Model Predictive Control for Voltage Regulation in DC Vehicular Microgrids under Load Disturbances

This paper proposes a multi-source coordinated voltage regulation strategy based on model predictive control (MPC) to address the voltage stability issue of direct current vehicular microgrids (DC-VMG) under load disturbances. A system-level dynamic model is first established, incorporating the engine-generator set (EGS), energy storage system (Energy Storage Batteries(ESB) and Supercapacitor(SC)), and the drive motor, thereby revealing the coupling relationships among source, network, and load. On the control level, a Maximum Torque Per Ampere (MTPA) strategy is introduced to optimize the generator-side torque output, while a dual-model predictive control (D-MPC) algorithm is designed to achieve voltage regulation and frequency-domain decoupling of the energy storage system(ESS). In this scheme, the ESB handles low-frequency power demands, whereas the SC responds to high-frequency fluctuations. Simulation results show that, compared with the conventional dual-PI (D-PI) method, the proposed control strategy significantly improves DC bus voltage stability, dynamic response speed, and energy scheduling efficiency, thereby enhancing the overall robustness and energy utilization of the system. This research provides a theoretical foundation and control approach for the stable operation of DC-VMG under dynamic operating conditions.