2026-04-28T19:35:35Zhttps://uvadoc.uva.es/oai/request

oai:uvadoc.uva.es:10324/833212026-03-04T20:02:52Zcom_10324_38col_10324_852

AI-Driven Efficiency Optimization of a Dual Active Bridge for Bidirectional EV Charging San José Vega, Diego Mazaeda Echevarría, Rogelio Universidad de Valladolid. Escuela de Ingenierías Industriales The rapid global transition toward electromobility and renewable energy systems has highlighted the need for efficient, scalable, and intelligent power conversion technologies. The Dual Active Bridge (DAB) converter, known for its high efficiency and bidirectional operation, has become a popular component in the field of electric vehicle (EV) chargers, where it is used as a DC-DC stage. This research focuses on optimizing the design of the DAB converter for bidirectional EV chargers by integrating advanced computational tools, such as artificial intelligence (AI), to enhance efficiency and reduce the engineering workload in converter design. The primary objective of this project is to develop a practical and generalized workflow that leverages on AI to optimize the efficiency of a DAB converter under a real-time control scenario for a continuous range of load conditions. By utilizing PLECS simulations and AI supervised prediction models, this work aims to identify the most combination of control parameter to achieve peak efficiency operating points for the converter, thereby improving the overall performance of EV chargers. The proposed methodology combines both classification and regression models to predict power losses and efficiency, using a comprehensive dataset derived from simulated DAB converter behaviour. Through the application of this AI-driven workflow, the study demonstrates significant improvements in the efficiency of the converter across a wide range of operating conditions typical of an EV charger with bidirectional nature. The results indicate that AI can effectively enhance the design of power electronics, making the optimization process faster, more accurate, and less reliant on manual engineering interventions. The findings contribute to the ongoing development of high-performance and cost-effective solutions for the next generation EV chargers and renewable energy systems. La transición global hacia la electromovilidad exige conversión de potencia eficiente, escalable e inteligente. 2026-03-04T09:53:50Z 2026-03-04T09:53:50Z 2025 info:eu-repo/semantics/bachelorThesis https://uvadoc.uva.es/handle/10324/83321 eng info:eu-repo/semantics/openAccess http://creativecommons.org/licenses/by-nc-nd/4.0/ Attribution-NonCommercial-NoDerivatives 4.0 Internacional