2014-05-162024-05-14https://scholars.lib.ntu.edu.tw/handle/123456789/660056摘要：都市防洪是高度開發城市中的重要議題，近年來由於都市化和氣候變遷導致洪峰流量快速上升，使得此項工作面臨極大挑戰。因此，建立一有效率且準確的模式來預測洪汛時期市區抽水站之前池水位有其迫切性與重要性。本計畫以臺北市玉成與新民權抽水站區域為研究對象。近年來，人工智慧技術運用於處理高度複雜之非線性系統有出色的處理能力，此計畫運用該技術建立抽水站前池水位預報模式。首先，蒐集取自於各種水文計量站之資料，如雨量站雨量、抽水站內外水位以及下水道系統之水位，透過相關性分析和考量實際狀況，進而決定雨量對前池水位之影響時距；其次，以Gamma test篩選出顯著影響前池水位之有效雨量站點；接著以倒傳遞類神經網路(BPNN)、Elman類神經網路與NARX類神經網路，分別建構多時刻水位預報模式; 最後評估多時刻水位預報模式之準確度與穩定性。結果顯示，各模式之水位預報準確度與穩定性皆相當高，可有效率且準確的預測洪汛時期臺北市玉成與新民權抽水站之前池水位提供有關政府單位進行都市防洪作業。<br> Abstract: Urban flood control is a crucial task in developed cities, which face a great challenge of fast rising peak flows resulting from urbanization and climate change. To mitigate future flood damages, it is imperative to construct an accurate real-time model to forecast inundation levels during &#64258;ood periods. Artificial intelligence techniques possess an outstanding ability to handle highly non-linear complex systems and are implemented to make real-time water level forecasts in this project. The Yu-Cheng and the New-Ming-Chuan pumping stations located in Taipei City are selected as the study areas. Firstly, historical hydrologic data such as rainfall, water levels of the floodwater storage pond (FSP) at the pumping station, water levels in the sewerage system, and the amount of pumped water are fully explored by statistical techniques to identify the time span of rainfall affecting the rise of the water level in the floodwater storage pond (FSP) at each pumping station. Secondly, effective factors (rainfall stations) that significantly affect the FSP water level are extracted by the Gamma test (GT). Thirdly, the backpropagation neural network (BPNN), the Elman neural network (Elman NN) and the nonlinear autoregressive network with exogenous inputs (NARX) network are used to construct intelligent real-time FSP water level forecast models. Finally, models are evaluated with respect to forecast accuracy and model reliability. The results demonstrate that each of the proposed models has an extraordinary ability with high accuracy in real-time FSP water level forecasts, which suggests that the propose models can be valuable and beneficial to the government authority for urban flood control.類神經網路NARX類神經網路Gamma test洪水預報都市防洪抽水站操作。Arti&#64257cial neural networks (ANNs)Nonlinear autoregressive network with exogenous inputs (NARX)Gamma testFlood forecastUrban flood controlPumping station operation.