机电工程技术2024，Vol.53Issue(11)：215-219,5.DOI:10.3969/j.issn.1009-9492.2024.11.046
基于改进ISTA算法的高速列车轴箱轴承故障诊断方法研究
Research on Fault Diagnosis Method for High-speed Train Axle Box Bearings Based on the Improved ISTA Algorithm
摘要
Abstract
Addressing the issue of effectively identifying the overlapping characteristics in vibration signals such as transient shock components,background noise,and harmonic components,an improved ISTA algorithm for fault diagnosis of high-speed train axle box bearings is proposed.Firstly,the optimal wavelet atom corresponding to the vibration fault signal is obtained from the original signal using the correlation filtering method,and an overcomplete dictionary with sparse representation is constructed.Then,considering that the reconstruction of a single Laplace wavelet by ISTA under a strong noise background can lead to severe distortion,the sparsity of the sparse representation coefficients is improved by modifying the threshold shrinkage function,enhancing the reconstruction ability of the fault vibration signal.Finally,simulation signals and experiments show that the improved ISTA algorithm clearly restores the original signal in terms of signal reconstruction,effectively suppressing background noise.In terms of fault feature extraction,it exhibits superior feature extraction capabilities,and the fault characteristic frequency components are also better enhanced in envelope analysis,effectively achieving fault diagnosis of the bearing outer ring.It has certain practical significance for the fault diagnosis of high-speed train axle box bearings.关键词
轴承/稀疏表示/Laplace小波原子/改进ISTA算法/故障诊断Key words
bearings/sparse representation/Laplace wavelet atoms/improved LEIST algorithm/fault diagnosis分类
交通工程引用本文复制引用
宁善平,赵晨,武文星..基于改进ISTA算法的高速列车轴箱轴承故障诊断方法研究[J].机电工程技术,2024,53(11):215-219,5.基金项目
广东省教育厅青年创新人才基金(2022KQNCX191) （2022KQNCX191）
