The modular multilevel converter (MMC) is a potentialcandidate for medium/high-power applications, specificallyfor high-voltage direct current transmission systems. One ofthe main challenges in the control of an MMC is to eliminate/minimize the circulating currents while the capacitor voltagesare maintained balanced. This paper proposes a controlstrategy for the MMC using finite control set model predictivecontrol (FCS-MPC). A bilinear mathematical model of the MMCis derived and discretized to predict the states of the MMC onestep ahead. Within each switching cycle, the best switching stateof the MMC is selected based on evaluation and minimization ofa defined cost function. The defined cost function is aimed at theelimination of the MMC circulating currents, regulating the armvoltages, and controlling the ac-side currents. To reduce the calculationburden of the MPC, the submodule (SM) capacitor voltagebalancing controller based on the conventional sorting method iscombined with the proposed FCS-MPC strategy. The proposedFCS-MPC strategy determines the number of inserted/bypassedSMs within each arm of the MMC while the sorting algorithm isused to keep the SM capacitor voltages balanced. Using this strategy,only the summation of SM capacitor voltages of each armis required for control purposes, which simplifies the communicationamong the SMs and the central controller. This paperalso introduces a modified switching strategy, which not onlyreduces the calculation burden of the FCS-MPC strategy evenmore, but also simplifies the SM capacitor voltage balancingalgorithm. In addition, this strategy reduces the SM switchingfrequency and power losses by avoiding the unnecessary switchingtransitions. The performance of the proposed strategies for a20-level MMC is evaluated based on the time-domain simulationstudies.
展开▼
