Minimization of system-level losses in VSI-based induction motor drives: Offline strategies

Srikanthan Sridharan, Philip T. Krein

Research output: Contribution to journalArticlepeer-review

Abstract

Applications of motor drives, ranging from automotive traction to more-electric aircraft, have widely varying speed and torque demands. With extensive use of inverter-fed drives, loss minimization techniques for such applications have evolved over the past few decades. Previous work on loss minimization has mainly focused on specific drive components. Component-level loss minimization, however, will not guarantee minimum total loss in the drive system. In this paper, a system-level loss minimization method is developed using a comprehensive loss model, to achieve true minimum total loss in the system. The total loss obtained is lower for the proposed system-based minimization than the component-based methods such as machine loss minimization, torque maximization per stator ampere, and dc-link loss minimization. Simulation and experimental results for a sample drive cycle are presented to verify the benefits of the proposed scheme for an induction motor drive. Results suggest system loss reduction of about 7%, when compared to machine-based optimization, in high-variability applications.

Original languageEnglish (US)
Article number7752804
Pages (from-to)1096-1105
Number of pages10
JournalIEEE Transactions on Industry Applications
Volume53
Issue number2
DOIs
StatePublished - Mar 1 2017

Keywords

  • Active damping
  • drive cycle
  • energy savings
  • field-oriented control (FOC)
  • flux maps
  • induction motor drive
  • loss minimization
  • loss models
  • optimization
  • voltage source inverter (VSI)

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Industrial and Manufacturing Engineering
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Minimization of system-level losses in VSI-based induction motor drives: Offline strategies'. Together they form a unique fingerprint.

Cite this