2002-08-012024-05-17https://scholars.lib.ntu.edu.tw/handle/123456789/684004摘要：本研究乃是參考新近發展的理論，分三階段有系統性的對颱洪時期水庫操作及管理工作進行研究，同時選定石門水庫為研究對象，以準確預測入庫洪水量、提高水庫操作效率並降低水庫操作風險。第二年度之目標除運用前一年度之準確洪水預測量外，再結合智慧型控制理論以提高洪水時期水庫操作之效率，建構一個智慧型水庫颱洪操作模式，不僅可自歷史颱洪事件學習，同時可結合專家經驗與防颱政策，即時地反應洪水流量歷程。<br> Abstract: This study is split into three stages and the Shi-man Reservoir is chosen as the study area. In the second year, intelligent control theory will be introduced to the optimization of a real time reservoir operation model during flood seasons. The ANFIS is built to create the fuzzy inference system, to construct the suitable structure and parameters of the network, and then to estimate the optimal water stage according to the current reservoir watermark and inflow situation. This model is learning through the historical data and can provide valuable output for the reservoir operators.水庫操作調適性網路模糊推論系統智慧型控制理論reservoir operationadaptive-network-based fuzzy inference systemintelligent control theory