一种基于连续控制集电流的模型预测控制方法在带LCL滤波器的三相两电平AC/DC变换器中的应用

This article presents a developed novel continuous control set current-based model predictive control (CCS C-BMPC) for an ac–dc grid converter connected to the grid through an LCL filter. Most CCS MPC algorithms control the converter current by regulating the reference voltage derived from the prediction algorithm in the final control stage. The proposed algorithm enables direct control of the grid current by regulating the converter current and the voltage on the filter capacitor. The algorithm uses optimized prediction to compensate for the grid current vector error, minimizing total harmonic distortion (THD) at a relatively low switching frequency. It exhibits high robustness to power grid disturbances and avoids current distortions during zero crossing, especially at low current amplitudes. The advantages of the proposed method have been confirmed in simulations and laboratory tests in steady-state and transient conditions. It confirms that the developed control technique shows a high quality of the grid current (low THDi value), high dynamic performance, and immunity in case of weak grid conditions. The proposed method was compared with a method that uses a typical proportional-resonant current controller.