Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (12): 233-235.DOI: 10.3778/j.issn.1002-8331.2009.12.074

• 工程与应用 • Previous Articles     Next Articles

Novel generating algorithm of immune detector and its convergence analysis

CEN Jian1,2,ZHANG Qing-hua3,XU Bu-gong1,GAO Ting-yu3,4,LI Hong-fang3,4   

  1. 1.College of Automation,South China University of Technology,Guangzhou 510640,China
    2.College of Automation,Guangdong Polytechnic Normal University,Guangzhou 510635,China
    3.College of Electronic Information and Computer,Maoming University,Maoming,Guangdong 525000,China
    4.College of Information Engineering,Taiyuan University of Technology,Taiyuan 030024,China
  • Received:2008-12-30 Revised:2009-02-16 Online:2009-04-21 Published:2009-04-21
  • Contact: CEN Jian

新的免疫检测器生成算法及其收敛性分析

岑 健1,2,张清华3,胥布工1,高廷玉3,4,李红芳3,4   

  1. 1.华南理工大学 自动化科学与工程学院,广州 510640
    2.广东技术师范学院 自动化学院,广州 510635
    3.茂名学院 计算机与电子信息学院,广东 茂名 525000
    4.太原理工大学 信息工程学院,太原 030024
  • 通讯作者: 岑 健

Abstract: A novel algorithm which generates immune detectors is proposed.Passing immune vaccine,immune learning and immune response of fault signal,fault pattern is identified and diagnosed.The algorithm uses of k-nearest neighbor classification method to decide fault type and has a good dynamic adaptive behavior.Regarding the set of each era antibodies mutated in the system learning as a random sequence,the condition of convergence of the series and a proof are presented.The algorithm’s astringency is proved.The machine unit experimental results show that the method is effective.

Key words: immunology, fault detector, k-nearest neighbor classification, clonal selection, convergence

摘要: 提出一种基于免疫学原理的免疫检测器生成算法,通过对故障信号(即有害抗原)的免疫疫苗、免疫学习和免疫应答进行故障模式识别和诊断,利用k-nearest neighbor分类法确定该模式的故障类型,该算法有较好的动态性、自适应性。把各代抗体集合作为随机序列,给出序列的收敛条件及证明,证明了所提出的算法是概率1收敛,并通过机组试验验证了由该算法产生的检测器对故障检测的有效性。

关键词: 免疫学, 故障检测器, k-nearest neighbor分类, 克隆选择, 收敛性