• DocumentCode
    151129
  • Title

    On the benefit of long-horizon direct model predictive control for drives with LC filters

  • Author

    Geyer, Tobias ; Karamanakos, Petros ; Kennel, Ralph

  • Author_Institution
    ABB Corp. Res., Baden-Dättwil, Switzerland
  • fYear
    2014
  • fDate
    14-18 Sept. 2014
  • Firstpage
    3520
  • Lastpage
    3527
  • Abstract
    Even though direct model predictive control (MPC) schemes almost exclusively use a prediction horizon of one in power electronics applications, the use of longer horizons offers significant performance benefits. This statement is underlined in this paper for a medium-voltage variable speed drive system, which consists of a three-level inverter, an LC filter and an induction machine. The proposed MPC controller simultaneously regulates the inverter current, capacitor voltage and stator current along given references, by manipulating the switch positions of the inverter. As will be shown, extending the prediction horizon significantly reduces the oscillations due to the filter resonance. For sufficiently long horizons, such as ten, low total harmonic distortions of the stator current can be achieved at low device switching frequencies. An additional active damping loop is not required, adding to the conceptual simplicity of the proposed control scheme.
  • Keywords
    asynchronous machines; invertors; machine control; power filters; predictive control; stators; variable speed drives; MPC schemes; THD; capacitor voltage; conceptual simplicity; induction machine; inverter current; lilter resonance; long-horizon direct model predictive control; low device switching frequencies; medium-voltage variable speed drive system; power electronics applications; stator current; switch positions; three-level inverter; total harmonic distortions; Capacitors; Decoding; Inverters; Rotors; Stators; Switches; Switching frequency;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Energy Conversion Congress and Exposition (ECCE), 2014 IEEE
  • Conference_Location
    Pittsburgh, PA
  • Type

    conf

  • DOI
    10.1109/ECCE.2014.6953879
  • Filename
    6953879