基于拓扑感知和双视图分类器的旋转机械故障诊断方法

PDF(3138 KB)

振动与冲击 ›› 2025, Vol. 44 ›› Issue (1) : 151-162.

故障诊断分析

基于拓扑感知和双视图分类器的旋转机械故障诊断方法

陈子旭1,2，余文念*1,2，杜伟涛3，林正宇1,2

作者信息 +

Fault diagnosis method for rotating machinery based on topology perception and dual-view classifier

CHEN Zixu1,2, YU Wennian*1,2, DU Weitao3, LIN Zhengyu1,2

Author information +

文章历史 +

摘要

针对旋转机械不同工况下数据分布不同，以及故障数据稀缺使得样本类别不均衡，导致故障诊断模型性能退化这一问题，提出一种基于拓扑感知和双视图分类器的故障诊断方法。该方法以一个图卷积网络为诊断框架，提出的非参数化拓扑感知模块能自适应更新图数据拓扑结构，约束不同域数据获取近似的消息传递路径，通过图卷积网络有效提取域一致故障特征；利用二分类器和多分类器构建双视图分类器，并计算二元输出和多元输出的相似度对训练数据进行重加权，避免了类别不均衡下模型的有偏训练以及对少数类样本识别能力不强的现象。利用公开的西安交通大学齿轮故障数据集、MAFAULDA旋转机械故障数据集及自制的滑动轴承故障模拟数据进行实验。结果表明，提出的方法能有效提升类别不均衡下变工况故障诊断的性能。

Abstract

Aiming at the degraded performance of diagnostic models due to the uneven distribution of data under different working conditions and the imbalanced classes caused by the scarcity of fault data, a topology-aware and dual-view classifier based diagnostic method was proposed. The graph convolutional network (GCN) was taken as the framework. The non-parametric topology-aware module can adaptively update the graph topology, obtaining approximate message passing paths for cross-domain data, and extracting domain-invariant features through GCN. The dual-view classifier was constructed using binary and multiple classifiers, and the output similarities were calculated to reweight the training data, which avoids biased training with imbalanced data and poor recognition of the minority classes. Experiments were conducted using publicly available datasets (Xi'an Jiaotong University gear fault dataset, MAFAULDA machinery fault dataset) and a self-collected journal bearing fault dataset. The results show that the proposed method can improve the diagnostic performance under variable working conditions and imbalanced data.

导出引用

陈子旭1, 2, 余文念1, 2, 杜伟涛3, 林正宇1, 2. 基于拓扑感知和双视图分类器的旋转机械故障诊断方法[J]. 振动与冲击, 2025, 44(1): 151-162

CHEN Zixu1, 2, YU Wennian1, 2, DU Weitao3, LIN Zhengyu1, 2. Fault diagnosis method for rotating machinery based on topology perception and dual-view classifier[J]. Journal of Vibration and Shock, 2025, 44(1): 151-162

