Page 121 - 《振动工程学报》2026年第5期
P. 121
第 5 期 陆 平,等:压缩感知匹配追踪波达时间差提取及轴承损伤定位研究 1325
的信号复杂,信噪比较低。 sion analysis for wind turbine blade bearing fault detection
文献 [12] 采用了 3 个声发射传感器来进行低速 under time-varying low-speed and heavy blade load condi-
轴承试验台上的轴承损伤检测,在定位 120°损伤源 tions[J]. IEEE Transactions on Industry Applications,2021,
位置时,平均误差为 8.5°,略高于本文的 3.0°误差。 57(3):2791-2800.
[6] 张瑞,李可,宿磊,等. 深度稀疏最小二乘支持向量机故
本研究在相同条件下,仅使用 1 个声发射传感器进
障诊断方法研究 [J]. 振动工程学报,2019,32(6):1104-
行测量,误差水平仍能满足实际应用需求。
1113.
ZHANG Rui,LI Ke,SU Lei,et al. Fault diagnosis method
4 结 论 of deep sparse least squares support vector machine[J]. Jour-
nal of Vibration Engineering,2019,32(6):1104-1113.
本研究表明,通过轴承材料 Lamb 波频散曲线的 [7] 周陈林,董绍江,李玲,等. 滚动轴承多状态特征信息的
交叉点频率信息,可由带通滤波得到含有两次到达 改进型卷积神经网络故障诊断方法 [J]. 振动工程学报,
事件的信号。利用构造的感应字典通过 CS-OMP 方 2020,33(4):854-860.
法,能够准确匹配到二次到达事件,并表示在稀疏系 ZHOU Chenlin,DONG Shaojiang,LI Ling,et al. Method
数中,从而可计算出时间差。当两次到达信号有部 to improve convolutional neural network in rolling bearing
fault diagnosis with multi-state feature information[J]. Journal
分重叠时,得到的时间差仍较为准确。结合 Lamb 波
of Vibration Engineering,2020,33(4):854-860.
对应频率的波速,只使用 1 个传感器进行两次定位,
[8] NAUMANN J. Acoustic emission monitoring of wind turbine
即能确定低速轴承损伤源的位置。
bearings[D]. Sheffield:University of Sheffield,2016.
在试验验证阶段,断铅试验和故障试验的绝对
[9] JONES M R,ROGERS T J,MARTINEZ I,et al. Bayesian
误差分别为 1.7°和 3.0°。结果表明,该方法具有较高 localisation of acoustic emission sources for wind turbine bear-
的定位精度。所提方法既不依赖轴承的转速信息, ings[C]//Proceedings of Health Monitoring of Structural and
同时又减少了传感器的使用数量,为低速轴承的损 Biological Systems XV. SPIE,2021:78.
伤定位提供了一种不同的途径。 [10] MA Z P,ZHAO M,LUO M R,et al. An integrated moni-
未来研究应扩展到内圈损伤的识别,探索不同 toring scheme for wind turbine main bearing using acoustic
损伤位置对波形特征的影响,从而进一步研究径向 emission[J]. Signal Processing,2023,205:108867.
载荷应力对波速和试验结果的影响,以提高诊断精 [11] 柳小勤,汤林江,侯凯泽,等. 基于声发射的滚动轴承损
伤定位方法研究 [J]. 振动与冲击,2020,39(15):176-182.
度。同时,考虑滚动体至测点传递路径的干扰,对信
LIU Xiaoqin, TANG Linjiang, HOU Kaize, et al. Fault
号进行精细分析和处理,以减少噪声干扰,提高故障
localization for rolling bearing based on AE[J]. Journal of
定位准确性。
Vibration and Shock,2020,39(15):176-182.
[12] TANG L J,LIU X Q,WU X,et al. An acoustic emission
参考文献: event filtering method for low-speed bearing outer race defect
localization[J]. IEEE Sensors Journal, 2022, 22( 16) :
[1] VAN HECKE B,YOON J,HE D. Low speed bearing fault 16293-16305.
diagnosis using acoustic emission sensors[J]. Applied Acous- [13] CIAMPA F, MEO M. Acoustic emission localization in
tics,2016,105:35-44. complex dissipative anisotropic structures using a one-channel
[2] BEN ALI J,FNAIECH N,SAIDI L,et al. Application of reciprocal time reversal method[J]. The Journal of the Acousti-
empirical mode decomposition and artificial neural network cal Society of America,2011,130(1):168-175.
for automatic bearing fault diagnosis based on vibration [14] JIAO J P,HE C F,WU B,et al. Application of wavelet
signals[J]. Applied Acoustics,2015,89:16-27. transform on modal acoustic emission source location in thin
[3] HOU B C, WANG D, KONG J Z, et al. Understanding plates with one sensor[J]. International Journal of Pressure
importance of positive and negative signs of optimized weights Vessels and Piping,2004,81(5):427-431.
used in the sum of weighted normalized Fourier [15] EBRAHIMKHANLOU A, SALAMONE S. Acoustic emis-
spectrum/envelope spectrum for machine condition monitor- sion source localization in thin metallic plates:a single-sensor
ing[J]. Mechanical Systems and Signal Processing, 2022, approach based on multimodal edge reflections[J]. Ultrason-
174:109094. ics,2017,78:134-145.
[4] HAN T,LIU Q N,ZHANG L,et al. Fault feature extrac- [16] 缪祥垚,伍星,柳小勤,等. 基于声发射 Lamb 波频散特性
tion of low speed roller bearing based on Teager energy opera- 的轴承损伤单传感器定位 [J]. 振动与冲击,2023,42(6):
tor and CEEMD[J]. Measurement,2019,138:400-408. 196-201.
[5] LIU Z P,YANG B Y,WANG X F,et al. Acoustic emis- MIAO Xiangyao, WU Xing, LIU Xiaoqin, et al. Single-

