Page 115 - 《振动工程学报》2026年第5期
P. 115

第 39 卷第 5 期                       振 动 工 程 学 报                                       Vol. 39 No. 5
               2026 年  5 月                     Journal of Vibration Engineering                       May 2026



                               压   缩   感   知    匹   配   追   踪   波    达   时   间   差   提    取

                                           及    轴   承   损   伤   定    位   研   究



                                         陆 平 , 柳小勤 , 汤林江 , 伍 星                     1,2
                                                                       1
                                                            1
                                                 1
                         (1. 昆明理工大学机电工程学院,云南 昆明 650500; 2. 云南机电职业技术学院,云南 昆明 650203)

              摘要:本文针对故障声发射信号沿轴承座圈中顺时针和逆时针两条传递路径造成的信号二次到达特性,提出了一种基于压缩
              感知匹配追踪的波达时间差提取方法,进而实现单传感器轴承损伤定位。通过                           Lamb 波频散特性找到多种传播模式的交叉点
              频率,再根据交叉点频率对声发射信号进行带通滤波,从而确定波速。结合传感器拾取到两个声发射事件的稀疏特性,根据
              半高宽得到的原子构造字典,利用压缩感知的正交匹配追踪算法提取沿两个方向传播的声发射信号到达脉冲,并计算到达信
              号的时间差。根据波速和时间差,估计两条传播路径上的路程差。考虑到环形结构的对称性会导致信号传播方向的不确定
              性,通过调整传感器位置进行二次定位。根据两次定位结果的重叠情况,确定声发射源的唯一位置。通过断铅试验和对低速
              轴承故障信号定位的测试,验证了所提方法的有效性。
              关键词: 声发射;Lamb 波;损伤定位;单传感器定位;压缩感知
                             +
              中图分类号:TH165 .3;TH133.33        文献标志码:A        DOI:10.16385/j.cnki.issn.1004-4523.202405033


                         Compressed sensing matching pursuit of time difference extraction
                                          and bearing defect localization study

                                                           1
                                               1
                                                                         1
                                         LU Ping ,LIU Xiaoqin ,TANG Linjiang ,WU Xing 1,2
                 (1.Faculty of Mechanical and Electrical Engineering,Kunming University of Science and Technology,Kunming 650500,China;
                            2.Yunnan Vocational College of Mechanical and Electrical Technology,Kunming 650203,China)
              Abstract:This paper focuses on the twice arrival characteristics of two transmission paths, clockwise and counterclockwise caused by fault
              acoustic emission signals, in the bearing housing ring. A wave arrival time difference extraction method based on compressive sensing matching
              pursuit is proposed to achieve single-sensor bearing defect localization. Firstly, the intersection frequency of multiple propagation modes is
              determined  based  on  the  Lamb  wave  dispersion  property,  and  the  acoustic  emission  signal  is  band-pass  filtered  based  on  the  intersection
              frequency to determine the wave speed. Secondly, by exploiting the sparse nature of two acoustic emission events picked up by sensors and
              employing an atom constructed dictionary obtained from full width at half maximum (FWHM), the compressive sensing orthogonal matching
              pursuit  algorithm  is  utilized  to  extract  the  arrival  events  of  acoustic  emission  signals  propagating  in  two  directions  and  compute  the  time
              difference of arrival. Then, the difference in distances along the two propagation paths is estimated based on the wave speed and time difference.
              Finally, considering the uncertainty of the signal propagation direction due to the symmetry of the ring structure, secondary localization is
              performed by adjusting the sensor position. Based on the overlap of the two localization results, the unique location of the acoustic emission
              source is determined. The effectiveness of the proposed method is verified by the pencil lead breakage experiments and tests on the localization
              of low-speed bearing fault signals.
              Keywords:acoustic emission;Lamb wave;defect localization;single sensor localization;compressed sensing


                  低速回转轴承被广泛应用于各种重要的大型机                          承部件的状态监测,对减少停机时间和提升设备运
              械装备,例如盾构机的主轴承、吊机回转轴承、混合                           行的安全性具有普适价值 。
                                                                                       [2]
              机拖轮轴承和风机叶轮轴承等,它们通常在低速或                                轴承故障诊断方法主要分为两类 ,一种是提取
                                                                                                 [3]
              非连续的状态下运行。这些轴承一旦损坏可能导致                            故 障 特 征 频 率, 另 一 种 是 基 于 机 器 学 习 的 状 态 识
              较长的维修时间和巨大的经济损失 。因此,开展轴                           别。HAN    等  [4]  采用了  Teager 能量算子对声发射信号
                                             [1]


                  收稿日期:2024-05-17;修订日期:2024-07-24
                  基金项目:国家自然科学基金资助项目             (52265069);云南省科技计划重点项目       (202002AC080001)
   110   111   112   113   114   115   116   117   118   119   120