Abstract:Cutters of shield boring machine (TBM) generally work in a complex geological environment with heavy load and impact, which can easily result in failure due to the partial wear of the cutter. Therefore, it is of particular importance to monitor the wear situation of the cutter to diagnose the cutter partial wear fault and guide the operations of the equipment based on data-driven techniques. In this study, aiming at the cutter partial wear problem, a novel diagnostic model based on wavelet time-frequency analysis and Inception-BiGRU is proposed, which can improve the fault diagnosis efficiency compared with existing models. The disc cutter is selected to perform linear rock breaking experiments on a multi-functional scaled rock-breaking test machine, during the process the vibration signals are collected. After that continuous wavelet transform (CWT) is used to obtain the wavelet time-frequency diagram that reflects the time-frequency domain characteristics of the vibration signals, and then the multi-scale spatial information is adaptively extracted using the convolutional kernels of different sizes of the Inception model. Furthermore, bidirectional gated recurrent units (BiGRU) are added to enable the model to learn the rich time-dependent relationships in the time-frequency diagram more accurately, and the hyperparameters of the model are determined by Bayesian optimization algorithm. The four diagnostic experiments of different cutter wear degree indicate that the proposed model can effectively extract the partial wear characteristics from the time-frequency diagram and diagnose the partial wear state of the cutting tools, which can realize end-to-end fault diagnosis of the TBM cutter's partial wear. The proposed model shows an average diagnostic accuracy up to 98.5%, whose diagnostic accuracy and stability are both indicated to outperform other algorithms, which proves the superiority of the proposed method.
樊翔翔1,2,项载毓1,2,孙瑞雪1,2,张敏1,莫继良1,2. 基于小波时频分析和Inception-BiGRU模型的盾构滚刀偏磨故障诊断[J]. 振动与冲击, 2023, 42(15): 232-240.
FAN Xiangxiang1,2, XIANG Zaiyu1,2, SUN Ruixue1,2, ZHANG Min1, MO Jiliang1,2. Fault diagnosis of TBM hob eccentric wear based on wavelet time-frequency analysis and Inception-BiGRU model. JOURNAL OF VIBRATION AND SHOCK, 2023, 42(15): 232-240.
[1] Jafar H. Development of an empirical model to estimate disc cutter wear for sedimentary and low to medium grade metamorphic rocks[J]. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research, 2018, 75: 90-99.
[2] Jin D, Shen Z, Yuan D. Effect of Spatial Variability on Disc Cutters Failure During TBM Tunneling in Hard Rock[J]. Rock Mechanics and Rock Engineering, 2020, 53(10): 1-13.
[3] Ren D J, Shen S L, Arul A, et al. Prediction Model of TBM Disc Cutter Wear During Tunnelling in Heterogeneous Ground[J]. Rock Mechanics and Rock Engineering, 2018,51: 3599-3611.
[4] Wang L, Kang Y, Zhao X, et al. Disc cutter wear prediction for a hard rock TBM cutterhead based on energy analysis[J]. Tunnelling and Underground Space Technology, 2015, 50: 324-333.
[5] Hassanpour J, Rostami J, Azali S T, et al. Introduction of an empirical TBM cutter wear prediction model for pyroclastic and mafic igneous rocks; a case history of Karaj water conveyance tunnel, Iran[J]. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research, 2014, 43: 222-231.
[6] Guo J B, Wang Y J, Liu J Z, et al. The Research and Design of a New Shield Cutter Wear Detection System[J]. Advanced Materials Research, 2013, 711: 381-384.
[7] Gong Q M, Wu F, Wang D J, et al. Development and Application of Cutterhead Working Status Monitoring System for Shield TBM Tunnelling[J]. Rock Mechanics and Rock Engineering, 2021, 54: 1731-1753.
[8] Lan H, Xia Y, Ji Z, et al. Online monitoring device of disc cutter wear-Design and field test[J]. Tunnelling and Underground Space Technology, 2019, 89: 284-294.
[9] 李宏波, 孙振川, 周建军, 等. 基于声发射和改进灰关联度分析的TBM滚刀磨损状态评估方法[J]. 中国铁道科学, 2019, 40(03): 65-71.
LI Hongbo, SUN Zhenchuan ZHOU Jianjun, et al. Wear Condition Evaluation Method of TBM Hob Based on Acoustic Emission and lmproved Grey Correlation Analysis[J]. China Railway Science, 2019, 40(03): 65-71.
