基于改进CEEMDAN和TEO的轴承故障特征提取方法  

Bearing Fault Feature Extraction Method Based on Improved CEEMDAN and TEO

作　　者：陆彦希 曹乐 LU Yanxi;CAO Le(School of Electrical and Electronic Engineering,Shanghai University of Engineering Science,Shanghai 201620,China)

机构地区：[1]上海工程技术大学电子电气工程学院,上海201620

出　　处：《噪声与振动控制》2020年第2期109-114,共6页Noise and Vibration Control

基　　金：国家自然科学基金资助项目(61703270)。

摘　　要：故障特征的提取是检测与识别故障类型的关键。为优化自适应噪声完整集成经验模态分解(CEEMDAN)的降噪效果,提出一种基于综合评价模型(SDEA)改进的CEEMDAN降噪方法,优化CEEMDAN的去噪效果。该方法首先建立综合评定模型,然后通过峭度准则、相关系数筛选特征模态重构信号,由SDEA评价信号去噪效果。经过多次迭代,选择综合指标最高的迭代次数作为最优降噪信号,再利用能量贡献率选取最优降噪信号的IMF进行重构,最后通过Teager能量算子解调对信号进行包络谱分析,进而得到故障特征频率。实测数据证明,此方法能够准确提取故障特征频率,实现对故障信号的识别,且相较于现有方法可提高信噪比和运算效率。因此采用该方法可为早期轴承故障诊断提供一种有效的解决方案。Extraction of fault features is the key for detecting and identifying fault types.In order to optimize the noise reduction effect of CEEMDAN,an improved CEEMDAN noise reduction method based on the comprehensive evaluation model(SDEA)was proposed to optimize the denoising effect of the CEEMDAN.First of all,the comprehensive evaluation model is established.Then,the characteristic modal reconstruction signals are filtered by using kurtosis criterion and correlation coefficient.And the signal denoising effect is evaluated by SDEA.After several iterations,the iteration number with the highest comprehensive index is selected as the optimal noise reduction signal,and the IMF of the optimal noise reduction signal is selected by using the energy contribution rate for reconstruction.Finally,the envelope spectrum of the signal is demodulated by Teager energy operator(TEO)analysis,and the fault characteristic frequency is obtained.The measurement data verifies that this method can accurately extract the fault characteristic frequency to realize the fault recognition of the signal,and improve the signal-to-noise ratio and the operation efficiency compared with the existing methods.Therefore,this method provides an effective solution for early bearing fault diagnosis.

关 键 词：故障诊断 CEEMDAN 综合评价模型 TEAGER能量算子 故障特征提取 

分 类 号：TH13[机械工程—机械制造及自动化]