Voltage and Frequency Droop Control in Low Voltage Grids by Distributed Generators with Inverter Front-end

In recent years, an ongoing increase of grid-connected distributed and renewable electricity generation technologies is clearly noticeable. From an environmental perspective, the increase of renewable technologies is highly desirable, as they do not emit greenhouse gasses and thus not contribute to climate change. Nevertheless, their increase poses increasing challenges on the reliable operation of the grid, due to their variable nature, which is on top often hard to predict. This work proposes, develops and implements techniques to cope with variable generation technologies both in islanded and interconnected electricity grids, while sustaining a high reliability and minimizing costs. The focus is on the control of power-electronic inverters, being the interface between the generation source and the grid. First, a profound review of existing voltage and frequency droop control methods for inverters is conducted. The use of an improved stability analysis technique is introduced, revealing severe stability problems under certain grid conditions and suggesting the need for new, improved control techniques. Subsequently, some fundamental control concepts are introduced, which allow for the design of a new, virtual impedance droop controller. After the detailed development of this controller, the performance of both voltage and frequency control is investigated under various grid conditions both interconnected with the mains and in islanded grids, for a single inverter unit and for several parallelconnected inverters. Hereto, simulations as well as laboratory experiments are used. Finally, the developed techniques are applied in an autonomous electricity grid, consisting of four parallel-connected inverters in isolation from the grid. The droop control algorithm is supplemented with an optimizing algorithm, ensuring that the isolated grid is always operated at minimal marginal cost. This is obtained through the continuous exchange of data between the inverters, using an innovative, fully distributed gossiping algorithm. Again, laboratory experiments are included to show the performance under various scenarios, imitating variable generation resources and consumption profiles.