Loading [a11y]/accessibility-menu.js
Maximizing Anomaly Detection Performance Using Latent Variable Models in Industrial Systems | IEEE Journals & Magazine | IEEE Xplore

Maximizing Anomaly Detection Performance Using Latent Variable Models in Industrial Systems


Abstract:

In conventional process monitoring, a latent variable model (LVM) is first learned in offline training and the statistics related to extracted latent features and residua...Show More

Abstract:

In conventional process monitoring, a latent variable model (LVM) is first learned in offline training and the statistics related to extracted latent features and residuals are then used for online monitoring. However, such a practice ignores the dynamic interaction between modeling and monitoring, rendering useful online samples underexplored. This study proposes a novel LVMs-based monitoring framework that exploits the interaction using a weighting strategy and the maximum likelihood method to improve the monitoring performance with online information. The key idea is to integrate a weighting vector to components which contribute to the fault detection indices for more effective online fault information extraction. We use the maximum likelihood ratio to optimize the weighting vector and construct a new fault detection index accordingly. Case studies on a numerical example and a three-phase flow facility demonstrate the effectiveness of our approach. Note to Practitioners—A large number of anomaly detection methods in industrial systems has emerged in recent years. Latent variable models are the dominant branch of anomaly detection methods with substantial research and practice. However, fault detection performance is still unexpected especially for some minor faults that cause underwhelming fluctuation. Based on LVM models, we investigate the performance maximization method of industrial fault detection. The main mechanism is a weighting strategy that connects the normal data and the online sample to be monitored. We do not attach any additional conditions besides the existing requirements for LVM models. Also, it is the class of LVM methods that can be benefited from this novel strategy rather than a specific approach, which has been verified using principal component analysis and canonical correlation analysis. Moreover, the fault detection performance has boosted for all kinds of fault types. Notice we use general linear LVM models for deriving the methodology....
Published in: IEEE Transactions on Automation Science and Engineering ( Volume: 21, Issue: 3, July 2024)
Page(s): 4808 - 4816
Date of Publication: 14 August 2023

ISSN Information:

Funding Agency:

References is not available for this document.

I. Introduction

Data-driven process monitoring is gaining increasing attention, owing to the development of large-scale and complex modern industrial processes [1], [2], [3]. Operators can collect and store large amounts of data via the burgeoning distributed control system (DCS) to extract valuable features for distinguishing between normal and faulty status [4], [5]. Thus, data-driven process monitoring is more versatile and convenient for industrial processes,in contrast with the model-based counterpart that depends on accurate process mechanisms.

