摘要
为了消除振动信号中离散频率分量和强背景噪声对提取滚动轴承故障特征频率的干扰,提出了一种新的基于倒谱编辑(Cepstrum Editing Procedure,CEP)预白化和形态学自互补Top-Hat变换的方法用于滚动轴承的故障特征提取。CEP能够去除故障振动信号中的周期性频率成分,剩余只包含背景噪声和碰撞损伤引起的非平稳冲击成分的白化信号,通过分析构造的形态学自互补Top-Hat变换滤波器,提出采用故障特征幅值能量比(Feature Amplitude Energy Radio,FAER)的方法自适应确定最优结构元素的尺度,预白化信号经过形态学滤波有效消除了背景噪声的干扰,提取了较为清晰的轴承故障特征频率。对实测轴承滚动体、内圈故障信号进行了分析,结果表明该方法可有效提取滚动轴承故障冲击成分并抑制噪声。
In order to eliminate interferences from discrete frequency components and strong background noises in vibration signals to extract fault features of rolling bearings,a new method based on combining the pre-whitening technology using cepstrum editing procedure（ CEP） and the morphology self-complementary Top-Hat（ STH）transformation theory was presented for extracting fault features of rolling bearings. CEP could eliminate periodic frequency components in fault vibration signals,and remain pre-whitening signals containing only non-stationary impact components from collision damage and background noise. Then,through analyzing the filter constructed with the morphology selfcomplementary Top-Hat transformation, a novel method named fault feature amplitude energy radio（ FAER） was presented to adaptively select the optimal structural element（ SE） scale. The interferences of background noise in prewhitening signals were eliminated effectively with morphological filtering,and the clearer fault feature frequencies of rolling bearings were extracted. The result showed that the proposed method is effective to extract fault impact components of roll bearings and suppress noises by analyzing vibration signals of rolling bearings with ball and inner race faults.
出处
《振动与冲击》
EI
CSCD
北大核心
2015年第15期77-81,149,共6页
Journal of Vibration and Shock
基金
国家自然科学基金项目(51307058)
河北省自然科学基金(E2014502052)
中央高校基本科研业务专项资金项目(2014XS83)
关键词
滚动轴承
倒谱编辑
信号预白化
自互补Top-Hat变换
特征幅值能量比
rolling bearing
cepstrum editing procedure
signal pre-whitening
self-complementary Top-Hat transformation
feature amplitude energy radio
