基于深度迁移学习的水润滑轴承跨工况状态辨识

doi:10.3901/JME.260606

摘要/Abstract

摘要： 水润滑轴承作为关键的传动部件，其运行状态的稳定性对船舶的安全和效率至关重要。针对船舶水润滑轴承在运行过程中可能发生故障且单一信号诊断结果不可靠等问题，提出了一种综合利用动力学信息和摩擦学信息的Transformer-1D CNN模型，该模型通过自注意力机制有效捕捉长距离依赖关系，结合卷积网络的局部特征提取能力，能够对多种传感器数据(如摩擦系数、温度、振动信号等)进行综合分析，并结合迁移学习提出了一种跨工况的状态辨识方法，在高性能水润滑轴承综合试验台上进行试验获取数据。试验结果表明，在固定工况下的状态辨识任务中，所提出模型在测试集的准确率达99.97%；在跨工况的辨识任务中，基于特征的迁移学习方法结果优于基于参数的迁移学习方法，在三种跨工况辨识任务中准确率分别达到94.91%、91.38%、96.96%，相比于未使用迁移学习准确率分别提升了17.43%、18.03%、13.59%，证明了所提出的Transformer-1D CNN模型及其结合迁移学习的方法在提高船舶水润滑轴承故障诊断准确性方面的有效性，能够为船舶运营安全提供可靠保障。

关键词: 水润滑轴承, 状态辨识, 深度学习, 迁移学习

Abstract: As a critical transmission component，the operational stability of water-lubricated bearings is essential for the safety and efficiency of ships. To address the potential failures of ship water-lubricated bearings during operation and the unreliability of diagnostic results based on single signals, this study proposes a Transformer-1D CNN model that integrates dynamic and tribological information. The model leverages the self-attention mechanism to effectively capture long-range dependencies and combines the local feature extraction capabilities of convolutional networks. It enables comprehensive analysis of multi-sensor data, such as friction coefficient, temperature, and vibration signals. Additionally, a cross-condition state identification method based on transfer learning is introduced. Experimental data are obtained from a high-performance water-lubricated bearing comprehensive testing platform. The results show that for state identification tasks under fixed conditions, the proposed model achieves an accuracy of 99.97% on the test set. For cross-condition identification tasks, the feature-based transfer learning approach outperforms the parameter-based method，achieving accuracies of 94.91%, 91.38%, and 96.96% in three cross-condition tasks, representing improvements of 17.43%, 18.03%, and 13.59%, respectively, compared to non-transfer learning methods. These findings demonstrate the effectiveness of the proposed Transformer-1D CNN model and its integration with transfer learning in enhancing the accuracy of fault diagnosis for ship water-lubricated bearings, providing a reliable safeguard for ship operational safety.

Key words: water-lubricated bearings, state identification, deep learning, transfer learning

中图分类号:

U664

冯润麒, 郭智威, 薛恩驰, 袁成清. 基于深度迁移学习的水润滑轴承跨工况状态辨识[J]. 机械工程学报, 2026, 62(11): 445-459.

FENG Runqi, GUO Zhiwei, XUE Enchi, YUAN Chengqing. Cross-condition State Identification of Water-lubricated Bearings Based on Deep Transfer Learning[J]. Journal of Mechanical Engineering, 2026, 62(11): 445-459.

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: http://www.cjmenet.com.cn/CN/10.3901/JME.260606

