Frequency-adaptive Current Control of a Grid-connected Inverter Based on Incomplete State Observation under Severe Grid Condition

Grid-connected inverter (GCI) plays a crucial role on facilitating a stable and efficient power delivery, especially under severe and complex grid conditions. Harmonic distortions and imbalance of the grid voltages may degrade the grid-injected current quality. Moreover, inductive-capacitance (LC) grid impedance and the grid frequency fluctuation also degrade the current control performance or stability. In order to overcome such an issue, this study presents a frequency-adaptive current control strategy of a GCI based on incomplete state observation under severe grid condition. When LC grid impedance exists, it introduces additional states in a GCI system model. However, since the state for the grid inductance current is unmeasurable, it yields a limitation in state feedback control design. To overcome such a limitation, this study adopts a state feedback control approach based on incomplete state observation by designing the controller only with the available states. The proposed control strategy incorporates feedback controllers with ten states, an integral controller, and resonant controllers for robustness of inverter operation. To reduce the reliance on additional sensing devices, a discrete-time full-state current observer is utilized. Particularly, with the aim of avoiding the grid frequency dependency of system model as well as complex online discretization process, observer design is developed in the stationary reference frame. Additionally, a moving average filter (MAF) based phase-locked loop (PLL) is incorporated for accurate frequency detection against distortions of grid voltages. For evaluating the performance of the designed control strategy, simulation and experiment are executed with severe grid conditions including the grid frequency changes, unbalanced grid voltage, harmonic distortion, and LC grid impedance.