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.

Process Monitoring Fault Detection and Isolation (FDI) Autoencoder Variational Autoencoder Data Augmentation

Venkatsubramanian V, Rengaswamy R, Yin K, Kavuri SN, Comput. Chem. Eng., 27(3), 293 (2003)
Lee J, Yoo C, Wook S, Vanrolleghem PA, Lee I, 59, 223 (2004).
Yan W, Guo P, Li Z, Chemom. Intell. Lab. Syst., 15, 31 (2016)
Holden AJ, Science, 313, 504 (2006)
Lv F, Wen C, Bao Z, Liu M, 2016 Am. Control Conf., 2, 6851 (2016).
Fan J, Wang W, IEEE, 1001 (2017).
Vincent P, Larochelle H, Lajoie I, Bengio Y, Manzagol PA, J. Mach. Learn. Res., 11, 3371 (2010)
Rifai S, Vincent P, Muller X, Glorot X, Bengio Y, Proc. 28th Int. Conf. Mach. Learn. ICML 2011, 1, 833 (2011).
Jiang L, Song Z, Ge Z, Chen J, Ind. Eng. Chem. Res., 56, 26 (2017)
Heo S, Lee JH, Processes, 7, 7 (2019)
Zhang ZH, Jiang T, Li SH, Yang YP, J. Process Control, 64, 49 (2018)
Yu W, Zhao C, IEEE Trans. Control Syst. Technol., 1 (2019).
Zhao H, Chemom. Intell. Lab. Syst., 176, 11 (2018)
Simard PY, Steinkraus D, Platt JC, Proc. Int. Conf. Doc. Anal. Recognition, ICDAR, 958 (2003).
Han H, Wang WY, Mao BH, in Lecture Notes in Computer Science, 3644, 878 (2005).
Wong SC, Mcdonnell MD, 2016 Int. Conf. Digit. Image Comput. Tech. Appl., 1 (2016).
Devries T, Taylor GW, arXiv Prepr. arXiv1702.05538, 1 (2017).
Wan Z, Zhang Y, He H, IEEE, 1 (2017).
Etworks N, Storkey A, Edwards H, arXiv preprint arXiv:1711.04340, 1 (2017).
Jorge J, Paredes R, Sanchez JA, Bened M, VISIGRAPP (5:VISAPP), 96 (2018).
Hsu WN, Zhang Y, Glass J, IEEE Autom. Speech Recognit. Underst. Work., 1, 16 (2017).
Gao X, Deng F, Yue X, Neurocomputing, 396, 487 (2019)
Lim SK, Loo Y, Tran N, Cheung N, Roig G, Elovici Y, 2018 IEEE Int. Conf. Data Min., 1122 (2018).
Mellon C, Berkeley UC, arXiv preprint arXiv:1606.05908, 1 (2016).
Krogh A, Hertz JA, Adv. Neural Inf. Process. Syst. (1992).
Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R, J. Mach. Learn. Res., 15(1), 1929 (2014)
Ioffe S, Szegedy C, Int. Conf. Mach. Learn. (2015).
Han S, Pool J, Tran J, Dally WJ, arXiv preprint arXiv:1506.02626 (2015).
Heo S, Lee JH, Comput. Chem. Eng., 127, 1 (2019)
Baldi P, Hornik K, Neural Networks, 2(1), 53 (1989)
Zhao S, Song J, Ermon S, Proc. AAAI Conf. Artif. Intell., 33(1), 5885 (2019)
Downs JJ, Company EC, Comput. Chem. Eng., 17, 3 (1993)
Bathelt A, Ricker NL, Jelali M, IFAC-PapersOnLine, 48(8), 309 (2014).
Ricker NL, J. Process Control, 6(4), 205 (1996)
Lee H, Kim C, Lim S, Min J, Comput. Chem. Eng., 142, 107064 (2020)
Venkatasubramanian V, Rengaswamy R, Kavuri SN, Yin K, Comput. Chem. Eng., 27(3), 327 (2003)
Samuel RT, Cao Y, Syst. Sci. Control Eng., 4(1), 165 (2016)
Olson DL, Delen D, Advanced data mining techniques, Springer, December (2013).
Kim C, Lee H, Kim K, Lee Y, Lee WB, Ind. Eng. Chem. Res., 57(39), 13144 (2018)
Makhzani A, Frey B, Goodfellow I, arXiv Prepr. arXiv1511.05644 (2014).