基于异构输入和改进集成学习的滚动轴承故障诊断

PDF(2396 KB)

振动与冲击 ›› 2024, Vol. 43 ›› Issue (19) : 174-182.

论文

基于异构输入和改进集成学习的滚动轴承故障诊断

赵柄锡1，虞磊1，陈景阳1，马梁2，王俊3

作者信息 +

Fault diagnosis of rolling bearings based on heterogeneous inputs and improved ensemble learning

ZHAO Bingxi1, YU Lei1, CHEN Jingyang2, MA Liang2, WANG Jun3

Author information +

文章历史 +

摘要

针对传统滚动轴承故障诊断无法利用多维故障信息以及仅关注总体正确率而较少考虑类别间诊断均衡性的问题，构建了一种具有异构输入的滚动轴承集成学习故障诊断方法。为此，首先发展了基于多域异构输入和特征融合的复合结构网络模型（H-NET），相比单网络结构模型增加了输入的多样性和特征稳健性；其次，针对传统Adaboost集成学习未考虑类别间分类均衡性的问题，引入类别权重提出了H-NET训练策略，将其与Adaboost算法融合建立了改进的集成学习方法。验证结果表明：采用H-NET相对单网络结构模型可以提高故障诊断的正确率和均衡性；同时，基于改进后集成学习构建的故障诊断模型在保证总体正确率同时提升了强分类模型的稳健性。最后与文献中方法进行对比，进一步表明了本文方法的优越性。

Abstract

A rolling bearing fault diagnosis method with heterogeneous inputs was constructed to address the issues of traditional rolling bearing fault diagnosis that cannot utilize multidimensional fault information and only focuses on overall accuracy without considering the balance of diagnosis between categories. To this end, a hybrid network model (H-NET) based on multi-domain heterogeneous inputs and feature fusion was first developed, which increases input diversity and feature robustness compared to single structured models; Secondly, in response to the problem of traditional Adaboost ensemble learning not considering the balance of classification between categories, category weights are introduced and an H-NET training strategy is further proposed, which is integrated with the Adaboost algorithm to establish an improved ensemble learning method. The validation results indicate that using H-NET relative to single structure models can improve the accuracy and balance of fault diagnosis; Meanwhile, the fault diagnosis model constructed based on improved ensemble learning ensures overall accuracy while enhancing the stability of the strong classification model. Finally, a comparison with the methods in the literature further demonstrates the superiority of the proposed method.

导出引用

赵柄锡1, 虞磊1, 陈景阳1, 马梁2, 王俊3. 基于异构输入和改进集成学习的滚动轴承故障诊断[J]. 振动与冲击, 2024, 43(19): 174-182

ZHAO Bingxi1, YU Lei1, CHEN Jingyang2, MA Liang2, WANG Jun3. Fault diagnosis of rolling bearings based on heterogeneous inputs and improved ensemble learning[J]. Journal of Vibration and Shock, 2024, 43(19): 174-182

