基于EMD和TVAR模型的滚动轴承故障诊断方法

摘要

针对经验模态分解（EMD）的固有模态函数（IMF）选择问题,提出了一种基于能量阀值的自适应筛选方法。对筛选出的IMF分量进行信号重构后,提取重构信号时变自回归模型的时变参数特征进行支持向量机分类器（SVM）的智能故障识别。仿真实验结果表明,所提的基于能量阀值的IMF分量选择方法能够有效提取滚动轴承的故障信息,有利于滚动轴承运行状态的识别。%A self-adapting solution based on an energy threshold value is proposed in view of the problem of selecting intrinsic mode function （IMF） of empirical mode decomposition （EMD）. The time-varying autoregressive parameters in signal reconstruction of the selected IMFs are extracted as the feature. Finally, the feature is put in the support vector machine classifier to recognize the state of rolling bearings.