Adaptive Control of Grid-Connected Inverters With Nonlinear LC Filters

Abstract

Distributed generation is an effective solution to clean, reliable, and cost-efficient energy supply. In this framework, inverters play a key role as energy converters. Accordingly, this article investigates the utilization of nonlinear inductive–capacitive (LC) filters for grid-connected inverters. Nonlinear LC filters are made of inductors with nonlinear B–H curves, which makes them have a smaller volume, less dissipation loss, and lower cost. The main drawback of using these filters, however, is their nonlinear current-dependent characteristics. To harness the undesirable impacts of nonlinear filters on the system performance, an adaptive observer-based control scheme is developed in this article. The proposed controller is based on a concurrent estimation of nonlinear inductor values, capacitor voltages, and grid-side currents. Based on its fast adaptation to plant changes, this scheme provides an improved stability margin and an inherent resonant damping characteristic under grid impedance uncertainty. Compared with the conventional approaches, this method requires no extra sensors. A comparative simulation and experimental analysis, based on a three-phase 7-kW inverter, is provided to validate the efficacy of the proposed control strategy.