Moosavi Torshizi, Seyed Morteza (2017-12). Soft-Switching High-Frequency Link Converters with Reduced Voltage Stress. Doctoral Dissertation. Thesis uri icon

abstract

  • Power converters with stacked cells or stacked semiconductor switches have the widely-known advantage of achieving higher voltage or current levels by using standard low-voltage or low-current switches. This dissertation presents several galvanically isolated power electronic converters for various applications and is mainly focused on reducing voltage stress on the semiconductor switches. The proposed topologies convert power in a single stage and achieve soft-switching throughout their range of operation without utilizing snubbing or auxiliary networks. The proposed topologies are divided into three categories; DC-DC, DC-AC, and AC-AC. In each category, two power conversion topologies are presented. All proposed converters include at least one multiple-winding transformer and a resonant capacitor on each winding to form a partial-resonant AC-link which facilitates soft turn-on and turn-off transitions for all semiconductor devices. The magnetizing inductance of the transformer is the key component in energy transfer. Due to non-appearance of a DC link, the proposed converters do not require bulky electrolytic capacitors to convert power. Since the AC-link operates at several kilo-hertz, the transformer size and weight are significantly smaller than the conventional line-frequency transformer. Reduction of switch voltage stress is achieved through a sequential connection of cells. Each cell is composed of one transformer winding, one resonant capacitor, and one or more semiconductor devices. The reduction factor in the voltage stress is directly proportional to the number of cells. The energy stored in winding leakage inductances is fed to the AC capacitors to mitigate switching spikes without using auxiliary snubbing switches. The transformer design is simple because the inverter operation and soft-switching are not affected by the transformer non-idealities

publication date

  • December 2017