Real-time switching thirteen-level modified CHB-Multilevel inverter using artificial neural network technique based on selective harmonic elimination

Abstract

Recently, global research is giving more attention to the renewable energy sources due to its sustainability and environmentally friendly nature. The necessity of DC to AC conversion to integrate these sources to the well established AC power system create a significant development of multi-level inverter with its advantages of operating at higher rating system with lower component rating as well as better harmonics performance at it output voltage and current. In this research paper, a modified topology of CHB-MLI to provide 13-level output AC waveform is proposed based on the SHE-PWM strategy using the ANN optimization technique. The system modeling is done with Matlab Simulink software and verification are carried out by both simulation and experiment. Results show that the ANN technique able to reduce the THD significantly as the level of waveform is increased as low as 5.16% THD for the 13-level output voltage. Results from the experiment shows a good agreement with the simulation, thus verifying the effectiveness of the proposed ANN technique as an optimization method.