信号瞬变成分检测与提取及其在故障诊断中的应用

Detection and Extraction of Signal Transients and Their Applications in Fault Diagnosis

研究基于小波变换的信号瞬变成分检测与提取方法及其在机械故障诊断中的应用 .在分析信号的连续小波变换的模极大值理论的基础上 ,指出连续小波变换系数的模与信号瞬变成分的关系 ;通过分析小波函数的性质 ,分析小波函数对信号的连续小波变换的影响 ;在信号瞬变特征的提取过程中 ,提出基于门限值的特征重建方法 .将该方法应用于齿轮箱振动信号中瞬变成分的检测与重建 .结果表明基于连续小波变换方法能有效检测到信号中的瞬变成分 ,瞬变成分的重建结果有效地表示了机械的故障状态 .

The detection and extraction of transients through continuous wavelet transform are studied for gearbox fault diagnosis. The modulus local maximum theory of continuous wavelet transform is introduced for the analysis of the relation between the transform coefficients and the transients. The feature of mother wavelets and its influence on the continuous wavelet transform is presented. To extract the transients,the threshold is set in the reconstruction of signal transient feature. The proposed approaches are utilized in the detection and extraction of transient component in the gearbox vibration for fault diagnosis and the outcomes indicate that continuous wavelet transform is quite effective in detecting the existence of the transients and the extracted transients in the gearbox vibration reveal the fault of the gearbox effectively.