介绍了Mallat算法,提出了将Mallat算法应用于多种混叠的振动信号的特征提取。建立了转子不对中模拟实验台,把采集到的支承振动信号进行Mallat分解和重构,重构后的信号能够有效地体现支承振动的细部特征,有效地提高信号的分辨率。结合实测的支承负荷和轴心轨迹很好地辨识了转子不对中故障。
Algorithm of Mallat was introduced briefly, and a method that algorithm of Mallat was applied to character extraction from a variety of aliasing signal was presented. A test rig was set up to simulate mis -alignment faults, vibration signals of bearings collected were decomposed and reconstructed. Reconstructed signal can reflect detail characters of bearings vibration, effectively improve resolution of signal. Combining with load of bearings and orbits of axis has identified mis - alignment faults.