基于图域泛化的直升机尾传动系统故障诊断方法

陈国旺; 唐倩; 何刘

doi:10.7527/S1000-6893.2025.32642

航空学报 >

0 1 - 0

DOI: https://doi.org/10.7527/S1000-6893.2025.32642

基于图域泛化的直升机尾传动系统故障诊断方法

陈国旺 ,
唐倩 ,
何刘

展开

重庆大学

收稿日期: 2025-07-31

修回日期: 2025-12-28

网络出版日期: 2025-12-29

基金资助

国家自然科学基金重点资助项目

收起

Fault diagnosis method of helicopter tail transmission systems based on graph domain generalization

CHEN Guo-Wang ,
TANG Qian ,
HE Liu

Expand

Received date: 2025-07-31

Revised date: 2025-12-28

Online published: 2025-12-29

Fold

摘要

针对直升机尾传动系统高精度故障数据难获取和传统迁移学习对未知故障的知识泛化不足问题，本文提出了用于直升机尾传动系统的自适应权重多级融合图域泛化(AMFGDG)故障诊断框架。首先，结合集中质量法与有限元法，生成直升机尾传动系统不同健康状态下的多节点振动数据，并用实验验证了该方法的有效性。基于所生成数据方法构建多源域的带有限元节点的图样本，同时以不同传感器的测试数据作为目标域的图样本。然后，采用自注意力机制完成图网络边权重自适应更新，实现节点间影响强度精准表征。引入混合专家门控策略完成跨图层级的信息融合，缓解多层图网络堆叠导致的过平滑现象。最后，结合分类损失、深度聚类损失及局部分布对齐损失实现诊断模型更新，完成故障诊断分类。基于直升机尾传动试验台私有数据集和西储大学轴承公开数据集的诊断结果对比和消融实验，验证了提出方法具有低数据依赖性和高泛化能力，诊断精度高达95.4%。

关键词： 故障诊断; 直升机尾传动系统; 图卷积网络; 域泛化; 动力学建模

本文引用格式

陈国旺 , 唐倩 , 何刘 . 基于图域泛化的直升机尾传动系统故障诊断方法[J]. 航空学报, 0 : 1 -0 . DOI: 10.7527/S1000-6893.2025.32642

Abstract

To address the challenges of limited access to high-quality fault data and the poor generalization performance of traditional transfer learning methods on unseen fault types, this paper proposes an Adaptive-weight Multi-level Fusion Graph Domain Generalization (AMFGDG) framework for fault diagnosis of helicopter tail transmission systems. Specifically, multi-node vibration signals under various health conditions are generated by integrating the concentrated mass method with the finite element method, and their validity is experimentally verified. Graph-structured samples with finite element nodes are constructed from the generated data for multiple source domains, while test data from different sensors are formulated as target domain graphs. A self-attention mechanism is employed to adaptively update edge weights, enabling precise modeling of inter-node influence. Additionally, a mixture-of-experts gating strategy is introduced to fuse information across graph layers and mitigate the over-smoothing effect in deep GNNs. The model is optimized via a joint loss function combining classification loss, deep clustering loss, and local distribution alignment loss. Diagnostic results and ablation studies on a private helicopter tail transmission test bench dataset and the public CWRU bearing dataset demonstrate the proposed method’s low data dependency and strong generalization capability, achieving an accuracy of up to 95.4%.

Key words： Fault diagnosis; Helicopter tail transmission system; Graph convolutional networks; Domain generalization; Dynamic modeling