参考文献

[1] 王辉，徐佳文，严如强. 基于多尺度注意力深度强化学习网络的行星齿轮箱智能诊断方法[J]. 机械工程学报，2022，58(11): 133-142.
WANG Hui, XU Jiawen, YAN Ruqiang. Multi-Scale Attention Based Deep Reinforcement Learning for Intelligent Fault Diagnosis of Planetary Gearbox [J]. Journal of Mechanical Engineering, 2022, 58(11): 133-142.
[2] 王玉静，夏林，康守强，等. 基于多源域异构模型迁移的滚动轴承故障诊断方法[J]. 振动与冲击，2023，42(24): 257-266. DOI:10.13465/j.cnki.jvs.2023.24.030.
WANG Yujing, XIA Lin, KANG Shouqiang, et al. A fault diagnosis method of rolling bearings based on multi-source domain heterogeneous model transfer [J]. Journal of Vibration and Shock, 2023, 42(24): 257-266. DOI:10.13465/j.cnki.jvs.2023.24.030.
[3] Lei Y, Yang B, Jiang X, et al. Applications of machine learning to machine fault diagnosis: A review and roadmap[J]. Mechanical Systems and Signal Processing, 2020, 138: 106587.
[4] 李彦夫，韩特. 基于深度学习的工业装备PHM研究综述[J].振动.测试与诊断，2022，42(05): 835-847+1029. DOI:10.16450/j.cnki.issn.1004-6801.2022.05.001.
LI Yanfu, HAN Te. Deep Learning Based Industrial Equipment Prognostics and Health Management: a Review [J]. Journal of Vibration,Measurement & Diagnosis, 2022，42(05): 835-847+1029. DOI:10.16450/j.cnki.issn.1004-6801.2022.05.001.
[5] 孙洁娣，刘保，温江涛，等. 类别标签辅助改进稠密网络的变工况轴承故障诊断[J]. 振动与冲击，2022，41(17): 204-212. DOI:10.13465/j.cnki.jvs.2022.17.025.
SUN Jiedi, LIU Bao, WEN Jiangtao, et al. Fault diagnosis of variable condition bearing based on improved dense network aided by class labels [J]. Journal of Vibration and Shock, 2022, 41(17): 204-212. DOI:10.13465/j.cnki.jvs.2022.17.025.
[6] Chen Z, He G, Li J, et al. Domain adversarial transfer network for cross-domain fault diagnosis of rotary machinery[J]. IEEE Transactions on Instrumentation and Measurement, 2020, 69(11): 8702-8712.
[7] Chen Z, Yu W, Kong C, et al. A dual-view network for fault diagnosis in rotating machinery using unbalanced data[J]. Measurement Science and Technology, 2023, 34(11): 115107.
[8] Li X, Zhang W, Ma H, et al. Partial transfer learning in machinery cross-domain fault diagnostics using class-weighted adversarial networks[J]. Neural Networks, 2020, 129: 313-322.
[9] 王军辉，雷文平，刘华杰，等. 基于深度动态域适应的轴承故障诊断研究[J]. 振动与冲击，2023，42(14): 245-250. DOI:10.13465/j.cnki.jvs.2023.14.029.
WANG Junhui, LEI Wenping, LIU Huajie, et al. Bearing
fault diagnosis based on deep dynamic domain adaptation [J]. Journal of Vibration and Shock, 2023, 42(14): 245-250. DOI:10.13465/j.cnki.jvs.2023.14.029.
[10] 周长巍，李国勇，任密蜂，等. 基于区分性联合概率分布的域适应故障诊断[J]. 振动与冲击，2023，42(07): 170-179. DOI:10.13465/j.cnki.jvs.2023.07.020.
ZHOU Changwei, LI Guoyong, REN Mifeng, et al. Domain adaption fault diagnosis based on discriminative joint probability distribution difference [J]. Journal of Vibration and Shock, 2023, 42(07): 170-179. DOI:10.13465/j.cnki.jvs.2023.07.020.
[11] 董绍江，周存芳，陈里里，等. 基于判别性特征提取和双重域对齐的轴承跨域故障诊断[J]. 中国机械工程，2023，34(15): 1856-1863.
DONG Shaojiang, ZHOU Cunfang, CHEN Lili, et al. Cross-domain Fault Diagnosis of Bearings Based on Discriminant Feature Extraction and Dual-domain Alignment [J]. China Mechanical Engineering, 2023, 34(15): 1856-1863.
[12] Kipf T N, Welling M. Semi-supervised classification with graph convolutional networks[J]. arXiv preprint arXiv:1609.02907, 2016.
[13] 陈起磊，蒋亦悦，唐瑶，等. 基于时频图与改进图卷积神经网络的异步电机故障诊断方法[J]. 振动与冲击，2022，41(24): 241-248. DOI:10.13465/j.cnki.jvs.2022.24.030.
CHEN Qilei, JIANG Yiyue, TANG Yao, et al. An induction motor fault diagnosis method based on the time-frequency image method and an improved graph convolutional network [J]. Journal of Vibration and Shock, 2022, 41(24): 241-248. DOI:10.13465/j.cnki.jvs.2022.24.030.
[14] Li T, Zhao Z, Sun C, et al. Domain adversarial graph convolutional network for fault diagnosis under variable working conditions[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 1-10.
[15] Ranjan E, Sanyal S, Talukdar P. Asap: Adaptive structure aware pooling for learning hierarchical graph representations[C]. Proceedings of the AAAI Conference on Artificial Intelligence. 2020, 34(04): 5470-5477.
[16] Zhang J, Yang X, Yu S, et al. Roughening of hollow glass microspheres by NaF for Ni electroless plating[J]. Surface and Coatings Technology, 2019, 359: 62-72.
[17] Elhassan T, Aljurf M. Classification of imbalance data using tomek link (t-link) combined with random under-sampling (rus) as a data reduction method[J]. Global J Technol Optim S, 2016, 1: 2016.
