基于WT特征增强的cICA带式输送机齿轮箱故障诊断

发布时间：2019-03-26 20:59

【摘要】：约束独立分量分析对于测量信号中的传感器噪声(测量噪声)具有很强的免疫能力,但对源噪声的免疫性却很差。针对这个问题,提出了小波变换特征增强的约束独立分量分析的齿轮箱故障特征提取方法。通过对测量信号小波分解,有针对性地选择某子频段小波系数重构,有利于提高信噪比,增强信号的统计独立性和非高斯性,从而增强约束独立分量分析方法提取齿轮故障特征的效果;而未经小波变换除噪时,约束独立分量分析的效果不佳。通过仿真分析和在矿用带式输送机齿轮箱故障诊断的应用结果综合表明,该方法能有效降低源噪声的影响,准确提取出齿轮故障特征,尤其是微弱低频故障特征。为矿用齿轮箱多通道振动状态监测与故障诊断提供了一种新的有效手段和途径。
[Abstract]:Constrained independent component analysis has strong immunity to sensor noise (measurement noise) in the measurement signal, but poor immunity to the source noise. In order to solve this problem, a wavelet transform feature enhanced constrained independent component analysis method for gear box fault feature extraction is proposed. Through the wavelet decomposition of the measured signal, it is beneficial to improve the signal-to-noise ratio (SNR) and enhance the statistical independence and non-Gaussian property of the signal by selecting the wavelet coefficients of a certain sub-band. In order to enhance the effect of extracting gear fault features by constrained independent component analysis method; Without wavelet transform de-noising, the effect of constrained independent component analysis is not good. The results of simulation analysis and application in fault diagnosis of belt conveyor gearbox show that this method can effectively reduce the influence of source noise and extract the fault characteristics of gears, especially the weak low frequency fault characteristics. It provides a new effective method for multi-channel vibration monitoring and fault diagnosis of mine gearbox.
【作者单位】：河南理工大学机械与动力工程学院;
【基金】：国家自然科学基金资助项目(U1304523) 煤炭工业协会指导性基金资助项目(MTKJ2015-261) 河南理工大学博士基金资助项目
【分类号】：TD528.1

【相似文献】