基于有效奇异值数量规律的滑动轴承转子轴心轨迹提纯研究  

Purification of the axis trajectory of a rotor supported by sliding bearings based on the number rule of effective singular values

作　　者：郭明军 李伟光[1] 杨期江 赵学智[1] GUO Mingjun;LI Weiguang;YANG Qijiang;ZHAO Xuezhi(School of Mechanical&Automotive Engineering,South China University of Technology,Guangzhou 510640,China;School of Marine Engineering,Guangzhou Maritime College,Guangzhou 510725,China)

机构地区：[1]华南理工大学机械与汽车工程学院,广州510640 [2]广州航海学院轮机工程学院,广州510725

出　　处：《振动与冲击》2019年第22期155-161,共7页Journal of Vibration and Shock

基　　金：国家自然基金(51875205;51875216);广东省自然科学基金(2018A030310017);广州市科技计划(201904010133);广东省教育厅项目(2017KQNCX145)

摘　　要：针对实测的主轴位移信号存在噪声污染的问题,提出基于有效奇异值数量规律的轴心轨迹提纯方法,并将其应用于大型滑动轴承试验台转子的轴心轨迹提纯,效果优于传统算法(EMD、谐波小波)及其改进算法(基于EMD的改进谐波小波算法),成功识别了转子的不对中故障。此外发现,当奇异值差分谱的首个峰值和紧随其后的差分谱幅值很接近(≥97.16%)时,根据差分谱方法提纯的轴心轨迹会发生畸变,并从能量损失的角度对这种现象进行了分析,分析指出,在这种情况下应该以第2个差分谱峰值对应的分量个数重构信号才能提纯到正确的轴心轨迹,从而进一步完善了差分谱理论。研究结果表明,利用改进的差分谱法的轴心轨迹提纯效果与该研究提出的有效值法的效果相当。Aiming at the noise pollution of the measured displacement signals of a rotor,a method for axis trajectory purification was proposed based on the number rule of singular values.The effect of the method proposed was compared with that of the improved harmonic wavelet algorithm based on EMD.In addition,when the first peak of the difference spectrum of singular values be close to the next one,the signal should be purified using the components corresponding to the second peak.The results by the proposed method is equivalent to the one based on the theory of difference spectrum of singular values.Although the improved harmonic wavelet algorithm based on EMD can successfully identify the misalignment fault of the rotor,its effect is worse than that of the effective singular value method.

关 键 词：奇异值分解 有效奇异值 经验模式分解(EMD) 谐波小波 轴心轨迹 

分 类 号：TH113.1[机械工程—机械设计及理论] TH165.3