基于声信号的滚动轴承故障诊断研究

摘要
图/表
参考文献(17)
相关文章 (15)

全文: PDF (2209 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要结合小波包短时能量散布熵、回溯搜索算法以及学习矢量神经网络，提出一种基于声信号的滚动轴承故障诊断新方法。首先利用小波包分解结合短时能量对声信号进行脉冲能量提取，突出与故障相关的时频子空间的能量分布，再通过计算各子空间短时能量序列的散布熵，构造特征矩阵。利用t-分布随机邻域嵌入方法对所获特征进行降维聚类，显示所提取的特征具有较好的聚类性能。然后采用回溯搜索算法优化学习矢量量化建立神经网络故障诊断模型，对轴承故障进行识别，并与多种诊断方法进行比较，实验结果表明，加入短时能量散布熵后，本模型提升了声信号的能量特性，优化了特征矩阵，诊断性能最佳。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	陈剑1
	2
	徐庭亮1
	黄志1
	孙太华1
	李雪原1
	季磊1
	杨惠杰1

关键词 ：轴承故障诊断, 声信号, 短时能量散布熵（STE-DE）, 学习矢量量化（LVQ）, 回溯搜索算法（BSA）

Abstract：Combined with the short-term energy dispersion entropy of wavelet packets（WPD-STE-DE）, backtracking search algorithm（BSA） and learning vector neural network（LVQ）, a rolling bearing fault diagnosis model based on acoustic signal is proposed. The wavelet packet decomposition（WPD）combined with STE is firstly used to extract the pulse energy of the acoustic signal, highlight the energy distribution of the time-frequency subspace associated with the fault, and then construct the characteristic matrix by calculating the STE-DE of each subspace. The feature extracted by t-SNE has better clustering performance. Then the BSA is used to optimize LVQ to establish a neural network fault diagnosis model, identify the bearing fault, and compare it with a variety of diagnostic methods, the experimental results show that after adding STE-DE, the model improves the energy characteristics of the acoustic signal, optimizes the feature matrix, and the diagnostic performance is the best.

Key words： Bearing fault diagnosis acoustic signal short-term energy dispersion entropy（STE-DE） Learning vector quantization（LVQ） Backtracking search algorithm（BSA）

收稿日期: 2022-06-20 出版日期: 2023-11-15

引用本文:

陈剑1,2，徐庭亮1，黄志1，孙太华1，李雪原1，季磊1，杨惠杰1. 基于声信号的滚动轴承故障诊断研究[J]. 振动与冲击, 2023, 42(21): 237-244.
CHEN Jian1,2, XU Tingliang1, HUANG Zhi1, SUN Taihua1, LI Xueyuan1, JI Lei1, YANG Huijie1. Fault diagnosis of rolling bearings based on acoustic signals. JOURNAL OF VIBRATION AND SHOCK, 2023, 42(21): 237-244.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2023/V42/I21/237

