Optimal Current Reference Calculation for MMCs Considering Converter Limitations

Abstract

The paper addresses an optimization-based reference calculation method for Modular Multilevel Converters (MMC) operating in normal and constrained situations (when the converter needs to prioritize its quantities as it has reached voltage or current limitations, e.g. during system faults). The optimization problem prioritizes to satisfy the external AC active and reactive current set-points demanded by the grid operator through the corresponding grid code. If the operator demands are fulfilled, it uses the available MMC degrees of freedom to minimize the arm inductance losses. Otherwise, if the operator demanded AC setpoints cannot be accomplished, the optimization attempts to minimize the error prioritizing between either AC active or reactive currents. The optimization problem constraints are imposed through a steady-state model considering simultaneously the external and internal AC and DC magnitudes of the converter. The steady-state model also includes the voltage variation in the equivalent arm capacitors (considering the ripple). Then, the imposed limitations are the maximum allowed grid and arm currents, the maximum allowed arm voltages and the sub-module capacitor maximum voltages. The paper presents a detailed formulation of the optimization problem and applies it to several case studies where it is shown that the presented approach can be potentially used to obtain the MMC references both in normal and fault conditions.

On the Roles and Interactions of the MMC Internal Energy Balancing Degrees of Freedom for Three-Wire Three-Phase Connections

“…In contrast to classical two-and three-level Voltage Source Converters (VSCs), the mathematical description and analysis of the MMC presents increased complexity since it has additional voltage components that can be applied into the converter's arms, which will be refereed as DOFs, allowing the performance of the converter to be improved [8]. Conventionally, the MMC's DOFs are described based on additive x sum and differential x dif f quantities, allowing the DC-and AC-side related components of the converter to be conveniently analyzed separately [9].…”