Rolling bearing fault diagnosis method based on FSC-MPE and BP neural network
-
摘要:
目的 提出一种从强背景噪声、非平稳、非线性的复杂设备滚动轴承早期冲击故障振动信号中有效提取故障特征并进行故障模式识别的方法。 方法 首先，利用快速谱相关（FSC）分析提取原始振动信号的故障特征，并利用多尺度排列熵（MPE）对故障特征进行量化；然后，将故障特征数据输入BP神经网络进行故障诊断模型训练与测试；最后，对变速情况下的滚动轴承故障模拟实验数据和美国凯斯西储大学公开的轴承故障试验数据集进行故障识别研究。 结果 结果显示：所提方法对不同类型的故障具有较高的辨识精度，可达97%以上。 结论 研究验证了基于FSC-MPE与BP神经网络的滚动轴承故障诊断方法的可行性和优越性，可为滚动轴承健康状态评估提供技术支持。 Abstract:Objectives This paper proposes a method for effectively extracting fault features and identifying fault patterns from the early impact vibration signals of the rolling bearings of complex equipment which is non-stationary, nonlinear and has strong background noise. Methods First, the fault features of the original vibration signals are extracted via fast spectral correlation analysis and quantified via multi-scale permutation entropy (FSC-MPE). The fault feature data is then input into a BP neural network for fault diagnosis model training and testing. Finally, fault identification research is carried out on the rolling bearing fault simulation experimental data under variable speed and the public bearing fault test dataset of Case Western Reserve University. Results The results show that the proposed method has high identification accuracy for different types of faults, reaching more than 97%. Conclusions The feasibility and superiority of the proposed rolling bearing fault diagnosis method based on FSC-MPE and BP neural network are verified, and it can provide technical support for rolling bearing health evaluation. -
图 7 离散小波变换和EMD得到的故障特征
Figure 7. Fault characteristics obtained by discrete wavelet transform and EMD
图 9 试验轴承振动信号增强包络谱
Figure 9. Enhanced envelope spectrum of vibration signal of test bearing
图 10 神经网络交叉熵损失与测试集识别准确率
Figure 10. Accuracy of neural network cross entropy loss and test set recognition
表 1 轴承数据样本统计
Table 1. Sample statistics of bearing data
故障类型 损伤尺寸/in 样本数量/个 标签 正常情况 150 0 内圈故障 0.07 150 1 0.14 150 2 0.21 150 3 外圈故障 0.07 150 4 0.14 150 6 0.21 150 7 总计 1 050 
下载: 导出CSV
表 3 轴承故障诊断结果
Table 3. Diagnosis results of bearing fault
故障类型 损伤尺寸/in 正确数/总数 准确率/% 正常情况 25/25 100 内圈故障 0.07 24/25 96 0.14 24/25 96 0.21 25/25 100 外圈故障 0.07 24/25 96 0.14 24/25 96 0.21 25/25 100 总计 171/175 97.71
下载: 导出CSV
-
[1] 杨大春. 基于集合经验模态分解的滚动轴承振动信号希尔伯特谱分析方法[J]. 机械制造, 2019, 57(8): 29–32, 79. doi: 10.3969/j.issn.1000-4998.2019.08.009YANG D C. Hilbert spectrum analysis method of rolling bearing vibration signal based on set empirical mode decomposition[J]. Machinery, 2019, 57(8): 29–32, 79 (in Chinese). doi: 10.3969/j.issn.1000-4998.2019.08.009 [2] 彭畅. 旋转机械轴承振动信号分析方法研究[D]. 重庆: 重庆大学, 2014.PENG C. Vibration signal analysis of bearings in the rotating mechanical[D]. Chongqing: Chongqing University, 2014 (in Chinese). [3] 何洋洋, 王馨怡, 董晶. 基于经验小波变换与谱峭度的船舶轴系故障特征提取方法[J]. 中国舰船研究, 2020, 15(增刊1): 98–106. doi: 10.19693/j.issn.1673-3185.01771HE Y Y, WANG X Y, DONG J. Fault feature extraction method for marine shafting based on empirical wavelet transform-spectral kurtosis[J]. Chinese Journal of Ship Research, 2020, 15(Supp1): 98–106 (in Chinese). doi: 10.19693/j.issn.1673-3185.01771 [4] CHEN Y L, ZHANG P L. Bearing fault detection based on SVD and EMD[J]. Applied Mechanics and Materials, 2012, 184/185: 70–74. doi: 10.4028/www.scientific.net/AMM.184-185.70 [5] 张仲良, 朱晓军, 彭飞, 等. 基于HHT的船体结构应力监测数据特征分析和去噪方法[J]. 中国舰船研究, 2019, 14(增刊1): 158–164. doi: 10.19693/j.issn.1673-3185.01509ZHANG Z L, ZHU X J, PENG F, et al. Characteristic analysis and de-noising method of stress monitoring data of hull structures based on HHT[J]. Chinese Journal of Ship Research, 2019, 14(Supp1): 158–164 (in Chinese). doi: 10.19693/j.issn.1673-3185.01509 [6] 张永祥, 朱杰平, 张帅. 基于EEMD的滚动轴承故障诊断方法[J]. 