DC Bus Stability Improvement Using Dynamic Voltage Feedback Model Predictive Control Method

This paper proposes a dynamic voltage feedback model predictive control (MPC) method for a motor-driven system with a Supercapacitor (SC). This method is suitable for improving the stability of DC bus during the regenerative energy tracking process. The MPC is used to improve the response performance of the SC. The dynamic feedback of the voltage aims to calculate more accurate current reference values. The discrete state-space model is established for both the SC and bidirectional DC/DC converters to predict the voltage and current of the SC. The voltage prediction value of the SC is dynamically used as feedback to calculate the current reference values, ensuring the correct energy flow between the SC and the motor. Finally, a cost function is introduced to minimize the error between the predicted SC current and the current reference values while adhering to specific constraints. Simulation and experiment were conducted to compare the stability of the DC bus under different control methods. The DC bus voltage regulation accuracy is 1.62% in the dynamic voltage feedback MPC method, while it is 8.65% in the power control based on PI during the regenerative energy tracking process.