摘要
针对经验小波变换在频谱划分时检测的边界过于集中在频谱中幅值较大频率段的问题,对经验小波变换在频域内利用检测极大值来进行频谱划分的方式进行研究,提出基于频谱趋势的改进型经验小波变换,利用仿真信号和汽车座椅水平驱动器振动信号进行测试。研究结果表明,该方法能够使啮合频率及其谐波与边频带作为一个模态被分解出来。通过对分量信号进行Hilbert变换解调分析,实现对齿轮箱故障的定位。
When the EWT method is used for frequency spectrum classification,the boundary of detection is overconcentrated in the frequency range of larger amplitude.To solve the problem,the spectrum segment manner in the frequency domain by extracting the maximum value in the EWT is studied and the improved EWT based on spectrum trend is proposed.The simulation signal and the vibration signal of vehicle’s seat horizontal driver are tested by this method.The results indicate that the method can extract the meshing frequency and its octave with its side band as one modal.Through demodulation analysis of the modulated signal by performing Hilbert transform to the signal components,the fault localization of gearboxes can be realized.
作者
王昌明
张征
李峰
鲁聪达
WANG Changming;ZHANG Zheng;LI Feng;LU Congda(School of Mechanical Engineering,Zhejiang University of Technology,Hangzhou 310032,China;Ningbo Shuanglin Auto Parts Co.Ltd.,Shanghai 201799,China)
出处
《噪声与振动控制》
CSCD
2018年第5期167-172,共6页
Noise and Vibration Control
基金
国家自然科学基金资助项目(51675485
11672269)
浙江省杰出青年基金资助项目(LR18E050002)
浙江省科技计划公益基金资助项目(2016C31040)
关键词
振动与波
经验小波变换
频谱趋势
频谱划分
解调分析
vibration and wave
empirical wavelet transform(EWT)
spectrum envelope
spectrum segmentation
demodulation analysis
