基于迭代学习的模型预测控制用于直流微电网参数不确定性解耦与补偿

The control performance of power converters in energy storage systems (ESSs) critically determines the power quality of DC microgrids. However, parameter mismatches during operation may lead to significant voltage and current fluctuations, particularly under pulse load integration. To address this challenge, this article proposes the iterative learning decoupling model predictive control (ILD-MPC) method for DC–DC converters. The key innovation lies in decoupling and estimating uncertain parameters from periodic model prediction errors by simple equations. These errors are treated as tracking errors and leveraged to iteratively update the MPC prediction model, thereby refining subsequent control actions. Simulation and experimental results demonstrate that the proposed method achieves superior performance compared to conventional MPC (C-MPC), reducing the steady-state voltage ripple by 37.5%–73% and the settling time by 33.3%.