A Time-Domain Virtual-Flux Based Predictive Control of Modular Multilevel Converters for Offshore Wind Energy Integration
Yuanxiang Sun, Zhenbin Zhang, Yuzhe Zhang, Yu Li, Zhen Li
Abstract
Modular multilevel converter-based high-voltage direct current (MMC-HVDC) transmission has become a promising solution for offshore wind energy integration. Direct model predictive control (DMPC) has emerged as an effective control alternative for such a system. However, classical predictive control of the grid-side MMC relies on the measurement accuracy of grid voltage, reducing the system robustness against measurement noises/bias. This work proposes a virtual-flux based grid voltage sensorless DMPC scheme for the grid-side MMC. The proposed solution is designed in the time-domain, achieving fast estimation during the transient phases, good robustness to capacitor voltage measurement noises/bias and grid frequency variation, and strong adaptivity under weak and distorted grid conditions. Additionally, seamlessly combining with predictive control frame, the proposed solution achieves precise grid voltage estimation and fast power control dynamics, which are validated with both simulation and experimental data.