Design Methodology of Highly Efficient Soft-Switching Two-Stage Power Converter

Abstract

Recently, interest and demand for renewable energy such as solar cells and wind turbines, batteries, and fuel cells have increased. For this reason, in the industry, various attempts and efforts have been actively conducted to incorporate renewable energy into one system in connection with a system composed of an existing distribution network. For this purpose, a two-stage power conversion system with cascaded boost converter and inverter, which can cover a wide input voltage range and can be connected to the grid, is mainly used in these applications. Meanwhile, in the field of power electronics, the demand for high power density such as miniaturization, light weight and small volume is continuously required in the power conversion system in order to be applied to various applications. The simplest and easiest way to achieve low volume is to reduce the size of the output filter consisting of inductor and capacitor by increasing the switching frequency. However, increasing the switching frequency can not only increase the switching loss, but also reduce the efficiency of the overall system. In addition, power density may not be improved because larger heat dissipation facilities are required to remove heat caused by high switching losses. For this reason, in order to overcome the problem of simply increasing the switching frequency, various studies have been conducted to improve the power density of power conversion systems. One of the methods to increase the power density is to construct a power conversion system using WBG devices such as Gallium nitride(GaN) and Silicon carbide(SiC), which have recently emerged. WBG devices are attracting attention as a substitute for Si devices because of their relatively low switching and conduction losses compared to Si devices, but they are still lacking in commercialization due to their high price and immature driving technology compared to Si devices. For this reason, soft switching technologies such as zero-voltage transition(ZVT) and zero-current transition(ZCT), which increase power density by reducing switching losses, have been studied for decades. This paper includes studies of a highly efficient soft-switching two-stage power converter and their design methodology. The proposed high efficiency two-stage power converter consists of a front-end ZCT boost converter and a back-end ZVT single-phase full-bridge inverter. First, in front-end of the two-stage power converter, by analyzing the operation of the conventional ZCT boost converter, analysis of additional conduction losses caused by the operation of the auxiliary circuit is performed when soft switching is applied, and a design technique is proposed to minimize additional conduction losses. The operation of the auxiliary circuit part consisting of auxiliary resonant parameters L and C creates an additional deformed current region compared to the circuit when soft switching is not applied. The design of an erroneous auxiliary circuit that does not sufficiently consider the added current region not only does not improve the switching loss, but also reduces the overall efficiency due to the additional conduction loss, so a design considering the additional current region is necessary. In this paper, additional conduction losses are calculated by mathematically analyzing the shape of the deformed current due to the operation of the auxiliary circuit using the existing ZCT boost converter. In addition, to properly apply ZCT, a ZCT boost converter design method is proposed to minimize additional conduction losses based on mathematical analysis. To verify the effectiveness of the proposed design method, additional conduction losses of the ZCT boost converter at various design points using auxiliary resonance parameters are compared by various experiments and analysis. In back-end of the two-stage power converter, a bipolar switching single-phase full-bridge ZVT inverter using an additive polarity coupled inductor is proposed. The proposed circuit uses a coupled inductor, so there is no voltage imbalance caused by the split capacitor. Also, applying ZVT to one leg usually requires one auxiliary circuit part, while the proposed circuit requires only one auxiliary circuit part to apply ZVT to two main legs. Therefore, the proposed circuit may be simpler and more economical in terms of cost because the number of elements is reduced in order to apply ZVT compared to the existing circuit. In this paper, all operation steps of the proposed circuit are explained and analyzed by each mode analysis of one switching cycle. Then, design considerations for implementing the proposed circuit are discussed. Finally, the feasibility of the proposed circuit is verified by various experiments and simulations.