基于EEMD能量熵及LS-SVM滚动轴承故障诊断

摘要

针对滚动轴承振动信号的非平稳特性和现实中难以获得大量典型故障样本的实际情况，提出基于集合经验模态分解（EEMD）能量熵和最小二乘支持向量机（LS-SVM）的滚动轴承故障诊断方法。首先通过EEMD分解将非平稳的原始振动信号分解成若干个平稳的固有模态函数（IMF）；滚动轴承同一部位发生不同严重程度的故障时，在不同频带内的信号能量值会发生改变，因此可通过计算振动信号的EEMD能量熵判断发生故障的严重程度；从包含主要故障信息的IMF分量中提取的能量特征作为输入来建立支持向量机，判断滚动轴承的技术状态和故障严重程度，并选用不同核函数对诊断效果进行分析比较。实验结果表明，该方法能有效地应用于滚动轴承的故障诊断。%In view of the non-stationary features of vibration signals of rolling bearings and the difficulty to collect a large number of fault samples in practice, a fault diagnosis scheme based on ensemble empirical mode decomposition (EEMD) energy entropy and least square support vector machine (LS-SVM) is proposed. First of all, original acceleration vibration signals are decomposed into several intrinsic mode functions (IMFs). Since the energy of vibration signal changes in different frequency bands when fault occurs, the fault pattern and condition can be identified by calculating and analyzing the EEMD energy entropy. The energy features extracted from several IMFs, which contain the most dominant fault information, serve as the input vector of the support vector machine to judge the technical condition of the bearing and seriousness of the fault. The diagnosis results are analyzed and compared with different kernel functions. The experimental results show that the proposed method is effective.