頁籤選單縮合
題名 | 適應性網路模糊推論系統在洪水演算之研究=A Study of Adaptive-Network-Based Fuzzy Inference System for River Flood Routing |
---|---|
作者姓名(中文) | 段鏞; 傅金城; | 書刊名 | 農業工程學報 |
卷期 | 50:3 2004.09[民93.09] |
頁次 | 頁71-81 |
分類號 | 443.42 |
關鍵詞 | 適應性網路模糊推論系統; 洪水演算; 複合式學習; Adaptive-network-based fuzzy inference system; Flood routing; Hybrid-learning; |
語文 | 中文(Chinese) |
中文摘要 | 本文將複合式學習的適應性網路模糊推論系統應用於河川洪水演算,由複合式學習方式,建構規則庫及優選分佈之歸屬函數,應用於求解河川之洪水演算問題。在複合式學習的過程中,適應性網路模糊推論系統可以利用模糊規則建構輸入與輸出間的對應關係,可適用於非線性函數與非線性向量之系統。本文利用四場颱洪事件,以新海橋、中正橋和河口之即時水位觀測值為邊界條件預測台北橋水位,經由模擬結果顯示以即時水位及第1小時預報結果較佳。 |
英文摘要 | This study presents the application of hybrid-learning Adaptive-Network-Based Fuzzy Inference System (ANFIS) in river flood routing. By using the hybrid-learning algorithm in ANFIS model, the rule database and optimal distribution of membership functions have been constructed to solve the river flood routing problems. Based on the fuzzy rules, the ANFIS model establishes a mapping relationship between input and output in the processes of the hybrid-learning algorithm. The algorithm can be applied to nonlinear functions or nonlinear components. Four typhoon flood events are simulated to predict the water stage of Taipei Bridge using the real-time observed stages of the Hsinhai Bridge, Chungcheng Bridge and river mouth. The results show that the ANFIS model has good accuracy for real-time and 1-hour stage forecasting in the Tanshui River. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。