机械制造与自动化2025，Vol.54Issue(3)：23-27,5.DOI:10.19344/j.cnki.issn1671-5276.2025.03.004
基于RF特征优选和EEMD-SSAE的行星齿轮箱故障诊断
Fault Diagnosis of Planetary Gear Box Based on RF Feature Optimization and EEMD-SSAE
摘要
Abstract
For the low recognition rate in planetary gearbox fault diagnosis due to insufficient feature extraction,a planetary gearbox fault diagnosis method combining RF feature optimization and EEMD-SSAE is studied.The time domain signal is decomposed by EEMD.IMF components with large correlation coefficients are selected based on Pearson correlation coefficient,and time-domain features,frequency domain features and original signal features are extracted to construct the dataset.RF is used to eliminate redundant features,a new dataset is constructed as input to the SSAE network,and with softmax classifier,fault classification is implemented.The results show that the proposed method is superior to other models in terms of accuracy and robustness under mixed conditions and noise interference.关键词
故障诊断/行星齿轮箱/堆栈稀疏自编码器/总体平均经验模态分解/特征优选Key words
fault diagnosis/planetary gear box/stacked sparse autoencoder/ensemble empirical mode decomposition/feature optimization分类
机械制造引用本文复制引用
刘维团,王友仁,蒋浩宇..基于RF特征优选和EEMD-SSAE的行星齿轮箱故障诊断[J].机械制造与自动化,2025,54(3):23-27,5.基金项目
航空科学基金资助项目(20183352030) （20183352030）
