噪声与振动控制2025，Vol.45Issue(2)：132-138,7.DOI:10.3969/j.issn.1006-1355.2025.02.021
基于SIEDL的交通机电轴承故障特征提取研究
Fault Feature Extraction for Traffic Electromechanical Bearings Based on SIEDL
摘要
Abstract
To address the issue that noise can interfere atoms easily influence the effectiveness of feature extraction in classical shift-invariant dictionary learning(SIDL),this paper proposed a shift-invariant enhanced dictionary learning(SIEDL)method for feature extraction in traffic electromechanical bearings.Firstly,SIDL was employed to learn periodic impulse atoms,and the optimal selection was made based on the gini index(GI)as the selection criterion.Secondly,the max-imum second-order cyclostationary blind deconvolution(CYCBD)algorithm was introduced to enhance the features of the selected atoms.Finally,the fault characteristic signals were reconstructed based on the optimized atoms,and envelope analy-sis was performed.Verification using simulation signals and engineering data demonstrates that SIEDL can effectively ex-tract fault features of traffic electromechanical bearings under low signal-to-noise ratio conditions,and it has certain advan-tages compared to classical SIDL,analytical dictionary algorithms and adaptive CYCBD methods.关键词
故障诊断/轴承/特征提取/字典学习/基尼指数/CYCBDKey words
fault diagnosis/bearing/feature extraction/dictionary learning/GI/CYCBD分类
机械制造引用本文复制引用
郭华,褚惟,张孟,王宽,李应董..基于SIEDL的交通机电轴承故障特征提取研究[J].噪声与振动控制,2025,45(2):132-138,7.基金项目
云南省交通科学研究院有限公司自立资助项目(JKYZLX-2021-20、JKYZLX-2023-14) （JKYZLX-2021-20、JKYZLX-2023-14）
