Development of Artificial Intelligence Based Multilevel Inverter Using MATLAB/Simulink

This study focuses on the development and evaluation of a multilevel inverter, comparing it with the latest iterations of multilevel inverters and verifying its performance using ONLT-PWM. Multilevel inverters offer several advantages but suffer from limitations such as harmonic injection and low output voltage, making them unsuitable for certain projects. To address these limitations, the proposed inverter topology incorporates a reduced number of switches and a sophisticated controller. Two control approaches, a traditional Proportional-Integral (PI) controller and an Artificial Intelligence-based Artificial Neural Network (ANN) with a feed-forward architecture, are trained and implemented. The proposed multilevel inverter, along with the controller-based system, aims to eliminate the shortcomings associated with traditional multilevel inverters. Renewable energy applications, particularly solar photovoltaic systems, are identified as suitable areas of application for the proposed multilevel inverter. The performance of the new inverter design is demonstrated and validated using either hardware or MATLAB/Simulink simulations, considering a range of input values.