Real-Time Identification and Adaptive Control of Grid-Connected Three-Phase Inverters

Abstract

Grid-connected inverters are typically applied to integrate renewable energy, such as solar and wind, to the power system. Therefore, proper operation of the inverters is essential for efficient utilization of the renewable energy. Rapidly increasing the number of grid­connected inverters has started to affect power system dynamics and stability, particularly in areas where the majority of the power is generated through the inverters. One issue studied is the harmonic resonance between the inverter and the grid.

The stability of the grid-connected system can be evaluated by an impedance-based stabil­ity analysis, where the system stability is assessed by the ratio between the inverter and the grid impedances. Previous studies have presented multiple methods for designing the grid­connected inverter so that the impedance-based stability issues can be avoided under certain grid conditions. However, the grid impedance usually remains unknown in the inverter de­sign process and may vary over time; this typically means it is not possible to customize the inverter to the grid-connection point or ensure the impedance-based stability.

This thesis has presented methods for real-time measurements of the grid-connected sys­tems. Applying the methods, the grid impedance can be obtained in real time and the impedance-based stability of the grid-connected system can be assessed under varying grid conditions. Additionally, the measurements can be utilized to improve the inverter perfor­mance in the grid-connection point. In this thesis, the grid-connected inverter is configured to adaptively re-adjust its own control parameters based on the real-time measurements of the grid conditions. This is shown to effectively avoid the impedance-based stability issues and to improve the inverter control performance under varying grid conditions.