Model-Free Predictive Control based on the Execution Time of Active Vectors Using Ultra-Local Model

Model-free predictive control based on the execution time of active vector (ETAV-MFPC) is proposed in this paper, and the traditional model-free predictive control uses the entire sampling period time with all candidate voltage vectors according to the principle of minimizing the cost function. The delay problem in the predictive controller may degrade the controller's performance. With the proposed control, the optimal vector is chosen only during the execution time of the active vector, which solves the delay problem which reduces the resulting current ripple and lower harmonics. The execution time is calculated using the ultra-local model (ULM) equation. In this work, the unknown part in ULM is estimated by the integrated sliding mode observer (ISMO) with an enhanced design to reduce the chattering phenomenon. The simulation results confirm the efficacy of the proposed ETAV-MFPC method for output current, demonstrating reduced current ripples and enhanced current quality.