Application of Model Predictive Control in Modular Multilevel Converter for MTPA Operation and SOC Balancing

In this thesis, a one-step horizon model predictive control strategy (MPC) is implemented in a multilevel modular converter (MMC) to control the speed of an electric vehicle (EV) motor. Maximum torque per ampere (MTPA) and field weakening (FW) control strategies are used to generate reference signals for maximum torque output. The proposed control scheme aims to track the reference signal by independently regulating voltages from the MMC modules. To achieve this, the switches of the MMCs are directly controlled, eliminating the need for a pulse width modulator. A one-step horizon implementation of MPC ensures the robustness of the control system by making the real-time implementation possible. It leads to favorable performance under asymmetrical loads. The phase voltage is supplied to the motor through the MMC architecture which is composed of a large number of battery cells connected in series to supply the motor drive. Due to the non-identical characteristics of the battery, the state of charge (SOC) and the terminal voltage of the cells vary significantly at different operating conditions. The given control scheme is also incorporating a voltage balancing property that ensures the terminal voltages of all the battery cells in the MMC architecture are equalized. Finally, simulation results are presented to show the effectiveness of this control strategy and hardware is under development to validate the system performance