工程科学学报，第39卷，第6期：909-916,2017年6月 Chinese Journal of Engineering,Vol.39,No.6:909-916,June 2017 D0L:10.13374/j.issn2095-9389.2017.06.014;htp:/journals.ustb.edu.cn 形态分量分析在滚动轴承故障诊断中的应用 刘永兵，周亚凯，冯志鹏四 北京科技大学机械工程学院，北京100083 ☒通信作者，E-mail:fengzp@usth.edu.cn 摘要滚动轴承局部故障振动信号中的周期性冲击是识别故障的关键特征.形态分量分析在由多种形态原子组成的过完 备字典基础上提取信号中的不同形态成分，基于这种思想提出了一种基于新型过完备复合字典的形态分量分析方法.依据 滚动轴承故障振动信号中分量间的形态差异性，改进字典后该方法可以更具针对性地提取出包含故障特征的冲击分量，配合 包络谱分析准确提取故障特征频率，诊断滚动轴承局部故障.对比基于快速谱峭度法的轴承故障诊断方法，该方法可以避免 人为选择共振带产生的不准确性和非最优问题，提高了故障诊断效果.通过轴承仿真信号和故障实验信号分析验证了该方 法的有效性 关键词滚动轴承：故障诊断：形态分量分析：冲击 分类号TP165·.3 Application of morphological component analysis for rolling element bearing fault diagnosis LIU Yong-bing,ZHOU Ya-kai,FENG Zhi-peng School of Mechanical Engineering.University of Science and Technology Beijing,Beijing 100083,China XCorresponding author,E-mail:fengzp@ustb.edu.cn ABSTRACT Periodical impulses in vibration signals are key features in rolling element bearing fault diagnosis.Based on an over- complete dictionary composed of different morphological atoms,morphological component analysis can be used to extract the signal components of different types of morphologies.A new morphological component analysis method based on a novel over-completed dic- tionary was proposed herein.According to morphological differences between components in rolling element bearing fault vibration sig- nal,the method after improved dictionary could more targeted to extract impulse components containing fault feature.Then through en- velope spectrum analysis,the fault characteristic frequency was extracted accurately,and rolling element bearing local faults were di- agnosed.Compared with the Fast Kurtogram method for bearing fault diagnosis,the new method could avoid non-accuracy and non-op- timality problems caused by artificial choice of resonance band,and improve the effectiveness of fault diagnosis.By analyzing both the simulation signal and the experimental dataset of rolling element bearing faults,the proposed method is validated. KEY WORDS rolling element bearing;fault diagnosis;morphological component analysis;impulse 滚动轴承内圈、外圈、滚动体等关键元件出现局部的分析，从而诊断轴承故障. 损伤时，振动信号中出现周期性冲击，有效提取冲击成 对于共振解调方法，如何自适应地找出共振峰和 分，并准确分析它们的重复频率是诊断滚动轴承局部 共振带是关键问题.基于谱峭度的滚动轴承故障诊断 故障的关键.许多学者提出了大量的信号分量分离方 方法依据谱峭度大小，自适应地选取共振带，有效诊断 法，包括共振解凋、Hilbert-huang变换[)、集合经验 了滚动轴承故障.谱峭度的概念最初是由Dwyert提 模式分解和交叉能量算子]等.这些方法的基本思路 出的，后由Antoni对其进行系统定义，并提出了Fast 都是预先提取振动信号中的故障特征分量再做进一步 Kurtogram算法，成功应用于轴承故障诊断).此后， 收稿日期：2016-07-13工程科学学报,第 39 卷,第 6 期:909鄄鄄916,2017 年 6 月 Chinese Journal of Engineering, Vol. 39, No. 6: 909鄄鄄916, June 2017 DOI: 10. 13374 / j. issn2095鄄鄄9389. 2017. 06. 014; http: / / journals. ustb. edu. cn 形态分量分析在滚动轴承故障诊断中的应用 刘永兵, 周亚凯, 冯志鹏苣 北京科技大学机械工程学院, 北京 100083 苣通信作者, E鄄mail: fengzp@ ustb. edu. cn 摘 要 滚动轴承局部故障振动信号中的周期性冲击是识别故障的关键特征. 形态分量分析在由多种形态原子组成的过完 备字典基础上提取信号中的不同形态成分,基于这种思想提出了一种基于新型过完备复合字典的形态分量分析方法. 依据 滚动轴承故障振动信号中分量间的形态差异性,改进字典后该方法可以更具针对性地提取出包含故障特征的冲击分量,配合 包络谱分析准确提取故障特征频率,诊断滚动轴承局部故障. 对比基于快速谱峭度法的轴承故障诊断方法,该方法可以避免 人为选择共振带产生的不准确性和非最优问题,提高了故障诊断效果. 通过轴承仿真信号和故障实验信号分析验证了该方 法的有效性. 关键词 滚动轴承; 故障诊断; 形态分量分析; 冲击 分类号 TP165 + 郾 3 Application of morphological component analysis for rolling element bearing fault diagnosis LIU Yong鄄bing, ZHOU Ya鄄kai, FENG Zhi鄄peng 苣 School of Mechanical Engineering, University of Science and Technology Beijing, Beijing 100083, China 苣Corresponding author, E鄄mail: fengzp@ ustb. edu. cn ABSTRACT Periodical impulses in vibration signals are key features in rolling element bearing fault diagnosis. Based on an over鄄 complete dictionary composed of different morphological atoms, morphological component analysis can be used to extract the signal components of different types of morphologies. A new morphological component analysis method based on a novel over鄄completed dic鄄 tionary was proposed herein. According to morphological differences between components in rolling element bearing fault vibration sig鄄 nal, the method after improved dictionary could more targeted to extract impulse components containing fault feature. Then through en鄄 velope spectrum analysis, the fault characteristic frequency was extracted accurately, and rolling element bearing local faults were di鄄 agnosed. Compared with the Fast Kurtogram method for bearing fault diagnosis, the new method could avoid non鄄accuracy and non鄄op鄄 timality problems caused by artificial choice of resonance band, and improve the effectiveness of fault diagnosis. By analyzing both the simulation signal and the experimental dataset of rolling element bearing faults, the proposed method is validated. KEY WORDS rolling element bearing; fault diagnosis; morphological component analysis; impulse 收稿日期: 2016鄄鄄07鄄鄄13 滚动轴承内圈、外圈、滚动体等关键元件出现局部 损伤时,振动信号中出现周期性冲击,有效提取冲击成 分,并准确分析它们的重复频率是诊断滚动轴承局部 故障的关键. 许多学者提出了大量的信号分量分离方 法,包括共振解调[1] 、Hilbert鄄鄄 huang 变换[2] 、集合经验 模式分解和交叉能量算子[3]等. 这些方法的基本思路 都是预先提取振动信号中的故障特征分量再做进一步 的分析,从而诊断轴承故障. 对于共振解调方法,如何自适应地找出共振峰和 共振带是关键问题. 基于谱峭度的滚动轴承故障诊断 方法依据谱峭度大小,自适应地选取共振带,有效诊断 了滚动轴承故障. 谱峭度的概念最初是由 Dwyer [4] 提 出的,后由 Antoni 对其进行系统定义,并提出了 Fast Kurtogram 算法,成功应用于轴承故障诊断[5] . 此后