Select All
1.
D. Wu, D. Zhou, J. Zhang, and M. Chen, “Multimode process monitoring based on fault dependent variable selection and moving window-negative log likelihood probability,” Comput. Chem. Eng., vol. 136, May 2020, Art. no. 106787.
2.
J. Huang and X. Yan, “Quality-driven principal component analysis combined with kernel least squares for multivariate statistical process monitoring,” IEEE Trans. Control Syst. Technol., vol. 27, no. 6, pp. 2688–2695, Nov. 2019.
3.
J. Huang, O. K. Ersoy, and X. Yan, “Fault detection in dynamic plant-wide process by multi-block slow feature analysis and support vector data description,” ISA Trans., vol. 85, pp. 119–128, Feb. 2019.
4.
L. Qin, C. Tong, T. Lan, and Y. Chen, “Statistical process monitoring based on just-in-time feature analysis,” Control Eng. Pract., vol. 115, Oct. 2021, Art. no. 104889.
5.
X. Ma, Y. Si, Y. Qin, and Y. Wang, “Fault detection for dynamic processes based on recursive innovational component statistical analysis,” IEEE Trans. Autom. Sci. Eng., vol. 20, no. 1, pp. 310–319, Jan. 2023.
6.
J. Zhang, H. Chen, S. Chen, and X. Hong, “An improved mixture of probabilistic PCA for nonlinear data-driven process monitoring,” IEEE Trans. Cybern., vol. 49, no. 1, pp. 198–210, Jan. 2019.
7.
K. Wang, J. Chen, and Z. Song, “Concurrent fault detection and anomaly location in closed-loop dynamic systems with measured disturbances,” IEEE Trans. Autom. Sci. Eng., vol. 16, no. 3, pp. 1033–1045, Jul. 2019.
8.
F. Destro, P. Facco, S. G. Muñoz, F. Bezzo, and M. Barolo, “A hybrid framework for process monitoring: Enhancing data-driven methodologies with state and parameter estimation,” J. Process Control, vol. 92, pp. 333–351, Aug. 2020.
9.
Y. Qin, Y. Yan, H. Ji, and Y. Wang, “Recursive correlative statistical analysis method with sliding windows for incipient fault detection,” IEEE Trans. Ind. Electron., vol. 69, no. 4, pp. 4185–4194, Apr. 2022.
10.
S. J. Qin, “Statistical process monitoring: Basics and beyond,” J. Chemometrics, vol. 17, nos. 8–9, pp. 480–502, 2003.
11.
J. Li, D. Ding, and F. Tsung, “Directional PCA for fast detection and accurate diagnosis: A unified framework,” IEEE Trans. Cybern., vol. 52, no. 11, pp. 11362–11372, Nov. 2022.
12.
H. Lahdhiri and O. Taouali, “Reduced rank KPCA based on GLRT chart for sensor fault detection in nonlinear chemical process,” Measurement, vol. 169, Feb. 2021, Art. no. 108342.
13.
R. Fazai, “Multiscale Gaussian process regression-based generalized likelihood ratio test for fault detection in water distribution networks,” Eng. Appl. Artif. Intell., vol. 85, pp. 474–491, Oct. 2019.
14.
S. Ding, “Data-driven design of fault-tolerant control systems,” IFAC Proc. Volumes, vol. 45, pp. 1323–1328, Jan. 2012.
15.
J. V. Kresta, J. F. Macgregor, and T. E. Marlin, “Multivariate statistical monitoring of process operating performance,” Can. J. Chem. Eng., vol. 69, no. 1, pp. 35–47, Feb. 1991.
16.
W. Ku, R. H. Storer, and C. Georgakis, “Disturbance detection and isolation by dynamic principal component analysis,” Chemometric Intell. Lab. Syst., vol. 30, no. 1, pp. 179–196, Nov. 1995.
17.
I. Jaffel, O. Taouali, M. F. Harkat, and H. Messaoud, “Moving window KPCA with reduced complexity for nonlinear dynamic process monitoring,” ISA Trans., vol. 64, pp. 184–192, Sep. 2016.
18.
Z. Ge and Z. Song, “Performance-driven ensemble learning ICA model for improved non-Gaussian process monitoring,” Chemometric Intell. Lab. Syst., vol. 123, pp. 1–8, Apr. 2013.
19.
J. Dong, K. Zhang, Y. Huang, G. Li, and K. Peng, “Adaptive total PLS based quality-relevant process monitoring with application to the Tennessee eastman process,” Neurocomputing, vol. 154, pp. 77–85, Apr. 2015.
20.
Z. Chen, S. X. Ding, T. Peng, C. Yang, and W. Gui, “Fault detection for non-Gaussian processes using generalized canonical correlation analysis and randomized algorithms,” IEEE Trans. Ind. Electron., vol. 65, no. 2, pp. 1559–1567, Feb. 2018.
21.
B. Guo, C. Zhang, J. Liu, and X. Ma, “Improving text classification with weighted word embeddings via a multi-channel TextCNN model,” Neurocomputing, vol. 363, pp. 366–374, Oct. 2019.
22.
J. Li, C. Wu, R. Song, Y. Li, and F. Liu, “Adaptive weighted attention network with camera spectral sensitivity prior for spectral reconstruction from RGB images,” in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Workshops (CVPRW), Jun. 2020, pp. 1894–1903.
23.
X. Yuan, B. Huang, Y. Wang, C. Yang, and W. Gui, “Deep learning-based feature representation and its application for soft sensor modeling with variable-wise weighted SAE,” IEEE Trans. Ind. Informat., vol. 14, no. 7, pp. 3235–3243, Jul. 2018.
24.
K. Li, C. Qiu, Y. Lin, M. Chen, X. Jia, and B. Li, “A weighted adaptive transfer learning for tool tip dynamics prediction of different machine tools,” Comput. Ind. Eng., vol. 169, Jul. 2022, Art. no. 108273.
25.
A. Ahrari, S. Elsayed, R. Sarker, D. Essam, and C. A. C. Coello, “Weighted pointwise prediction method for dynamic multiobjective optimization,” Inf. Sci., vol. 546, pp. 349–367, Feb. 2021.
26.
Q. Jiang and X. Yan, “Chemical processes monitoring based on weighted principal component analysis and its application,” Chemometric Intell. Lab. Syst., vol. 119, pp. 11–20, Oct. 2012.
27.
D. Wu, D. Zhou, and M. Chen, “Performance-driven component selection in the framework of PCA for process monitoring: A dynamic selection approach,” IEEE Trans. Control Syst. Technol., vol. 30, no. 3, pp. 1171–1185, May 2022.
28.
C. Tong, T. Lan, and X. Shi, “Fault detection and diagnosis of dynamic processes using weighted dynamic decentralized PCA approach,” Chemometric Intell. Lab. Syst., vol. 161, pp. 34–42, Feb. 2017.
29.
Y. Wang, D. Zhou, M. Chen, and M. Wang, “Weighted part mutual information related component analysis for quality-related process monitoring,” J. Process Control, vol. 88, pp. 111–123, Apr. 2020.
30.
M.-F. Harkat, M. Mansouri, M. N. Nounou, and H. N. Nounou, “Fault detection of uncertain chemical processes using interval partial least squares-based generalized likelihood ratio test,” Inf. Sci., vol. 490, pp. 265–284, Jul. 2019.

Contact IEEE to Subscribe

References

References is not available for this document.