Diagnosis of Fluid Leaks in Pipelines Using Dynamic PCA

被引：27

作者：

Santos-Ruiz, I. ^{[1
]}

Lopez-Estrada, F. R. ^{[1
]}

Puig, V. ^{[2
]}

Perez-Perez, E. J. ^{[1
]}

Mina-Antonio, J. D. ^{[3
]}

Valencia-Palomo, G. ^{[4
]}

机构：

[1] TURIX Dynam Diag & Control Grp, Tecnol Nacl Mexico, Inst Tecnol Tuxtla Gutierrez, Carr Panamer Km 1080, Tuxtla Gutierrez 29050, Chiapas, Mexico

[2] UPC, Dept Automat Control ESAII, Rambla de St Nebridi 10, Terrassa 08222, Spain

[3] CENIDET, Tecnol Nacl Mexico, Interior Internado Palmira S-N, Cuernavaca 62490, Morelos, Mexico

[4] Inst Tecnol Hermosillo, Tecnol Nacl Mexico, Ave Tecnol & Periferico Poniente S-N, Hermosillo 83170, Sonora, Mexico

来源：

IFAC PAPERSONLINE | 2018年 / 51卷 / 24期

关键词：

Fault diagnosis; Principal component analysis; Data-driven fault detection; Pipelines; Fluid leaks; PRINCIPAL COMPONENT ANALYSIS;

D O I：

10.1016/j.ifacol.2018.09.604

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

In this paper, a data-driven system based on PCA is described to detect and quantify fluid leaks in an experimental pipeline. A dynamic PCA implementation (DPCA) was used to capture the process dynamics because the system variables are time-correlated. To detect leaks online, the Hotelling's T-2 statistic and the squared prediction error (SPE) were used as residuals, which are compared against statistically defined thresholds from a set of training data. To determine the number of delays to be included in the DPCA model as well as the number of principal components to be used, a tuning process was executed to find the residual with the optimal number of delays and components that showed the best correlation between the residuals and the leakage size. This allowed the construction of a regression model to estimate the flow rate of the leaks directly from the residual. (C) 2018, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.

引用

页码：373 / 380

页数：8

共 21 条

[11]

Hernandez N, 2017, TECHNICAL REPORT

[12] CONTROL PROCEDURES FOR RESIDUALS ASSOCIATED WITH PRINCIPAL COMPONENT ANALYSIS [J].

JACKSON, JE ;

MUDHOLKAR, GS .

TECHNOMETRICS, 1979, 21 (03) :341-349

[13] Disturbance detection and isolation by dynamic principal component analysis [J].

Ku, WF ;

Storer, RH ;

Georgakis, C .

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 1995, 30 (01) :179-196

[14] Analysis and comparison of an improved unreconstructed variance criterion to other criteria for estimating the dimension of PCA model [J].

Mnassri, Baligh ;

El Adel, El Mostafa ;

Ouladsine, Mustapha .

JOURNAL OF PROCESS CONTROL, 2016, 44 :207-223

[15] Reconstruction-based contribution approaches for improved fault diagnosis using principal component analysis [J].

Mnassri, Baligh ;

El Adel, El Mostafa ;

Ouladsine, Mustapha .

JOURNAL OF PROCESS CONTROL, 2015, 33 :60-76

[16] Generalization and analysis of sufficient conditions for PCA-based fault detectability and isolability [J].

Mnassri, Baligh ;

El Adel, El Mostafa ;

Ouladsine, Mustapha .

ANNUAL REVIEWS IN CONTROL, 2013, 37 (01) :154-162

[17]

OECD, 2016, OECD STUD WAT, DOI [10.1787/9789264251090, DOI 10.1787/9789264251090]

[18]

Oswald U, 2017, REV MEX CIENC POLIT, V62, P155, DOI 10.1016/S0185-1918(17)30020-X

[19] Statistical process monitoring: basics and beyond [J].

Qin, SJ .

JOURNAL OF CHEMOMETRICS, 2003, 17 (8-9) :480-502

[20]

Quinones Grueiro M., 2016, C NAC CONTR AUT CNCA, P276

← 1 2 3 →