Double voltage vector model predictive control for grid-connected cascade H-bridge multilevel inverter with fixed...

Area of Science:

• Electrical Engineering
• Power Electronics
• Control Systems

Background:

• Model predictive control (MPC) offers advanced dynamic, nonlinear, and multi-objective control but suffers from dispersed harmonic spectra due to unfixed switching frequencies, complicating output filter design.
• Cascade H-bridge (CHB) multilevel inverters are crucial for power applications, but efficient and stable control remains a challenge.

Purpose of the Study:

• To present a novel double voltage vector model predictive control (DVV-MPC) algorithm for grid-connected CHB multilevel inverters.
• To address the challenge of dispersed harmonic spectra by achieving a fixed switching frequency.
• To improve the overall control performance and simplify filter design for CHB inverters.

Main Methods:

• Implementation of a DVV-MPC algorithm utilizing two adjacent voltage vectors within a single switching cycle.
• Minimization of an additional cost function to determine the duty cycle for the associated voltage level.
• Application of adjacent regions prediction and time-delay compensation methods to reduce computational burden and delay.

Main Results:

• The proposed DVV-MPC algorithm successfully achieves a fixed switching frequency for the CHB inverter.
• Demonstrated advantages in both dynamic and steady-state performance through simulations and experiments on a single-phase 7-level CHB inverter.
• Simplified output filter design due to the predictable harmonic spectrum.

Conclusions:

• The DVV-MPC strategy effectively combines the benefits of MPC with the advantage of a fixed switching frequency for CHB inverters.
• The method is suitable for cascade inverters with 'n' cells and offers improved control and easier filter implementation.
• Validated effectiveness for grid-connected applications, paving the way for more efficient power conversion systems.