摘要
机械系统中滚动轴承局部故障会导致振动信号中出现冲击响应成分,为有效提取轴承故障冲击特征,基于信号稀疏表示理论,提出了基于改进Laplace字典的轴承故障诊断方法。通过快速Fourier变换及相关滤波法确定基底函数的参数,构造Laplace小波基并通过错位拓展转换为Toeplitz矩阵,以此矩阵作为稀疏表示字典,实现振动信号在Laplace小波基下的稀疏表示,提取信号中瞬态冲击成分特征。对仿真信号及轴承故障信号的验证结果表明:基于改进Laplace字典的稀疏表示方法可有效提取冲击信号特征,实现轴承典型故障诊断。
The local faults of rolling bearings in mechanical system tend to result in impact response component in vibration signals.In order to effectively extract fault impact characteristics of the bearings,based on theory of signal sparse representation,the fault diagnosis method for the bearings is proposed based on improved Laplace dictionary.The parameters of basis function are determined by fast Fourier transform and correlation filtering method.The Laplace wavelet base is built and converted to Toeplitz matrix dislocation expansion.The matrix is taken as sparse representation dictionary to ensure sparse representation of vibration signals under Laplace wavelet base,the transient impact component characteristics in signals are extracted.The verification results of simulation signals and fault signals of the bearings show that the sparse representation method based on improved Laplace dictionary is able to extract characteristics of impact signals effectively and realize typical fault diagnosis of the bearings.
作者
李景乐
谢馨
王华庆
LI Jingle;XIE Xin;WANG Huaqing(School of Mechanical and Electrical Engineering,Beijing University of Chemical Technology,Beijing 100029,China)
出处
《轴承》
北大核心
2018年第11期57-60,共4页
Bearing
基金
国家自然科学基金项目(51675035
51375037)
关键词
滚动轴承
故障诊断
稀疏表示
Laplace小波
相关滤波
rolling bearing
fault diagnosis
sparse representation
Laplace wavelet
correlation filtering
