Evolutionary algorithm based selective harmonic elimination for three-phase cascaded H-bridge multilevel inverters with optimized input sources

Abstract

Multilevel inverters are finding wide application in electric drives, traction, flexible AC transmission systems (FACTS) and renewable energy systems. A cascaded H-bridge type multilevel inverter (CHBMLI) produces a near sinusoidal output voltage with lower switching stress and a higher conversion efficiency than the other types of MLIs. The Selective Harmonic Elimination (SHE) strategy is used to eliminate lower-order harmonic profiles and to regulate the fundamental component in the output voltage. SHE has the advantages of low switching frequency, low switching losses and low stress. In this paper, the modulation index and input voltage values are also considered as optimization variables along with the conventional switching angles to analyze the performance improvement in selective harmonic elimination. Heterogeneous Comprehensive Learning Particle Swarm Optimization (HCLPSO) and Gravitational Search Algorithm (GSA) algorithms are used to find the optimal switching angles, modulation index and input voltage source values for minimizing the lower-order harmonics present in the output voltage of seven-level and eleven-level CHBMLIs, while maintaining the fundamental component of the output voltage. The results obtained from MATLAB simulations and an experimental setup clearly indicate that the proposed HCLPSO-based multilevel inverter provides better performance when compared with GSA, firefly and Differential Search Algorithm (DSA)-based MLIs.