沈阳航空航天大学学报2024，Vol.41Issue(5)：15-25,11.DOI:10.3969/j.issn.2095-1248.2024.05.002
基于多特征参数融合降维的主轴承故障识别方法
Main bearing fault recognition method based on multi-feature parame-ters fusion and dimensionality reduction
摘要
Abstract
In response to the complexities of fault signal transmission path,instability and difficulties in extracting fault feature for aircraft engine main bearing,a fault recognition method was proposed based on the fusion of time-domain feature parameters,frequency-domain feature parameters and intrin-sic mode function(IMF)energy moment feature parameters for dimensionality reduction.Firstly,60 groups of bearing rolling element fault,inner ring fault,outer ring fault and bearing without fault data were selected respectively then time-domain,frequency-domain and energy moment features were ex-tracted from these instances.Addressing the issue of high dimensionality,extensive data and redundant information of the fusion vector composed of three parameters,principal component analysis(PCA)was employed to reduce the dimensionality of these data and effective principal components were ex-tracted based on cumulative contribution rates of principal components.Finally,the dimensionality re-duction feature vectors were input into the support vector machine(SVM)for pattern recognition to di-agnose the types of bearing faults.The results demonstrate that compared to models employing single feature parameters,this method effectively extracts fault feature vectors from complex signals.Subse-quently,it accurately identifies and classifies fault types using these feature vectors,achieving a fault recognition rate of 98.75%.关键词
主轴承/多特征参数/主成分分析方法/支持向量机/故障诊断/航空发动机Key words
main bearing/multi-feature parameters/principal component analysis/support vector ma-chine/fault diagnosis/aircraft engine分类
航空航天引用本文复制引用
沙云东,赵俊豪,栾孝驰,马煜..基于多特征参数融合降维的主轴承故障识别方法[J].沈阳航空航天大学学报,2024,41(5):15-25,11.基金项目
中国航发产学研合作项目(项目编号:HFZL2018CXY017) （项目编号:HFZL2018CXY017）
