基于多信息融合的INFO-VMD-CNN的齿轮箱故障诊断方法

PDF(2636 KB)

振动与冲击 ›› 2025, Vol. 44 ›› Issue (13) : 309-316.

故障诊断分析

基于多信息融合的INFO-VMD-CNN的齿轮箱故障诊断方法

吴胜利*1, 郑子润1, 邢文婷2

作者信息 +

Gearbox fault diagnosis method based on INFO-VMD-CNN of multi-information fusion

WU Shengli*1, ZHENG Zirun1XING Wenting2

Author information +

文章历史 +

摘要

针对齿轮箱振动信号复杂多变，导致现有的齿轮箱故障诊断方法诊断精度不高、较弱故障特征容易被噪声淹没等问题，提出了一种基于向量加权平均优化算法（INFO）、变分模态分解（variational mode decomposition，VMD）和卷积神经网络（convolutional neural network，CNN）的齿轮故障诊断方法。该方法首先采用熵权法将不同位置的振动传感器信号信息进行融合，利用向量加权均值算法（INFO）对VMD算法中参数进行优化，并设计一个复合评价指标作为参数优化的评价标准，使用奇异峭度差分谱（SKD）的方法对敏感分量进行重构；其次，从重构的信号中提取时域、频域特征并输入到CNN模型中进行分类；最后通过Shap（Shapley Additive Explanations）值法对模型输入特征的重要性进行排序，分析不同特征组合对模型分类和特定故障识别的影响。在东南大学行星齿轮数据集上进行验证，结果表明，利用所提特征组合进行故障诊断，CNN模型故障诊断准确率为98.24%，高于其他特征组合，为行星齿轮箱的故障诊断提供了一组有效的特征指标。

Abstract

To address the issues with complex and variable vibration signals in gearboxes, which led to low diagnostic accuracy of existing gearbox fault diagnosis methods and the risk of weak fault features being overwhelmed by noise, a new fault diagnosis method was proposed. Firstly, entropy weight fusion algorithm was used to fusion vibration sensor signals at different locations, and vector weighted mean value algorithm (INFO) was used to optimize parameters in the variational mode decomposition (VMD) algorithm. A composite evaluation index was designed as the evaluation standard for parameter optimization. The Singular Kurtosis Differential Spectrum (SKD) method was employed to reconstruct the sensitive components. What’s more, the time domain and frequency domain features were extracted from the reconstructed signals and fed into the CNN model for classification. Finally, Shapley Additive Explanations (Shap) value method was used to rank the importance of input features. The impact of different feature combinations on model classification and specific fault identification was analyzed. The proposed method was validated on the Planetary Gearbox Dataset from Southeast University. It was shown that using the proposed feature combination for fault diagnosis, the CNN model achieves an accuracy of 98.24%, which is higher than other combinations. This provides an effective set of feature indicators for planetary gearbox fault diagnosis.

导出引用

吴胜利1, 郑子润1, 邢文婷2. 基于多信息融合的INFO-VMD-CNN的齿轮箱故障诊断方法[J]. 振动与冲击, 2025, 44(13): 309-316

WU Shengli1, ZHENG Zirun1XING Wenting2. Gearbox fault diagnosis method based on INFO-VMD-CNN of multi-information fusion[J]. Journal of Vibration and Shock, 2025, 44(13): 309-316

