Machine Learning-Based Harmonic Optimization and Control of Cascaded H-Bridge Multilevel Inverters
Contributors
Dr Santhosh Kumar Thota
Prof.Sai Kiran Oruganti, Faculty of Engineering and Built Science, Lincoln University College-KL,Malaysia.
Keywords
Proceeding
Track
General Track
License
Copyright (c) 2026 Sustainable Global Societies Initiative

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Abstract
The increasing integration of renewable energy sources, electric transportation, and advanced electronic devices has created a growing need for reliable and high-performance power conversion systems. Conventional inverter control methods may experience limitations in harmonic suppression, output voltage stability, and responsiveness to variations in load conditions. To address these challenges, this study investigates a machine learning-assisted control framework for cascaded H-bridge multilevel inverters (CHB-MLIs), with particular emphasis on improving output waveform quality. Different learning algorithms, including convolutional neural networks (CNNs), K-nearest neighbors (KNN), and recurrent neural networks (RNNs), are considered for waveform characterization, adaptive pulse-width modulation (PWM) parameter selection, load identification, and early fault prediction. The proposed framework aims to enhance inverter operation by using data-driven techniques to support control decisions under changing operating conditions. Its performance is evaluated through simulation and experimental analysis, with harmonic distortion, voltage regulation, and operational reliability considered as key assessment criteria. The study explores the potential of integrating machine learning into multilevel inverter control to improve power quality and support applications in renewable energy conversion, smart-grid infrastructure, and industrial power systems.