基于HOSVD方法的轴承故障多传感器诊断分析  

Multi-sensor Diagnosis Analysis of Bearing Fault Based on HOSVD Method

作　　者：戚德慧[1] 尹静文[1] Qi Dehui;Yin Jingwen(College of Intelligent Manufacturing,Xinxiang Vocational and Technical College,Xinxiang Henan 453006,China)

机构地区：[1]新乡职业技术学院智能制造学院,河南新乡453006

出　　处：《山西电子技术》2024年第6期45-47,共3页Shanxi Electronic Technology

摘　　要：不同故障同时产生时单一通道信号识别精度明显降低,为了进一步降低轴承多通道信号同时滤波干扰的影响,设计了一种基于截断高阶奇异值分解(HOSVD)的多轴承故障诊断方法。以HOSVD方法为基础,提出多通道故障降噪方法,对多通道实施滤波处理,在很大程度上加强其检测效率。仿真信号分析表明,从时域层面故障信号难以实现对故障特征信息的提取。利用本文方法进行降噪处理以后获得的信号,在降噪方面具有良好成效,最终结果验证其可以有效将脉冲的周期特征提取出来。When different faults occur at the same time,the recognition accuracy of single channel signal is obviously reduced.In order to further reduce the influence of simultaneous filtering interference of bearing multi-channel signal,a multi-bearing fault diagnosis method based on truncated high order singular value decomposition(HOSVD)is designed.Based on the HOSVD method,a multi-channel fault noise reduction method is proposed,and the multi-channel filter is implemented to enhance the detection efficiency to a great extent.Simulation signal analysis shows that it is difficult to extract fault feature information from time domain fault signal.The signal is obtained after noise reduction with the method presented in this paper.It has a good effect on noise reduction,and the final results show that it can effectively extract the periodic characteristics of the pulse.

关 键 词：轴承 多通道信号 故障诊断 截断高阶奇异值分解 降噪 

分 类 号：TH133[机械工程—机械制造及自动化]