An extensive critique on machine learning techniques for fault tolerance and power quality improvement in multilevel inverters
K. Sakthivel, S. Albert Alexander
Abstract
Multilevel inverters (MLI) perform a significant role in microgrids to overcome the power demand for various load conditions due to violations of load day by day in recent centuries and enhance the reliability of Renewable energy sources (RES) for higher power and higher voltage applications. In the review paper to enhance the reliability of Multilevel inverters, a detailed investigation of the study of Power quality improvement using Machine learning techniques is presented. Fault tolerance, fault detection, reduced number of switches, reduced harmonics from the output voltage for calculating (THD) Total harmonic distortion from the output waveform , and ensuring the Modulation index (MI) from the input waveform , are implemented in the review for various applications. In the review, the comparison of Fault tolerance for various levels of Multilevel inverters and a thorough investigation of the comparison of different Machine learning techniques, highlight the accuracy of each Machine learning technique. The research review article explores innovative switching capacitor inverter technology, parameter analysis, significant outcomes, research requirements, and applications in renewable energy integration with the grid, isolated systems, and carbon reduction in grids. The merits, limitations, and methodology of MLI are comprehensively explored. The presentations, evaluations, difficulties, and recommendations assist academics, engineers, and decision-makers in employing renewable integrated inverter solutions to mitigate grid emissions and promote sustainability through Machine learning algorithms .