Multilevel Dual Active Bridge Leakage Inductance Selection for Various DC-Link Voltage Spans

The leakage inductance of the transformer in a dual active bridge (DAB) dc-dc converter directly impacts the ac current waveforms and the power factor; thus, it can be considered a design requirement for the transformer. In the existing literature, a choice is made to either ensure soft switching in nominal power or to minimize the RMS current of the transformer. The inductance is typically obtained using optimization procedures. Implementing these optimizations is time-consuming, which can be avoided if a closed-form equation is derived for the optimum leakage inductance. In this paper, analytical formulas are derived to estimate the desired leakage inductance such that the highest RMS value of the current in the operation region of a DAB is kept to its minimum value. The accuracy and sensitivity of the analytical solutions are evaluated. It is shown that in a large design domain, the solution for the YY-connected MFT has a less than 3% error compared to the results obtained from an optimization engine. As an example of the importance of selecting the leakage inductance correctly, it is shown that for 11% deviations in the dc link voltages, a 10% deviation from the desired leakage inductance value can cause 2% higher RMS currents in the converter.

Automated summary

This paper proposes deriving analytical formulas to estimate the desired leakage inductance for the efficient operation of a dual active bridge (DAB) dc-dc converter. The accuracy and sensitivity of the analytical solutions are evaluated and compared to optimization results. It is shown that the analytical solutions have less than 3% error compared to the optimization engine, highlighting the importance of selecting the correct leakage inductance for optimal converter performance.