(Chung Yuan Christian University, Taiwan, Province of China)(2) Yi Shan Lee
(Chung Yuan Christian University, Taiwan, Province of China)(3) * Junghui Chen
(Chung Yuan Christian University, Taiwan, Province of China)*corresponding author
AbstractThe traditional fault diagnosis models cannot achieve good fault diagnosis accuracy when a new unseen fault class appears in the test set, but there is no training sample of this fault in the training set. Therefore, studying the unseen cause-effect problem of fault symptoms is extremely challenging. As various faults often occur in a chemical plant, it is necessary to perform fault causal-effect diagnosis to find the root cause of the fault. However, only some fault causal-effect data are always available to construct a reliable causal-effect diagnosis model. Another worst thing is that measurement noise often contaminates the collected data. The above problems are very common in industrial operations. However, past-developed data-driven approaches rarely include causal-effect relationships between variables, particularly in the zero-shot of causal-effect relationships. This would cause incorrect inference of seen faults and make it impossible to predict unseen faults. This study effectively combines zero-shot learning, conditional variational autoencoders (CVAE), and the signed directed graph (SDG) to solve the above problems. Specifically, the learning approach that determines the cause-effect of all the faults using SDG with physics knowledge to obtain the fault description. SDG is used to determine the attributes of the seen and unseen faults. Instead of the seen fault label space, attributes can easily create an unseen fault space from a seen fault space. After having the corresponding attribute spaces of the failure cause, some failure causes are learned in advance by a CVAE model from the available fault data. The advantage of the CVAE is that process variables are mapped into the latent space for dimension reduction and measurement noise deduction; the latent data can more accurately represent the actual behavior of the process. Then, with the extended space spanned by unseen attributes, the migration capabilities can predict the unseen causes of failure and infer the causes of the unseen failures. Finally, the feasibility of the proposed method is verified by the data collected from chemical reaction processes.
KeywordsConditional variational autoencoder; Fault diagnosis; Signed directed graph; Zero-shot learning
|
DOIhttps://doi.org/10.26555/ijain.v9i3.1434 |
Article metricsAbstract views : 375 | PDF views : 58 |
Cite |
Full Text Download
|
References
[1] Y. Ma, B. Song, H. Shi, and Y. Yang., “Fault detection via local and nonlocal embedding,” Chem. Eng. Res. Des., vol. 94, pp. 538–548, Feb. 2015, doi: 10.1016/j.cherd.2014.09.015.
[2] Z. Zhu et al., “A review of the application of deep learning in intelligent fault diagnosis of rotating machinery,” Measurement, vol. 206, p. 112346, Jan. 2023, doi: 10.1016/j.measurement.2022.112346.
[3] G. Wu, J. Tong, L. Zhang, Y. Zhao, and Z. Duan, “Framework for fault diagnosis with multi-source sensor nodes in nuclear power plants based on a Bayesian network,” Ann. Nucl. Energy, vol. 122, pp. 297–308, Dec. 2018, doi: 10.1016/j.anucene.2018.08.050.
[4] J. Xu, S. Liang, X. Ding, and R. Yan, “A zero-shot fault semantics learning model for compound fault diagnosis,” Expert Syst. Appl., vol. 221, p. 119642, Jul. 2023, doi: 10.1016/j.eswa.2023.119642.
[5] J. Ma and J. Jiang, “Applications of fault detection and diagnosis methods in nuclear power plants: A review,” Prog. Nucl. Energy, vol. 53, no. 3, pp. 255–266, Apr. 2011, doi: 10.1016/j.pnucene.2010.12.001.
[6] S. Gajjar, M. Kulahci, and A. Palazoglu, “Real-time fault detection and diagnosis using sparse principal component analysis,” J. Process Control, vol. 67, pp. 112–128, Jul. 2018, doi: 10.1016/j.jprocont.2017.03.005.
[7] Y. Han, G. Song, F. Liu, Z. Geng, B. Ma, and W. Xu, “Fault monitoring using novel adaptive kernel principal component analysis integrating grey relational analysis,” Process Saf. Environ. Prot., vol. 157, pp. 397–410, Jan. 2022, doi: 10.1016/j.psep.2021.11.029.
[8] M. R. Maurya, R. Rengaswamy, and V. Venkatasubramanian, “Fault Diagnosis by Qualitative Trend Analysis of the Principal Components,” Chem. Eng. Res. Des., vol. 83, no. 9, pp. 1122–1132, Sep. 2005, doi: 10.1205/cherd.04280.
[9] M. Žarković and Z. Stojković, “Analysis of artificial intelligence expert systems for power transformer condition monitoring and diagnostics,” Electr. Power Syst. Res., vol. 149, pp. 125–136, Aug. 2017, doi: 10.1016/j.epsr.2017.04.025.
[10] M. A. Kramer and B. L. Palowitch, “A rule‐based approach to fault diagnosis using the signed directed graph,” AIChE J., vol. 33, no. 7, pp. 1067–1078, Jul. 1987, doi: 10.1002/aic.690330703.
