Dynamic Phasor Model Predictive Control of Switched Mode Power Converters

Abstract

This paper considers model predictive control (MPC) of switched mode power converters and uses a hybrid prediction model that represents the switching ripple of the converter state. The goal of the approach is to achieve both high dynamic performance and low harmonic distortion of the output at steady state. It is argued that the previously published approaches to hybrid MPC of power converters are sensitive to plant/model mismatch, and an alternative problem formulation is suggested. The new formulation describes the converter state in terms of dynamic Fourier coefficients. This allows to quantify and feed back more information about the switched converter waveform, and can thus reduce the dependence on the system model. The suggested approach is evaluated in both simulation and experiment, and it is shown to be less sensitive to plant/model mismatch than approaches from the literature.