参考文献

[1] Attoui I, Fergani N, Boutasseta N, et al. A new time-frequency method for identification and classification of ball bearing faults[J]. Journal of Sound and Vibration, 2017, 397:241-265.
[2] 李军宁, 罗文广, 陈武阁. 面向振动信号的滚动轴承故障诊断算法综述[J]. 西安工业大学学报, 2022,42(02):105-122.
LI Junning, LUO Wenguang, CHEN Wuge. Overview of Algorithms for Rolling Bearing Fault Diagnosis Based on Vibration Signal[J]. Journal of Xi'an Technological University. 2022,42(02):105-122.
[3] 陈鹏. 基于振动信号的滚动轴承故障诊断方法综述[J]. 轴承, 2022(000-006).
Chen Peng. Review on Fault Diagnosis Methods for Rolling Bearings Based on Vibration Signals[J]. Bearing. 2022(000-006).
[4] 郭艳平, 解武波, 龙涛元,等. 拟合振动信号模型实现滚动轴承故障诊断[J]. 兰州交通大学学报, 2017, 36(3):5.
Guo Yanping, Xie Wubo, Long Taoyuan. Fault Diagnosis of Rolling Bearing by Vibration Signal Model fitting[J]. Journal of Lanzhou Jiaotong University. 2017, 36(3):5.
[5] Dolenc B , Boskoski P , Juricic D . Distributed bearing fault diagnosis based on vibration analysis[J]. Mechanical systems and signal processing, 2016, 66-67(JAN.):521-532.
[6] Prieto MD, Cirrincione G, Espinosa AG, et al. Bearing Fault Detection by a Novel Condition-Monitoring Scheme Based on Statistical-Time Features and Neural Networks[J]. IEEE Transactions on Industrial Electronics, 2013, 60(8): 3398-3407.
[7] 戴邵武, 陈强强, 丁宇. 基于时域特征的滚动轴承寿命预测[J]. 计算机测量与控制, 2019, 027(010):60-63.
Dai ShaoWu, Chen Qiangqiang, Ding Yu. Prediction for Rolling Bearing Remaining Life Based on Time Domain Feature[J]. Computer Measurement & Control. 2019, 027(010):60-63.
[8] 王平, 廖明夫. 滚动轴承故障特征信息的自动提取方法研究[J]. 机械强度, 2003, 25(6):604-608.
Wang Ping, Liao Mingfu. Auto Extraction of fault features in rolling element bearing fault diagnosis[J]. Journal of Mechanical Strength. 2003, 25(6):604-608.
[9] 韩业锋, 仲涛, 石磊. 基于包络谱分析的滚动轴承故障诊断分析[J]. 机械研究与应用, 2010, 23(004):118-119.
Han Yefeng, Zhong Tao, Shi Lei. Rolling element bearing fault diagnosis based on envelope analysis[J]. Mechanical Research & Application. 2003, 25(6):604-608.
[10] Yang BS, Widodo A. Introduction of intelligent machine fault diagnosis and prognosis[M]. Nova Science Publishers, 2009.
[11] Lerch A. An introduction to Audio Content Analysis: Application in signal processing and music informatics[M]. Wiley-IEEE Press, 2012.
[12] Dawei Z, Wei G, Da H. An Intelligent Fault Diagnosis Method based on STFT and Convolutional Neural Network for Bearings Under Variable Working Conditions[C]. 2019 Prognostics and System Health Management Conference (PHM-Qingdao), Qingdao, China, 2019.
[13] Shuang Z, Maohua X, Petr B, et al. Remaining Useful Life Prediction and Fault Diagnosis of Rolling Bearings Based on Short-Time Fourier Transform and Convolutional Neural Network[J]. Shock and Vibration, 2020, 2020(8857307):1-14.
[14] 朱军. 滚动轴承非平稳信号故障诊断若干方法研究[D]. 中国科学技术大学, 2016.
Zhu Jun. Research on Several Methods for Fault Diagnosis of Non stationary Signals in Rolling Bearings[D]. University of Science and Technology of China, 2016.
[15] 陈刚, 廖明夫. 基于小波分析的滚动轴承故障诊断研究[J]. 科学技术与工程, 2007, 7(12):2810-2814.
Chen Gang, Liao Mingfu. Fault Diagnosis for a Rolling Element Bearing Based on Wavelet Analysis[J]. Science Technology and Engineering,2007, 7(12):2810-2814.
[16] Zhang J, Yan R, Gao RX, et al. Performance enhancement of ensemble empirical mode decomposition[J]. Mechanical Systems and Signal Processing, 2010, 24(7):2104-2123.
[17] Shen F , Chen C , Yan R , et al. Bearing fault diagnosis based on SVD feature extraction and transfer learning classification[C]. Prognostics and System Health Management Conference. IEEE, 2016.
[18] Lu W, Wang X, Yang C, et al. A novel feature extraction method using deep neural network for rolling bearing fault diagnosis[C]. The 27th Chinese Control and Decision Conference (2015 CCDC). IEEE, 2015.
[19] Zhao L, Wang X. A Deep Feature Optimization Fusion Method for Extracting Bearing Degradation Features[J]. IEEE Access, 2018:19640-19653.
[20] Xu G, Liu M, Jiang Z, et al. Bearing Fault Diagnosis Method Based on Deep Convolutional Neural Network and Random Forest Ensemble Learning[J]. Sensors, 2019, 19(5).
[21] 陈法法, 杨晶晶, 肖文荣等. Adaboost-SVM集成模型的滚动轴承早期故障诊断[J]. 机械科学与技术. 2018, 037(002): 237-243.
Chen Fafa, Yang Jingjing, Xiao Wenrong. Adaboost-SVM Integrated Model for Early Fault Diagnosis of Rolling Bearings[J]. Mechanical Science and Technology. 2018, 037(002): 237-243.
[22] Li H, Zheng H. Bearing Fault Detection Using Envelope Spectrum Based on EMD and TKEO[C]// International Conference on Fuzzy Systems and Knowledge Discovery. IEEE, 2008:142-146.
[23] Wang X, Li Y, Rui T, et al. Bearing fault diagnosis method based on Hilbert envelope spectrum and deep belief network[J]. Journal of Vibroengineering, 2015, 17(3):1295-1308.
[24] Gong XY, Han J, Chen H, et al. A Bearing Fault Diagnosis Using Wavelet Envelope Spectrum Based on Full Vector Spectrum Technology[J]. Applied Mechanics and Materials, 2012, 190-191:873-879.
[25] Wade AS, Robert, BR. Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark study. Mechanical Systems and Signal Processing. 2015, 64–65:100–131.
[26] Dongyang D, Jian J, Yuling W, et al. A rule-based classifier ensemble for fault diagnosis of rotating machinery[J]. Journal of Mechanical Science and Technology, 2018, 32(6):2509-2515.
[27] Gangavva, Mangai J A, Mohit B. An Investigation of Ensemble Learning Algorithms for Fault Diagnosis of Roller Bearing. Computer Science. 2022.