[11] C. Kang, “A Bayesian belief network-based advisory system for operational availability focused diagnosis of complex nuclear power systems,” Expert Syst. Appl., vol. 17, no. 1, pp. 21–32, Jul. 1999, doi: 10.1016/S0957-4174(99)00018-4.
[12] N. Liu, M. Hu, J. Wang, Y. Ren, and W. Tian, “Fault detection and diagnosis using Bayesian network model combining mechanism correlation analysis and process data: Application to unmonitored root cause variables type faults,” Process Saf. Environ. Prot., vol. 164, pp. 15–29, Aug. 2022, doi: 10.1016/J.PSEP.2022.05.073.
[13] C. C. Pantelides and J. G. Renfro, “The online use of first-principles models in process operations: Review, current status and future needs,” Comput. Chem. Eng., vol. 51, pp. 136–148, Apr. 2013, doi: 10.1016/j.compchemeng.2012.07.008.
[14] Y.-K. Liu, G.-H. Wu, C.-L. Xie, Z.-Y. Duan, M.-J. Peng, and M.-K. Li, “A fault diagnosis method based on signed directed graph and matrix for nuclear power plants,” Nucl. Eng. Des., vol. 297, pp. 166–174, Feb. 2016, doi: 10.1016/j.nucengdes.2015.11.016.
[15] Z. Zhou and Q. Zhang, “Model Event/Fault Trees With Dynamic Uncertain Causality Graph for Better Probabilistic Safety Assessment,” IEEE Trans. Reliab., vol. 66, no. 1, pp. 178–188, Mar. 2017, doi: 10.1109/TR.2017.2647845.
[16] X. Bu, H. Nie, Z. Zhang, and Q. Zhang, “An Industrial Fault Diagnostic System Based on a Cubic Dynamic Uncertain Causality Graph,” Sensors, vol. 22, no. 11, p. 4118, May 2022, doi: 10.3390/s22114118.
[17] L. Feng and C. Zhao, “Fault Description Based Attribute Transfer for Zero-Sample Industrial Fault Diagnosis,” IEEE Trans. Ind. Informatics, vol. 17, no. 3, pp. 1852–1862, Mar. 2021, doi: 10.1109/TII.2020.2988208.
[18] K. Hadad, M. Pourahmadi, and H. Majidi-Maraghi, “Fault diagnosis and classification based on wavelet transform and neural network,” Prog. Nucl. Energy, vol. 53, no. 1, pp. 41–47, Jan. 2011, doi: 10.1016/j.pnucene.2010.09.006.
[19] K. Mo, S. J. Lee, and P. H. Seong, “A dynamic neural network aggregation model for transient diagnosis in nuclear power plants,” Prog. Nucl. Energy, vol. 49, no. 3, pp. 262–272, Apr. 2007, doi: 10.1016/j.pnucene.2007.01.002.
[20] J. Qian, Z. Song, Y. Yao, Z. Zhu, and X. Zhang, “A review on autoencoder based representation learning for fault detection and diagnosis in industrial processes,” Chemom. Intell. Lab. Syst., vol. 231, p. 104711, Dec. 2022, doi: 10.1016/j.chemolab.2022.104711.
[21] M. Sun, H. Wang, P. Liu, S. Huang, and P. Fan, “A sparse stacked denoising autoencoder with optimized transfer learning applied to the fault diagnosis of rolling bearings,” Measurement, vol. 146, pp. 305–314, Nov. 2019, doi: 10.1016/j.measurement.2019.06.029.
[22] C. Zhang, D. Hu, and T. Yang, “Anomaly detection and diagnosis for wind turbines using long short-term memory-based stacked denoising autoencoders and XGBoost,” Reliab. Eng. Syst. Saf., vol. 222, p. 108445, Jun. 2022, doi: 10.1016/J.RESS.2022.108445.
[23] K. Wang, M. G. Forbes, B. Gopaluni, J. Chen, and Z. Song, “Systematic Development of a New Variational Autoencoder Model Based on Uncertain Data for Monitoring Nonlinear Processes,” IEEE Access, vol. 7, pp. 22554–22565, 2019, doi: 10.1109/ACCESS.2019.2894764.
[24] X. Yan, D. She, and Y. Xu, “Deep order-wavelet convolutional variational autoencoder for fault identification of rolling bearing under fluctuating speed conditions,” Expert Syst. Appl., vol. 216, p. 119479, Apr. 2023, doi: 10.1016/j.eswa.2022.119479.
[25] A. Mishra, S. K. Reddy, A. Mittal, and H. A. Murthy, “A Generative Model for Zero Shot Learning Using Conditional Variational Autoencoders,” in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Jun. 2018, vol. 2018-June, pp. 2269–22698, doi: 10.1109/CVPRW.2018.00294.
[26] A. A. Pol, V. Berger, C. Germain, G. Cerminara, and M. Pierini, “Anomaly Detection with Conditional Variational Autoencoders,” in 2019 18th IEEE International Conference On Machine Learning And Applications (ICMLA), Dec. 2019, pp. 1651–1657, doi: 10.1109/ICMLA.2019.00270.
