基于VAE的工业过程不平衡数据故障诊断
Fault Diagnosis of Imbalanced Data in Industrial Processes Based on VAE
DOI: 10.12677/pm.2024.144113, PDF,    国家自然科学基金支持
作者: 王 敖, 田 颖*：上海理工大学光电信息与计算机工程学院，上海
关键词: 工业过程；故障诊断；变分自编码器；不平衡数据；数据增强；Industrial Process； Fault Diagnosis； Variational Autoencoder； Imbalanced Data； Data Augmentation
摘要: 深度学习技术的快速发展为工业过程故障诊断问题提供了越来越多的解决方法，但在训练数据集不平衡的情况下，深度学习方法的表现往往不能令人满意。因此，本文提出了一种应用于工业过程中的不平衡数据故障诊断策略。首先使用有限的故障样本训练VAE模型并生成增强样本，之后将生成的样本用于丰富不平衡数据集，最终使用深度神经网络进行故障诊断。采用PRONTO数据集对提出的方法进行验证，实验结果表明本文提出的故障诊断策略能够有效提高模型的故障诊断性能。
Abstract: The rapid development of deep learning technology provides more and more solutions to industrial process fault diagnosis problems, but when the training data set is imbalanced, the performance of deep learning methods is often unsatisfactory. Therefore, this paper proposed a fault diagnosis strategy for imbalanced data applied in industrial processes. First, limited fault samples are used to train a VAE and generate augmentation samples. The generated samples are then used to enrich the imbalanced data set, and finally a neural network is used for fault diagnosis. The PRONTO benchmark dataset is used to verify the proposed method. The experimental results show that the fault diagnosis method proposed in this paper can effectively improve the fault diagnosis performance of the model.
文章引用：王敖, 田颖. 基于VAE的工业过程不平衡数据故障诊断[J]. 理论数学, 2024, 14(4): 80-90. https://doi.org/10.12677/pm.2024.144113

参考文献

[1] Zhang, J.S., Zhang, K., An, Y.Y., Luo, H. and Yin, S. (2023) An Integrated Multitasking Intelligent Bearing Fault Diagnosis Scheme Based on Representation Learning under Imbalanced Sample Condition. IEEE Transactions on Neural Networks and Learning Systems. [Google Scholar] [CrossRef] 
[2] Al-Haddad, L.A. and Jaber, A.A. (2023) An Intelligent Fault Diagnosis Approach for Multirotor UAVs Based on Deep Neural Network of Multi-Resolution Transform Features. Drones, 7, Article No. 82. [Google Scholar] [CrossRef] 
[3] Tang, S.N., Yuan, S.Q. and Zhu, Y. (2020) Deep Learning-Based Intelligent Fault Diagnosis Methods toward Rotating Machinery. IEEE Access, 8, 9335-9346. [Google Scholar] [CrossRef] 
[4] Zhou, T.T., Han, T. and Droguett, E.L. (2022) Towards Trustworthy Machine Fault Diagnosis: A Probabilistic Bayesian Deep Learning Framework. Reliability Engineering & System Safety, 224, Article ID: 108525. [Google Scholar] [CrossRef] 
[5] Krizhevsky, A., Sutskever, I. and Hinton, G. (2012) ImageNet Classification with Deep Convolutional Neural Networks. NIPS Advances in Neural Information Processing Systems 25, Lake Tahoe, 3-6 December 2012, 25.
[6] 赵志宏, 吴冬冬, 窦广鉴, 杨绍普. 一种基于CNN-BiGRU孪生网络的轴承故障诊断方法[J]. 振动与冲击, 2023, 42(6): 166-171, 211.
[7] 王连云, 陶洪峰, 徐琛, 杨慧中. 基于多层训练干扰的CNN轴承故障诊断[J]. 控制工程, 2022, 29(9): 1652-1657.
[8] 王泳欣, 张大斌, 车大庆, 吕建秋. 面向不平衡数据集分类的LDBSMOTE过采样方法[J]. 统计与决策, 2022, 38(18): 58-63.
[9] 褚菲, 丁珮宽, 马小平, 王福利. 基于CWGANs数据增强的离心压缩机迁移建模方法[J]. 控制工程, 2023, 30(2): 292-299.
[10] Shorten, C. and Khoshgoftaar, T.M. (2019) A Survey on Image Data Augmentation for Deep Learning. Journal of Big Data, 6, Article No. 60. [Google Scholar] [CrossRef] 
[11] Chawla, N.V., Bowyer, K.W., Hall, L.O. and Kegelmeyer, W.P. (2002) SMOTE: Synthetic Minority Over-Sampling Technique. Journal of Artificial Intelligence Research, 16, 321-357. [Google Scholar] [CrossRef] 
[12] Soltanzadeh, P. and Hashemzadeh, M. (2021) RCSMOTE: Range-Controlled Synthetic Minority Over-Sampling Technique for Handling the Class Imbalance Problem. Information Sciences, 542, 92-111. [Google Scholar] [CrossRef] 
[13] Arora, J., et al. (2022) MCBC-SMOTE: A Majority Clustering Model for Classification of Imbalanced Data. Computers, Materials & Continua, 73, 4801-4817.
[14] 张俊杰, 曹丽. 基于代价敏感神经网络集成模型的类别不平衡问题研究[J]. 合肥工业大学学报(自然科学版), 2023, 46(11): 1573-1579.
[15] Frumosu, F.D., Khan, A.R., Schioler, H., Kulahci, M., Zaki, M. and Westermann-Rasmussen, P. (2020) Cost-Sensitive Learning Classification Strategy for Predicting Product Failures. Expert Systems with Applications, 161, Article ID: 113653. [Google Scholar] [CrossRef] 
[16] Ren, Z., et al. (2022) Adaptive Cost-Sensitive Learning: Improving the Convergence of Intelligent Diagnosis Models under Imbalanced Data. Knowledge-Based Systems, 241, Article ID: 108296. [Google Scholar] [CrossRef] 
[17] Xu, Q., Lu, S., Jia, W. and Jiang, C. (2020) Imbalanced Fault Diagnosis of Rotating Machinery via Multi-Domain Feature Extraction and Cost-Sensitive Learning. Journal of Intelligent Manufacturing, 31, 1467-1481. [Google Scholar] [CrossRef] 
[18] Kingma, D.P. and Welling, M. (2014) Auto-Encoding Variational Bayes.
[19] 范振杰, 罗娜. 基于改进VAE的时间序列数据增强方法[J]. 华东理工大学学报(自然科学版), 2023, 50(3): 1-11.
[20] Liu, Y., et al. (2023) Cloud-VAE: Variational Autoencoder with Concepts Embedded. Pattern Recognition: The Journal of the Pattern Recognition Society, 140, Article ID: 109530. [Google Scholar] [CrossRef] 
[21] Stief, A., Tan, R., Cao, Y., Ottewill, J.R., Thornhill, N.F. and Baranowski, J. (2019) A Heterogeneous Benchmark Dataset for Data Analytics: Multiphase Flow Facility Case Study. Journal of Process Control, 79, 41-55. [Google Scholar] [CrossRef] 

为你推荐