论文部分内容阅读
在对已有克隆选择算法的抗体行为特征分析的基础上,提出了一种新的偏心动态免疫克隆算法(EDICA,Eccentric Dynamic Immune Clone Algorithm).利用进化过程中子代抗体比父代抗体更靠近最优解的启发性信息,提出偏心变异策略,使抗体更快地靠近最优解域.引入控制因子,通过动态调整变异搜索半径的方法,在进化初期加大步长以加快搜索速度,而在后期减小搜索粒度以提高优化精度.采用超球体混沌变异策略以克服各向异性的不利影响并提高全局搜索能力.实验结果表明:EDICA不仅能够准确地找到静态函数的多个最优点,而且还能以较高的精度锁定和跟踪动态函数的最优点.
Based on the analysis of the behavioral characteristics of the existing clonal selection algorithms, a new Eccentric Dynamic Immune Clone Algorithm (EDICA) is proposed, in which the progeny antibody is more closely related to the progeny antibody The optimal solution of the heuristic information, put forward eccentric mutation strategy, the antibody faster near the optimal solution domain.Adopting the control factor, through the dynamic adjustment of mutation search radius, increase the step size in the early evolution to speed up the search speed, and Reduce the search granularity in the later stage to improve the precision of the optimization.The hyper-sphere chaos mutation strategy is adopted to overcome the adverse effect of anisotropy and improve the global search ability.The experimental results show that EDICA can not only find the most optimal points of the static function accurately, It also locks and tracks the best of dynamic functions with high precision.