Optimal DC Source Influence on Selective Harmonic Elimination in Multilevel Inverters Using Teaching-Learning-Based Optimization

This paper presents an optimization method for selective harmonic elimination in a cascaded multilevel inverter using teaching-learning-based optimization (TLBO). The main objective in selective harmonic elimination (SHE) strategy is to eliminate low-order harmonics by solving nonlinear equations and reaching optimal solution, while the fundamental component is satisfied. In this paper, in one side, the influence of optimal dc sources is investigated to reach the SHE goal, and in another side, comparing optimization methods is considered. In this paper, the TLBO as a recently emerged nature-inspired algorithm is presented to provide better results for the SHE in comparison with genetic algorithm, artificial bee colony, imperialistic competitive algorithm, harmony search, ant colony optimization, particle swarm optimization, and differential evolution. For better comparison of those methods and influence of optimal dc sources, 5-, 9-, and 15-level inverters are chosen and MATLAB software is used for optimization. Simulation results show the superiority of TLBO, higher precision and probability of convergence than other mentioned algorithms. Finally, to validate the influence of optimal dc sources and the accuracy of TLBO results, the experimental setup is conducted for a 5-level cascaded H-bridge inverter with optimal dc sources.