單相市電併聯換流器簡化控制策略研究

Abstract

由於對環境的關注，推動了再生能源的發展，例如燃料電池、太陽能和風能，上述能源可以當作分佈式發電，而一個微電網可由上述分佈式發電，結合當地發電和儲能設施所組成。在微電網中，並網逆變器可以轉換電力，因此扮演重要的角色，其已廣泛用於住宅太陽能應用中。在傳統控制器中，需要複雜的數學計算才能計算出參考信號，因此，需要複雜且昂貴的數位信號處理控制器來執行程式運算。本論文提出了一種簡化的控制算法，以降低參考信號計算的複雜度，使用提出的簡化控制算法，可以使用低成本之微控制器單元。本論文中對於提出之簡化控制算法的工作原理和數學方程式有詳細推導，可證明其精準度與傳統控制法相比非常接近，僅有0.33%的誤差。此外，提出的簡化控制算法透過MATLAB/Simulink進行模擬，以及TMS320F28335進行編譯，同時製作並測試一台1 kVA之並網電壓源逆變器原型電路，透過電腦模擬和實驗結果驗證提出的簡化控制算法之可行性和其性能，亦證實此簡化控制算法之理論結果。藉由提出的簡化控制算法，可以減少52.3 %的計算量。

The environmental concerns drive renewable energy resources such as fuel cells, photovoltaic, and wind energy as the distributed generation associated with local generation and storage facilities to form a microgrid. In a microgrid, a grid-connected inverter plays an important role in exchanging power to the grid, and it is widely used in residential photovoltaic applications. In a conventional controller, the algorithm requires a complex mathematical calculation to compute the reference signal. Thus, a sophisticated and more costly digital signal processor controller is required to perform the algorithm process. In this dissertation, a simplified control algorithm is developed to reduce the complexity of the reference signal calculation. With the proposed simplified control algorithm, the usage of a low-cost microcontroller unit becomes possible. The operation principle and the mathematical equation of the proposed simplified control algorithm are derived thoroughly. The mathematical derivation validation shows that the accuracy of the proposed simplified control algorithm is very close to the conventional algorithm, with 0.33 % of error percent. Furthermore, the mathematical derivation of the proposed simplified control algorithm is adopted for simulation via MATLAB/Simulink and for digital signal processor programming in the TMS320F28335. A 1 kVA grid-connected voltage source inverter prototype is built and tested. The simulation and experimental results obtained from the prototype verify the feasibility and the performance of the proposed simplified control algorithm. Computer simulations and experimental results are presented to confirm the theoretical results. By using the proposed simplified control algorithm, 52.3 % of the computational burden can be reduced.