小波包能量谱-稀疏核主元在故障检测中的应用

Application of wavelet packet energy spectrum and sparse kernel principal component in fault detection

针对滚动轴承故障检测的问题,提出一种基于小波包能量谱-稀疏核主元的滚动轴承故障检测方法。对振动信号进行小波包分解,提取信号的能量频谱,用增量式样本基构造方法,提取能量频谱的样本基,以此样本基建立核主元模型,来分析轴承振动信号能量频谱的变化。通过实验仿真来说明此算法的有效性。

For the problem of rolling bearing fault detection, a method of rolling bearing fault detection is proposed, which is based on wavelet packet energy spectrum and sparse kernel principal component. The vibration signal is decomposed by wavelet packet, in order to extract the energy spectrum of the signal. Then the sample base of energy spectrum is extracted through the method of incremental sample base. A kernel principal component model is built by the sample base for the analysis of the energy spectrum of the bearing vibration signal. The experimental simulation is presented to illustrate the effectiveness of the algorithm.