• DocumentCode
    420649
  • Title

    Error treatment in constrained control and its application to MPC of rectifier

  • Author

    Xionglin, Luo ; Xin, Zuo ; Dianlin, Du

  • Author_Institution
    Res. Inst. of Autom., Univ. of Pet., Beijing, China
  • Volume
    1
  • fYear
    2004
  • fDate
    15-19 June 2004
  • Firstpage
    701
  • Abstract
    A new error treatment was applied to the model predictive control (MPC) of a commercial ethylene-ethane rectifier. The present method simply considers the error as zero if it is in the constrained zone, otherwise, the error is equaled to ysp-y. There is a jump (0 to ysp-y) at the constrained boundary. While, the new method presented in this paper avoids the jump at the constrained boundary, and ensures the measure of controlled variables to approach the setpoint in multivariable control. The error exponential coefficient, namely index was estimated by the simulating experiments of ethylene rectifier, ranged from 2 to 4. In the commercial ethylene-ethane rectifier application, the setpoint control of side draw products purity comes true and the concentration of bottom ethylene is controlled within a certain scope. Hence, the optimization of economy can be got from the proper choice of the setpoint and the scope of constrained zone.
  • Keywords
    error analysis; multivariable control systems; optimisation; organic compounds; predictive control; process control; constrained control; error exponential coefficient; error treatment; ethylene-ethane rectifier; model predictive control; multivariable control; optimization; setpoint control; Automatic control; Automation; Constraint optimization; Control systems; Error correction; Petroleum; Predictive control; Predictive models; Process control; Rectifiers;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Intelligent Control and Automation, 2004. WCICA 2004. Fifth World Congress on
  • Print_ISBN
    0-7803-8273-0
  • Type

    conf

  • DOI
    10.1109/WCICA.2004.1340675
  • Filename
    1340675