Answers are grounded in this paper and cite their sources.
Experimental study demonstrates 96.98% efficiency in an AI-controlled universal power converter for hybrid electric vehicles, highlighting improved power density and reduced component stress.
This paper presents an optimized reduced-switch universal power converter for renewable-energy-assisted electric vehicle charging applications. Unlike conventional universal converters that require a large number of semiconductor devices for different power conversion functions, the proposed topology integrates AC–DC rectification, DC–DC buck/boost conversion, DC–AC inversion, and AC–AC voltage regulation within a single compact architecture using six active switches, four thyristors, two inductors, and two capacitors. The reduced-switch configuration decreases switching stress, conduction losses, and maximum switch current while maintaining bidirectional power conversion capability. A supervised feed-forward artificial neural network (ANN) controller is employed as a supervisory controller to perform operating-mode selection, duty-ratio optimization, firing-angle regulation, and switching-state generation under varying source and load conditions. The ANN is trained offline using MATLAB/Simulink-generated datasets and executed online with low computational complexity, enabling real-time implementation at a 20 kHz switching frequency. The converter is validated through detailed MATLAB/Simulink simulations and a 250 W laboratory prototype operating in all four conversion modes. Experimental investigations demonstrate stable voltage regulation under load variations, a maximum conversion efficiency of 96.98% , and an output total harmonic distortion (THD) of 3.02% , satisfying high-quality power conversion requirements. Compared with conventional universal converters, the proposed converter achieves lower component count, reduced switching stress, and improved conversion efficiency, making it a promising compact power-conditioning solution for renewable-energy-assisted hybrid electric vehicle energy management and multifunctional power conversion systems.
No takes yet. Share an insight, caveat, or question.
Suresh et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: