摘要
为了准确地进行滚动轴承故障诊断,针对故障振动信号的低信噪比、非线性、非平稳的特征,提出了奇异值分解降噪、局部特征尺度分解和1.5维谱相结合的故障诊断方法。该方法首先运用奇异值分解降噪技术降低信号中的噪声,然后对降噪信号进行局部特征尺度分解,得到若干个内禀尺度分量,并进行Hilbert变换求取包络信号,最后求取包络信号的1.5维谱提取故障特征。通过轴承内圈故障数据分析,验证了该方法的有效性。
In order to accurately diagnosis of fault for rolling bearings, aiming at low signal - noise rate, nonlinear and non- stationary characteristics of fault vibration signals, a fault diagnosis method is proposed based on singular value decomposition（SVD） denoising, local characteristic -scale decomposition （LCD） and 1. 5 dimension spectrum. First- ly, the noise of signal is reduced by SVD denoising technology. Secondly, the denoising signal is decomposed into several intrinsic scale components （ISC） by LCD, and then the envelope signal is obtained by Hilbert transform. At last, the 1.5 dimension spectrum of envelope signal is obtained to extract fault feature. The effectiveness of method is verified through analysis of fault data for inner rings of bearings.
出处
《轴承》
北大核心
2016年第1期54-58,64,共6页
Bearing
基金
国家部委预研基金项目(9140A27020214JB1446)
关键词
滚动轴承
故障诊断
局部特征尺度分解
奇异值分解
1.5维谱
rolling hearing
fault diagnosis
local characteristic -scale decomposition
singnlar value decomposition
1.5 dimension spectrum
