基于免疫学的在线故障检测算法研究及其工程应用.pdfVIP

25 卷 24 期 中 国 电 机 工 程 学 报 Vol.25 No.24 Dec. 2005 2005 年12 月 Proceedings of the CSEE 2005 Chin.Soc.for Elec.Eng. 文章编号：0258-8013 （2005 ）24-0149-05 中图分类号：TH 165 文献标识码：A 学科分类号：47040 基于免疫学的在线故障检测算法的研究及应用 茂，周晓军，孟庆华 （浙江大学机械与能源工程学院，浙江省杭州市 310027 ） STUDY AND APPLICATION OF ON-LINE FAULT TESTING ALGORITHM BASED ON IMMUNOLOGY PANG Mao, ZHOU Xiao-jun, MENG Qing-hua （The College of Mechanical and Energy Engineering, Zheji ang University, Hangzhou 310027, Zheji ang Province, China ） [4-5] [6] [7] ABSTRACT: According to the human being immune 算法和遗传编程 、粗糙集合 、模糊集合 和专 mechanism, an immune algorithm named as vaccine algorithm [8] 家系统 等。这些方法为准确检测诊断设备存在的 was presented. The mathematic model of the vaccine algorithm 故障问题起到了重要作用，但也存在以下不足： was built. The process of the algorithm is introduced in detail. （1）需要使用系统仿真模型，对于复杂系统， The algorithm uses resampling to get a fault’s sufficient 模型难以准确建立，诊断速度慢；对于不同的领域， samples by use of a few samples. When an unknown fault 仿真模型各异， 难统一。 appears, the algorithm can get the fault feature automatically （2 ）需要大量的先验知识 （如故障样本等）。 and gives some information to experts for analysis. When the fault appears again, the algorithm can respond. The algorithm 对于复杂的设备来说，这些先验知识是非常难以获 is a dynamic system which can test and study at the sam