参考文献

[1] Li X, Ma Z, Kang D, et al. Fault diagnosis for rolling bearing based on VMD-FRFT [J]. Measurement, 2020, 155: 107554.
[2] Rapur J S, Tiwari R. Experimental fault diagnosis for known and unseen operating conditions of centrifugal pumps using MSVM and WPT based analyses [J]. Measurement, 2019, 147: 106809.
[3] Yu D, Cheng J, Yang Y. Application of EMD method and Hilbert spectrum to the fault diagnosis of roller bearings [J]. Mechanical Systems and Signal Processing, 2005, 19(2): 259-270.
[4] Xu L, Chatterton S, Pennacchi P. Rolling element bearing diagnosis based on singular value decomposition and composite squared envelope spectrum [J]. Mechanical Systems and Signal Processing, 2021, 148: 107174.
[5] Wu Z, Huang N E. Ensemble empirical mode decomposition: a noise-assisted data analysis method [J]. Advances in adaptive data analysis, 2009, 1(01): 1-41.
[6] Dragomiretskiy K, Zosso D. Variational mode decomposition [J]. IEEE transactions on signal processing, 2013, 62(3): 531-544.
[7] Yang Y, Liu H, Han L, et al. A feature extraction method using VMD and improved envelope spectrum entropy for rolling bearing fault diagnosis [J]. IEEE Sensors Journal, 2023, 23(4): 3848-3858.
[8] He X, Zhou X, Yu W, et al. Adaptive variational mode decomposition and its application to multi-fault detection using mechanical vibration signals [J]. ISA Transactions, 2021, 111: 360-375.
[9] Yang K, Wang G, Dong Y, et al. Early chatter identification based on an optimized variational mode decomposition [J]. Mechanical Systems and Signal Processing, 2019, 115: 238-254.
[10] Li H, Li S, Sun J, et al. Ultrasound signal processing based on joint GWO-VMD wavelet threshold functions [J]. Measurement, 2024, 226: 114143.
[11] Jin Z, He D, Wei Z. Intelligent fault diagnosis of train axle box bearing based on parameter optimization VMD and improved DBN [J]. Engineering Applications of Artificial Intelligence, 2022, 110: 104713.
[12] Tang J, Zheng G, Wei C, et al. Signal-transformer: A robust and interpretable method for rotating machinery intelligent fault diagnosis under variable operating conditions [J]. Ieee Transactions on Instrumentation and Measurement, 2022, 71: 1-11.
[13] Borghesani P, Herwig N, Antoni J, et al. A Fourier-based explanation of 1D-CNNs for machine condition monitoring applications [J]. Mechanical Systems and Signal Processing, 2023, 205: 110865.
[14] 陈钱, 陈康康, 董兴建, 等. 一种面向机械设备故障诊断的可解释卷积神经网络 [J]. 机械工程学报, 2024, 60(12): 65-76.
Chen Q,Chen KK, Dong XJ, et al. Interpretable Convolutional Neural Network for Mechanical Equipment Fault Diagnosis [J]. Journal of Mechanical Engineering, 2024, 60(12):65-76.
[15] Du M, Wei Y, Zheng X, et al. Multi-feature based network for multivariate time series classification [J]. Information Sciences, 2023, 639: 119009.
[16] 王冉, 石如玉, 胡升涵, 等. 基于声成像与卷积神经网络的轴承故障诊断方法及其可解释性研究 [J]. 振动与冲击, 2022, 41(16): 224-231.
Wang R, Shi RY, Hu SH, et al. An acoustic fault diagnosis method of rolling bearings based on acoustic imaging and convolutional neural network [J]. Journal of Vibration and Shock，2022, 41(16): 224-231.
[17] Wang Y, Wang P, Tansey K, et al. An interpretable approach combining Shapley additive explanations and LightGBM based on data augmentation for improving wheat yield estimates [J]. Computers and Electronics in Agriculture, 2025, 229: 109758.
[18] Kumar M, Singh S, Bowles J. Variational Bayesian deep fuzzy models for interpretable classification [J]. Engineering Applications of Artificial Intelligence, 2024, 132: 107900.
[19] Shen S, Lu H, Sadoughi M, et al. A physics-informed deep learning approach for bearing fault detection [J]. Engineering Applications of Artificial Intelligence, 2021, 103: 104295.
[20] Zhang S, He M, Zhong Z, et al. An industrial interference-resistant gear defect detection method through improved YOLOv5 network using attention mechanism and feature fusion [J]. Measurement, 2023, 221: 113433.
[21] Feng Z, Shen X, Li P, et al. Performance optimization and scheme evaluation of liquid cooling battery thermal management systems based on the entropy weight method [J]. Journal of Energy Storage, 2024, 80: 110329.
[22] 李帅永, 毛维培, 程振华, 等. 基于VMD和K-SVD字典学习的供水管道泄漏振动信号压缩感知方法 [J]. 仪器仪表学报, 2020, 41(03): 49-60.
Li YS, Mao WP, CHENG ZH, et al, Compressed sensing method for leakage vibration signal in water-supply pipelines based on VMD and K-SVD dictionary learning [J]. Chinese Journal of Scientific Instrument, 2020, 41(03): 49-60.
[23] Ahmadianfar I, Heidari A A, Noshadian S, et al. INFO: An efficient optimization algorithm based on weighted mean of vectors [J]. Expert Systems with Applications, 2022, 195: 116516.
[24] Golafshan R, Sanliturk K Y. SVD and Hankel matrix based de-noising approach for ball bearing fault detection and its assessment using artificial faults [J]. Mechanical Systems and Signal Processing, 2016, 70: 36-50.
[25] Alomari Y, Andó M. SHAP-based insights for aerospace PHM: Temporal feature importance, dependencies, robustness, and interaction analysis [J]. Results in Engineering, 2024, 21: 101834.