Model-based adaptive control of modular DAB converter for EV chargers

This paper presents the discrete-time modelling and control of modular input-parallel–output-parallel (IPOP) dual-active-bridge (DAB) converters for electric vehicle (EV) charging. The proposed adaptive control system ensures adequate current-sharing among parallel modules while minimizing DAB current stress by adopting dual phase-shift modulation. Driven by the growing need for fast EV charging options, the paper highlights the importance of achieving top-notch control performance, especially with varying load conditions. Specifically, it introduces a discrete-time model for adjusting controller parameters adaptively, which simplifies the typically cumbersome manual tuning process associated with these systems. The proposed PI formulae are derived to satisfy specifications on the frequency domain as phase margin and the gain crossover frequency of the open loop gain transfer function, ensuring stability and robustness in operation. Moreover, the implementation of these formulae in discrete microcontrollers facilitates seamless PI autotuning for precise current, voltage, or power control. Notably, the proposed control strategy effectively mitigates current overshot issues commonly encountered during module engagement and shedding operations in modular EV chargers. To validate its efficacy, the proposed controller is evaluated through extensive testing and comparisons within the PLECS environment, particularly focusing on a two-module IPOP-DAB converter scenario, and including comparisons with classical offline model-based pole placement methodology. Furthermore, real-time hardware-in-the-loop experiments are conducted to confirm the feasibility and performance of the proposed controller under realistic EV charging profiles.