基于压缩感知理论和密度聚类算法的全回转推进器滚动轴承故障诊断  

Fault Diagnosis of Rolling Bearing in Full Rotary Propeller Based on Compressive Sensing Theory and Density Clustering Algorithm

作　　者：崔忞慜 许汪歆 田忠殿 CUI Minmin;XU Wangxin;TIAN Zhongdian(The First Military Representative Office of Shanghai Haizhuang Bureau in Nanjing,Nanjing 210001,China;Shanghai Marine Equipment Research Institute,Shanghai 200031,China)

机构地区：[1]海装上海局驻南京第一军事代表室,江苏南京210001 [2]上海船舶设备研究所,上海200031

出　　处：《工业技术创新》2022年第3期98-103,共6页Industrial Technology Innovation

摘　　要：滚动轴承振动信号具有非线性、非平稳性等特性,且往往存在噪声干扰,信号特征难以提取,传统故障工况诊断方法受到应用限制。以全回转推进器为研究对象,提出一种基于压缩感知理论和密度聚类算法的滚动轴承故障诊断算法。首先,将振动信号分解,进行稀疏变换,得到稀疏域下的振动信号;其次,采用压缩感知理论进行信号重构,建立绝对值、偏斜度、方差以及裕度等特征指标;最后,采用密度聚类算法DBSCAN,将不同特征指标之间两两组合进行分析,成功区分了故障工况。为滚动轴承故障的无损化、智能化诊断提供了理论依据。Vibration signals of rolling bearings are nonlinear and non-stationary,and often have noise interference,which makes it difficult to extract signal features,so the traditional fault diagnosis methods are limited in application.Taking the full rotary propeller as the research object,a fault diagnosis algorithm of rolling bearing based on compressive sensing theory and density clustering algorithm was proposed.Firstly,the vibration signal is decomposed and sparse transformed to obtain the vibration signal in sparse domain.Secondly,the compressive sensing theory was used to reconstruct the signal,and the characteristic indexes such as absolute value,skewness,variance and margin were established.Finally,the density clustering algorithm DBSCAN was used to analyze the pairwise combinations of different characteristic indexes,and the fault conditions were successfully distinguished.Provided is a theoretical basis for nondestructive and intelligent diagnosis of rolling bearing faults.

关 键 词：压缩感知理论 密度聚类算法 滚动轴承 特征指标 故障诊断 

分 类 号：TH133.3[机械工程—机械制造及自动化]