参考文献

[1] ZHANG C, ZHU R P, CHEN W F, et al. Dynamic modeling and analysis of the supercritical tail drive system of a helicopter considering the mounting platform deformation[J]. Nonlinear Dynamics, 2024, 112(11): 9063-9094.
[2] ELASHA F, GREAVES M, MBA D. Planetary bearing defect detection in a commercial helicopter main gearbox with vibration and acoustic emission[J]. Structural Health Monitoring, 2018, 17(5): 1192-1212.
[3] ZHOU L H, DUAN F, MBA D, et al. Using frequency domain analysis techniques for diagnosis of planetary bearing defect in a CH-46E helicopter aft gearbox[J]. Engineering Failure Analysis, 2018, 92: 71-83.
[4] SAWALHI N, WANG W Y, BLUNT D. Helicopter planet gear rim crack diagnosis and trending using cepstrum editing enhanced with deconvolution[J]. Sensors, 2024, 24(8): 2593.
[5] 张新, 赵艺珂, 王家序, 等. 基于最大平均峭度盲解卷积的直升机故障诊断[J]. 振动、测试与诊断, 2024, 44(3): 480-485, 617.
ZHANG X, ZHAO Y K, WANG J X, et al. Helicopter fault diagnosis based on maximum kurtosis blind deconvolution[J]. Journal of Vibration，Measurement & Diagnosis, 2024, 44(3): 480-485, 617 (in Chinese).
[6] KONG Y, QIN Z Y, WANG T Y, et al. Data-driven dictionary design–based sparse classification method for intelligent fault diagnosis of planet bearings[J]. Structural Health Monitoring, 2022, 21(4): 1313-1328.
[7] 谢习华, 徐雷, 谭耀, 等. 基于PSO-GRNN的直升机旋翼不平衡故障诊断[J]. 噪声与振动控制, 2015, 35(2): 175-179.
XIE X H, XU L, TAN Y, et al. Helicopter rotor imbalance fault diagnosis based on PSO-GRNN[J]. Noise and Vibration Control, 2015, 35(2): 175-179 (in Chinese).
[8] LI T, ZHAO Z, SUN C, et al. Adaptive channel weighted CNN with multisensor fusion for condition monitoring of helicopter transmission system[J]. IEEE sensors journal, 2020, 20(15): 8364-8373.
[9] SUN K, YIN A. Multi-sensor graph adaptive federated generalization for helicopter transmission system fault diagnosis[J]. IEEE Transactions on Instrumentation and Measurement, 2024.
[10] 余志锋, 熊邦书, 熊天旸, 等. 基于VMD-CWT和改进CNN的直升机轴承故障诊断[J]. 航空动力学报, 2021, 36(5): 948-958.
YU Z F, XIONG B S, XIONG T Y, et al. Helicopter bearing fault diagnosis based on VMD-CWT and improved CNN[J]. Journal of Aerospace Power, 2021, 36(5): 948-958 (in Chinese).
[11] 李恒, 唐倩, 陈国旺, 等. 数字孪生辅助的直升机尾传动系统轴承与传动轴故障诊断[J]. 航空动力学报, 2024.
LI H, TANG Q, CHEN G W, et al. Digital twin-assisted fault diagnosis of helicopter tail transmission system bearings and drive shafts[J]. Journal of Aerospace Power, 2024 (in Chinese).
[12] LI W, ZHONG X, SHAO H D, et al. Multi-mode data augmentation and fault diagnosis of rotating machinery using modified ACGAN designed with new framework[J]. Advanced Engineering Informatics, 2022, 52: 101552.
[13] GUO Y, LI X Y, ZHANG J D, et al. SDCGAN: A CycleGAN-based single-domain generalization method for mechanical fault diagnosis[J]. Reliability Engineering & System Safety, 2025: 110854.
[14] LI X, ZHANG W, MA H, et al. Domain generalization in rotating machinery fault diagnostics using deep neural networks[J]. Neurocomputing, 2020, 403: 409-420.
[15] FAN Z H, XU Q F, JIANG C X, et al. Deep mixed domain generalization network for intelligent fault diagnosis under unseen conditions[J]. IEEE Transactions on Industrial Electronics, 2024, 71(1): 965-974.
[16] ZHAO C, SHEN W M. A domain generalization network combing invariance and specificity towards real-time intelligent fault diagnosis[J]. Mechanical Systems and Signal Processing, 2022, 173: 108990.
[17] SHI Y W, DENG A D, DENG M Q, et al. Domain transferability-based deep domain generalization method towards actual fault diagnosis scenarios[J]. IEEE Transactions on Industrial Informatics, 2023, 19(6): 7355-7366.
[18] JIA L S, CHOW T W S, YUAN Y X. Causal disentanglement domain generalization for time-series signal fault diagnosis[J]. Neural Networks, 2024, 172: 106099.
[19] LI T F, ZHOU Z, LI S N, 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.
[20] CHEN H, WANG X B, YANG Z X. Semi-supervised self-correcting graph neural network for intelligent fault diagnosis of rotating machinery[J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: 1-11.
[21] YIN P Z, NIE J, LIANG X Y, et al. A multiscale graph convolutional neural network framework for fault diagnosis of rolling bearing[J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: 1-13.
[22] WANG H, WANG J W, ZHAO Y K, et al. Few-shot learning for fault diagnosis with a dual graph neural network[J]. IEEE Transactions on Industrial Informatics, 2023, 19(2): 1559-1568.
