一种基于分段幂函数插值法的经验模态分解方法及其应用研究

摘要

针对经验模态分解(EMD)方法的分解不完全问题,提出一种改进EMD算法.该算法采用分段幂函数插值法代替原EMD算法中的三次样条插值法,实验表明其分解效果更充分更完全.在此基础上,结合时间序列分析中的AR模型,提出一种基于EMD和AR模型的故障诊断方法,并将其应用到电磁换向阀的故障诊断中.实验结果表明,该方法能够正确有效地实现电磁换向阀的故障诊断.%An improved EMD algorithm was proposed in this paper to overcome the deficiency of the empirical mode decomposition(EMD) method. The algorithm uses the piecewise power function interpolation algorithm instead of the cubic spline interpolation in the original EMD algorithm to generate the envelope. Comparative experiments show the advantages of the improved algorithm. Combined with the AR model in time series analysis, the method of fault feature extraction based on EMD and AR model was proposed and adopted to the fault feature extraction of the electromagnetic valves. The experimental results show that the method can correctly and efficiently extract the fault characteristics of the electromagnetic directional valve.