A process monitoring method for autoregressive-dynamic inner total latent structure projection

doi:10.23919/JSEE.2024.000105

Journal of Systems Engineering and Electronics ›› 2024, Vol. 35 ›› Issue (5): 1326-1336.doi: 10.23919/JSEE.2024.000105

• • 上一篇

收稿日期:2022-11-07 接受日期:2023-11-24 出版日期:2024-10-18 发布日期:2024-11-06

A process monitoring method for autoregressive-dynamic inner total latent structure projection

Yalin CHEN¹^,²(), Xiangyu KONG¹^,*(), Jiayu LUO¹()

¹ School of Missile Engineering, Rocket Force University of Engineering, Xi’an 710025, China
² AVIC Chengdu Caic Electronics Co., Ltd, Chengdu 610091, China

Received:2022-11-07 Accepted:2023-11-24 Online:2024-10-18 Published:2024-11-06
Contact: Xiangyu KONG E-mail:cyl2318959445@163.com;xiangyukong01@163.com;540629964@qq.com
About author:
CHEN Yalin was born in 1997. She received her B.E. degree from Henan Institute of Technology, Xinxiang, China, in 2020, and M.E. degree in control engineering from Rocket Force University of Engineering, Xi’an, China, in 2005. Her research interests include dynamic feature extraction and multivariate statistical process monitoring. E-mail: cyl2318959445@163.com

KONG Xiangyu was born in 1967. He received his B.E. degree from the Beijing Institute of Technology, Beijing, China, in 1990, M.E. degree from the High-Tech Institute of Xi’an, Xi’an, China, in 2000, and Ph.D. degree in control science and engineering from Xi’an Jiaotong University, Xi’an, in 2005. He is currently a professor with the High-Tech Institute of Xi’an. His research interests include adaptive signal processing, system modeling, and fault diagnosis. E-mail: xiangyukong01@163.com

LUO Jiayu was born in 1994. He received his B.E. degree from Hunan University, Changsha, China, in 2017, and M.E. degree from the High-Tech Institute of Xi’an, Xi’an, China, in 2019, where he is currently pursuing his Ph.D. degree with the High-Tech Institute of Xi’an. His research interests include feature extraction and process monitoring. E-mail: 540629964@qq.com
Supported by:
This work was supported by the National Natural Science Foundation of China (62273354; 61673387; 61833016).

摘要/Abstract

Abstract:

As a dynamic projection to latent structures (PLS) method with a good output prediction ability, dynamic inner PLS (DiPLS) is widely used in the prediction of key performance indicators. However, due to the oblique decomposition of the input space by DiPLS, there are false alarms in the actual industrial process during fault detection. To address the above problems, a dynamic modeling method based on autoregressive-dynamic inner total PLS (AR-DiTPLS) is proposed. The method first uses the regression relation matrix to decompose the input space orthogonally, which reduces useless information for the prediction output in the quality-related dynamic subspace. Then, a vector autoregressive model (VAR) is constructed for the prediction score to separate dynamic information and static information. Based on the VAR model, appropriate statistical indicators are further constructed for online monitoring, which reduces the occurrence of false alarms. The effectiveness of the method is verified by a Tennessee-Eastman industrial simulation process and a three-phase flow system.

Key words: dynamic characteristic, fault detection, feature extraction, process monitoring, projection to latent structure (PLS), quality-related, spatial partitioning

. [J]. Journal of Systems Engineering and Electronics, 2024, 35(5): 1326-1336.

Yalin CHEN, Xiangyu KONG, Jiayu LUO. A process monitoring method for autoregressive-dynamic inner total latent structure projection[J]. Journal of Systems Engineering and Electronics, 2024, 35(5): 1326-1336.

