- 學位論文
考慮最小成本與需量反應之分散式電源規劃及單日空調排程
Distributed Generation Planning Considering Cost Minimization and Demand Response and Day-ahead kW Scheduling of Air-conditioning
摘要
近年來隨者氣候變遷所帶來的環境議題影響日漸加重，使得分散式電源與需量反應的研究與發展更加熱絡。本論文使用模糊C均值法 (Fuzzy-c-Mean, FCM)、馬可夫模型及內部點法求解小型電力系統之再生能源電源規劃與單日最佳室內溫度排程問題。首先應用FCM分類法將風力發電機、太陽能電池發電量、非彈性負載及室外溫度分類，以便進行馬可夫模型分析。使用馬可夫模型可計算其機率、頻率及持續時間，可加快內部點法運算速度。最後考慮舒適度與需量反應，並結合內部點法求出符合最小成本之再生能源電源規劃。再應用自適應演算法與內部點法，對確定和不確定性的氣候與負載做單日最佳化空調排程(夏、冬季)。本論文假設環境為50戶的小型社區(非彈性負載)，每1戶裝有1台冷暖氣機，共50台(彈性負載)。本論文應用馬可夫模型可使內部點法模擬時間較時間依序法減少近3-5倍且最小成本之誤差皆於6.4%以內。
並列摘要
In recent years, because the impact of global climate change on the environment is a crucial issue, research and development of distributed generation and demand response become important. In this thesis, the Fuzzy-c-Mean (FCM), Markov model, and interior point method are used for sizing renewable energy power of a small power system, and for studying the day-ahead indoor temperature scheduling problems. Using FCM, wind power, solar power, inelastic load, and outdoor temperature are clustered. The Markov model can be used to calculate their probability, frequency, and duration and thus speed up the computational speed of the internal point method. Then, the interior point method considering the comfort and demand response is used to calculate the minimum cost of renewable power generation. Finally, by applying self-adaptive algorithms and interior point method, the optimal kW-consumption scheduling of air-conditioners in a single day (summer and winter) is studied. In this thesis, a small community including 50end-users who have an individual air-conditioner was studied. The difference between the results obtained by Markov model-based and time sequence-based interior point method is less than 5%. The CPU time used by the proposed method can be less 3-5times.