基于自适应CYCBD和互相关谱的滚动轴承复合故障诊断方法

强背景噪声环境下,复合故障信号中的多特征提取与分离是实现滚动轴承复合故障诊断的重点与难点。提出了基于自适应最大二阶循环平稳盲卷积(CYCBD)和互相关谱的滚动轴承复合故障特征提取方法。首先,基于故障信号特点,通过设置CYCBD中不同的循环频率,提取不同频率的故障冲击成分,并以最大谐波显著性指数(HSI)为依据,自适应选取CYCBD的最优滤波器长度;然后,利用互相关分析进一步抑制信号中的噪声,提高信噪比;最终通过快速傅里叶变换(FFT)实现滚动轴承故障特征的提取。仿真和实测信号的分析结果表明,该方法能够去除背景噪声的干扰、提取滚动轴承复合故障特征,验证了方法的有效性。

Compound faults diagnosis of rolling element bearing using adaptive CYCBD and cross-correlation spectrum

The separation and extraction of multi-features from the bearing compound faults signals is the key and difficult point for the identification of compound faults of rolling element bearing, especially when the signals are masked by strong background noise. Hence, a new compound faults diagnosis method is proposed in this paper based on the combination of adaptive maximum second-order cyclostationarity blind deconvolution (CYCBD) and cross-correlation spectrum. First, based on the characteristic of fault signals, various fault features are separated by using different cycle frequencies in CYCBD, the optimal filter length is determined based on the HSI; then cross-correlation is calculated to further suppress the noise; finally, fast fourier transform (FFT) is employed to acquire the cross-correlation spectrum where the fault features can be detected. Verification is performed on both simulation and experimental signals, results show that the proposed method is suitable for detecting compound faults in rolling element bearing.