基于改进小波去噪预处理和EEMD的采煤机齿轮箱故障诊断

发布时间：2018-02-21 23:50

本文关键词： 采煤机齿轮箱故障特征分解效率改进小波去噪集合经验模态分解行星轮模态混叠　出处：《中南大学学报(自然科学版)》2016年10期 　论文类型：期刊论文

【摘要】：针对采煤现场强噪声背景下采煤机齿轮箱振动信号集合经验模态分解(EEMD)故障特征不明显和分解效率较低的问题,提出基于改进小波去噪预处理和EEMD的故障诊断方法。采用小波改进阈值函数法对振动信号进行去噪预处理,与传统小波阈值函数法相比能够有效地提高信号的信噪比。对去噪后的信号进行EEMD分解得到若干个本征模态分量(IMF),计算各IMF分量的相关度并剔除虚假分量。将该方法应用于采煤机齿轮箱行星轮的故障诊断,通过对真实的IMF分量进行频谱分析并提取信号的故障特征频率,与未去噪的信号进行对比。研究结果表明:该方法能够突出故障特征频率,使分解效率提高17.35%,并能进一步减小模态混叠现象。
[Abstract]:In view of the problem that the fault characteristics of EEMDD of shearer gear box vibration signal set are not obvious and the decomposition efficiency is low under the background of strong noise in coal mining field, A fault diagnosis method based on improved wavelet denoising preprocessing and EEMD is proposed. Compared with the traditional wavelet threshold function method, the signal-to-noise ratio (SNR) of the signal can be improved effectively. Several intrinsic mode components are obtained by EEMD decomposition of the de-noised signal, the correlation of each IMF component is calculated and the false component is eliminated. The method is applied to fault diagnosis of planetary gear in shearer gear box. By analyzing the true IMF component and extracting the fault characteristic frequency of the signal, the research results show that the method can outshine the fault feature frequency. The decomposition efficiency is increased by 17.35 and the mode aliasing can be further reduced.
【作者单位】：中南大学机电工程学院;高性能复杂制造国家重点实验室;深海矿产资源开发利用技术国家重点实验室;
【基金】：国家重点基础研究发展计划(973计划)项目(2014CB046305) 国家大洋专项项目(DY125-14-T-03)~~
【分类号】：TD421.6

【相似文献】