Robust Control Design of MMC-HVDC Systems Using Multivariable Optimal Guaranteed Cost Approach

The modular multilevel converter (MMC) represents the important technological innovation that emerged among the diverse available topologies of voltage source converter and is avowedly the most suitable solution for converters in high voltage direct current (HVDC) transmission and multiterminal direct current grids. Special focus is given through this paper to the dynamic performance of an MMC-based, point-to-point HVDC system. Using an optimal guaranteed cost control theory, a robust control approach is designed in order to reject the impact of the unmodeled uncertainty, mainly in the ac side of the MMC. For this aim, a small-signal state-space linear model is derived for the control design of an advanced local controller of each MMC station. Furthermore, a new optimal guaranteed cost controller is proposed based on the convex optimization problem using linear matrix inequality optimization approach. The proposed strategy leads to regulate simultaneously the ac grid and differential currents as well as total stored energy per phase in abc frame. To ensure the energy balancing between upper and lower arm per phase, an outer control loop is used in order to control the energy difference per phase between upper and lower arms of MMC. For the MMC linked to HVDC system, the active power reference is generated through an outer classical dc voltage controller. This combined control strategy between classic and advanced robust regulation methods allows exploiting the advantages of both control methods. Effectiveness of the proposed optimal robust control strategy for point-to-point MMC-HVDC system is evaluated across accurate and skillful simulation study under MATLAB/SimPowerSystem environment. The simulation results convince satisfactory dynamics responses of a two-terminal MMC-HVDC system based on the robust control approach under various operating conditions, even under unbalanced ac grid conditions (e.g., asymmetrical fault).