• DocumentCode
    605099
  • Title

    Design of a double manifold Sliding Mode observer for speed and state estimation of the induction motor

  • Author

    Kachkovskiy, V. ; Comanescu, Mihai

  • Author_Institution
    Penn State Altoona, Altoona, PA, USA
  • fYear
    2013
  • fDate
    22-25 April 2013
  • Firstpage
    1123
  • Lastpage
    1126
  • Abstract
    The paper discusses the problem of speed and state estimation for the induction motor (IM) drive and presents a double-manifold Sliding Mode observer. This is obtained by adding extra feedback terms to a single-manifold observer design. Both designs are based on the model of the IM in the stationary reference frame. The observer uses compound manifolds which are constructed as a combination of the estimated states and mismatches. Observers with compound manifolds have not been widely investigated because they cannot be designed using a standard procedure; however, they may have very interesting properties. The paper investigates the behavior of a single-manifold observer first; it is shown that this only converges partially and provides an approximate speed estimate. Based on this, a double-manifold observer is designed. This estimates the fluxes, drives the current mismatches to zero and yields an accurate speed estimate. The method is applicable in a sensorless IM drive control algorithm where the speed, the flux magnitude and the rotor flux angle are needed. The theoretical developments are supported with simulations.
  • Keywords
    angular velocity control; control system synthesis; induction motor drives; power system state estimation; rotors; variable structure systems; IM drive control algorithm; compound manifolds; flux magnitude; induction motor drive; manifold sliding mode observer design; rotor flux angle; speed estimation; state estimation; Induction motors; Manifolds; Observers; Rotors; Switches;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Power Electronics and Drive Systems (PEDS), 2013 IEEE 10th International Conference on
  • Conference_Location
    Kitakyushu
  • ISSN
    2164-5256
  • Print_ISBN
    978-1-4673-1790-0
  • Electronic_ISBN
    2164-5256
  • Type

    conf

  • DOI
    10.1109/PEDS.2013.6527188
  • Filename
    6527188