A New Generalized Robust Predictive Current Control for Grid-Connected Inverters Compensates Anti-Aliasing Filters Delay

Abstract

In this paper, a new predictive control for grid-connected inverters is presented, which provides the best performance in terms of transient response and stability margins when analog filters in current sensing circuits are used. These filters are necessary to avoid aliasing in the A/D conversion process, which causes an important ripple on injected current into the grid and increases the total harmonic distortion. However, predictive controls are very sensitive to delays on the acquisitions, so when such anti-aliasing filters are used a reduction of stability margins is produced and the transient response is affected. The proposed predictive control allows to compensate the effect of the filter, for any order and cutoff frequency, providing the best transient response in the reference tracking and the highest stability margins. Furthermore, a modification on the adaptive strategy of current error is proposed, which allows to work with single-phase grid-connected inverters. The proposed predictive control has been tested on a 3.3-kVA single-phase grid-connected inverter with LCL filter, using different anti-aliasing filters, obtaining in all cases, better results than those offered by other predictive controls recently published.