Prediction-based diagnosis and loss prevention using qualitative multi-scale models

  • Authors:

  • E. Németh;R. Lakner;K. M. Hangos;I. T. Cameron

  • Affiliations:

  • Department of Computer Science, University of Pannonia, Veszprém, Hungary and Systems and Control Laboratory, Computer and Automation Research Institute, Hungarian Academy of Sciences, Kende ...;Department of Computer Science, University of Pannonia, Veszprém, Hungary;Systems and Control Laboratory, Computer and Automation Research Institute, Hungarian Academy of Sciences, Kende u.13-17, 1111 Budapest, Hungary;Division of Chemical Engineering, School of Engineering, The University of Queensland, Brisbane 4072, Australia

  • Venue:
  • Information Sciences: an International Journal
  • Year:
  • 2007

Quantified Score

Hi-index 0.07

Visualization

Abstract

A prototype prediction based intelligent diagnostic system that is capable of integrating qualitative and quantitative process models and operational experience in the form of HAZOP result tables is proposed in this paper. The diagnostic system utilizes Gensym's real time G2 expert system software. Its diagnostic ''cause-effect'' rules and possible actions (suggestions) are extracted from the results of standard HAZOP analysis. The knowledge base of the system is organized in a hierarchical way following the hierarchy levels of a multi-scale model of the process system. This supports focusing used by fault detection and loss prevention and thus decomposes the otherwise computationally hard problem. Prediction by simplified dynamic models are used to reduce ambiguity in case of multiple possible causes and/or multiple possible mitigating actions. The system is illustrated on the example of a commercial fertilizer granulator circuit using a simulation test bed.