2003-05-132024-05-14https://scholars.lib.ntu.edu.tw/handle/123456789/656078摘要：台灣地區夏季易遭受颱風侵襲，常伴隨著豪雨出現，中上游多為高山及丘陵地帶，致使水流急促，往往挾帶大量砂石沖刷而下，然下游則因平原地形使坡度驟減故流速減緩而宣洩不良，氾濫成災，因此，建置河川流量/水位預報模式將有助於流域的洪災預警功能。 本計畫為建置流域「智慧型洪災分析模式」之先驅研究，以為台灣地區各流域之示範模式。本研究以目前廣被使用的類神經網路建立蘭陽溪流域之水位預測模式，並就模式架構之精確性、穩定性與實用性進行探討。同時，為落實洪災預警模式的可行性，將串聯結合模式推估、即時資料自動傳輸與水文知識與新技術，並透過線上模式分析，提供即時洪災預測與研擬決策之洪災資訊系統，達成水文資訊之即時化。<br> Abstract: Taiwan has a subtropical climate where typhoons, usually coupled with heavy rainfall, hit the island around four times a year, causing downstream flooding within a few hours. Consequently, streamflow forecasting is crucial for flood warning system. This pioneer study, which constructs an intelligent flood forecasting system of the Lan Young river, would provide a pattern for constructing similar systems of other rivers in the future. In this study, we apply the artificial neural networks (ANNs) as the rainfall-runoff model and explore the accuracy, stability and practicability of this model. For the practicable purpose, the forecasting analysis model, the coming data and new technology are integrated to provide the flood information for the decision-maker through on-line analysis.智慧型洪災分析模式類神經網路蘭陽溪intelligent flood forecasting systemartificial neural networksLan Young river