Recognition of dynamic patterns in DC-DC switching converters

Techniques are presented for analyzing the dynamic patterns of power flow in high-frequency DC-DC switching power converters. Emphasis is placed on exploring methods for developing reduced-order models of power electronic circuits for simulation and state-space averaging. Tools from selective modal analysis are used to develop the dynamic pattern recognition algorithm that facilitates the classification of circuit modules. The algorithm is intended for the analysis of converters derived from the canonical switching call, i.e., the buck, boost, and flyback converters.<
>