Discrete-Time Multi-Resolution Modeling of Switching Power Converters Using Wavelets

Abstract

Modeling of power electronics circuits presents significant challenges due to the non-linear time-varying behaviour of this kind of system. It is common practice to operate with families of models of different complexity depending on the analysis goal. In this paper we propose a wavelet-based approach as a unifying framework able to provide multi-resolution capabilities and consequently significant flexibility to the modeling process. The comparison with more classical approaches to power converter modeling demonstrates how this approach can be considered an extension of formalizations such as state-space averaging. Simulation results for both DC-DC and DC-AC power converters, including soft-switching configurations, are presented as sample applications. The discrete-time formulation makes this approach particularly suitable for design and analysis of converters operated with digital controllers.