[18] Peng T, Shen C, Sun S, et al. Fault feature extractor based on bootstrap your own latent and data augmentation algorithm for unlabeled vibration signals[J]. IEEE Transactions on Industrial Electronics, 2021, 69(9): 9547-9555.
[19] 谢莹，刘雪伟，鲁振杰. 基于改进辅助分类生成对抗网络的小样本轴承故障诊断方法[J/OL]. 轴承, 1-12[2024-01-13]. http://kns.cnki.net/kcms/detail/41.1148.TH.20231220.1742.004.html.
XIE Ying, LIU Xuewei, LU Zhenjie. Small Sample Bearing Fault Diagnosis Method Based on Modified Auxiliary Classifier Generative Adversarial Network [J]. Bearing, 1-12[2024-01-13]. http://kns.cnki.net/kcms/detail/41.1148.TH.20231220.1742.004.html.
[20] 熊美景. 基于变分自编码器的数据增强方法在旋转设备故障诊断中的应用[D].广东工业大学，2021. DOI:10.27029/d.cnki.ggdgu.2021.001845.
XIONG Meijing. Application of Data Augmentation Method in Fault Diagnosis of Rotating Equipment Using Variational Auto-encoder [D]. Chinese Master's Theses Full-text Database, 2021. DOI:10.27029/d.cnki.ggdgu.2021.001845.
[21] Kim J, Kang J, Sohn M. Ensemble learning-based filter-centric hybrid feature selection framework for high-dimensional imbalanced data[J]. Knowledge-Based Systems, 2021, 220: 106901.
[22] Xu Q, Lu S, Jia W, et al. Imbalanced fault diagnosis of rotating machinery via multi-domain feature extraction and cost-sensitive learning[J]. Journal of Intelligent Manufacturing, 2020, 31(6): 1467-1481.
[23] Kuang J, Xu G, Tao T, et al. Class-imbalance adversarial transfer learning network for cross-domain fault diagnosis with imbalanced data[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 71: 1-11.
[24] Chen Z, Yu W, Wang L, et al. A Dual-View Style Mixing Network for unsupervised cross-domain fault diagnosis with imbalanced data[J]. Knowledge-Based Systems, 2023, 278: 110918.
[25] Defferrard M, Bresson X, Vandergheynst P. Convolutional neural networks on graphs with fast localized spectral filtering[J]. Advances in neural information processing systems, 2016, 29.
[26] Ding K, Xu Z, Tong H, et al. Data augmentation for deep graph learning: A survey[J]. ACM SIGKDD Explorations Newsletter, 2022, 24(2): 61-77.
[27] Zhu Y, Xu Y, Yu F, et al. Graph contrastive learning with adaptive augmentation[C]. Proceedings of the Web Conference 2021. 2021: 2069-2080.
[28] Han X, Jiang Z, Liu N, et al. G-mixup: Graph data augmentation for graph classification[C]. International Conference on Machine Learning. PMLR, 2022: 8230-8248.
[29] Rong Y, Huang W, Xu T, et al. Dropedge: Towards deep graph convolutional networks on node classification[C]. International Conference on Learning Representation. 2019.
[30] Chen Z, Ke H, Xu J, et al. Multichannel Domain Adaptation Graph Convolutional Networks-Based Fault Diagnosis Method and With Its Application[J]. IEEE Transactions on Industrial Informatics, 2023, 19(6): 7790-7800.
[31] Chen D, Lin Y, Li W, et al. Measuring and Relieving the Over-Smoothing Problem for Graph Neural Networks from the Topological View[C]. Proceedings of the AAAI conference on artificial intelligence, 2020, 34(04): 3438-3445.
[32] Wen X, Lu G, Liu J, et al. Graph modeling of singular values for early fault detection and diagnosis of rolling element bearings[J]. Mechanical Systems and Signal Processing, 2020, 145: 106956.
[33] Zhang B, Sconyers C, Byington C, et al. A probabilistic fault detection approach: Application to bearing fault detection[J]. IEEE Transactions on Industrial Electronics, 2011, 58(5): 2011-2018.
[34] Hui K H, Lim M H, Leong M S, et al. Dempster-Shafer evidence theory for multi-bearing faults diagnosis[J]. Engineering Applications of Artificial Intelligence, 2017, 57: 160-170.
[35] 陈子旭，朱振杰，卢国梁. 一种新的图谱域滚动轴承早期故障检测与识别方法[J]. 振动与冲击，2022，41(06): 51-59. DOI:10.13465/j.cnki.jvs.2022.06.008.
CHEN Zixu, ZHU Zhenjie, LU Guoliang. Novel early fault detection and diagnosis for rolling element bearings in graph spectrum domain [J]. Journal of Vibration and Shock, 2022, 41(06): 51-59. DOI:10.13465/j.cnki.jvs.2022.06.008.
[36] Li T, Zhou Z, Li S, et al. The emerging graph neural networks for intelligent fault diagnostics and prognostics: A guideline and a benchmark study[J]. Mechanical Systems and Signal Processing, 2022, 168: 108653.
[37] Marins, Matheus A, Ribeiro, et al. Improved similarity-based modeling for the classification of rotating-machine failures[J]. 2018, 355(4): 1913-1930.
[38] Hamilton W, Ying Z, Leskovec J. Inductive representation learning on large graphs[J]. Advances in neural information processing systems, 2017, 30.