HYBRID OPTIMIZATION AND CNN-BASED CONTROL FOR HARMONIC REDUCTION IN MULTILEVEL INVERTERS APPLIED TO MICROGRIDS

High-power ratings and improved output voltage waveforms are made possible by the modular construction of multi-level inverters (MLIs). Because of their excellent performance, modular design, and low harmonic distortion, these systems are increasingly being used in a variety of industries and renewable energy applications. However, the integration of RES is severelyhampered by their intricate management and coordination needs. To improve performance and reduce Total Harmonic Distortion (THD), this research suggests a novel 31-level MLI design that integrates a hybrid control technique. A CNN controller optimized by a hybrid algorithm fully tunes the inverter's performance. The suggested hybrid method optimizes the MLI's switching pulses by combining a CNN controller with the Pelican optimization and Pufferfish optimization algorithms. Thegoal of this optimization is to reduce voltage stress and THD across the switches. To properly utilize RES, the MLI is also integrated with wind and solar energy sources. With MATLAB/SIMULINK, the effectiveness of the recommended technique is confirmed and put into practice. The results confirm that in the proposed 31-level inverter topologies, the recommended inverter has a THD of 0.94% for voltage and 0.79% for current. Additionally, they demonstrate that the suggested approach achieves better results when compared to earlier studies.