Modified Model Predictive Control and Optimized Nearest Level Control on Balance of the Submodule Voltage Control for MMC-HVDC

Abstract

This paper introduces an optimized nearest level control (NLC) method and a modified model predictive control (MPC) method with the decreasing number of considered states for modular multilevel converter (MMC) as to compare their effects on the compensation of the voltage sorting algorithm. The NLC method fits for for MMC-HVDC with large number of sub-modules (SM), and as an advanced control strategy, the MPC method could control multiple variables via a cost function owning advantages of direct modeling and fast dynamic responding performances. Number rounding algorithm for SMs and the capacitor voltage sorting algorithm are two methods being normally considered for the modified NLC and MPC methods to track the referenced voltage for guaranteeing the voltage balancing effects among different capacitors of SMs. However, errors which are resulted from the number rounding algorithm, and errors which are generated from the deviation of the capacitor voltage of SMs, bringing a significant difference between the output arm voltage and the referenced one cannot be neglected. In this paper, two compensated methods brought from modified NLC and MPC methods are given to compare the effects concerning improvement on the performance to dampen the deviation of the output converter voltage at valve side from the referenced one, to decrease the harmonics and to avoid the oscillation etc. Finally, two proposed algorithms will be verified with the help of RTLAB simulation with the model of Zhangbei VSC - HVDC project.