Research on the Prediction Method of Remaining Useful Life of Gears by Combining Fault Mechanism and Convolutional Neural Network

doi:10.3901/JME.2024.12.116

Abstract

Abstract: In the process of fatigue degradation of gears, there are usually steady-type faults and shock-type faults. These two types of fault features contain rich information about gear degradation. Most of the current end-to-end prediction methods on remaining useful life are based on the semantic features extracted by neural network. Basically, the establishment of the prediction model has a low degree of correlation with the gear, and the physical interpretability is poor. To this end, combined with the steady-type and shock-type fault response mechanism and the convolutional neural network, a convolution kernel in time domain and angle domain fused with fault prior is designed, and the feature extraction ability of the neural network is used to mine the steady-type and shock-type faults in the vibration signal. At the same time, an attention mechanism is introduced to suppress noise, dynamically capture different fault components and degraded information in signals at different times. Finally, the weighted optimization of the prediction error is realized by the noise covariance matrix of the model. The simulation and experimental results show that the method can effectively mine the fault information of the original signal, has better physical interpretability, and has higher prediction accuracy and better robustness on remaining useful life than the traditional intelligent prediction method, which verifies the effectiveness of the proposed method.

Key words: gear, steady-type faults, shock-type faults, neural network, remaining useful life

CLC Number:

TN911

LIU Huakai, DING Kang, HE Guolin, LI Weihua, LIN Huibin. Research on the Prediction Method of Remaining Useful Life of Gears by Combining Fault Mechanism and Convolutional Neural Network[J]. Journal of Mechanical Engineering, 2024, 60(12): 116-125.

References

[1] DEUTSCH J，HE D. Using deep learning-based approach to predict remaining useful life of rotating components[J]. IEEE Transactions on Systems，Man，and Cybernetics：Systems，2017，48(1)：11-20.
[2] 宋仁旺，王萪峰，石慧. 基于深度置信网络集成的齿轮剩余寿命预测[J]. 组合机床与自动化加工技术，2021(3)：70-73，79. SONG Renwang，WANG Kefeng，SHI Hui. Gear remaining useful life based on ensembel deep belief network[J]. Modular Machine Tool & Automatic Manufacturing Technique，2021(3)：70-73，79.
[3] LO C C，LEE C H，HUANG W C. Prognosis of bearing and gear wears using convolutional neural network with hybrid loss function[J]. Sensors，2020，20(12)：3539.
[4] JING L，ZHAO M，LI P，et al. A convolutional neural network based feature learning and fault diagnosis method for the condition monitoring of gearbox[J]. Measurement，2017，111：1-10.
[5] ZHANG W，PENG G，LI C，et al. A new deep learning model for fault diagnosis with good anti-noise and domain adaptation ability on raw vibration signals[J]. Sensors，2017，17(2)：425.
[6] QIN Y，XIANG S. Macroscopic–microscopic attention in LSTM networks based on fusion features for gear remaining life prediction[J]. IEEE Transactions on Industrial Electronics，2019，67(12)：10865-10875.
[7] YAN H，QIN Y，XIANG S，et al. Long-term gear life prediction based on ordered neurons LSTM neural networks[J]. Measurement，2020，165：108205.
[8] LI Tianfu，ZHAO Zhibin，SUN Chuang，et al. Wavelet Kernel net：An interpretable deep neural network for industrial intelligent diagnosis[J]. IEEE Transactions on Systems，Man，and Cybernetics：Systems，2021，51(1)：1-11.
[9] NASCIMENTO R G，VIANA F A C. Fleet prognosis with physics-informed recurrent neural networks[J]. arXiv preprint arXiv，2019：1901.05512.
[10] 雷亚国，汤伟，孔德同，等. 基于传动机理分析的行星齿轮箱振动信号仿真及其故障诊断[J]. 机械工程学报，2014，50(17)：61-68. LEI Yaguo，TANG Wei，KONG Detong，et al. Vibration signal simulation and fault diagnosis of planetary gearbox based on transmission mechanism analysis[J]. Journal of Mechanical Engineering，2014，50(17)：61-68.
[11] 何国林. 复合齿轮传动系统振动响应调制机理及稀疏分离方法研究[D]. 广州：华南理工大学，2015. HE Guolin. Research on vibration response modulation mechanism and sparse separation method of compound gear transmission system[D]. Guangzhou：South China University of Technology，2015.
[12] 左泽敏，李舜酩，郑娟丽. 相关分析在机械振动信号处理中的应用[J]. 机械制造与自动化，2009，38(1)：75-79. ZUO Zemin，LI Shunming，ZHEN Juanli. Application of correlation analysis in mechanical vibration signal processing[J]. Machine Building & Automation，2009，38(1)：75-79.
[13] STEVENSON J D. Structural damping values as a function of dynamic response stress and deformation levels [J]. Nuclear Engineering & Design，1980(2)：211-237.
[14] HE G，DING K，LI W，et al. A novel order tracking method for wind turbine planetary gearbox vibration analysis based on discrete spectrum correction technique[J]. Renewable Energy，2016，87：364-375.
[15] HU J，SHEN L，SUN G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2018：7132-7141.
[16] HOCHREITER S，SCHMIDHUBER J. Long short- TermMemory[J]. Neural Computation，1997，9(8)：1735-1780.
[17] LY A，MARSMAN M，VERHAGEN J，et al. A tutorial on Fisher information[J]. Journal of Mathematical Psychology，2017，80：40-55.
[18] HE K，ZHANG X，REN S，et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016：770-778.
[19] AL-DULAIMI A，ZABIHI S，ASIF A，et al. NBLSTM：Noisy and hybrid convolutional neural network and BLSTM-Based deep architecture for remaining useful life estimation[J]. Journal of Computing and Information Science in Engineering，2020，20(2)：021012.
[20] SHI X，CHEN Z，WANG H，et al. Convolutional LSTM network：A machine learning approach for precipitation nowcasting[J]. Advances in Neural Information Processing Systems，2015(1)：802-810.
[21] CHEN D，HONG W，ZHOU X. Transformer network for remaining useful life prediction of lithium-ion batteries[J]. IEEE Access，2022，10：19621-19628.