| Summary: | 碩士 === 國立中山大學 === 電機工程學系研究所 === 102 === With the development of smart grids, more and more Distributed Generation Systems (DGSs) and Renewable Energy Generation Systems (REGSs) will be interconnected to power grids in the foreseeing future. Improving the output efficiency of REGSs is one of the most important issues to increase the electrical energy of REGSs. Due to the limited capacity of a single inverter, parallel inverters are often used for the interconnection of DGSs and REGSs. The equal current sharing is the commonly-used control method for parallel inverters; however, the output efficiency based on the equal current sharing is low at light loads. Therefore, this thesis proposes control strategies to improve the output efficiency of parallel inverters. The proposed control strategies is designed based on the efficiency output curve of a single inverter and then a Particle Swarm Optimization (PSO) algorithm is used to find the optimal output efficiency of parallel inverters. The Texas Instruments (TI) TMS320F28335 is used to implement a fully digital inverter with the rated output voltage and output power 220V and 2 kW, respectively. Two implemented inverters are used to validate the performance of the proposed control strategies for parallel inverters. Experimental results show that the proposed control strategies can improve the output efficiency of parallel inverters at light loads. The integration of the proposed control strategies into a Photovoltaic Generation System (PVGS) with 100kWp rated output power interconnected by 10 inverters with 10kW rated output is also simulated in this thesis. Simulation results demonstrate that the proposed control strategies can really increase the electrical energy of PVGS.
|
|---|
