基于滚动轴承故障诊断的AR预测滤波器阶数问题研究

PDF(1591 KB)

振动与冲击 ›› 2012, Vol. 31 ›› Issue (4) : 44-47.

论文

基于滚动轴承故障诊断的AR预测滤波器阶数问题研究

从飞云; 陈进; 董广明

作者信息 +

Order selection research of AR predict filter for rolling bearing diagnosis

Cong Fei-yun, Chen Jin, Dong Guang-ming

Author information +

文章历史 +

摘要

通过对滚动轴承故障原理的研究，详细讨论的了利用AR预测滤波器对滚动轴承故障进行降噪处理时其最优化阶数选择的问题。根据滚动轴承故障信号的冲击衰减模型，提出了共振衰减整周期截取的定阶思路。通过对仿真信号的研究分析，揭示了应用AR预测滤波器分析故障信号时阶数选取和信号的信噪比、衰减阻尼比、采样频率、结构共振频率之间的关系，为深入分析AR预测滤波器的阶数选择问题提供了良好的基础。同时，利用最大峭度方法实现了最优化阶数的确定，通过对实验数据的分析，验证了本定阶方法的有效性

Abstract

According to the research on the fault mechanism of rolling bearings, this paper gives a detail discussion on the optimal order selection of AR predict filter when applied to denoise process of rolling bearing’s fault signal. Based on the impact decay model in the rolling bearing’s fault signal, we propose an order selecting thinking called entire cycle interception for resonance decay signals. According to the analysis of the simulated signals, we reveal the relation between the order selection problem and signal-to-noise ratio, decay damping ratio, sampling rate, structural resonance frequency when applying AR predict filter for the denoise of rolling bearing’s fault. It provides a good foundation for deep analysis of the AR predict filter’s order selection problem. Further more, we determine the optimal order by minimizing the kurtosis target. The experiment study shows the effectivity of proposed order selection method.