增强组合差分乘积形态学滤波的轴承故障特征提取方法

针对滚动轴承故障信号的非线性、非平稳、强噪声特性导致的常规时频域特征提取方法受限问题，提出一种增强组合差分乘积形态学滤波的轴承故障特征提取方法。在分析数学形态学4种基本运算的正、负冲击脉冲提取特性的基础上，运用级联、差分、乘积构造的一种新的组合差分乘积算子(combination difference multiply operator, CDMO)具备了同时提取正、负冲击脉冲的能力，并发挥梯度乘积运算对脉冲提取更敏感的优势，实现故障信息的充分提取。引入故障特征频率比指标优化CDMO结构元素参数，修正待处理信号的几何特征，提取与结构元素相匹配的信号特征信息。在CDMO滤波的基础上，借助三阶累积量切片谱技术能够抑制高斯噪声、突出二次耦合分量的优势，准确提取故障特征频率及其倍频，增强轴承故障特征提取能力并抑制噪声干扰。依托2种不同来源的工程实际信号并与经典故障特征提取方法对比分析，验证了所提方法的有效性。

Bearing fault feature extraction method based on enhanced combination difference multiply morphological filter

To address the limitations of conventional time-frequency domain feature extraction methods when dealing with the non-linear, non-stationary and strongly noisy characteristics of rolling bearing fault signals, a bearing fault feature extraction method based on an enhanced combination difference multiply morphological filter is proposed in this study. Based on the understanding of the positive and negative shock pulse extraction characteristics of the four basic operations of mathematical morphology, a new combination difference multiply operator (CDMO) is constructed. This CDMO has the ability to simultaneously extract positive and negative shock pulses by combining cascade, difference and multiply operations. The gradient multiply operation that is more sensitive to pulse extraction is utilized to achieve comprehensive fault information extraction. The fault characteristic frequency ratio index is introduced to optimize the parameters of the CDMO structural elements. This optimization modifies the geometric characteristics of the signal to be processed, allowing for the extraction of signal characteristic information that matches the structural elements. Following CDMO filtering, third-order cumulant slice spectrum technology is employed to suppress Gaussian noise and highlight the advantages of secondary coupling components. This enhances the ability to accurately extract fault feature frequencies and their multiplications, thus improving bearing fault feature extraction and suppressing noise interference. The proposed method’s effectiveness is verified by relying on actual engineering signals from two different sources and comparing its performance with classic fault feature extraction methods.