Abstract
Process monitoring plays an essential role in safe and profitable operations. Various data-driven fault detection models have been suggested, but they cannot perform properly when the training data are insufficient or the information to construct the manifold is confined to a specific region. In this study, a process monitoring framework integrated with data augmentation is proposed to supplement rare but informative samples for the boundary regions of the normal state. To generate data for augmentation, a variational autoencoder was employed to exploit its advantage of stable convergence. For the construction of the process monitoring system, an autoencoder that can extract useful features in an unsupervised manner was used. To illustrate the efficacy of the proposed method, a case study for the Tennessee Eastman process was applied. The results show that the proposed method can improve the monitoring performance compared to the autoencoder without data augmentation in terms of fault detection accuracy and delay, particularly within the feature space.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
V. Venkatasubramanian, R. Rengaswamy, K. Yin and S. N. Kavuri, Comput. Chem. Eng., 27(3), 293 (2003).
J. Lee, C. Yoo, S. Wook, P. A. Vanrolleghem and I. Lee, 59, 223 (2004).
W. Yan, P. Guo and Z. Li, Chemom. Intell. Lab. Syst., 158, 31 (2016).
A. J. Holden, Science, 313, 504 (2006).
F. Lv, C. Wen, Z. Bao and M. Liu, 2016 Am. Control Conf., 2, 6851 (2016).
J. Fan and W Wang, IEEE, 1001 (2017).
P. Vincent, H. Larochelle, I. Lajoie, Y. Bengio and P. A. Manzagol, J. Mach. Learn. Res., 11, 3371 (2010).
S. Rifai, P. Vincent, X. Muller, X. Glorot and Y. Bengio, Proc. 28th Int. Conf. Mach. Learn. ICML 2011, 1, 833 (2011).
L. Jiang, Z. Song, Z. Ge and J. Chen, Ind. Eng. Chem. Res., 56, 26 (2017).
S. Heo and J. H. Lee, Processes, 7, 7 (2019).
Z. Zhang, T. Jiang, S. Li and Y. Yang, J. Process Control, 64, 49 (2018).
W Yu and C. Zhao, IEEE Trans. Control Syst. Technol., 1 (2019).
H. Zhao, Chemom. Intell. Lab. Syst., 176, 11 (2018).
P. Y. Simard, D. Steinkraus and J. C. Platt, Proc. Int. Conf. Doc. Anal. Recognition, ICDAR, 958 (2003).
H. Han, W Y. Wang and B. H. Mao, in Lecture Notes in Computer Science, 3644, 878 (2005).
S. C. Wong and M. D. Mcdonnell, 2016 Int. Conf. Digit. Image Comput. Tech. Appl., 1 (2016).
T. Devries and G. W. Taylor, arXiv Prepr. arXiv1702.05538, 1 (2017).
Z. Wan, Y. Zhang and H. He, IEEE, 1 (2017).
N. Etworks, A. Storkey and H. Edwards, arXiv preprint arXiv:1711. 04340, 1 (2017).
J. Jorge, R. Paredes, J. A. Sanchez and M. Bened, VISIGRAPP (5: VISAPP), 96 (2018).
W. N. Hsu, Y. Zhang and J. Glass, IEEE Autom. Speech Recognit. Underst. Work., 1, 16 (2017).
X. Gao, F. Deng and X. Yue, Neurocomputing, 396, 487 (2019).
S. K. Lim, Y. Loo, N. Tran, N. Cheung, G. Roig and Y. Elovici, 2018 IEEE Int. Conf. Data Min., 1122 (2018).
C. Mellon and U. C. Berkeley, arXiv preprint arXiv:1606.05908, 1 (2016).
A. Krogh and J. A. Hertz, Adv. Neural Inf. Process. Syst. (1992).
N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever and R. Salakhutdinov, J. Mach. Learn. Res., 15(1), 1929 (2014).
S. Ioffe and C. Szegedy, Int. Conf. Mach. Learn. (2015).
S. Han, J. Pool, J. Tran and W. J. Dally, arXiv preprint arXiv:1506. 02626 (2015).
S. Heo and J. H. Lee, Comput. Chem. Eng., 127, 1 (2019).
P. Baldi and K. Hornik, Neural Networks, 2(1), 53 (1989).
S. Zhao, J. Song and S. Ermon, Proc. AAAI Conf. Artif. Intell., 33(1), 5885 (2019).
J. J. Downs and E. C. Company, Comput. Chem. Eng., 17, 3 (1993).
A. Bathelt, N. L. Ricker and M. Jelali, IFAC-PapersOnLine, 48(8), 309 (2014).
N. L. Ricker, J. Process Control, 6(4), 205 (1996).
H. Lee, C. Kim, S. Lim and J. Min, Comput. Chem. Eng., 142, 107064 (2020).
V. Venkatasubramanian, R. Rengaswamy, S. N. Kavuri and K. Yin, Comput. Chem. Eng., 27(3), 327 (2003).
R. T. Samuel and Y. Cao, Syst. Sci. Control Eng., 4(1), 165 (2016).
D. L. Olson and D. Delen, Advanced data mining techniques, Springer, December (2013).
C. Kim, H. Lee, K. Kim, Y. Lee and W. B. Lee, Ind. Eng. Chem. Res., 57(39), 13144 (2018).
A. Makhzani, B. Frey and I. Goodfellow, arXiv Prepr. arXiv1511.05644 (2014).
Acknowledgement
This work was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (MSIP) (NRF-2016R1A5A1009592).
Author information
Authors and Affiliations
Corresponding authors
Additional information
Conflict of Interest
The authors declare no conflict of interest.
Rights and permissions
About this article
Cite this article
Lee, H., Kim, C., Jeong, D.H. et al. Data-driven fault detection for chemical processes using autoencoder with data augmentation. Korean J. Chem. Eng. 38, 2406–2422 (2021). https://doi.org/10.1007/s11814-021-0894-1
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11814-021-0894-1