[27] Y. Wei, D. Wu, and J. Terpenny, “Learning the health index of complex systems using dynamic conditional variational autoencoders,” Reliab. Eng. Syst. Saf., vol. 216, p. 108004, Dec. 2021, doi: 10.1016/j.ress.2021.108004.
[28] K. A. Reay and J. D. Andrews, “A fault tree analysis strategy using binary decision diagrams,” Reliab. Eng. Syst. Saf., vol. 78, no. 1, pp. 45–56, Oct. 2002, doi: 10.1016/S0951-8320(02)00107-2.
[29] Z. Masalegooyan, F. Piadeh, and K. Behzadian, “A comprehensive framework for risk probability assessment of landfill fire incidents using fuzzy fault tree analysis,” Process Saf. Environ. Prot., vol. 163, pp. 679–693, Jul. 2022, doi: 10.1016/j.psep.2022.05.064.
[30] R. M. Sinnamon and J. D. Andrews, “Improved efficiency in qualitative fault tree analysis,” Qual. Reliab. Eng. Int., vol. 13, no. 5, pp. 293–298, Sep. 1997, doi: 10.1002/(SICI)1099-1638(199709/10)13:5<293::AID-QRE110>3.0.CO;2-Y.
[31] S. Shao, S. McAleer, R. Yan, and P. Baldi, “Highly Accurate Machine Fault Diagnosis Using Deep Transfer Learning,” IEEE Trans. Ind. Informatics, vol. 15, no. 4, pp. 2446–2455, Apr. 2019, doi: 10.1109/TII.2018.2864759.
[32] Y. Pan, F. Mei, H. Miao, J. Zheng, K. Zhu, and H. Sha, “An Approach for HVCB Mechanical Fault Diagnosis Based on a Deep Belief Network and a Transfer Learning Strategy,” J. Electr. Eng. Technol., vol. 14, no. 1, pp. 407–419, Jan. 2019, doi: 10.1007/s42835-018-00048-y.
[33] “Zero-data learning of new tasks,” in Proceedings of the 23rd national conference on Artificial intelligence, 2008, pp. 646–651. [Online]. Available at: https://dl.acm.org/doi/10.5555/1620163.1620172.
[34] S. Rahman, S. Khan, and F. Porikli, “A Unified Approach for Conventional Zero-Shot, Generalized Zero-Shot, and Few-Shot Learning,” IEEE Trans. Image Process., vol. 27, no. 11, pp. 5652–5667, Nov. 2018, doi: 10.1109/TIP.2018.2861573.
[35] J. Yang, C. Wang, and C. Wei, “A novel Brownian correlation metric prototypical network for rotating machinery fault diagnosis with few and zero shot learners,” Adv. Eng. Informatics, vol. 54, p. 101815, Oct. 2022, doi: 10.1016/j.aei.2022.101815.
[36] C. H. Lampert, H. Nickisch, and S. Harmeling, “Learning to detect unseen object classes by between-class attribute transfer,” in 2009 IEEE Conference on Computer Vision and Pattern Recognition, Jun. 2009, pp. 951–958, doi: 10.1109/CVPR.2009.5206594.
[37] M. Mou, X. Zhao, K. Liu, and Y. Hui, “Variational autoencoder based on distributional semantic embedding and cross-modal reconstruction for generalized zero-shot fault diagnosis of industrial processes,” Process Saf. Environ. Prot., vol. 177, pp. 1154–1167, Sep. 2023, doi: 10.1016/j.psep.2023.07.080.
[38] B. Li and C. Zhao, “Federated Zero-Shot Industrial Fault Diagnosis With Cloud-Shared Semantic Knowledge Base,” IEEE Internet Things J., vol. 10, no. 13, pp. 11619–11630, Jul. 2023, doi: 10.1109/JIOT.2023.3243401.
[39] Y. Zhang, S. Zhang, X. Jia, X. Zhang, and W. Tian, “A novel integrated fault diagnosis method of chemical processes based on deep learning and information propagation hysteresis analysis,” J. Taiwan Inst. Chem. Eng., vol. 142, p. 104676, Jan. 2023, doi: 10.1016/j.jtice.2023.104676.
[40] H. Ali, A. S. Maulud, H. Zabiri, M. Nawaz, H. Suleman, and S. A. A. Taqvi, “Multiscale Principal Component Analysis-Signed Directed Graph Based Process Monitoring and Fault Diagnosis,” ACS Omega, vol. 7, no. 11, pp. 9496–9512, Mar. 2022, doi: 10.1021/acsomega.1c06839.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
___________________________________________________________
International Journal of Advances in Intelligent Informatics
ISSN 2442-6571 (print) | 2548-3161 (online)
Organized by UAD and ASCEE Computer Society
Published by Universitas Ahmad Dahlan
W: http://ijain.org
E: info@ijain.org (paper handling issues)
andri.pranolo.id@ieee.org (publication issues)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0