As the problem of the fault signal of rolling bearing is a multi-component and non-stationary vibration signal, which is difficult to diagnose when the signal has weak initial features. A fault diagnosis method based on variational mode decomposition (VMD) and spectral kurtosis (SK) was proposed in the paper. Firstly, the vibration signal was decomposed into several component signals by VMD, and the component which had the maximum kurtosis and had the most fault impact components was selected as the optimal component. Then, the fast SK was computed to the optimal component for band pass filtering and highlighting the fault impact components. Finally, the fault was diagnosed by analyzing the fault frequency appeared in the filtered signal envelope spectrum. The experimental analysis results show that the proposed method can diagnose bearing incipient faults effectively and it has certain engineering application value.%针对滚动轴承故障信号为多分量非平稳振动信号、故障早期特征微弱诊断困难的问题,该文提出变分模态分解(VMD)结合谱峭度的滚动轴承早期故障诊断方法.首先对振动信号进行VMD分解得到若干分量信号,选择峭度最大分量作为最优分量,然后对最优分量进行快速谱峭度计算并进行带通滤波、凸显故障冲击成分,通过分析滤波信号包络谱中故障频率成分实现故障诊断.实验数据分析结果表明该方法能有效诊断轴承早期故障,有一定的工程应用价值.
展开▼