图/表 13

参考文献 35

1	YAO L, SHAO W M, GE Z Q, Hierarchical quality monitoring for large-scale industrial plants with big process data. IEEE Trans. on Neural Networks and Learning Systems, 2021, 32(8): 3330−3341.
2	KONG X Y, YANG Z Y, LUO J Y, et al. Extraction of reduced fault subspace based on KDICA and its application in fault diagnosis. IEEE Trans. on Instrumentation and Measurement, 2022, 71: 1−12.
3	ZHU Q Q Dynamic autoregressive partial least squares for supervised modeling. IFAC-PapersOnLine, 2021, 54 (7): 234- 239. doi: 10.1016/j.ifacol.2021.08.364
4	XU B, ZHU Q Q Online quality-relevant monitoring with dynamic weighted partial least squares. Industrial & Engineering Chemistry Research, 2020, 59 (48): 21124- 21132.
5	HU C H, LUO J Y, KONG X Y, et al Novel fault subspace extraction methods for the reconstruction-based fault diagnosis. Journal of Process Control, 2021, 105, 129- 140. doi: 10.1016/j.jprocont.2021.07.008
6	CHEN C, WANG Y J, ZHANG Y, et al Indoor positioning algorithm based on nonlinear PLS integrated with RVM. IEEE Sensors Journal, 2018, 18 (2): 660- 668. doi: 10.1109/JSEN.2017.2772798
7	ZHOU K, LI D J, CUI A J, et al Sparse flight spotlight mode 3-D imaging of spaceborne SAR based on sparse spectrum and principal component analysis. Journal of Systems Engineering and Electronics, 2021, 32 (5): 1143- 1151. doi: 10.23919/JSEE.2021.000098
8	WANG Y, ZHENG J H, ZENG J Z Dynamic shared segment protection algorithm with differentiated reliability in GMPLS networks. Journal of Systems Engineering and Electronics, 2009, 20 (1): 178- 184.
9	MA W N, ZHAO F, LI X, et al Joint optimization of inspection-based and age-based preventive maintenance and spare ordering policies for single-unit systems. Journal of Systems Engineering and Electronics, 2022, 33 (5): 1268- 1280. doi: 10.23919/JSEE.2022.000120
10	ZHANG L P, SUN L S, LI W J, et al A joint Bayesian framework based on partial least squares discriminant analysis for finger vein recognition. IEEE Sensors Journal, 2022, 22 (1): 785- 794.
11	FAISAL A I, MONDAL T, COWAN D, et al Characterization of knee and gait features from a wearable tele-health monitoring system. IEEE Sensors Journal, 2022, 22 (6): 4741- 4753. doi: 10.1109/JSEN.2022.3146617
12	LUO J Y, KONG Y X, HU C H, et al Key-performance- indicators-related fault subspace extraction for the reconstruction-based fault diagnosis. Measurement, 2021, 186 (8): 110- 119.
13	SI Y, WANG Y, ZHOU D H Key-performance-indicator- related process monitoring based on improved kernel partial least squares. IEEE Trans. on Industrial Electronics, 2021, 68 (3): 2626- 2636. doi: 10.1109/TIE.2020.2972472
14	HU J, WEN C L, LI P, et al Direct projection to latent variable space for fault detection. Journal of the Franklin Institute, 2014, 351 (3): 1226- 1250.
15	ZHOU D H, LI G, QIN S J Total projection to latent structures for process monitoring. AIChE Journal, 2010, 56 (1): 168- 178. doi: 10.1002/aic.11977
16	QIN S J, ZHENG Y Quality-relevant and process-relevant fault monitoring with concurrent projection to latent structures. AIChE Journal, 2013, 59 (2): 496- 504. doi: 10.1002/aic.13959
17	YIN S, DING S X, ZHANG P, et al Study on modifications of PLS approach for process monitoring. IFAC Proceedings Volumes, 2011, 44 (1): 12389- 12394. doi: 10.3182/20110828-6-IT-1002.02876
18	PENG K X, ZHANG K, YOU B, et al Quality-relevant fault monitoring based on efficient projection to latent structures with application to hot strip mill process. IET Control Theory & Applications, 2015, 9 (7): 1134- 1145.
19	KASPAR M H, RAY W H Dynamic PLS modelling for process control. Chemical Engineering Science, 1993, 48 (20): 3447- 3461. doi: 10.1016/0009-2509(93)85001-6
20	QIN S J, MCAVOY T J. A data-based process modeling approach and its applications-sciencedirect. Dynamics and Control of Chemical Reactors, Distillation Columns and Batch Processes. 1993, 5(25): 93−98.
21	LAKSHMINARAYANAN S, SHAH S L, NANDAKUM- AR K Modeling and control of multivariable processes: Dynamic PLS approach. AIChE Journal, 1997, 43 (9): 2307- 2322. doi: 10.1002/aic.690430916
22	LI G, LIU B S, QIN S J, et al Quality relevant data-driven modeling and monitoring of multivariate dynamic processes: the dynamic T-PLS approach. IEEE Trans. on Neural Networks, 2011, 22 (12): 2262- 2271. doi: 10.1109/TNN.2011.2165853
23	DONG Y N, QIN S J Dynamic-inner partial least squares for dynamic data modeling. IFAC-PapersOnLine, 2015, 48 (8): 117- 122. doi: 10.1016/j.ifacol.2015.08.167
24	DONG Y N, QIN S J Regression on dynamic PLS structures for supervised learning of dynamic data. Journal of Process Control, 2018, 68, 64- 72. doi: 10.1016/j.jprocont.2018.04.006
25	YUAN Z Y, MA X, QIN Y H, et al. Two-step partial least squares for monitoring processes. Proc. of the 33rd Chinese Control and Decision Conference 2021, 2808−2813.
26	XU B, ZHU Q Q Concurrent auto-regressive latent variable model for dynamic anomaly detection-sciencedirect. Journal of Process Control, 2021, 108 (3): 1- 11.
27	ZHU Q Q Latent variable regression for supervised modeling and monitoring. IEEE/CAA Journal of Automatica Sinica, 2020, 7 (3): 800- 811. doi: 10.1109/JAS.2020.1003153
28	LI G, LIU B, ZHOU D H A new method of dynamic latent-variable modeling for process monitoring. IEEE Trans. on Industrial Electronics, 2014, 61 (11): 6438- 6445. doi: 10.1109/TIE.2014.2301761
29	DOWNS J J, VOGEL E F A plant-wide industrial process control problem. Computers & Chemical Engineering, 1993, 17 (3): 245- 255.
30	HU C H, LUO J Y, KONG X Y, et al Novel fault subspace extraction methods for the reconstruction-based fault diagnosis. Journal of Process Control, 2021, 105 (1): 129- 140.
31	CAPACI F, VANHATALO, KULAHCI E M, et al The revised Tennessee Eastman process simulator as testbed for SPC and DoE methods. Qual. Eng, 2019, 31 (2): 221- 229.
32	DHIBI K, FEZAI R, MANSOURI M, et al A hybrid approach for process monitoring: improving data-driven methodologies with dataset size reduction and interval-Valued representation. IEEE Sensors Journal, 2020, 20 (17): 10228- 10239. doi: 10.1109/JSEN.2020.2991508
33	JIAO J F, YU H, WANG G A quality-related fault detection approach based on dynamic least squares for process monitoring. IEEE Trans. on Industrial Electronics, 2016, 63 (4): 2625- 2632.
34	RUIZ-CARCEL C, CAO Y, MBA D, et al Statistical process monitoring of a multiphase flow facility. Control Engineering Practice, 2015, 42, 74- 88. doi: 10.1016/j.conengprac.2015.04.012
35	SUN C Y, YIN Y Z, KANG H B, et al. A distributed principal component regression method for quality-related fault detection and diagnosis. Information Sciences, 2022, 600: 301−302.

