2004-01-012024-05-14https://scholars.lib.ntu.edu.tw/handle/123456789/660824摘要：台灣地處西太平洋颱風路徑之要衝，經常在每年七、八月間面臨颱風之侵襲，因此，颱風降雨量預報是台灣地區天氣預報與災害防範作業中相當重要之一環。本計畫建立類神經網路颱風降雨預報模式，第一年先以八個颱風特徵因子來作為類神經網路模式之輸入，未來再以降雨空間分佈資訊來強化模式的預測能力，而半變異圖則用來篩選雨量站資料。研究成果顯示類神經網路網路具有處理非線性關係的能力，成功地將複雜的颱風降雨機制記憶在類神經網路架構內，以達到推估颱風降雨之目的，再經由電子計算機的高速運算能力，能即時推估出目標時段的颱風降雨量，增進颱風降雨預報之效率與準確度。本計畫之研究成果可提供政府相關部門進行颱風降雨預報之參考。<br> Abstract: Typhoon rainfall forecasting has always been a challenging task for hydrologist, water resources engineers and managers in Taiwan. In this project, a neural network will be developed to forecast the typhoon rainfall. During the first year, the model configuration has been evaluated using eight typhoon characteristics. For the next years, the influence of spatial rainfall information on the rainfall forecasting will be considered for improving the model design. Besides, the semivariogram will be applied to determine the required number of nearby rain gauges whose rainfall information will be used as input to the model too. With the typhoon characteristics and the spatial rainfall information as input to the model, the forecasting model will be able to produce reasonable forecasts. The results of this project can serve as references for typhoon rainfall forecasting in Taiwan.颱風降雨類神經網路預報Typhoon rainfallArtificial neural networksforecast