Total Harmonic Distortion Optimization of Multilevel Inverters Using Genetic Algorithm: Experimental Test on NPC Topology with Self-Balancing of Capacitors Voltage Using Multilevel DC-DC Converter

Multilevel converter plays an important role in high- and medium-power applications, such as renewable energy generation systems and power transmission. One of the most known problems of power electronics converter is harmonics. Furthermore, optimization algorithms can solve complex issues in electrical engineering such as the harmonics in power converters. In this paper, the optimization of total harmonic distortion using genetic algorithm is performed for three-phase three, five, seven, and nine levels inverter for several numbers of switching angles and different modulation index values. The best values of the total harmonic distortion of the three-level inverter were less than 23.60% and about 5% for the nine levels inverter in this study. The results obtained in the optimization process have been applied to a neutral point clamped multilevel inverter. Another challenge for this inverter is the DC-link capacitors voltage unbalance; thus, the multilevel boost DC-DC converter is proposed as a solution. The proposed system is validated experimentally using a three-phase three-level neutral point clamped inverter and four-Level (or two-Level) DC-DC boost converter.

Automated summary

Multilevel converters are crucial in high- and medium-power applications, but harmonics pose a common issue. This study optimizes total harmonic distortion in various-level inverters using genetic algorithms and proposes a multilevel boost DC-DC converter to address DC-link capacitor voltage imbalances.