Identification of faulty sensors using principal component analysis

Dunia, Ricardo; Qin, S. Joe; Edgar, Thomas F.; McAvoy, Thomas J.

doi:10.1002/aic.690421011

Cited by 481 publications

(273 citation statements)

References 17 publications

Supporting

Mentioning

256

Contrasting

Unclassified

Order By: Relevance

“…Neben Mittelwert-und Varianzschätzungen und weiteren stochastischen Methoden kommen dabei häufig Verfahren aus dem Bereich der Mustererkennung zum Einsatz, insbesondere für die Isolation von Fehlern. Neben der sogenannten Principal Component Analysis [61] werden dabei Support Vector Ma- chines [256], statische Neuro-Fuzzy-Systeme [166] und fortgeschrittene Clustering-Algorithmen [10] angewendet. Der zusätzliche Aufwand wird also in den Softwareteil verlagert, was gegenüber redundanter Hardware wesentlich kostengünstiger ist.…”

Section: Signalbasierte Fehlerdiagnoseunclassified

Entwurf robuster modellbasierter Fehlerisolationsfilter

Wahrburg

2014

At - Automatisierungstechnik

View full text Add to dashboard Cite

Section: Signalbasierte Fehlerdiagnoseunclassified

Entwurf robuster modellbasierter Fehlerisolationsfilter

Wahrburg

2014

At - Automatisierungstechnik

View full text Add to dashboard Cite

“…Neural networks are an important class of non-statistical classifiers (Leonard and Kramer, 1990;Kavuri and Venkatasubramanian, 1994). PCA/PLS and statistical pattern classifiers form a major component of statistical feature extraction methods (Kramer, 1991;Nomikos and MacGregor, 1994;Dong and McAvoy, 1996;Dunia et al, 1996).…”

Section: Process History-based Methodsmentioning

confidence: 99%

Prognostic and diagnostic monitoring of complex systems for product lifecycle management: Challenges and opportunities

Venkatasubramanian

2005

Computers & Chemical Engineering

View full text Add to dashboard Cite

Modern technological advances have resulted in a myriad of complex systems, processes and products. These increasingly complicated processes, systems and products pose considerable challenges in their design, analysis, manufacturing and management for successful operation and use over their life cycles. In the process industries, for example, the maintenance and management of complex process equipment and processes, and their integrated operation, play a cruicial role in ensuring the safety of plant personnel and the environment as well as the timely delivery of quality products. In the discrete parts industries, such as the auto industry, many product malfunctions are due to unanticipated dynamic interactions, due to repeated use or misuse of components. These interactions thrive in complex systems when the combined effects of uncertainty and operational adversity are not properly addressed either in design or in operation. Given the size, scope, and complexity of the systems and inretactions it is becoming increasingly difficult for plant personnel to anticipate, diagnose and control serious abnormal events in a timely manner. Hence, it should come as no surprise that human operators tend to make erroneous decisions and take actions which make matters even worse, as reported in the literature. All these cost the companies and consumers in billions of dollars every year in product lifecycle management. Businesses and federal organizations are increasingly required to manage their entire products' life cycles to avoid costly failure or degradation in performance through service/maintenance, more robust design and control, and so on. These product life cycle management (PLM) issues present us with both major challenges and opportunities. There exist considerable incentives in developing appropriate prognostic and diagnostic methodologies for monitoring, analyzing, interpreting, and controlling such abnormal events in complex systems and processes. People in the process and product industries view this as the next major challenge in control systems research and application. In this paper, we will present an overview of these challenges and the opportunities. Recent progress has promising implications on the use of intelligent systems for product lifecycle management applications in the chemical, petrochemical, pharmaceutical and discrete parts manufacturing industries.

show abstract

“…Some examples of typical faults for feedback controllers are a burned-out thermocouple, a broken transducer or a stuck valve 14 . For sensors such as a temperature or a flow measurement, the most typical fault types are a bias fault, a complete failure, a drifting fault and a precision degradation fault 15 . Therefore, increasing the robustness of individual control system components may not be sufficient to maintain a high level of control performance.…”

Section: Data-based Fault-tolerant Model Predictive Controlmentioning

confidence: 99%

Data-Based, Fault-Tolerant Model Predictive Control of a Complex Industrial Dearomatization Process

Kettunen

Jämsä‐Jounela

2011

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Abstract:The main focus of this paper is on the development of an active data-based fault-tolerant model predictive controller (FTMPC) for an industrial dearomatization process. Three different fault-tolerant control (FTC) strategies are presented; these comprise data-based fault detection and diagnosis (FDD) methods and fault accommodation-and controller reconfiguration-based FTC methods. These three strategies are tested with the simulated industrial dearomatization process. According to the validation and performance testing, the FTMPC performs efficiently and detects and prevents the effects of the most common faults in the analyser, flow and temperature measurements as well as the controller actuators. The reliability of the model predictive controller (MPC) is increased and the profitability is enhanced due to the lower off-spec production.2

show abstract

Identification of faulty sensors using principal component analysis

Cited by 481 publications

References 17 publications

Entwurf robuster modellbasierter Fehlerisolationsfilter

Entwurf robuster modellbasierter Fehlerisolationsfilter

Prognostic and diagnostic monitoring of complex systems for product lifecycle management: Challenges and opportunities

Data-Based, Fault-Tolerant Model Predictive Control of a Complex Industrial Dearomatization Process

Contact Info

Product

Resources

About