摘要
生物免疫系统在识别和清除抗原的过程中,免疫细胞之间信息交互和协作,能够快速适应环境变化,具有很强的学习和自适应控制能力。基于此,本文提出了针对高维动态函数优化的免疫算法。该算法的主要特点是采用Gray码编码、采用不同的克隆繁殖策略、对抗体实施不同概率的超变异和多细胞编辑等操作,提高算法寻优能力和种群的多样性。通过与几种典型算法进行比较,仿真结果证明该算法对动态优化性能及跟踪能力有明显的改善。
In the process of identifying and removing antigens in the immune system, immune cells interact ano collaborate between each other, which quickly adapt to the environment change and have very strong learning and self-adapting control ability. In this paper, by adopting Gray code, using different cloning strategies, imple- menting hyper-mutation and multiple- cell editing with different probabilities on antibodies, the author presents an immune algorithm for high-dimension dynamic function optimization to improve the tracking capacity and diversity of antibodies. Compared with several typical algorithms, the simulation results show the performance and tracking capacity of the proposed algorithm for dynamic function optimization is improved obviously.
出处
《华东交通大学学报》
2012年第4期57-63,共7页
Journal of East China Jiaotong University
基金
江西省自然科学基金项目(2007GZS0883)
江西省教育厅科技项目(GJJ08239)
关键词
免疫算法
高维动态函数
环境跟踪
immune algorittma
dynamic function optimization
environment tracking
