摘要
作者针对一类决策空间的维数随时间变化的动态多目标优化问题,借鉴免疫应答蕴含的动态进化机制,提出了一种动态多目标优化免疫算法。算法设计中,依据抗体学习机理,设计几种具有自适应能力的免疫算子进化当前抗体群,以及借助免疫系统的识别功能设计环境识别规则,用于加速相似环境的寻优过程。另外,借助两个性能评价指标,经由比较性的数值试验,获得该算法具有较好的搜索效果以及较稳定的环境跟踪能力。
This work, based on the mechanisms of dynamic evolution of immune response, is to investigate a dynamic multiobjective optimization immune algorithm for a class of dynamic multi-objective optimization problems with the time-varying dimension of decision space. In designs of the algorithm, several adaptive immune operators relying on the metaphors of antibody learning are designed to evolve the current evolving population, while an environmental recognition rule, in terms of the function of recognition in the immune system, is developed to step up the process of optimization. In addition, depending upon the two performance indexes proposed, numerical experiments show that the proposed algorithm has satisfactory searching effects and the ability of stable environmenttracking.
出处
《贵州大学学报(自然科学版)》
2007年第5期486-490,共5页
Journal of Guizhou University:Natural Sciences
基金
国家自然科学基金(60565002)
关键词
动态多目标优化
环境跟踪
免疫应答
免疫算法
Dynamic multiobjective optimization, Environmenttracking, Immune response, Immune algorithm