Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (8): 18-20.

• 博士论坛 • Previous Articles     Next Articles

A Self-Adaptive Fault-Diagnoses Based-on OCSVM-CPSO

ZiXing Cai   

  1. 1.Center of Intelligent system and Software,School of Information Science and Engineering,Central South University,Chang Sha 410083,China;
    2. School of Computer and Communication, Hunan University, Chang Sha Hunan 410082 )
  • Received:2006-11-09 Revised:1900-01-01 Online:2007-03-11 Published:2007-03-11
  • Contact: ZiXing Cai

基于OCSVM-CPSO的自适应故障诊断

钟清流 蔡自兴   

  1. 湖南大学计算机与通信学院 中南大学信息科学与工程学院
  • 通讯作者: 蔡自兴

Abstract: A new model for Self-Adaptive Fault-Diagnoses Based-on OCSVM-CPSO has been put forward, which takes OCSVM as basic detecting module and CPSO as searching module for optimizing parameter, and start up CPSO to search new parameters when the accuracy detected has been under a assured threshold, but maintains detecting process by new parameters when he accuracy detected has been above the assured threshold. The simulation-experiments have shown, it can be effectively used for online detecting tasks with higher accuracy .

Key words: One-class SVM, Chaos Particle Swarm Optimization, Self-Adaptive, Fault-Diagnoses

摘要: 提出一种基于OCSVM-CPSO自适应故障诊断模型,它用OCSVM作为基本检测模块,而用CPSO作为最优参数的搜索模块.当在线运行的检测准确率低于某确定的阈值时,启动CPSO搜索新的参数.而准确率达标时,用新参数继续后续检测过程.实验表明:此法能够有效地实现高准确率在线检测任务.

关键词: 一类支持向量机, 混沌粒子群优化, 自适应, 故障诊断