变工况下滚动轴承的故障特征提取  

Fault Feature Extraction of Rolling Bearings under Variable Operating Conditions

作　　者：李书乐 马洁 LI Shule;MA Jie(School of Automation,Beijing Information Science&Technology University,Beijing 100192,China;School of Electrical and Mechanical,Beijing Information Science&Technology University,Beijing 100192,China)

机构地区：[1]北京信息科技大学自动化学院,北京100192 [2]北京信息科技大学机电学院,北京100192

出　　处：《机械科学与技术》2022年第1期1-8,共8页Mechanical Science and Technology for Aerospace Engineering

基　　金：国家自然科学基金项目(61973041);国家重点研发计划项目(2019YFB1705403)。

摘　　要：滚动轴承常被用于风力涡轮机、发动机等旋转机械中,由于负载、电流变化等因素将导致旋转设备中的滚动轴承在变速条件下运行。在变转速的工况下,现有时频分析、共振解调等故障诊断方法并不能有效提取故障特征,且考虑到强大背景噪声下存在故障特征提取困难的问题,本文提出了一种基于广义变分模态分解(Generalized variational mode decomposition,GVMD)和分数阶傅里叶变换(Fractional fourier transform,FRFT)的变工况故障特征提取方法。首先将在变工况下故障特征频率呈非线性分布的原始振动信号广义解调为近似线性分布,其次对解调后的信号进行变分模态分解(Variational mode decomposition,VMD)得到本征模态函数分量(Intrinsic mode functions,IMF),根据相关系数准则选取最优的分量进行分数阶域的滤波,最后通过分析滤波后信号的1.5维包络谱提取故障特征频率。通过滚动轴承仿真数据和实验数据的验证表明本文所提方法能够有效提取变工况下滚动轴承的故障特征频率。Rolling bearings are often used in rotating machinery such as wind turbines and engines.Due to factors such as load and current changes,rolling bearings in rotating equipment will operate under variable speed conditions.Under variable speed conditions,the existing fault diagnosis methods such as time-frequency analysis and resonance demodulation cannot effectively extract fault features,and considering the difficulty of extracting fault features in strong background noise,a variable operating condition fault feature extraction method based on generalized variational mode decomposition(GVMD)and fractional Fourier transform(FRFT)is proposed in this paper.First,the original vibration signal with a non-linear distribution of fault characteristic frequencies under variable operating conditions is generalizedly demodulated to an approximate linear distribution;and then the demodulated signal is subjected to variational mode decomposition(VMD)to obtain several components of the intrinsic mode functions(IMF),the optimal component is selected according to the correlation coefficient criterion for filtering in the fractional order domain;finally,the characteristic frequency of the fault is extracted by analyzing the 1.5 dimensional envelope spectrum of the filtered signal.The results of applying this method to rolling bearing simulation data and actual test data show that this method can effectively extract the fault characteristic frequency of rolling bearings under variable operating conditions.

关 键 词：故障诊断 特征提取 广义变分模态分解 分数阶傅里叶变换 滚动轴承 

分 类 号：TH165.3[机械工程—机械制造及自动化]