University of Tasmania
Browse
whole_GalePaulH1998_thesis.pdf (14.03 MB)

An expert system based aid for anode production at an aluminium smelter

Download (14.03 MB)
thesis
posted on 2023-05-26, 22:55 authored by Gale, Paul H
In the smelting of aluminium, a high current is passed between an anode and a cathode across a molten bath of electrolyte. As the anode is consumed in the electrolytic process, the manufacture of anodes represents a high operational cost. Better control of the anode production process could lead to a reduction in the variability of anode qualities, improved anode performance and a resulting longer anode life with a concomitant reduction in operational costs. The production process of anodes is controlled by operators through a Supervisory Control and Data Acquisition (SCADA) system that continuously polls the front line program logic control of the anode plant. Information relating to potential problems or causes of existing problems is available, which an experienced operator can obtain by querying the SCADA system. As there is too much data for the operator to assimilate to properly control the process, the operator tends to rely on the control system itself to control the process within its own limitations, since more pressing problems usually divert him from this fine tuning task. Thus, anodes are produced that generally meet specification, but with some degree of variability. The work presented in this thesis addresses the use of artificial intelligence techniques to develop a decision support system for use by the anode plant operators. Research focused on the following issues: ‚Äö identification of the process parameters that critically affect an anode's properties; ‚Äö determination of the problems, and their causes, that affect those process parameters; ‚Äö determination of the most appropriate way to present information to the operator; ‚Äö development of the required software; ‚Äö determination of the best methods, in terms of speed and PC memory, of exchanging data with SCADA; ‚Äö development of a robust application that runs continuously, without failure, for a ten day production run. The developed application, based on Level5 Object software, continuously interrogates the process control system and provides on-line analysis to operators of ten process variables identified as critically affecting anode properties. For each process variable, are displayed a triplet of statistical process control charts and messages advising of trends, potential problems and possible causes of those problems. Utilising hypertext, the operator can view and print out a Standard Operating Procedure to rectify said problem. Properly used, the monitor provides the operator with a high level tool to enable anodes to be made to a consistent quality within tight process limits. The research has established the necessary techniques to design a system to continuously oversee and monitor a complex production process, to analyse large amounts of data in real time and to provide meaningful and timely information to the operator to enable the application of the necessary corrective action. The developed system has been installed for live monitoring of the process by technical personnel during 1996 and for operator use in 1997. Two refereed papers have been published and presented at international conferences.

History

Publication status

  • Unpublished

Rights statement

Copyright 1997 the Author - The University is continuing to endeavour to trace the copyright owner(s) and in the meantime this item has been reproduced here in good faith. We would be pleased to hear from the copyright owner(s). Addresses the use of artificial intelligence techniques to develop a decision support system for use by anode plant operators. The developed application, based on Level15 Object software, continuously interrogates the process control system and provides on-line analysis to operators of ten process variables identified as critically affecting anode properties. No access until 1.6.1998. Thesis (M.Eng.Sc.)--University of Tasmania, 1998. Includes bibliographical references. Addresses the use of artificial intelligence techniques to develop a decision support system for use by anode plant operators. The developed application, based on Level15 Object software, continuously interrogates the process control system and provides on-line analysis to operators of ten process variables identified as critically affecting anode properties

Repository Status

  • Open

Usage metrics

    Thesis collection

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC