Computer Engineering and Applications ›› 2015, Vol. 51 ›› Issue (18): 261-265.

Previous Articles     Next Articles

Application of feature extraction based on S energy spectrum in bearing fault diagnosis

WANG Zijia1,2, FAN Yugang1,2, WU Jiande1,2   

  1. 1.Faculty of Information Engineering & Automation, Kunming University of Science and Technology, Kunming 650500, China
    2.Engineering Research Center for Mineral Pipeline Transportation, YN, Kunming 650500, China
  • Online:2015-09-15 Published:2015-10-13



  1. 1.昆明理工大学 信息工程与自动化学院,昆明 650500
    2.云南省矿物管道输送工程技术研究中心,昆明 650500

Abstract: To accurately extract the features of vibration signal is essential in detecting rolling bearing fault. Therefore, a method named feature extraction based on S energy spectrum is brought up in this paper. By performing S transform on the vibration signals, time-frequency matrix is obtained and S energy spectrum is constructed. S energy spectrum is decomposed into singular values that can reflect the energy distribution and analyses with the help of Singular Value Decomposition (SVD). Through utilizing the Variable Predictive Model based Class Discriminate(VPMCD) and comparing the interrelation among singular value vectors of S energy spectrum, the fault identification model is constructed. The experimental results prove that the proposed method applied to the bearing fault diagnosis acquires a better correction rate.

Key words: S energy spectrum, Singular Value Decomposition(SVD), Variable Predictive Model based Class Discriminate(VPMCD), fault diagnosis

摘要: 准确提取振动信号的特征,是滚动轴承故障检测的关键问题,为此提出一种基于S能量谱特征提取的故障诊断方法。该方法对振动信号进行S变换,得到时频矩阵,并构建S能量谱,对S能量谱进行奇异值分解(Singular Value Decomposition,SVD)分析,得到能够反映S能量谱特征的奇异值,利用变量预测模型(Variable Predictive Model based Class Discriminate,VPMCD)方法,通过建立特征值之间的内在关系,构建故障识别模型。将所提方法应用于滚动轴承故障检测,实验结果表明,S能量谱特征提取轴承故障诊断方法具有较高的正判率。

关键词: S能量谱, 奇异值分解(SVD), 基于变量预测模型的模式识别(VPMCD), 故障诊断