Reduced-order modeling of power electronics components and systems

Davoudi, Ali

Permalink

https://hdl.handle.net/2142/16097 Copy

Description

Title

Reduced-order modeling of power electronics components and systems

Author(s)

Davoudi, Ali

Issue Date

2010-05-19T18:34:46Z

Director of Research (if dissertation) or Advisor (if thesis)

Chapman, Patrick L.

Doctoral Committee Chair(s)

Chapman, Patrick L.

Committee Member(s)

• Krein, Philip T.
• Sauer, Peter W.
• Cangellaris, Andreas C.

Department of Study

Electrical & Computer Eng

Discipline

Electrical & Computer Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Energy Systems
• Power Electronics
• Energy Conversion
• Magnetic
• Model order reduction
• Modeling and simulation

Abstract

This dissertation addresses the seemingly inevitable compromise between modeling fidelity and simulation speed in power electronics. Higher-order effects are considered at the component and system levels. Order-reduction techniques are applied to provide insight into accurate, computationally efficient component-level (via reduced-order physics-based model) and system-level simulations (via multiresolution simulation). Proposed high-order models, verified with hardware measurements, are, in turn, used to verify the accuracy of final reduced-order models for both small- and large-signal excitations. At the component level, dynamic high-fidelity magnetic equivalent circuits are introduced for laminated and solid magnetic cores. Automated linear and nonlinear order-reduction techniques are introduced for linear magnetic systems, saturated systems, systems with relative motion, and multiple-winding systems, to extract the desired essential system dynamics. Finite-element models of magnetic components incorporating relative motion are set forth and then reduced. At the system level, a framework for multiresolution simulation of switching converters is developed. Multiresolution simulation provides an alternative method to analyze power converters by providing an appropriate amount of detail based on the time scale and phenomenon being considered. A detailed full-order converter model is built based upon high-order component models and accurate switching transitions. Efficient order-reduction techniques are used to extract several lower-order models for the desired resolution of the simulation. This simulation framework is extended to higher-order converters, converters with nonlinear elements, and closed-loop systems. The resulting rapid-to-integrate component models and flexible simulation frameworks could form the computational core of future virtual prototyping design and analysis environments for energy processing units.

Graduation Semester

2010-5

Permalink

http://hdl.handle.net/2142/16097

Copyright and License Information

Copyright 2010 Ali Davoudi

Owning Collections