基于最大相关峭度反褶积的齿轮箱复合故障特征提取-噪声与振动控制.PDF

第37 卷 第3 期 噪 声 与 振 动 控 制 Vol 37 No.3 2017 年6 月 NOISE AND VIBRATION CONTROL Jun. 2017 文章编号：1006-1355(2017)03-0173-04 基于最大相关峭度反褶积的齿轮箱 复合故障特征提取 王志坚，寇彦飞，王俊元，张纪平，齐明思，赵志芳 ( 中北大学 机械与动力工程学院，太原 030051 ) 摘 要：提出一种基于最大相关峭度反褶积（Maximum correlated kurtosis deconvolution ，MCKD ）的复合故障特征提 取方法，通过MCKD 对原信号降噪，提取感兴趣的周期成分，同时将此方法与最小熵反褶积对比研究，验证该方法的强 降噪效果。将该方法运用于齿轮箱复合故障诊断中，可成功提取出各个故障特征。 关键词：振动与波；最大相关峭度反褶积；最小熵反褶积；复合故障；故障检测 中图分类号：TP 17；TP206 文献标识码：A DOI 编码：10.3969/j.issn. 1006-1355.2017.03.034 A Feature Extraction Method for Gearboxes with Compound Faults Based on MCKD WANG Zhi-jian , KOU Yan-fei , WANG Jun-yuan , ZHANG Ji-ping , QI Ming-si , ZHAO Zhi-fang ( College of Mechanical and Power Engineering, North University of China, Taiyuan 030051, China ) Abstract : A feature extraction method for gearboxes with compound faults is proposed. The maximum correlated kurtosis deconvolution (MCKD) method is used to reduce the noise of the original signal and extract the related periodic components. Comparing the result of this method with that of the minimum entropy deconvolution (MED) method, the superiority of the MCKD method on noise reduction is proved. Applying the proposed method to the compound fault diagnosis of the gearboxes, the fault features can be extracted successfully. Key words : vibration and wave; maximum correlated kurtosis deconvolution (MCKD); minimum entropy deconvolution (MED); multi-fault; fault detection 齿轮和轴承作为旋转零部件，其健康状况一直 量小而常被淹没，导致误诊断或漏诊断。因此当复 备受关注，当齿轮箱出现微故障时，为了评估设备的 合故障共