【摘要】 为了提高齿轮箱状态进行监测能力，通过麻雀搜索算法(SSO)优化模态分解(VMD)算法对其开展故障诊断控制分析。研究结果表明：选择的断齿信号作为测试对象，通过SSO-VMD算法实施降噪，采用SSO算法进行VMD参数寻优得[708,3],同时获得全局最优适应度为2.685 2,再对以上IMF分量实施重构，获得降噪故障信号。采用多源域特征提取方法达到了最大识别精度，为68.43%,相对尺寸、时域、频域等表现出了更优的提取性能。I-Imap算法达到了最大降维性能指标与最优故障识别性能，具备理想的多源域故障特征降维性能，对行星齿轮箱工况达到100%的高精度故障识别率，此模型满足有效性。该研究可以拓展到相关的自动控制领域，具有很好的推广应用价值。更多还原

【Abstract】 In order to improve the gear box condition monitoring ability, the fault diagnosis and control analysis are carried out by sparrow search optimized VMD algorithm. The selected broken tooth signal is taken as the test object.Noise reduction is implemented by SSO-VMD algorithm. The SSO algorithm is used to optimize the VMD parameters [708,3], and the global optimal fitness is obtained as 2.6852. Then the above IMF components are reconstructed to obtain the fault signal after noise reduction. Multi-domain feature extraction method is applied to achieve the maximum recognition accuracy of 68.43%, which shows better performance in relative size, time domain and frequency domain. The I-Imap algorithm is used to achieve the maximum dimension reduction performance index and the optimal fault recognition performance, and has the ideal multi-domain fault feature dimension reduction performance. For planetary gearbox working conditions, the high-precision fault recognition rate reaches 100%, and this model meets the validity. This research can be extended to the related field of automatic control.更多还原