Online grid impedance identification for adaptive control of grid-connected inverters

Abstract

This paper presents the online implementation of a grid impedance identification method embedded as an ancillary function within the DSP of a grid-connected inverter. Online identification of the grid impedance is required for assessing in real-time the stability of the interconnected system formed by the grid-connected inverter and the ac grid. In this paper, a non-parametric estimation of the grid impedance is acquired on a cycle-to-cycle basis by injecting a current impulse from the grid-connected inverter and performing an online DFT analysis of the impulse response. The impulse response is decomposed into its positive- and negative-sequence components to fully characterize the impedance of balanced three-phase grids. The paper discusses as well some of the design considerations of the injected impulse, the online DFT implementation, and the sequence component decomposition. Experimental results are presented to compare the online identification against offline analysis methods, as well as grid impedance measurements based on frequency response analyzer readings.