Extended ISMO-Based Two-Step Prediction Horizon Model-Free Predictive Control for Power Converters
Zeyu Zhang, Jien Ma, Lin Qiu, Xing Liu, Bowen Xu, Youtong Fang
Abstract
Model predictive control is widely used in the control of power converters due to its simple implementation, excellent performance, and fast dynamic response. However, the conventional model predictive control relies on load parameters for prediction. Hence, it exhibits high sensitivity. Furthermore, high switching frequency (SF) also contributes to increased unnecessary losses. To solve this issue, a novel integral sliding mode observer (SMO)-based robust finite control-set model predictive control (FCS-MPC) methodology for power converters is proposed in this article, aiming to enhance the robustness of the control system while reducing its SF. In this proposal, the current prediction is accomplished by the implementation of an extended ISMO for ultra-local model observation, which successfully mitigates the impact of load parameter perturbations. Meanwhile, the two-step prediction horizon structure extends the optimization range to two control periods, thereby enhancing the probability of applying the same voltage vector in consecutive control periods, which significantly reduces the SF. The notable advantage of the proposed control method lies in enhancing the robustness of the system and reducing sensitivity to load parameters, while maintaining the SF within a lower range, concurrently. Finally, simulation and experimental results are obtained to verify the robustness and low-switching-frequency performance of the proposed approach.