基于1-DISVM的聚类模型及直升机齿轮箱故障诊断应用  

Unsupervised 1-DISVM Based Clustering Model for Fault Diagnosis of Helicopter Gearbox

作　　者：柳新民[1] 刘冠军[1] 邱静[1] 胡茑庆[1] 

机构地区：[1]国防科技大学机电工程与自动化学院,湖南长沙410073

出　　处：《航空学报》2006年第3期453-458,共6页Acta Aeronautica et Astronautica Sinica

基　　金：国家自然科学基金(50375153)资助项目

摘　　要：针对当前故障诊断中存在的训练样本少、知识难获取的问题,结合SVM小样本学习的特点,提出一种基于SVM的自学习聚类模型。通过改进无监督1-SVM算法上的不足,形成一种改进决策1-SVM(1-DIS-VM)算法,由此构建了多模式训练与分类算法,并设计出基于1-DISVM的自学习聚类模型。最后对其进行仿真验证,并应用于直升机齿轮箱的故障诊断,结果表明该方法能从少量样本中自学习输入模式的内在规律,自适应地对未知故障模式进行准确地分类识别。To solve the problems of insufficient fault-samples and diagnosis-knowledge, and according to tire merit of Support Vector Machines （SVM） that carl be trained with small-sample, a SVM based unsupervised clustering model is presented. By modifying the decision-function of One-Class Support Vector Machine （1 SVM）, which has the ability to find outliers from a dataset without any class of information but rarely is ap plied to pattern-recognition for its algorithm limits, a Decision-Improved I-SVM （1-DISVM） is formed, Based on it, muhi-pattern training and classing method is designed, then an unsupervised clustering model is constructed. The simulation and diagnostic experiment results of a helicopter＇s gearbox show that this clustering model can not only recognize the unknown fault patterns adaptively and precisely, hut also learn the nature of the input-patterns from small samples and diagnose the faults successfully.

关 键 词：故障诊断 聚类 支持向量机 无监督学习 

分 类 号：TH165.3[机械工程—机械制造及自动化] O235[理学—运筹学与控制论]