2008-08-012024-05-18https://scholars.lib.ntu.edu.tw/handle/123456789/705119摘要：總計畫： 本整合型計畫是「都市積淹水即時預報」的先導研究。根據預報系統的需求，原本邀請這方面的專家共同參與，分為總計畫與七項子題，擬以三年期間完成系統雛型研發；第一年審查通過四項子題：(1)精密降水估計雷達觀測策略研擬，(2)應用微機電感測器與Zigbee無線監測網路於即時水位監測之研發，(3)渠道管網演算與資料同化模式研發，(4)分散式都市淹水模擬與支援決策系統框架之研發。採用台北市中港抽水站集水區作為「自然實驗室」，設置水位與流量觀測儀器，利用中央大學的雙偏極化都卜勒雷達，透過密集觀測，了解都市降雨－逕流歷線轉換關係，以及可能發生積淹水的過程。同時，研究發展數學模式，逐步達成「都市積淹水即時預報」必須具備的各項科學技術、模擬程式、演算系統和展示工具等。 因應逕流模式、淹水模式和淹水展示等三個子計畫被刪除的計畫執行方式：(1)子計畫三利用地貌型逕流模式開發山區逕流，利用SWMM模式和密集觀測結果推估降雨－逕流關係；(2)將淹水模式限制下水道滿管的一維市街流模式，納入子計畫三執行；(3)總計畫擬利用協助水利署制訂網頁服務（web service）即時水文氣象資料格式、以標準介面模組化洪水預報水文模式，和以可延伸標示語言（eXtensible Markup Language, XML）編寫工作流程（workflow）的經驗；規劃與制訂「都市積淹水潛勢預報系統」的a.網頁服務（web service）即時水文氣象資料格式與資料庫的概要綱目（schema），b.水文模組之間資料交換的標準介面XML檔案，以利模組化都市積淹水即時預報水文模式，c.執行不同預報水文模組與指令的XML工作流程，以及d.元件化展示介面。 子計畫： 暴雨時期地表水與下水道內之水流交換頻繁，地表因可能跨區漫淹而沒有清楚的集水區邊界，因此都市積淹水的即時模擬，必須在利用精密降水估計結果估計地表逕流後，同時考量地表水與下水道垂直方向及都市區塊間水平方向的交換機制，才能得到合理的計算結果。 由於都市地文條件複雜且缺乏經常性觀測且不受迴水影響的流量站，傳統集水區降雨轉換為逕流歷線的經驗方法不能適用，必須透過密集水文試驗方可取得高時間、空間解析度之經驗關係。本研究擬使用地貌型模式與SWMM RUNOFF模組分別估計山區逕流與都市地表逕流，並採用黑盒分析法配合即時水位觀測資料，發展推估無密集水文觀測集水區降雨－逕流經驗關係的方法。 由於降雨觀測和降雨—逕流歷線演算均存在不確定性，其影響都將反映在下水道水理模式模擬結果之中。加上都市集水區面積小、渠道管網容量小的特性，模擬計算結果與觀測資料的誤差往往不可忽略。本計畫擬藉由以NewC數值法開發之下水道管網與市街流聯合演算模式，採用資料同化技術，結合伴隨狀態法(adjoint state method，簡稱ASM)與「擬牛頓法(quasi-Newton method)」，利用下水道水位觀測資料修正模式狀態，避免因為不確定因素的影響使誤差持續累積。 本研究將以台北市中港抽水站之集水範圍進行模式測試與結果分析，並評估模式平行化計算之可行性。最後對於建置「都市積淹水即時預報系統」提供詳實之建議。<br> Abstract: 總計畫： This is an integrated pilot study for the development of the Urban Inundation-Potential Nowcasting System. It is planned to establish all the observations and hydrologic models needed to construct a prototype system. Originally, there are 7 topics, but only 4 are approved. The focus of these four are: (1) Develop radar scanning strategies for accurate estimate of quantitative precipitation. (2) Device and install MEMS pressure sensors and transducers in storm sewer for the intensive observation of stage hydrograph to enable the development of black-box rainfall-runoff relationships. (3) Develop pipe/open channel simulation model able to do trans-pressurized/free surface flow and sewer/street network coupled flow, and able to assimilate stage data using Adjoint State Method. (4) Design a distributed computation system. Using the Jhonggang pumping station catchment area and sewer system of Taipei City as a “natural laboratory”, a few sites will be selected to construct stage measuring instruments and conduct rainfall-runoff intensive observations. Radar QPE data and hydrograph data are used to derive and calibrate rainfall/runoff relationships. The mathematical models and graphical display modules needed to establish an Urban Inundation-Potential Nowcasting System prototype are to be developed. The adjustments corresponding to the omission of Runoff model and Inundation model development as well as the 4D Display development 3 projects are: (1) The new Project 3 will utilize the Geomophological model to estimate hilly catchment runoff and SWMM model for sewer stage and discharge computation. (2) The new Project 3 will be responsible to develop inundation model caused by sewer overflowing and it is able to simulate 1-D street network inundation flow. (3) Based on the experiences developing Distributed Flood Forecasting System for Water Resources Agency, the Overall Project is obligated to establish the followings for urban flash flood forecasting system: 1. The web-service real-time meteo-hydrology data standard format and schema. 2. The standardized data exchange XML files among hydrologic models. 3. The workflow using XML. 4. The reusable display modules showing inundation maps and time series data, etc. 子計畫： In urban area, 2-way vertical flow between surface runoff and storm sewer is not rare. Horizontal exchange among urban catchments may also occur during inundation. A storm sewer numerical model must have the capability to simulate these situations to be able to correctly reproduce inundation events. Owing to the complex land-use, the lack of densed gauging stations and back-water effect in urban sewer system, the rainfall-runoff black-box relationships cannot be derived in the current sewer system. Intensive Observation Experiments are needed to establish them. In this study, water stage gauges will be constructed in sewer system, and the observation data are to be utilized to adjust the runoff hydrographs derived from the terrain-based runoff model and SWMM RUNOFF module. For the catchment areas and the storage capacities are small for urban sewer, simulated stage hydrographs using observed rainfall may differ from the measured ones. In order to avoid error accumulation, a data-assimilation scheme using the adjoint state method and quasi-Newton solution technique are to be developed for the coupled model of sewer pipe and urban street flow. Lateral discharge is considered uncertain and can be adjusted such that the differences between the observed and simulated stage are minimized. The catchment area of Jhonggang pumping station in Taipei City is chosen be the case study area. Feasibility of parallel computation using PC cluster will be studied. Results of this research will be used to provide sensible suggestions for establishing a `real-time urban inundation forecast system`.網頁服務可延伸標示語言工作流程NewC法地貌型逕流模式市街流與下水道水理數值模式資料同化伴隨狀態法都市積淹水即時預報系統web serviceXMLwork flowNewC schemeterrain-based runoff modelSWMM RUNOFFurban street flow and storm sewer pipe flow modeldata assimilationadjoint state methodurban inundation forecast system