Enhanced low-voltage ride-through capability of flying capacitors inverter using model predictive control

This paper proposes an enhancement of the low-voltage ride-through (LVRT) capability of multilevel flying capacitors inverter using an improved model predictive control (MPC). The proposed solution is a multiobjective control strategy ensuring simultaneous capacitors' voltages balancing and grid current injection with reduced total harmonics distortion (THD) and unity power factor even under severe voltage sags. The proposed MPC technique makes use of a multioptimization solution for the common problem of weight dissimilarity of the control objectives. The first optimization is achieved by the inclusion of the state variables normalization within the cost function computation. This will provide to the proposed MPC technique more robustness with the ability to take into consideration external disturbances (grid voltage sag) while ensuring the meeting of the state variables constraints. The second one is achieved through accurately tuning the weighting factor based on minimizing the grid current total harmonics distortion and the maximal errors on the capacitors' voltages. Theoretical analysis and simulation results are presented to prove that the presented solution guarantees continuous and smooth power transfer to the grid even under severe voltage sags. The obtained results are validated on a 1-kW experimental prototype.