[10] 梁波. 基于小波包和希尔伯特包络分析的盾构机主轴承故障诊断方法研究[D]. 兰州理工大学, 2018.
Liangbo, Research on fault diagnosis method of main bearing of shield machine based on wavelet packet and Hilbert envelope analysis[D]. Lanzhou University of Technology, 2018.
[11] 谷玉海, 朱腾腾, 饶文军, 等. 基于EMD二值化图像和CNN的滚动轴承故障诊断[J]. 振动.测试与诊断, 2021, 41(01): 105-113+203.
GU Yuhai, ZHU Tengteng, RAO Wenjun, et al. Fault Diagnosis for Rolling Bearing Based on EMD Binarization lmage and CNN[J]. Journal of Vibration, Measurement & Diagnosis, 2021, 41(01): 105-113+203.
[12] 李奕江, 张金萍, 李允公. 基于VMD-HMM的滚动轴承磨损状态识别[J]. 振动与冲击, 2018, 37(21): 61-67.
LI Yijiang, ZHANG Jinping, LI Yungong. Wear state recognition of rolling bearings based on VMD-HMM[J]. Journal of Vibration and Shock, 2018, 37(21): 61-67.
[13] 贾继德, 吴春志, 张玲玲, 等. 基于时频相干分析的曲轴主轴承磨损故障诊断研究[J]. 振动与冲击, 2018, 37(02): 114-120.
JIA Jide, WU Chunzhi, Zhang Lingling, et al. Wear fault diagnosis for crankshaft main bearing based on time-frequency coherence analysis[J]. Journal of Vibration and Shock, 2018, 37(02): 114-120.
[14] 吴晨芳, 杨世锡, 黄海舟, 等. 一种基于改进的LeNet-5模型滚动轴承故障诊断方法研究[J]. 振动与冲击, 2021, 40(12): 55-61.
WU Chenfang, YANG Shixi, HUANG Haizhou, et al. An improved fault diagnosis method of rolling bearings based on LeNet-5[J]. Journal of Vibration and Shock, 2021, 40(12): 55-61.
[15] 米珂. 基于虚拟仪器的盾构机故障预测与健康管理研究[D]. 石家庄铁道大学.
MI Ke. Research on Prognostics and Health Management System of Shiled Machine Based on Virtual Instrument[D]. Shijiazhuang Tiedao University, 2019.
[16] 张存吉, 姚锡凡, 张剑铭, 等. 基于深度学习的刀具磨损监测方法[J]. 计算机集成制造系统, 2017, 23(10): 2146-2155.
ZHANG Cunji, YAO Xifan, ZHANG Jianming, et al. Tool wear monitoring based on deep learning[J]. Computer Integrated Manufacturing Systems, 2017, 23(10): 2146-2155.
[17] 陈启鹏, 谢庆生, 袁庆霓, 等. 基于深度门控循环单元神经网络的刀具磨损状态实时监测方法[J]. 计算机集成制造系统, 2020, 26(07): 1782-1793.
CHEN Qipeng, XIE Qingsheng, YUAN Qingni, et al. Real-time monitoring method for wear state of tool based on deep bidirectional GRU model[J]. Computer Integrated Manufacturing Systems, 2020, 26(07): 1782-1793.
[18] 石茂林, 孙伟, 宋学官. 隧道掘进机大数据研究进展:数据挖掘助推隧道挖掘[J]. 机械工程学报, 2021, 57(22):344-358.
SHI Maolin, SUN Wei, SONG Xueguan. Research Progress on Big Data of Tunnel Boring Machine: How Data Mining Can Help Tunnel Boring[J]. Journal of Mechanical Engineering, 2021, 57(22): 344-358.
[19] 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.
[20] 张训杰, 张敏, 李贤均. 基于二维图像和CNN-BiGRU网络的滚动轴承故障模式识别[J]. 2021, 40(5):113-118.
ZHANG Xunjie, ZHANG Min, LI Xianjun. Rolling bearing fault mode recognition based on 2D image and CNN-BiGRU[J]. 2021, 40(5):113-118.
[21] 陈保家, 陈学力, 沈保明,等. CNN-LSTM深度神经网络在滚动轴承故障诊断中的应用[J]. 西安交通大学学报, 2021, 55(6): 28-36.
CHEN Baojia,CHEN Lixue,SHEN Baoming,et al.An application of convolution neural network and long short-term memory in rolling bearing fault diagnosis[J]. Journal of Xi'an Jiaotong University, 2021, 55(6): 28-36.