Parameter Identification and Adaptive Control of a High Power Medium Voltage Dual Active Bridge

A unified control software is typically designed and tuned for structurally identical Dual Active Bridges (DABs) based on worst-case hardware parameters. However, improving dynamic behavior, efficiency and resource utilization requires precise knowledge of each individual unit, since even identically built systems may exhibit different characteristics due to component tolerances and varying operating conditions. This necessitates either additional sensors or individual testing and adaptation of each DAB’s control algorithm. To address this limitation, this work presents an online parameter identification method that estimates the leakage inductance with less than 2% deviation without requiring extra sensors. Additionally, the identified inductance is used to adjust the control algorithm in real time, thereby improving efficiency and enabling higher maximum output power by relaxing conservative software constraints, while ensuring operation within the safe operating area.