOA were better than of CSA.