基于包络线算法改进EMD的旋转机械故障诊断方法研究

Study on Rotating Machinery Fault Diagnosis Method Based on Envelopes Fitting Algorithms EMD

针对经验模态分解（EMD）中包络线拟合算法存在的过冲／欠冲以及端点问题，提出采用分段三次Hermite插值解决包络线过冲／欠冲问题，并结合基于斜率再优化的极值点延拓加窗函数的端点效应抑制方法，来优化EMD分解中包络线拟合的结果，得到更准确内禀模态函数（IMF）的同时也可以通过Hilbert变换得到更准确的边界谱和Hilbert谱，提高EMD分解质量。通过仿真分析和滚动轴承内圈故障诊断实例研究表明，该方法有良好的效果，实现旋转机械故障的有效诊断。

Aiming at the problems of the end effects and the overshoot and undershoot during envelope fitting in empirical mode decomposition, a new method is proposed by using pieeewise Hermite interpolation to solve the overshoot and undershoot problems. Combining with the extreme point extension window function, the end effects problem is solved based on slope re-optimization. The results are that the envelope fitting is optimized, the IMFs are more accurate, the Hilbert spectrum and boundary spectrum are also more accurate by Hilbert transform and finally the quality of EMD algorithm is improved. The simulations and the rolling bearing fault diagnosis case show that the new method has a good effect and can effectively diagnose rotating machinery faults.