Robust Sequential Model-Free Predictive Control of a Three-level T-type Shunt Active Power Filter

Three-level T-type shunt active power filter (3LT²SAPF) has been widely used to improve the power quality of power grid under nonlinear load, and their harmonic compensation performance and reliability have been widely noticed. Model predictive control (MPC) offers the advantage of convenient multi-objective optimization, quick response and a simple principle. However, once the parameter mismatch occurs due to some reasons (e.g., temperature change, noise from the sensors, etc.), the output performance of MPC system will be seriously deteriorated, sometimes may even endanger the system stability. Moreover, for cases when the circuit relation is unknown, the prediction model of the control object could not be established, rendering the traditional model based MPC (TMPC) inapplicable. To address these challenges and enhance the robustness of the 3LT²C SAPF system, a linearly fitting sequential model free predictive control method (SMFPC) is proposed. First, the principle of linear fitting method for current prediction is briefly analyzed. Second, a current difference matrix for each switch sequence has been established, which is utilized for harmonic current tracking. Finally, the cost functions for NP voltage and DC-bus voltage are designed. Numerous experimental results under different scenarios verify the correctness and effectiveness of the proposed method.