http://www.cjmenet.com.cn/CN/Y2026/V62/I11/445

参考文献

[1] WU Zumin，SHENG Chenxing，GUO Zhiwei，et al. Surface texture processing for tribological performance improvement of UHMWPE-based water-lubricated bearings[J]. Industrial Lubrication and Tribology，2018，70(7)：1341-9.
[2] WU Zumin，GUO Zhiwei，YUAN Chengqing. Influence of polyethylene wax on wear resistance for polyurethane composite material under low speed water-lubricated conditions[J]. Wear，2019，426：1008-17.
[3] 李群，刘晓玲，张政，等. 低速工况下水润滑轴承材料的混合润滑性能[J]. 轴承，2024，(5)：84-91. LI Qun，LIU Xiaoling，ZHANG Zheng，et al. Mixed lubrication performances of water-lubricated bearing materials under low speed conditions[J]. Bearing，2024(5)：84-91.
[4] ZHANG Xiaohan，YU Tao，JI Hao，et al. Analysis of water-lubricated journal bearings assisted by a small quantity of secondary lubricating medium with Navier–Stokes equation and VOF model[J]. Lubricants，2024，12(1)：16.
[5] 解忠良，焦见，杨康. 舰船用水润滑轴承微观界面润滑机理研究[J]. 机械工程学报，2022，58(21)：186-200. XIE Zhongliang，JIAO Jian，YANG Kang，et al. Investigation on the micro interface lubrication mechanism of water-lubricated-bearing of large ships [J]. Journal of Mechanical Engineering，2022，58(21)：186-200.
[6] 贾谦，杨帅，刘哲，等. 水润滑推力轴承摩擦学性能的分析和试验研究[J]. 动力工程学报，2023，43(5)：535-542. JIA Qian，YANG Shuai，LIU Zhe，et al. Analytical and experimental research on tribological properties of water-lubricated thrust bearings[J]. Journal of Chinese Society of Power Engineering，2023，43(5)：535-542.
[7] WU Zumin，YUAN Chengqing，GUO Zhiwei，et al. Effect of the groove parameters on the lubricating performance of the water-lubricated bearing under low speed[J]. Wear，2023，522：204708.
[8] WU Chuang，WANG Yinbo，CHEN Feng，et al. A mixed-lubricated friction model of water-lubricated- bearing based on fractal theory[J]. Proceedings of the Institution of Mechanical Engineers，Part J：Journal of Engineering Tribology，2023，237(10)：1913-1929.
[9] 崔博超，陈平，杨潍旭，等. 微织构排布方式对水润滑轴承启停过程摩擦学性能的影响[J]. 中国表面工程，2024，37(1)：160-168，178. CUI Bochao，CHEN Ping，YANG Weixu，et al. Effect of microtexture arrangement on tribological properties of water-lubricated bearings during start-stop process[J]. China Surface Engineering，2024，37(1)：160-168，178.
[10] QU Chunhui，WANG Tingmei，WANG Qihua，et al. A novel ternary interpenetrating polymer networks based on NBR/PU/EP with outstanding damping and tribological properties for water-lubricated bearings[J]. Tribology International，2022，167：107249.
[11] PENG Engao，LIU Zhenglin，ZHOU Xincong，et al. Application of vibration and noise analysis in water-lubricated rubber bearings fault diagnosis[J]. Advanced Materials Research，2011，328：1995-1999.
[12] 张圣东，龙志林，金勇，等. 基于持续同调机器学习的尾轴承黏滑振动研究[J]. 振动测试与诊断，2021，41(4)：756-61，834. ZHANG Shengdong，LONG Zhilin，JIN Yong，et al. Stick-slip vibration of water-lubricated rubber stern tube bearing based on persistent homology based machine learning[J]. Journal of Vibration，Measurement & Diagnosis，2021，41(4)：756-61，834.
[13] 刘祺霖. 基于试验信息的偏载全尺寸水润滑轴承动力学建模研究[D]. 武汉：武汉理工大学，2022. LIU Qilin. Study on dynamic modeling of full-size water-lubricated bearing under bias load based on test information[D]. Wuhan：Wuhan University of Technology，2022.
[14] 王裕，徐起秀，郭智威，等. 基于随机森林算法的改性水润滑轴承摩擦性能预测[J]. 润滑与密封，2022，47(8)：150-155. WANG Yu，XU Qixiu，GUO Zhiwei，et al. Friction performance analysis of modified water-lubricated bearing based on random forest algorithm[J]. Lubrication Engineering，2022，47(8)：150-155.
[15] 周兴康，余建波. 基于深度一维残差卷积自编码网络的齿轮箱故障诊断[J]. 机械工程学报，2020，56(7)：96-108. ZHOU Xingkang，YU Jianbo. Gearbox fault diagnosis based on one-dimension residual convolutional auto- encoder[J]. Journal of Mechanical Engineering，2020，56(7)：96-108.
[16] 姜洪开，邵海东，李兴球. 基于深度学习的飞行器智能故障诊断方法[J]. 机械工程学报，2019，55(7)：27-34. JIANG Hongkai，SHAO Haidong，LI Xingqiu. Deep Learning Theory with Application in Intelligent Fault Diagnosis of Aircraft.[J]. Journal of Mechanical Engineering，2019，55(7)：27-34.
[17] LEI Yaguo，ZUO Mingjian. Gear crack level identification based on weighted K nearest neighbor classification algorithm[J]. Mechanical Systems and Signal Processing，2009，23(5)：1535-47.
[18] SU Zuqiang，JIANG Weilong，CHEN Kai，et al. Multi-adversarial deep transfer network for multi-source open-set fault diagnosis of rotating machinery with category shift[J]. Knowledge-Based Systems，2023，282：111106.
[19] VASHISHTHA G，KUMAR R. Unsupervised learning model of sparse filtering enhanced using wasserstein distance for intelligent fault diagnosis[J]. Journal of Vibration Engineering & Technologies，2023，11(7)：2985-3002.
[20] LITWIN W. Problems of water-lubricated sliding bearings[M]. Amsterdam：Elsevier，2024.
[21] 国家市场监督管理总局，中国国家标准化管理委员会. GB/T 39982—2021，水润滑径向滑动轴承承载能力测试方法[S].北京：中国标准出版社，2021. State Administration for Market Requlation，Standardization Administration of the People’s Republic of China. GB/T 39982—2021 Water lubricated journal bearings—Test methods of load carrying capacity[S]. Beijing：Standards Press of China，2021.
[22] 王松，郑卫刚. 水润滑橡胶尾轴承摩擦、磨损试验研究[J]. 机械制造，2011，49(08)：98-100.
WANG Song，ZHENG Weigang. The friction and wear test study of water-lubricated rubber tail bearings[J]. Machinery，2011，49(8)：98-100.
[23] MA Shangjun，CAI Wei，LIU Wenkai，et al. A lighted deep convolutional neural network based fault diagnosis of rotating machinery[J]. Sensors，2019，19(10)：2381.