Title: Analytical process and system design of integrated fault diagnostic system

Authors: Hossam A. Gabbar, Hanaa E. Sayed, Ajiboye S. Osunleke, Masanobu Hara

Addresses: Faculty of Energy Systems and Nuclear Science, University of Ontario Institute of Technology (UOIT), 2000 Simcoe Street North, Oshawa, ON L1H 7K4, Canada. ' Department of Intelligent Systems Engineering, Division of Industrial Innovation Science, Okayama University, 1-1 Tsushima Naka 3-chome, Okayama, 700-8530, Japan. ' Department of Intelligent Systems Engineering, Division of Industrial Innovation Science, Okayama University, 1-1 Tsushima Naka 3-chome, Okayama, 700-8530, Japan. ' Business Consulting, AspenTech Japan Co., Ltd. Kojimachi Crystal City 10F, Kojimachi, Chiyoda-ku, Tokyo, 102-0083, Japan

Abstract: There is an increasing interest to find an effective mechanism for fault diagnosis, which is essential for safe plant operation and optimised maintenance. This paper presents an integrated framework for qualitative and quantitative fault diagnosis mechanism where qualitative fault models are developed for the underlying plant process and equipment and linked with quantitative methods on the basis of trend analysis. The proposed fault diagnosis process is described using detailed activity models. The proposed integrated fault diagnosis framework is illustrated using a case study experimental plant G-Plant, which showed improved fault diagnosis capabilities in terms of root cause and consequence analysis.

Keywords: plant OOM; process OOM; object oriented modelling; fault diagnosis; fault propagation analysis; quantitative fault models; qualitative fault models; fault semantic networks; FSN; process design.

DOI: 10.1504/IJPSE.2009.028001

International Journal of Process Systems Engineering, 2009 Vol.1 No.1, pp.66 - 81

Published online: 26 Aug 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article