On Faster-Than-Real-Time Dynamic Simulation of AC-DC Grids

Abstract

The proportion of renewable energy versus conventional generation has been increasing significantly worldwide due to enhanced environmental standards, bringing with it a host of technical and operational challenges. Restrictions on available transfer capacities leading to severe congestion of major transmission and distribution corridors have been the most common problem, with severe consequences related to voltage and frequency regulation, and transient stability and reliability of the grid. Building new transmission facilities is the only viable long-term solution to assuage the impacts of renewable energy penetration, and increasingly electric utilities are opting to construct new direct current (DC) transmission lines. With modular multi-level converter (MMC) technology as the foundation, multi-terminal DC grids have enabled the transfer of large amounts of power, and interconnection of asynchronous alternating current (AC) systems. In this scenario, real-time dynamic simulation of integrated AC-DC networks is indispensable for maintaining system stability and resiliency. Currently, available dynamic simulation tools adopt sequential compute hardware and therefore turn out to be time-consuming. On the other hand, field-programmable gate arrays (FPGAs) with large hardware resource capacity allow parallel processing of various components. As a result, they shorten the process of finding the whole network solution by optimizing the hardware latency, which is why they are the ideal platform to attain faster-than-real-time (FTRT) execution. In this thesis, practical modeling techniques, numerical solution strategies, and hardware implementation schemes are proposed for FTRT dynamic emulation of integrated AC-DC grids on FPGAs.