- 學位論文
遺傳規劃法於模糊建模與負載排程之應用
Application of Genetic Programming to Fuzzy Modeling and the Schedule of Direct Load Control
摘要
遺傳規劃法能被設計在搜尋結構解的應用上，主要是遺傳規劃法擁有著強大得搜尋能力，且其透過樹狀結構的交配演化而得到更好的結構解，故在本研究中將利用遺傳規劃法來解決兩個最佳化組合的問題。 (1).遺傳規劃法於模糊建模之應用：本篇論文中是談論有關以普遍的方法去發展模糊建模，而模糊建模與數值建模最大的不同在於，模糊建模能在一個模糊集合中透露一些訊息給我們，且這個觀點也提醒著我們，在建模的同時，除了其精確度的問題外，尚須考慮到一個模糊建模的透明度的問題。 (2).遺傳規劃法於負載排程之應用：其主要是利用遺傳規劃法搜尋結構的特性，在尖峰負載時找出合適的空調來受控，而此受控情形即對空調負載作短暫的停機動作，由於部分空調系統受控，故能有效的抑制尖峰負載。
並列摘要
Based on a good searching ability of structure, Genetic Programming is designed to search the structural solution and to utilize crossover mechanism of tree structure to get the better structure. In this thesis, we will use the performance of Genetic Programming to solve two optimal questions. (1).Application of Genetic Programming to fuzzy modeling: This study is concerned with a general methodology of identification of fuzzy models. Unlike other numeric models, fuzzy models operate at a level of information granules (fuzzy sets), and this aspect brings up an important requirement of design on about the transparency of the model. (2).Application of Genetic Programming to the schedule of direct load control: Based on the searching ability of Genetic Programming, we can find an optimal control strategy to reduce the peak load.