自适应匹配追踪算法在齿轮故障特征提取的应用

Application of Adaptive Matching and Tracking Algorithm in Gear Fault Feature Extraction

为了提高齿轮故障信号特征提取时处理稀疏性差信号的能力,设计了一种应用稀疏度自适应匹配追踪(SAMP)算法对齿轮故障信号进行处理方法。通过对仿真的齿轮故障信号数学模型和齿轮出现点蚀时的实验数据分析表明:经过SAMP处理后齿轮啮合频率以及半频和转频更加明显,边频带也更加突出,干扰成分降到最低。证明SAMP算法能够提取主要齿轮故障特征信息,有效降低噪声影响。相对于OMP算法处理齿轮故障信息,进过SAMP算法处理的故障信号,故障特征更加明显,且重构信号精度更高,表明SAMP算法重构故障信号相对于正交匹配追踪算法能够更好的提取主要齿轮故障特征。

In order to improve the ability of processing sparsity difference signals during feature extraction of gear fault signals,a Sparsity Adaptive Matching and Tracking(SAMP)Algorithm is designed to process gear fault signals.Through the simulation of the mathematical model of the gear fault signal and the analysis of the experimental data of the pitting corrosion of the gear,the results show that:after the treatment of SAMP,the gear meshing frequency,half frequency and rotation frequency are more obvious,the side frequency band is more prominent,and the interference component is reduced to the minimum.It is proved that the SAMP algorithm can extract the main gear fault characteristic information and effectively reduce the influence of noise.Compared with OMP algorithm for processing gear fault information,the fault features of the fault signals processed by SAMP algorithm are more obvious,and the reconstruction signal accuracy is higher,indicating that the reconstruction fault signals of SAMP algorithm can better extract the main gear fault features compared with the orthogonal matching tracking algorithm.