Minimum-Current-Stress Scheme of Three-Level Dual-Active-Bridge DC–DC Converters With the Particle Swarm Optimization

Current stress is essential for the reliability of power switching devices, especially for high-voltage applications, such as three-level NPC-based dual-active-bridge (3LNPC-DAB) converters. With the conventional single phase shift (SPS) control, the current stress of power devices in 3LNPC-DAB converters is high, which not only affects the conversion efficiency for a wide operating range but also endangers the power devices, which is harmful for the healthy operation of 3LNPC-DAB converters. To solve this problem, this article proposes a minimum-current-stress scheme using particle swarm optimization (PSO) strategy with the aim to reduce the current stress and improve the efficiency of 3LNPC-DAB converters. Different from SPS, two phase shift variables are adopted in the proposed algorithm: one is set between the primary and secondary bridges, and the other is set within the same bridge. Thus, not only the control flexibility is expanded, but also main performance indexes, such as the transmission power, the current stress of each switch, and the efficiency for the whole operating range, are improved. Moreover, the PSO strategy requires fewer parameters to adjust, less computational burden, and faster convergence speed, which simplifies the theoretical calculation and the practical implementation. Finally, the main experimental results for various operating conditions are provided to validate the proposed scheme.