基于小波包和解调分析的多类故障综合诊断方法研究  

Research on the multi-fault comprehensive diagnosis method based on wavelet packet and demodulation

作　　者：杨国安[1] 许飞云[2] 吴贞焕[3] 高金吉[1] 

机构地区：[1]北京化工大学设备诊断工程研究中心,北京100029 [2]东南大学机械工程系,南京210096 [3]北京化工大学信息科学与技术学院,北京100029

出　　处：《东南大学学报（自然科学版）》2004年第1期42-45,共4页Journal of Southeast University：Natural Science Edition

摘　　要：提出了一种基于小波包分解和解调分析相结合的旋转机械故障综合诊断方法 .它能诊断复杂机械的多类故障 ,针对振动信号的非平稳和调制特点 ,首先采用小波包将信号进行分解 ,再分别以不同的方法分析不同的分解频段 .利用解调技术分析小波包分解的高频段 (调制频段 ) ,用来诊断轴承、齿轮等零部件的各类故障 ;利用FFT频谱分析小波包分解的低频段 ,用来诊断转轴的各类故障 .理论分析和仿真结果表明该方法优于仅做FFT频谱分析或Hilbert解调分析 .A comprehensive fault diagnosis method based on the wavelet packet decomposition and demodulation is proposed. The fault signals which are non stationary and modulated are decomposed by the wavelet packet, then the fault features in different frequency bands are extracted respectively. The low frequency bands are analyzed by FFT spectra, and the high frequency bands are demodulated by Hilbert transformation. This makes it possible to diagnose the different faults of machines. The shaft faults can be diagnosed in the low frequency bands, and the bearing and gear faults can be diagnosed in high frequency bands (modulated frequency). By simulation and real example, the results show that the method proposed is superior to the FFT spectra or the Hilbert transform.

关 键 词：小波包 解调 离心泵 故障诊断 

分 类 号：TB52[理学—物理] TN911[理学—声学]