说明：  摘　要:提出一种新的基于Pareto多目标进化免疫算法(PMEIA)。算法在每一代进化群体中选取最优非支配抗体保存到记忆细胞文档中 同时引入Parzen窗估计法计算记忆细胞的熵值,根据熵值对记忆细胞文档进行动更新,使算法向着理想Pareto最优边界搜索。此外,算法基于点在目标空间分况进行克隆选择,有利于得到分布较广的Pareto最优边界,且加快了收敛速度。与已有算法相比, PMEIA在收敛性、多样性,以及解的分布性方面都得到很好的提高。
(Abstract:This paperproposed a new pareto-based multi-object evolutionary mi mune algorithm(PMEIA). PMEIA selected optmi alnon-dominated antibodieswhichwere then reserved inmemory cellarchive, and introducedParzenwindow to calculate entropy ofmemory cells. Updated thememory cell archive according to entropy ofmemory cells. This guarantees the conver- gence to the true Pareto fron.t Moreover, the performance of clone selection was dependent on distribution in the objective space, whichwas favorable forgetting awidely spreadPareto frontand mi proving convergence speed. Comparedwith the exis- ted algorithms, the obtained solutions ofPMEIA havemuch betterperformance in the convergence, diversity and distribution.)