基于正交匹配追踪算法的齿轮箱轴承故障诊断及实验验证

为了提高传动振动过程中齿轮箱轴承内外圈故障诊断能力，采用正交匹配追踪(OMP)算法建立了故障诊断模型，并开展仿真分析及实验验证。研究结果表明：通过OMP算法对轴承外圈故障仿真加噪信号进行处理，能够看到信号呈周期性波动，通过频率及其倍频呈现逐步衰减，故障特征明显；经过OMP算法处理的轴承内圈故障仿真纯净信号呈周期性波动，能够看到滚动轴承的故障，轴承内圈通过频率和倍频以及边频带呈现逐步衰减，故障特征明显。为了进一步验证OMP算法处理齿轮故障的有效性，搭建封闭式功率流齿轮箱试验台，OMP重构故障信号谱图中啮合频率360 Hz峰值较低，边频带被完全掩盖，不存在大量的干扰成分。经OMP算法处理过的故障信号的谱图能很好地体现故障特征。

Fault Diagnosis and Experimental Verification of Gearbox Bearing Based on Orthogonal Matching Tracking Algorithm

In order to improve the fault diagnosis ability for gearbox bearing inner and outer rings fault during transmission vibration, the orthogonal matching tracking (OMP) algorithm was used to establish the fault diagnosis model, and simulation analysis and experimental verification were carried out.The results show that the OMP algorithm is used to deal with the fault simulation noise signal of bearing outer ring, the signal appears periodic fluctuation, and the frequency and its multiplier frequency attenuate gradually, so the fault features are obvious.The pure signal of bearing inner ring fault simulation processed by OMP algorithm shows periodic fluctuation, and the fault of rolling bearing can be seen. The through frequency, multiplier frequency and the side band of the bearing inner ring attenuate gradually, and the fault features are obvious.In order to further verify the effectiveness of OMP algorithm in dealing with gear faults, a closed power flow gear box test bed was built. In the spectrum of OMP reconstructed fault signal, the peak of meshing frequency 360 Hz was low, the side band was completely covered, and there was no large amount of interference components.The spectrum of fault signal processed by OMP algorithm can reflect the fault characteristics well.