条件分布域适应下数模混动齿轮箱故障诊断

王冉1, 韩海保1, 颜福成1, 余亮2, 3

振动与冲击 ›› 2025, Vol. 44 ›› Issue (3) : 182-190.

PDF(3293 KB)
PDF(3293 KB)
振动与冲击 ›› 2025, Vol. 44 ›› Issue (3) : 182-190.
故障诊断分析

条件分布域适应下数模混动齿轮箱故障诊断

  • 王冉1, 韩海保1,颜福成1,余亮*2,3
作者信息 +

Fault diagnosis of digital-analog hybrid transmission gearbox under conditional distribution domain adaptation

  • WANG Ran1, HAN Haibao1, YAN Fucheng1, YU Liang*2,3
Author information +
文章历史 +

摘要

齿轮箱的故障诊断对于确保机械系统的可靠性、安全性和经济可行性至关重要。在工业实际中,齿轮箱通常运行在正常状态下,因此故障状态发生较少,且由于获取有标签的故障数据的成本较高,导致齿轮箱的健康状态监测面临着有标签故障数据稀缺的问题。然而,现有的深度迁移诊断方法存在数据生成质量不均匀和过度依赖少数类信息等局限性。为了克服这一挑战,提出条件分布域适应下数模混动齿轮箱故障诊断方法。首先基于集中参数法构建不同齿轮故障的动力学模型以扩充少标签源域的故障数据。其次,类条件分布最大均值差异(Class-Conditional Maximum Mean Discrepancy, CMMD)被嵌入诊断模型中,在再生希尔伯特核空间中(Reproducing kernel Hilbert space, RKHS)显式构建了故障特征与故障标签的关系,以减小源域数据和目标域数据的分布差异。同时,为保证目标域样本建立可靠的伪标签,熵损失被引入模型训练过程中。最后,通过两个实验验证了所提出方法的有效性和可行性。

Abstract

Gearbox fault diagnosis is crucial for ensuring the reliability, safety, and economic viability of mechanical systems. In industrial practice, gearboxes typically operate under normal conditions, resulting in fewer occurrences of faults. Moreover, the high cost associated with acquiring labeled fault data leads to a scarcity of labeled fault data for gearbox health monitoring. However, existing deep transfer diagnosis methods suffer from limitations such as uneven data generation quality and excessive reliance on minority class information. To overcome this challenge, we propose a method for fault diagnosis of hybrid transmission gearboxes under conditional distribution domain adaptation. Firstly, dynamic models of different gear faults are constructed based on the centralized parameter method to augment the limited labeled fault data in the source domain. Secondly, the Class-Conditional Maximum Mean Discrepancy (CMMD) is embedded into the diagnostic model to explicitly establish the relationship between fault features and fault labels in the Reproducing Kernel Hilbert Space (RKHS), reducing the distribution discrepancy between the source and target domain data. Meanwhile, to ensure reliable pseudo-labels for target domain samples, entropy loss is introduced during model training. Finally, the effectiveness and feasibility of the proposed method are validated through two experiments.

关键词

齿轮箱故障诊断;动力学建模;条件最大均值差异  /

Key words

Gearbox fault diagnosis / dynamic modeling / conditional maximum mean discrepancy

引用本文

导出引用
王冉1, 韩海保1, 颜福成1, 余亮2, 3. 条件分布域适应下数模混动齿轮箱故障诊断[J]. 振动与冲击, 2025, 44(3): 182-190
WANG Ran1, HAN Haibao1, YAN Fucheng1, YU Liang2, 3. Fault diagnosis of digital-analog hybrid transmission gearbox under conditional distribution domain adaptation[J]. Journal of Vibration and Shock, 2025, 44(3): 182-190

