改进DenseNet的干气密封摩擦润滑状态识别研究

PDF(2590 KB)

振动与冲击 ›› 2025, Vol. 44 ›› Issue (4) : 313-321.

故障诊断分析

改进DenseNet的干气密封摩擦润滑状态识别研究

张帅1，丁雪兴*1，王世鹏2，力宁3，张兰霞1

作者信息 +

Study of dry gas seal friction lubrication state identification with improved DenseNet

ZHANG Shuai1, DING Xuexing*1, WANG Shipeng2, LI Ning3, ZHANG Lanxia1

Author information +

文章历史 +

摘要

为了克服干气密封运行中端面接触状态参数（膜厚、端面开启时间）测量困难的问题，提出自注意力机制融合稠密连接网络（DenseNet-CBAM）的干气密封端面摩擦润滑状态识别方法。根据斯特里贝克曲线和干气密封运行规律分析端面可能出现的摩擦润滑状态：流体润滑，边界润滑、混合润滑。通过声发射传感器采集密封系统运行时的声发射信号，通过滤波、时域分析、频域分析得出能够表征各种摩擦润滑状态的特征分量，获取三维连续小波（3D-CWT）时频图，最终基于深度学习模型Densenet-CBAM识别时频图，实现密封系统摩擦润滑状态识别。与其他二维时频特征图作为输入端相比，3D-CWT时频图提高了状态识别的准确率。同时，相较于其他深度学习模型，该方法对干气密封摩擦润滑状态识别精度高，达到了99.27%。

Abstract

In order to address the difficulties in measuring key contact state parameters (such as membrane thickness and open time of end faces) during the operation of dry gas seals, a method for recognizing the friction lubrication states of dry gas seal end faces is proposed by integrating a self-attention mechanism with a DenseNet-CBAM (Convolutional Block Attention Module) network. Based on the Stribeck curve and the operating laws of dry gas seals, the potential friction lubrication states of the end faces are identified as fluid lubrication, boundary lubrication, and mixed lubrication. Acoustic emission signals generated during the operation of the seal system are collected using sensors, and characteristic components representing various friction lubrication states are extracted through filtering, time-domain analysis, and frequency-domain analysis. A 3D Continuous Wavelet Transform (3D-CWT) is applied to generate time-frequency maps, and ultimately, the deep learning model DenseNet-CBAM is employed to recognize these maps and identify the friction lubrication states of the seal system. The 3D-CWT time-frequency map improves the accuracy of state identification compared to other 2D time-frequency feature maps as inputs.Compared to other deep learning models, this approach achieves a high precision of 99.27% in identifying the friction lubrication states of dry gas seals.

导出引用

张帅1, 丁雪兴1, 王世鹏2, 力宁3, 张兰霞1. 改进DenseNet的干气密封摩擦润滑状态识别研究[J]. 振动与冲击, 2025, 44(4): 313-321

ZHANG Shuai1, DING Xuexing1, WANG Shipeng2, LI Ning3, ZHANG Lanxia1. Study of dry gas seal friction lubrication state identification with improved DenseNet[J]. Journal of Vibration and Shock, 2025, 44(4): 313-321