Subspace	Dimension	Description
$ {{\boldsymbol{S}}_y} $	${{{A}}_y}$	A predictable mass-dependent dynamic subspace in X-space
$ {{\boldsymbol{S}}_r} $	${{{A}}_r}$	Irrelevant subspaces in the predictable dynamic space of X that do not contribute to the predicted output
$ {{\boldsymbol{S}}_{{\mathrm{er}}}} $	${{A}} - {{{A}}_y} - {{{A}}_r}$	A noisy subspace in the predictable dynamic space of X
$ {{\boldsymbol{S}}_{{\mathrm{ep}}}} $	${{m}} - {{A}}$	Unpredictable subspaces in X space

Fault	DMPLS		DTPLS			AR-DiTPLS
Fault	$T_{y\_{\rm{new}}}^2$	${Q_{r\_{\rm{new}}} }$	$T_{y\_{\rm{new}}}^2$	$T_{r\_{\rm{new}}}^2$	${Q_{r\_{\rm{new}}} }$	$T_{y\_{\rm{new}}}^2$	$T_{r\_{\rm{new}}}^2$	${Q_{r\_{\rm{new}}} }$
1	84.77	98.63	99.00	99.87	99.25	99.00	34.96	99.75
2	98.25	96.75	97.87	98.75	98.38	92.1	89.47	98.62
5	99.50	60.80	15.41	35.97	13.50	29.57	14.41	24.28
6	99.75	99.63	97.37	99.62	99.38	99.62	97.74	100.0
7	56.18	35.33	28.20	100.0	100.0	39.34	22.68	100.0
8	72.78	92.01	72.18	98.12	91.25	83.08	67.42	98.00
10	80.27	26.97	10.78	31.08	08.00	36.21	07.27	31.91
12	93.38	95.38	71.55	98.62	84.38	88.97	67.92	98.37
13	83.65	90.89	88.47	95.99	94.75	91.97	81.95	94.74

Fault	DMPLS		DTPLS			AR-DiTPLS
Fault	$T_{y\_{\rm{new}}}^2$	${Q_{r\_{\rm{new}}} }$	$T_{y\_{\rm{new}}}^2$	$T_{r\_{\rm{new}}}^2$	${Q_{r\_{\rm{new}}} }$	$T_{y\_{\rm{new}}}^2$	$T_{r\_{\rm{new}}}^2$	${Q_{r\_{\rm{new}}} }$
3	4.24	2.50	9.02	17.04	4.00	3.50	1.88	1.75
4	11.36	2.12	6.77	69.67	87.88	2.38	100.0	88.74
9	4.12	2.00	9.40	14.91	5.63	1.38	2.38	1.63
11	10.61	10.24	12.16	64.66	64.75	1.63	64.41	63.08
15	2.50	4.87	10.03	16.04	5.75	3.38	2.50	5.26

Variable	Location	Specific description
$ {X_1} $/MPa	PT408	Differential pressure at the top
$ {X_2} $/(kg/s)	FT407	Top traffic
$ {X_3} $/m	LI405	Top of the stage separator
$ {X_4} $/(kg/s)	FT406	Output of top separator
$ {X_5} $/(kg/s)	LI504	Gas and liquid three-phase separator
$ {X_6} $/%	VC501	VC501 valve location
$ {X_7} $/%	VC302	VC302 valve location
$ {X_8} $/A	PO1	The size of the pump current
$ Y $/(kg/m³)	FT407	Highest density upper riser

A process monitoring method for autoregressive-dynamic inner total latent structure projection

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 35

相关文章 0

编辑推荐

Metrics

本文评价