基于小波包分解与权重包络谱的滚动轴承故障特征增强
发布时间：2019-05-24 23:54
【摘要】：针对小波包分解(wavelet packet decomposition,WPD)在对低信噪比信号处理时带内噪声较大,无法准确提取故障信息的问题,提出一种基于小波包分解与权重包络谱的滚动轴承故障特征增强方法。首先对原始信号进行三层小波包分解,并选其最优系数以初步减少噪声干扰,使故障特征信息得到一次增强;而后基于非局部均值算法(Non-Local Means,NLM)对最优系数加权运算得到权重包络曲线,使故障冲击在权重角度得到二次增强;最后对权重包络曲线包络谱分析诊断出故障类型。仿真信号及实验室信号验证了本文方法的有效性及实用性。
[Abstract]:In order to solve the problem that wavelet packet decomposition (wavelet packet decomposition,WPD) can not extract fault information accurately when dealing with low signal-to-noise ratio (SNR) signals, a fault feature enhancement method for rolling bearings based on wavelet packet decomposition and weight envelope spectrum is proposed. Firstly, the original signal is decomposed by three-layer wavelet packet, and the optimal coefficient is selected to reduce the noise interference and enhance the fault feature information once. Then, based on the non-local mean algorithm (Non-Local Means,NLM), the weight envelope curve is obtained by weighted operation of the optimal coefficient, so that the fault impact is enhanced twice at the weight angle. Finally, the fault type is diagnosed by the envelope spectrum analysis of the weight envelope curve. The effectiveness and practicability of the proposed method are verified by simulation signals and laboratory signals.
【作者单位】： 华东交通大学机电与车辆工程学院;
【基金】：国家自然科学基金资助项目(51665013,51265010) 载运工具与装备教育部重点实验室资助课题(15JD02) 江西省青年科学基金资助项目(20161BAB216134)
【分类号】：TH133.33
本文编号：2485298
[Abstract]:In order to solve the problem that wavelet packet decomposition (wavelet packet decomposition,WPD) can not extract fault information accurately when dealing with low signal-to-noise ratio (SNR) signals, a fault feature enhancement method for rolling bearings based on wavelet packet decomposition and weight envelope spectrum is proposed. Firstly, the original signal is decomposed by three-layer wavelet packet, and the optimal coefficient is selected to reduce the noise interference and enhance the fault feature information once. Then, based on the non-local mean algorithm (Non-Local Means,NLM), the weight envelope curve is obtained by weighted operation of the optimal coefficient, so that the fault impact is enhanced twice at the weight angle. Finally, the fault type is diagnosed by the envelope spectrum analysis of the weight envelope curve. The effectiveness and practicability of the proposed method are verified by simulation signals and laboratory signals.
【作者单位】： 华东交通大学机电与车辆工程学院;
【基金】：国家自然科学基金资助项目(51665013,51265010) 载运工具与装备教育部重点实验室资助课题(15JD02) 江西省青年科学基金资助项目(20161BAB216134)
【分类号】：TH133.33
【相似文献】
相关期刊论文 前8条
1 钱苏翔;杨世锡;焦卫东;胡红生;;基于独立分量分析与小波包分解的混叠声源信号波形恢复[J];中国机械工程;2010年24期
2 刘晓波,黄其柏,孙康;小波包分解及其在旋转机械故障诊断中的应用[J];风机技术;2004年06期
3 李自国;郝伟;李凌均;;基于小波包分解和支持向量数据描述的故障诊断方法[J];机械强度;2007年03期
4 任世美;;基于小波包分解的转子叶片裂纹故障特征分析[J];冶金设备;2005年06期
5 王轩;王莉;魏蔚;;瞬时功率小波包分解法在轴承故障诊断中的应用[J];轴承;2010年10期
6 叶瑞召;李万红;;基于小波包分解和BP神经网络的轴承故障诊断[J];轴承;2012年10期
7 王洪刚,郑海起,马吉胜,栾军英;变速箱故障声压信号的小波包分解与诊断[J];振动与冲击;2001年02期
8 ;[J];;年期
相关会议论文 前1条
1 王志刚;吕勇;李友荣;李方;;基于自适应谐波小波包分解的齿轮故障诊断方法[A];第六届全国信息获取与处理学术会议论文集（2）[C];2008年
，本文编号：2485298
本文链接：https://www.wllwen.com/jixiegongchenglunwen/2485298.html