[23] WANG Z, WU Z Y, LI X Q, et al. Attention-aware temporal–spatial graph neural network with multi-sensor information fusion for fault diagnosis[J]. Knowledge-Based Systems, 2023, 278: 110891.
[24] LIU R N, ZHANG Q H, LIN D, et al. Causal intervention graph neural network for fault diagnosis of complex industrial processes[J]. Reliability Engineering & System Safety, 2024, 251: 110328.
[25] LIU Z, PENG Z R. Few-shot bearing fault diagnosis by semi-supervised meta-learning with graph convolutional neural network under variable working conditions[J]. Measurement, 2025, 240: 115402.
[26] LI C Y, MO L F, KWOH C K, et al. Noise-robust multi-view graph neural network for fault diagnosis of rotating machinery[J]. Mechanical Systems and Signal Processing, 2025, 224: 112025.
[27] 李晖, 陈银超, 孙绍山, 等. 基于联邦图网络的转子系统故障诊断方法[J]. 航空学报, 2024, 45(17): 207-220.
LI H, CHEN Y C, SUN S S, et al. Rotor system fault diagnosis method based on federated graph network[J]. Journal of Aerospace, 2024, 45(17): 207-220 (in Chinese).
[28] QIN Y, LIU H Y, WANG Y, et al. Inverse physics–informed neural networks for digital twin–based bearing fault diagnosis under imbalanced samples[J]. Knowledge-Based Systems, 2024, 292: 111641.
[29] YU Y N, TANG L H, LIU Z P, et al. A novel bearing fault data generation strategy combining physical modeling and CycleGAN variant for fault diagnosis without real samples[J]. IEEE Transactions on Instrumentation and Measurement, 2025, 74: 1-17.
[30] MING Z, TANG B P, DENG L, et al. Digital twin-assisted fault diagnosis framework for rolling bearings under imbalanced data[J]. Applied Soft Computing, 2025, 168: 112528.
[31] 张锐奇, 孙弋, 于耀翔, 等. 数模联合驱动的动态对抗自适应轴承故障诊断方法[J]. 振动与冲击, 2023, 42(12): 256-263, 282.
ZHANG R Q, SUN Y, YU Y X, et al. Dynamic adversarial adaptive bearing fault diagnosis method driven by data and model joint[J]. Vibration and Shock, 2023, 42(12): 256-263, 282 (in Chinese).
[32] 于树博, 刘占生, 赵辰. 动力学仿真数据驱动的域自适应智能诊断方法[J]. 中国机械工程, 2023, 34(23): 2832-2841.
YU S B, LIU Z S, ZHAO C. Domain adaptive intelligent diagnostic method driven by dynamic simulation data[J]. China Mechanical Engineering, 2023, 34(23): 2832-2841 (in Chinese).
[33] XU W H, WANG H, JIN J B, et al. Dynamic model-based intelligent fault diagnosis method for fault detection in a rod fastening rotor[J]. Engineering Applications of Artificial Intelligence, 2025, 139: 109499.
[34] 张龙, 张号, 王朝兵, 等. 显式动力学驱动的轴承变工况故障诊断方法[J]. 中国机械工程, 2023, 34(8): 982-992.
ZHANG L, ZHANG H, WANG Z B, et al. Bearing fault diagnosis method under varying operating conditions driven by explicit dynamics[J]. China Mechanical Engineering, 2023, 34(8): 982-992 (in Chinese).
[35] WANG H, WANG S H, YANG R G, et al. A numerical simulation enhanced multi-task integrated learning network for fault detection in rotation vector reducers[J]. Mechanical Systems and Signal Processing, 2024, 217: 111525.
[36] 肖刚, 顾海瑞, 董锦锦, 等. 仿真数据驱动的长期服役电梯导轨故障迁移诊断方法[J]. 中国机械工程, 2024, 35(1): 125-135.
XIAO G, GU H R, DONG J J, et al. Transfer fault diagnosis method for elevator guide rails under long-term service driven by simulation data[J]. China Mechanical Engineering, 2024, 35(1): 125-135 (in Chinese).
[37] ZHU Y C, ZHUANG F Z, WANG J D, et al. Deep subdomain adaptation network for image classification[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(4): 1713-1722.
[38] CARON M, BOJANOWSKI P, JOULIN A, et al. Deep clustering for unsupervised learning of visual features[C]//Proceedings of the European conference on computer vision (ECCV). 2018: 132-149.
[39] Case western reserve university (CWRU) bearing data center. 2019, [Online]. Available: https://csegroups. case.edu/bearingdatacenter/pages/ download-data-file/. [Accessed September 2019].
[40] LI H L, PAN S J, WANG S Q, et al. Domain generalization with adversarial feature learning[C]// Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 5400-5409.
[41] SUN B C, SAENKO K. Deep CORAL: Correlation alignment for deep domain adaptation[C]//HUA G, JéGOU H. Computer Vision – ECCV 2016 Workshops. Cham: Springer International Publishing, 2016: 443-450.
[42] YANG Y T, YIN J C, ZHENG H L, et al. Learn generalization feature via convolutional neural network: A fault diagnosis scheme toward unseen operating conditions[J]. IEEE Access, 2020, 8: 91103-91115.
[43] RAGAB M, CHEN Z H, ZHANG W Y, et al. Conditional contrastive domain generalization for fault diagnosis[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 1-12.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献