参考文献

[1] 雷亚国, 贾峰, 孔德同, 等. 大数据下机械智能故障诊断的机遇与挑战[J]. 机械工程学报, 2018, 54(5): 94-104.
LEI Yaguo,JIA Feng,KONG Detong,et al.Opportunities and challenges of machinery intelligent fault diagnosis in big data era[J].Journal of Mechanical Engineering, 2018, 54(5): 94-104.
[2] 雷亚国, 贾峰, 周昕, 等. 基于深度学习理论的机械装备大数据健康监测方法[J]. 机械工程学报, 2015, 51(21): 49-56.
LEI Yaguo,JIA Feng,ZHOU Xin,et al.A deep learning-based method for machinery health monitoring with big data[J].Journal of Mechanical Engineering, 2015, 51(21): 49-56.
[3] Yan S, Shao H, Min Z, et al. FGDAE: A new machinery anomaly detection method towards complex operating conditions[J]. Reliability Engineering & System Safety, 2023, 236: 109319.
[4] Kumar A, Kumar R. Oscillatory behavior-based wavelet decomposition for the monitoring of bearing condition in centrifugal pumps[J]. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology, 2018, 232(6): 757-772.
[5] Gu J X, Albarbar A, Sun X, et al. Monitoring and diagnosing the natural deterioration of multi-stage helical gearboxes based on modulation signal bispectrum analysis of vibrations[J]. International Journal of Hydromechatronics, 2021, 4(4): 309-330.
[6] Hilbert M, Smith W A, Randall R B. The effect of signal propagation delay on the measured vibration in planetary gearboxes[J]. Journal of Dynamics, Monitoring and Diagnostics, 2022, 1(1): 9-18.
[7] Li W, Huang R, Li J, et al. A perspective survey on deep transfer learning for fault diagnosis in industrial scenarios: Theories, applications and challenges[J]. Mechanical Systems and Signal Processing, 2022, 167: 108487.
[8] Sun G, Wang Y, Sun C, et al. Intelligent detection of a planetary gearbox composite fault based on adaptive separation and deep learning[J]. Sensors, 2019, 19(23): 5222.
[9] You W, Shen C, Wang D, et al. An intelligent deep feature learning method with improved activation functions for machine fault diagnosis[J]. IEEE Access, 2019, 8: 1975-1985.
[10] Chen P, Li Y, Wang K, et al. An automatic speed adaption neural network model for planetary gearbox fault diagnosis[J]. Measurement, 2021, 171: 108784.
[11] Yan R, Shen F, Sun C, et al. Knowledge transfer for rotary machine fault diagnosis[J]. IEEE Sensors Journal, 2019, 20(15): 8374-8393.
[12] YANG Bin,LEI Yaguo,JIA Feng,et al. An intelligent fault diagnosis approach based on transfer learning from laboratory bearings to locomotive bearings[J]. Mechanical Systems and Signal Processing, 2019,122:692-706.
[13] N. Lu, Z. Cui, H. Hu, T. Yin. Multi-view and multi-level network for fault diagnosis accommodating feature transferability. Expert Systems with Applications 213 (2023) 119057.
[14] Y. Qin, Q. Yao, Y. Wang, Y. Mao. Parameter sharing adversarial domain adaptation networks for fault transfer diagnosis of planetary gearboxes. Mechanical Systems and Signal Processing 160 (2021) 107936.
[15] Dong Y, Li Y, Zheng H, et al. A new dynamic model and transfer learning based intelligent fault diagnosis framework for rolling element bearings race faults: Solving the small sample problem[J]. ISA transactions, 2022, 121: 327-348.
[16] Zhu P, Dong S, Pan X, et al. A simulation-data-driven subdomain adaptation adversarial transfer learning network for rolling element bearing fault diagnosis[J]. Measurement Science and Technology, 2022, 33(7): 075101.
[17] 邵海东, 肖一鸣, 颜深. 仿真数据驱动的改进无监督域适应轴承故障诊断[J]. 机械工程学报, 2023, 59(3): 76-85.
SHAO Haidong,XIAO Yiming,YAN Shen.Simulation data-driven enhanced unsupervised domain adaptation for bearing fault diagnosis[J].Journal of Mechanical Engineering,2023,59(3): 76-85.
[18] Yu K, Fu Q, Ma H, et al. Simulation data driven weakly supervised adversarial domain adaptation approach for intelligent cross-machine fault diagnosis[J]. Structural Health Monitoring, 2021, 20(4): 2182-2198.
[19] Yi Y, Huang K, Xiong Y, et al. Nonlinear dynamic modelling and analysis for a spur gear system with time-varying pressure angle and gear backlash[J]. Mechanical Systems and Signal Processing, 2019, 132: 18-34.
[20] 万志国, 訾艳阳, 曹宏瑞, 等. 时变啮合刚度算法修正与齿根裂纹动力学建模[J]. 机械工程学报, 2013, 49(11): 153-160.
WAN Zhiguo,ZI Yanyang,CAO Hongrui,et al.Time-varying mesh stiffness algorithm correction and tooth crack dynamic modeling[J].Journal of Mechanical Engineering,2013,49(11): 153-160.
[21] Chaari F, Fakhfakh T, Haddar M. Analytical modelling of spur gear tooth crack and influence on gearmesh stiffness[J]. European Journal of Mechanics-A/Solids, 2009, 28(3): 461-468.
[22] Song L, Huang J, Smola A, et al. Hilbert space embeddings of conditional distributions with applications to dynamical systems[C]//Proceedings of the 26th Annual International Conference on Machine Learning. 2009: 961-968.  
[23] Baker C R. Joint measures and cross-covariance operators[R]. North Carolina State University. Dept. of Statistics, 1971.
[24] Ren Y, Zhu J, Li J, et al. Conditional generative moment-matching networks[J]. Advances in Neural Information Processing Systems, 2016, 29.
[25] Wang R, Yan F, Yu L, et al. Joint Wasserstein distance matching under conditional probability distribution for cross-domain fault diagnosis of rotating machinery[J]. Mechanical Systems and Signal Processing, 2024, 210: 111121.
[26] Grandvalet Y, Bengio Y. Semi-supervised learning by entropy minimization[J]. Advances in neural information processing systems, 2004, 17.

PDF(3293 KB)

Accesses

Citation

Detail

段落导航
相关文章

/