并行RSSD和改进MOMEDA的齿轮箱故障诊断  

Gearbox Fault Diagnosis Based on Parallel RSSD and Improved MOMEDA

作　　者：尹志安[1] 孙文龙 王凯[1] YIN Zhi-an;SUN Wen-long;WANG Kai(Nanchang Institute of Science and Technology,Jiangxi Nanchang 330108,China)

机构地区：[1]南昌工学院,江西南昌330108

出　　处：《机械设计与制造》2024年第9期196-204,共9页Machinery Design & Manufacture

基　　金：江西省教育厅科学技术研究项目(GJJ202508)。

摘　　要：为了克服传统共振稀疏信号分解与矩量法的局限性,提高其提取微弱故障特征的能力,提出了一种并行双参数优化RSSD和改进MOMEDA的行星齿轮箱故障诊断方法。首先,并行双参数优化RSSD构造了与不同故障特征相匹配的小波基函数,并将复合故障信号自适应分解为不同的谐振分量,实现了复杂故障特征的解耦。其次,利用改进MOMEDA对共振分量进行去卷积滤波,有效地消除了复杂传输路径和强环境噪声的影响,增强了与弱故障相关的脉冲。最后,通过对行星齿轮箱实验平台的实际故障信号的分析,证明了提出的方法不仅具有良好的解耦性能以及提取弱故障信号能力,且能够全面、准确地提取不同类型的故障。In order to overcome the limitations of traditional resonance sparse signal decomposition and mom,and improve its ability to extract weak fault features,a parallel dual parameter optimization RSSD and improved MOMEDA method for plan⁃etary gearbox fault diagnosis was proposed.Firstly,the parallel two parameter optimization RSSD constructed wavelet basis func⁃tions matching with different fault characteristics,and decomposed the composite fault signal into different resonance compo⁃nents adaptively to realize the decoupling of complex fault features.Secondly,the resonance component was deconvoluted by the improved MOMEDA,which effectively eliminated the influence of complex transmission path and strong environmental noise,and enhanced the weak fault related pulse.Finally,through the analysis of the actual fault signals of the planetary gearbox ex⁃perimental platform,it is proved that the proposed method not only has good decoupling performance and the ability to extract weak fault signals,but also can comprehensively and accurately extract different types of faults.

关 键 词：共振稀疏信号分解 多点最优最小熵反褶积 行星齿轮箱 故障诊断 

分 类 号：TH16[机械工程—机械制造及自动化] TH133.33[自动化与计算机技术—控制理论与控制工程] TP18[自动化与计算机技术—控制科学与工程]