Generalization of the Transient DC-bias Current Elimination Scheme in Dual Active Bridge Converters using the Real-Time Zero Current Crossing Prediction Algorithm

The Dual Active Bridge (DAB) is a power electronic converter widely used in areas such as transportation, renewable energy sources, and battery-based energy storage systems. This type of converter typically serves as an interface between voltage sources, providing galvanic isolation while operating under varying transferred power. As a result, transient DC-bias current may arise, leading to increased losses and potentially causing saturation of magnetic components. Typically, to overcome the aforementioned negative effects, proper converter control is utilized. One recently proposed method for eliminating DC-bias current is based on the idea of shifting the voltage waveforms in phase so that the current waveforms, at the beginning of the switching cycle, start from zero. This approach enables a seamless power transition. However, the aforementioned method has been implemented only for the specific modes of TPS modulation. Another aspect is the dead-time effect, for which compensation has not been shown in a straightforward manner. Given the above, the algorithm called Real-Time Zero Current Prediction (RT-ZCCP) is proposed to augment the known solution with an arbitrary modulation scheme compatibility, along with a simple dead-time compensation technique. This paper outlines the RT-ZCCP algorithm in a step-by-step manner, including the detailed analysis of the dead-time effect, and proposes an optimal method to choose the corrective time value. The circuit-parameter-dependent robustness analysis of the proposed algorithm has also been performed. The RT-ZCCP algorithm was implemented on the TMS320F28379D microcontroller and has been successfully verified on the 1.3 kW laboratory prototype with SiC-based MOSFET modules, operating at a switching frequency of 40 kHz. What is more, the experimental results comparison with the other competitive methods has been performed to prove the effectiveness of the proposed RT-ZCCP algorithm.