Design of a novel robust adaptive PI controller for dual active bridge converter enhanced by transfer learning algorithm and snake optimization

Author Identifier (ORCID)

Mehrdad Ghahramani: https://orcid.org/0000-0002-5926-0996

Daryoush Habibi: https://orcid.org/0000-0002-7662-6830

Asma Aziz: https://orcid.org/0000-0003-3538-0536

Abstract

This paper presents a novel robust adaptive Proportional-Integral (PI) controller for a Dual Active Bridge (DAB) converter, enhanced through Transfer Learning and the Snake Optimization algorithm. The DAB converter exhibits complex nonlinear dynamics due to its bidirectional power flow, high-frequency operation, and resonant elements, posing significant challenges for conventional control strategies. To address these challenges, a flexible and adaptive control framework is developed that dynamically adjusts to variations in operating conditions and system parameters. The proposed controller combines an adaptive PI structure with Transfer Learning (TL) to utilize previously acquired control knowledge for improving learning efficiency and robustness across different scenarios. The Snake Optimization Algorithm (SOA) is employed to fine-tune the controller gains, ensuring superior transient performance and minimal steady-state error. Simulation results under diverse conditions - including parameter uncertainties and load changes - demonstrate the effectiveness of the method. Furthermore, experimental validation using a Hardware-in-Loop (HIL) setup confirms the real-time feasibility and performance of the proposed control strategy in managing the complex dynamics of the DAB converter.

Keywords

Dual active bridge converter, PI controller, robust, snake optimization, transfer learning algorithm

Document Type

Conference Proceeding

Date of Publication

1-1-2025

Publication Title

2025 4th International Conference on Power Systems and Electrical Technology (PSET)

Publisher

IEEE

School

School of Engineering

Comments

Ghamari, S. M., Ghahramani, M., Habibi, D., & Aziz, A. (2025). Design of a novel robust adaptive PI controller for dual active bridge converter enhanced by transfer learning algorithm and snake optimization. In 2025 4th International Conference on Power Systems and Electrical Technology (PSET) (pp. 108–113). IEEE. https://doi.org/10.1109/PSET66242.2025.11296490

Copyright

subscription content

First Page

108

Last Page

113

Share

 
COinS
 

Link to publisher version (DOI)

10.1109/PSET66242.2025.11296490