Gearbox Fault Diagnosis Based on Improved Time Series Gray Scale Image and Deep Learning

doi:10.3778/j.issn.1002-8331.2304-0143

Abstract

Abstract: A gear box fault diagnosis method based on improved time series gray image and deep learning is put forward in order to solve the problems such as complex actual working environment of gear box, inadequate performance of extracting features by traditional methods and gray image extraction features. After the vibration signal is decomposed into several intrinsic mode components (IMFs) by EEMD, the IMFs are divided into high-frequency and low-frequency components by cumulative mean criterion, in which the high-frequency components are denoised by wavelet threshold. The high frequency IMFs and low frequency IMFs after noise reduction are reconstructed, and the reconstructed signal is coded by gray image method. Two-dimensional improved time series gray image is sent to convolution neural network for training, so as to exert the advantages of convolution network in feature extraction of pictures and display the results by confusion matrix. Finally, the model results and different gray scale images are compared with traditional diagnosis methods. The results show that compared with common gray image and global denoising gray image, the method proposed in this paper improves the accuracy of gear box fault diagnosis by 4 and 1.8 percentage points respectively, and the convergence speed is significantly faster. Compared with BP neural network and ELM diagnosis method, the method proposed in this paper significantly improves the accuracy of gear box fault diagnosis.

Key words: ensemble empirical mode decomposition, fault diagnosis, improved time series gray scale image, deep learning

摘要： 针对齿轮箱实际工作环境复杂、传统方法提取特征以及灰度图提取特征性能不足，提出了一种基于改进时序灰度图和深度学习的齿轮箱故障诊断方法。用EEMD（ensemble empirical mode decomposition）将振动信号分解为若干个本征模态分量（IMF）后，通过累计均值准则将IMFs划分为高频和低频分量，其中高频分量采用小波阈值降噪进行处理；重构降噪后的高频IMFs与低频IMFs，并利用灰度图方法对重构信号进行编码。将二维改进时序灰度图送入卷积神经网络进行训练，以发挥卷积网络对图片特征提取优势，并由混淆矩阵显示结果。最后将模型结果和不同灰度图与传统诊断方法进行对比。结果表明：相对于普通灰度图、全局去噪灰度图，所提方法对齿轮箱故障诊断准确率分别提高4、1.8个百分点，且收敛速度明显加快；相对于BP神经网络以及ELM诊断方法，所提方法对齿轮箱故障诊断准确率显著提高。

关键词: 集合经验模态分解, 故障诊断, 改进时序灰度图, 深度学习

XIE Fengyun, LI Gang, WANG Linglan, LIU Hui, WANG Gan. Gearbox Fault Diagnosis Based on Improved Time Series Gray Scale Image and Deep Learning[J]. Computer Engineering and Applications, 2024, 60(13): 338-344.

谢锋云, 李刚, 王玲岚, 刘慧, 汪淦. 改进时序灰度图和深度学习的齿轮箱故障诊断[J]. 计算机工程与应用, 2024, 60(13): 338-344.

