Full-Power-Range Multiobjective Optimization Scheme for Three-Level ANPC-DAB-Based Distributed Generation System

Due to high-voltage withstanding ability, high control flexibility, and bidirectional power transmission ability, inside distributed generation (DG) system, a three-level active neutral-point-clamped dual active bridge (ANPC-DAB) converter can be employed to connect dc bus and energy storage modules. In this article, first, the operation principle, power characteristics, and current characteristics under different ANPC-DAB modes are discussed. For efficiency promotion strategy, rms current optimization will narrow the soft-switching region greatly. Therefore, second, a full-power-range multiobjective optimization scheme is proposed to reduce the current level and widen the soft-switching region. Based on nondominated sorting genetic algorithm II (NSGA-II), rms current and peak current can be optimized simultaneously. Meanwhile, the current Pareto-optimal curve can be established. All switches zero voltage switching (ZVS) condition can also be realized and ZVS boundary can also be established. Through comprehensive comparisons among existing modulations, multiobjective optimization based on NSGA-II can achieve better efficiency promotion. Finally, an ANPC-DAB prototype is established and the effectiveness and correctness of the proposed method can be proved.