基于性能衰退评估的轴承寿命状态识别方法研究

PDF(1725 KB)

振动与冲击 ›› 2021, Vol. 40 ›› Issue (5) : 186-192.

论文

董绍江1，吴文亮1，贺坤1，潘雪娇1，蒙志强1，汤宝平2，赵兴新3

作者信息 +

Bearing life state recognition method based on performance degradation evaluation

DONG Shaojiang1, WU Wenliang1, HE Kun1, PAN Xuejiao1, MENG Zhiqiang1, TANG Baoping2, ZHAO Xingxin3

Author information +

文章历史 +

摘要

针对滚动轴承退化性能难以评估、寿命状态难以识别的难题，提出一种基于性能衰退评估的轴承寿命状态识别新方法，该方法基于卷积自编码器（convolutional autoencoder，CAE）与多维尺度分析（multidimensional scaling, MDS）算法构建轴承性能衰退指标，再根据构建指标和改进卷积神经网络（convolutional neural network，CNN）建立轴承寿命状态识别模型，实现轴承寿命状态识别。首先，将轴承信号样本输入CAE，实现轴承寿命状态特征的自动提取与表达，再将所提取的特征通过MDS算法进行约简获得低维特征，在低维特征空间构造欧氏距离作为轴承性能衰退指标，依据指标实现轴承数据标签化。然后，使用标签化的轴承数据训练CNN，建立轴承寿命状态识别模型。在训练过程中，为抑制过拟合，对原始训练样本进行加噪处理，为提高模型抗干扰能力，将Leaky ReLU（LReLU）函数和dropout作为激活函数。最后，运用轴承全寿命试验数据对识别模型进行检验，通过对比验证，结果表明所提出的轴承寿命状态识别方法能更准确的实现轴承寿命状态识别。

Abstract

Aiming at the difficulty in evaluating the degradation performance and identifying the life state of rolling bearings, a new method for bearing life state recognition based on improved convolutional neural network with anti-interference is proposed. First, the life characteristics of bearing signal is extracted and expressed by the CAE. Then the feature is reduced by the MDS. The Euclidean geometrical distance method was constructed in the low-dimensional feature space as an indicator of bearing performance degradation. After that, the CNN bearing life state recognition model is constructed based on the tagged bearing data. In order to suppress the over-fitting of the model, the original training samples are added with noise. The Leaky ReLU (LReLU） function and dropout are used as the activation function to improve the anti-interference in the model. The results of test show that the proposed method can more accurately realize the bearing life state recognition.

导出引用

董绍江1，吴文亮1，贺坤1，潘雪娇1，蒙志强1，汤宝平2，赵兴新3. 基于性能衰退评估的轴承寿命状态识别方法研究[J]. 振动与冲击, 2021, 40(5): 186-192

DONG Shaojiang1, WU Wenliang1, HE Kun1, PAN Xuejiao1, MENG Zhiqiang1, TANG Baoping2, ZHAO Xingxin3. Bearing life state recognition method based on performance degradation evaluation[J]. Journal of Vibration and Shock, 2021, 40(5): 186-192

参考文献

[1] 李恒, 张氢, 秦仙蓉, 等. 基于短时傅里叶变换和卷积神经网络的轴承故障诊断方法[J]. 振动与冲击, 2018, 37(19): 124-131.
LI Heng, ZHANG Qin, QIN Xianrong, et al. Fault diagnosis method for rolling bearings based on short-time Fourier transform and convolution neural network[J]. Journal of Vibration and Shock, 2018, 37(19): 124-131.
[2] 郑小霞, 周国旺, 任浩翰, 等. 基于变分模态分解和排列熵的滚动轴承故障诊断[J]. 振动与冲击, 2017, 36(22): 22-28
ZHENG Xiaoxia, ZHOU Guowang, REN Haohan, et al. A rolling bearing fault diagnosis method based on variational mode decomposition and permutation entropy [J]. Journal of Vibration and Shock, 2017, 36(22): 22-28.
[3] 钱林, 康敏, 傅秀清, 等. 基于VMD的自适应形态学在轴承故障诊断中的应用[J]. 振动与冲击, 2017, 36(03): 227-233.
QIAN Lin, KANG Min, FU Xiuqing, et al. Application of adaptive morphology in bearing fault diagnosis based on VMD [J]. Journal of Vibration and Shock, 2017, 36(03): 227-233.
[4] Y. Wang, Y. Z. Peng, Y. Y. Zi, et al. A Two-Stage Data-Driven-Based Prognostic Approach for Bearing Degradation Problem[J]. Ieee Transactions on Industrial Informatics, 2016, 12(3): 924-932.
[5] P. Boskoski, M. Gasperin, D. Petelin, et al. Bearing fault prognostics using Renyi entropy based features and Gaussian process models[J]. Mechanical Systems and Signal Processing, 2015, 52-53: 327-337.
[6] X. F. Chen, Z. J. Shen, Z. J. He, et al. Remaining life prognostics of rolling bearing based on relative features and multivariable support vector machine[J]. Proceedings of the Institution of Mechanical Engineers Part C-Journal of Mechanical Engineering Science, 2013, 227(12): 2849-2860.
[7] 张康, 黄亦翔, 赵帅, 等. 基于t-SNE数据驱动模型的盾构装备刀盘健康评估[J/OL]. 机械工程学报:1-8. http://kns.cnki.net/kcms/detail/11.2187.TH.20190124.1320.072.html
ZHANG Kang, HUANG Yixiang, ZHAO Shuai, et al. Health Assessment of Shield Equipment Cutterhead Based on t-SNE Data-driven Model[J/OL]. Journal of Mechanical Engineering:1-8. http://kns.cnki.net/kcms/detail/11.2187.TH.20190124.1320.072.html
[8] J. Ben Ali, B. Chebel-Morello, L. Saidi, et al. Accurate bearing remaining useful life prediction based on Weibull distribution and artificial neural network[J]. Mechanical Systems and Signal Processing, 2015, 56-57: 150-172.
[9] T. Benkedjouh, K. Medjaher, N. Zerhouni, et al. Remaining useful life estimation based on nonlinear feature reduction and support vector regression[J]. Engineering Applications of Artificial Intelligence, 2013, 26(7): 1751-1760.
[10] D. Jamroz. APPLICATION OF MULTIDIMENSIONAL SCALING TO CLASSIFICATION OF VARIOUS TYPES OF COAL[J]. Archives of Mining Sciences, 2014, 59(2): 413-425.
[11] X. M. Zhao, Y. Li, Q. H. Zhao. Mahalanobis distance based on fuzzy clustering algorithm for image segmentation[J]. Digital Signal Processing, 2015, 43: 8-16.
[12] Hai Qiu, Jay Lee, Jing Lin, et al. Wavelet filter-based weak signature detection method and its application on rolling element bearing prognostics[J]. Journal of Sound & Vibration, 2006, 289(4): 1066-1090.
[13] Alex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton. ImageNet Classification with Deep Convolutional Neural Networks[C]. 2012.
[14] Xavier Bouthillier, Kishore Konda, Pascal Vincent, et al., Dropout as data augmentation[C]. //International Conference on Learning Representations, Puerto Rico, 2016.
[15] M. Gan, C. Wang, C. A. Zhu. Construction of hierarchical diagnosis network based on deep learning and its application in the fault pattern recognition of rolling element bearings[J]. Mechanical Systems and Signal Processing, 2016, 72-73: 92-10