Artificial Intelligence Techniques for Dual-Stage Interleaved Electric Vehicle Onboard Chargers: Trends and Challenges

Abstract

The rapid global adoption of electric vehicles has increased the demand for efficient, compact, and intelligent onboard charging systems. Dual-stage interleaved onboard chargers have emerged as a preferred architecture due to their ability to reduce current ripple, improve power density, enhance thermal performance, and achieve high efficiency with low harmonic distortion.

This paper presents a comprehensive review of advanced control and optimization techniques for dual-stage interleaved onboard chargers. It highlights the use of PIDD2-PD controllers, which extend classical PID structures by incorporating higher-order derivatives for improved transient response, disturbance rejection, and steady-state accuracy. The study further explores the Hybrid Adaptive Genghis Khan Shark Gold Rush (HAGKSGR) algorithm for optimizing controller parameters and switching strategies, enabling efficient handling of nonlinear and multi-objective optimization challenges in dynamic charging environments.

Applications include Level 1 and Level 2 electric vehicle charging systems, focusing on improving power factor, minimizing total harmonic distortion, and enhancing overall system efficiency. Simulation and hardware-in-the-loop studies demonstrate superior performance compared to conventional control and optimization methods. However, challenges such as computational complexity, real-time implementation, and scalability remain. This review emphasizes the potential of integrating advanced control architectures with hybrid metaheuristic optimization to develop high-performance onboard charging systems for next-generation electric vehicles.