[1] Jiaoyi Hou, Hongyu Sun, Aoyu Xu, et al. Fault diagnosis of synchronous hydraulic motor based on acoustic signals[J]. Advances in Mechanical Engineering, 2020, 12(4).
[2] Chaoren Qin, Dongdong Wang, Zhi Xu, Gang Tang. Improved Empirical Wavelet Transform for Compound Weak Bearing Fault Diagnosis with Acoustic Signals[J]. Applied Sciences, 2020,10(2).
[3] Chao Wang, Fanrang Kong, Qingbo He, Fei Hu, Fang Liu. Doppler Effect removal based on instantaneous frequency estimation and time domain re-sampling for wayside acoustic defective bearing detector system[J]. Measurement, 2014, 50.
[4] Adam Glowacz. Fault diagnosis of single-phase induction motor based on acoustic signals[J]. Mechanical Systems and Signal Processing, 2019,117.
[5] Yao Jiachi, Liu Chao, Song Keyu, el at. Fault diagnosis of planetary gearbox based on acoustic signals[J]. Applied Acoustics, 2021,181.
[6] 张思阳，徐敏强，王日新，等. EMD与样本熵在往复压缩机气阀故障诊断中的应用[J]. 哈尔滨工程大学学报，2014(6):696-700.
Zhang Siyang, Xu Minqiang, Wang Rixin ,et al. Application of EMD and SampEn to the fault diagnosis of reciprocating compressor valves[J]. Journal of Harbin Engineering University, 2014(6):696-700.
[7] 孟宗，王亚超，胡猛. 基于改进LMD和IED-SampEn的齿轮故障特征提取方法[J]. 机械工程学报，2016,52(5):169-176.
Meng Zong, et al. Fault Feature Extraction Method of Gear Based on improved Local Mean Decomposition and instantaneous Energy Distribution-sample Entropy [J]. Journal of Mechanical Engineering, 2016,52(5):169-176.
[8] 胡爱军，朱丽佳，向玲. 基于ITD和排列熵的行星齿轮箱故障诊断方法[J]. 中国工程机械学报，2021,19(6):542-548.
Hu Aijun, Zhu Lijia, Xiang Ling. Fault diagnosis method of planetary gearbox based on ITD and entropy [J]. Chinese Journal of Construction Machinery, 201,19(6):542-548.
[9] 吴守军，冯辅周，吴春志等. 基于VMD-DE的坦克行星变速箱故障诊断方法研究[J]. 振动与冲击，2020,39(10):170-179.
Wu Shoujun, Feng Fuzhou, et al. Research on fault diagnosis method of tank planetary gearbox based on VMD-DE[J]. Journal of Vibration and Shock, 2020,39(10):170-179.
[10] 张雄，张逸轩，张明，万书亭，何玉灵，豆龙江. 基于小波包散布熵与Meanshift概率密度估计的轴承故障识别方法研究[J].湖南大学学报(自然科学版)，2021,48(08):133-140.
Zhang Xiong, Zhang Yixuan, et al. Research on Bearing Fault Identification Method Based on Wavelet Packet Dispersion Entropy and Meanshift Probability Density Estimation[J]. Journal of Hunan University (Natural Science), 2021,48(08):133-140.
[11] Ping Yao, Kang Zhou. Application of short time energy analysis in monitoring the stability of arc sound signal[J]. Measurement, 2017, 105.
[12] Xiaoan Yan, Minping Jia. A novel optimized SVM classification algorithm with multi-domain feature and its application to fault diagnosis of rolling bearing[J]. Neurocomputing, 2018,313.
[13] 唐秋华，刘保华，陈永奇，等. 结合遗传算法的LVQ神经网络在声学底质分类中的应用[J].地球物理学报,2007，50(01):313-319.
Tang Q H，Liu B H，Chen Y Q，et al．Application of LVQ neural network combined with the genetic algorithm in acoustic seafloor classification．Chinese Journal of Geophysics，2007，50（1）：313～319.
[14] 朱振杰，周梅.狼群优化LVQ神经网络的齿轮箱故障诊断应用研究（英文）[J].机床与液压,2020,48(12):125-130.
Zhen-jie ZHU，Mei ZHOU．Application of gearbox fault diagnosis by LVQ neural network optimized by wolves［J］．Machine Tool & Hydraulics，2020，48( 12) : 125 － 130．
[15] Wang Yu and Shang Pengjian. Complexity Analysis of Time Series Based on Generalized Fractional Order Refined Composite Multiscale Dispersion Entropy[J]. International Journal of Bifurcation and Chaos, 2020, 30(14).
[16] 杨国安，钟秉林，黄仁，等. 机械故障信号小波包分解的时域特征提取方法研究[J]. 振动与冲击,2001,20(2):25-28,31.
Yang Guoan, Zhong Binglin, Huang Ren, et al. Research on the extraction method of time-domain symptoms based on wavelet packet decomposition of mechanical fault signal[J] Journal of Vibration and Shock,2001,20(2):25-28,31.
[17] 王超，高扬，刘超，等. 基于回溯搜索优化算法求解带时间窗和同时送取货的车辆路径问题[J]. 计算机集成制造系统,2019,25(9):2237-2247.
Wang Chao, Gao Yang, et al. Vehicle routing problem with simultaneous delivery and pickup problem solving by backtracking search optimization algorithm[J]. Computer Integrated Manufacturing Systems，2019,25(9):2237-2247.