基于高阶累积量自适应算法的列车轴承的故障诊断

论述三阶累积量递推最小二乘自适应算法（CDRLS）和三阶累积量最小均方误差自适应算法（CDEFWLMS），及其在列车轴承故障诊断中的应用。通过对列车滚动轴承的保持架断裂和滚子掉块典型故障信号分析，得出高阶累积量自适应算法具有良好的降噪性，CDEFWLMS算法比CDRLS算法处理后的特征频率更突出明显。研究表明：用高阶累积量自适应滤波算法提取信号特征，容易地分离正常轴承信号、保持架断裂和滚子掉块故障信号，从而验证了高阶累积量自适应滤波在故障诊断、检测中具有良好的应用特性。

Rolling bearing faults diagnosis based on adaptive filter of third-order cumulants

Both adaptive CDRLS filter and adaptive CDEFWLMS filter, which are both based on the third-order cumulants are applied to the diagnosis of rolling bearing faults. Via analyses on the typical signals of the locomotive bearings, such as bearing sound running signal, bearing cage broken and bearing ball flake, the results show that the adaptive filters based on the third-order cumulants have nice qualities and higher SNR, and CDEFWLMS filter is more effective than CDRLS filter in practice and its peak value in spectrum is more sharp and distinct. By means of FCM in the spot, it is found that the memberships for the signals given by CDEFWLMS are of higher degrees than those given by CDRLS. In a word, these results show both adaptive filters based on the third-order cumulants are helpful for bearing fault diagnosing and monitoring.