Improved Model Predictive Control Methods with Natural Capacitor Voltage Balancing for the Four Level-Single Flying Capacitor (4L-SFC) Inverter

In this paper, two control schemes based on finite control set-model predictive control (FCS-MPC) are proposed for a reduced-capacitor multilevel converter, referred to as four level-single flying capacitor (4L-SFC) converter. By using the notion of reconstructing phase voltage levels, natural capacitor voltage balancing can be achieved for the 4L-SFC converter, which is not fulfilled by the PI-based conventional modulation methods. Besides, two proposed strategies present a weighting factor-less approach to achieve fixed switching frequency, lower current distortion, and reduced ripple of capacitor voltages as well as lightening the computational burden. A single-objective cost function based on voltage vector error is defined in both methods to accomplish the reference current tracking. The second proposed method employs modulated model predictive control (M2PC) to guarantee a well-concentrated harmonic spectrum and lower current distortion compared to the first strategy. Simulation studies are conducted to validate and compare the performance of the proposed methods in terms of steady-state and transient-state response.