参考文献

[1] Wang Y, Yang H, Wang J, et al. Theoretical analyses and field applications of gas-film lubricated mechanical face seals with herringbone spiral grooves[J]. Tribology Transactions, 2009, 52(6): 800-806.
[2] 李晓晖,傅攀,曹伟青,等.机械密封端面接触状态的声发射监测研究[J].振动与冲击,2016,35(08):83-89.
LI Xiaohui, FU Pan, CAO Weiqing, et al. Research on acoustic emission monitoring of mechanical seal end face contact state[J]. Journal of Vibration and Shock ,2016,35(08):83-89.
[3] Salant R , Wolff P , Navon S .Electronically Controlled Mechanical Seal for Aerospace Applications—Part I: Design, Analysis, and Steady State Tests©[J].Tribology Transactions, 1994,37:189-175
[4] Green Z I .Clearance control of a mechanical face seal[J].Tribology Transactions, 1999, 42(3):535-540.
[5] Reddyhoff T, Dwyer-Joyce RS, Harper P. A New Approach for the Measurementof Film Thickness in Liquid Face Seals[J]. Tribol Trans ,2008,51(2):140–9.
[6] Yin Y, Huang W, Liu X, et al. Analysis of the dynamic friction of a gas face seal based on acoustic emissions[J]. Tribology Letters, 2018, 66: 1-10.
[7] Orcutt F K .An Investigation of the Operation and Failure of Mechanical Face Seals[J].Journal of Tribology, 1969, 91(4):713.
[8] Juha Miettinen, Veli Siekkinen. Acoustic emission in monitoring sliding contact behaviour[J].Wear, 1995, 181–183(1):897-900.
[9] Towsyfyan H., Wei N., Gu F., et al. Identification of lubrication regimes in mechanical seals using acoustic emission for condition monitoring. In: Proceedings of the 54th Annual Conference on British Institute of Non-Destructive Testing, vol. 42,pp. 219–225: (2015).
[10] 高志,林尤滨,黄伟峰,等.干气密封启动过程中的声发射信号特征[J].清华大学学报(自然科学版),2013,53(03):319-322+329.
Gao Z, Lin Yubin, Huang Weifeng, et al. Characterization of acoustic emission signals during the startup of dry gas seals[J]. Journal of Tsinghua University(Natural Science Edition),2013,53(03):319-322+329.
[11] Huang W , Lin Y , Gao Z ,et al.An Acoustic Emission Study on the Starting and Stopping Processes of a Dry Gas Seal for Pumps[J].Tribology Letters, 2013, 49(2):379-384.
[12] Xu L, Wu JH, Yuan X, et al. Influence of inlet pressure disturbance on transient performance of liquid oxygen lubricated mechanical seal and rub-impact phenomenon caused by excitation overload[J]. Tribology International,2023,178:108058.
[13] Hu F, He Q, Huang W, et al. Face grooves for state monitoring in dry gas seals[J]. Tribology International, 2023, 189: 108950.
[14] Zhiqin Zhu, Yangbo Lei, Guanqiu Qi,et al.A review of the application of deep learning in intelligent ault diagnosis of rotating machinery[J].Measurement,2023,206,112346.
[15] Tamilselvan P, Wang P. Failure diagnosis using deep belief learning based health state classification[J]. Reliability Engineering & System Safety, 2013, 115: 124-135.
[16] Wang G, Zhang Y, Zhang F, et al. An ensemble method with DenseNet and evidential reasoning rule for machinery fault diagnosis under imbalanced condition[J]. Measurement, 2023, 214: 112806.
[17] Liu Z, Mao L, Hu Z, et al. A novel densely connected neural network for proton exchange membrane fuel cell fault diagnosis[J]. International Journal of Hydrogen Energy, 2022, 47(94): 40041-40053.
[18] Yu J, Zhou X, Lu L, et al. Multiscale dynamic fusion global sparse network for gearbox fault diagnosis[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 1-11.
[19] Ha J, Ram J P, Kim Y J, et al. Data-Driven Two-Stage Fault Detection and Diagnosis Method for Photovoltaic Power Generation[J]. IEEE Transactions on Instrumentation and Measurement, 2024,73:1:13
[20] An Z, Wu F, Zhang C, et al. Deep learning-based composite fault diagnosis[J]. IEEE Journal on Emerging and Selected Topics in Circuits and Systems, 2023, 13(2): 572-581.
[21] Xiaohui Li , Pan F , Kan C ,et al.The Contact State Monitoring for Seal End Faces Based on Acoustic Emission Detection[J].Shock and Vibration,2015, 2016:1-8.
[22] Sun X, Zheng X, Xu P, et al. Research on friction state monitoring method of liquid film seal during start-up based on acoustic emission[J]. Applied Acoustics, 2023, 210: 109424.
[23] Huang G, Liu Z, Van Der Maaten L, et al. Densely connected convolutional networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 4700-4708.
[24] Woo S, Park J, Lee J Y, et al. Cbam: Convolutional block attention module[C]//Proceedings of the European conference on computer vision (ECCV). 2018: 3-19.
[25] Huang T, Zhang Q, Tang X, et al. A novel fault diagnosis method based on CNN and LSTM and its application in fault diagnosis for complex systems[J]. Artificial Intelligence Review, 2022, 55(2): 1289-1315.
[26] Zang B, Gao H, Yang X, et al. Fault diagnosis of charging facilities based on improved RNN[C]//Journal of Physics: Conference Series. IOP Publishing, 2022, 2301(1): 012008.
[27] Hou S, Lian A, Chu Y. Bearing fault diagnosis method using the joint feature extraction of Transformer and ResNet[J]. Measurement Science and Technology, 2023, 34(7): 075108.
[28] Yuan J, Liu H, Liu W, et al. A method for melt pool state monitoring in laser-based direct energy deposition based on DenseNet[J]. Measurement, 2022, 195: 111146.
[29] Zhou Q, Zhou Z, Chen C, et al. Grading of hepatocellular carcinoma using 3D SE-DenseNet in dynamic enhanced MR images[J]. Computers in biology and medicine, 2019, 107: 47-57.
[30] Towsyfyan H, Gu F, Ball A.D, et al. Tribological behavior diagnostic and fault detection of mechanical seals based on acoustic emission measurements[J]. Friction7, 572-586 (2019).