基于小波理论的故障特征提取

在对齿轮进行故障诊断时,采样信号不可避免地受到各种噪声和干扰的污染,所测信号属于典型的非平稳信号.信号的降噪和特征提取是齿轮状态监测和故障诊断的关键环节.小波理论对于非平稳信号的处理非常有效.在MATLAB环境下,利用小波理论对减速器齿轮箱的采样数据进行去噪实验和分析,提取齿轮大周期故障的特征指标,为进一步进行故障诊断奠定基础.

Fault Feature Based on Wavelet Transform

In the gear fault diagnosis,the sampling signal is inevitably polluted by all kinds of noise and interference,and the measured signals are typical non-stationary signals.Therefore,de-noising and feature extraction are the key step in the condition monitoring and fault diagnosis of gearbox.Traditional signal processing methods are unsatisfactory in signal denoising.Wavelet Transform as a signal processing method is developed rapidly in recent years,and it is very effective to analyze non-stationary signals.In this paper,the wavelet transform was applied to the analysis and de-noises on the sampling data of a reducer gearbox using wavelet theory based on the MATLAB environment and the characteristic of the cycle fault of the gear was extracted,which established a foundation for the further fault diagnosis.