EEMD和TFPF联合降噪法在齿轮故障诊断中的应用
发布时间：2019-06-08 12:33
【摘要】：为了消除噪声对齿轮传动系统故障特征提取的影响,提出了一种基于集成经验模态分解(ensemble empirical mode decomposition,简称EEMD)和时频峰值滤波(time-frequency peak filtering,简称TFPF)相结合的降噪方法。针对TFPF算法在窗长的选择方面受到限制的问题,采用了EEMD方法对其进行改进,使得信号在噪声压制和有效信号保真两方面得到权衡;含噪声的信号经过EEMD分解后,得到一系列频率成分从高到低的本征模态函数(intrinsic mode functions,简称IMFs),计算出各IMFs间的相关系数,判断需要滤波的IMFs。对不同的IMFs选择不同的窗长进行TFPF滤波,把过滤后的IMFs和剩余的IMFs重构得到最终的降噪信号。用模拟仿真信号和齿轮齿根故障信号对该方法进行验证,可见EEMD+TFPF能有效地去除噪声,成功提取齿根裂纹故障特征。
[Abstract]:In order to eliminate the influence of noise on fault feature extraction of gear transmission system, a noise reduction method based on integrated empirical mode decomposition (ensemble empirical mode decomposition, (EEMD) and time-frequency peak filtering (TFPF) is proposed. In order to solve the problem that the selection of window length of TFPF algorithm is limited, the EEMD method is used to improve it, which makes the signal balance between noise suppression and effective signal fidelity. After the signal with noise is decomposed by EEMD, a series of intrinsic modal functions (intrinsic mode functions, with frequency components from high to low are obtained to calculate the correlation coefficients among IMFs, and the IMFs. that needs to be filtered is judged. Different IMFs window lengths are selected for TFPF filtering, and the filtered IMFs and the remaining IMFs are reconstructed to obtain the final noise reduction signal. The simulation signal and gear root fault signal are used to verify the method. It can be seen that EEMD TFPF can effectively remove noise and successfully extract the fault characteristics of tooth root crack.
【作者单位】： 太原理工大学机械工程学院;太原科技大学机械工程学院;
【基金】：国家自然科学基金资助项目(50775157) 山西省基础研究资助项目(2012011012-1)
【分类号】：TH132.41;TH17
本文编号：2495283
[Abstract]:In order to eliminate the influence of noise on fault feature extraction of gear transmission system, a noise reduction method based on integrated empirical mode decomposition (ensemble empirical mode decomposition, (EEMD) and time-frequency peak filtering (TFPF) is proposed. In order to solve the problem that the selection of window length of TFPF algorithm is limited, the EEMD method is used to improve it, which makes the signal balance between noise suppression and effective signal fidelity. After the signal with noise is decomposed by EEMD, a series of intrinsic modal functions (intrinsic mode functions, with frequency components from high to low are obtained to calculate the correlation coefficients among IMFs, and the IMFs. that needs to be filtered is judged. Different IMFs window lengths are selected for TFPF filtering, and the filtered IMFs and the remaining IMFs are reconstructed to obtain the final noise reduction signal. The simulation signal and gear root fault signal are used to verify the method. It can be seen that EEMD TFPF can effectively remove noise and successfully extract the fault characteristics of tooth root crack.
【作者单位】： 太原理工大学机械工程学院;太原科技大学机械工程学院;
【基金】：国家自然科学基金资助项目(50775157) 山西省基础研究资助项目(2012011012-1)
【分类号】：TH132.41;TH17
【相似文献】
相关硕士学位论文 前1条
1 王少君;基于EEMD的滚动轴承微弱故障特征提取方法研究[D];石家庄铁道大学;2016年
，本文编号：2495283
本文链接：https://www.wllwen.com/jixiegongchenglunwen/2495283.html