海军工程大学学报, 2014(6): 90–94.ZHANG Y X, ZHU J P, ZHANG S. Rolling element bearing feature extraction based on EEMD[J]. Journal of Naval University of Engineering, 2014(6): 90–94 (in Chinese). [7] 唐贵基, 王晓龙. 变分模态分解方法及其在滚动轴承早期故障诊断中的应用[J]. 振动工程学报, 2016, 29(4): 638–648.TANG G J, WANG X L. Variational mode decomposition method and its application on incipient fault diagnosis of rolling bearing[J]. Journal of Vibration Engineering, 2016, 29(4): 638–648 (in Chinese). [8] 王海龙, 夏筱筠, 孙维堂. 基于EMD与卷积神经网络的滚动轴承故障诊断[J]. 组合机床与自动化加工技术, 2019(10): 46–48, 52.WANG H L, XIA X J, SUN W T. Rolling bearing fault diagnosis based on EMD and convolutional neural network[J]. Modular Machine Tool and Automatic Manufacturing Technique, 2019(10): 46–48, 52 (in Chinese). [9] 何江江, 李孝全, 赵玉伟, 等. 基于改进EEMD的卷积神经网络滚动轴承故障诊断[J]. 重庆大学学报, 2020, 43(1): 82–89.HE J J, LI X Q, ZHAO Y W, et al. Fault diagnosis of rolling bearing based on improved EEMD and convolutional neural network[J]. Journal of Chongqing University, 2020, 43(1): 82–89 (in Chinese). [10] 吕阳, 廖与禾, 王报祥, 等. 基于VMD和CNN的滚动轴承故障定量诊断方法[J]. 中国科技论文, 2020, 15(7): 735–742. doi: 10.3969/j.issn.2095-2783.2020.07.020LÜ Y, LIAO Y H, WANG B X, et al. Quantitative diagnosis method for rolling bearing faults based on VMD and CNN[J]. China Sciencepaper, 2020, 15(7): 735–742 (in Chinese). doi: 10.3969/j.issn.2095-2783.2020.07.020 [11] 方立华. 基于振动特征的风电机组传动系统机械故障诊断研究[D]. 吉林: 东北电力大学, 2018.FANG L H. Research on mechanical fault diagnosis of wind turbine transmission system based on vibration feature[D]. Jilin: Northeast Electric Power University, 2018 (in Chinese). [12] GARDNER W A, BROWN W, CHEN C K. Spectral correlation of modulated signals: Part II-digital modulation[J]. Communications IEEE Transactions on, 1987, 35(6): 595–601. doi: 10.1109/TCOM.1987.1096816 [13] ANTONI J, XIN G, HAMZAOUI N. Fast computation of the spectral correlation[J]. Mechanical Systems & Signal Processing, 2017, 92: 248–277. [14] 陈东宁, 张运东, 姚成玉, 等. 基于变分模态分解和多尺度排列熵的故障诊断[J]. 计算机集成制造系统, 2017, 23(12): 2604–2612.CHEN D N, ZHANG Y D, YAO C Y, et al. Fault diagnosis method based on variational mode decomposition and multi-scale permutation entropy[J]. Computer Integrated Manufacturing Systems, 2017, 23(12): 2604–2612 (in Chinese). [15] 薛红新. 基于机器学习方法的分类与预测问题研究[D]. 太原: 中北大学, 2019.XUE H X. Research on the problems of classification and prediction based on machine learning methods[D]. Taiyuan: North University of China, 2019 (in Chinese). [16] 陈伟南, 黄连忠, 张勇, 等. 基于BP神经网络的船舶主机能效状态评估[J]. 中国舰船研究, 2018, 13(4): 127–133, 160.CHEN W N, HUANG L Z, ZHANG Y, et al. Evaluation of main engine energy efficiency based on BP neural network[J]. Chinese Journal of Ship Research, 2018, 13(4): 127–133, 160 (in Chinese). [17] AZIZ W, ARIF M. Multiscale permutation entropy of physiological time series[C]//International Multitopic Conference. IEEE, 2005. [18] 杨武俊. 基于神经网络的人脸图像识别算法[J]. 网络安全技术与应用, 2016(1): 98,100.YANG W J. Face image recognition algorithm based on neural network[J]. Network Security Technology and Application, 2016(1): 98,100 (in Chinese). [19] 王靖, 陈特放, 黄采伦. 基于等角度采样的列车频带变化类故障诊断方法研究[J]. 中国机械工程, 2012, 23(16): 1957–1961. doi: 10.3969/j.issn.1004-132X.2012.16.015WANG J, CHEN T F, HUANG C L. Research on diagnosis method to train's vibration frequency band—varying fault based on uniform angle sampling[J]. China Mechanical Engineering, 2012, 23(16): 1957–1961 (in Chinese). doi: 10.3969/j.issn.1004-132X.2012.16.015 [20] URBANEK J, BARSZCZ T, ANTONI J. A two-step procedure for estimation of instantaneous rotational speed with large fluctuations[J]. Mechanical Systems and Signal Processing, 2013, 38(1): 96–102. doi: 10.1016/j.ymssp.2012.05.009 -
点击查看大图
图(11) / 表(3)
计量
- 文章访问数: 293
- HTML全文浏览量: 105
- PDF下载量: 14
- 被引次数: 0