References

[1] 周浩轩, 温广瑞, 黄鑫, 等. 多尺度复合稀疏的齿轮箱复合故障诊断研究[J]. 振动. 测试与诊断, 2023, 43(2): 215-222.
ZHOU H X, WEN G R, HUANG X, et al. Fault diagnosis of gearbox compound fault based on multi-scale compound regularized convolutional sparse coding[J]. Journal of Vibration, Measurement & Diagnosis, 2023, 43(2): 215-222.
[2] 谢锋云, 董建坤, 王二化, 等. 基于双隐含层RWPSO-BP神经网络的齿轮箱故障诊断研究[J]. 现代制造工程, 2021(6): 155-160.
XIE F Y, DONG J K, WANG E H, et al. Research on gearbox fault diagnosis based on double hidden layer RWPSO-BP neural network[J]. Modern Manufacturing Engineering, 2021(6): 155-160.
[3] 王卫国, 孙磊. 基于EEMD-CWD的齿轮箱振动信号故障特征提取[J]. 兵工学报, 2014, 35(8): 1288-1294.
WANG W G, SUN L. Gearbox vibration signal fault feature extraction based on ensemble empirical mode decomposition and Choi-Williams distribution[J]. Journal of Military Engineering, 2014, 35(8): 1288-1294.
[4] 胡爱军, 马万里, 唐贵基. 基于集成经验模态分解和峭度准则的滚动轴承故障特征提取方法[J]. 中国电机工程报, 2012, 32(11): 106-111.
HU A J, MA W L, TANG G J. Rolling bearing fault feature extraction method based on ensemble empirical mode decomposition and kurtosis criterion [J]. China Journal of Electrical Engineering, 2012, 32(11): 106-111.
[5] CHAO L, FENG Z, YAN L. GPS/Pseudolites technology based on EMD-wavelet in the complex field conditions of mine[J]. Procedia Earth and Planetary Science, 2009(1): 1293-1300.
[6] 胡春生, 李国利, 赵勇, 等. 变工况滚动轴承故障诊断方法综述[J]. 计算机工程与应用, 2022, 58(18): 26-42.
HU C S, LI G L, ZHAO Y, et al. Summary of fault diagnosis methods for rolling bearings under variable working conditions[J]. Computer Engineering and Applications, 2022, 58(18): 26-42.
[7] WEN L, LI X, GAO L, et al. A new convolutional neural network-based data-driven fault diagnosis method[J]. IEEE Transactions on Industrial Electronics, 2018, 65(7): 5990-5998.
[8] 胡茑庆, 陈徽鹏, 程哲, 等. 基于经验模态分解和深度卷积神经网络的行星齿轮箱故障诊断方法[J]. 机械工程学报, 2019, 55(7): 9-18.
HU M Q, CHEN H P, CHENG Z, et al Fault diagnosis for planetary gearbox based on EMD and deep convolutional neural networks[J]. Journal of Mechanical Engineering, 2019, 55(7): 9-18.
[9] 蒲悦逸, 王文涵, 朱强, 等. 基于CNN-ResNet-LSTM模型的城市短时交通流量预测算法[J]. 北京邮电大学学报, 2020, 43(5): 9-14.
PU Y Y, WANG W H, ZHU Q, et al. Urban short-term traffic flow prediction algorithm based on CNN-ResNet-LSTM mode[J]. Journal of Beijing University of Posts and Telecommunications, 2020, 43(5): 9-14.
[10] 朱渔, 李丹, 李晓明, 等. 基于EEMD和BLSTM算法的齿轮泵行星轮典型故障诊断[J]. 机械设计与研究, 2022, 38(4): 198-201.
ZHU Y, LI D, LI X M, et al. Typical fault diagnosis of gear pump planetary wheel Based on EEMD and BLSTM[J]. Mechanical Design and Research, 2022, 38(4): 198-201.
[11] LV D, WANG H, CHE C. Multiscale convolutional neural network and decision fusion for rolling bearing fault diagnosis[J]. Industrial Lubrication and Tribology, 2021.
[12] 宋世林, 张学军. 脑电信号多特征融合与卷积神经网络算法研究[J]. 计算机工程与应用, 2024, 60(8): 148-155.
SONG S L, ZHANG X J. Algorithm research based on multi-feature fusion of EEG signals with convolutional neural networks[J]. Computer Engineering and Applications, 2024, 60(8): 148-155.
[13] SHAO S, MCALEER S, YAN R, et al. Highly accurate machine fault diagnosis using deep transfer learning[J]. IEEE Transactions on Industrial Informatics, 2019, 15(4): 2446-2455.
[14] WANG H, XU J, SUN C, et al. Intelligent fault diagnosis for planetary gearbox using time-frequency representation and deep reinforcement learning[J]. IEEE/ASME Transactions on Mechatronics, 2022, 27(2): 985-998.
[15] 杨枫. 基于深度学习的行星齿轮箱故障诊断研究[D]. 济南: 山东大学, 2020.
YANG F. Research on fault diagnosis of planetary gearboxes based on deep learning[D]. Jinan: Shandong University, 2020.