• DocumentCode
    2222003
  • Title

    Optimisation of a polymer plant by an embedded hybrid architecture

  • Author

    Meert, Kurt ; Van Bael, Patrick ; Rijckaert, Marcel

  • Author_Institution
    Expert Syst. Appl. Dev. Group, Katholieke Univ., Leuven, Heverlee, Belgium
  • Volume
    1
  • fYear
    1998
  • fDate
    4-8 May 1998
  • Firstpage
    719
  • Abstract
    In this paper a hybrid architecture is developed to optimise set-point changes between different polyethylene brands. This architecture combines two different AI-techniques: a recurrent neural network, which is used to model the polyethylene plant; and a simulated annealing-based optimisation algorithm. Several experiments were performed to demonstrate the capabilities of this hybrid optimisation architecture
  • Keywords
    chemical variables control; neurocontrollers; plastics industry; process control; real-time systems; recurrent neural nets; simulated annealing; embedded hybrid architecture; optimisation; plastics industry; polyethylene plant; polymer; process control; recurrent neural network; set-point control; simulated annealing; Expert systems; Inductors; Neural networks; Optimal control; Polyethylene; Polymers; Predictive models; Production; Recurrent neural networks; Simulated annealing;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on
  • Conference_Location
    Anchorage, AK
  • ISSN
    1098-7576
  • Print_ISBN
    0-7803-4859-1
  • Type

    conf

  • DOI
    10.1109/IJCNN.1998.682369
  • Filename
    682369