Modular Multilevel Converter with a Sensing and Balancing (SAB) Technique for Capacitor Voltage Balancing

Modular multilevel converters (MMCs) are powerful tools of the future for high- and medium-power applications. The system is composed of cascaded connections of individual submodules. Balancing of submodule capacitor voltage is a key issue while designing MMC. The converter's reliability will suffer because of increased voltage stress on power components due to the imbalanced capacitor voltage. This work addresses that problem using a novel capacitor voltage sensing and balancing technique. Using two control loops one for finding the direction of the arm current and the other for balancing of capacitor voltage with a low-complexity algorithm, the capacitor voltage will be well balanced. The converter dynamics of the aforementioned method have been studied by comparison of parameters like circulating current, grid voltage, submodule capacitor voltage, switching losses, and total harmonic distortion with that of MMC using the conventional sorting algorithm. The submodule capacitor voltage is typically balanced by the conventional sorting approach, but the proposed method adds the feature that the balancing of capacitor voltage is continued till the MMC output voltage matches the required terminal voltage of the high voltage DC (HVDC) line. The proposed model is scalable to any number of submodules without further computational complexity. The proposed system's simulation results reveal that the circulating current is decreased by 50%, the grid voltage is improved by 13%, switching losses are reduced by 49%, and total harmonic distortion is reduced by 13% compared to conventional MMC. The features of the proposed system are reduced complexity of the balancing algorithm, fast stabilization of output voltage without any external controller, and superior performance which can be utilized to feed HVDC/medium voltage DC systems.