- 學位論文
使用實數編碼之演化式計算應用於配電系統多時段饋線重構問題之研究
Application of real-coding on evolutionary computation for distribution system feeder reconfiguration problems under load variations.
摘要
由於環保意識的抬頭，節能減碳的理念逐漸受到重視，如何讓配電系統在正常運行的情況下，使其運轉處於最佳的狀態以降低能源的消耗是非常重要的課題，為達此一目的，饋線重構為相當重要的一項技術。饋線重構主要是在滿足輻射狀的情況之下，透過開關的切換以改變配電系統的架構。因此饋線重構可視為一種組合性最佳化的問題。由於配電系統上開關數量眾多，面對此一問題，可行解空間開關操作策略的數量龐大，有效找尋到最佳的開關操作策略來達成饋線重構的目的變成相當的重要。有鑒於實際配電系統之負載是隨著時間而有所變動，本論文之相關模擬分別考慮配電系統在單時段以及多時段的情況之下，進行單目標以及多目標之分析，探討基因演算法搭配實數編碼方式應用於饋線重構問題上之可行性，並且與其他編碼方式比較其效能，於多目標分析部分本論文改良了既有的TOPSIS多屬性決策法以計算演算法之適應值，期望能有效求解多目標饋線重構之問題。
並列摘要
As the concepts of environmental protection and energy saving bring more attention, how to reduce the energy loss during the normal operations of distribution system becomes an important issue. Feeder reconfiguration is a very important technique that can be used to deal with different types of distribution system problems. By changing the distribution system structure the distribution system can be operated in a more efficient way during normal and contingency operations. Feeder reconfiguration is a typical combinatorial optimization problem. Due to the large amount of switches on a distribution system, the possible solutions of the switching operation plans increase dramatically. Therefore, searching for the best switching operation plan to accomplish the feeder reconfiguration becomes an important issue. This paper applies real-coding of Genetic Algorithm for single- and multi-objectives feeder reconfiguration under fixed load and various load conditions. The searching efficiency and stability with other coding methods are compared. In the multi-objectives feeder reconfiguration problems, the improved TOPSIS is applied to calculate the fitness value of the Genetic Algorithm in order to effectively solve the feeder reconfiguration problems.