COVID-19 Update: We are currently shipping orders daily. However, due to transit disruptions in some geographies, deliveries may be delayed. To provide all customers with timely access to content, we are offering 50% off Science and Technology Print & eBook bundle options. Terms & conditions.
On-Line Fault Detection and Supervision in the Chemical Process Industries 1998 - 1st Edition - ISBN: 9780080432335, 9780080964201

On-Line Fault Detection and Supervision in the Chemical Process Industries 1998

1st Edition

Authors: P.S. Dhurjati Sylvie Cauvin
eBook ISBN: 9780080964201
Paperback ISBN: 9780080432335
Imprint: Pergamon
Published Date: 20th October 1998
Sales tax will be calculated at check-out Price includes VAT/GST

Institutional Subscription

Secure Checkout

Personal information is secured with SSL technology.

Free Shipping

Free global shipping
No minimum order.

Table of Contents

Section headings and selected papers: Plenary Papers. Contributions from the community IT programme (P. Corsi). Trend Analysis. Process trend analysis using wavelet-based de-noising (A. Bakhtazad et al.). Observers. Non-linear observer based method for fault detection and isolation (F. Armanet et al.). Detect unexpected changes of particle size distribution in paper-making white water systems (H. Wang). Applications. Observing the sugar-beet quality using process and signal analysing methods (A. Arenz et al.). Different Approaches. A generic fault propagation modeling approach to on-line diagnosis and event correlation (G.M. Stanley, R. Vaidhyanathan). Can chemical process industry benefit from analog electronics diagnosis methods? (P. Taillibert). Neural Networks. Fault detection in paper making (a neural network approach) (Y. Bissessur et al.). Statistics/Reconciliation. Univariate and multivariate process monitoring and improvement (A. King et al.). An industrial application of principal component test to fault detection and identification (H. Tong, D. Bluck). Qualitative. Automated interpretation of PCA-based process monitoring and fault diagnosis using signed digraphs (H. Vedam, V. Venkatasubramanian). Fault diagnosis expert system with probability calculations (D. Leung, J. Romagnoli). Supervision/Control/Alarm. Early detection of alarm situations using model predictions (B.C. Juricek et al.). Poster Papers. Suboptimal conditions for leakage detectability in pipelines (C. Verde, F. Ibinarriaga). Comparison of the performances of parametric estimation, time-frequency, wavelet and segmentation with detection of abrupt changes in non-destructive evaluation (S. Femman, N.K. M'Sirdi). New intrinsic runaway criteria and their application to the model-based on-line monitoring of continuous and discontinuous reactors (T. Obertopp et al.).


The field of "On-Line Fault Detection and Supervision in the Chemical Process Industries" is relatively young. Major activity in this area has taken place only in the last fifteen years.
The goals of the first workshop in Delaware were to discuss various methodologies necessary for solving industrial problems in fault diagnosis/supervision and to encourage interactions between academia and industry. This workshop also focused on development and evaluation of methodologies for on-line fault detection and supervision in the chemical process industries. It addressed theory, application, validation, performance and evaluation of methodologies such as parameter estimation, observers, parity equations, signal analysis methods, classification, rule-based systems with probabilistic approaches, fuzzy logic and neural networks.

There are several trends that make the topic of this workshop especially relevant in today's world. The first is the tremendous advances made in automation and information technology that can potentially bring in an ever-increasing amount of information on to computer screens in the operating room of a plant. Avoiding problems of information overload and converting plant data to "on-line useful knowledge" is a key challenge. In some respects, one can draw parallels here to biological evolution where, over billions of years, human beings have evolved "mental models" to interpret the huge amount of information received through their senses. In the absence of the time advantage that evolution has had, we have to rely on methodologies such as those presented in this workshop to provide assistance to operators and engineers in interpreting plant information.

A second trend that makes this field relevant in today's world is the increasing emphasis on environment and safety. Community activism and accidents such as those in Bhopal, India have caused media spotlights to be turned on the smallest of toxic releases or loss of life due to chemical accidents. The negative publicity generated by such events as well as the need to maintain the image of an environmentally conscious company make industry more sensitive to the issues of early detection of faults.

The third trend that makes this field very relevant is that of the globalization of the world economy. Increasing globalization of the chemical process industry puts pressure on economic competitiveness and higher productivity. This implies reduced down-time due to faults, quick and flexible response of production to supply and demand changes, increasing reliance on automation and reduced personnel.


For both researchers and practitioners working within the fields of chemical, industrial, systems and control engineering.


© Pergamon 1998
20th October 1998
eBook ISBN:
Paperback ISBN:

Ratings and Reviews

About the Authors

P.S. Dhurjati

Affiliations and Expertise

Department of Chemical Engineering, University of Delaware, Newark, USA

Sylvie Cauvin

Affiliations and Expertise

IFP (Institut Français du Pétrol), France